CONTROL STRATEGIES FOR IN-USE VEHICLES
U. S. Environmental Protection Agency
Office of Air and Water Programs
Mobile Source Pollution Control Program
Washington, D.C. 20460
November 1972
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
November 29, 1972
This report sets forth the findings and conclusions of work
sponsored by the Environmental Protection Agency, or available to
the EPA from other sources, on the technological feasibility,
effectiveness, and costs of reducing the emission of air pollutants
from automobiles currently in use. Some of the work on which this
report is based was initiated as far back as 1969, while some other
work was initiated later and was completed only in the last few months.
While many individuals participated in and deserve credit for
various aspects of designing and carrying out the experimental work on
which this report is based, the data that was generated from the
individual projects was translated into the format of this report by
Mr. Joseph Merenda, Assistant to the Director, Mobile Source Pollution
Control Program, EPA; and by Mr. Steven Kuhrtz, who worked as a
Staff Assistant in that office as a part of his graduate training
program at Dartmouth College, Hanover, New Hampshire.
Additional copies of this report are available from the Mobile
Source Pollution Control Program, Office of Air and Water Programs,
Environmental protection Agency, Washington, D.C. 20460.
Eric 0. Stork
Director
Mobile Source Pollution Control Program
-------
11
TABLE OF CONTENTS
Page
PREFACE i
CHAPTER 1 - OVERVIEW 1-1
1.1 INTRODUCTION 1-1
1.2 SUMMARY OF MAJOR FINDINGS 1-2
1.2.1 INSPECTION/MAINTENANCE 1-3
1.2.2 RETROFIT 1-6
1.2.3 GASEOUS FUEL CONVERSION 1-8
CHAPTER 2 - INSPECTION AND MAINTENANCE OF IN-USE VEHICLES 2-1
2.1 INTRODUCTION 2-1
2.1.1 TYPES OF INSPECTION/MAINTENANCE APPROACHES 2-1
2.1.2 EVALUATION OF INSPECTION/MAINTENANCE APPROACHES 2-4
2.1.3 ORANIZATION OF CHAPTER 2 2-6
2.2 DESCRIPTION OF INSPECTION PROCEDURES 2-7
2.2.1 EMISSION INSPECTION PROCEDURES 2-7
2.2.2 ENGINE PARAMETER INSPECTION PROCEDURES 2-20
2.3 FEASIBILITY EVALUATION OF INSPECTION/MAINTENANCE PROGRAMS 2-23
2.3.1 EXHAUST EMISSION INSPECTION 2-26
2.3.2 ENGINE PARAMETER INSPECTION 2-54
2.3.3 MANDATORY MAINTENANCE 2-60
2.4 COST-EFFECTIVENESS OF THE INSPECTION/MAINTENANCE STRATEGY 2-65
2.4.1 COST-EFFECTIVENESS CONSIDERATIONS 2-65
2.4.2 ANALYTICAL APPROACH 2-72
2.4.3 TIME AVERAGED EFFECTIVENESS 2-76
-------
m
2.4.4 ECONOMIC ESTIMATES 2-81
2.4.5 COST-EFFECTIVENESS SUMMARY 2-84
2.5 CORRELATION ANALYSIS 2-88
2.5.1 DESCRIPTION OF TEST PROCEDURES EVALUATED 2-89
2.5.2 DESCRIPTION OF CORRELATION ANALYSIS 2-91
2.5.3 RESULTS OF THE CORRELATION ANALYSIS 2-94
2.5.4 CONSIDERATIONS OF THE NEED FOR CORRELATION 2-99
2.6 CONCLUSIONS 2-116
CHAPTER 3 - RETROFIT OF EMISSION CONTROL TO IN-USE VEHICLES 3-1
3.1 INTRODUCTION 3-1
3.2 SURVEY OF RETROFIT APPROACHES 3-6
3.2.1 DESCRIPTION OF PROGRAM 3-6
3.2.2 RETROFIT TYPES IDENTIFIED 3-7
3.2.2.1 EXHAUST EMISSION CONTROL SYSTEMS 3-7
3.2.2.2 CRANKCASE BLOWBY EMISSION CONTROL SYSTEMS 3-11
3.2.2.3 EVAPORATIVE EMISSION CONTROL SYSTEMS 3-12
3.2.2.4 EMISSION CONTROL COMBINATIONS 3-13
3.2.3 RETROFIT EVALUATION METHODOLOGY 3-13
3.2.4 RESULTS OF PRELIMINARY SCREENING 3-16
3.2.4.1 EXHAUST EMISSION CONTROL SYSTEMS 3-16
3.2.4.2 CRANKCASE EMISSION CONTROL SYSTEMS 3-18
3.2.4.3 EVAPORATIVE EMISSION CONTROL SYSTEMS 3-18
3.2.5 TEST PROGRAM 3-18
3.2.6 RESULTS FOR RETROFIT SYSTEMS TESTED IN THE PROGRAM 3-21
3.2.6.1 GENERAL RESULTS 3-21
3.2.6.2 AIR BLEED TO INTAKE MANIFOLD 3-23
3.2.6.3 CATALYTIC CONVERTER WITH DISTRIBUTOR VACUUM
ADVANCE DISCONNECT 3-26
-------
IV
3.2.6.4 IGNITION TIMING MODIFICATION WITH LEAN IDLE
ADJUSTMENT 3-30
3.2.6.5 SPEED-CONTROLLED EXHAUST GAS RECIRCULATION
WITH DISTRIBUTOR VACUUM ADVANCE DISCONNECT 3-33
3.3 FLEET TEST OF GENERAL MOTORS RETROFIT SYSTEM 3-37
3.3.1 BACKGROUND 3-37
3.3.2 DESCRIPTION OF TEST PROGRAM 3-38
3.3.3 RESULTS OF EVALUATION 3-40
3.4 COST AND EFFECTIVENESS OF RETROFIT STRATEGIES 3-45
3.4.1 INTRODUCTION 3-45
3.4.2 EVALUATION OF OVERALL EFFECTIVENESS AND COST OF
RETROFIT STRATEGIES 3-46
3.4.2 COMPARISON OF EFFECTIVENESS AND COST OF SELECTED
RETROFIT STRATEGIES 3-47
3.5 CONCLUSIONS 3-55
CHAPTER 4 - CONVERSION OF IN-USE VEHICLES FOR GASEOUS FUEL OPERATION 4-1
4.1 INTRODUCTION 4-1
4.2 SUMMARY OF GASEOUS FUEL TECHNOLOGY 4-2
4.2.1 DESCRIPTION OF GASEOUS FUEL SYSTEMS 4-2
4.2.2 EMISSION REDUCTIONS ATTAINABLE THROUGH GASEOUS FUEL
CONVERSION 4-3
4.2.3 COST OF GASEOUS FUEL CONVERSION 4-6
4.2.4 OTHER ASPECTS OF GASEOUS FUEL CONVERSION 4-7
4.3 USEFULNESS OF GASEOUS FUEL CONVERSION AS AN IN-USE VEHICLE
EMISSION CONTROL STRATEGY 4-7
4.4 CONCLUSIONS 4-9
-------
1-1
Chapter 1 OVERVIEW
1.1 INTRODUCTION
BACKGROUND
The Clean Air Act^ provides two basic strategies for controlling air
pollutant emissions from motor vehicles. The United States Environmental
Protection Agency is empowered to promulgate and enforce emission standards
applicable to new motor vehicles; this has been done on an increasingly
stringent basis since the 1968 model year. The States are responsible for
the establishment and enforcement of emission control strategies which
apply to vehicles in use, to the extent that limitations on vehicular
emissions beyond those resulting from the Federal new vehicle standards
are necessary to allow each State to achieve and maintain the
National Ambient Air Quality Standards.
In preparing their implementation plans for achievement of the air
quality standards, a substantial number of States have determined that the
application of control strategies to in-use vehicles will be necessary if
the objectives to the Clean Air Act are to be fully achieved. A number of
alternatives are potentially available to the States to control emissions
from in-use vehicles. These include programs of periodic inspection and
maintenance of vehicles to minimize excessive emissions that result from
inadequate or improper vehicle maintenance; the retrofitting
of emission control systems to vehicles not originally so equipped, or the
installation of more effective emission control systems on already-controlled
vehicles; the conversion of motor vehicles to permit their operation using
gaseous fuels; restrictions on the use and parking of vehicles and modification
of traffic flow patterns; improvements and expansion of public transportation
-------
1-2
systems together with incentives or restrictions to ensure the more extensive
use of those systems in place of private vehicles; modification of social
patterns which influence transportation patterns, such as work schedules;
and restrictions on land use to influence transportation needs and patterns.
The first three approaches have in common their dependency on the application
of emission control limitations to individual vehicles without necessarily
altering the mode or frequency of use of those vehicles; these may be referred
to as "hardware" approaches. In contrast, the other approaches seek to
modify vehicle useage patterns without altering the emission characteristics
of the individual vehicles and can be seen to be complimentary to the "hardware"
approaches.
SCOPE
This document deals with the "hardware" approaches to in-use vehicle
emission control described above; namely, inspection/maintenance programs,
retrofit programs, and conversion of vehicles to permit the use of gaseous
fuels. The "non-hardware" approaches are discussed in other EPA reports. ^»^
This document presents the major results of recent studies and evaluations
made by the Environmental Protection Agency of the feasibility, emission
reduction effectiveness, and costs of the various "hardware" approaches to
in-use vehicle emission control.
In general, the emphasis of this document is on providing basic emission
reduction and cost data which may be useful to the States in evaluating the
alternative approaches to in-use vehicle emission control, as those approaches
may be applied to their particular air quality requirements. No attempt has
been made to identify a "best"in-use vehicle emission control approach or an optimum
-------
1-3
combination of approaches. Such evaluations must be performed on a
region-by-region basis, taking account of the time period over which
emission reductions must be achieved and the magnitudes of the needed
reductions, and using vehicle population data representative of the area
under consideration.
ORGANIZATION
This document is organized into four chapters. Chapters 2,3, and 4
deal, respectively, with inspection/maintenance programs, retrofit approaches,
and gaseous fuel conversion. Each of those chapters provides a detailed
discussion of the data and conclusions relating to that aspect of in-use
vehicle emission control. This chapter presents an overview of the subject
by presenting the major findings with respect to each approach.
1.2 SUMMARY OF MAJOR FINDINGS
1.2.1 INSPECTION/MAINTENANCE
Inspection/maintenance programs aim at reducing emissions from in-use
vehicles through ensuring that the emission levels of those vehicles are
not permitted to deteriorate5through inadequate or improper maintenance,
substantially beyond the levels of which the vehicles were capable when new.
Inspection/maintenance programs can accomplish emission reductions only to
the extent that voluntary maintenance is inadequate in maintaining the
vehicles in good condition.
Studies conducted using representatives samples of privately-owned in-use
automobiles have demonstrated that significant reductions in the aggregate
emissions of those groups of vehicles can be obtained through additional
maintenance. It has been found that approximately 50% of the vehicles in a
typical sample had malfunctions or maladjustments which, when corrected,
resulted in a decrease in emissions. When maintenance was performed on those
-------
1-4
vehicles, the average point-in-fime reductions in emissions for the entire
sample (including the vehicles not maintained) ranged up to 25% for exhaust
hydrocarbons (HC) and up to 19% for carbon monoxide (CO). No statistically
significant changes in the average emissions of oxides of nitrogen (NOx) were
found. These studies lead to the conclusion that current voluntary maintenance
practices are inadequate to keep the entire in-use vehicle population at the
minimum levels of emissions of which it is capable; and, therefore, that
implementation of a required inspection/maintenance strategy could result in
significant reductions in HC and CO emissions.
The actual emission reductions that can be achieved over time through
an inspection/maintenance program will depend upon the rate and extent of
emission control deterioration between the required inspection/maintenance
events. A study of this factor is being conducted,but sufficient data are
not yet available to define the influence of deterioration on the average
reductions achievable by an inspection/maintenance program over time. However,
based on an analysis of the frequency and distribution of current voluntary
maintenance of emission-related components, a preliminary estimate of the
average effectiveness over time has been made. It has been estimated that
inspection/maintenance on an annual basis may be expected to achieve emission
reductions of U£. tp_ 12% in light duty vehicle exhaust HC emissions and up to
10% in light duty vehicle CO emissions. More frequent inspection/maintenance
could achieve larger reductions.
The emission reduction effectiveness of several alternative approaches to
inspection/maintenance has been evaluated. These include: emission testing at
idle only, emission testing using a loaded test cycle, engine parameter diagnosis,
and mandatory maintenance of specific emission-related components or adjustments.
While the mean reductions that were observed for the emission testing approaches
-------
1-5
were the largest, the empirical data currently do not permit differences in
the effectiveness of the various types of approaches to be distinguished at
a 90% confidence level.
This is not to say, however, that all of these approaches show equal
promise for effective application in inspection/maintenance programs for the
general population of in-use light duty vehicles. Engine parameter diagnosis
and mandatory maintenance approaches must be designed to correspond to the
specific emission control systems employed on each type of vehicle to be inspected,
Thus, a specific form of one of these approaches may not be broadly applicable
to the entire population of in-use light duty vehicles; particularly, to those
produced in future model years.
Emission inspection approaches, on the other hand, can be considered more
generally applicable, to the extent that the emission measurements made in the
inspection test accurately reflect the vehicles' emissions during typical
urban driving. The relatively poor correlation for current vehicles between
idle emission measurements and emission rates resulting from typical urban
operation indicates that vehicles may be adjusted so as to satisfy idle emission
inspection standards without achieving reductions in true mass emissions. Loaded
emission tests provide a significantly better measure of the emission levels
which the vehicle would produce in typical driving, although all short emission
tests are hampered by their inability to assess emissions from a cold start.
In general, it appears that, among the inspection/maintenance approaches which
are generally applicable to the current in-use light duty vehicle population,
emission testing using a loaded test cycle has the greatest likelihood of
achieving maximum emission reductions in practice. This conclusion is not,
however, intended to rule out the possibility that other inspection/maintenance
approaches of comparable effectiveness may be designed for specific segments
of the in-use vehicle population.
-------
1-6
Estimates of the cost of vehicle inspection and of the additional
maintenance required through an inspection/maintenance program have been
made. It has been estimated that emission testing using a loaded test cycle
in State-operated inspection lanes would cost approximately $2 per vehicle.
Extensive engine parameter diagnosis has been estimated to cost approximately
$8 per vehicle. It has been estimated that the cost of the additional
maintenance required by an annual inspection/maintenance program in which
30% of the vehicles failed inspection would cost an average of about $2 per
vehicle in the population. Of course, the cost per vehicle failing the
inspection would be higher and would vary significantly among failing
vehicles, depending upon the nature of the malfunction.
The above conclusions regarding inspection/maintenance approaches apply
specifically to the types of vehicles that are currently a part of the in-use
vehicle population. It is not possible at this time to define the effectiveness
which inspection/maintenance programs may achieve if applied to future model
year vehicles using substantially different types of emission control systems.
1.2.2 RETROFIT
Retrofit approaches go beyond the attempt made by inspection/maintenance
approaches to keep in-use vehicles at minimum levels consistent with their
original design. In a retrofit approach, the goal is to reduce an in-use
vehicle's emissions below 1ts "well-maintained" levels, through adding
new emission control devices or through modifying the original design to
achieve lower emissions.
Evaluation of available retrofit devices has indicated that retrofit
systems exist which can achieve substantial reductions in emissions of HC,
CO, and NOx from light vehicles that were not originally subject to Federal
-------
1-7
emission standards (pre-controlled vehicles). Maximum concurrent reductions
of approximately 70% in exhaust HC, 65% in CO, and 50% in NOx, relative to
the vehicle's emissions when in a properly maintained condition, have been
observed. Although concurrent reductions of these magnitudes can be obtained
only with relatively expensive devices, similar reductions in CO and NOx are
achievable individually using much less costly retrofit approaches. Devices
are also available which achieve lesser reductions in all three pollutants at
lower cost.
Application of retrofit approaches to pre-controlled light duty vehicles
can achieve emission reductions of up to the magnitudes stated above without
major compromise in vehicle performance that would render the vehicles
unsafe. Effects of retrofit devices on fuel economy appear to vary markedly
among different approaches but have not been extensively evaluated. The
installed cost of various retrofit approaches ranges from about $20 to as
much as $175.
Limited durability testing of light duty vehicles equipped with various
retrofit approaches has led to the conclusion that actual achievement
and maintenance of emission reductions similar to those stated above
can be ensured only if retrofitted vehicles are maintained in good
condition; emission-related components and adjustments that are not themselves
a part of the retrofit system must be maintained as well as the retrofit.
Thus, periodic inspection and maintenance of retrofitted vehicles appears
to be a necessary part of any retrofit strategy.
Retrofit approaches are aiso potentially applicable to light duty vehicles
already equipped with emission control systems if different or more effective
systems are added. Empirical data quantifying the emission reductions
achievable through retrofitting such vehicles are very limited; however, based
upon the effectiveness of retrofit approaches when applied to pre-controlled
light duty vehicles, it appears reasonable that emission reductions of up to
-------
1-8
50% in exhaust HC and CO per vehicle could be achieved through retrofitting
1968 through 1974 models. It is expected that such reductions could be
achieved only through the use of relatively costly ($80 to $160 per vehicle,
on average) devices requiring the use of unleaded gasoline. Reductions of up
to 40% in NOx per vehicle could probably be achieved through retrofitting
1968 through 1972 models using devices in the $30 to $50 price range. Applica-
bility of such approaches to various model years in California would differ
because of different new vehicle emission requirements there.
The emission reductions cited above as feasible through retrofitting
light duty vehicles refer to the reduction per vehicle. Since retrofit
approaches apply, in general, only to a portion of the in-use vehicle
population, the overall emission reductions achieved by retrofit programs
would be substantially less than those figures. Calculation of the overall
effectiveness of retrofit programs must be performed on a region-bynregion
basis since significant differences tn vehicle population statistics exist.
However, based upon nationwide average vehicle population data, it can beconcluded
that exhaust emission retrofit programs for pre-controlled and controlled
light duty vehicles could be expected to achieve significant impacts on total
light duty vehicle emissions throughout the 1975 to 1980 time period.
1.2.3 GASEOUS FUEL CONVERSION
Conversion of motor vehicles to permit their operation using gaseous
fuels can be considered a special case of retrofit, but implementation of
such an approach to in-use vehicle emission control involves a number of
considerations beyond those required in evaluating retrofit approaches which
simply change the engine design. Those additional considerations include
the availability of adequate supplies of the gaseous fuel, the availability
of5or feasibility of constructing,an adequate number of fueling facilities
-------
1-9
for the converted vehicles, and the possible impact on emissions from
stationary sources of diverting gaseous fuel supplies to vehicular use.
Based upon such considerations, it has generally been concluded that
conversion of large numbers of privately-owned vehicles for gaseous fuel operation
is not feasible, and that gaseous fuel conversion strategies are feasible only
for fleet-operated vehicles.
In spite of the large numbers of vehicles which have been converted to use
gaseous fuels, data on the emission reductions attainable through such
conversions are relatively limited. Available data indicate that
substantial reductions in emissions of HC, CO, and NOx can be achieved
through gaseous fuel conversion, although in a number of cases gaseous fuel
conversion had actually increased emission levels. Data indicate that when
1970 model year and earlier light duty vehicles are converted for operation
on gaseous fuel, and ignition timing and air/fuel ratio are optimized for low
emissions, reductions of up to 80% in HC and CO and up to 60% in NOx are
achievable per vehicle.
The initial cost of gaseous fuel conversion is substantial. Costs for
converting a single vehicle to gaseous fuel operation are estimated in the
range of $450 to $800 or more, depending on the type of fuel and the vehicle
range between fueling stops desired. This cost is exclusive of any cost
necessary to provide fueling facilities. On the other hand, decreased operating
and maintenance costs when using gaseous fuels have motivated a large number
of the conversions made to date, although the extent of such savings may vary
widely.
As with other retrofit strategies, the overall effectiveness of a gaseous
fuel conversion strategy must be evaluated on a region-by-region basis.
-------
1-10
Beyond the differences in vehicle population data among regions and the
different magnitudes of emission reductions required, the availability of fuels
and impact of fuel switching is particularly regional in nature. Therefore,
no general conclusion can be drawn regarding the feasibility and desirability
of gaseous fuel conversion as an in-use vehicle emission control strategy.
However, evaluation of such strategies in a few specific instances has indicated
that unless fleet vehicles contribute a large portion of the total vehicular
emissions in a region, and unless the fleet to be converted is relatively
stable in the sense of not being rapidly replaced by new vehicles, emission
reductions through gaseous fuel conversion of fleet vehicles cannot be
expected to be large or long-lived.
REFERENCES - CHAPTER 1
1. Clean Air Act, 42 U.S.C. 1857 et seq., as amended by P.L. 91-604
December 31, 1970.
2. Evaluating Transportation Controls to Reduce Motor Vehicle Emissions
in Major Metropolitan Areas: An Interim Report, Institute of Public
Administration, Washington, 1T.C., March 16, 1972.
3. Prediction of the Effects of Transportation Controls on Air Quality
in Major Metropolitan Areas (Six Cities Transportation Study),
TRW Systems Group, McLean, Va., July, 1972.
-------
2-1
Chapter 2 INSPECTION AND MAINTENANCE OF IN-USE VEHICLES
2.1 INTRODUCTION
The degradation of automotive emission control resulting from improper
or insufficiently frequent maintenance suggests the possibility of achieving
significant reductions of current levels of motor vehicle emissions through
the periodic inspection and enforced maintenance of in-use vehicles. This
chapter presents the currently available information concerning the
feasibility of alternative test procedures for the periodic inspection
of extensive populations, the potential effectiveness in reducing automotive
emission levels through emission related maintenance procedures, and the
potential cost burden attributable to the implementation of alternative
approaches for the inspection/maintenance strategy.
2.1.1 TYPES OF INSPECTION/MAINTENANCE APPKuaChES
All inspection/maintenance approachs include, conceptually, two phases:
an inspection phase, used to screen the in-use vehicle population to
determine which of those vehicles should be required to receive maintenance;
and a maintenance phase, in which appropriate corrective maintenance is
performed on the selected vehicles. Alternative inspection/maintenance
approaches can be categorized according to the nature of their inspection
phases as follows:
Emission Inspection Approaches - Each vehicle included in the program
is subjected to an emission test and the results compared with a set of
in-use vehicle emission standards. Vehicles with emissions in excess of
the standards are considered to have failed, and are required to have
maintenance performed. An emissions retest may be required after the
-------
2-2
maintenance to ensure that the faired vehicle has been brought into
compliance with the emission standards.
Engine Parameter Inspection Approaches - Each vehicle included in
the program is subjected to a sequence of diagnostic tests which seek to
evaluate the mechanical condition of various emission-related vehicle
systems and determine if malfunctions or maladjustments are present.
Vehicles showing measurements outside of accepted tolerance ranges are
considered to have failed,and are required to have corrective maintenance
performed. This approach bypasses the question of each vehicle's emission
levels, although in some cases emission measurements may be made to evaluate
the state of certain vehicle systems (e.g., measurement of idle CO
concentration to evaluate proper idle air/fuel ratio adjustment).
Mandatory Maintenance Approaches - Each vehicle, independent of
its emission levels or mechanical condition, is required to have specific
maintenance operations performed at required intervals. Thus, the inspection
phase is simply eliminated; and the appropriate maintenance is explicitly
specified for each type of vehicle and identical for all vehicles of that
type, rather than being whatever maintenance is necessary to achieve
compliance with an emission standard or to ensure that specific vehicle
systems pass diagnostic checks.
The requirements for the identification of the optimal engine components
to be tested in a functional, or diagnostic, inspection approach also exists
for the identification of the components to be periodically replaced or
adjusted in a mandatory maintenance program. Any mandatory adjustments
must rely upon some measurement of functional performance to ensure the
-------
2-3
proper setting. If the adjustment is not made to the specifications
of the maintenance policy, the regulations have not been properly executed.
For these reasons, there are only subtle differences between the mandatory
maintenance approach and an engine parameter inspection approach. These
exist primarily within the enforcement implications rather than the
technologies involved.
Three alternative configurations characterize the operational
format of the inspection program. These alternatives are described
below. In each case, it is anticipated that the private service garages
and dealerships comprising the automotive repair industry will provide the
requisite maintenance. Although the selection of a specific alternative
configuration should not conceptually have an impact on the effectiveness
of the strategy, each configuration may have substantial effects on the
feasibility of providing effective enforcement, the total cost burden of
the initial capital investment, or the inconvenience imposed upon the
vehicle owner.
Operational Configurations for the Inspection Program
* Publicly Operated Lane System - The testing functions are provided
directly by the appropriate state or
municipal agency in a system of
publicly owned and operated facilities
designed exclusively for the testing
of motor vehicles.
-------
2-4
* Privately Operated Lane System - The testing is performed by a
private organization under contract
to the appropriate government
agency in a system of inspection
*
lanes designed exclusively for the
testing of motor vehicles.
* Licensed Garage System - The testing is performed by existing
private service or repair agencies
within the maintenance industry.
Each inspection facility is certified,
licensed, and controlled by the
appropriate government agency.
2.1.2 EVALUATION OF INSPECTION/MAINTENANCE APPROACHES
The optimal design of the inspection/maintenance strategy requires the
assessment of numerous interacting factors which have a substantial impact
upon the effectiveness and the incremental economic burden imposed by
the implementation of such programs. Certain of these factors may be evaluated
to provide an initial assessment of the feasibility and the relative performance
characteristics of the various inspection/maintenance approaches, thereby
screening the multitude of potential alternatives. Additional parameters must
then be evaluated in order to assess the actual cost-effectiveness of each
approach as a function of time.
The following factors are the major determinants of the potential
feasibility, effectiveness, and cost-effectiveness of the inspection/maintenance
strategy.
-------
2-5
* The extent to which the lack of voluntary maintenance by the
general public causes in-use vehicles to differ from the original
manufacturer's specifications.
* The magnitude of the emission increases which result from such
deviations from manufacturer's specifications.
* The effectiveness with which the inspection testing procedure can
identify from among the entire vehicle population those vehicles
with excessive emissions resulting from engine malfunctions and
maladjustments.
* The information value of the inspection test results in identifying
the specific maintenance actions required to restore the vehicle to
a condition which is in compliance with the inspection standards.
* The effectiveness of the repair industry in applying proper
maintenance to reduce emissions from those vehicles identified as
having malfunctions or maladjustments.
* The rate and extent of the deterioration of emission control
following maintenance.
* The correlation between the inspection test procedure and the
Federal Certification Test Procedure as it relates to the impact
of vehicular emissions upon ambient air quality.
* The initial capital investment and the annual operating costs of
the inspection facilities and program administration.
* The maintenance costs and any indirect costs or savings such as
changes in fuel economy or the substitution of required maintenance
for existing voluntary maintenance schedules.
-------
2-6
2.1.3 ORGANIZATION OF CHAPTER 2
The remainder of Chapter 2 is devoted to presenting the results of
evaluations of the feasibility, effectiveness, and cost-effectiveness
factors identified above. Section 2.2 provides an overview of the
characteristic test procedures considered feasible for use in the inspection
process. Section 2.3 presents an assessment of the feasibility and relative
performance characteristics of three representative approaches for the
inspection/maintenance strategy. The empirical results of three independent
sample fleets are evaluated.
One factor which remains to be quanitified in a definitive manner is
the rate and extent of emission control deterioration following maintenance.
Studies are presently being conducted to provide an assessment of this
parameter such that the true effectiveness of the inspection/maintenance
strategy can be defined. However, until such information is developed,
available data must be used to provide an estimate of the expected average
effectiveness over time. Accordingly, an analysis is presented in Section 2.4
which sets forth a methodology for estimating the impact of the inspection/
maintenance strategy utilizing existing empirical data.
Section 2.5 identifies the degree of correlation with the Federal
Certification Test Procedure demonstrated by numerous alternative emission
inspection procedures, and discusses the need for such correlation to ensure
the continued effectiveness of inspection/maintenance programs.
Section 3.6 summarizes the results and presents the major conclusions
concerning the existing state of technology and the potential impact of
the inspection/maintenance strategy.
-------
2-7
2.2 DESCRIPTION OF INSPECTION PROCEDURES
The inspection phase of an inspection/maintenance program is a
screening procedure applied to the entire population of in-use vehicles
subject to inspection/maintenance. Its function is to maximize the
cost-effectiveness of the inspection/maintenance program by identifying
for required maintenance only those vehicles likely to exhibit significant
emission reductions if given additional maintenance. The inspection pro-
cedure may also serve to provide a check on the quality of the maintenance
performed on the failed vehicles if they are required to be retested after
maintenance.
The investigation of alternative procedures that satisfy the
performance requirements of the inspection process leads to the
identification of two general categories: direct emission measurement
tests and engine parameter, or diagnostic, tests. A detailed description
of the principal technical factors which must be considered when selecting
a particular inspection procedure is presented in the following subsections,
2.2.1 EMISSION INSPECTION PROCEDURES
The feasibility and effectiveness of the emission inspection
approach depends upon the availability of emission test procedures which
can reliably identify vehicles whose emissions can be reduced through
maintenance. First, let us briefly review the major characteristics
of emission test procedures.
Typical urban driving patterns are composed of four basic types
of operating modes: idle, acceleration, deceleration, and constant speed
or cruise modes. The characteristic emission rates of current in-use
-------
2-8
vehicle .types vary substantially among the various operating modes.
Accordingly, the total emissions generated while a vehicle is operated
over a driving cycle are sensitive to the extent of vehicle operation in
the various modes during the driving cycle. In general, if the emission
measurement is to provide a meaningful indication of the vehicle's
emissions during typical urban driving, the driving cycle must accurately
reflect urban driving patterns typical of the current vehicle population.
The measurement of vehicular emissions is also known to be affected
by tne methodology utilized for sampling and analysis of the emissions
produced during vehicle testing, and by the pre-test vehicle conditioning
procedures employed. Thus, the definition of an exhaust emission test
must include a description of the vehicle pre-conditioning procedures, the
driving cycle which prescribes the vehicle operation during the collection
of the exhaust gas sample, the collection or sampling procedures, and the
analytical techniques used to measure the amount of pollutants in the
exhaust sample. Alternative methods which have been employed for each
of these emission test procedure components are summarized in Table 2-1.
The Federal Certification Test Procedure ^ (FTP) is the standard
method used to quantify light duty vehicle emissions. Although the FTP would
provide the most reliable measure of excessive emissions from in-use light
duty vehicles, it is far too costly and time consuming to be feasible
for the periodic testing of large vehicle populations. Aside from the
expense of the necessary equipment, the test pre-conditioning requires
that the vehicle remain inoperative for at least twelve hours prior to
initiating the emission test, and the driving cycle, derived from data
obtained through operating instrumented vehicles over a typical urban
route, is 23 minutes in length. Therefore, it is necessary to consider
-------
2-9
C\J
CO
oo
UJ
ID
f~\
UJ
CJ
o
rv^
CL.
1—
OO
UJ
1—
Z.
o
»— i
oo
oo
i — i
UJ
U-
o
oo
I—
z.
UJ
•z.
o
a.
s:
o
CJ
1
•z.
o
p
2
Ll-J
Q_
o
— 1
«=t
CJ
1—
>-
«=C
^^
<
CD
^^
t— 1
_l
D-
21
=£
OO
to
• 1 —
(ft
>^
r—
O3
0
O
— 1
O
o
1—
LU
^"
C
o;
i — i
z
•
i —
4^
>-
CD
1
O
Q
O
~T"|
i—
UJ
s:
to
rs
03
c~
X
UJ
r-^
UJ
_l
CJ
UJ CJ
CJ
1 — 1
-]~
CD
•z.
UJ|—
"O
c: a)
0) S-
E 3
OJ tO
i- fO
3 a>
CD CJ
s: :E
O)
JO
o
s-
Q.
E
03
ai
s-
.4_)
oo
• •
to
•I—
(/)
>^
C
Q
U_
•
CVJ
•!->
C
03
-(->
-(->
d
OJ
E
a>
s_
3
03
OJ
S
OJ
^
^
f—
O
^>
+
o
•z
(/) X
•r- O
r^M-
03 O
10 c
C < ^_>
O c
-jQ ^> QJ
^ ~~i E
S i £
O — 3
*- Qi tO
"P i— ' 03 ^-'
-^ § S ^
•
CO
en
C"
• 1—
r—
Q.
^
03
OO
a;
-a
o to
s: QJ
T3
Q) O
-l-> •<->
oo c
QJ
!>^ •! —
T3 to
03 E
OJ re
00 1—
.
ra -Q
to
•i- X
CO O
03 i—
C *"O
o
C
QJ t|_
(J O
t/1
O) -M
C C
•r- QJ
E E
3 QJ
^§0
E co ^
QJ 03
-C QJ to
o s: 03
•
^
CD
CJ i—i
1-1 Q
UJ O
>• CJ
CtL
o_
a>
•*-> c
S- -O T-
03 to a) ai
4-> T- -l-> C
oo nso)
+-> •(-
T3 to -M T3
r— QJ -i- •—
O 4-> C O
C_) 1- o
S- OJ
O +J
tO •!—
••- S- 01
a. c
OJ •(-
s- -o +->
-t-> 3 QJ 03
S_ 40 M -i-
03 QJ 03 -i- -4-J
+•> E i. i— ••-
OO -i- OJ -i- C
cn Q-JO •!-
+-> C E 03
O QJ QJ +-> O
C\J
-------
2-10
whether alternative test procedures are available which would be
feasible for in-use vehicle emission inspection testing.
VEHICLE PRE-CONDITIONING
There are two basic alternatives for vehicle pre-conditioning: cold
start and hot start testing. In cold start testing, the vehicle is allowed
to stand unoperated for an extended period (typically twelve hours or
more) at a specified ambient temperature to allow all of the vehicle
systems to attain an ambient temperature. Then the vehicle testing is
initiated with the cold vehicle and emissions are normally measured from
the beginning of the vehicle start up procedure. Cold start pre-conditioning
simulates the emissions from a vehicle being started after a period of
extended shutdown, as, for example, when first started after overnight
parking. Cold start testing allows evaluation of systems (such as choke)
which do not function when the vehicle has attained normal operating
temperatures; and, therefore, is considered the best type of test to employ
for extensive evaluation of vehicle emissions. However, the extended time
required for pre-conditioning makes a cold start test unfeasible for in-use
vehicle emission inspection.
The hot start test is conducted with the vehicle starting from a
temperature at or near normal operating temperature. One version, the
hot-restart test employed in the 1975 Federal Test Procedure,1 requires
restarting the vehicle after a ten-minute soak period at ambient temperature
following vehicle operation. For this test, emissions are measured from
the beginning of the vehicle restart procedure. Another type of hot start
test which is often considered for in-use vehicle emission inspection
testing does not measure emissions from the beginning of the vehicle
start up procedure, but requires operation of the vehicle for a sufficient
-------
2-11
period to stabilize operating temperatures and then measures emissions
from the stabilized, operating vehicle.
VEHICLE OPERATING CYCLES
Vehicle operating cycles (driving cycles) are characterized
according to the types of driving modes which are simulated during the
test. Of interest are the idle mode test cycle, cruise mode test
cycles, and transient mode test cycles; the latter two may be classified
as dynamic mode cycles.
The measurement of emissions only during the idle mode requires
the least time and equipment, and is the most convenient operating mode
for an emission measurement. The pollutant levels within the exhaust gas
are measured during normal idle operation of the vehicle. There is no
external power loading imposed on the engine. The idle procedure must
specify whether the emission measurement is to be made with the transmission
in "drive" or in "neutral" when testing vehicles with an automatic
transmission. An idle mode test typically requires about two minutes to
complete.
The operational requirements for emission measurements during dynamic
mode driving cycles are substantially more sophisticated. A chassis
dynamometer is required to simulate the loading which is imposed on the
engine during actual driving conditions. The alternatives within this
category may be composed only of cruise modes, or they may include the
transient (acceleration and deceleration) operating modes.
The Clayton Key Mode2 procedure is a typical cruise mode test cycle.
The vehicle is positioned on the dynamometer and the driver accelerates
until an indicated speed of 44 to 50 mph is attained; a steady cruise
-------
2-12
is maintained while the emission measurement is taken. The driver then
decelerates to a cruise speed of 22 to 30 mph and the emission measurement
is repeated. The final measurement is recorded while the vehicle is
idling. The dynamometer provides a power absorption which simulates the
loading due to the various sources of fluid and mechanical friction present
during constant speed operation of the vehicle on the road. The power
absorption used for the Key Mode Test is proportional to the third power
of the road speed, providing a 30 horsepower load at 50 mph. Cruise
mode tests typically require about three minutes to complete.
The inclusion of acceleration and deceleration in transient mode
cycles provides a driving pattern which is clearly more representative
of actual vehicle operation. It must, therefore, be expected that the
associated emission measurements provide the best measure of vehicle
emissions on the road. A transient mode test cycle requires additional
loading capabilities to simulate the inertial effect of the vehicle mass.
The requisite dynamometer and related equipment are substantially more
sophisticated and costly than those required for a cruise mode cycle.
A transient mode driving pattern frequently used for emission testing
is the Seven Mode Cycle.3 The speed-time trace is shown in Figure 2-1.
The vehicle is positioned on the dynamometer and driven over the prescribed
test cycle. The dynamometer must be calibrated to apply a road load
effect and an intertial load effect which are proportional to the weight
of the vehicle. The exhaust emissions may be measured for the entire
operating cycle or limited to specific segments of each individual mode.
About five minutes is typically required to complete a transient mode
test using the Seven Mode Cycle.
-------
2-13
o
(Nl
O
O
to
-o
c
o
O
co
i
CM
Di
CJ3
CXL
a
oo
D_
ct
o
C\J
o
CO
o
C\J
—*o
o
(qdui) Q33dS 310IH3A
-------
2-14
SAMPLING METHODOLOGY
The techniques used to collect the exhaust gas sample for analysis
have a direct impact on the emission measurement results. The least
expensive technique for sampling the exhaust gas utilizes a collection probe
which is inserted directly into the tailpipe. A slight vacuum is applied
to the sample line and the exhaust gas is transferred to the analytical
system for continuous analysis during the sampling period. The resulting
analysis provides a measure of the concentration of pollutants within the
exhaust mixture as a function of time.
An evaluation of the impact of automotive pollution requires a
measure of the mass emission rate as a function of vehicle usage. The
concentration measurement obtained using the probe sampling technique
provides no indication of the mass of pollutants emitted without a
concurrent measure of the exhaust flow rate. Analysis of vehicle exhaust
volumes has shown that the mass flow rate over a given driving cycle is
primarily a function of the vehicle weight and the measured concentration
of the pollutant within the exhaust gas. This method provides an
acceptable measure of the mean emission rate for large samples of
vehicles; however, substantial error often results when calculating the
mass emission rate for an individual vehicle.
A recently developed sampling technique^ is used to directly
determine the mass of pollutants being emitted from an individual test
cycle. The procedure utilizes a Constant Volume Sampling (CVS) system
which eliminates the need for the empirical conversion formula. The CVS
technique provides an integrated measure of the total exhaust flow over
the test cycle, and the sample which is analyzed provides an integrated
measure of the average pollutant concentration during the entire cycle.
-------
2-15
The mass emission rate as a function of vehicle usage (grams per mile)
is then directly calculated.
Either sampling approach can potentially be used with any'of the
specific test cycles discussed previously.
ANALYTICAL METHODOLOGY
Three primary components of automotive exhaust are currently
regulated. These are: unburned hydrocarbons (HC), carbon monoxide (CO),
and oxides of nitrogen (NOx). However, the composite exhaust from the
automobile engine is a complex mixture of many different gaseous (and
particulate) reaction products of the combustion process. Accordingly,
the emission measurement must be based upon analytical methodology highly
specific for each component of concern and essentially insensitive to the
presence of other components within the exhaust mixture.
Each specific test procedure may have unique constraints for the
performance characteristics of the analytical system. In general, however,
the time required to analyze the sample should be minimized. The total
labor time required for an emission measurement is often the most
significant cost element, due to the high skill level necessary for the
personnel involved in the testing procedure. In addition, an excessive
delay time between the collection of the sample from the exhaust stream
and the completion of the analysis may confound the emission measurement
due to the potential reactivity of the many gaseous components.
An obvious attribute desirable for any testing system is operational
simplicity; thus, minimizing the probability of erratic measurement error.
The broad range of analytical technology potentially applicable to
exhaust emission testing is beyond the scope of this report. The
following text provides a brief description of the basic instrumentation
-------
2-16
currently used to fulfill the analytical requirements of motor vehicle
emission test procedures.
•Nondispersive Infrared Analysis - the operational principle of the
nondispersive infrared (NDIR) analyzer relies upon the infrared energy
absorption characteristics exhibited by certain molecular species.
The attenuation of an infrared light source passing through a gaseous
mixture is proportional to the concentration of the absorbing species
within the mixture.
NDIR analysis is appropriate for any molecular species exhibiting
a strong characteristic absorption peak within a narrow frequency band.
The absorption peak must provide adequate resoltuion to avoid any
interference resulting from the presence of other exhaust components
which absorb at a similar frequency in the infrared.
NDIR analysis is typically used for the measurement of carbon
monoxide and nitric oxide (NO). The water content of the exhaust sample
generates interference within the absorption band of CO. This is
generally overcome by passing the sample stream through a drying agent
prior to entering the analyzer.
Exhaust gas contains a large number of unique hydrocarbon species.
Many of the hydrocarbon molecules with complex bonding structures cannot
be detected using NDIR analysis. A number of simple hydrocarbons, known as
paraffins, exhibit strong infrared absorption peaks. Current motor
vehicle exhaust compositions exhibit a reasonably constant proportionality
between the paraffin content and the total hydrocarbon content. The
NDIR instrument is sensitized for n-Hexane and a measure of the paraffin
content is obtained which can then be related to total hydrocarbons
using the proportionality constant.
-------
2-17
Nondispersive Ultraviolet Analysis - in principle, the NDUV analyzer
relies upon the same optical absorption phenomenon discussed above. The
molecular species which can be analyzed using this approach exhibit
characteristic absorption peaks within the ultraviolet frequency range.
The utility of NDUV analysis for exhaust emission testing is limited
to the measurement of nitrogen dioxide (NC^). The production of NC^
occurs as the exhaust gas effluent cools, favoring the reduction of any
residual oxygen by nitric oxide formed during combustion. A composite
NDIR/NDUV system provides a useful measure of the total oxides of nitrogen
(NO and N02) composition of the exhaust.
Analysis by Flame lonization Detector - Flame ionization detectors provide
a direct measure of total hydrocarbons. The sample is passed through
a hydrogen flame ionizer which dissociates the hydrocarbon molecules,
creating an ionized gas of charged carbon atoms. An electro-potential
measurement across the gas is proportional to the concentration of carbon
ions present. The measurement is directly related to total hydrocarbon
content of the initial gas mixture. FID analysis is not affected by
the presence of CO, C02, NOX, or water. However, the procedure is
extremely sensitive to the precise control of the sample flow rate
through the analyzer.
Chemiluminescent Analysis - The reaction of nitric oxide (NO) and ozone
produces chemiluminescent radiation (emission of energy in the red and
infrared frequency range) which is proportional to the initial concentration
of NO in the reaction mixture. Preceding the actual analysis, the
exhaust gas sample is passed through a thermal converter to ensure the rapid
conversion of any N02> which may be present, to nitric oxide. Injection
of ozone into the sample stream initiates the reaction. The analyzer is
-------
2-18
calibrated to relate the intensity of the chemiluminescence to the
initial concentration of nitric oxide.
SUMMARY OF EMISSION TEST PROCEDURES
Although a large number of alternative emission test procedures
are possible through combinations of the components shown in Table 2-1 ,
not all of these combinations would be practical. For example, selection
of a simple driving cycle, such as a single idle mode, limits the
practical choices of sampling and analytical methodology since the limited
information content of an idle mode emission measurement would not justify the
use of highly precise and accurate sampling and analytical techniques.
Table 2-2 outlines three alternative test procedures that are currently
considered applicable for State inspection programs. These specific
alternatives are representative of the range of costs and accuracy which can
be expected for the present state of emission testing technology, and the
existing vehicle population.
The present population, with the exception of post-1970 model year
vehicles sold in California, has not been designed to comply with standards
for NOX emissions. Therefore, analytical requirements for NOx are not
included among the alternatives shown in Table 2-2. Future vehicles (1973
and later model years) may necessitate the inclusion of NOx measurement
capabilities.
-------
2-19
c
o
•I—
•p
re
* 4->
co c
h- OJ
Z O
LU C
SI O
i— i to
ro 10
CTCJ3
LU
o; -4->
s-
O- O
s-
CU
N
>•*->
r— CU
re E
c o
< E
to
O C
o >-.
o: co
I-H re
a .c
z o
O)
0>
i— -a
re re
=3 O
•a _i
re
" S-
O O)
(_) C
re
•p
en
-a a>
o
o -r-
•p >
cu
cn-o
o
r— +->
re ^
c o
•a: -M
c
• • T—
{A S-
E Q.
cu
.. cu
to E
s- o
cu E
N re
>>
re Q
c
«=C to
C£. to
>—i re
to s-
>> cu
co -P
3
cn Q.
c E
—*T- o
to co u
cu co
'i— CU *
o s-
•r- O CU
•— S- -P
•>- 0. S_
-Q CU
re re >
re re o
O Q O
CVJ
C\J
CQ
UJ
o:
cr
UJ
o:
o;
o
UJ
o
o
oo
co
to
cu
to
-
CO
ai
13
to
CU
O)
-o
O
-a cu
re i—
o -o
to
cu
o
to
cu
o
re
cu
-a
o
cu
to
CD
-a
o
-p
c
QJ
•r^
CO
re
s-
cu
-P
to
>»
to
en
c:
•r—
to
in
cu
o
o
a.
re
•p
re
•o
cu
3
CT
CU
CO
CU
S-
3
T3
CU
en o
c: o
•r- S-
.— Q.
Q.
E r-
re re
CO O
a> -P
3 i—
i— re
o c
> re
•p cu
c _c
re -P
•P
to s-
c: o
o M-
o
•o
s- cu
o s-
•
(/>
C"
0
40
re
S-
cu
Q.
0
r^
re
3
C
re
E
s_
o
CO
-t_>
C
cu
E
cu
S-
•r—
3
O"
CU
s-
E
3
•o
3
O
^
•p
3 •
Q. CU
_<~ ^~
en o
3 >,
O C_5
S-
JZ CU
-P "O
o
cu s:
13 -P
Jz cu
0) T-
> to
c
-c: re
cn s-
•r- 1—
-d
CU
S- J=
O -P
c^«
i_
CO O
CU 4-
S-
3 E
•o 'S
CU O
O -C
O to
S-
Q. CU
CO
-o o
cu -c
•p -p
re
E 0
0 -P
3 S-
re re
•p cr
c cu •
cu s- to
E -P
Q- CO C
••- -r- CU
3 E
cr cu cu
cu •— s-
O -r-
i — -1-3
re J^ cr
c: cu cu
O > s-
•r-
•P S- -P
••- cu c
-o Q. cu
•a E
re co 3
cu s-
CU -P -P
•P 3 tO
re c c
-P •!- •!-
•r- E
co S-
CO 00 O
CU 1
CJ CM i—
cu cu
C i— C
re c
to c: o
•P O co
c: ••- s-
CU -P CU
E -i- CL
cu -a
s- -a i—
3 re re
to c
re c o
O) < -r-
E -P
• r-
c • -a
O CO T3
•i- cu re
CO -i-
10 -P CU
••- ••- s_
Er- re
CU -r-
un
CVJ
E E
cu ••-
CO CO
_a cu
to re s-
co a. cu
re re .c
2: o I—
-------
2-20,
2.2.2 ENGINE PARAMETER INSPECTION PRECEDURES
The second major classification of alternatives for the inspection
process is the engine parameter approach. Parameter inspection provides
a functional evaluation of selected engine components and adjustments.
Both the effectiveness and the cost-effectiveness of this approach
are highly sensitive to the selection of the optimal parameters to be
evaluated during each periodic inspection. The selection criteria for these
engine parameters must reflect the extent and frequency of various types
of malfunctions (or rates of maladjustment) and their impact on exhaust
emissions.
Table 2-3 outlines the characteristics of an extensive engine
parameter inspection. Three basic engine subsystems are identified for
evaluation; and within each subsystem, there are identified individual
engine components or adjustments which are important in the control of
exhaust emissions.
TABLE 2-3
ENGINE PARAMETER INSPECTION*
Equipment Inspection Time
Subsystem Engine Parameter Requirements (Man hours)
Idle Adjustments % Idle CO NDIR CO Analyzer 0.15
RPM, Timing Tachometer, Timing
Light
Secondary Ignition Plugs, Wires, Electronic Engine 0.25
Distributor Analyzer
Induction Air Cleaner Air Cleaner Tester 0.25
PCV Valve Pressure Gauge
Air Injection Air Flowmeter
System
* The alternatives shown here were identified in the APRAC/CAPE - 13 study
(see Reference 5).
-------
2-21
The individual inspection parameters shown in this table are only
appropriate for the present vehicle population, and its associated
emission control systems. It cannot be assumed that these specific
procedures are either necessary or sufficient for future model year vehicles
with advanced emission control systems. As new vehicles enter the population,
parameter inspection procedures must be re-evaluated to ensure that the
program remains effective and cost-effective.
The failure limits for each inspection parameter are generally functional
criteria which reflect the manufacturer's specifications. For the most
part, this approach does not rely on the direct measurement of exhaust
emissions. The only exception being the measurement of the idle mode
volumetric concentration of carbon monoxide. Within the context of the
engine paramenter inspection, % idle CO is considered a functional
evaluation of the idle air/fuel ratio.
Although the procedures outlined in Table 2-3 would provide the most
thorough inspection of a vehicle's state of maintenance, it would not necessarily
provide the most cost-effective approach from the standpoint of reducing
vehicle exhaust emissions. Certain of these engine parameters may
malfunction infrequently and cause only a marginal increase of the exhaust
emission levels. The incremental cost required to inspect such a component
may be disproportionate with respect to the resultant impact upon the
emission levels of the total vehicle population.
There are no operational constraints which restrict the specific
combination of parameters selected for an inspection program. Logically,
consideration should first be concentrated on those parameters which are
-------
2-22
most accessible, easily tested, and easily adjusted and which have the
largest impact on emissions. If larger reductions are required, the
additional inspection parameters, requiring more complex procedures, would
be warranted. This implies that the design of a parameter inspection pro-
cedure should consider in an ordinal manner; idle adjustment inspection,
ignition system inspection, and induction system inspection.
Since it depends upon checking the mechanical condition of the vehicle,
rather than a direct evaluation of the vehicle's emission levels, the
engine parameter approach would be less reliable in identifying vehicles
with high emission levels than would an emission inspection approach. On
the other hand, an engine parameter inspection approach can minimize the
diagnosis required by the repair agency in contrast with an emission
inspection whose results can provide only general guidelines for determining
the proper repair actions. In addition, an emission inspection approach
requires, in general, a reinspection of failed vehicles after maintenance
to ensure compliance with the standards. In contrast, the results of an
engine parameter test may specifically define the appropriate maintenance
and, if the maintenance is properly performed, compliance with the parameter
inspection requirement is achieved by definition. Reduced emissions of
the repaired vehicle depend upon the validity of the relationship between
the parameter evaluated and emission levels.
-------
2-23
2.3 FEASIBILITY EVALUATION OF INSPECTION/MAINTENANCE PROGRAMS
This section presents a preliminary evaluation of the feasibility
and the potential effectiveness of the inspection/maintenance strategy.
The results of three independent studies are analyzed to characterize
the attributes of exhaust emission inspection, engine parameter inspection,
and mandatory maintenance. The evaluation of the first approach includes
a study of both an idle mode emission inspection and a dynamic mode emission
inspection.
The information set forth in the following discussion provides an
assessment of the extent "to which excessive emissions are caused Dy an inadequate
level of maintenance for vehicles currently in service, the feasibility
of identifying individual vehicles which achieve substantial emission
reductions following maintenance, the capability of specifying the proper
repair and adjustment procedures to achieve these reductions, the cost
of emission-related maintenance, and the cost of implementing and
operating the inspection program. The quantification of these
characteristic parameters currently provides the only empirical basis
for evaluating the potential impact of the inspection/maintenance strategy.
The salient observations are shown in Table 2-4 for each of the
three inspection/maintenance approaches. The emission inspections and the
engine parameter inspection were evaluated utilizing composite test fleets which
were statistically representative of the current vehicle population with
respect to the distribution by model year, emission control groups,
manufacturer, and engine size. Every vehicle in the sample fleet was
tested using the appropriate inspection procedure, and those failing
-------
2-24
T3
•P 0)
to O
O -r- OJ
„, <-> > •—
Q> s- o
CJ> S_ O) -r-
rO -I- CO J=
t- (O O)
QJ 0. V. >
> QJ Q)
i— OJ
U <1)
•r— r^
-C U_
CD
> *->
l/l
T3 0)
QJ I—
U
•i- <4-
> O
0) ci-S
co —•
to X
c o
o
LO
LO
S
o
o
o
+
o
o
(U
3
-o
ai
o
o
a.
•P
to
O)
ro
S-
O)
"D
O)
E
ro
o
CO -r-
o
CO
CTl
° =5
CQ
l—i
CO
a>
a:
(O
O) O
S- 0)
-l-> Q
CO
LO
CVJ
o
o
E
O
-t-> o 4-1
tO -r- O
3 to QJ
ro to D-
-C 'i— to
X E E
UJ UJ I—I
o
LO
i- a
O) T-
-p -l->
O) O) O
CEO)
•i— ro Q-
CT) S- tO
C ra E
UJ Q- i—i
LO
CO
a>
(J
>> c
S- (O
O E
4J O)
(O •(->
-a c
CVJ
cr>
i—
O)
E
•i—
to
"^2
T3
CD
S_
^
to
ro
QJ
E
to
E
0
to
to
'i
QJ
to
^
ro
.E
X
>1
1—
03
O
•r—
+->
C/)
•r-
_i_ *
+^
ro
4-)
to
to
E
O
•i—
to
to
• 1—
E
QJ
X
O
—^
^_
E
QJ
O1
E
ro
JZ
O
-------
2-25
the inspection standards received diagnosis and maintenance
according to the defined methodology of the test program. The third
alternative (mandatory maintenance) was evaluated using only pre-emission
controlled vehicles. In this case, there is no inspection required;
every vehicle receives identical maintenance wherein specific engine
adjustments are required and specific engine components are replaced.
A description of the study methodology and a more definitive evaluation
of the observations for each test program are presented in the following
text.
-------
2-26
2.3.1 EXHAUST EMISSION INSPECTION
Two representative exhaust emission test procedures were independently
evaluated in a preliminary pilot study. The idle mode test and a loaded
cruise mode test were selected as the inspection procedures for two separate
vehicle test fleets composed of privately-owned automobiles. Each sample
fleet was tested using the appropriate inspection procedures, and those
vehicles which did not comply with the emission standards were repaired
in accordance with the objective of reducing exhaust emission levels.
Hereafter, this test program will be referred to as the Short Cycle
Study. 7
STUDY METHODOLOGY
The purpose of the Short Cycle Study was to determine, as accurately as
possible, the effects which could be expected immediately following the actual
implementation of an enforced emission inspection/maintenance program.
Accordingly, the sample fleet was composed of privately-owned vehicles
randomly selected in such a manner as to ensure that it was statistically
representative of the real vehicle population.
Private service garages and independent automobile dealerships were
solicited to participate in the program to provide the necessary repairs for
those vehicles which failed the inspection standards. The service facilities
were selected to provide a representative cross section of the present maintenance
industry. The test program was duplicated in two geographic areas to evaluate
the existence of any potential differences which may occur among regional
vehicle populations and regional automotive service industries.
The idle mode emission test and a constant velocity cruise mode emission
test were used to evaluate the effectiveness of alternative inspection
procedures. These two particular procedures were selected to assess the
-------
2-27
range of the impact which can be expected as a result of measuring
emissions during loaded operation of the engine, versus measuring emissions
from a free running engine with no external loading imposed. The evaluation
of the two procedures was carried out independently using separate vehicle
sample fleets and separate groups of repair facilities. The vehicle selection
was matched to ensure that the two test fleets were statistically equivalent.
The emission test sequence which was performed for every vehicle in
both inspection fleets was designed to provide three types of information.
The Federal Test Procedure was used to provide a measure of the true mass
emission level and was the defined performance standard for evaluating the
impact of the program, the inspection test measurement provided the information
necessary to identify excessively emitting vehicles which were to receive
maintenance, and numerous additional short test emission measurements were made
to obtain data to evaluate the range of correlation which can be achieved
between the Federal Test Procedure and the various emission tests which are
considered feasible alternatives for the inspection process. The correlation
analysis will be discussed in greater detail in Section 2.5.
The effectiveness measure which defined the performance of the
inspection/maintenance programs is the change in the total fleet mean emission
level of HC, CO, and NOX) as measured by the FTP. Although NOx emissions
were not used as failure criteria, the impact of the program upon this
emission species was evaluated to assess the potential deleterious effects
on NOx typically associated with tune-up procedures designed to reduce
HC and CO.
The idle mode inspection fleet used standard procedures for the
volumetric measurement of HC and CO during free running idle operation.
The vehicle was tested with the transmission in neutral. A vehicle failed
-------
2-28
the inspection if it exceeded the emission standards for either HC or
CO.
2
The cruise mode inspection fleet was tested usina the Key Mode test
procedure as described in Section 2.2.1. The loading imposed on the engine
during the low and high cruise modes varies according to the weight
class of the vehicle. The load factor is controlled by altering the
simulated speed for the vehicle being tested. Table 2-5a shows the speed
and load factors for the different weight classes.
The emission standards were selected to fail approximately 50% of each
test fleet. A vehicle failed the inspection if it exceeded the emission
standard for either HC or CO in any test mode.
The emission failure criteria for the Idle Mode fleet and the Key
Mode fleet are summarized in Table 2-5b.
Following maintenance, the failed vehicles were retested using the
complete test sequence described earlier. If the vehicle again failed to
comply with the inspection standards, it was returned to the service garage
for additional maintenance. Failure following the second service resulted
in the termination of the vehicle and a complete diagnosis to determine
the cause of the failure. If at any point in the test program the vehicle
was shown to require repairs exceeding $100, the vehicle was terminated
without performing the repair. The emission test data remained in the fleet
statistics, and the vehicle was diagnosed to determine the cause of failure.
The flow diagram in Figure 2-2 outlines the operational design of the
test program.
The repair facilities were selected to represent average capabilities
within the service industry in the two regions. Prior to initiating the
-------
2-29
TABLE 2-5a
LOAD AND SPEED FACTORS FOR KEY MODE TEST
Low Cruise Mode High Cruise Mode
Weight Class Speed Load Speed Load
Less than 2800 Ibs 23 mph 5 hp 37 mph 14 hp
2800-3800 Ibs 30 mph 9 hp 45 mph 23 hp
Greater than 3800 Ibs 33 mph 11 hp 49 mph 29 hp
-------
2-30
o
cu
o
-!->
c
o
o
£
O
•r—
tO
tO
oo
o
o
CO
s« LO &s
LO r~- ur>
CM CM
E E £
CL CL CL
CL CL O-
o
*d-
O CO LO
LO CM CM
IT)
Q.
Q.
LO O O LO
1-^ O O CM
CO CO CO CO
c?s un as
O i--. LO
• • •
*d- CM CM
LO
CO
Q. Q. Q.
Q. Q. Q.
O O O
o o o
«3- co co
O 00 LO
«* CM CM
Q.
Q.
CD
LO
CO
CO
CO
Q. CL O.
CL CL Q.
O O O
O O O
CO CO CO
CL
CL
o
LO
CM
Q. Q. Q.
CL Q. CL
O O O
000
co co co
CL
CL
O
LO
CM
.a
LO
CM
OQ
a:
o
in
03
o
E ••-
o -c
•i- CU
CO >
to
•r- -a
E co
LU i—
I ,—
CO o
i. S-
a. •»->
c
o
O CM
LO
co
CM LO LO
r^ ^" CO
o
CO
CL CL CL
CL Q. CL
LO LO LO
CM CM CM
00 LO LO
CL
Q.
O
O
I--.
CL CL CL
CL Q. CL
O O O
O LO LO
CO LO LO
CL
CL
O
O
-l-> CO
to T)
CO O
CO
co to
to -r-
•i- 'S
CO 3 i-
,— i. O
•oo
i— I _C
2 01
O •!-
CO
cu to
tO •!-
CO
r-
-o
CO
s- o
T3 CJ
i— i ^:
S en
O '1-
CO
r—
-a
O CO
CO CO
CLi—
to U_
rO
•r-
E
o
(O
o
CO
CO
CO
CO
-o
c o
fa s:
o>
(J
CO
•o
CO
T3
-------
2-31
CD 14- _J
-------
2-32
test program, every garage was assigned to work with vehicles from one of
the two inspection fleets. Adequate provisions were made to ensure that
every garage in both inspection regions had an infrared hydrocarbon and
carbon monoxide analyzer.
The service mechanics were required to attend a briefing session
during which they were provided with the guidelines for the maintenance
program and the administrative requirements. In addition, they were given
instruction in the causes of excessive emissions and the relationship between
excessive emission levels and specific engine malfunctions. The briefing
sessions were held independently for the garages associated with each of
the two test fleets. The garages were instructed in the interpretation of
the diagnostic content of the respective inspection procedures, and the
limitations on the types of repairs which could be justified with each
inspection test.
The test program was duplicated in two phases; the purpose of the
second phase was primarily to expand the sample size. The study methodology
was identical in both phases. During Phase II, however, a new group of
service garages was used and the failure criteria were lowered to obtain a
greater number of serviced vehicles. The sample size for the test fleets and
various sub-groups are shown in Figure 2-3.
-------
2-33
>-
O
CO
i—i
o;
1— I—
oo oo
Q LU
S -- 1
LU O <-J
—I 4- >-
D- O
I OO
Ll_ LU
O _l
LU ID (_3
CD i— 1—1
o 3:
PRE
CONTROLLED
80
VEHICLES
OO
UJ
•—i 00
z: uj
CSL O _1
O O CJ
u_ ro i—i
o
LU
LU O I
Q LO O
O i— H-.
^: :E
LU
I
C\J
CD
_l O _1
u. u-> o
-------
2-34
FLEET EMISSION REDUCTIONS
The initial emission reductions resulting from maintenance of the
failed vehicles are summarized in Table 2-6. Each emission reduction reflects
the change in the pooled mean emission level of the total tesc fleet,
or vehicle sub-group, as appropriate. For purposes of comparison, the <
composite reductions of the pooled data for the combined inspection fleets
in each State, and the total composite reductions are also presented.
The 90% confidence intervals, based upon the variance of the fleet emission
reductions, are illustrated in Figure 2-4 (a, b, & c) for hydrocarbons,
carbon monoxide, and oxides of nitrogen respectively. These intervals are
an estimate of the range of effectiveness which can be expected to occur for
the real vehicle population at the 90% level of confidence. This uncertainty
in the test results reflects the inherent risk associated with the assumption
that each sample fleet was truly representative of the vehicle population.
There is no significant difference between the initial emission
reductions observed for the Idle Mode and Key Mode fleets tested in Michigan.
In California, the mean, or expected, initial reductions observed for the
Key Mode test fleet were substantially larger than the reductions observed
for the Idle Mode fleet; although the current data do not support such a
hypothesis at the 90% confidence level.
To further evaluate the initial effectiveness of the two emission
inspection approaches, the characteristics of the inspection and maintenance
phases are considered separately below.
-------
2-35
TABLE 2-6
SHORT CYCLE STUDY EMISSION REDUCTIONS
VEHICLES
TEST REGIME SERVICED
CALIFORNIA TEST FLEET 50%
Idle Mode Inspection Regime 44%
Pre-1966 Vehicles 53%
Post-1966 Vehicles 34%
Key Mode Inspection Regime 56%
Pre-1966 Vehicles 61%
Post-1966 Vehicles 52%
MICHIGAN TEST FLEET 50%
Idle Mode Inspection Regime 50%
Pre-1968 Vehicles 50%
Post-1968 Vehicles 50%
Key Mode Inspection Regime 49%
Pre-1968 Vehicles 54%
Post-1968 Vehicles 44%
Composite Total 50%
HC_
29%
23%
24%
14%
34%
38%
26%
23%
21%
25%
16%
22%
27%
14%
25%
MEAN REDUCTION OF
EXHAUST EMISSIONS
cp_
20%
15%
16%
14%
24%
29%
21%
19%
19%
18%
20%
18%
17%
20%
19%
N0x+
6.5%
0
0
- 2.5%
-13.0%
-21.0%
- 8.0%
- 2.5%
- 5.0%
0
- 8.7%
0
3.0%
- 2.4%
- 4.0%
+ Negative reduction indicates an emission increase.
NOx changes were not statistically significant.
-------
2-36
O •
u. o:
oo to
8
CO
CO
CO
O
CO
cc
<£
O
O
cc
O
oi
IDLE-MODE
FLEET
90% CONF.LIMITS*
SHOWN FOR AFTER
SERVICE
EMISSIONS
KEY-MODE
FLEET
IDLE-MODE
FLEET
KEY-MODE
FLEET
CALIFORNIA
MICHIGAN
FIGURE 2-4a
SHORT CYCLE STUDY EMISSIONS
BEFORE AND AFTER SERVICE
-------
2-37
100
o
x
tO
O
to
to
X
o
o
CD
ce
80
60
40
20
90% CONFIDENCE LIMITS
SHOWN FOR AFTER
SERVICE EMISSIONS
IDLE- MODE
FLEET
KEY-MODE
FLEET
IDLE-MODE
FLEET
KEY-MODE
FLEET
CALIFORNIA
MICHIGAN
FIGURE 2-4b
SHORT CYCLE STUDY EMISSIONS
BEFORE AND AFTER SERVICE
-------
2-38
(_> l-l
CO
O
CO
CO
UJ
O
O
or
X
O
0
90% CONFIDENCE LIMITS
SHOWN FOR AFTER
SERVICE EMISSIONS
IDLE-MODE
FLEET
KEY-MODE
FLEET
IDLE-MODE
FLEET
KEY-MODE
FLEET
CALIFORNIA
MICHIGAN
FIGURE 2-4c
SHORT CYCLE STUDY EMISSIONS
BEFORE AND AFTER SERVICE
-------
2-39
FAILED VEHICLE CHARACTERISTICS
The test failures within each fleet were evaluated in order to
characterize the nature of the serviced vehicle sub-group identified by
the two inspection techniques. The incidence of failures due to excessive
HC and/or CO emissions within each test mode is summarized in Table 2-7.
The results are shown individually for the two inspection fleets in each
State, and a composite distribution is shown for each inspection approach by
combining the appropriate inspection fleet data for the two States.
Among all of the fleets tested in the study, excessive CO emissions
in the idle mode was seen to be the most prevalent type of failure; about
70% to 80% of the failed vehicles had idle mode CO emissions in excess of
the inspection standards. Excessive idle mode HC levels were observed for
as many as 50% of the vehicles failed in the various fleets. Excessive
HC or CO levels in the cruise modes were found in about 35% to 45% of the
failed vehicles; but in nearly all cases, excessive cruise mode emission
levels ocurred in conjunction with excessive emission levels in the idle mode.
In an average of only 4% of the cases were vehicles failed only on the basis
of excessive cruise mode CO levels; and the number of vehicles failed only
because of excessive HC emissions in the cruise modes was insignificant.
A potential advantage of a loaded mode inspection test over one which
measures emissions only in the idle mode is the ability of cruise mode
testing to identify vehicles having malfunctions which are evidenced only when
the vehicle engine is under load. Partial blockage of the air cleaner and
certain types of ignition system failures are typical of this. However, the
above analysis indicates that, for the vehicle fleets tested, approximately
95% of the vehicles failed using the Key Mode inspection approach would also
have been failed had only an Idle Mode inspection been performed.
-------
2-40
—I CU
O S-
*~^ ^3
CJ
OO
CM
tO
00
00
CM
C_J
00
CM
O i—
o
o o
oo
I I I ,—.
—I T-
CO
«C M-
I=So
U_
i—i '
_l
<:
o
O IO
uo
CM
LO IO
CO
in o r>>
i
CM
LU
CD
RACTERISTICS
FAILURE C
'0|
O
cj o
CO CD
LU S-
_J 3
O i—
V—i «r-
_L_ (O
LU M-
>
M-
_l O
a
CO
LT>
O
1C E
O
Lf)
CM
LU
1=1
O
LU
OL
OJ
O)
O
cu
Q.
CU
•a
cu
•o
cu
-o
o
cu
^
cu
cu
o
cu
Q.
cu
•a
-------
2-41
This explains partially why similar initial emission reductions were
observed for both the Idle Mode and Key Mode fleets, since both inspection
approaches selected essentially equivalent groups of vehicles to be subjected
to maintenance. However, substantial changes in the prevalence of
cruise-mode-only failures, as might result when an idle emission inspection
program ensured maintenance of idle parameters without necessarily causing
power system malfunctions to be corrected, would result in significant differ-
ences between the vehicle groups failed by idle and loaded mode inspection
tests.
MAINTENANCE AND DIAGNOSTIC PERFORMANCE
Ultimately, an inspection/maintenance program can accomplish a
reduction in vehicular emissions only to the extent that it causes more
and/or better vehicle maintenance to be performed than would have occurred
voluntarily. Clearly, the quality of the maintenance performed in response
to an emission inspection failure will have an important impact on the
effectiveness and cost of the inspection/maintenance program. The perform-
ance of the service garages participating in the Short Cycle Study has
been evaluated from two diametric viewpoints. One has considered the extent
to which the repair agencies were capable of performing maintenance which
would permit the vehicles to comply with the inspection standards and,
thereby, achieve the primary goal of reducing exhaust emissions; and the
second, the extent of excessive repair, and the resultant unnecessary cost
burden, which can be attributed to poor diagnostic performance.
The adequacy of the maintenance may be judged by considering the
test history of the Tailed vehicles following the first and second service
events for the Key Mode test fleets and the Idle Mode test fleets, which
is shown in Table 2-8. The maintenance performance is determined to be
-------
2-42
O LU
LU h-
cc:
oo
«=t —I
D- i—i
LU CC
Di U_
cc: i—i
o s;
cj o;
CO
co
o
D_
C/)
LU
D;
co
i
o-
LU
cc: ct:
i—i
>
ct:
I— LU
o oo
LU
o; o
o: 2:
o o
c_> c_>
s: LU
i—i OO
o
CM
CO
03
o
Ol
oo
cu
o
CU
oo
o
o
O)
oo
o
OJ
Q-
CO
O)
O)
cu
0
cu
oo
cu
u
cu
oo
"O
E
O
O
(L)
00
-------
2-43
adequate only on the basis of the inspection criteria; if the vehicle
complies with the standards, the repair action is considered satisfactory.
Approximately 70% of the failed vehicles in both inspection fleets received
sufficient maintenance during the first servicing to comply with the inspec-
tion standards on retest. About 10% more passed after a second servicing.
13% of the initial failures received inadequate diagnosis and were terminat-
ed as failures. In general, these vehicles were judged to have been capable
of achieving compliance with reasonable maintenance; however, the excess
cost incurred due to the inadequacy of the diagnosis prevented their repair
for $100 or less. The remaining 7% to 10% of the failures required major
repairs in excess of $100.00. These vehicles were terminated as failures
and judged to be unservicable.
The second consideration in analyzing the repair actions requires an
assessment of the excessive or unjustified maintenance. Initially, the
test program personnel conducted a diagnosis of the inspection failures
to identify the maintenance actions which would be necessary to comply
with the emission standards. The repair action of the service garages was
then analyzed and a determination of excess repair was made. Table 2-9
summarizes the maintenance requirements for each inspection fleet. Table
2-10 summarizes the actual incidence of excessive repair that was identified
for the following maintenance categories:
-------
2-44
*
s_
CU
03
O
t- T-
o c
CM
CO
td
00
UJ
LU
Q
LU
C£
>—i
ID
cr
e>3 S-
E re
O) Q-
-P CU
CO Qi
oo s-
o
C. -P
o cu
•P 3
•i- -Q
C. S-
01 re
i—i CJ
s-
o
cu ••-
s- re
=3 Q.
o CU
S- QL
re
o
O E
•,- CU
-P -P
•i— co
OX/V
•P S-
s- re
3 Q.
-Q CU
s- o;
re
CJ
oo
oo
OO
co
co
CM
CM
vo
CM
S-
o
CM
LU
_J
CQ
O E S-
•r- CU •!-
O
oo
I—I
I/O
o
CD
«=t
cnoo
Q.
CU
co
cu
O)
T3
•=C
tn
CM
00
re
Q.
a>
to
o
o
cu
o
•"-5
co
CU CD
r— S-
u- rs
r—
CD -i-
-o re
0 4-
•>
Cf —
CD O
"O ^5
i — i — -
S-
re
ex
co
CD
T3
^_
O
E
1—1
*
CU
S- E
O -t->
c: co
•' — ^3
T3
<:
o _c s-
4-> •!-
>>-r- re
o S d
£Z CU
CO CO Od
rj cu
CTi — CO
CU O CO
s- -i- cu
u. ^r o
cu x
>• LU
:
§^
0
un
C
0
^—«
LU
LU
_J LU
U_ OO
c^
•z. zc
O Q.
1 — 1
1— 1—
cj oo
LU LU
Q- h-
OO
KH
•i—
-P
o
cu
CD-
CO
C CU
i— i C
o
cu
-o cu
O co
:>_ re
c~*
>} Q_
O)
^2
c?^
CM
CM
0
2
h-
cu
co
re
JZ
Q-
'
o~^
<*o
CO
C
o
•r—
4->
o
cu
CL
CO
C
h— I CD
E
CD 0
-a
o cu
s: co
re
CU -E
i— Q-
-o
1 — 1
^
<3~^
CM
CO
0
3
h-
cu
CO
ra
^:
Q-
-------
2-45
*Minor Adjustment - routine idle adjustment of the carburetor
and distributor.
*Minor Replacements - replacement of minor parts such as the
air filter, PCV valve, heat riser, vacuum
lines, gaskets, etc.
*Carburetor - repair, rebuild, or replace the carburetor.
*Major Ignition Tuneup - replacement of plugs, points, condenser,
distributor, and associated adjustments.
*Major Mechanical - major work involving the repair or replacement
of items such as rings, valves, etc. up to a
total of $100 for parts and labor.
The guidelines for the test program required that any vehicles
needing repair work in excess of $100.00 be classified as unservicable
and terminated from the program. The exhaust emission data, however,
remained in the fleet statistics.
The maintenance requirements identified in Table 2-9 exhibit the
same trends observed for the modal failure characteristics shown in
Table 2-7. The incidence of excessive idle mode emissions was the pre-
dominant cause for failure; therefore, the need for idle adjustment would
logically be expected to dominate the other repair categories.
The frequency of excessive repair is presented individually for
Phase One and Phase Two of the test program. Prior to initiating Phase Two,
a second group of service garages was selected for participation in the
test program. The orientation and instruction techniques were re-eva'luated
in an attempt to ensure that the maintenance personnel understood the
diagnostic capability of the emission tests. An effort was also made to
stress the importance of minimizing unjustifiable repairs.
-------
2-46
The total incidence of excessive repair was highest for the Key Mode
inspection fleet during Phase One. A comparison of the ignition system
maintenance required (Table 2-9) and the associated incidence of excess
repair indicates that extensive impact; is incurred from this repair category
alone. This observation is attributed to a strong tendency for the maint-
enance personnel to rely upon conventional maintenance practices for re-
pairing the individual vehicles.
The effort to improve the garage performance during Phase Two effected
a substantial improvement for the Key Mode fleet, reducing the frequency
of excess repair by more than one-half. The change observed for the
Idle Mode fleet was insignificant. This effect would be expected, based
upon the assumption that excess repair among the Idle Mode service
garages results from the insufficient diagnostic content of the idle
emission test. The excess repair performed by the Key Mode service garages
would be attributed to an inadequate diagnostic interpretation of the
test results.
COST OF EMISSION-RELATED REPAIR
One factor in the cost of an inspection/maintenance program is the
cost of the maintenance required for the vehicles failed by the inspection
test. This cost includes both the cost for diagnosing the failed vehicles
to determine how to bring them into compliance with the inspection standards,
and the cost for labor and parts to perform those repair actions. In
practice, the out-of-pocket cost for repair of failed vehicles will
probably exceed the net additional maintenance cost imposed by the
inspection/maintenance program, since some of the maintenance performed on
failed vehicles is likely to substitute for voluntary maintenance which
-------
2-47
would otherwise have been performed. The effect on net inspection/
maintenance program cost of substitution of required maintenance for
voluntary maintenance is discussed in Section 2.4; the maintenance cost
figures presented in this section refer only to the actual incurred costs
for repair of failed vehicles.
Table 2-11 summarizes the maintenance costs per serviced vehicle
observed in each phase of the Short Cycle Study for both the Idle Mode
and Key Mode inspection fleets. In addition to the actual incurred cost,
a corrected cost is presented for each fleet and phase which has been
derived from the incurred cost by subtracting the cost of the excessive
repair identified in Table 2-9. Averaged over both phases of the study,
the cost of excessive maintenance represents about 30% of the observed
cost for both inspection fleets.
In all categories, the Key Mode fleet service costs were higher than
those for the Idle Mode fleet; when both were corrected for excessive re-
pair, the Key Mode maintenance costs averaged about 20% higher than those
for the Idle Mode fleet. This observation is not unexpected, since the
diagnostic capabilities of the loaded mode test provide justification for
more comprehensive, and generally more costly, repair actions.
To facilitate comparison with other in-use vehicle emission control
approaches, the maintenance costs are also presented as an average cost
per vehicle in the fleet, including both serviced and non-serviced vehicles.
It can be seen from Table 2-11 that the average maintenance cost per vehicle
in the fleet might be expected to range from about $9 to $14, depending
on the inspection approach and the extent of excessive repair.
-------
to
O
2-48
cu
cu
cu
T3
O
CVJ
CM
•foO-
o
CM
*d-
CM
o
o
cu
^
cu
to
o
Q.
E
O
cu
S-
o
t/1
O
C\J
CO
OJ
LT>
CM
o
LO
00
CM
LD
CM
Qi
cu
cu
cu
T3
O
T3
O>
•t-)
O
CU
s-
o
10
o
CO
Lf>
tn
co
CO
00
CM
CQ
CO
o
o
cu
•o
I—I
cu
O T3
Q. CU
O i-
CM
LO
•d-
CM
•bO-
CM
LO
CM
CM
cu
I—
o
cu
-a
cu
o
CU CU
c to
o
cu cu
in Q.
(O
cu
cu
-a
cu
o
CU
to
s-
a> cu
to
(O
0)
O
-C CU
£ CU O)
fO > (O
*- s-^.
CDT3 CU ,
O CU >
i- 0 <^
> +» '!
Q-
_. cu
-------
2-49
INSPECTION COSTS
In addition to the cost for maintenance of vehicles failed by an
inspection/maintenance program, the costs associated with inspecting the
vehicle population must be evaluated. These costs include both the in-
vestment costs to provide inspection facilities and equipment, and operat-
ing costs for carrying out the inspections and administering the inspection/
maintenance program.
There are two basic types of inspection system configurations which
may be considered: State-operated inspection lanes, which would be devoted
exclusively to vehicle inspection activities on a high-throughput basis;
and licensed private garages, which would conduct emission inspections in
addition to their normal maintenance and repair activities.
In a licensed private garage system, the inspection cost per vehicle
will depend strongly upon the number of vehicles inspected per garage
and the administrative costs; these, in turn, will depend upon the size
and regional distribution of the vehicle population, the number of garages
licensed, and the possible previous existence of an administrative frame-
work for safety inspection. Since these factors will vary significantly
among the regions which may be considering the implementation of emission
inspection programs, cost estimates for licensed private garage systems
must be made on a regional basis and no generally applicable inspection
cost data for such approaches can be presented here. Estimates which have
been made in specific cases range from about $3 to $6 per vehicle for an
8
idle mode emission inspection.
Table 2-12 presents estimates of the investment and annual operating
costs for State-operated inspection lanes using the Idle Mode and Key
Mode procedures. Estimates for both approaches provide for semi-automated
-------
2-50
TABLE 2-12
o
Inspection Station Cost Estimates0
Cost Element Station Type
Idle Mode Key Mode
Investment Costs 1 lane 2 lane 1 lane 2 lane
Inspection Equipment $11,200 $22,400 $14,000 $28,000
Administration $1,000 $1,700 $1,000 $1,700
Site Acquisition
($2/sq. ft.) $14,380 $20,000 $21,800 $30,220
Construction $10,960 $16,320 $16,320 $24,480
($8/sq. ft.)
TOTAL $37,540 $60,420 $53,120 $84,000
Operating Cost (1st year)
Personnel Salaries $22,000 $44,000 $22,000 $44,000
Supplies & Maintenance $ 1,748 $ 3,186 $ 2,216 $ 3,994
TOTAL $23,748 $47,186 $24,216 $47,994
Annual Capacities of Inspection Lanes: Idle Mode - 32,000 vehicles/lane
Key Mode - 25,000 vehicles/lane
-------
2-51
processing of emission test data. Since the data acquisition systems are
the major equipment cost and both test regimes require similar data
acquisition systems, the equipment investment costs per lane for the Idle
Mode and Key Mode inspection approaches differ by only about 20%. Labor
is seen to be the predominant operating cost and this is considered to
be the same per lane for both the Idle Mode and Key Mode approaches.
In comparing the total costs of inspection programs using the two approaches,
however, the higher capacity of the Idle Mode lanes must be taken into
account.
Regional cost estimates for an inspection program must include
additional expenditures required for training inspection personnel ,
program planning, and the initial qualification and certification of the
inspection facilities. The operating cost element must also consider the
maintenance and depreciation of the inspection facilities, and the total
system administration and enforcement requirements.
An operational analysis of the inspection station design for each
test procedure, and an analysis of the distribution of vehicle population
8
densities was performed for the State of California. The total program
cost elements are summarized in Table 2-13. The testing capacity of the
inspection system is based on a yearly inspection of the total California
population of 10 million vehicles.
-------
2-52
TABLE 2-13
Total Program Costs +
(Thousands of Dollars)
Cost Element Idle Mode Key Mode
Investment Costs
Site Acquisition & Construction 7,117 (59)* 12,445 (63)
Equipment & Installation 4,090 (33) 6,270 (32)
Planning & Training 112 ( 1) 193 ( 1)
Qualification & Certification 745 ( 7) 912 ( 4)
TOTAL 12,064 19,820
Operating Cost (1st year)
Salaries of Inspection Personnel 6,635 (67) 6,648 (61)
Salaries of Administrative
Personnel 631 ( 7) 981 ( 9)
Equipment Maintenance &
Depreciation 1,214 (12) 1,473 (14)
Facility Maintenance &
Depreciation 256 ( 2) 516 ( 4)
Program Administrative Costs 1,243 (12) 1,301 (12)
TOTAL 9,979 10,919
*Percentage of Total
+Estimated for California vehicle population (10 million)
The initial investment burden of the inspection system can potentially
be pre-financed by increasing the vehicle registration fee prior to the
program implementation; alternatively, the total investment can be
amortized on a fixed capital return schedule. Table 2-14 compares the
initial investment cost and the annual capital return schedule required
for the two test regimes in the example discussed above.
-------
Total
(Thousands)
$12,064
$ 2,160
Cost per
Vehicle Test
$1.21
$ .22
Total
(Thousands)
$19,820
$ 3,549
Cost per
Vehicle Test
$1.98
$0.35
2-53
TABLE 2-14
DISTRIBUTED ANNUAL INVESTMENT RETURN
Idle Mode Inspection Key Mode Inspection
Tc
(Thoi
Total Investment
Annual
Capital Return*
*Capital investment amortized over 10 years at 6%.
Discounting the annual capital return plus the inflated annual operat-
ing cost for the first ten years of operation yields an average annual cost
per vehicle test in 1972 dollars. The average annual cost figures are
$1.04 for the Idle Mode test and $1.22 for the Key Mode test.
Assuming that a retest would be required following maintenance, the
cost per vehicle test should be proportionately scaled to account for the
failure rate. This provides an estimate of the average annual cost per
vehicle which facilitates a more convenient estimate of regional costs.
The estimated annual cost per vehicle is assumed to be valid for
any region which has a vehicle population density and an urban-rural
distribution similar to that of California. The validity of this assump-
tion relies on the dominance of the total system cost by the labor cost
element. Because the total labor time is directly proportional to the
number of vehicles subject to the inspection, an estimate of the total
program cost may be calculated based on the regional vehicle population
and the annual vehicle cost for the appropriate inspection procedure.
-------
2-54
2.3.2 ENGINE PARAMETER INSPECTION
The engine parameter inspection requires a functional test of
specific engine components and adjustable parameters which are
determined to have a significant effect upon vehicle emissions, and which
exhibit an extensive rate of deterioration. As a general class, the engine
parameter inspection technique may include any combination of diagnostic
procedures for emission related subsystems of the engine. In addition
to sJletting the most cost-effective parameters to be tested, the optimal
design of an inspection policy requires a determination of the appropriate
failure criteria.
FLEET EMISSION REDUCTION
The data presented in this section are applicable to the extensive engine
parameter inspection which was described in Section 2.2. Both the inspection
and resultant maintenance were performed by an experienced emission testing
laboratory. Table 2-15 summarizes the results for each control group.
The 1971 California vehicles are listed separately to reflect the new
control mechanisms which were incorporated to comply with the California
emission standard for oxides of nitrogen.
The large number of individual parameters that were inspected resulted
in the extremely high rejection rate, although many vehicles only required
very minor adjustments.
Figures 2-5 (a&b) present the 90% confidence intervals for the fleet mean exhaus
emissions after maintenance. HC and CO emissions are shown for the three
separate control groups. The data indicate that the engine parameter inspection
approach acheived a significant reduction of HC emissions at the 95% confidence
level, whereas CO emissions were not significantly reduced above the 80% confidence
level. The change in NOx emissions was not statistically significant.
-------
2-55
TABLE 2-15
INITIAL EFFECTIVENESS OF ENGINE PARAMETER INSPECTION/MAINTENANCE
Exhaust Emissions*
Test Fleet Rejection Rate HC CO NOx
(Percent)
Pre-Control Vehicles 95
As Received (gpm)
After Inspection and
Maintenance (gpm)
% Reduction
11.94
10.80
9.5%
130
122
6.2%
3.97
4.22
-6.3%**
1966-1970 Control 95
Vehicles
As Received (gpm) 7.26 86.6 6.34
After Inspection
and Maintenance (gpm) 6.24 80.5 6.50
% Reduction 14% 7% -2.5%**
1971 NOx Control 95
Vehicles
As Received (gpm) 4.43 63.4 6.00
After Inspection
and Maintenance (gpm) 4.36 59.2 5.88
**
% Reduction 1.6% 6.6% 2.0%
* As measured by 1972 Federal Certification Test Procedure.
** A negative reduction represents an increase. NOx changes were not statistically
significant.
ESTIMATED VEHICLE OWNER COSTS
The cost elements for the parameter inspection are attributed to the
inspection time required for the engine diagnosis and the labor and parts
required for proper maintenance. The labor cost is based on a burdened
overhead rate of $10 per hour. It is assumed that the inspection system would
be composed of existing privately owned automobile service facilities licensed
by the state.
-------
2-56
12
10
Q.
O
to
to
CO
o
m
tr.
«t
o
o
a:
8
co
90% CONFIDENCE LIMITS
SHOWN FOR AFTER
SERVICE EMISSIONS
PRE-1966 1966-1970 1971
CALIFORNIA VEHICLES
FIGURE 2-5a
FLEET EMISSIONS BEFORE AND AFTER
SERVICE ENGINE PARAMETER INSPECTION
-------
2-57
14
12
o
- 10
X
Q.
O
CO
CO
CO
X
o
o
CD
cr
-------
2-58
Table 2-16 summarizes the cost elements of the extensive
engine parameter inspection. The inspection time estimates
are based upon the observed performance of the project
personnel during the actual inspection of the test flett. The
total inspection fee includes an additional $1.00 per vehicle
to cover state administrative and enforcement cost for the
licensed garage system.
TABLE 2-16
ENGINE PARAMETER INSPECTION TEST COST
Subsystem Inspection Time Inspection Cost
Idle Adjustments 0.15 hours $1.50
Secondary Ignition System 0.25 hours $2.50
Induction System 0.25 hours $2.50
Total Inspection Fee per Vehicle = $7.50 (includes $1.00 Processing
Cost)
The minimum maintenance cost was determined from the
extent and frequency of repair that was observed for the test fleet.
The labor time required for the repair of each engine parameter was
extracted from Chilton's flat rate manual. Table 2-17 lists the
repair cost for each parameter that was i nspected. Heighting the
cost for each parameter by the frequency of malfunction yields the
expected cost per vehicle for that parameter.
-------
$ 1.00
$ 1.00
$ 1.00
40%
50%
20%
$ .40
$ .50
$ .20
?-59
TABLE 2-17
VEHICLE REPAIR COST FOR ENGINE PARAMETER INSPECTION
Repair Cost Frequency of Expected Cost
Engine Parameter (Parts and Labor) Occurrence per Vehicle
A/F Ratio
Timing
Idle RPM
Misfire $27.00 5% $1.35
Air Pump $54.00 3% $1.62
PCV Valve $ 2.80 20% $ .56
Air Cleaner $ 6.00 20% $1.20
Heat Riser $ 2.50 10% $ .25
Choke Blade $ 2.50 5% $ .12
Total Fleet Average Cost $6.29
Average Cost for Serviced Vehicles Only (90% of Total Fleet) $7.00
The Chilton's Flat Rate Manual was used to obtain an index for the service
cost estimate which was then adjusted to reflect the performance of the test
program personnel who actually performed the repairs. Because of their
familiarity with the requisite diagnostic and repair functions, the maintenance
estimates in Table 2-17 are expected to be considerably below the initial cost
that would actually be incurred in a typical service garage. An empirical
evaluation of garage effectivenss was conducted to determine the capabilities
of the private service industry to diagnose and repair specific engine
9
malfunctions . The results of this study provide a means of adjusting the cost
estimates shown in Table 2-17 to account for the frequency and extent of unnecessary
repair costs which are likely to occur in practice. It was found that the actual
incurred costs for repairs typical of those needed to comply with the inspection
program (i.e., idle adjustments plus a component replacement in either the induction
or the ignition system) was $22.00. Therefore, during the initial implementation of
the inspection program, the annual cost of repair may be up to three times the
optimal figures outlined above.
-------
2-60
2.3.3 MANDATORY MAINTENANCE
The feasibility of an enforced mandatory tune-up requires the
identification of specific engine components and adjustment
parameters which have a si gni fi cant effect upon vehicle
emissions and whose malfunctioning can be predicted with a reasonable
degree of certainty. The cost-effectiveness of mandatory
maintenance is highly sensitive to the identification of the
optimal tune-up specifications for each engine component and the
reliability of predicting the frequency of malfunction.
10
The test fleet used to evaluate this strategy was composed
of pre-1966 vehicles representative of the domestic vehicle
population in California. The repair action included the
replacement of the major components in the secondary ignition
system (spark plugs, spark plug wires, breaker points, condenser, and
distributor rotor), the air cleaner filter, and the PCV valve, and adjustment
of the idle parameters (air-fuel ratio, rpm, and timing). The estimated cost
for the required parts and labor was $55 per vehicle. A 15% average reduction
of exhaust HC emissions and an 11% average reduction of CO emissions were
observed following the tuneup of the entire sample fleet. There
was no significant change in the fleet mean NOx emissions.
Every vehicle in the test fleet received identical maintenance
regardless of the as-received exhaust emissions. Figures 2-6 and
2-7 present the total fleet average emissions after tune-up
as a function of the percentage of vehicles receiving maintenance.
For this analysis, the vehicles were ranked according to the as-
-------
2-61
11
s
P.
00
w
u
w
M
o
§
U
PC
10.5
10.0 _
9.5 -
9.0 -
8.5
20
80
40 60
PERCENT OF POPULATION
FIGURE 2-6
HYDROCARBON REDUCTION vs. VEHICLE REJECTION RATE
100
-------
e
a,
oo
w
u
w
M
O
ps
o
o
o
iz:
<
w
H
W
W
2-62
120 ._
115
110
105
100
95
90
j
20
I
j
80
40 60
PERCENT OF POPULATION
FIGURE 2-7
CARBON MONOXIDE REDUCTION vs. VEHICLE REJECTION RATE
100
-------
2-79
emission level is calculated on the basis of the elapsed time between
maintenance events.
It is asssumed that the enforcement of the inspection program will
require the remaining vehicles to receive maintenance as frequently as
the inspection interval. The time averaged emission level is calculated
for these vehicles, and the weighted sum of the two segments of the
population represents the steady state emission baseline for the
appropriate inspection interval. The calculations are defined as
follows:
Pp [Eo + (DF x MJp x %)] = Ep
Pf [Eo + (DF x II x l/2)]=Ef
(Pp x Ep) f (Pf x Ef) = Et.
ER = (Eni - Eti) /Eni
where: Pp = fraction of vehicle population currently serviced more
frequently than the inspection interval.
Pf = fraction of vehicle population currently serviced less
frequently than the inspection interval.
Eo = minimum emission level immediately following maintenance.
DF = deterioration factor in terms of grams per mile per
month.
MIp= current average maintenance interval of vehicles passing
the inspection test.
Ep = time averaged emission level of vehicles passing inspection
test.
II = inspection interval
Ef = time averaged emission level of vehicles failing
inspection test.
1/2: constant to determine average emission level between
maintenance events assuming a linear deterioration rate.
-------
2-80
Eti= emission baseline for total vehicle population
following the implementation of an inspection/
maintenance program.
Eni= emission baseline for total vehicle population prior
to the implementation of an inspection/maintenance
program.
ER = emission reduction relative to the initial emission
level (Eni).
The effectiveness of a twelve month and a six month inspection interval
corresponding to each of the three deterioration factors are shown in
Table 2-20. Only the reduction of HC and CO emissions were calculated
because the results of the maintenance programs have not demonstrated a
statistically significant change in the emissions of nitrogen oxides.
TABLE 2-20
TIME AVERAGED EFFECTIVNESS OF ENFORCED INSPECTION/MAINTENANCE PROGRAMS
Exhaust Emission
Inspection
Interval
12 months
6 months
Annual Vehicle Rejection
Rate (% of total fleet)
30%
81%
Deterioration
Factor
High DF
Median DF
Low DF
High DF
Median DF
Low DF
Reduction
HC
7.4%
11.9%
16.7%
11.2%
14.7%
18.1%
CO
5.72%
9.5%
12.5%
8.6%
11.6%
13.9%
-------
2-81
2.4.4 ECONOMIC ESTIMATES
The incremental costs which can be attributed to the enforcement of
an inspection program include the capital investment requirement and
the annual operating cost of the inspection program, and the
additional economic burden placed on the vehicle owner as a result
of the additional maintenance required to comply with the test standards.
To provide the widest range of applicability, the inspection costs
are estimated for the publicly operated inspection lane configuration.
The cost estimates presented in Section 2.2 indicate that an annual
cost of $1.22 per vehicle inspection would be adequate to account for the
administrative expense of a dynamic mode emission inspection utilized in
a high throughput lane configuration. An assumed inspection fee of $1.50
per vehicle inspection is used in the following calculations to ensure
the coverage of any additional hidden cost factors. The incremental
maintenance costs are based on the higher frequency of maintenance imposed
by the inspection program. There is no attempt to account for a cost
increase per maintenance event. It is assumed that the improvement in
maintenance procedures to achieve the maximum reductions does not require
the use of repair techniques which are any more comprehensive than those
generally recommended by the vehicle manufacturer.
The methodology used to calculate the incremental maintenance cost
is similar to the emission baseline calculation. Conservation of existing
maintenance is assumed for that portion of the vehicle population which
currently receives maintenance more frequently than would be required by
the inspection policy. Those vehicles which receive maintenance less
frequently, receive the incremental cost burden as the maintenance
-------
2-82
interval is reduced to comply with the inspection test. The cost
burden imposed on this fraction of vehicles is then distributed over
the entire population such that the cost figure per vehicle is comparable
to economic estimates for other emission control strategies.
The estimated cost for existing maintenance is provided by the
Champion Survey data which indicates an average of $36.86 per service
event. The cost for incremental emission related maintenance is assumed
to be $28.52 per service event. This figure reflects the pooled average
repair cost per serviced vehicle in the Short Cycle Study. The annual
rejection rate will determine the number of additional tests required
for the after service compliance test. The added cost for the compliance
testing is assumed to be included in a single inspection fee for all vehicles
subject to the enforcement program. The cost calculation is defined as
follows:
Cm = Pf (II x 1/12 x Cem) - Pf (MIf x 1/12 x Ccm)
CI = Cvt [(II x 1/12) + Pf ]
where:
Cm = the incremental annual cost per vehicle for additional maintenance
requirements.
Cem= cost per service event for emission related maintenance.
Ccm= cost per service event for present conventional maintenance.
Pf = fraction of vehicle population currently serviced less frequently
than the inspection interval.
II = inspection interval in months.
MIf= current average maintenance interval of vehicles failing
inspection test.
CI = annual inspection cost per vehicle.
-------
2-83
Cvt= annual inspection cost per vehicle test.
The incremental cost estimates for a twelve month and a six month
inspection program are summarized in Table 2-21.
TABLE 2-21
INCREMENTAL COST ATTRIBUTES FOR
AN ENFORCED INSPECTION
PROGRAM
Inspection Interval
12 months 6 months
Inspection Cost per Vehicle $2.00 $4.22
Incremental Maintenance $2.56 $15.40
Cost per Vehicle
Total Average Cost per $4.56 $19.66
Vehicle
Total Average Annual
Cost, per Vehicle
for 1975-1980
(present worth) $3.97 $17.08
-------
2-84
2.4.5 COST-EFFECTIVENESS SUMMARY
The cost and effectiveness results estimated by this analysis are
summarized in Table 2-22. Also shown are values of a cost-effectiveness
index derived by equally weighting the percentage reductions for all three
pollutants and dividing the weighted reduction by the program cost. Because
the same basic assumptions apply to both HC and CO emission reductions, the
relative values of the cost-effectiveness index for the various choices of
inspection interval and deterioration factor are independent of the
pollutant weighting factors chosen; therefore, equal weighting was chosen
for simplicity. For the entire range of deterioration factors considered
in the analysis, a twelve-month inspection interval is seen to be less
effective, but more cost-effective, than a six-month interval. That is, the
additional reductions obtained by decreasing the inspection interval from
twelve to six months are more costly than the reductions first obtained by
implementing a twelve-month interval program. However, as is shown by
Table 2-23, the relative effectiveness and cost-effectiveness of twelve
and six-month inspection intervals vary significantly with the assumed
deterioration factor.
Thus, a principle parameter in the cost-effectiveness analysis is the
effectiveness of current conventional maintenance relative to that which
may be performed in response to an enforced inspection/maintenance program;
for this relative effectiveness provides the basis for quantifying the
deterioration factor. If it is assumed that current maintenance is
effectively performed (a relative effectiveness factor of 1.00), the results
obtainable through tuning up test fleets indicate thatr.the rate of
-------
2-85
CM
CM
1
CM
LU
_J
CO
e£
r—
>-
LU
r-
e^
Qf^
H-
00
1 i 1
LJ_I
o
"^
^^
z
LU
t^
i — i
«=C
s:
•z.
o
I-H
1—
0
LU
O-
00
•z.
\ — 1
U_
O
00
00
LU
z.
LU
•^^
I-H
1—
O
LU
U_
U_
LU
1
I—
00
o
o
to
CO
CO
c
CO
•r-
-p
o
CO
LU
1 X
co -o
0 C
O I-H
-o o
CO T-
•p -p
.C 0
CT> 3
CO CU
3 C£
O
CO
cn
*~ X
3; co O
o c z:
C£. ^3 O
O 0 <4- T-
u. ce o co
a: co
1— C T-
0 S
LO T- LU
r-N 4-> o
en c_> -P o
i— 3 tO
-o rs
CO rO
fV f"
X
LU O
*T~
C
0
•r—
40
ra
i-
O
•r- i.
i- O
CU 4J
4-> CJ
CD
O CO
c o
I-H (_J
c
0
•i— r—
4-> ra
o >
CU S-
Q. CU
co -P
c c
I-H H- 1
r- CO «* *3" LO tD
r— I— CM O O O
^ r— 1 — CO 00 f--.
^j- i — « en to oo o
^—
O 0 0 0 0 O
^? ^^ ^^ ^R ^s ^^
\£> O CM CT» C\J "sj-
i—1 r*^ r™~ i~™
^^ ^^ ^^ ^^. ^^ ^^.
rs c\j r^ i — LO co
r~" r— • r— ^— r~-
U_ u_
^"^ ^^
U_ LJ_
O C U- O C ti-
ro L^ rO C^
jz T- j: •!-
cn T3 2 O) -o S
•r- CO O -r- CO O
3r s: — i =c 2: — i
0 0
o o
• •
^~ ^«
1 —
•to- -faO-
to to
C"" f~
4-> 4J
c c:
o o
E E
CM to
r—
oo
C£
LU
1—
LU
^c
ry*
^£
a.
oo
CO
LU
co z
CM LU
1 i— <
I—
CM 0
LU
Lu
LU LU
LU
1 |
h-
CQ OO
O
< 0
I™™* ^^
o
>—
I—
I— »
>.
1— «
1—
t-H
OO
^^
LU
00
O
ir>
*
o
LT>
f*^
•
o
o
o
•
r—
CO
U
c
rO
C
CO CO
in -i- *
V) -r-*
c. c
CO ••-
> ro
"^ ^"
4J
O r—
CO ro
t)^T £^
H- O
1 1 1 «f«
4->
CO C
> CO
•I- >
4-> C
ro O
r— O
CO
0; s_
o
T3 <+-
cu
E X
3 CU
to -a
CO C
«=t I-H
to
>
s_
CO
4J
C
•r-
C
O
•r-*
4->
U
CO
Q.
CO
C
4->
C
O
E
CM
^~
tO
cn
C
'£
•r—
3
cr
CO
E
ro
S-
CD
i— LO O
cn co s-
Q.
O CO
CO
o
c
ro
r— in C
00 "H- CO
4->
O CO C
rO
£
^•x^
to co c:
to i — o
• • *^
O CM 4->
O
CU
Q.
CO
c~
• ^
C
rO
S-
o
q-
c
0
4->
4J U
O 3
co -a
if- i — i i— i CU
Ct_ 1— 1 I-H S_
LU
• • f~
• o o o
4-> LU tO
V 3 to O to CO
^^ • «i^
N. ^^ E
• ^^. ^^k
•P ^\^ ^^v "CJ
O \ ^\ CO
CU KH l-H 4J
t|— I — I i — I _C
M- cn
LU . •!-
o o cu
• E * E 3
-a 1-1
4-> CM LU CM
3 r- Or- +
•
^™
ra
>
s_
co
c
•I—
c
o
•r"
4->
o
CO
a.
CO
c
•r-
^
4->
O
E
CM
p—
ra
en
C
•If-
i.
•r-
cr
cu
i.
E
IB
Cn
O
a.
CO
u
c
rO
C
CO
4J
c
•r—
ro
E
r-
O
•i —
4J
U
CO
Q.
to
c~
•i —
C
ra
S-
o
H-
X
cu
•a
c
•r-
to
CO
ai
c
cu
•r—
4->
O
CO
"4-
tu.
a>
|
4J
to
O
t^_J
*
-------
2-86
deterioration as a function of time is high. In that case, a short
inspection interval would be necessary to keep vehicle emissions at
a minimum, but the present performance of the maintenance industry would
appear adequate to achieve the expected reductions.
On the other hand, if the emission reductions typically achieved
through conventional maintenance are small (a relative effectiveness
factor of 0.50), the empirical test fleet data imply a low rate of
deterioration. In that event, only minimal additional emission reductions
would be obtained by shortening the inspection interval. However,
substantial improvements in the diagnostic and repair performance of
the maintenance industry would be necessary to ensure the effectiveness
of the inspection/maintenance program.
There is a limited amount of information in the literature which
provides some guidance as to the relative effectiveness of current
5
conventional maintenance practices. The garage effectiveness study
discussed in Section 2.3.2 indicates1 that the diagnostic and repair
performance of the current maintenance industry achieves reductions between
60% and 80% of the maximum reductions that could be achieved through
14
emission-related maintenance. On the other hand, an earlier study
indicated'that conventional tune-up practices achieve reductions of less
than one-half the magnitudes obtained when optimal emission tune-ups were
performed. However, in this latter study, the optimal tune-up procedures used
as a reference included adjusting the vehicles to the minimum idle CO levels
achieveable without misfire, rather than to manufacturer's specifications.
Based upon the available data, it appears that an intermediate
effectiveness factor of 0.75 is a reasonable assumption to reflect the relative
-------
2-87
effectiveness of current voluntary maintenance in reducing "emissions.
Accordingly, the expected average effectiveness over time of an
annual inspection/maintenance program would be as follows:
HC exhaust emission reduction 12%
CO exhaust emission reduction 10%
NOx exhaust emission reduction no significant change
-------
2-88
2.5 CORRELATION ANALYSIS
In the context of the need for transportation control measures, the
single purpose of a periodic inspection program is to achieve reductions
in the average rate of vehicular emissions during operation. Since the
Federal Test Procedure is the single standard of measure for defining
typical urban emission levels, an effective inspection test should pass
and fail individual vehicles in a manner similar to the way in which the
same vehicles would pass or fail the FTP. Testing any two vehicles, the
inspection test should preferentially identify the highest emitting
vehicle to receive maintenance. In addition, compliance with the
inspection test failure criteria following corrective maintenance must,
in fact, represent a significant reduction of true mass emission levels
as measured by the FTP if the program is to accomplish its goal.
There are two basic methods for evaluating the potential effectiveness
of inspection procedures in fulfilling the objectives outlined above. The
first approach is to directly measure the change in the average emission
levels for a large test fleet which has been subjected to the appropriate
inspection and maintenance procedures. The results of studies using
this type of approach were presented in Section 2.3.
Alternatively, a statistical analysis can be performed to evaluate
the correlation between the inspection emission measurement and the actual
vehicular emission levels as measured by FTP. Conceptually, this type of
approach is an attempt to quantify the information content of the
inspection test procedure. The degree of correlation indicates the
accuracy with which the inspection test is monitoring true emission rates,
and provides a measure of the assurance that compliance with the inspection
will actually result in emission reductions.
-------
2-89
The data used for this analysis were obtained as a parallel task
in the Short Cycle Study. In each of the four test fleets, independent
emission measurements were made using the FTP and a number of alternative
hot start test procedures and individual test modes. A regression analysis
was performed, and the degree of correlation between the FTP and alternative
short test procedures was determined using the pooled test data from each
fleet. Hereafter, the term short test will refer to any composite test
procedure, or to any individual test modes, which are considered feasible
alternatives for an inspection test procedure.
2.5.1 DESCRIPTION OF TEST PROCEDURES EVALUATED
A wide range of composite test procedures and individual test modes
were evaluated in order to investigate the full spectrum of feasible
alternatives which have been proposed for use in an emission inspection
program. The intent of the analysis was to identify the emission test
characteristics which have the most substantial effect on the correlation
with the Federal Test Procedure. Accordingly, the analysis evaluated
composite transient mode tests, a number of constant velocity cruise modes,
and the idle mode. Both mass emission measurements and volumetric (concen-
tration) emission measurements were also evaluated. The basic characteristics
of each test procedure are summarized below.
1972 Federal Test Procedure! - This is the standard of measure for defining
vehicular emission levels during typical urban driving patterns. The
FTP is a cold start test (the vehicle remains inoperative for a minimum
of twelve hours prior to initiating the test) which utilizes a 23-minute
non-repetitive transient mode driving cycle, and provides a CVS mass
emission measurement of total hydrocarbons, carbon monoxide, and total oxides
of nitrogen.
-------
2-90
Federal Short Cycle Test - The Federal Short Cycle is a composite
transient mode cycle composed of nine operating modes including
accelerations, declerations, cruise modes, and idle operation. The
duration of the test cycle is 125 seconds. Mass emission rates of
total hydrocarbons, carbon monoxide, and total oxides of nitrogen are
measured. The dynamometer load factors are set according to the weight
class of the individual vehicles.
Hot Seven Mode Test - Two repetitions of the standard Seven Mode test
cycle (see Figure 2-1, Section 2.2.1) were made from a hot start. Both
mass emission and volumetric emission measurements were made simultaneously.
The volumetric emission rates provide a measure of total hydrocarbons as
n-hexane (using NDIR), carbon monoxide, and oxides of nitrogen as NO
(using NDIR). The duration of each Seven Mode Cycle is 136 seconds. The
power absorption unit on the dynamometer was uniformly set for 10 hp at
50 mph for every test vehicle.
Steady State Test Modes - Mass emission rates and volumetric emission
rates were measured at constant cruise speeds of 10, 20, 30, 40, 50, and
60 mph. Mass and volumetric measurements were made at zero mph with the
transmission in neutral; this was defined as the idle mode measurement.
The power absorption unit was set to correspond to the weight class of
the individual test vehicle. Hereafter, these cruise modes will be referred
to as the steady state modes to differentiate them from the Key Mode cruise
modes which use a higher loading factor.
Key Mode Test - Mass emission rates and volumetric emission rates were
measured during the idle mode and each of the two cruise modes which
comprise the Key Mode2 vehicle operating cycle. The power absorption unit
-------
2-91
was set for 30 hp at 50 mph. In comparison, the road load setting for a
3000 pound vehicle using the Federal Test Procedure is approximately 10 hp
at 50 mph.
An emission measurement was made for every vehicle in the Short Cycle
Study both before and after service using each of the test procedures
described above.
2.5.2 DESCRIPTION OF CORRELATION ANALYSIS
A regression analysis of the data obtained from the Short Cycle Study
was performed to formulate a functional relationship between the various
short tests and the Federal Test Procedure using the method of least
squares.^ This regression equation provides a means of predicting the
true mass emission (FTP) rate as a function of the short test emission
measurement. The simple regression equation is of the general form:
A
Y = a + bX
A
where Y (the dependent variable) is the predicted FTP value and X (the
independent variable) is the measured emission level from the single
mode measurement or the composite emission level from a short test cycle.
The regression was developed independently for HC, CO, and NOX using
the data from each of the steady state modes, the two transient mode test
cycles, and the Key Mode test.
If the two related emission measurements (X and Y) are plotted
graphically, the individual data points will exhibit some degree of
scatter about the regression line. The extent to which the data points
are scattered is reflected by the standard error of the estimate. This
parameter represents the square root of the mean of the squares of the
deviations about the regression line:
SEE = Lz(Y - Y)2/nj^
-------
2-92
The summation is made over all "n" data points. The standard error of
the estimate provides a measure of the variation in the data and is
used as the basis for establishing confidence limits within which a given
percentage of the predicted values can be expected to fall.
Associated with each of the simple regressions is a coefficient of
correlation which ranges in value from zero to one. This parameter is a
measure of the degree of association between the two variables relative
to the range of values observed for the Y variable, or FTP measurements.
The following equation provides a convenient operational definition of
the coefficient of correlationjalthough it is not usually calculated in
this manner:
r = [1 - {SEE2/VY)3 *
where Vy -js tne variance of the FTP data (or the square of the standard
deviation) and SEE is the standard error of the estimate of the regression
equation. A value of one indicates perfect correlation, and zero indicates
the lack of any association between the two variables. The coefficient
not only represents the accuracy of the regression, but also the significance
of the fit associated with the dispersion or variance of the particular
data set observed. For any given deviation about the regression line, the
significance of the correlation improves (coefficient of correlation
increases) as the variance of the observed data set increases.
The standard error of the estimate does not always provide a
measure of comparison for different data sets because the standard error
is expressed in the original units of the dependent variable (Y). For
example, the standard error of a regression of HC measurements cannot
be directly compared to the standard error of a regression of CO emission
-------
2-93
measurements. Alternatively, the coefficient of correlation is an
absolute term expressed as a fraction. Therefore, this parameter
eliminates the problems incurred when comparing two associated series
of data which have different dispersions and different units.
A multiple regression analysis was performed for the multiple mode
emission measurements wherein each mode is entered into the regression
equation. The multiple regression equation is of the form:
A
Y = a + bX-| + cX2 + dX3 +
A
where Y again represents the predicted FTP value and the X. represent
the independent emission measurements from each individual measurement
mode. Parameters analagous to those calculated for the simple regressions
were determined for the multiple regressions. Thus,the standard error
of the estimate reflects the average deviation between the predicted FTP
value and the true FTP value,providing a measure of the accuracy of the
prediction, or the confidence interval associated with the prediction. A
coefficient of multiple determination (R) is calculated which represents
the degree of scatter around the regression relative to the variance of
the dependent variable (the measured FTP emissions) observed for the test
fleet.
The regression analysis of the six steady state modes included a
multiple stepwise regression where the individual measurement modes are
added to the regression equation in an iterative or stepwise fashion. If
any one mode does not add a statistically significant amount of information
to the regression (i.e., substantially reduce the standard error of the
estimate), it is dropped from the equation.
-------
2-94
2.5.3 RESULTS OF THE CORRELATION ANALYSIS
The results of the correlation analysis are summarized in Table 2-24.
For each test procedure or individual test mode evaluated, the standard
error of the estimate and the appropriate correlation coefficient (either
the simple coefficient of correlation, r, or the multiple coefficient
of determination, R) are presented for hydrocarbons, carbon monoxide,
and oxides of nitrogen. The results are presented for only one of the
four test fleets,although the parameters are summarized for the before
service and the after service emission measurements individually. The
results for the remaining three test fleets did,not differ from those
presented here to the extent that the conclusions based upon the data
presented here would be contradicted.
In general, the results of the regression analysis indicate that the
best correlation is always achieved for the dynamic (loaded) mode tests
which provide a measure of the mass emission rate. The correlation
coefficients indicate that there is no significant advantage of the
transient mode tests over the steady state modes or the Key Mode test
for the measurement of either HC or CO. However, the transient mode tests
do appear to achieve somewhat better correlation for the NOX emission
measurements.
The same relative trends among the alternative test procedures are
observed for the volumetric emission measurements of all three pollutants.
Among all of the test measurements, the idle mode volumetric measurement
consistently achieves the lowest degree of correlation with the Federal
Test Procedure. The difference is most substantial for the measurement
of oxides of nitrogen.
-------
2-95
CM
I
CM
ca
<=c
oo
Lul
Q
UJ
o
o
cc
(X
UJ
ZD
oo
UJ
s:
o
I—I
oo
oo
UJ
O
n:
00
oo
O —i o
UJ
oo
a;
o
O Cl-
i—i oo
oo ~z.
OO I—i
Qi
CM
o
o
o
I—I
h-
o:
a:
a;
O
UJ
_l
Q
a;
o
CO
uj cu
<+- ^
o 2:
s- s-
0 0)
i- D-
s-
UJ CO
E
~O fO
S- S-
ro CS
-0
(TS
4->
OO
4->
c
cu
u
C
M-
cu
o
0
c
0
rtf
'cu
i-
o
o
X
—yr
f-~
O
o
0
^
X
o
z
o
(J
— r-
-1_
0)
S-
^s
CO
(O
cu
s:
c
o
CO
co
UJ
cu
S-
-o
cu
o
0
i_
Q_
4_>
CO
cu
h-
I—
CM
co
LO
•
CM
P"x.
•
o
p—
CO
•
o
^-
CT>
o
CO
co
(O
•SL
,
o
>}
o
4_>
S-
o
-C
oo
T—
rO
Sb.
O)
T3
cu
l_l_
t—
00
CO
,_
•
CO
o
^^.
•
o
o
p-^.
•
o
r—
CTl
C5
CO
CO
t&
s:
CU
'o
>J
o
0)
T3
~>
C
cu
>
cu
oo
CM
C\J
CO
CM
•
CM
CO
VO
*
0
^—
CO
•
o
UD
cr>
C5
CO
CO
r<3
S
f N
C
o
CO
co
CU
S-
o>
cu
S-
cu
'o.
•i —
4-)
j—
^3
P^
• — "
O)
-a
o
s:
>j
cu
CM
CM
CO
CM
•
C\J
r—
p""-.
*
O
C\J
00
*
o
CO
CTl
O
CO
CO
TJ
-l_i
00
>^
T3
(X3
CU
4_>
00
CO
CM
^~
^j-
•
*"
LT>
r-~
»
o
CM
CO
•
O
o
CO
o
CO
to
to
s:
cu
-a
O
^j>~
00
IO
CO i—
o o o o
CO CM
co O
IT)
o
(J
4->
cu
cu
o
o
cu
T3
o
cu
cu
oo
co
UD
CO
10
r— LO
co co
o
'si
cu
e
'o
>
^ — ^
c~
0
CO
co
CU
CD
CU
S-
cu
r—
Q.
£
"5
>~,
CU
-o
o
>}
cu
^
o
»l—
S-
CD
e
3
O
>•
c
o
•^*
(/>
CO
cu
s-
cn
cu
S-
•
-p
3
e
CO
cu
•o
o
cu
a
4->
oo
>>
(C
cu
4->
OO
O
£
cu
E
3
O
>•
cu
*T3
o
s:
cu
r—
-a
i — i
-------
2-96
LU
CO
UJ
u_
et
I
to
LU
QJ
t|— i—
O T-
E
O S- CD
S- QJ (_)
s- a.
-a E
S- ro
n3 S_
-a 01
c:
rO
4-1 O
•x.
CD
QJ
O
QJ
O
<_> O
S-
S-
o
<-> CJ
CXI
oo
00
UD
O
LO
UD
OO
UD
CM
O
UD
CM
UD
r— CT,
00 CM
o
CM
CM
UD
CO
CTl
CM
OO
CM
UD
O
r-.
O
^j- oo 00
o o o
co .—
co en
TABLE 2-24
(continued
NSPECTION FLEE'
1— I
LU
_I
Q
t— (
i — i
•z.
o
Lu
t_t
_l
f i
ion Measuremen
to
l/l
• f—
E
LU
to
to
ro
5:
r^
O
>^
O
QJ
^
-o
QJ
O
O
i.
Q-
1 *
to
QJ
1—
~s_
0
c~
CO
i —
ro
S-
O)
-a
QJ
LU
to
to
ro
s:
QJ
i —
O
O
QJ
-a
o
^~
£Z
QJ
>
QJ
CO
to
to
rO
s:
, — ^
c
o
., —
to
to
QJ
Oi
QJ
S-
QJ
f-~
a.
•i—
-t-*
^
^
- —
QJ
O
^~
>^
QJ
:*i
to
1/1
ro
^
c:
o
•i —
to
to
QJ
i_
en
QJ
S-
•
-i->
rs
E
to
QJ
~O
O
s:
QJ
4-)
ro
+J
CO
>^
-a
ro
QJ
4.)
CO
l/l
to
ro
s:
QJ
"O
O
2:
QJ
i —
-a
i — i
o
il
QJ
E
=3
i —
O
>
QJ
i —
O
C_3
QJ
T3
O
21
C
QJ
>
QJ
CO
O
•1 —
S-
QJ
E
=3
i —
O
>
*- — -»
c
0
•1 —
to
to
QJ
cn
QJ
s_
QJ
i — .
Q-
•i —
-P
Z5
__£_
QJ
T3
O
s:
>Sj
QJ
ii
u
s_
QJ
E
13
i —
O
>
E
o
• 1 —
to
to
QJ
S-
a>
OJ
S-
•
4->
3
E
to
QJ
•^
O
S
QJ
4_)
rO
CO
>^
T3
QJ
4-3
CO
O
si
QJ
E
13
i —
O
>
QJ
T3
O
21
QJ
i —
-o
i— <
-------
2-97
In general, the correlation coefficients are lower for all test
procedures after service. This is not unexpected, however, since the
effect of maintenance tends to reduce the variance, or the dispersion,
of the fleet emission levels. The standard error for CO and NOx
measurements is not significantly changed following service; whereas, the
standard error of the HC measurements is reduced for all test procedures
except the Federal Short Cycle and the idle mode mass emission
measurement.
The results of the multiple stepwise regression are summarized in
Table 2-25. The coefficients of multiple determination are shown for each
multivariate regression using various combinations of the six steady state
modes and the idle mode (0 mph) with volumetric measurements. The individual test
modes are entered into the regression equation such that the maximum improvement in
the correlation is achieved for each iteration of the regression. The simple
coefficients of correlation are also shown for each test mode independently.
The multivariate regressions indicate only a marginal improvement of the
information content by adding more than two or three operating modes to the
regression equation. The same trends are observed for both HC and CO measurements.
With respect to an inspection program, the additional time and cost incurred by
including emission measurements in all seven modes would not be justified.
The single modes demonstrating the best correlation with the FTP are
consistently observed to be loaded cruise modes; the simple r values being
substantially higher than the idle mode ( 0 mph ) r value. With respect
to both HC and CO values, the correlation coefficient for the best loaded mode
is around 0.7 as opposed to 0.4 to 0.5 for the idle mode. Better correlation
is achieved for HC measurements than for CO measurements.
-------
2-98
LO
CM
1
CM
z:
o
oo
oo
LU
CJ3
LU
Qi
LU
_J
Q-
h-
rD
s:
i , i
oo
1— i
s
0-
LU
1—
OO
OO
1 1 t
O
s*.
UJ
& LU
^^ LLJ
1 — c_j>
OO •"—*
>
Q LU
I—
_J LU
O LU
> _l
LU
•
S-
cu
Q.
E
OO
OJ
-O
OO O
i— s:
LU
O3 O
CO LO
LO oo
CM O
LO 10
<£> LO
CO
o
O
oo
Q.
Q-
O O O O
CM Rt- i—
Q. Q.
E E
O O
r- CO
o.
E
o
CM
o.
o
ID
oo
CQ
o:
LU ro LU
|— Q Q.
eC LU CO
1—02:
oo o >->
>- Q- LU
Q _1
LU OO •— <
1— LU
OO |—
c~
LU
CO
cu
!__
^(^
rd
i_
fO
>•
cu
'o.
•i —
-(->
i —
3
to
c
co
• 1 —
CO
>,
^~
fX3
C
<
c
o
«i —
CO
cu
s-
Ol
r-~
^
^:
Q-
E
0
CO
f,
0
r—
f>
0
^"
p^
r^.
to
-C
a.
E
O
CM
•t
0
CO
rv
o
I—
A
0
^3"
OO
r-~
to
-C
Q.
E
O
Lp
ft
o
LO
rt
o
^
o
CO
n
o
CM
rt
o
r-~
A
0
-------
2-99
2.5.4 CONSIDERATIONS OF THE NEED FOR CORRELATION
The results observed in the Short Cycle Study indicate that an idle
mode emission inspection and a loaded mode emission inspection would be
equally effective for an in-use vehicle inspection program. However, the
poor correlation between the idle emission measurements and the FTP raises
concern regarding the continued effectiveness of an idle inspection program
for a number of reasons.
• The sizeable emission reductions observed in the idle inspection
fleets appear to be due, in large part, to the prevalence of
excessive maintenance. It is expected that competitive and consumer
pressure will eventually constrain the maintenance procedures such
that only repairs and adjustments necessary to achieve compliance
with the idle test will be performed.
• Over time, the mechanics and vehicle owners are likely to learn
how to "beat" the idle test by inducing certain adjustments which
lower idle concentration measurements and which have no effect, or
perhaps increase, true mass (FTP) emissions.
• Further evaluation of alternative inspection approaches must
consider the capability of the specific test procedures for testing
the advanced control systems which may be used in future model
year vehicles. There is little assurance that the idle test will
be adequate for evaluating the emission control performance of
new motor vehicles (post-1973). As these model year groups
comprise an increasing fraction of the vehicle population, the
effectiveness of an idle inspection program will be reduced
accordingly.
-------
2-100
Effectiveness for Current In-Use Vehicles
The low level of maintenance necessary to comply with the idle
mode emission test is not expected to achieve the sizeable reductions
observed in the Short Cycle Study. The discrepancy between the
expected effect and the observed results may be explained by the
tendency of the mechanics participating in the study to perform a
substantial amount of excessive repair.
The Short Cycle Study was designed to evaluate the capability
of the two inspection procedures in screening vehicles with excessive
emissions attributable to engine malfunctions. To approach the
actual conditions which would exist immediately following
implementation, a sample of private garages was selected from the
present service industry to perform the reguired maintenance on the
"failed" vehicles. The mechanics were provided with instruction
in emission diagnostic procedures only to the extent that could
be expected prior to the implementation of a real State program.
A detailed analysis of the repair actions revealed that the mechanics
consistently resorted to conventional repair procedures rather than
utilizing the diagnostic content of the emission test results
provided with each vehicle they received.
Assuming that the test fleets were statistically valid samples
of the vehicle population, the fleet emission distribution would be
similar to the distribution of elapsed time since the last conventional
maintenance or tune-up, i.e. low emissions would correlate with recent
maintenance. Therefore, the use of any two screening tests which
reflect the elapsed time without maintenance and which are designed
to fail a large fraction of the test fleet should identify two
-------
2-101
equivalent vehicle sub-sets with respect to the distribution of both
emissions and malfunctions. This phenomenon appears to have occurred
in the Short Cycle Study; both the idle mode test and the cruise mode
test failed statistically equivalent sub-sets within each of the
independent test fleets. Since the sub-sets of failed vehicles were
identical between the two fleets, and the repair agencies utilized
identical maintenance techniques, equivalent emission reductions would
have to be expected.
Although the mechanics were not informed of the specific purpose
of the study, they were coginzant of some type of performance
evaluation. Knowing that the program in some way was evaluatina
the reduction of motor vehicle pollution which could be achieved by
proper repair and adjustment of the automobile, the mechanics would
naturally try to demonstrate the importance of requrinq more frequent
maintenance. Therefore, their effort to provide any and all maintenance,
which in their judgement was necessary, is understandable. This
tendency was further enhanced by the design of the test program which
did not require any direct interface between the repair agency and
the vehicle owner.
Irrespective of the Short Cycle Study results, it is reasonable
to assume that the competitive and consumer pressure that would
be present in a real program would tend to eliminate this "shotgun"
maintenance approach and eventually support only repair and
adjustment of those components necessary to comply with the inspection
test standards.
-------
2-102
At present, the empirical data base is insufficient to define
the level of effectiveness which can be expected as "shotgun" maintenance
procedures are gradually eliminated. However, the regression equations
derived from the results of the Short Cycle Study provide a means
for estimating the resultant mass emission reduction that may be achieved
following the implementation of an idle inspection program. The
methodology involves utilizing the idle mode regression equations derived
from the fleet emission levels before and after service to predict
fleet mean emission levels as a function of the observed idle mode
emission levels before and after service. The absolute difference between
the before-service and after-service predicted emission levels is divided
by the measured as-received mass (FTP) emission level to obtain an
index of effectiveness. This analytical approach relies upon the
following assumptions:
• The change in idle mode emissions observed in the Short
Cycle Study will be achieved by an idle inspection program.
• The mechanics will not perform "shotgun" maintenance, but will
perform repairs and adjustments necessary to comply with the
idle mode emission test.
' The repair actions taken by the mechanics will not be those
which "beat" the idle test.
The predicted mass emission reductions and the observed mass
emission reductions based on the Short Cycle Study results are
summarized in Table 2-26 for each of the four independent test fleets.
The average reduction for the four fleets indicates that the relative
-------
2-103
TABLE 2-26
COMPARISON OF OBSERVED FLEET EMISSION REDUCTION^-/
and
PREDICTED EMISSION REDUCTION USING IDLE REGRESSION EQUATION
Hydrocarbons Carbon Monoxide
Before After Before After
Service Service Service Service
California Fleet I
Observed Level
Predicted Level
Observed Change
Predicted Change
California Fleet II
Observed Level
Predicted Level
Observed Change
Predicted Change
Michigan Fleet I
Observed Level
Predicted Level
Observed Change
Predicted Change
Michigan Fleet II
Observed Level
Predicted Level
Observed Change
Predicted Change
I = idle fleets (150 vehicles each)
II = key mode fleets (150 vehicles each)
7
6
8
7
8
8
7
7
.46
.97
.18
.08
.52
.53
.69
.09
22%
6.7%
34%
12%
21%
20%
22%
10%
5.82
6.47
5.43
6.10
6.69
6.88
6.00
6.31
89
87
94
87
97
94
94
89
.88
.56
.81
.14
.73
.88
.8
.52
1
8
24%
12%
19%
15%
18%
11.
76
80
5%
.3%
71
75
78
80
77
78
5%
.31
.13
.78
.28
.95
.56
.39
.60
Total Relative Effectiveness
HC = 12.2 =.49
2O"
CO = 12 =.63
I/ 1972 Federal Test Procedure (CVS-C)
-------
2-104
effectiveness of the idle mode inspection approach would be 0.49 for
HC and 0.63 for CO; i.e., the magnitude of the HC emission reduction
would be approximately one-half that achieved by a loaded mode inspection
program.
As stated previously, the above analysis assumes the inspection
procedures prevent the possibility of "beating" the idle mode test.
Certain idle adjustment parameters are known to effect a reduction of
idle concentration levels while concurrently increasing mass emission
levels in other operating modes. Accordingly, many adjustments made
to achieve compliance with an idle inspection may, in fact, have a negative
effect on air quality.
To evaluate the effect on emission levels induced by various engine
adjustments and component malfunctions, a controlled study was carried
out in support of the APRAC/CAPE-13 Project.9 Emissions response
coefficients were calculated for specific engine parameters. These
response coefficients represent the unit emission change per unit
parameter change. The parameter change may result from an adjustment
or from mechanical deterioration over time. The coefficients are
shown in Table 2-27 for idle mode concentration measurements, cruise mode
:oncentration measurements, and FTP mass measurements.
Any parameter change causing an idle emission response which is
significantly different (in sign or magnitude) from the mass emission
response is capable of being used to "beat" the idle test. Table 2-28
places these effects in perspective for some typical engine adjustments
and malfunctions. The total effects and the percent change for each
case are relative to the tuned emission level for the 1971 California
vehicle.
-------
2-105
Parameter
% Idle CO
Idle rpm
Basic Timing
Misfire +
NOx Device
PCV Valve
Air Cleaner
Parameter
% Idle CO
Idle rpm
Basic Timing
Misfire
NOx Device
PCV Valve
Air Cleaner
Parameter
% Idle CO
Idle rpm
Basic Timing
Misfire
NOx Device
PCV Valve
Air Cleaner
TABLE 2-17
EMISSION RESPONSE COEFFICIENTS
— Hydrocarbon Response —
Idle Mode
Concentration
(ppm)
50 mph Cruise
Mode Concentration
(ppm)
13
11
137
105
-2
.43
.34
.55
.60
.80
.97
4
128
11
-6
-.09
.60
.02
.05
.00
.20
.85
.09
— Carbon Monoxide Response —
Idle Mode
Concentration
(Percent)
1.03
00
00
.03
-.01
00
50 mph Cruise Mode
Mode Concentration
(Percent)
.01
00
-.01
-.02
-.14
00
— Oxides of Nitrogen Response —
Idle Mode
Concentration
(ppm)
-38.3
.39
.43
40.00
-3.94
-.04
50 mph Cruise Mode
Concentration
(ppm)
-1 .8
-.04
91 .62
313.6
53.78
-.70
FTP
Mass
(gpm)
.10
00
.10
1.80
.60
-.18
00
FTP
Mass
(gpm)
11.
-ll
-7,
-9,
43
08
35
30
75
11
FTP
Mass
(gpm)
-.03
00
.16
.72
.13
-.003
— Tuned Emission Level for 1971 California Vehicles —
1972 CVS
Loaded Cruise
Idle Mode
HC
4.36gpm
125ppm
190ppm
CO
59.2gpm
0.83%
2.1
Change in emissions per unit parameter change (Delta E/Delta P)
Representative of 1971 California vehicles.
Data from APRAC-CAPE-13 Project, TRW and Scott Laboratories (Ref.
Ignition timing modifications.
9)
NOx
5.88gpm
2520ppm
177ppm
-------
2-105
00
CM
CM
CO
OO
z.
o
HH
1—
O
Lu
_J
^^
s:
OL
O
Q
Z.
oo
1 —
LU
oo
;=j
Q
i
__1
<
0
i — i
Q-
>-
1—
O
1—
LU
Q
LU
OO
z.
0
Q.
00
LU
Z.
o
1 — 1
00
00
1 — 1
LU
&g.
Lu en
oo
LU
Q
i — i CU
X CO
o ••-
Z2&
LU C_3 Q.
00 • 1 IO &«
r~- co i • i^
• r— .— CM
11 II
\p i^ \p
l£) CO 1 O IjD
«* i — 1 CTl CO
II II
CJ3
O
o:
i— i CU
2^i.
I-H Q-
IO CM 1 *d~ i —
in co i t^. ^3-
Q- E
I— Q.
LU Lu en
Q
CO CO
CD-I 6^
• un i o i—
en i — co LO
X
O CU
Z. CO
O -r- &5
i-
z. <->
LO &§
O 5-S 1 i — CM
. U3 i . , —
"T~ + + T
o
CO
o; cu
O T3
I— i S^
CM 65
LO co i o o
1 1
>-p
Q- E
1- °-
Lu C7^
IO
LO • co o s«
LO CM • CO CO
i" T T +" "+
OO CU
Z. CO
o •.-
CO 3 E
Q: s- Q-
«=C CO O-
CO
CM OO CM •=!- ?-~
CM i — i — CO l£>
II I 1 1
1 T^ T T-
o
o
ce:
Q
>- CU
-T- _,. ^
^ i.
M O.
co
• ^a co &s
CO CM jjC1^ IO "d-
cn LO T-I — 'd- CM
1 I g+ II
E
O CO
O 3 cn
I-H 'i-j CU
cn^^ ^3 "O
O) LO =£
-o • o o
O O oC O i —
OJ
en
c
S- 03
0 -C
CJ)
1 >
C -(->
CU C
E CU
+-> c
CO O
3 Q-
••-) E
T3 O
eC C_5
O LO 1 CM 1
i— i c: cu +
•4->+OLU OS- CDLU
c~ rr>f ^ cXi '*~ 'r- E C CD
CU E C 1— 'Z. +-> <4- O-T- 1— Z.
E Q-'i- CUCJci OCO S-ECJet
+-> s- E r— LU 3: c-i- -^Lun:
CO -r--OLuO 321 CU 1— Lu (_>
3CUI — •— iLu 4- "CO Lu
•r-jcO LU| — i — i — Q.n3-OLU| —
•O03T3CU Z. (O033CUS- "Z^
eCCUS_O_ILU SI -r- 1 S-03—1LU
S-033CCO +JCUO+-)=CC_)
— ~ o +-> -o I— c£ ' — s- E c CD i— o:
O3CCUCUOLU ^30331— iQiOLU
-l-fQiQil— Q. Q- 1— 1— CL,
co s^
• LO
1 1
O 5-9
r-. co
t 1
«3- LO
«=t CM
r— CTl
T T
CTl S«
CM LO
• CO
"> I
1 ^
O CO
+' +'
CD CD
CTi r^
' '
_
^ T^
CU ' '
T3
o
0
^~
CU — -
en
03 S-
^ CU
o c
o z.
r— =t
to s- nr
+J e£
S- h-
ro 4- H- z:
Q- O O LU
LU O
^-^ u_ o:
O Lu LU
^_-^ 1 1 i O-
-------
2-107
Case "a" shows the resultant impact on emission levels for a typical
combination of parameter adjustments. The most significant effect to
be observed here is the discrepancy between the idle mode CO measurement
and the FTP CO measurement. Note that the cruise mode measurement of
HC and CO is reasonably representative of the FTP measurement.
Case "b" represents a hypothetical situation to demonstrate
the ease with which the idle test can be "beaten." A partial
misfire is shown to cause excessive idle HC emissions. Rather than
repairing the ignition system failure, the idle mode emissions may be
reduced by adjustment of idle rpm and basic timing. The true effect,
however, as measured by the FTP, would be 15% and 14% increases of
HC and CO respectively.
The diagnostic potential of a leaded mode test is exemplified in
case "c." The nominal malfunction is the partial blockage of the air
cleaner. The idle mode CO emission response is insufficient to warrant
failure of the vehicle; thus the malfunction would remain undetected.
Alternatively, the loaded mode CO measurement does indicate an induction
system malfunction and thereby provides adequate justification for
failure.
-------
2-108
Jtility of Emission Test Procedures for Inspection of Future Motor Vehicles
Until the operational characteristics of emission control systems
which will be employed in future model year vehicles are firmly defined,
it is not possible to determine with certainty the adequacy of various
emission test procedures in identifying malfunctions of those systems; the
the relative importance of identifying various types of malfunctions
cannot be determined until operating experience with substantial numbers
of those vehicles has been gained. However, some general conclusions
can be drawn based upon the general characteristics of various test
procedures.
Because it is considered most representative of typical light
duty vehicle operation in urban areas, the Federal Certification Test
Procedure (FTP) is the standard for measuring vehicle emissions. The
limited number of vehicle operating conditions included in an idle mode
emission test, as compared with the FTP, makes it improbable that this
inspection approach will be useful for detecting certain types of
malfunctions in future motor vehicles.
The potential shortcomings of the idle mode test for future control
systems are exemplified by its inability to diagnose malfunctions of
exhaust gas recirculation (EGR) systems which are currently being used
by most automobile manufacturers to ensure compliance with the 1973 Federal
NOx emission standards. When the EGR valve is functioning
properly, there is no recirculation of the exhaust gas during idle
operation; therefore, the system provides no reduction of idle NOx
emissions. A malfunction of the EGR system causing an increase in NOX
-------
2-109
emissions during loaded operating modes would not result in a concurrent
increase in idle mode emissions. The malfunction would,therefore,
remain undetected by an idle test measurement.
A loaded emission test, on the other hand, includes a wider
range of operating conditions and should be more generally useful in
testing future vehicles, although all current short emission tests are
hampered by their inability to measure cold-start emissions, which will
become increasingly important for vehicles equipped with catalytic
and thermal reactor emission control systems.
The same type of considerations would also apply to the choice
of emission inspection procedures to be used for vehicles retrofitted
with exhaust emission control systems, which must receive periodic
inspection and maintenance to ensure the effectiveness of the retrofit
emission controls (see Chapter 3).
Additional Considerations
The evaluation of alternative inspection procedures must also
consider their relationship to enforcing the warranty provisions set forth
in Section 207 of the Clean Air Act. That section authorizes the EPA to
establish regulations requiring automobile manufacturers to warrant the
.emission control performance of every new motor vehicle for the vehicle's
useful life. To implement this provision, Section 207 requires that there
be available short test procedures which achieve adequate correlation
with the FTP. While the definition of adequate correlation is yet to be
established, it is clear that those short tests which achieve the
highest degree of correlation will be most likely to satisfy the require-
ments for adequate correlation. The correlation analyses have
-------
2-110
CO
H
CO
§
HI
H
M
CO H
K &>
O W
M C_>
CO
en i-J
W
Q
-------
2-111
I
Cxi
W
w
u
u
H
CO
w
H
O
M
H
* 3
M
CO PH
O H
M P4
CO W
CO O
W W
n Q
M W
X fu
o
3 to
O D
a w
a
M
< a
o u
>^
u
CO
13
w
Q
O
o
a.
H
Z
W
W
Pi
CO
<
W
O
M
H
-------
2-112
consistently shown that for current vehicles the dynamic (loaded) tests,
as a general category, achieve significantly higher correlation with the
FTP than do the idle mode tests.
The States are not required to consider the feasibility of enforcing
the warranty provisions in the design of their transportation control
plans, and the warranty requirement (Section 207(b)) would not apply to
current vehicles in any case. However, any enforcement program which
imposes a burden of responsibility upon the private citizen, should
also provide adequate protection for the vehicle owner to ensure that
the burden of non-compliance is only placed upon those who are truly
liable. Accordingly, the enforcement of the warranty provision may
directly affect the public acceptability of any enforced in-use vehicle
inspection program.
The selection of an individual inspection test requires the
development of criteria for determining what degree of correlation is
adequate to satisfy the warranty provisions. The following analysis
provides a qualitative means of making such a determination.
For illustrative purposes, it is assumed that the points marked "a"
in Figures 2-13 and 2-14 represent the Federal emission standard for all
the vehicles in the sample fleet. The points marked "b", "c", "d", and
"e", represent hypothetical cutpoints for a State inspection program.
A higher outpoint results in a lower rejection rate, and thereby reduces
the fleet emission reduction potential of the program. Any vehicle
which is above the inspection cutpoint, and is to the left of point "a", is
defined as an error of comission. These vehicles are erroneously identified
as excessive emitters. Any vehicle which fails the inspection criteria
and is to the right of point "a" is a valid failure.
-------
2-113
The feasibility of enforcing the warranty will be determined by the
frequency of comission errors among the vehicles which fail the short
test. The probability of a commission error can be reduced by
raising the inspection test failure criteria. At any cutpoint, a
commission error is still defined as any failed vehicle to the left of
point "a." Therefore, a tradeoff exists between the feasibility of
enforcing the warranty and the fleet emission reduction achieved by the
inspection strategy. The degree of correlation between the two test
procedures is a measure of the extent to which the short test failure
criteria must be raised to reduce the errors of commission to an
acceptable level.
Table 2-29 presents the results of applying this type of analysis
for the Federal Short Test procedure and the Idle Mode test procedure.
The rejection rate, the frequency of commission errors, and the fleet
emission reductions are shown for selected short test cutpoints.
The results of this analysis are not intended to provide sufficient
information to determine the failure criteria which should be used in a
State program. The test fleet used to demonstrate this analysis was
composed of the total model year mix in the present California vehicle
population. The individual failure criteria would have to be determined
for each model year such that the commission errors were reduced to an
acceptable level. However, Table 2-29 does demonstrate the impact of
the tradeoff between commission errors and the fleet emission reduction
potential for two levels of correlation.
-------
2-114
TABLE 2-29
ERRORS OF COMMISSION FOR TWO REPRESENTATIVE SHORT TEST PROCEDURES
Test Type
Federal Short Cycle
(Corr Coef = 0.721)
Idle Mode Test
(Corr. Coef. = 0.375)
Frequency of
Commission Errors
Fleet Emission Reduction
Rejection
Rate
50%
40%
30%
20?;
10%
50%
40%
30%
20%
10%
5%
% of Failed
Vehicles
32%
22%
18%
10%
7%
43%
40%
30%
30%
27%
14%
After Maintenance
(CO Emissions)
17%
15%
14%
8%
5%
17%
15%
10%
12%
8%
4%
-------
2-115
It appears that the Federal Short Cycle, which exhibits a correlation
coefficient of 0.721, can achieve a significant emission reduction while
maintaining only a 10% to 15% chance of failing a low-emitting vehicle.
Alternatively, the probability of a commission error is greater than 20%
for Idle Mode failure criteria which only achieve a rejection rate of 10%.
Conclusions
The results of the Short Cycle Study comparing the idle emission
inspection procedure and a loaded cruise mode emission inspection
procedure are somewhat misleading and may indicate erroneous conclusions
regarding the relative effectiveness of the two inspection approaches.
Although the idle inspection fleets and the cruise mode inspection fleets
achieved statistically equivalent emission reductions, the correlation
studies support a number of reasons for questioning the effectiveness of
the idle inspection approach:
1) The low level of maintenance required to comply with the idle mode
emission test will result in substantially smaller reductions than
those expected for a loaded mode test.
2) The tendancy to perform engine adjustments which "beat" the idle mode
test rather than effecting proper repair of malfunctioning components
will degrade the reductions achieved by an idle inspection program.
3) The idle mode emission test is least likely to be of utility for the
evaluation of the emission control performance of future motor vehicles,
-------
2-116
2.6 CONCLUSIONS
This chapter has discussed, based upon currently available data, the
feasibility, emission reduction effectiveness, and cost of inspection/
maintenance approaches to reducing emissions from in-use vehicles. The
objective of this chapter has been to provide estimates of the effectiveness
and cost of such programs which will be useful to the States in the initial
evaluation and planning of motor vehicle inspection/maintenance programs.
Prior to, and during, the implementation of such programs, additional
investigation and evaluation of a number of factors will be necessary to
ensure that the extensive resources which may be committed to an inspection/
maintenance program will provide an adequate return in terms of emission
reductions.
The orinciple conclusions which can be drawn at this time reaardina
inspection/maintenance approaches are summarized in the following
paragraphs.
Effectiveness
* Significant reductions in the overall emissions of HC and CO from the
current light duty vehicle population can be achieved through requiring
additional maintenance beyond that which the vehicles now receive voluntarily.
Major changes in NOX emission levels from such vehicles are not anticipated as a
result of providing the additional maintenance. At any one time, significant
reductions in the emission levels of approximately 50% of the current in-use
light duty vehicle population could be achieved through a requirement for
additional maintenance.
* The inspection and maintenance of sample fleets has achieved initial emission
reductions of up to 25% in exhaust HC and up to 19% in CO averaged over the entire
fleet. No statistically significant changes in the average emissions of NOx were
found.
-------
2-117
* Deterioration of vehicle emission control performance between periodic
inspection/maintenance events will cause the average emission reductions
achieved by an inspection/maintenance program over time to be less than the
initial reductions observed for the test fleets. A preliminary estimate,
based upon an analysis of the current frequency and distribution of voluntary
maintenance of emission-related components, indicates that inspection/
maintenance on an annual basis may be expected to achieve average reductions
over time of up to 12% in light duty vehicle exhaust HC emissions and up to
10% in light duty vehicle CO emissions. Larger reductions could be achieved
with more frequent inspection/maintenance.
* Mean initial emission reductions observed for emission inspection
approaches are larger than those observed for engine parameter inspection or
mandatory maintenance approaches; however, current data do not permit the
relative effectiveness of these approaches to be differentiated at a 90%
confidence level.
Cost
* Annual emission inspection in State-operated lanes using a short, loaded-
mode test procedure is estimated to cost approximately $2 per vehicle,
including amortization of capital investment and operating costs. Extensive
engine parameter inspection utilized in a licensed private garage system
is estimated to cost approximately $8 per vehicle.
* Repair costs observed in fleet studies of inspection/maintenance approaches
indicate that an average repair cost of $20 to $30 is typical for vehicles
failing an emission or engine parameter inspection, although an extensive
mandatory maintenance requirement could cost up to $60 per vehicle.
-------
2-118
* The true net cost of maintenance required by an inspection/maintenance
program will depend upon the number of vehicles failing the inspection and
the extent to which maintenance required by the program is substituted for
maintenance which would normally have been performed voluntarily. It has
been estimated that, averaged over the entire light duty vehicle population,
the additional cost of maintenance required by an annual inspection/maintenance
program in which 30% of the vehicles failed inspection would be about $2 per
vehicle.
Comparison of Inspection/Maintenance Approaches
* To the extent that their results relate adequately to emission
levels produced during normal urban driving, emission inspection approaches
are generally applicable to all in-use light duty vehicles since they directly
identify those vehicles having high emission levels.
* Among vehicle fleets tested to date, the frequent coincidence of
malfunctions or maladjustments causing high idle mode emissions with
malfunctions causing high emission levels under load has resulted in idle
emission testing beinq nearly equivalent to loaded mode emission testing
in identifying high emitting vehicles for maintenance. However, the
generally poor correlation between idle mode emission measurement and emission
levels during typical urban driving indicates that, in practice, some vehicles
may be brought into compliance with idle emission inspection standards without
significantly reducing their true emission contributions. Correlation with
emission levels during typical urban driving is substantially better for
loaded mode emission tests.
-------
2-119
* Engine parameter inspection and mandatory maintenance approaches
depend upon the identification of a relationship between various
specific mechanical malfunctions or maladjustments and excessive emission
levels. This requires that such approaches be designed for specific model
year groups of vehicles, giving consideration to their design and based
upon data on the frequency and impact on emissions of various types of
malfunctions.
-------
2-120
REFERENCES - CHAPTER 2
1. Federal Register, Volume 35, Number 219, Part II, November 10, 1970,
"Control of Air Pollution from New Motor Vehicles and New Motor
Vehicle Engines."
2. Cline, E.L. and Tinkbam, L., "A Realistic Vehicle Emission Inspection
System, " June 1968, Clayton Manufacturing Company, El Monte, California,
3. Federal Register, Volume 31, Number 61, Part II, March 30, 1966, "Con-
trol of Air Pollution from New Motor Vehicles and New Motor Vehicle
Engines."
4. Federal Register, Volume 33, Number 108, Part II, June 4, 1968, "Con-
trol of Air Pollution from New Motor Vehicles and New Motor Vehicle
Engines."
5. TRW Systems Group, "The Economic Effectiveness of Mandatory Engine
Maintenance for Reducing Vehicle Exhaust Emissions," in support of
APRAC Project Number CAPE-13, One Space Park, Redondo Beach, California;
preliminary final report, July 1972.
6. Federal Register, Volume 36, Number 128, Part II, July 2, 1972, "Con-
trol of Air Pollution from New Motor Vehicles and New Motor Vehicle
Engines."
7. Olson Laboratories, study performed in support of EPA contract number
68-01-0410; contractor's report unpublished as of November 1972.
8. Northrop Corporation, "Mandatory Vehicle Emission Inspection and
Maintenance, Volumbe III, prepared under contract ARB 1522 with the
State of California Air Resources Board, May 1971, Anaheim, California.
9. TRW Systems Group/Scott Research Laboratories, studies performed in
support of APRAC project number CAPE -13-68
10. In-House Evaluation; MSPCP, OAWP, EPA; 2565 Plymouth Road, Ann Arbor,
Michigan.
-------
2-121
11. Federal Register, Volume 36, Number 158, Part II, August 14, 1971,
"Regulations for the preparation, Adoption and Submittal of Implementa-
tion Plans."
12. Clean Air Act (42 U.S.C. 1857 et seq.), Amendments of 1970 (P.L. 91-
604); December 31, 1970.
13. Champion Spark Plug Company; "American and Foreign Car Survey of the
Domestic Market"; Toledo, Ohio; March 1972. Additional data supplied
by Champion Spark Plug Company.
14. Brubacher, M.L. and Olson, D.R.: "Smong Tune-up for Older Cars"; SAE paper
S403; April 1964.
-------
3-1
Chapter 3 RETROFIT OF EMISSION CONTROL TO IN-USE VEHICLES
3.1 INTRODUCTION
Continuing advances in motor vehicle emission control technology
suggest that retrofitting more effective emission control systems to
in-use vehicles may be useful to accelerate reductions in vehicular
emissions. This chapter discusses, in the light of recent EPA evaluations
of retrofit emission control technology, the feasibility, emission reduction
effectiveness, and costs of retrofit approaches to in-use vehicle emission
control.
DEFINITION OF RETROFIT
A retrofit approach can be defined as the addition of any device or
system and/or any modification or adjustment, beyond that of regular main-
tenance, which is made to a motor vehicle after its initial manufacture
to reduce emissions. Emission control programs based upon periodic
maintenance to restore vehicles to original design specifications are
not considered retrofit approaches and are discussed in Chapter 2 of
this document. Included in this definition of retrofit would be the
conversion of in-use vehicles for operation on gaseous fuels such as
propane or natural gas. However, since the evaluation of strategies using
alternative fuels requires the consideration of a number of factors not
involved in other retrofit approaches, gaseous fuel conversion is mentioned
only briefly in this chapter and is discussed in more detail in Chapter 4.
FACTORS WHICH AFFECT THE USEFULNESS OF A RETROFIT APPROACH
The major factors which determine the usefulness of a given retrofit
approach can be categorized as follows:
* Applicability to vehicle population
* Emission control potential
-------
3-2
* Installation requirements
* Reliability and inspection/maintenance requirements
* Cost
* Acceptability of safety and driveability effects
The potential usefulness of a general retrofit approach (e.g., oxidizing
catalytic converter) can be estimated by evaluating representative retrofit
systems or devices incorporating that approach according to these criteria.
However, before a retrofit program is implemented, extensive testing and
evaluation of the specific retrofit systems or devices which will be
installed must be performed to ensure that the anticipated emission reductions
will be obtained^and that use of the devices will result in no unacceptable
effects.
In evaluating a retrofit system, a factor of primary importance is the
extent to which its use can reduce emissions. First, the applicability
of the retrofit system must be determined. This includes determining the
different classes of vehicles (classified according to vehicle weight,
engine type, and the types of emission control systems which may already
be installed on the vehicles) to which the retrofit system is applicable.
The extent of the vehicle population within each of those classes to which
the retrofit system can be fitted (e.g., is the retrofit applicable only
to specific makes or models) must also be determined. It is desirable
that a retrofit system be flexibly designed so as to allow application to
a large segment of the vehicle population. Then, the ability of the retrofit
system to control each of the three regulated automotive pollutants
(hydrocarbons, carbon monoxide, and nitrogen oxides) must be determined; it
is also important that the retrofit system not result in emissions of any
additional harmful pollutants. The control potential of the system must
-------
3-3
be established for each different class of vehicles to which it may be
applied.
Before a specific system is selected for widespread use, the factors
influencing its installation must be evaluated. These include availability
of an adequate supply of retrofit units, of facilities equipped to install
the system, and of manpower capable of competently installing the system.
Factors which may affect the continued effectiveness of the system in
use must also be considered. The reliability of the system must be established
and the need for periodic inspection and maintenance of the retrofit system
determined. In addition, the need for normal maintenance, such as periodic
tune-up, of vehicle systems not part of the retrofit shoudl be considered.
To the consumer, the cost is the sum of three separate terms:
initial installed cost, recurring maintenance costs, and associated costs,
such as fuel consumption penalties. The cost of the retrofit program implemen-
tation to the local jurisdiction for administration and for facilities
and manpower needed for inspection is an additional cost factor.
The final area of concern is the retrofit system's acceptability.
This analysis involves safety considerations and adverse driveability
effects. Possible decreased vehicle life resulting from installation of a
retrofit system could be a major consideration in determining system
acceptability.
USEFULNESS OF RETROFIT PROGRAMS AS IN-USE VEHICLE EMISSION CONTROL STRATEGIES
Beyond the feasibility, effectiveness, and cost of using retrofit
approaches to reduce emissions from individual in-use vehicles, the overall
cost and emission reductions resulting from implementation of programs for
installing those devices must be evaluated. In addition to the effectiveness
and cost of the devices, this depends upon the number of vehicles to be
retrofitted and the magnitude of their Collective emissions relative to
-------
3-4
other sources. These factors vary substantially between air quality
control regions and with time; so a thorough analysis of the usefulness
of retrofit strategies must be made on a region-specific basis. Therefore,
the principal concern of this chapter is with the technical feasibility of
applying various retrofit approaches to in-use vehicles, with the cost of
typical devices, and the emission reductions per vehicle which can be obtained.
As a qualitative example, however, some calculations of overall cost and
effectiveness of retrofit strategies based upon the average nationwide
Vehicle distribution are presented in Section 3.4 of this chapter.
STATUS OF RETROFIT TECHNOLOGY EVALUATIONS
Potentially, retrofitting of emission controls could be considered for
in-use light duty and heavy duty vehicles, both gasoline and diesel-powered.
However, except for evaluation of certain techniques for reducing smoke
and odor emissions from diesel-powered buses , EPA evaluations of retrofit
emission control have to date considered only gasoline-powered light duty
vehicles. The principal reason for this emphasis has been the predominant
contribution of that class of vehicles to vehicular emissions in most
major urban areas. As a result, while retrofit approaches may be applicable
to certain classes of heavy duty vehicles, empirical data on the emission
reductions attainable through such modifications are not available at
this time. Information is also lacking on the application of retrofit
approaches to light duty diesel-powered vehicles; but the extremely small
number and generally low emission levels of such vehicles makes them an
unlikely candidate for a cost-effective retrofit strategy.
The Environmental Protection Agency has completed two major studies
directed toward evaluating the feasibility and emission reduction potential
of various retrofit approaches. The approaches have been evaluated and
tested specifically in terms of their applicability to light duty vehicles
-------
3-5
not originally equipped with emission control systems. However, some of
the results may be useful in judging the applicability of those approaches
to controlled vehicles as well. One of these studies was designed as an
overall survey of potential retrofit approaches and employed no extensive
vehicle fleet testing, although limited testing of several retrofit systems
was performed. This study provides a comprehensive assessment of the types
of technology currently being applied in developing retrofit devices,as
well as a limited evaluation of the range of emission reductions that can
be obtained by applying the more promising of those approaches to pre-con-
trolled light duty vehicles. The major findings of this study are discussed
in Section 3.2.
The second study was a more extensive evaluation of one relatively
inexpensive retrofit device. This study involved the evaluation of the
retrofit system through testing it on a fleet of over one hundred pre-controlled
automobiles. The major results of that study are discussed in Section 3.3.
-------
3-6
3.2 SURVEY OF RETROFIT APPROACHES
3.2.1 DESCRIPTION OF PROGRAM
Through a contract, the Environmental Protection Agency has conducted
a survey of currently available retrofit emission control approaches for
in-use light duty vehicles. The purpose of the survey was to identify
which approaches are feasible for retrofit emission control and to provide
a preliminary evaluation of their effectiveness and cost in reducing emissions
from pre-controlled automobiles. The study considered all potential retro-
fit approaches except systems employing lean idle mixture adjustment and
ignition timing modification developed by major U.S. automobile manufacturers.
This approach was excluded from consideration in the study because a
representative device of that type was concurrently being evaluated in a
more extensive testing program (see Section 3.3).
The study was initiated by performing a thorough search for all sources
of information on retrofit methods and developers. Each source was sent a
letter describing the purpose of the program and requesting their partici-
pation in the program. Each respondent expressing interest in participating,
and who was an actual candidate retrofit method developer, was sent a
request to provide data on his device. These data were used to screen
the devices by type of retrofit method and to rank them based on their
feasibility. The most feasible and representative devices available were
selected for a test program.
-------
3-7
A fleet of twenty vehicles representative of the ore-controller!
used car population was used for testina the oerformanr.p nf thp
devices. The cars were tested for emissions and driveability both with
and without retrofit.
From the devices selected for testing, four were further screened
for additional tests which included durability tests.
To evaluate and compare the different retrofit systems, a
methodology was developed that provides quantitative figures of
merit or indexes for feasibility criteria, performance, and cost-
effectiveness.
3.2.2 RETROFIT TYPES IDENTIFIED
Retrofit devices which are designed to control emissions from
gasoline-powered motor vehicles can be classified according to the
sources of vehicle emissions they control:
Group 1: Exhaust Emission Control Systems
Group 2: Crankcase Blowby Emission Control Systems
Group 3: Fuel Evaporative Emission Control Systems
Group 4: Combinations of these groups
Table 3-1 shows a more detailed classification structure used to
categorize retrofit devices studied in this program.
Some principles and descriptions associated with the principal
retrofit systems follow.
3.2.2.1 EXHAUST EMISSION CONTROL SYSTEMS
In controlling exhaust emissions, retrofit devices may be
designed either to work on the exhaust gases after they leave the
combustion chambers and enter into the exhaust system or to decrease
-------
3-8
TABLE 3-1 Retrofit Approach Categories
Group 1. Exhaust Emission Control Systems
1.1 Exhaust Gas Control Systems:
Catalytic Reactor
Thermal Reactor
Exhaust Gas Afterburner
Exhaust Gas Filter
"".2 Induction Control Systems:
Air Bleed to Intake Manifold
Exhaust Gas Recirculation
Intake Manifold Modification
Carburetor Modification
Turbocharger
Fuel Injection
1.3 Ignition Control Systems:
Ignition Timing Modification
Ignition Spark Modification
1.4 Fuel Variations:
Alternative Fuel Conversion
Fuel Additives
Fuel Conditioners
Group 2. Crankcase Emission Control Systems
Closed Systems
Open Systems
Group 3. Evaporative Emission Control Systems
Crankcase Storage
Canister Storage
Group 4. Emission Control Combinations
-------
3-9
emission formation by modifications to the induction system or the
ignition system, or by the use of fuel conversions or fuel additives.
Within these broad categories there are a number of different
approaches which can be pursued and for which devices were evaluated
in the program.
EXHAUST GAS CONTROL SYSTEMS
One approach for reducing HC and CO is to subject the exhaust to
an oxidation process. Among retrofit devices, this is done by using
either a catalytic reactor, a thermal reactor, or an afterburner.
In the catalytic reactor approach, the exhaust gas is passed
through a canister containing a catalyst for oxidizing HC and CO to
carbon dioxide (CX^) and water. The catalyst is not consumed in the
oxidation reaction but deterioration may result from use of certain
fuels (such as leaded gasoline). The heat required to initiate
oxidation comes from the exhaust gas itself. The oxygen needed for
oxidation in the catalytic converter is provided by leaning the fuel
mixture at the carburetor or by the addition of suoDlementary air
through installation of an air pump.
The thermal reactor works in much the same way. The reactor is
located as close'as possible to the combustion chambers where the
exhaust gas temperature is usually high enough to support oxidation of
HC and CO without having to use a catalyst. In the case of a thermal
reactor for rich mixture, adequate oxygen availability is provided by
means of air pumped directly into the exhaust manifold, near the
exhaust valves.
The exhaust gas afterburner oxidizes the HC and CO in a muffler-
type container installed in the exhuast system where an ignition
-------
3-10
source is provided, such as a spark plug. It is designed to operate
with rich fuel mixture gases and, therefore, additional air must be
supplied also.
In some designs, reactor approaches to oxidizing exhaust gas
HC and CO also indirectly reduce NO. Frequently, a rich fuel mixture
is set at the carburetor with the result of inhibited NO formation
because of the decreased availability of oxygen in the combustion
chambers of the engine.
The purpose of gas filters is to eliminate or reduce particulate
emissions such as lead or soot from the exhaust stream. Relatively
little work has been done on gas filters in general, and particularly
for retrofit purposes. Cyclone separators, glass fiber filters, and
scrubber devices would be possible approaches.
INDUCTION CONTROL SYSTEMS
Many retrofit devices of this type operate, in general, on the
basis of either leaner air-fuel mixture ratio or Improved distribution
of the mixture. Lean fuel mixtures provide HC and CO reduction due to
the increased oxygen availability. Air bleed to intake manifold,
carburetor modifications, and intake manifold modifications are
representative methods that produce lean mixtures and/or improved
mixture distribution.
Recirculating a portion of the exhaust gases back into the
induction system is an effective method to control oxides of nitrogen.
The recirculated gas reduces the peak temperature of the combustion
process which results in less NO formation.
-------
3-11
IGNITION CONTROL SYSTEMS
Most ignition control systems use the principle of retarding
the ignition spark. This retardation results in lower peak temperatures
in the combustion chamber, and higher temperatures in the exhaust
gases; and, therefore, less NO is formed in the engine and more HC is
oxidized in the exhaust system.
A simole wav of retarding the spark is by disconnecting the
vacuum advance of the distributor. This can be done automatically
during the appropriate driving modes with different degrees of
sophistication according to the type of device.
FUEL VARIATIONS
The substitution of normally gaseous fuels for gasoline is another
approach to emission control. Substitute fuels normally considered
are liquified petroleum gas (LPG), liquified natural gas (LNG), and
compressed natural gas (CNG). Use of gaseous fuels allows leaner
mixture operation and can result in lower emissions than for gasoline.
In addition, the photochemical reactivity of hydrocarbons in the
exhaust gas from gaseous fuels can be substantially less than when
gasoline is used; however, there is at present no Federal reactivity
scale which allows for quantitative adjustment for this factor.
Fuel additives may also have some potential to reduce emissions.
Certain additives, for instance, may reduce engine deposits, and
therefore, decrease the tendency for CO and HC emissions to increase
as mileage is accumulated.
3.2.2.2 CRANKCASE BLOWBY EMISSION CONTROL SYSTEMS
Engine blowby results when the air- fuel mixture in the cylinder
-------
3-12
escapes past the piston rings during the compression and power strokes.
The vapors enter the crankcase and subsequently escape to the atmosphere.
Crankcase control systems provide a means of circulating ventilation
air through the crankcase, mixing with the blowby gases, and recirculating
the mixture into the intake manifold, usually through a variable orifice
control valve. The flow rate through the valve is normally controlled
by intake manifold vacuum.
Typically, ventilation air is drawn either directly from the engine
compartment (referred to as an open system), or from the engine air
cleaner through a tubing into the crankcase (a closed system).
3.2.2.3 EVAPORATIVE EMISSION CONTROL SYSTEMS
These systems control emissions from fuel that is evaporated from
the fuel tank and carburetor. Gasoline tanks and carburetors are vented
to the atmosphere on pre-1970 vehicles sold new in California and on
pre-1971 vehicles sold new nationally. Losses at the carburetor occur
almost entirely during the hot soak period after shutting off a hot
engine. Residual heat causes the temperature of the fuel bowl to rise to
150-200°F, resulting in substantial boiling and vaporization of the fuel.
Devices in this group to retrofit used vehicles are not currently
available. However, the types of systems employed to control
evaporative emissions from new cars are described briefly below.
In one type of evaporative emission control system, the crankcase
is used as a storage volume for vapors from the fuel tank and carburetor.
During the hot soak period after engine shutdown, the declining tempera-
ture in the crankcase causes a reduction in crankcase pressure
sufficient to induct vapors. During this period, vapors emanating from
-------
3-13
the carburetor are drawn into the crankcase. Vapor formed in the fuel
tank is carried to a condenser and liquid-vapor separator; the
condensate returns to the fuel tank, and remaining vapors are drawn
into the crankcase. When the engine is started, the crankcase is
purged of vapors by the action of the positive crankcase ventilation
sys tern.
In the absorption-regeneration system, a canister of activated
carbon traps the vapors and holds them until such time as they can be
drawn back into the induction system for burning in the combustion
chamber. During a hot soak period, vapor from the fuel tank is routed
to a condenser and separator, and liquid fuel is returned to the tank.
The remaining vapor, along with fuel vapor from the carburetor, is vented
through the canister filled with activated carbon which traps the fuel
vapor. When the engine is started, fresh air is purged through the
canister and removes the trapped fuel vapor from the activated carbon
and carries it to the combustion chamber.
3.2.2.4 EMISSION CONTROL COMBINATIONS
In many cases retrofit systems combine two or more of the retrofit
generic types summarized above. Thus, for instance, methods such as
"Exhaust Gas Recirculation with Distributor Vacuum Disconnect,"
"Catalytic Converter with Distributor Vacuum Disconnect," are basically
just combinations of ignition timing modification with exhaust gas
recirculation and catalytic converter, respectively.
3.2.3 RETROFIT EVALUATION METHODOLOGY
To evaluate and compare the performance of the various retrofit
systems relative to each other, a methodology was developed that
provides quantitative figures of merit or indexes for feasibility
-------
3-14
criteria, performance, and cost-effectiveness. This methodology is described
briefly below; a more detailed description may be found in the contract
report 2_
CRITERIA INDEX
The Criteria Index can be expressed by a product of terms each of
which can be given the value of either one or zero. It provides an
indication as to whether a device will meet the various constraints or
limiting values specified for each performance parameter. If the
Criteria Index is zero it means the device did not pass one or more of
the specified requirements. In that case, the device is sub-standard
for at least one of the set of Criteria used. If the Criteria Index
is one, it means the device has met all criteria factors.
The Criteria Index includes the following factors:
a. Emission levels for HC, CO, NOx.
b. Safety
c. Critical driveability
d. General driveability
e. Installation cost
f. Recurrent cost
g. Reliability
h. Maintainability
i. Availability
Stated simply, the "Criteria Index" is just a check list to
verify if a retrofit passes specified performance requirements.
PERFORMANCE INDEX
The Performance Index provides a more quantitative evaluation of
a device, in contrast to the Criteria Index which screens devices for
a "yes" or "no" answer as to their basic feasibility. The Performance
-------
3-15
Index is a measure of the emission reduction benefit of a device
relative to its cost and driveability penalties, and it is expressed by
the following equation:
[Emission] (Driveability) (Cost
C1 V Index I 'C2 V Index I ~C3 V Index
Performance Index = \ / \ / \ ^
CT + C2 + C3
This equation weights, according to the weighting coefficeints
(C-|, C2, 03), the three terms related to emissions, driveability, and
cost. These three terms are expressed by corresponding indexes which
include scaling factors to bring their values within the same order of
magnitude.
The Emission Index is a weighted sum of the reductions of each
of the considered pollutants. In applying the methodology, pollutant
weighting factors must be chosen based on specific air quality problems
to be solved. The Driveability Index is based upon a driveability
score determined by assessing what might be considered demerits for a
variety of abnormal driving characteristics (rough idle, detonation,
surge, etc,). The Cost Index combines those cost parameters which
determine the initial cost of the device and the recurrent costs.
To compute the Performance Index, careful judgment must be
exercised in assigning the three weighting coefficients (C-j, 03, 03)
because the weights given to emission reduction, driveability, and
cost can greatly influence the value for the Performance Index and
the relative ranking of the devices that might be compared. These
weighting coefficients can very well be different depending on the
requirements or judgment of the specific air pollution control agency using
-------
3-16
this evaluation methodology.
COST EFFECTIVENESS INDEX
The Cost-Effectiveness Index provides an additional information
item to complement the Performance Index. Cost-Effectiveness is
usually defined as the ratio of the desired results or the desired
output versus the required cost input. In this case the CEI is
defined as the ratio of the Emission Index to the Cost Index:
Cost-Effectiveness Index = Emission Index (per unit reduction)
Cost Index ($/100 miles)
3.2.4 RESULTS OF PRELIMINARY SCREENING
The data obtained from all sources were analyzed to select the retrofit
methods tested in the program. Following the different generic
retrofit groups, this section lists, briefly, which retrofit methods
were selected for testing and which were evaluated without testing,
indicating the fundamental reasons for this distinction. Also, in this
section are presented general comments on some of the retrofit
methods not tested in the program. Additional details may be
found in the contract report , where some sixty retrofit devices are
evaluated. The results for the retrofit systems tested in the program
are presented in Section 3.2.6.
3.2.4.1 EXHAUST EMISSION CONTROL SYSTEMS
EXHAUST GAS CONTROL SYSTEMS
Catalytic Reactor: A retrofit of this kind, with distributor
vacuum disconnect was selected for testing. Results are given in
Section 3.2.6.
Thermal Reactor: There were no developers of thermal reactors
interested in retrofit application. Thus, no thermal reactor was
-------
3-17
tested, but the contractor evaluated a representative type, finding that
it would cost approximately $375.
Exhaust Gas Afterburner: Some devices of this sort were evaluated,
but none tested. Study shows that this type device is less reliable
than other types of retrofits.
Exhaust Gas Filter: None was selected for testing. Currently there
are no regulations for particulate emissions from automobiles. The
devices offered to the program were evaluated.
INDUCTION CONTROL SYSTEMS
Two different Air Bleed to Intake Manifold, three distinct Exhaust Gas
Recirculation, and three different Carburetor Modification systems were
selected for testing, and their results are presented in Section 3.2.6. Other
less representative induction control retrofits were also evaluated but not
tested in the program.
IGNITION CONTROL SYSTEMS
Two different Ignition Control approaches were selected for testing.
Their results are presented in Section 3.2.6. Other less representative
systems were evaluated but not tested.
FUEL VARIATIONS
Alternative Fuel Conversion: Typically, substantial emission
reductions can be obtained but, in general, the very high cost of
the system, and the limited availability of fuels would limit the
application only for special fleets or circumstances. Devices offered
to the program were evaluated. The feasibility of gaseous fuel
conversion as a vehicle emission control strategy is discussed further
in Chapter 4.
Fuel Additives or Conditioners: The time constraints of the test
program did not permit the mileage accumulation needed to evaluate fuel
-------
3-18
additives and, therefore, none were tested. However, some data offered
to the project on these methods were evaluated.
3.2.4.2 CRANKCASE EMISSION CONTROL SYSTEMS
Considering that there is already substantial information on this
emission control system, none was tested. It is expected that
reliable devices can eliminate crankcase emissions from uncontrolled
cars. However, only some twenty-three percent of the current total
car population do not have control for crankcase emissions. Devices
offered to the program were evaluated. The conventional crankcase
retrofit costs up to $40 installed.
3.2.4.3 EVAPORATIVE EMISSION CONTROL SYSTEMS
No device was tested for evaporative emission control. In this
category there were no found devices to retrofit used vehicles. It
has been estimated that reliable devices could largely eliminate
evaporative emissions from uncontrolled cars; about eighty-five percent
of the current total car population do not have evaporative control.
The contractor made some evaluation on the basis of the systems supplied
in new vehicles. An estimate of $140 was made for the cost to
retrofit a used car with evaporative control should such retrofits be
made available.
3.2.5 TEST PROGRAM
A summary of the test vehicle fleet, and of the various tests
conducted for the retrofit program fleet follows.
TEST VEHICLE FLEET
A total fleet of twenty pre-controlled used cars was used for
testing the performance of the retrofits. This fleet was constituted of
two replicate subfleets which were bought and operated in Anaheim,
California and Taylor, Michigan respectively. The rationale for the
-------
3-19
replicate fleets was to evaluate differences in retrofit performance
between the two fleets which could possibly be attributable to
driving conditions, geographical location, and vehicle climatoloqical
exposure history at two disparate locations in the United States, or to
bias in testing facilities and personnel.
The criteria for selecting the test vehicles were based mainly
on the number of vehicles on the road by make and model year orior to
1968, and on the popularity of engine size and type of transmission
within that vehicle population. The contract report^ provides details
of the test fleet composition.
Appropriate inspection was applied to ensure that all the cars
bought for the test fleet were in a condition reoresentative of
normally-maintained vehicles for their corresponding model year.
EMISSION TESTS
As each vehicle was procured, an "as received" exhaust emission
test and a driveability test were conducted. The vehicle was then
tuned to the manufacturer's specifications to minimize the possibility
of engine malfunction during the subsequent retrofit system tests and
to provide a reproducible baseline. The vehicle then received a
series of baseline exhaust emission tests and driveability tests.
After each baseline test, the vehicle was equipped with a candidate
retrofit system for exhaust emission and driveability tests. The tests
alternated between the baseline and retrofit system tests until
testing of all candidate systems was completed.
The 1972 Federal Test Procedure 3 was used to measure the exhaust
emissions during the baseline and retrofit tests. The Federal exhaust
emission test consists of prescribed sequences of fueling, parking
(cold soak), dynamometer operation, sampling, and analytical calculations.
-------
3-20
The exhaust test is designed to determine hydrocarbons, carbon monoxide,
and oxides of nitrogen on an actual mass emission basis while the
vehicle is simulating an average urban trip of 7.5 miles.
Fuel consumption was measured during the baseline and retrofit
exhaust emission tests. For this purpose the fuel consumed during the
driving cycle was measured by weight. This measurement is representative
of fuel consumption in urban driving; however, the statistical
significance of these test results is not known.
DRIVEABILITY TESTS
The Automobile Manufacturer's Association (AMA) standard driveability
test procedure was used to evaluate the operating characteristics of
the vehicles on the road. Basically, the procedure consists of a
cold start driveaway following an overnight soak period; and then a hot
start driveaway following the cold start portion of the test. The cold
start evaluations consist of engine startup, idle , and part throttle
and full throttle acceleration modes up to 30 mph. The hot start
evaluations consist of a series of cruises, accelerations, idle modes of
operation, and hot start restarts. The quality of each driving mode
was noted by the driver and recorded by an observer during each mode
of operation. Vehicle performance was determined at wide open
throttle from 0 to 60 mph by measuring the elapsed time. As explained
the Section 3.2.6, four devices were tested more extensively, and
in this case some driveability tests were also performed to determine
whether environmental extremes (such as high altitude) had any significant
performance effect on vehicle driveability when a retrofit device was
installed.
DURABILITY TESTS
For the four devices that were tested more extensively (see Section
-------
3-21
3.2.6), durability tests were also performed. These tests consisted of
driving retrofit equipped cars for 25,000 miles and measuring the
exhaust emission by the 1972 Federal Test Procedure at 5,000 mile
increments. Mileage accumulation was performed on a test route which
consisted of freeway, urban, and city driving, at an average speed of
approximately 35 mph. Fuel consumption, and any significant occurences
during the mileage accumulation were recorded daily. Durability tests
were made with only one car for each device, except in the case of the
catalytic converter retrofit which had durability tests on two vehicles.
This device requires the use of unleaded fuel, and in one of the two
cars only this fuel was used; in the other car a tank refill with leaded
premium gasoline was made every 2500 miles to determine the influence
of occasional exposure to leaded fuel on the performance of the
catalytic converter.
3.2.6 RESULTS FOR RETROFIT SYSTEMS TESTED IN THE PROGRAM
3.2.6.1 GENERAL RESULTS
As identified in Section 3.2.4, eleven retrofit devices were
selected for testing in this program. Table 3-2 lists these devices.
from these, four devices were selected for more extensive testing. These
four devices were selected to cover representative retrofit types,
considering their performance potential, and also the availability of
retrofits to test in all of the fleet cars.
All eleven devices were submitted to exactly the same test pro-
cedures for emissions and driveability, the only difference being that
the four devices tested more extensively received up to eighteen complete
-------
3-22
TABLE 3-2 RETROFIT SYSTEMS TESTED
Contract Report ^
Devices Receiving more than 3 Tests (up to 18) Identification' Number
* Air Bleed to Intake Manifold (controlled
by a poppet valve) 1
* Catalytic Converter with Distributor Vacuum
Advance Disconnect 96
* Ignition Timing modification with Lean Idle
Adjustment 175
* Speed-Controlled Exhaust Gas Recirculation with
Distributor Vacuum Advance Disconnect 246
Devices Receiving up to 3 Tests
* Throttle-Controlled Exhaust Gas Recirculation with
Distributor Vacuum Advance Disconnect 10
* Carburetor Modification of Main Nozzle Differential
Pressure 33
* Air Bleed to Intake Manifold (controlled by compressible
plastic balls valve) 42
* Electronic-Controlled Distributor Vacuum Advance Dis-
connect and Careburetor Lean Idle Modification 69
* Variable Camshaft Timing 245
* Carburetor Main Discharge Nozzle Modification 288
* Variable Venturi Carburetor 295
-------
3-23
tests, while the other seven devices received up to three complete tests.
Descriptions of the four device and their more important results.
are presented separately in the following subsections. Table 3-3
lists the main results for all of the eleven devices tested in the
program.
3.2.6.2 AIR BLEED TO INTAKE MANIFOLD (Contract Report Identification No. 1)
DESCRIPTION
This tyoe of device is an air valve that enables the air-fuel
ratio to be increased by metering additional air to the intake mani-
fold in accordance with intake manifold vacuum.
The specific device tested consists of a cylinder incorporating an
oil-damped air valve, an air intake adapter plate, and an air bleed
hose. The air valve cylinder mounts in the engine compartment and
the adapter plate installs between the existing carburetor and the
intake manifold. The hose connects the air valve to the intake
adapter plate.
EVALUATION RESULTS
Emission
Percent Reduction
Pollutant Pooled Mean
HC
CO
NOX
Installed Cost:
Fuel consumption
21
58
-5
$56 to 64
, as measured during
Reductions
90% Confidence
Intervals for
Mean % Reduction
10 to 32
22 to 80
-15 to 5
1972 Federal Emission
Statistical
Significance
Significant Reduction
Significant Reduction
Insignificant Increase
Test Procedure:
4% improvement, i.e., more miles per gallon. Statistical significance of
change not known.
-------
H
hJ
c/j
os
H
cn
B
o
>
D
H
M
O
a;
H
w
!*T
JJ CO
CO S-i
i-H I
r-H v_^
4J QJ
co E
M H
QJ *O CO
u m C
c * S
2 T-. e
s > £
S«o
"I 4J cr
S C oj
* r-l Oi
4J
Irl fa
XJ H
CO M
•rH £
T-l S
PJ
>>
•rH
• rH
X)
CD
O)
i-i
O
*W<
0*
rH
CO ^^
CO CO
C O
M a
CO ,,
C
O CJ g
•iH C
JJ -H
CJ i-l
3 0)
TJ E to
0) O co O
Cri J-i CO U
U-l
OJ
C C
co b
s s
CO
• 4J
O MH co
S 0 OJ
JJ
OJ O
1— 1
oj
CJ
•H
>
S
in
CM
rH~
O
• o
E r-*
O -^
O •
O to
CM cn
I-H
o
o
o
o
m
r^-
^__
XI
OJ
rH
XI
CD
jj
CX
o
u
<;
S
O
m
^^
CD
00
in
rH
CN
00
rH
rH
i-H
O
MH
"p
1
OJ
CO
JJ
c
0
4J
TJ
0)
OJ
1— 1
XI
•H
•*
TJ
CJ
O
O
o*
m
^
XI
a>
rH
XI
CD
jj
CX
QJ
CJ
CJ
4J O
CO OJ
1-1 C
TJ C
O
JJ CO
•H -H
> TJ
to OJ
01 U
JJ C
to CO
OJ >
> TJ
C CD
o
0 i
3
0 3
-rH CJ
JJ ffl
i-H
CD to
JJ O
CD JJ
U 3
i— i
QJ
O
O
O
m
f~-
^_^
XJ
OJ
Xl
CO
4J
CX
CJ
CJ
o
^
in
in
CN
i-H
O
. m
6 r-
O "-*
O •
O 1-
* x:
CM m
O
O
o
o
in
xT
CU
i-H
X>
CD
JJ
cx
cj
o
CJ
TJ CO
01 CO
OJ 0)
rH to
•a a
rl O
rH
o
• o
E CJ
m
CN
CM
in
• CM
o ---
o -
O i-i
in m
CN
O
O
o
o
m
i-*
^^
xs
4)
3
CO
4J
a.
O)
o
o
•^
00
CM
r^
CM
1
m
r-H
in
CN
00
C
•iH
e
•rl
JJ
JJ
MH
CO
x:
CO
E
CO
OJ
r-H
CO
CD
in
CN
rH
0!
S •
1 1
0)
i-i
O
o
m
^_,
MH
rH
1
^
en
CN
00
00
CN
01
60
to
CO
Xl
o
to C
-rH O
TJ -rH
JJ
C CO
-H CJ
SiH
<4H
•H
O *O
4J E
01
It
0 o
o c
in
P^.
0
m
• CN
•rH
E ^
O \
o •
O to
CM m
o
o
o
m
^^
00
Ch
m
CM
o
CN
r-
m
i
,_(
m
CM
J-i
O
JJ
01
S-i
3
to
CO
CJ
3
4J
c
•*
OJ
i-H
X)
CO
•rH
CO
4J
C
CD
CJ
•r-t
UH
-H
C
00
CO
>,
sticall
4J
CD
4-1
CO
OJ
X5
4J
O
c
>»
E
CO
4-J
CO
JJ
OJ
0)
4H
^
O
01
0)
J-l
XJ
jj
x:
jj
CO
O)
CJ
01
i-l
o
MH
CO
4J
3
01
C
o
•H
CO
CO
•rl
E
CO
4-1
CO
O)
4J
MH
O
j-i
Xl
e
c
01
4J
§
a
3
CO
(3
g.
01
CO
CO
a>
x:
jj
MH
O
jj
•H
i— 1
•H
XI
CO
r-H
OJ
*
QJ
to
3
CO
MH
i— i
jj
• O
O OJ
O to
O
fXMH
?JS
3 00
JJ CO
OJ
OJ r-H
S -B
•rl E
00
C C
0) CO
OJ
01 E
3 TJ
i— 1 OJ
CJ JJ
c 2
•r-l
4J 4J
O CO
C Cd
Sh
o to
0 g
* +
J-l
^
rH
CO
CD
c
o
•H
4J
CO
3
X3
E
o
o
cx
o
x:
CO
CD
TJ
C
CO
CJ
•H
C
s
CJ
•H
JJ
jj
3
CO
CO
OJ
•H
3
cr
OJ
c
o
JJ
CO
r-H
rH
CO
4J
CO
rH
C
JJ
C C
O 0
CJ -H
jj
G t-i
•H O
4J • -rH
CO JJ JH
i-i C OJ
01 CD 4-J
CX CJ OJ
O -H TJ
MH
TJ -iH S-l
O C O
O 00 C
co E
-S
rH E
OJ rH O
x> CD co
o
TJ -H TJ
r-H JJ CU
3 CO g
XJ 4J XJ
CO CO CO
4J
OJ CO CO
C 4J
-H JJ CO
00 O Ol
C Z H
+ co XI
jj
CO
CO
o
rH
'g
0
o
o
CM
i-l
0)
CJ
TJ
0)
C
CD
0)
CJ
4>
XJ
O
4J
J-i
0)
JJ
• rH
CO -iH
O) MH
•H CX
4-J £
•H -J
CO J-i
•H
CO -rf
3
O •
S-i CO
o o»
x: c
CX -r-l
CO 00
0 C
J_l
4J OJ
3 TJ
O C
X -r*
4J rH
-H >,
S o
QJ
3 r*
MH 0
MH
TJ
o» m
CO •
r-H to
CD o
jj MH
CO
o o
CN
•
CO OJ
01 O JJ
co 3
OJ rH O
-H OJ CO
rH J-i
MH
JJ JJ O
CO CO
t^ >, JJ
rH r- 1 CO
CO CO O
jj jj a
U O CO
0 TJ
^
i-l
CD
CO
CO
O)
CJ
OJ
CO
-r-t
JJ
•H
C
3
S
0)
c
CO
x;
4-J
•r-l
S
c
s
0)
o
CO
i-H
CX
OJ
J-l
TJ
c
CO
CO
O
E •
OJ
a>
JH
3
r-H '
•H
CD
MH
0)
CJ
0) '
M-4 I
0
4J
C 'I
> -1
0) r
0»
JJ (
C J
•H
J
CU
rH J
Xl c
CO 4
CO J.
o :
cx c
4J
o c
C T
co a
•rH (.
r-< J.
-H J.
CD •
(D 4-1 V
pi CO
0)
• JJ
J-i
01 OJ •.
a C x
OJ ?
> JJ 4-
OJ C *r
T) OJ r-
S T-
OJ 01 X
rC > 1
JJ O
S-i
-Q1 E* T
•iH T
T)
OJ OJ r-
JJ E C
co 5 C
at co c
CTTJ T
01 01 i
o c
CO XI X
•H CO 0
Ol JJ J-
gco a
OJ 0
s H e-
01 MH 0
-------
3-25
Safety: There are no apparent safety hazards.
Driveability: Some minor deterioration. Acceptable.
Reliability: Estimated about 75,000 miles before total failure.
Installation: Engine should be in good operating condition. In-
stallation takes about 1.25 hours; requires automotive mechanic and
shop combustion analyzer.
Maintenance: About 0.3 hours every 12,000 miles, with cost of
approximately $7.00.
Applicability to vehicle population: Applicable to most uncontrolled
cars having conventional carburetion. In the case of vehicles that
already incorporate lean carburetion, installation of air bleed
retrofits might cause misfire or adverse driveability characteristics,
and therefore, criteria would have to be established to identify lean-
ness limits for cars on which the retrofit could be applied.
Durability results: At the beginning and end of the durability tests,
the percentage emission reductions caused by the device were the
following:
Zero Miles 25,000 Miles 25,000 Miles
(1 test without device, (3 tests with device, (Device reinstalled
1 test with device) 3 tests without device) and adjusted, 1 test.
3 tests without device)
HC 1% 7% -7% (increase)
CO -2% (increase) 7% 64%
NOX -16% (increase) 21% 9%
At the 20,000 mile tests, it was discovered that blowby deposits were
plugging the PCV (positive crankcase ventilation) tube at its connection
with the carburetor. Cleaning of this connection at that mileage reduced
-------
3-26
CO from 99 to 45 grams per mile. This problem was not caused by the
retrofit device, and it indicates the importance of proper vehicle
maintenance and periodic emission inspection to secure the benefits of the
emission controls. (The emission tests at zero durability miles
indicated unexpected results, however the durability mileage accumulation
was undertaken without determining the reasons for these irregularities.)
At PFi^OOO miles the Device '-'as reducim emissions. However, although
this is basically a device for reducing CO, it was reducing this emission
by only about 7%. On the basis of this poor reduction, the device was
reinstalled and adjusted. This brought the CO reduction up to about 64%
as indicated in the last column above.
The variability of results at 25,000 miles indicate the sensitivity
of the device to proper adjustment. This emphasizes the need for
proper installation procedures including the use of a combustion
analyzer for proper adjustment.
3.2.6.3 CATALYTIC CONVERTER WITH DISTRIBUTOR VACUUM ADVANCE DISCONNECT
(Contract Report Identification No.96)
DESCRIPTION
This device consists of a catalytic converter installed in the engine
exhaust system between the exhaust manifold and the muffler. The
converter is located as close as possible to the exhaust manifold. Six
cylinder engines use one converter and V-8 engines two converters, one
for each exhaust manifold. The converter chamber contains a platinum
pellet-type catalyst bed. The bed is filled through a removable plug
in the side of the chamber.
For vehicles older than 1968 model year, an air Dumo is required, in
general, to supply the air needed to support the oxidation process in
-------
3-27
the converter. The pump Is mounted on the front of the enqine where it
can be belt-driven from the drive shaft pulley. On later model cars,
the carburetor can generally be tuned to a sufficiently high air-fuel
ratio to provide the amount of air needed.
To protect the converter from overheating, a melt-out plug is
incorporated in the chamber. If the temperature rises above the level
at which the catalyst can operate without being damaged, the plug
melts and the exhaust gas is vented directly into the ambient air.
In this case the noise will indicate the failure and the necessity for
repair. In addition, a thermocouple installed in the converter is wired
to an electronic control which energizes a solenoid to divert the air
pump output away from the converter if the limit temperature is
exceeded.
To disconnect the standard distributor vacuum advance system, a
thermoswitch, installed in the radiator water return line is used. The
vacuum advance hose is connected through the thermoswitch, which is
normally closed, preventing the intake vacuum from actuating the
distributor vacuum advance mechanism.
EVALUATION RESULTS
Pollutant
HC
CO
NOX
Percent
Pooled
68
63
48
Emission Reductions
90% Confidence
Reduction Intervals for
Mean Mean % Reduction
53 to 91
37 to 97
17 to 64
Statistical
Significance
Significant Reduction
Significant Reduction
Significant Reduction
-------
3-28
Installed Cost: $143 for 6 cylinder engines or $175 for V-8 engines;
includes $85 for air pump.
Fuel consumption as measured during 1972 Federal Emission Test Procedure:
1% penalty. Statistical significance of change not known.
Safety: There are no appraent safety hazards (on the assumption that the
converter preventive overheat system would be reliable).
Driveability. Some minor deterioration. Acceptable.
Reliability: Estimated about 50,000 miles before total failure. This
does not extend to the catalyst which should be changed every 25,000
mi 1es.
Installation: Engine should be in good operating condition. Installation
takes about 4 hours; requires automotive mechanic, shop combustion analyzer.
Maintenance: Catalyst should be replaced every 25,000 miles. Cost
(including labor), and replacement time are estimated to be as follows:
$15, 0.25 hours, for 6 cylinder engines
$20, 0.30 hours, for V-8 engines
In addition, the pump air filter should be cleaned every 12,000 miles;
labor cost will be about $3.
The catalyst requires the use of gasoline-without lead or
phosphouous additives. However (see durability results below), it
appears that occasional use of leaded fuel impairs the efficiency
of the catalyst relatively little. Abnormal oil blowby due to worn piston
rings can affect the catalyst adversely because of certain oil additives.
-------
3-29
Applicability to vehicle population: Applicable to most domestic
models. Results presented here are for cars without exhaust controls,
however substantial emission reductions should be obtained on
controlled vehicles also. The retrofit was not available for VW cars,
the only foreign vehicle in the test fleet.
Durability Results: At the beginning and end of the durability tests,
the percentage reductions caused by the device were as follows:
Case of unleaded fuel, but one tahkful of
leaded premium gasoline every
2,500 miles
Zero Miles 25,000 Miles
(1 test without device (3 tests with device
1 test with device) 3 tests without device)
HC 81 % 80%
CO 94% 80%
NOY 47% 30%
A
The air bypass valve for overtemperature converter protection was
found to be failure prone with mileage accumulation; it had to be
replaced twice during the 25,000 mile durability evaluation.
Case of unleaded fuel only
Zero Miles
(1 test without device
1 test with device)
79%
97%
35%
20,000 Miles
(2 tests with device
1 test without device)
67%
56%
49%
HC
CO
NOX
Shortly after the 20,000 mile testing, the catalytic converters
were removed from the vehicle and inspected. 87% of the catalyst was
-------
3-30
recovered from the left converter; however, no catalyst was recovered
from the right converter due to high temperature failure of the catalyst
retaining screen. The car had been deteriorating rapidly, as indicated
by a large increase in engine oil consumption, and the excess burning of
oil might have accelerated the activity of the catalyst causing a
premature failure of the converter.
The durability tests were terminated after discovery of the loss
of catalyst from the right converter. This occurance points out the
necessity of periodic emission control system inspection to ensure continued
satisfactory operation of this type of retrofit system.
3.2.6.4 IGNITION TIMING MODIFICATION WITH LEAN IDLE ADJUSTMENT
(Contract Report Identification No. 175)
DESCRIPTION
This device consists of an ignition control assembly which is
attached to the fender wall under the vehicle hood. It includes the
following electronic circuits and components to retard, in a controlled
manner, the ignition timing of the engine:
1. A solenoid operated,valve which connects or disconnects the
distributor vacuum advance.
2. An ignition circuit which regulates the distributor point signal
to a retarded condition at vehicle speeds below 35 mph.
3. A sequencing circuit and switch which senses vehicle speed and
controls the regulation provided by the first two items.
The device makes three wire connections with the vehicle
ignition system, and two hose connections to attach to the
carburetor distributor vacuum port and distributor vacuum advance chamber.
-------
3-31
The lean idle mixture is obtained by adjusting the carburetor
idle mixture setting.
EVALUATION RESULTS
Emi ssi on
Percent Reduction
Pollutant Pooled Mean
HC 19
CO 46
NO,, 37
Reductions
90 ' Confidence
Intervals for
Mean "' Reduction
9 to 29
-8 to 77*
27 to 47
Statistical
Significance
Significant Reduction
_, _„ ., Significant Reduction
A
* There was a large difference between the results for CO found
in the two cities. In both cases the mean results indicated CO
reduction, but while in Taylor the 90% confidence intervals for the
mean percent reduction were 67 to 77% in Anaheim the limits were -8
to 23%, showing a reduction with no statistical significance. The CO
data were reviewed for anomalies to determine possible causes of the
conflict in results between the two cities, but none were found. However,
the operating principle of this device would not allow one to expect
large reductions of CO ; and,therefore,the mean reduction with lower
confidence intervals may be more representative for this device.
Installed Cost: $45
Fuel consumption, as measured during 1972 Federal Emission Test Procedure:
10% penalty. Statistical significance of change not known.
Safety: There are no apparent safety hazards.
DftVe'a'bi'llty':"' So~me~ffiinor~deteriofation. Acceptable.
Reliability: Estimated about 75,000 miles before failure.
-------
3-32
Installation: Engine should be in good operating condition.
Installation takes about 1 hour; requires automotive mechanic, shop
combustion analyzer.
Maintenance: None is required. Repair is not possible; in the
event of device failure, removal and replacement with new unit is
required.
Applicability to vehicle population: This retrofit device can be
applied to all domestic cars having engines over 140 cubic inches
displacement and not equipped with exhaust emission controls, from
1955 to 1967 model years (up to 1965 model year in California).
Durability Results: The device caused the following percentage
emission reductions at the beginning and end of the durability test:
Zero Miles 25,000 Miles
(1 test without device (3 tests with device
1 test with device) 3 tests without device)
HC 28% 8%
CO 18% -24%(increase)
NOx 48% 39%
High CO levels (up to 100% increase from baseline)
were measured at the 10,000, 15,000 and 20,000 mile points.
These were probably caused by faulty choke operation. The car
operated normally but it was found that the choke was partially
closed most of the time, and, therefore, the choke was repaired at
23,100 miles.
After 23,500 miles the vehicle alternator diodes failed. The
alternator was repaired and mileage accumulation was continued. Coincirlentally,
the vehicle developed moderate hesitation at heavy engine loads which
-------
3-33
was particularly noticeable when accelerating on the freeway. Following
the 25,000 mile emission test, a diagnosis was conducted to determine
the cause for the hesitation under heavy load. The retrofit device
was inspected by the developer, and it was revealed that two diodes
were malfunctioning in the electronic circuit of the device. It
could be that the CO increase recorded at 25,000 miles was due to
device malfunction originated by the alternator failure; and if this
was the case, the device should include some protection to prevent
such malfunctions in the event of vehicle electrical failures.
This shows the necessity of proper vehicle maintenance and
periodic emission control system inspection to ensure continued
satisfactory emission reductions using this device.
3.2.6.5 SPEED-CONTROLLED EXHAUST GAS RECIRCULATION WITH DISTRIBUTOR
VACUUM ADVANCE DISCONNECT (Contract Report Identification
No. 246)
DESCRIPTION
In the device tested, the exhaust gas recirculation (EGR) was
taken from an exhaust adapter in the exhaust pipe, passed through
an EGR valve, and introduced between the carburetor and the intake
manifold by means of an intake adapter. A more recent modification
by the developer introduces the recirculation exhaust gas through the
positive crankcase ventilation (PCV) system.
A speed control allows about 15% of the exhaust gas to be
recirculated to the intake manifold whenever the vehicle speed exceeds
26 mph, and it shuts off the recirculation whenever the speed drops
below approximately 12 mph. A deceleration switch is also provided to
stop recirculation whpno\/or thp accelerator pedal is released.
-------
3-34
The distributor vacuum advance unit is operated by a solenoid
valve which is interconnected in the vacuum advance tube between the
distributor and the intake manifold, and to the EGR valve. The
vacuum advance operates during exhaust gas recirculation at speeds
above 26 mph, and the vacuum advance is disconnected when exhaust
gas recirculation is terminated by the speed or accelerator pedal
controls.
The speed switch and cable are connected to the speedometer
tap on the transmission, and to the EGR valve. The solenoid valve
is connected to the speed switch electrically.
EVALUATION RESULTS
Pollutant
HC
CO
NOx
Installed
Emission
Percent Reduction
Pooled Mean
12
31
48
Cost: $35 (At the time
Reductions
90% Confidence
Intervals for
Mean % Reduction
3 to 21
6 to 60
43 to 52
of the study the anti
Statistical
Significance
Significant Reduction
Significant Reduction
Significant Reduction
cipated installed
cost was $89; however, since that time the manufacturer has announced that the
device will be made available at a price consistent with an installed cost of$35),
Fuel consumption, as measured during 1972 Federal Emission Test Procedure:
7% improvement, i.e., more miles per gallon. Statistical significance
of change not known.
Safety: There are no apparent safety hazards.
Driveability: Some minor deterioration. Acceptable
Reliability: Estimated about 75,000 miles before total failure.
Installation: Engine should be in good operating condition.
Installation takes about 2.25 hours; requires automotive mechanic, shop
combustion analyzer.
-------
3-35
Maintenance: Cleaning of exhaust gas recirculation valve and
solenoid vacuum valve filter recommended every 6 months/6,000 miles.
This would cost about $7.50.
Applicability to vehicle population: The device would be applicable
to most domestic cars, from the 1955 model year. Results presented
here are for cars without exhaust emission controls. These results
include testing on a VW car since the retrofit was available
for this foreign vehicle.
Durability Results: The device caused the following percentage
emission reductions at the beginning and end of the durability tests:
Zero Miles 25,000 Miles
(1 test without device (3 tests with device
1 test with device) 3 tests without device)
HC 46 -3 (increase)
CO -8 (increase) -55 (increase)
NOx 45 10
The car had choke operating problems which were not caused by
the retrofit . During the mileage accumulation it was necessary
to repair the choke six times. This may explain the bad results
for CO emissions.
Also, the speed control switch for the exhaust gas recirculation
unit of the retrofit had to be replaced twice during the mileage
accumulation; first because it was damaged, along with the speedo-
meter cable, at a car wash facility; but later this replacement became
inoperative and the speed switch had to be replaced again. Inadequate
control switch operation could explain the limited NOx reduction that
was being obtained at the 25,000 mile test point.
-------
3-36
Here too, adequate vehicle maintenance and periodic emission
control system inspection were shown to be necessary to ensure
continued satisfactory emission reductions using this device.
-------
3-37
3.3 FLEET TEST OF GENERAL MOTORS RETROFIT SYSTEM
3.3.1 BACKGROUND
Previous tests by the Environmental Protection Agency usinq a small
number of pre-controlled automobiles had indicated that a retrofit system
developed by the General Motors Corporation appeared capable of achieving
significant reductions in emissions of all three major automotive pollu-
tants (HC, CO, and NOx) at a relatively low cost. Similar systems have
been developed by other U.S. automobile manufacturers. As a result, a
program to obtain more extensive fleet test data for this type of system
evaluated using the Federal emission test procedure was initiated. The
General Motors system was selected for testing because of its general
applicability to domestically manufactured cars, and because of the previous
test experience with it. This evaluation was conducted roughly in parallel
with the survey of retrofit emission control technology discussed in
Section 3.2.
DESCRIPTION OF SYSTEM
This retrofit system consists principally of an exhaust emissions-
oriented engine adjustment procedure. The control specifies the use of
increased idle speed, leaner idle mixture and retarded ignition timing. A
thermostatic vacuum switch provides engine overheating protection by
restoring the normal ignition timing if high coolinq system temperatures
occur.
Installation of the system requires a competent mechnaic. Idle
air-fuel mixture is set at the ratio of 14 to 1. This can be accomplished
accurately using a combustion analyzer available in many shops, or
approximated by using an engine speed drop-off method detailed in the
device installation instructions. Idle speed is increased to 600 rpm to
-------
3-38
smooth the leaner engine operation. Normal manufacturer's recommended
ignition timing is specified and the vacuum timing advance is made inoperative
during normal operation. Through use of a thermostatic vacuum switch,
vacuum advance is restored if the engine collant temperature exceeds 205 F.
SYSTEM APPLICABILITY
The simplicity of the design of this retrofit approach enables almost
universal application to domestically manufactured pre-controlled vehicles
from 1955 through 1967 or to vehicles manufactured for sale in California
through 1965. The system is not applicable to the segment of this population
which utilized a distributor without centrifugal advance. Nor is the
system generally available for imported passenger cars, although its principle
of operation is applicable to many of these cars. The manufacturer does
not recommend application of the device to controlled vehicles.
3.3.2 DESCRIPTION OF TEST PROGRAM
A test program was designed to establish the emission control effective-
ness of this retrofit system. It was also desirable in conjunction with
this evaluation to obtain data on the control potential of standard tune-up
procedures, lean tune procedures, and combinations of these procedures with
the retrofit system. Both tune-up procedures specified the replacement of
spark plugs, spark plug wires, distributor rotor, distributor cap, ignition
points, condenser and air filter element. The procedures differed in
the fact that the lean tune-up set idle air-fuel mixture at a 14 to 1 level,
while the normal tune-up called for a mixture setting to give best minifold
vacuum at idle, effectively a richer fuel mixture. All other tune-up settings
were as specified by the vehicle manufacturers.
-------
3-39
A fleet of 110 automobiles statistically typical of the nationwide
population of pre-emission controlled domestic vehicles was used for the
testing. The sample was limited to model years 1962-1%7 inclusive. The
selection of models was based on model sales figures and included vehicles
with automatic and standard transmissions. Ootional engines for the highest
sales volume models were included. Because of the location of the study,
(Los Angeles) the cars typified maintenance states found in the southwest
United States. The vehicles were obtained through a contractor from pri-
vate individuals to assure a sample of typically operated and maintained
cars. No rental fleets, used car lots, or similar sources were employed.
Cars were rejected from the program that suffered from obvious gross mechani-
cal defects which would jeopardize their probability of completing the
test sequence, thus resulting in a sample somewhat biased toward better-main-
tained vehicles. As required in the emission testing procedures, the vehicles
were checked for leak-free exhaust systems.
All emission tests were conducted according to the 1972 Federal Test
Procedure. The instrumentation employed for the analysis of bag samples
for HC and CO was identical to that specified for certification testing;
Saltzman analysis was used to determine oxides on nitrogen on all tests.
Each vehicle received in the laboratory was subjected to a four phase
test sequence. Initially, "baseline" or "as received" emissions were
determined. This test was followed by installation and adjustment of the
retrofit system and subsequent emission testing. The vehicle was then "tuned,"
the retrofit device detached, and another test was performed. Finally the
retrofit was reactivated and the vehicle was tested in the combination
configuration of "tuned and retrofit."
-------
3-40
Fuel consumption over the LA4 driving cycle was measured during each
test to evaluate inherent benefits or penalties for each control technique.
On the first 25 vehicles driveability effects were qualitatively evaluated
by a short road test following each dynamometer test. Engine overheating
tendencies were monitored during the emission tests on all cars.
3.3.3 RESULTS OF EVALUATION
EMISSION REDUCTION EFFECTIVENESS
Table 3-4 indicates the relative effectiveness of the various con-
figurations compared to the emissions from the "as received" or "baseline"
vehicle. The "tuned with retrofit" configuration showed overall the
greatest control potential, with "retrofit alone" having somewhat lower
effectiveness. Statistically, the hydrocarbon and carbon monoxide levels of
these two configurations are significantly different. The oxides of nitrogen
reductions measured are not statistically different for the two configura-
tions. The emission control achieved by "normal tune-up" is presented in
Table 3-4. The "lean tune-up" is not included because the quantity of
test data was inadequate to predict control effectiveness with a reasonable
level of confidence.
The discussion above relates to the effectiveness of the retrofit
system alone, and in combination with tune-up, in reducing the emissions
of pre-controlled vehicles in typical states of owner maintenance. The
reductions cited would be the appropriate ones to consider in evaluating the
initial effectiveness of a retrofit program not combined with any mandatory
maintenance or inspection/maintenance program. It is also important to con-
sider the additional effectiveness which would be gained by applying this
retrofit system to pre-controlled vehicles already in a tuned-up condition.
This represents the additional effectiveness of adding a retrofit program to
a mandatory maintenance or inspection/maintenance program. The results of
this analysis are presented in Table 3-5.
-------
3-41
Table 3-4
Emission Reduction Data % Reduction
from "as Received" Baseline
Configuration
Number of vehicles
in test sample
Effectiveness
HC-Mean
95% Confidence
CO-Mean
95% Confidence
NO-Mean
95% Confidence
Retrofit
Alone
110
Normal
Tune
85
Tuned
with Retrofit
110
24%
14%
15%
26%
to 29%
16%
to 17%
22%
to 30%
11%
9%
-18%
15%
to 18%
11%
to 13%
-4% inc.
to 10%
33%
17%
13%
34%
to 38%
18%
to 20%
20%
to 27%
Table 3-5
Emission Reduction Data of Tune-up vs.
Tune-up and Retrofit Combined
% Reduction from Tuned Baseline
Configuration
Number of vehicles
in test sample
Effectiveness
(% Reduction from Tuned Baseline)
HC-Mean
95% Confidence
CO-Mean
95% Confidence
NOx-Mean
95% Confidence
Normal Tune and Retrofit
85
25%
22% to 28%
9%
6% to 11%
23%
15% to 29%
-------
3-42
When applied to tuned-up vehicles, the addition of the retrofit
system resulted in significant reductions in hydrocarbon, carbon monoxide,
and oxides of nitrogen levels.
RELIABILITY OF SYSTEM
An investigation into the durability of the system was conducted by
General Motors as part of their development effort. ^ A fleet of eight
vehicles was run 25,000 miles with the retrofit system installed. Mileage
was accumulated according to the AMA durability schedule for emission data.
Deterioration factors developed for this fleet were as follows:
General Motors Retrofit System
Deterioration Factors ^
HC 1.01
CO 1.20
NOx 1.00
These data reflect deterioration factors of the retrofit system alone.
Normal vehicle deterioration has been analytically removed. These factors
were developed from tests using the 1968 7-mode test procedure and may be
somewhat different than if they had been determined using the 1972 Federal
test procedure. It should be kept in mind that these are probably minimum
values as the mileage was accumulated over a very short time rather than
in typical customer usage. The rather large factor for carbon monoxide
means essentially that after 25,000 miles of operation, the retrofit system's
effect on that pollutant was negligible. Thus, a need for periodic readjust-
ment of the carburetor is indicated.
The hardware used in conjunction with the system includes simple standard
automotive parts, and reliability of those components is not anticipated
to present a problem.
-------
3-43
It should be kept in mind that the retrofit system does not aleviate
mechnical and electrical malfuctions that exist prior to installation.
Nor will it protect the engine from normal degradation problems associated
with automobiles. Thus, to ensure continued low emission levels for a
particular vehicle with the system installed, it would be necessary to
follow a normal vehicle maintenance schedule.
COST
While the magnitude of the emission reductions for this retrofit system
is not exceedingly high, the installed cost of the system is low. The
system retails at a cost of less than $10. Labor cost for installation
would be about $10. Thus the total initial cost to the consumer would
be less than $20.
Maintenance specifically related to the system itself would be
limited to a carburetor adjustment at a cost of about $5 per year. A fuel
consumption penalty of about 1% to 2% was associated with the retrofit
system. This would represent an additional cost to the average consumer
of about $3.00 per year.
SAFETY AND DRIVEABILITY EFFECTS
This retrofit system has no known adverse safety effects associated
with its installation. Potential engine overheating tendencies associated
with the vacuum advance disconnect are well guarded against through the
use of the thermostatic vacuum switch.
Engine overheating as a result of very heavy traffic conditions could
temporarily restore vacuum advance on a portion of the retrofitted vehicles.
This could result in decreased emission control effectiveness (principally
for oxides of nitrogen and hydrocarbons) under such driving conditions. No
data are available to indicate whether this effect occurs in actual use.
-------
3-44
The Environmental Protection Agency performed limited driveability
evaluations with the retrofit installed. The results of these tests tended
to confirm results previously reported by the device developer. ^
The device developer reported that cold start and driveaway ratings
of ten vehicles tested showed no real difference between "as received"
and "kit installed" configurations. Warm driveability ratings on 91
vehicles gave mixed results. Of the 91 vehicles, 20 cars demonstrated
worse driveability with the device, 14 were improved, and no effect on the
remaining 57 was noted. Of the 20 vehicles whose driveability deteriorated,
most of the problems were associated with a "stretchy" feel when the throttle
was depressed at light loads.
Thus, it appears that neither safety nor driveability effects result
in serious reservations as to the feasibility of this retrofit system.
-------
3-45
3.4 COST AND EFFECTIVENESS OF RETROFIT STRATEGIES
3.4.1 INTRODUCTION
Sections 3.2 and 3.3 have presented data on the per vehicle emission
reductions and costs associated with applying various retrofit approaches
to pre-controlled light duty vehicles. To the air pollution control
official, however, the subject of principal interest is the overall
effectiveness of retrofit programs in reducing vehicular emissions. The
per vehicle emission reductions and costs are basic data necessary
for evaluation of the effectiveness and cost of programs using such
approaches; and, except for possible effects of altitude on the levels
of emission reductions which are achievable, are generally applicable throughout
the nation. Overall program effectiveness and cost, on the other hand, depend
upon a number of region-specific factors such as the relative numbers and annual
mileages of vehicles of various model years. In addition, the relative
values of various retrofit or other in-use vehicle emission control
programs will depend upon the nature and magnitude of the air pollution
problems to be solved and the time period over which in-use vehicle
emission reductions are required. The result is that decisions on the
relative merits of various in-use vehicle emission control programs
must generally be made after the alternatives have been analyzed
for the specific region under consideration. However, to illustrate some
of the considerations involved, and to provide a perspective on the
overall costs and effectiveness of retrofit programs, this section
presents the results of some typical calculations of overall retrofit
program cost and effectiveness based upon nationwide average vehicle
population data.
-------
3-46
3.4.2 EVALUATION OF OVERALL EFFECTIVENESS AND COST OF RETROFIT STRATEGIES
The principal factors which influence the overall cost and emission
reduction effectiveness of a retrofit program are:
* Time period required to implement the retrofit program.
* Number of vehicles which will be retrofitted under the program.
* Fraction of total vehicular emissions contributed by the vehicles
to be retrofitted.
* Per vehicle emission reductions achievable with the retrofit
approach under consideration.
* Need for periodic inspection and maintenance of retrofitted
vehicles.
* Extent and rate of deterioration of retrofit emission control
performance over time.
* Per vehicle costs assocated with the retrofit approach under
consideration.
* Administrative costs associated with the implementation and
operation of the retrofit program.
* Costs associated with any inspection/maintenance program necessary
for the retrofitted vehicles.
* Rate of retirement of the retrofitted vehicles from the vehicle
population.
-------
3-47
As older vehicles which are potentially subject to retrofit approaches
are replaced by newer vehicles with lower initial emission levels, the
potential effectiveness of a retrofit strategy will decrease substantially.
Thus, the time period over which emission reductions are required plays
a key role in determining the usefulness of a retrofit strategy.
To evaluate the overall effectiveness and cost of retrofit strategies,
the factors mentioned above must be quantified using data specific to the
region for which retrofit is being considered. The key factors which vary
from one region to another are the number of vehicles to be retrofitted,
the fraction of total vehicular emissions emitted by those vehicles as a
function of time, and the rate of retirement of retrofitted vehicles
from the vehicle population; these factors are determined by the
composition of the vehicle population in the region.
3.4.3 COMPARISON OF EFFECTIVENESS AND COST OF SELECTED RETROFIT STRATEGIES
As noted earlier in this section, the calculation and comparison of
figures for the overall emission reduction effectiveness and cost of
retrofit strategies requires the use of data which can vary substantially
from one air quality control region to another. Presented below are the
results of calculations of overall effectiveness and cost for some retrofit
strategies which have been based on typical data. These calculations
illustrate the magnitudes of the emission reductions achievable by certain
retrofit approaches, but the results cannot be applied directly to the
needs of any specific air quality control region without a determination
that the data used are representative of the region under consideration.
It should also be noted that the calculations presented below are for
single retrofit strategies only; no attempt has been made to identify an
-------
3-48
optimum retrofit strategy which might combine the use of various
retrofit approaches on different segments of the in-use vehicle
population.
For the purpose of these calculations, it has been assumed that each
of the retrofit programsj and any adjunct inspection/maintenance program,
would be implemented by January 1, 1975. Vehicle population data used
have been nationwide average values; and an annual linear growth rate
of total vehicle miles travelled has been assumed at five percent.
In strategies employing exhaust emission retrofit, it has been assumed that
the retrofitted vehicles will be subjected to an annual inspection/maintenance
program, since the results presented in section 3.2 and 3.3 indicate that
continued effectiveness of exhaust emission retrofit devices can be assured
only by periodic inspection/maintenance of all exhaust emission related
systems on retrofitted vehicles, including those not specifically included
in the retrofit. The following types of retrofit strategies have been
considered:
CRANKCASE EMISSION RETROFIT
Crankcase emission retrofit of all pre-1968 model year light duty
vehicles.
EVAPORATIVE EMISSION RETROFIT
1. Evaporative emission retrofit of all pre-1971 model year light
duty vehicles.
2. Evaporative emission retrofit of all 1968 through 1970 model
year light duty vehicles.
-------
3-49
EXHAUST EMISSION RETROFIT - PRE-CONTROLLED VEHICLES
1. Exhaust emission retrofit of all pre-1968 model year light duty
vehicles with a maximally effective device and inspection/maintenance
of all retrofitted vehicles.
2. Exhaust emission retrofit of all pre-1968 model year light duty
vehicles with a device which is maximally effective for a specific
pollutant (CO was taken for this example) and inspection/maintenance of all
retrofitted vehicles.
3. Exhaust emission retrofit of all pre-1968 model year light duty vehicles
with a retrofit device which accomplishes smaller than maximum reductions of all
major automotive pollutants, but a relatively low cost, and inspection/maintenance
of retrofitted vehicles.
EXHAUST EMISSION RETROFIT - fWTRm I Fn VFHTf.l FS
1. Exhaust emission retrofit of all 1968 through 1972 model year
light duty vehicles with maximally effective devices and inspection/
maintenance of all retrofitted vehicles.
2. Exhaust emission retrofit of all 1973 and 1974 model year light
duty vehicles with maximally effective devices and inspection/maintenance
of all retrofitted vehicles.
Table 3-6 presents the emission reductions and costs per vehicle used
for these calculations. Emission reductions for exhaust emission retrofit
approaches are presented as reductions relative to a tuned vehicle
baseline since inspection/maintenance of retrofitted vehicles is assumed.
It is assumed that with annual inspection/maintenance of all retrofitted vehicles
the reductions cited in Table 3-6 will be maintained without deterioration.
Cost data are stated as the present cost "in 1972 dollars. In the calculations
of overall program cost, corrections are made for inflation at a five percent
-------
3-50
annual rate and the future costs are discounted to present worth using
and eight percent discount rate.
Emission reductions for crankcase and evaporative emission retrofit
systems are based on the effectiveness of systems now being installed on
new vehicles. Costs for those systems are based on estimates made in
2
EPA's survey of retrofit emission control methods. Emission reduction
and cost data for pre-controlled vehicle exhaust retrofit approaches are
based upon the data presented in sections 3.2 and 3.3. Data for controlled
vehicle exhaust retrofit are extrapolated from the available data for
pre-controlled vehicles. It has been assumed that an oxidizing catalytic
reactor capable of 50% reductions in both HC and CO can be applied to
all 1968 through 1974 model light duty vehicles and that an exhaust gas
recirculation system capable of achieving a 40% reductoion in NOx could
be applied to 1968 through 1972 models. Cost data for controlled vehicle
retrofit are based on those for pre-controlled vehicles. Emission
reduction and cost data for inspection/maintenance are discussed in
Chapter 2 of this document.
-------
3-51
QJ S- 03
> 0 S-
C QJ
S- •— i d
QJ CD
Q.
4-> T3
tO QJ
0 i—
O i—
03
to 4->
c to
i— i O
, —
03
•r—
C
i— i
UJ
— 1
CJ
;c 1/1
LU rs x
>• 03 O
.c z:
Qi X
Li-J I ' I
0-
CO 4->
1— to
CO 13
O 03 O
O -X -C (_>
QJ X
Q i — LU
z: o
.c +j
CO QJ tO
2: > 3
O 03 O
i— i s- -c: 3:
1— QJ X
O Q-UJ
Q to
LU CO)
o: o >
4J 4J
O U 03 O
i—" rs s- z:
CO T3 O
CO QJ Q.
I— l Qi 03
y ^
UJ C LU
O
| •!—
ec to QJ
O to to
I—I •!- (O
Q- E o (_>
1— C
O3
S-
c~
O
03
O
s-
Q.
Q.
e^
i
0
o
o
ft
r—
"^^
LO
OJ
"
o
•i/^-
o
^>
•faO-
^J-5
o
o
^5
0
&s
o
o
0
+J
•1
o
i-
4-*
O)
Qi
CU
t/)
fO
(J
C
(&
^~
c-3
• .,—
•r- E
E
0
CD O
0 0
CD
" r—
,— "^^
•^ o
LO O
CO
• 1 —
4^t -faO-
0 0
r— r—
•fa** -t^V
^5-9 o^
CD O
LO
S-9 S^
CD LO
tQ
^~9 ^^
O O
r^v
o-9 5-^
LO O
CTv
0 0
•!-> S-
•<- QJ O
4- > 4- >
O •••" Q
i- +•>+-> _J
•4-> O -i—
QJ OJ 4- T3
o; 4- o QJ
4- i- i—
QJ LU +-> i —
> QJ O
•i — >•} ry i_
4-) t— 4-^
03 i — +-> C
1- O3 to O
O E 13 0
Q- •(— 03 1
03 X -C QJ
> 03 X S-
1 1 1 ^~ 1 I 1 Q_
E
O
O
o
"
t—
•^^
0
LO
•
o
-t/^-
CD
UD
-t/>
c^9
O
O
UD
o~9
O
CM
3-9
O
O
I *
to >
^ Q
-C
X T3
LU O)
S- •—
(J O t—
•i- 4- O
4- S.
•r~ 40 4_>
0 -i- C
QJ 4- O
O. O 0
CO S- 1
4-> QJ
O QJ S-
o D; o-
•1 —
E
o
o
CD
*•
1 —
"• — ^
0
o
•
o
•*/}•
o
CM
CO
o-9
LO
CM
O
i —
o~9
LO
CM
3-9
O
0
^>
Q
to
13 T3
03 QJ
-C i. i —
X O •—
LU 4- O
S-
4_> 40 4->
(/}•!- C
O 4- O
c_) o o
i- 1
S 4-> QJ
O QJ S-
_1 C£ Q-
•r—
E
CD
O
CD
"
r^
~-^
O
o
•
CM
•f>
O
0
00
•fao-
3-9
O
^j-
o
LO
3-5
O
LO
3-9
O
o
S-
QJ O
> 4- >•
• r— t 1
O •!-
QJ 4- OJ
4- 0 (^
4- S- CTi
LU +J i —
QJ
>^ C£ ^
i— S-
i— 40 c~
03 to 4->
E =
•i- 03 00
X ^ <~O
03 X O1
s: LU ^
E
o
o
o
r.
i
*" — ^
0
o
•
f—
-t/>
o
VO
^—
v=>
^5
0
o
LO
CJ^i
o
LO
o-9
O
O
i_
QJ O
> 4-
•r— ^*
4-> 4-> O
O -i- 1
QJ 4-
4- O •=?•
4- S- 1^.
LU 4-> CT>
QJ r—
>^ cy
r— -0
i— 4-> C
03 to 03
E ^
•r- 03 CO
x -C r-~
03 X CTi
s: LU i —
s-
03
QJ
~^
O
CD
•
^j-
co
<^
~^f
o-9
CD
0 ^
|— ^
"QJ
CO
fO
^~5 ---t
^ ^
3
•*-" (/,
i_
03
O ^
3-S "^ O
0 QJ ~°
. OJ
+J ^~
"3 , —
'QJ c
^~ -r—
C1J
f 4-^
s^ i; to
0 n 0
to <->
O "^
« S
QJ -g tt
o r
c to
03 (— 03
^ ^~
QJ .° T3
"^ to ^
•i — . tO
03 E QJ
^~ ^ ^_
^-, ("i
c~ l ^ x
O tjn ^
•i — .ij
4_> ~~ 40
o 2 to
QJ "t? O
O- LU ^
to
C j, +:
i— i ^ -K
-------
3-52
Table 3-7 presents a comparison of the overall costs and effectiveness of
the various retrofit strategies considered. The cost figures are total cost
for each retrofit strategy over the time period 1975 through 1979 expressed in
1972 dollars, annualized, and distributed over the entire 1975 light duty
vehicle population. Included are the costs for: purchase and installation of
retrofit devices; regular maintenance of the retrofit systems; annual emission
inspection of the retrofitted vehicles and; maintenance required by the inspection/
maintenance program beyond that normally performed voluntarily. Administrative
costs for implementing the retrofit programs have not been specifically included
but, in the case of the exhaust emission retrofit strategies, these costs are'
probably covered by the costs for administration of the inspection/maintenance
program included in the assumed inspection cost. Fuel consumption savings
or penalties resulting from the programs have not been included because of
uncertainty in such figures.
Emission reduction effectiveness of each of the strategies is presented
for 1975, 1977 and 1980 as the projected light duty vehicle emissions in those
years if the strategy were implemented, normalized with respect to light
duty vehicle emissions in 1967. For comparsion, the projected emissions
without any in-use vehicle control program are also presented for each year.
The results presented in Table 3-7 lead to the following conclusions:
* The costs and emission reduction effectiveness of various retrofit
strategies vary substantially. The choice of a "best" retrofit
approach must be tailored to the specific pollutants whose emissions must
be reduced, the magnitude of the reductions required, and the time
period over which the reductions are needed.
-------
3-53
* Programs for retrofitting pre-controlled light duty vehicles could
have a significant influence on light duty vehicle emissions in 1975
and 1977 but would generally have only minimal effects in 1980 and beyond.
* Programs for retrofitting more effective exhaust emission controls
to pre-1975 model year controlled light duty vehicles could significantly
decrease light duty vehicle emissions even beyond 1980.
* Retrofitting of crankcase emission control systems to vehicles not
originally equipped with such devices would have a minimal effect on
light duty vehicle emissions in the 1975 to 1980 time period.
* Retrofitting evaporative emission control systems to light duty
vehicles not originally so equipped could bring about some decrease in HC
emissions during the 1975 to 1980 time period but does not appear
cost-effective relative to retrofitting exhaust emission controls
to the same vehicles.
Once again, it must be emphasized that these conclusions are for a
light duty vehicle population similar in age and mileage distribution to
the nationwide average. Markedly different age and mileage distributions
could alter these results.
-------
O
m
O
o
r-j
o
3-54
o
o
O
O
H
O
O r- o
•rH 01 CJ
CO rH
CO
•H
E
« a
u
x:
QJ
X
u 9
a z
jj
x: m
bC r-. o
•H Oi CJ
hJ rH
a
4J
W-Ji
o-;;
o cj i1
CO rH
0"! rO O
i-H CU 'H
1 N &
m ri cu
r-- rH >
CTN CD
£ OJ
C &
<
W
CU
jj
CD
S-i
jj
C/)
JJ
•H
M-l
O
1-1
JJ
QJ
5
O
CO
o
r--.
°^
o
vO
O
^_l
"^
o
o
B
CD
SH
bO
o
SH
cu
i-H
o
S-l
£
0
u
C
o
CO
CO
•rj
E
CD
QJ
i-H
o
•rl
CD
CU
CO
3
1
C
•rl
o
2;
5
o
o-i
o
r~-
°^
0
CO
MD
o
0
in
o
CN
JJ
bO
•H
i-H
S-i
CD
OJ
i-H
QJ
TJ
Q
E
CO
1
QJ
SH
a
I-H
CD
MH
o
JJ
•H
MH
o
JJ
CD
S-f
o
• r-l
CO
(fl
••H CO
E QJ
CU i-H
o
QJ -rl
co x:
CO 0)
0 >
CO -M
CJ TJ
5
o
CN
^
o
r~.
°^
o
^o
o
^
**
o
, — 1
S-i
CD
CD
CD
a.
OJ
ex
rH
rH
CD
MH
JJ
o
4-1
QJ
C
O CO
•H QJ
CO rH
CO O
•rl -rl
QJ QJ
QJ
> t*.
•H JJ
JJ 3
CD TJ
0 JJ
CO 6£>
> -H
5
o
^
o
r^.
°l
o
"O
^o
o
r-
-^
o
CO
r-H
CD
o
r--
r— 1
bo
o
oo
r-H
i-H
CO
M-l
JJ
•H
MH
0
JJ CO
CD QJ
S-1 rH
a
C -H
o x:
•H QJ
CO >
CO
B J-1
QJ 3
TJ
CD
&• JJ
•H x:
jj bO
CD 'rl
o
f> QJ
W >>
cr.
CO
o
o
CO
o
vD
00
0
CO
•st
o
0
i-H rH >
CO
CO SH CD
CD O 4J
J>,MH JJ
•H
i— 1 QJ M-*
QJ CJ O
B O) CD
TJ »H
00
v£> QJ M-l
& > 0
rH T-1
1 JJ QJ
QJ U O
SH CD C
&MH §
rH OJ QJ
CO >i £
rH -rl
O CD p
E -^
JJ -H C
•H X 0
MH CD -rl
o E J-1
IH CJ
JJ CD 0)
QJ CU
j-i jH co
jH C
pj >rH -r-l
o s
•H r(3
co co C
CO CD CD
•rl rH
B o to
CD -rl JJ
CO > JJ
3 3
CO >, rH
H TJ P-
O
O
CO
O
r-
cr-
0
cr.
O
^
<^
CD
^j-
•co-
jj en
x: o cu
bO CJ rH
•rl U
rH SH -H
0 J3
S-J MH CU
CD >
QJ CU
•rl QJ
i— 1 JJ JJ
CD U JJ
TJ QJ *rl
O MH MH
S M-I o
QJ in
OO JJ
\C >, QJ
CT* rH J-.
i-H rH
1 CD MH
CD CJ O
M -rl
d MH QJ
•r-l O
r-H O C
CO CU C
cn QJ
MH jj
O CD C
CJ -H
MH QJ ^.
VH 0
JJ CO -rl
CD 4J
JJ CD
O S CO
•r-t a
CO CO -H
CO QJ
•H rH TJ
B U C!
CD -rl CO
x:
•U QJ i— 1
W > 0
CD >, JJ
x: JJ cf
X 3 o
tJ TJ U
m
o"
CO
o
CM
CTv
O
in
c>
^
**
o
o
in
i — i
rH CO
CD 01
jj bo cj
x: C -H
•rl rH CD
rH rH >
o
J-i SH TJ
CD JJ CD
CD C JJ
>> O JJ
O -rl
Ol ^ O
TJ rH SH
O rH JJ
E CD CD
•rl SH
00 JJ
v£J U MH
CTi co O
rH CU
1 CD
CD CD CJ
JH 0 C
CU -H CO
> d
i-H CD CD
co C
JJ -iH
MH W CD
o o E
a ~^s.
J-J C
•rl & O
MH o *H
O rH JJ
SH O
JJ CD QJ
QJ p,
SH X! W
JJ C
C 'rl -rl
O &
-rl TJ
cn co C
CO QJ C0
•r-l rH
§O CO
•H JJ
JJ CD CD
cn > jj
3 3
CD >1 rH
X: JJ r-l
X 3 O
M TJ CU
CM
CO
O
vO
CM
O
1^.
^
O
^>
o
CO
CO
o*
CN
S-i
CO
QJ
rH TJ
U C
*O CD
B to
QJ
CM O
rH U
O -rl -r-l
JH JJ ^
x: o QJ
JJ 01 >
CO MH TJ
O\ JJ
rH >i JJ
i-H -H
r-H r-H MH
CD B ^
•rH JJ
MH X QJ
°g^
jj MH
•rl J^ O
MH JJ
O -H CiJ
in & O
JJ C
CD CO CO
rH QJ (3
rH 0)
C CJ JJ
O -H C!
•H x: i-i
0) CD CD
cn > B
-H "*s.
CD JJ O
3 -rl
JJ TJ JJ
cn o
3 jj a>
CD x! cu
J3 bO W
X i-l C
td I-H -r-1
OO
O
CO
CM
o"
1^,
cn
O
o
LO
o
CN
-*
o
o
in
CO
}H
CD c:
^i CD
T-H cn
CD CD
TJ 0
O -H
O>
stf TJ CO
r^ a)
(T. CD rH
rH p> CJ
TJ jj x:
COO)
CD CD >
M-f
CO MH t)
r^ QJ cu
Oi JJ
t-H ^ JJ
r-H -H
r-H rH MH
CD 1 M
•H JJ
MH X CU
O m (-*
JJ MH
•rl X! O
MH JJ
O -rl QJ
JJ ? C
a; w aj
SH 0) C!
rH CU
C CJ 4J
O -rl C
•rl X! 'H
CO CU CQ
w > B
•H "^
QJ JJ O
3 -rl
JJ 13 JJ
CO CJ
3 4J CU
CD J3 O-
X! OO co
W rH -rl
3 3
O TJ
JJ
co x:
C bO
O -H
•H rH
bO
QJ in
SH r-
o
rH rH
0
l-i QJ
C -rl
O JJ
CD
JJ SH
•H CD
rH >
to O
3
a
•H 3
CD xi
•rH
•H JJ
MH w
CJ TJ
CU
CUT)
co C
CD
0 TJ
MH QJ
N
cn -H
JJ rH
rH CQ
3 3
cn C
QJ C
SH CD
1 CO
co CD
JJ rH
CO rH
TJ 0
T)
c
O CN
JJ C^
CO rH
i-H
3 a
ft-H
0
CUTJ
O
CD -rl
i-H J-l
CJ CD
•rH CU
•rl
QJ JJ
bO
CD CTi
JH fs.
CJ O~i
> ^
x:
QJ bO
TJ 3
•H 0
c x:
0 4J
•H
jj m
CD fs.
r-H
00
C! rH
•H O
cn MH
3
jj
TJ cn
CD O
JJ CJ
CD
r-H r-H
3 co
0 JJ
i-H O
co H
•x *
-------
3-55
3.5 CONCLUSIONS
Retrofit Device Feasibility
Several exhaust emission control approaches appear feasible for
retrofit to pre-1968 (pre-1966 in California) model year light duty
vehicles. These approaches include air bleed to intake manifold,
catalytic converter with vacuum spark advance disconnect, exhaust gas
recirculation with vacuum spark advance disconnect, and lean idle
adjustment with vacuum spark advance disconnect. Other exhaust emission
control retrofit approaches may also be applicable to pre-controlled
vehicles but have not been tested as extensively as the above.
Retrofit application of some of the above approaches to controlled
(post-1967 models nationwide, post 1965 in California) light duty vehicles
may also be feasible but empirical test data are very limited. Based
upon emission control systems being developed for new vehicles, catalytic
converter and exhaust gas recirculation approaches appear likely to
be applicable to controlled vehicles. As retrofits, the catalytic
converter would be potentially applicable to light duty vehicles through
the 1974 model year and exhaust gas recirculation potentially applicable
through the 1972 model year (except for a limited number of earlier
models already employing this technique). Other retrofit approaches
may also be applicable to controlled vehicles but cannot be considered
feasible without further evaluation.
Exhaust emission control retrofit programs should not be implemented
for either pre-controlled or controlled light duty vehicles except in
conjunction with an inspection/maintenance program which will assure
continued satisfactory operation of emission-related systems of the
retrofitted vehicles.
-------
3-56
Retrofit of crankcase emission control systems to light duty
vehicles not already equipped with such systems is considered feasible.
Retrofit of evaporative emission control to light duty vehicles not
equipped with such control systems is potentially feasible but no
evaporative emission control retrofit systems are currently available.
Retrofit Device Effectiveness and Cost
Retrofit exhaust emission control systems appear capable of
accomplishing reductions of up to 70% in HC emissions, 65% in CO
emissions, and 50% in NOX emissions from pre-controlled light duty
vehicles. These maximum reductions can be achieved simultaneously
through use of a catalytic converter with vacuum spark advance disconnect.
Similar CO and NOx reductions can be achieved separately by air bleed to
intake manifold and exhaust gas recirculation with vacuum spark advance
disconnect retrofits respectively. No other retrofit system (except gaseous
fuel conversion) is known to rival the HC emission reductions achievable by
the catalytic converter with vacuum spark advance disconnect.
Retrofit crankcase emission control systems could be expected to
control essentially 100% of crankcase HC emissions. Based upon the
effectiveness of evaporative emission control systems now being installed
in new vehicles, retrofitting of similar systems to vehicles not
originally equipped with evaporative emission controls could be expected
to control approximately 95% of the evaporative HC emissions from those
vehicles as measured by current test procedures.
Empirical test data on the effectiveness of retrofit approaches in
reducing exhaust emissions from controlled light duty vehicles are very
limited. However, based upon the effectiveness of those approaches for
pre-controlled light duty vehicles and considering the nature of the
-------
o-57
emission control approaches already employed in controlled vehicles,
reductions of approximately 50% in HC and CO from 1968 through 1974
model year light duty vehicles through retrofit of an oxidizing
catalytic converter, and reductions of approximately 40% in NOX
from 1968 through 1972 model year light duty vehicles through retrofit
of exhaust gas recirculation, appear possible.
Costs of retrofit emission control devices vary substantially,
most falling within the range of $20 to $175 installed. In general,
devices achieving larger reductions and capable of controlling larger
numbers of pollutants effectively are among the more expensive. However,
for pre-controlled vehicles devices are available which can achieve
close to maximum reductions of either CO or NO/ singly at moderate or
intermediate costs.
Retrofit Strategy Effectiveness and Cost
In general, the relative emission reduction effectiveness and
costs of alternative retrofit strategies must be evaluated in terms of
the specific region for which retrofit is being considered, since
vehicle age and mileage distributions vary significantly throughout
the nation. Using nationwide average vehicle population data as a typical
case, it has been found that significant reductions in total light
duty vehicle emissions can be achieved during the period 1975 to 1980
through retrofit of exhaust emission controls to pre-1968 model year
light duty vehicles and even beyond 1980 if more effective exhaust
emission controls are retrofitted to pre-1975 model year controlled
light duty vehicles. Retrofit of crankcase emission controls to vehicles
not so equipped appears to have a negligible effect beyond 1975.
-------
3-58
Retrofit of evaporative emission controls to pre-1971 model year
light duty vehicles could significantly reduce HC emissions during
the 1975-1980 time period but appears substantially less cost-effective
than exhaust emission retrofit.
REFERENCES - CHAPTER 3
1. "Field Demonstration of General Motors Environmental Improvement
Proposal (EIP) - A Retrofit Kit for GMC City Busses," Interim Report
prepared under Environmental Protection Agency Contract No. PH-22-68-23
by Southwest Research Institute, June 1971.
2. "Analysis of Effectiveness and Costs of Retrofit Emission Control
Systems for Used Motor Vehicles," Final Report prepared under
Environmental Protection Agency Contract No. 68-04-0038 by Olson
Laboratories, Inc., in association with Northrop Corporation,
May 1972.
3. Federal Register, Volume 35, Number 219, Part II, Tuesday, November 10,
1970, "Control of Air Pollution from New Motor Vehicles and New
Motor Vehicle Engines."
4. SAE Paper #710069, "Exhaust Emission Control for Used Cars," G.W.
Niepoth, G.P. Ransom, J.H. Currie, International Automotive Engineering
Congress, January 11-15, 1971.
-------
4-1
Chapter 4 CONVERSION OF IN-USE VEHICLES FOR GASEOUS FUEL OPERATION
4.1 INTRODUCTION
The modification of in-use vehicles to permit their operation using
gaseous fuels falls within the general definition of retrofit approaches
discussed in Chapter 3 of this document. However, the feasibility
of gaseous fuel conversion as a cost-effective in-use vehicle emission
control strategy depends not only on the availability, emission reduction
effectiveness, and cost of the retrofit hardware, but also on a number of
other factors, such as the feasibility of providing the necessary fuel
distribution system, the availability of adequate fuel supplies, and the
impact of fuel diversion from other combustion sources. Therefore, the
conversion of in-use vehicles for operation on gaseous fuels is dealt with
separately in this chapter.
TYPES OF GASEOUS FUEL CONVERSIONS
There are three basic types of gaseous fuel conversions that may be
performed which differ according to fuel type. These are: Liquified Petroleum
Gas (LPG), Compressed Natural Gas (CNG), and Liquified Natural Gas (CNG).
Differences among the types manifest themselves principally in
such areas as the type of conversion hardware required, conversion
cost, and operating convenience. In addition to these three
basic types of conversions, a distinction can be made
between single-fuel conversions, in which the vehicle is modified
to operate exclusively on the gaseous fuel, and dual-fuel conversions in
which the modified vehicle is equipped to operate interchangably on either
the gaseous fuel or gasoline.
-------
4-2
APPLICABILITY OF GASEOUS FUEL CONVERSION
Conversion for use of gaseous fuel can be performed for both light
and heavy duty gasoline-powered vehicles. However, such conversions have
generally been performed for reasons other than emission reduction (usually
reduced operating or maintenance cost) and data evaluating the emission
reduction capability of such conversions are available only for light
duty vehicles. As a result, the emphasis in this chapter is placed on
light duty vehicle conversions.
4.2 SUMMARY OF GASEOUS FUEL CONVERSION TECHNOLOGY
4.2.1 DESCRIPTION OF GASEOUS FUEL SYSTEMS
Conversion of a vehicle for operation on a gaseous fuel requires the
installation of a new fuel tank designed to contain the fuel under pressure,
the installation of new fuel lines and appropriate control valves to ensure
that the fuel reaches the engine only when it is running, and the installation
of a new carburetor designed to meter the flow of the gaseous fuel into
the engine's induction system. Conversions for LPG and LNG also require
the installation of liquid-gas converters which vaporize the fuel;
and conversions for CNG and LNG require a series of pressure regulators
to reduce the high pressure at the fuel tank to a pressure suitable for
the carburertor. In single-fuel systems, the above components replace
the original fuel tank, fuel lines, fuel pump, and carburetor. In dual-
fuel systems, the new components must be added to the original comoonents,
since those are required when the vehicle is operated with gasoline fuel.
In addition to the new equipment which must be retrofitted to an
in-use vehicle to permit it to use gaseous fuel, certain engine adjustments
(such as ignition timing) may be required to achieve proper operation with
-------
4-3
the new fuel. Optimization of engine adjustments is necessary if
maximum emission reductions are to be achieved. For vehicles not originally
equipped with hardened valve seats, installation of such seats may be
necessary to prevent excessive wear with certain types of vehicle useage.
A number of systems for conversion of light duty vehicles to gaseous
fuel operation are currently available. It has been estimated that several
hundred thousand vehicles are currently being operated in the United States
using LPG and that several thousand vehicles using natural gas, mostly CNG,
are being operated experimentally.
4.2.2 EMISSION REDUCTIONS ATTAINABLE THROUGH GASEOUS FUEL CONVERSION
While a large number of vehicles have been converted for operation
on gaseous fuels, evaluations of the emission reductions obtained through
those conversions have been quite limited; particularly limited are
test data using the 1972 or 1975 Federal certification test procedures.
Table 4-1 summarizes some of the available test results. It can be
seen that substantial reductions in emissions of all three major automotive
pollutants are achievable. For example, initial reductions of approximately
80% in exhaust HC and CO emissions and 60% in NOx emissions appear to be
achievable concurrently when 1970 and earlier model year controlled light
duty vehicles are converted for gaseous fuel operation and optimized with
respect to ignition timing and air/fuel ratio. Alternatively, it can be
stated that exhaust emission levels of approximately 0.6 grams per mile HC,
3 grams per mile CO, and 3 grams per mile NOx, as measured by the 1972
Federal test procedure, are achievable with existing gaseous fuel conversion
systems. However, it must also be noted that in a number of cases either
very small reductions or even large increases in emissions resulted when
vehicles were converted for gaseous fuel operation.
-------
4-4
TABLE 4-1
Typical Emission Reductions Through Gaseous Fuel Conversion
1972 FTP Emissions
(grams per mile)
Vehicle HC** CO NOx Type Gonver.-
Converted 1968 Buick 350 3.5 4.7 8.9 LPG
Stock 1968 Buick 350 1.9 29.6 4.0
Percent Reduction (84)* 84 (123)*
Converted 1969 Ford 351 3.1 7.3 8.6 LPG dual fuel
Stock 1969 Ford 351 7.4 17.8 5.2
Percent Reduction 58 59 (65)*
Converted 1968 Ford 302 2.4 4.2 1.8 LPG dual fuel
Stock 1968 Ford 302 3.1 28.5 3.6
Percent Reduction 23 85 50
4 Converted 1969 Chrysler 318's 2.4 7.2 2.9 LPG dual fut-1
Stock 1969 Chrysler 318 3.4 30.5 3.6
Percent Reduction 29 76 19
2 Converted Rambler 343's 3.0 15.4 2.6 LPG dual fuel
Stock 1969 Rambler 343 3.0 31.5 3.1
Percent Reduction 0 51 16
Converted 1969 Ford 429 1.3 4.0 1.9 LPG
10 Converted 1970 Ford 250's 0.69 1.8 2.6 LPG
10 Stock 1970 Ford 250's 3.70 16.0 9.4
Percent Reduction 81 89" 72
10 Converted 1970 Rebel 232's .51 3.9 3.1 LPG
10 Stock 1970 Rebel 232's 2.7 22.1 6.9
Percent Reduction 81 82 55
-------
4-5
TABLE 4-1 (Cont.)
TYPICAL EMISSION REDUCTIONS THROUGH GASEOUS FUEL CONVERSIONS
1970 FTP Emissions
(grams per mile)
Vehicle
2 Converted 1968 Chevrolet 230
2 Stock 1968 Chevrolet 230 's
Percent Reduction
2 Converted 1968 Ford 250's
2 Stock 1969 Ford 250's
Percent Reduction
10 California State Cars
10 California State Cars
Same Cars on Gasoline
Percent Reduction
5 Los Angeles City Cars
Same Cars on Gasoline
Percent Reduction
HC** CO
s 1.1 9.5
5.7 58.2
70 84
NOx
0.9
2.6
65
1.5
3.1
52
1.4
2.9
52
7.8
25.3
69
1.5 10.5
6.7
42.9
84
5.0
31.0
84
1.4
1.2
3.2
63
3.0
3.5
14
Type Conversion
CNG dual fuel
CNG dual fuel
LPG
CNG dual fuel
CNG dual fuel
*
**
Figures in parentheses () reflect increases in emissions.
Although it is generally agreed that hydrocarbon emissions
from gaseous fueled vehicles are less photochemically reactive
than those from gasoline fueled vehicles, a Federal reactivity
scale has not been defined which would' allow quantitative
correction for this factor. Therefore, all hydrocarbon values
are reported on the same mass basis as gasoline.
-------
4-6
In general, it is to be expected that dual-fuel conversions will not be capable
of achieving as large emission reductions as are possible using single-fuel con-
versions, since compromises are required in various engine adjustments to permit
interchangable operation using both the gaseous fuel and gasoline.
Data on the deterioration of emission performance of vehicles converted for
gaseous fuel operation, as those vehicles accumulate time and mileage, are extremely
limited. However, it appears reasonable to assume that reductions comparable to
those cited above should be maintained if appropriate vehicle maintenance is practiced,
4.2.3 COSTS OF GASEOUS FUEL CONVERSION
Typical costs for gaseous fuel conversion hardware, excluding the
costs of fuel tanks and labor for the conversions, have been estimated ^
to be approximately $300 for LPG and CNG and $350 for LNG. The cost of
fuel tanks varies substantially with their size, which is determined by
the desired operating range of the converted vehicle between fuelings. The
same sources estimate fuel tank costs within the ranges $100 to $200 for LPG,
$60 to $200 for CNG, and $400 or more for LNG, depending upon size. The same sources
estimate installation labor requirements ranging from 4 hours to 12 hours, or
approximately $50 to $150. Thus, minimum conversion costs per vehicle, including
fuel tanks and labor, can be estimated at approximately $450 for LPG, $410 for
CNG, and $800 for LNG.
In addition to the cost of conversion hardware, several other factors
may play an important role in determining the cost of a gaseous fuel
conversion strategy. These are cost of refueling facilities, fuel cost, and
maintenance costs. The cost of refueling facilities depends upon the gaseous
fuel used; the number of vehicles to be serviced by the facility, and the
fueling speed which is desired. Fuel costs vary substantially throughout the
-------
4-7
country and must be assessed for the specific area in which gaseous fuel
conversion is contemplated. While experience with gaseous fuel conversions
generally shows a reduction in vehicle maintenance costs for such items as
oil and filter changes and spark plug replacement, available data on the
savings in maintenance costs achieved by gaseous fuel conversion vary
widely. 2,3
4.2.4 OTHER ASPECTS OF GASEOUS FUEL CONVERSION
Other factors which influence the feasibility of gaseous fuel conversion
approaches are effects on vehicle performance and safety considerations. In
general, experience with gaseous fuel conversion fleets has demonstrated that
acceptable vehicle performance (though often reduced from that obtained with
gasoline) can be obtained with gaseous fuels. While safely considerations
have restricted the use of gaseous fuels in some areas, enough experience
has been accumulated with gaseous fuel vehicles to demonstrate that under
closely controlled fleet operation the fuels can be used safely. Both
of these aspects of gaseous fuel conversion are discussed in other
1 ,3,4
documents.
4.3 USEFULNESS OF GASEOUS FUEL CONVERSION AS AN IN-USE VEHICLE EMISSION
CONTROL STRATEGY
In judging the usefulness of gaseous fuel conversion as an emission
control strategy for in-use vehicles, the following factors must be
considered: reductions in total vehicle population emissions achieveable
as a function of time by the conversion strategy; cost of implementing the
strategy; and impact of the diversion of gaseous fuels for vehicle use on
other fuel requirements, including possible increases in emissions from
stationary sources through fuel switching.
-------
4-8
Considerations of the cost of gaseous fuel conversions, the limited
availability and cost of refueling facilities, and limitations on the
availability of gaseous fuels in many areas have generally led investigators
to the conclusion that gaseous fuel conversion is most practical when applied
34 5
to controlled fleets of vehicles. ' A recent paper has estimated the
overall effectiveness and cost of fleet conversion programs for two cases of
some interest: taxicab fleets in New York City and in Washington, D.C.
The results of that analysis indicate that emission reductions expected to
result from converting the pre-1975 model year taxicab fleets in those two
cities to gaseous fuels would be relatively small compared to the
potential of other emission control strategies for in-use vehicles and
would be very short lived.
1 4
In another analysis ' , the overall emissions impact of diverting
natural gas from electric power generation to powering commercial vehicles
in New York City has been considered. It was determined that to operate
all commerical vehicles in New York City on natural gas would require the
diversion of approximately one-half of the quantity of that fuel currently
used for electric power generation. Based upon emission rates typical of the
1968 to 1970 time period, and assuming that the natural gas diverted
from power generation would be replaced by 1% sulfur fuel oil, it was
estimated that such a control strategy could result in net reductions in HC
and CO emissions equivalent to approximately 15% to 25% of the emissions
of those pollutants from all motor vehicles in New York City. However,
it was also found that a concurrent increase of emissions of sulfur oxides
equivalent to approximately 6% of the total New York City emissions of that
pollutant would result from implementing such a control strategy. This is
-------
4-9
an example of the type of pollutant control trade-offs which need to be
evaluated in regions where supplies of gaseous fuels are limited.
4.4 CONCLUSIONS
The conversion of fleet-operated vehicles for operation using gaseous
fuels is feasible with currently available technology. If properly
optimized to minimize emissions, such conversions can achieve larger
per-vehicle emission reductions than other, more generally applicable,
retrofit approaches. The initial installed cost for gaseous fuel
conversions is quite high relative to other retrofit approaches but in
some cases this high initial cost can be substantially offset through
savings in fuel and maintenance costs.
The usefulness of conversion of fleet vehicles for gaseous fuel
operation as an in-use vehicle emission control strategy must be evaluated
specifically for each region considering the approach since the
fraction of the total vehicle population converted, fuel availability,
fuel cost, and the impact of diversion of gaseous fuels from other air
pollution sources will depend upon the region considered. However, the
generally small portion of the total vehicle population which is
fleet-operated, and the generally rapid replacement of fleet vehicles by
newer models is likely to result in relatively small and short-lived
emission reductions being achievable through this approach.
REFERENCES - CHAPTER 4
1. Emission Reduction Using Gaseous Fuels for Vehicular Propulsion , Institute
of Gas Technology, Chicago, Illinois, June 1971 (Prepared under Contract No. 70-69
for the Environmental Protection Agency).
-------
4-10
2. Analysis of Effectiveness and Costs of Retrofit Emission Control Systems for
Used Motor Vehicles, Olson Laboratories, Inc., Anaheim, California, May 1972
(Prepared under Contract No. 68-04-0038 for the Environmental Protection
Agency).
3. Gas Power: The Fleet Owner's Gaseous Fuel Manual , California Institute of
Technology, Pasadena, California, March 1972.
4. Position Paper: Conversion of Motor Vehicles to Gaseous Fuel to Reduce
Air Pollution, Office of Air Program, Environmental Protection Agency,
May 1972.
5. The Effectiveness and Cost of Coversion of Fleet Vehicles to Gaseous
Fuel for Reducing Automobile Emissions in Selected Regions , Susan F. Mickey
and Joel Horowitz, Environmental Protection Agency, Washington, D.C.,
November 1972.
------- |