EPA-AA-TSS-83-10
Anti-Tampering and Anti-Misfueling
Programs To Reduce
In-Use Emissions from Motor Vehicles
December 31, 1983
Office of Mobile Sources
Office of Air and Radiation
U. S. Environmental Protection Agency
-------
TABLE OF CONTENTS
Section Page
1.0 INTRODUCTION 1
2.0 TAMPERING AND MISFUELING RATES 5
2.1 Current Rates 5
2.2 Future Rates 13
3.0 EFFECTS OF TAMPERING AND MISFUELING
AND COSTS OF REPAIRS 26
3.1 Air Pump 26
3.2 Catalyst 29
3.3 Habitual Misfueling 30
3.4 Positive Crankcase Ventilation 33
3.5 Evaporative Canister 35
3.6 Light-Duty Truck Effects 36
3.7 Oxygen Sensor Check 37"
4.0 METHOD FOR CALCULATION OF HC"
BENEFITS OF ANTI-TAMPERING ANDAHTT—
MISFUELING PROGRAMS 4-5 .
4.1 Discussion of Method 45
4.2 Example Calculation 50:
4.3 Excess Emissions Due to Tampering and
Misfueling: All Types 60
5.0 BENEFITS OF ANTI-TAMPERING AND
ANTI-MISFUELING PROGRAMS 65
5.1 l/M Programs 67
5.2 Periodic Inspection Programs 87
5.3 Other Anti-Tampering and Anti-
Misfueling Programs 93
6.0 ADJUSTMENTS TO LOCAL CONDITIONS: USE
OF MOBILE3 TO CALCULATE PROGRAM BENEFITS 113
REFERENCES 116
APPENDIX: DISABLEMENT DATA BASE
-------
1.0 INTRODUCTION
Since the 1960's when crankcase ventilation tubes on
automobile engines were rerouted to prevent the venting of
engine blowby gases directly into the atmosphere, automotive
designers have added to and redesigned various components of
the standard internal combustion engine to reduce its
emissions of hydrocarbons (HC), carbon monoxide (CO), and
nitrogen oxides (NOx). The success of their efforts is
evident in the fact that new passenger vehicles emit only a
small fraction of the HC, CO, and NOx emissions of
pre-controlled cars.
The full benefit of these modifications, however, is not
being realized in the field. EPA studies have shown
repeatedly that maladjustments, disablements, and component
failures in the emission control systems of automobiles occur
frequently and that the result is often emission levels many
times the design (certification) standards. This means that
the vehicle owners, who have paid for these emission control
components when their cars were purchased, and the public, in
general, have not been receiving the emission benefits of
this investment because of some form of tampering,
misfueling, malmaintenance or neglect. ..These emissions in
excess of design standards are a ma jor :-srtgrce . of HC.*,. CO;;, and
NOx from mobile sources and a signific ant^corrtritaut-ing.-^ fact or.
to air pollution in urban areas.
This report will specifically address the portion ofi. excess:
vehicle emissions due to tampering and mi sftieilng^
Tampering, in this report, will refer to any disablemenfcraf:;
any component of an emission control system whether it was
done deliberately, inadvertently, or through neglect.
Tampering can be as simple as losing (and not replacing) a
vehicle's gas cap or as deliberate as sawing off the
catalytic converter. This definition does not include
maladjustments which would increase emissions.
Misfueling and fuel switching in this report will mean any
introduction of leaded fuel into a vehicle originally
equipped with a catalytic converter. This can be done
deliberately by the vehicle owner by enlarging the fuel inlet
restrictor so that a leaded fuel nozzle fits, by fueling from
a leaded gasoline pump with an undersized nozzle, or by using
a funnel so that damaging the fuel inlet restrictor is not
necessary. The majority of fuel switching currently occurs
via tampered inlets. Fuel switching can also be done
inadvertently if the unleaded fuel supply at a particular
station or at a wholesale supplier becomes contaminated or
deliberately switched with leaded gasoline. Although EPA
estimates, however, that the nationwide contamination
-------
-2-
violation rate at retail gasoline stations is less than one
percent. There are many possible reasons why people misfuel,
but the primary reasons are thought to be price and the
perception of enhanced performance, since leaded fuel is both
cheaper and higher in octane rating than unleaded fuel.
EPA has been collecting data since 1978 on the occurrence of
tampering and misfueling to assess the magnitude of the
problem. Covert observation of vehicle owners at fueling
stations and direct inspection of individual vehicles in
roadside surveys have shown that nationally nearly one in
five in-use vehicles have at least one emission control
disablement and that a significant number of vehicle owners
misfuel. These figures are alarming in light of the fact
that it is a federal violation with large civil penalties for
repair garages, dealerships or fleet operators to remove or
disable emission control components and that many States have
laws which make such disablements by individual vehicle
owners illegal. Tampering and misfueling are, therefore,
significant problems which current efforts have not
adequately held in check.
Inspection and maintenance (i/M) programs are being
instituted in some areas to assure a better state of repair
for vehicles operated in large urban areas with air quality
problems. The Clean Air Act Amendments of 1977 require i/M
programs in urban areas which could not attain ozone or
carbon monoxide air quality standards by 1982. Although
these i/M programs will produce large reductions in HC and CO
emissions, most programs do not explicitly require that all
emission control components be in place and in good repair in
order to pass the i/M inspection. The simple idle test which
is used in most i/M programs is not designed to detect
specific component disablements. Such I/M programs alone,
therefore, will not completely solve that portion of the
excess emissions problem due to tampering and misfueling.
Additional emission reductions from reducing the occurrence
of tampering and misfueling are possible in all areas in
order to help meet or to maintain ambient air quality goals.
Tampering and misfueling, and thus the excess emissions
caused by them, can be reduced in a variety of ways:
o In areas with i/M programs, an ant i-tamperi ng and
anti-misfueling program could be added as part of the
emissions program.
o In areas with an existing safety or other periodic
inspection requirement, an anti-tampering and
anti-misfueling program can be added to the inspection
program. In areas without an existing inspection
requirement, a new requirement can be implemented
either on a periodic or change-of-ownership basis.
-------
-3-
o Various field enforcement efforts can also be used in
any area to deter tampering and misfueling.
Each of these three approaches is examined separately in
Section 5.0, which discusses implementation issues as well as
benefits.
In any approach, the potential benefits from anti-tampering
and anti-misfueling programs will be affected by: 1) how much
tampering and misfueling are occurring given existing
efforts, if any, to control them? 2) the effectiveness of the
program in reducing the observed rate of tampering and
misfuelng; and 3) the effects of tampering and misfueling on
the emissions from vehicles. There are two ways in which
anti-tampering and anti-misfueling programs reduce excess
emissions. First, a program will require repair and
replacement of damaged or missing emission control components
when they are discovered. Second, such a program will also
result in deterrence of tampering and misfueling which would
have occurred if the program had not been implemented-.. Any
program's benefits will be some mix of., thesev twrr elements
although the design of the program may rely- more on one than
the other for program benefits.
This report does not cover specific method's of detection'fbr.v
disablements. The report briefly describes what each
inspection would be like and covers general methods that can
be used to detect disablements. A twenty-hour tampering
detection training course is available from Colorado State
University. This course provides hands-on experience in
identifying the location and general functions of emission
control devices. Colorado State University has also recently
published a book titled "1968-1982 Automotive Emission.
Systems Application Guide". This book provides engine family
specific information on what emission control components are
original equipment on passenger vehicles and light-duty
trucks. Also, in-the-field training can be provided by EPA
inspectors to those jurisdictions interested in establishing
tampering and/or fuel switching enforcement programs that are
aimed at retail gasoline stations, fleet operations and
repair facilities.
Section 2.0 will discuss the current knowledge about
tampering and misfueling rates. Section 3.0 will examine the
effects of misfueling and disablement of individual emission
control components on vehicle emissions, discuss which
vehicles are equipped with each emission component, and
estimate the cost of repairs. Section 4.0 will discuss the
calculation approach which was developed for this report to
estimate the excess emissions caused by tampering and
misfueling. Effectiveness will depend on the particular
-------
-4-
program approach and will, therefore, be discussed for
individual approaches in Section 5.0. Section 6.0 explains
how to use the MOBILE3 emission factor model to calculate
program benefits.
This report analyzes four specific types of tampering—PCV,
evaporative control system, air pump, and catalyst
removal—plus misfueling. Catalyst removal and misfueling
affect HC, CO, and NOx emissions; and the remaining forms of
tampering affect only HC and CO. EGR tampering rates are
also presented, but the emission increases due to EGR
disablement and the potential benefits of an EGR tampering
inspection will be the subject of a later report.
The potential benefits of a check for disabled closed-loop
sensors have not been analyzed because of the uncertainty
associated with identifying a tampering rate for these
relatively new components. However, a gram per mile emission
effect for disablement is listed. Tailpipe I/M tests can
identify as much as 80% of the excess emissions associated
with oxygen sensor tampering. Thus, in i/M areas an oxygen
sensor check would have reduced benefits even if a
significant tampering rate existed. Future tampering surveys
will attempt to identify the existing closed-loop sensor
tampering rate.
The most cost-effective portion of the emission reductions
possible from a program to control tampering and misfueling
is the portion that results from preventing new instances of
tampering and misfueling, since no repair cost is incurred.
Some jurisdictions may wish to inspect only cars sold after
the program begins. For the convenience of such
jurisdictions, benefits are shown in all tables for 1984 and
later vehicles separately from those for older vehicles. One
possible compromise between the larger benefits and costs of
inspecting all model years and the reduced benefits of
inspecting only newer vehicles is to inspect all 1980 and
later model year vehicles. The tables have also separated
the 1980 through 1983 model years for this purpose.
Because 1987 is the deadline for attainment of the ozone and
carbon monoxide standards for areas which received extensions
beyond the 1982 deadline, benefits are calculated for
January 1, 1988.
-------
2.0 TAMPERING AND MISFUELING RATES
2.1 Current Rates
Since 1978, EPA has conducted surveys of in-use vehicles,
both passenger cars and trucks, in seventeen states and
collected data regarding emission component disablements and
misfueling from over 8,000 vehicles. The latest of these
surveys completed in 1982 [1]*, collected data from nearly
3,000 cars in ten states. All of the surveys were conducted
either at a roadside check in conjunction with a random
police roadside pullover or as a special, temporary addition
to a safety or i/M inspection at state-run or private
inspection stations. Although the inspections were
voluntary, efforts were made to assure as complete
participation as possible. Once a city and specific site in
the city were chosen, vehicles were chosen completely at
random, although the surveys since 1980 inspected only 1975
and later model year vehicles. Table 1 presents a summary of
the sample sizes collected in the various states in the 1982
tampering survey. Notation has been added to indicate-. i/M
areas with the program start date and the type of: vehicle
recruitment used in the survey at that site.;;^
The 1982 survey was chosen as the definftive^data base, with
which to calculate current and future-. . tampering rajfces...,-
Comparing the 1982 survey with the previous-survey shows that?,
tampering and misfueling behavior has changed'/.with time, and.;
therefore the latest survey will more clearly match future-
tampering and misfueling behavior. Also, the 1982 survey was
more successful than previous surveys in obtaining ..an.
essentially non-voluntary and therefore unbiased sample.
Table 2 shows the tampering rates observed for the 197 5 and
later vehicles in the 1982 survey. Table 2 indicates that
with the exception of EGR, PCV and evaporative canister
tampering, tampering rates are on average lower in cities
with i/M programs. Later in this chapter an analysis will be
made which more exactly identifies the impact of i/M programs
on tampering and misfueling rates.
Not all instances in which there was evidence of tampering or
misfueling are reflected in Table 2. For example, tampering
with air cleaner housings is not shown, nor is tampering with
crankcase fresh air hoses. Only those cases in which the
tampering was judged to be easily identifiable and to be
sufficient to cause substantial quantifiable increases in
emissions are included in Table 2. Consequently, Table 2 may
differ from other published summaries of the 1982 survey.
~Numbers in brackets refer to references at the end of the
report.
-------
-6-
Table 1
EPA 1982 Tampering
Survey Sample Sizes
Date of
Date I/M
Sample
Tampering
Type of
State
Program Started
Size
Survey
Recruitment
FL
n. a.
307
4/82
a
LA
n.a.
183
4/82
b
MN
n.a.
307
8/82
a
NV*
7/81*
275
9/82
d
NJ
2/74
290
7/82
a
OK
n.a.
282
5/82
b
OR
7/75
310
9/82
c
RI
1/79
324
7/82
a
TX
n.a.
293
4/82
b
WA
1/82
312
9/82
c
Total 2883
~Prior to 10/83, Nevada's i/M program required inspection
only on change of ownership,
a: Random roadside pullover.
b: As part of a centralized or decentralized safety
inspection.
c: As part of a centralized or decentralized i/M inspection,
d: Vehicles were recruited at a parking lot.
-------
-7-
Table 2
Current Tampering and Misfueling Rates*
From 1982 Tampering Survey
Emission Control
i/M Sites
Non-I/M
Sites
System
LDV
LDT
LDV
LDT
PCV
1.1%
3.4%
1.0%
4.4%
Evaporative
1.4%
3.4%
0.4%
6.1%
Air Pump
3.1%
2.9%
6.1%
13.8%
Catalyst
1.7%
5.0%
4.5%
19.5%
EGR
9.9%
10.8%
8.2%
15.2%
Habitual
Misf ueling^
5.3%
12.4%
9.4%
25.0%
Filler Inlet
Restrictor
3.2%
8.3%
6.6%
18.5%
For Comparison
Only:
All
Misfueling^*^
6.5%
12.4%
11.6%
31.0%
(Number of Vehicles)
(1055)
(146)
(1143)
(229)
~Grossly tampered cars only. See text.
~~Defined as an enlarged fuel inlet restrictor or leaded fuel
(lead content greater than 0.05 gm/gal) in tank. Catalyst
vehicles only. See text in Section 2.1 for discussion.
~~~Defined as an enlarged fuel inlet restrictor, leaded fuel
(lead content greater than 0.05 gm/gal) in tank, or lead
compounds detected in the tailpipe. Catalyst vehicles only.
The detection of lead deposits alone is not used as an
indication of habitual misfueling in this report for reasons
given in the text. A positive result on the test for lead
deposits is believed to be an accurate indication that at
least some leaded fuel has been used, however. The rates for
"all" misfueling shown in this table are for comparison only.
-------
-8-
The interpretation of the 1982 survey data to determine which
instances of tampering were sufficient to cause substantial
quantifiable increases in emissions was straightforward
except for misfueling. The survey examined three vehicle
parameters relative to misfueling: (1) whether the lead
content of the fuel in the tank was over the legal limit of
0.05 gram/gallon; (2) whether, the fuel inlet restrictor had
been enlarged enough to allow a leaded fuel nozzle to be
used? and (3) whether lead sensitive "Plumbtesmo" test
paper[2] detected lead deposits in the tailpipe. To result
in deactivation of the catalyst and substantial long term
emission increases, it is believed misfueling must be either
repeated at least three or four times in succession, or, if
not done consecutively, must occur with a fairly high
frequency over a long period of time. Such consecutive or
frequent misfueling is called habitual. As discussed below,
no combination of the parameters examined in the 1982 survey
are definitive indicators of this.
Each of the three misfueling parameters examined in the
survey has shortcomings in that habitually misfueled cars may
escape detection. Checking the inlet restrictor does not
detect: (1) vehicles whose owners have misfueled using
funnels or illegally small nozzles; (2) vehicles which are
victims of fuel mislabeling by gas stations or distributors?
or (3) vehicles which have otherwise used contaminated
gasoline. Fuel samples drawn on a one-time basis cannot
detect vehicles which were misfueled regularly in the past,
but for some reason, e.g., change of owners, have not been
misfueled recently. Available information does not rule out
the possibility that the lead sensitive Plumbtesmo test paper
may detect vehicles which have only been misfueled a couple
of times at wide intervals and have catalysts which are still
active. The test paper can also fail to detect vehicles
which have had tailpipe replacements since the last
misfueling episode.
The inlet restrictor check can be assumed to have few false
positives, since an owner is extremely unlikely to have
tampered with the restrictor for no reason. Past or current
habitual misfueling is therefore assumed whenever a tampered
inlet is found.
The check on fuel lead content is a definite identifier of
those cases where leaded fuel has been used recently.
Information on the observed lead concentrations of vehicles
over the legal limit but with intact inlet restrictors is
presented in Figure 1. Most of the vehicles with fuel over
the legal lead limit were well over it, so low level
contamination of unleaded fuel cannot possibly be the cause
in these cases. Many of the cars clearly had filled with
-------
-9-
leaded fuel at the last fillup. Based on EPA fuel
inspections and other fuel surveys, it is far more likely
that leaded fuel was purchased knowingly than that the
gasoline retailer had sold leaded fuel from a pump labeled
unleaded. Given that the owner knowingly bought leaded fuel
recently, itr is likely that the vehicle has been habitually
misfueled. Evidence that owners who use. leaded fuel once
tend to do so regularly is discussed in the last paragraph of
this section.
The only remaining issue, then, is whether a vehicle which
has a Plumbtesmo test paper result indicating misfueling, but
which does not have other indications of misfueling, has
actually been misfueled enough to deactivate the catalyst.
Since the fuel in the tank is below the legal lead limit, it
is certain that unleaded fuel has been used for at least the
last two or three fillups. The most plausible scenario for
earlier habitual misfueling would be that a previous owner
had misfueled extensively using a funnel or illegally small
nozzle but that the present owner does not». This. is: clea.rly
a possibility, particularly for older cars, but is* tempered
by the low rate of owner turnover. it is also possible that
a family car was or is raisfueled habitually by one member of
the family but not by the member who filled the tank the last
few times. A single vehicle operator may also have
habitually misfueled only during the last gasoline crisis, in
1979, when unleaded fuel may have been difficult to obtain.
Otherwise there is little reason to believe that the same
owner would stop habitual misfueling once he or she started.
The other possibility, as mentioned, is that leaded fuel has
been used only a couple of times, for whatever reason and
perhaps unknowingly.
Because of the uncertainty as to how to handle the vehicles
which failed only the test paper results, EPA has chosen for
this report to. include only the fuel lead content and inlet
restrictor as evidence for calculating habitual misfueling
rates. As can be seen in Figure 2, this decision reduces the
number of vehicles with any indication of misfueling that are
considered habitually misfueled by about 18% for the
passenger cars and 15% for the light-duty trucks. For the
reader's information, Table 2 shows the misfueling rate based
on these two indicators alone and on all three indicators.
EPA will be considering ways to reduce the uncertainty in
this area and may provide further information later.
An analysis of fueling habits was recently performed by a
Department of Energy contractor using data from detailed
diaries kept by families on their gasoline purchases[3].
(Data used for the fuel diary analysis is voluntary and
therefore may misrepresent the true incidence and patterns of
-------
-10-
misfueling.) The DOE analysis showed that among the families
keeping diaries, more than 85% of the leaded fuel purchased
was purchased by vehicle owners who misfuel more than 50% of
the time. This suggests that a given owner rarely stops his
or her habitual misfueling once started, but says nothing
about previous owners. This analysis supports the assumption
used in this report that evidence of deliberate misfueling,
such as a tampered filler inlet, usually indicates habitual
misfueling. The diaries have not yet been analyzed to
determine exactly how many vehicles were affected by serious
misfueling during the diary period.
-------
-11-
Figure 1
Distribution of Lead Content for Those
Passenger Cars in the 1982 Tampering Survey
Which Had FUel Lead Content
Greater Than 0.05 Grams Per Gallon
FUel Lead Content In Grams Per Gallon
-------
-12-
Figure 2
Passenger Cars;
All
Any
Tank
Inlet
Tailpipe
Overlap Among Indicators
of Misfueling in the 1982
EPA Tampering Survey*
Number
of Vehicles
2207
188
109
101
111
Inlet
Tailpipe
Number Inl
Light-Duty Trucks: of Vehicles
All = 351
Any = 7 7
Tank = 44
Inlet = 47
Tailpipe = 60
Tailpipe
*A11: All catalyst vehicles rn sample.
Any: All catalyst vehicles with any one or more of the
following indications of misfueling
Tank: All catalyst vehicles whose fuel sample indicates
a fuel lead content greater than 0.05 grams per
gallon.
Inlet: All catalyst vehicles whose fuel inlet restrictor
allows entry of a leaded fuel nozzle.
Tailpipe: All catalyst vehicles whose tailpipe lead deposits
indicate past use of leaded fuel.
-------
2.2 Future Rates
In order to estimate the excess emissions caused by tampering
and misfueling on a future date, January 1, 1988, for
example, it is necessary to predict the tampering and
misfueling rates when the average age of the vehicles will be
older than observed in the 1982 survey. Examination of the
data from the 1982 survey shows a marked increase in the
tampering rates of some components, including catalysts, and
in misfueling rates as the average mileage of the sample
increases. This increase is illustrated in Figures 3 through
9. Consequently, the dependence of tampering rates on
mileage must be accounted for.
To examine this issue, a linear regression equation on
mileage was fit to data from the 1982 EPA survey and appears
to reasonably explain the tampering and misfueling rates
observed in the surveys. Some of the regression lines are
also shown in Figures 3 through 9. Each linear equation is
defined by a zero mile rate and an increase in the rate for
every 10,000 miles of fleet average mileage. Other
non-linear equations did not seem to better explain the
increase. It was decided, therefore, to use the linear
equation to estimate the tampering and... aisfueling rates on
January 1, 1988, using standard EPA predictions of the
average age in miles of each model year on that date.
Least squares regression was used to estimate a line of the
form Y = bX+a, where Y is the proportion of tampered vehicles
at mileage X. The data used to generate estimates of the
regression coefficients, a and b, were the mileage and
whether the vehicle was tampered (Y=l) or not (Y=0) for each
vehicle in the 1982 tampering survey.
Least squares regression, as used in our case, requires
several assumptions concerning the distribution of Y for
fixed X in order to estimate the error variance of a and b.
Ordinarily, the Y values are assumed to be normally
distributed for each value of X. Further, it is assumed that
the variances for these Y distributions are equal at all
points along the line. Since the Y values in our data are
either zero or one, neither of these assumptions are met.
However, an investigation of the properties of the least
squares estimators has shown that they remain unbiased even
in the presence of a binary dependent variable. Since it is
unnecessary to obtain error estimates for the regression
coefficients for this application, it was determined that the
simple least squares regression approach is sufficient for
this application.
-------
-14-
In calculating equations to predict tampering and misfueling
rates, several factors have been considered. The rate of
tampering and misfueling among passenger cars and among
trucks is significantly different. Therefore, each of these
vehicle types were treated separately. Also I/M areas tend
to have lower tampering and misfueling rates than areas
without I/M programs. Each of these two classifications is,
therefore, also treated separately. All vehicles surveyed in
Portland were eliminated since Portland conducts a tampering
inspection in conjunction with its I/M emissions test.
Although local tampering and misfueling rates can vary
greatly, only one set of tampering rate equations is used in
this report. If a particular area has reason to believe, or
has data from the 1982 or 1983 tampering survey which show
that tampering or misfueling rates are higher in its area
than in the nation as a whole, EPA is willing to evaluate the
evidence and estimate benefits specific to that area. In
fact, EPA will require the use of valid local survey data
where it is available.
In order for a local survey to be considered valid, the
established EPA survey procedures must be followed.
Therefore, areas that are interested in using local survey
data should contact the Field Operations and Support Division
before conducting the survey to be sure that the results will
be acceptable. Also, in order to use local light-duty truck
tampering rates a sufficient sample of trucks will be
required.
If a state or local jurisdiction wishes to use a known or
estimated local VMT fraction for light-duty trucks that is
more than 25% higher than the national average VMT fraction,
EPA will assign a default tampering rate which will be
calculated from that subset of survey sites which have higher
than normal LDT populations. Alternatively, an EPA approved
local survey may be used to determine a local specific rate.
EPA tampering surveys have shown that light-duty truck
tampering rates are much higher for some components on a
national basis than those for light-duty vehicles. EPA is
unwilling to assume however, that areas with high light-duty
truck VMT have proportionally more serious truck tamperers.
Consequently, local data on tampering rates for light-duty
trucks will be required, unless the jurisdiction is willing
to use the default rate discussed above or assumes a national
average VMT fraction.
-------
-15-
Since there are no data in the 1982 survey from model years
before 1975, and since these vehicles should have little
effect on the overall benefits in 1987, it has been assumed
that tampering rates for pre-1975 cars are the same as for
1975 and later passenger cars at equal mileages. it is also
assumed that the tampering and misfueling behavior of 1981
and later model year passenger car owners will not be
significantly different in future years than the behavior of
pre-1981 passenger car owners, for those components treated
in this report. Both of these assumptions are unproverr, but
the data available are not adequate to treat these groups
separately. The assumption for 1981 and later vehicles may
not be necessary in future versions of this document, since
more data on 1981 and later vehicles at higher mileages will
be available then.
In addition, truck sample sizes are inadequate to estimate
the rate of increase of tampering and misfueling for trucks,
therefore, the rate of increase in tampering and misfueling
for passenger cars has been assumed for trucks also, although
the zero mile rates have been adjusted to reflect the
observed differences in the average tampering and misfueling
rates between trucks and passenger cars.
Table 3 presents the linear regression equ&C&or"coefficients'
calculated from the tampering survey datav:r*~The. equations.'
describe the relationship of tampering and lirisfaeling- rates-;
to vehicle mileage in the non-I/M areas. The light-duty.?
truck zero mile rate values were calculated using the overall
truck tampering and misfueling rates and average mileage and
projecting backwards to zero miles assuming the same increase
in rate as for passenger cars.
Table 4 presents the same information but for the I/M sample,
assuming no formal tampering check. Differences in the
design and history of the I/M programs had to be overlooked
in the interest of retaining a meaningful sample size.
Logically, an ordinary I/M program should have little affect
on EGR, PCV and evaporative canister tampering, since they
have little or no affect on idle HC and CO exhaust emissions
measured in I/M programs. Consequently, the tampering rates
for these components have been calculated using both I/M and
non-I/M areas combined. .
-------
-16-
In both Table 3 and Table 4, some linear equations contain
negative zero mile rates. Since these negative levels are
small, no effort has been made to force the equation through
zero. If, however, a tampering or misfueling rate for a
particular model year is calculated to be less than zero in
the evaluation year, that rate for that model year is set to
zero.
In both Table 3 and Table 4, overlap among tampering types is
ignored, so one car can contribute to several of the
regression equations. The overall tampering rate at a given
mileage is, therefore, less than the sum of these equations.
In estimating excess emissions due to tampering and the
benefits of controlling tampering, it is necessary to
explicitly account for vehicles with more than one form of
tampering, since tampering effects are not always additive.
Following sections will describe how this was done for each
case.
It is unlikely that a local tampering survey would inspect
enough vehicles to allow a reliable estimate to be made of
the dependence of tampering on vehicle mileage. Instead, the
outcome of a local survey is likely to be only the knowledge
that the tampering rate for a given component was a certain
value for a group of vehicles of a certain average mileage.
However, to estimate benefits, predictive equations that
include mileage effects are needed. EPA will address this
need as follows. If the local tampering rate for a vehicle
type (car or truck) is higher than the corresponding national
rate, the national equation will be adjusted by increasing
the zero-mile constant but the mileage coefficient will be
unchanged, as done above for light-duty trucks in the
national case. If the local rate is lower than the
corresponding national rate, both the zero-mile constant and
mileage coefficient will be reduced proportionally to the
rates of the local and national tampering rates at equal
mileages. Using an equal slope for rates lower than the
national averages would result in potentially unreasonable
mileage intervals before tampering begins. Using a
proportional adjustment for rates higher than the national
averages would create potentially unreasonable slopes for
tampering rates. The approach used moderates the potential
problems by using a different method for each case.
-------
-17-
Table 3
National Average Tampering Rate
Prediction Equations for Non-I/M Areas
Tampering Rate =
(zero
(A +
if mileage is
B x (mileage)
less than
otherwi se)
M0)
••M0"
(miles)
"A"
(%)
" B"
(%/10K)
Rate at
50,000
Miles (%)
Emission
Control
Component
LDV
LPT
LDV
LPT
Both
LDV
LPT
Air Pump
10,219
0
-2.71
4.89
2.652
10.55
18.15
Catalyst
12,104
0
-1.95
13.53
1.611
6.11
21.59
PCV System*
0
0
0.02
3 .08
0.248
1.26
4.32.
Evaporative*
Canister
14,328
0
-0.48
3.77
1.20
5.45.
Filler Inlet
7,072
0
-1.43
11.01
2 .022
8.68
21.12
Other
Mi sfueli ng**
0
0
1.65
6.96
0.559
4.45
9.76
EGR System
273
0
-0.06
5.02
2.199
10.94
16 .12
*EGR, PCV and evaporative canister tampering rates are assumed
to be the same in i/M and non-l/M areas.
~~Defined as leaded fuel (lead content greater than 0.05
gm/gal) in tank. Catalyst vehicles only. See text in Section
2.1 for discussion.
-------
-18-
Table 4
National Average Tampering Rate
Prediction Equations for I/M Areas
Tampering Rate =
Emission
Control
Component
(zero if mileage is less than M0)
(A + B x (mileage) otherwise)
«o
(miles)
LDV LDT
(%)
LDV
LD-]
"B"
(%/10K)
Both
Rate at
50,000
Miles (%)
LDV LDT
Air Pump
Catalyst**
PCV*
Evaporat ive*
Canister
Filler Inlet
9,091
2,397
0
14,328
9,001 -1.01
Other
Misfueling** 12,987
EGR System
273
0
0
0
0
0
0
-0.11
0.02
-0.48
-0.77
3.82
-0.06
-1.00
3.32
3.08
3.77
4.70
6.99
5.02
1.111
0.459
0.248
0.335
1.000
-0.211
2.199
4.55 4.56
2.19 5.62
1.26 4.32
1.20 5.45
4.23 9.70
2.77 5.94
10.94 16.12
*EGR, PCV and evaporative canister tampering rates are
assumed to be the same in i/M and non-l/M areas.
~~Defined as leaded fuel (lead content greater than 0.05
gm/gal) in tank. Catalyst vehicles only. See text in
Section 2.1 for discussion.
-------
-19-
Figure 3
Rate of PCV System Disablement
Versus Mileage
25
T3
£ 20
w 15
0)
^ 10-
Lsgend.
O Pqwngar Car
Raqr«wion
• Light-Outy Truck
5-
0J
0123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-20-
Figure 4
Rate of Evaporative Canister Disablement
Versus Mileage
25
T3
® 20
a
3"
a"
W
W 15
-------
-21-
Figure 5
Rate of EGR System Disablement
Versus Mileage
0123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-22-
Figure 6
Rate of Air Pump Disablements
Versus Mileage
25
T3
« 20
w 15
l/M RvgrMsion
~ NonH/MLOV
Nan—l/M R«w*«sion
• l/MUJT
¦ NonH/M LOT
0123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-23-
Figure 7
Rate of Catalyst Removal
Versus Mileage
Legend
O l/MLOV
l/M fteflrmvion
~ Non-H/M LOV
NonH/VI Raortssion
• l/M U3T
¦ NonH/MLDT
0123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-24-
Figure 8
Rate of Misfueling By Means of
Rid Filler Inlet Tampering
Versus Mileage
Legend
o t/y toy
l/M Rtflmsion
~ NonH/MLDV
NonH/M
• I/MU3T
¦ NonH/MLOT
0123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-25-
Figure 9
25-
Rate of Misfueling Other Than
By Means of FUel Filler Inlet Tampering
Versus Mileage
20-
W 15'
a)
10-
5-
O-1-
Legend
O l/MLOV
l/M R«4rniien
~ NonH/MLOV
NonH/M Rcgrmion
• l/M LOT
¦ NonH/M LOT
0 123456789 10
Vehicle Mileage In 10,000 Mile Increments
-------
-26-
3.0 EFFECTS OF TAMPERING AND MISFUELING AND COSTS OF
REPAIRS
The effect of a particular disablement of a specific emission
control component on vehicle emissions is not easy to
quantify. There are many different varieties of similar
emission control devices which can differ from manufacturer
to manufacturer and from model year to model year. Different
varieties can also have a different effect on vehicle
emissions depending on the engine type and overall state of
tune as well as the condition of other emission control
components. A testing program which would evaluate every
possible combination of all of these factors would require
immense resources. There has been some testing performed
over the years by EPA to assess the impact of disablements.
FTP and other tests were performed with and without a
particular emission control component disconnected. usually
all other emission control components were in operation and
the vehicles were in proper tune. The emission increases due
to disablement may vary for vehicles in less perfect
condition, however EPA believes that these tests provide the
best information available on the impact of in-the-field
tampering and misfueling on an individual vehicle's emissions.
In this report the individual vehicle benefits from repairs
of specific emission control component tampering are taken,
when possible, from these types of data. When practical, the
existing data are further divided into appropriate model year
technology groups to take into account changes in the design
and effectiveness of particular emission control components
in different model years. When adequate test data from
disablement testing are not available, estimates of the
benefits were made based on known controlled and uncontrolled
emission levels of vehicles of different model years.
The few jurisdictions with NOx attainment problems may want
to consider including an EGR check in an inspection program.
In fact, an under-the-hood tampering inspection which ignores
the EGR system - the most common tampering target - may lack
public credibility after its implementation even if NOx
reductions are not needed locally, since public understanding
of the differences between pollutants may be limited. NOx
emission reductions from anti-tampering and anti-misfueling
programs will be addressed at a later date.
3.1 Air Pump
The purpose of the air pump is to supply air to the engine's
exhaust in order to promote the oxidation of HC and CO to
harmless by-products. The air pump performs this function on
both catalyst and non-catalyst vehicles. The air pump is
-------
-27-
driven by means of a belt which transmits power from the
crankshaft as it rotates. This method of powering the air
pump is the same as that used to run the alternator and air
conditioner compressor. The air pump can, therefore, be
found near or on the same plane as the alternator or air
conditioning compressor. Its plumbing distinguishes it.
Some vehicles are equipped with pulse-air systems which also
supply supplemental air to the exhaust stream but without a
belt driven pump. Disablement of these systems is less
frequent than for air pump systems and identification of
disabled pulse air systems is not always as easily
accomplished; therefore, this section will deal solely with
disabled air pump systems..
The percentage of vehicles equipped with air pumps varies by
model year. An analysis of the occurrence of air pump
systems on passenger vehicles in the EPA Emission Factor data
base was used to establish estimates of the percentage of
vehicles in each model year group prior to 1984 equipped with
air. pump systems. The percentage for 1984 and later vehicles
was projected. The resulting estimates are shown in Table 12
through 14 at the end of this section.
There are three ways the air pump is .normally disabled-
First, the belt which drives the pump. .. can be removed.
Second, the entire unit — pump, belt, flexible hoses, steel
piping, and even mounting brackets — can be removed. Third,
the output hose from the air pump can be disconnected and/or
the air routing valve can be damaged. This last disablement
results in the air pump spinning freely and no air being
supplied to the exhaust. For purposes of this report, it is
assumed that all three of these forms of disablement can.be
readily identified by trained inspectors.
The repairs necessary for these various forms of disablement
are self-evident. In most cases, repair can be accomplished
by simply installing a new belt or reconnecting a hose. An
average repair cost of $20 has been assumed for this
analysis. This estimate takes into account the few cases in
which an expensive repair or reinstallation of an air pump is
expected to be required.
The HC and CO emission increases which accompany air pump
disablement for oxidation catalyst vehicles were quantified
by examining data from 13 vehicles (1975-1976 model years)
tested with and without their air pumps operational. All of
these vehicles came from the 300-car Restorative Maintenance
program. The vehicles are listed with their before and after
emission levels in the Appendix. The results are summarized
in Table 5. (One source of uncertainty in the analysis has
to do with the fact that the vehicles used to determine the
-------
-28-
emission effects of air pump disablement were all in tuned-up
condition. The emission increases due to air pump
disablement for vehicles in less perfect condition may vary.)
Table 5
Increase in HC and CO Emissions
Due to Air Pump Disablement
Technology
Increase in
HC Emissions
(gm/mi)
Increase in
CO Emissions
(gm/mi)
Oxidation Catalyst
3-way Catalyst
1.37
0.51
30.61
16.29
For three-way catalyst vehicles, the effects of air pump
disablement were quantified by examining the results of EPA
laboratory programs which took three vehicles representative
of three-way catalyst technology and tested them with and
without their air pumps operational. In addition, one
representative vehicle tested in an EPA surveillance program
in California was found to have its air pump disabled due to
having one of the vacuum control hoses kinked closed. This
vehicle was tested as-received (air pump disabled) as well as
after having the air pump repaired (vacuum hose unkinked).
Data from these four vehicles are listed in the Appendix and
summarized in Table 5.
There is some uncertainty as to the HC and CO effects of air
pump disablement for non-catalyst vehicles as no similar data
are available. However, these vehicles contribute only a
very small share of the fleet's emissions over the life of an
ant i-tampering program. They are assumed to show the same
absolute effect due to air pump tampering as for oxidation
catalyst vehicles. This assumption is reasonable and due to
the small contribution made by these vehicles, does not
significantly affect the analysis.
It is assumed that the effects described above apply to all
vehicles of the technology type regardless of the emission
standards. The available data do not allow a more detailed
analysis. It is likely, however, that any error in this
assumption is small.
-------
-29-
3.2 Catalyst
Automotive catalytic converters lower HC and CO emissions in
the exhaust by catalytically promoting the oxidation of HC
and CO to harmless by-products. (Catalysts on most 1981 and
later vehicles also help reduce NOx emissions.) Catalysts
are normally mounted on the underside of the vehicle, along
the exhaust pipe and before the muffler; however, a few
vehicles have catalysts mounted inside the engine
compartment. Tampering with the catalyst usually takes the
form of simple removal of the catalyst and replacement with
an exhaust pipe. Some automotive parts suppliers carry a
complete selection of catalytic converter "test pipes" which
can be bolted into the gap left in the exhaust pipe after the
converter is removed.
Using carefully placed mirrors or a mirror on an extension,
the underside of an inspected vehicle can be examined for the
presence of the converter. A catalytic converter is easily
distinguished from a muffler since it is made of stainless
steel and will not rust. If a catalyst is not observed by
checking underneath a 1975 or later model year vehicle, it
will be necessary to open the engine compartment hood and
either locate the catalyst there or confirm from the
emissions label put on every vehicle or from reference
literature that the vehicle was not equipped with a catalyst
at the factory. Colorado State University has recently
published a book which contains this information.[4] Tables
12 through 15 present the percentage of vehicles assumed to
have been equipped with a catalyst for this report.
Obviously repair will require installation of a new catalyst
(or reinstallation of the old one if it was saved). This
could be a relatively expensive repair. New catalysts now
cost between $172 and $320. Most of this cost is dealer and
distributor markup. However, most vehicles do not require
the more expensive converters. A market for lower-priced
non-OEM catalysts may also appear, if new OEM catalysts are
not a requirement of the program. Some certification of
catalyst efficiency should be required before non-OEM
catalysts are accepted since the benefits which appear later
in this report assume that the replacement catalyst is
operating properly. Lower-priced replacement catalysts are
possible if enough demand is created by a catalyst check. An
average cost of $200 per catalyst has been assumed for this
analysis.
The HC and CO emission increases which accompany catalyst
removal were determined by examining the engine-out (before
the catalyst) emissions of a number of vehicles involved in
several test programs. A listing of these vehicles is in the
-------
-30-
Appendix. These vehicles received both baseline tests (all
components functional) and tests with the catalyst removed or
bypassed. By comparing the results of the two tests, the
percentage increase in emissions which accompanies catalyst
removal can be calculated. Most catalysts are removed with
the intent of also using leaded fuel. There is evidence that
the use of leaded fuel itself will cause an increase in HC
emissions due to lead deposits in the engine. This effect
has been ignored in this analysis. Four vehicles with
oxidation catalysts and four vehicles with three-way
catalysts were tested. The results are summarized in Table 6.
Table 6
Increase in Emissions
Due to Catalytic Converter Removal
Increase in Increase in Increase in
HC Emissions CO Emissions NOx Emissions
Technology (gm/mi) (gm/mi) (gm/mi )
Oxidation Catalyst 3.05 28.01 0.00
3-way Catalyst 1.68 17.80 2.16
Insufficient testing has been conducted to determine how the
effect of catalyst removal varies with the average mileage of
a fleet. It is, therefore, assumed that the gram-per-mile
increase in emissions from catalyst removal remains the same
throughout a vehicle's life, regardless of mileage. This
will mean that the percent change due to catalyst removal
reduces with increased mileage. This makes sense since very
little of the deterioration of the fleetwide emission factor
is due to catalyst aging. Most is due to in-use
maladjustments and failures of other emission components.
Removing the catalyst on a vehicle that has high engine-out
emissions can be expected to have a smaller percentage effect
than removing a catalyst from a tuned vehicle, since there is
usually a relative shortage of oxygen in the exhaust of
maladjusted vehicles. This does mean that the estimates will
include some degree of uncertainty, especially when applied
to high mileage vehicles.
3.3 Habitual Misfueling
The use of leaded gasoline in a vehicle equipped with a
catalytic converter, referred to as "misfueling" in this
report, will cause a steady contamination of the catalyst
material resulting in lower and lower catalytic efficiency.
The result of continued misfueling will, therefore, be higher
-------
-31-
exhaust emission levels as the catalyst loses its ability to
convert pollutants into less harmful substances. It has been
estimated that after as few as three consecutive tankfuls of
leaded fuel, the majority of the catalyst's ability to
convert pollutants will be permanently lost, even if the
vehicle owner resumes use of unleaded fuel.
Determining the effects of misfueling is more difficult than
for most other checks described in this report, since the
increase in emissions is heavily dependent on catalyst
efficiency and thus the intensity of the misfueling.
MisS\ielir.g performed sporadically, often referred to as
"casual" ,-aisf ueli ng, nay not permanently destroy the
catalysr's function, although there will be some lasting
reduction in catalyst efficiency. This section estimates
only the effect of habitual misfueling, based on tests of
vehicles operated on leaded fuel for many tankfuls* There
are insufficient test data to estimate the long term effects
of casual misfueling, therefore casual misfueling is assumed
to have a comparatively negligible long term effect on fleet
emissions.
liven in cases of habitual misfueling/ some very low level of
catalyst efficiency may still remain. For this reason the
effect of misfueling is not as great as removal of the
catalyst on an individual basis. Since the overall rate of
misfueling is larger than that of catalyst removal, however,
the overall effect on emissions is more serious.
EPA has previously estinated the average effect on HC and CO
emissions of misfueling. These estimates were used in the
mobile source emission factors raodel (MOBILES f to adjust the
emissions of EPA's essentially misfueling-xree emission
factors test sample to reflect the extent of misfueling in
the fleet as a whole. These estimates were used in the form
of a percent increase over the average low-mileage emissions
of non-misfueled cars.
In this analysis all data now available were examined to
recalculate a gram per mile increase. These data included
data from four oxidation catalyst vehicles and seven 1981 and
later three-way catalyst vehicles. The emission increases
for 1981 and later model year vehicles include any effect
misfueling has on oxygen sensor performance in the
closed-loop vehicles in the sample. Most vehicles were run
on at least 10 tankfuls of leaded fuel. All of the vehicles
are listed in the Appendix.
Table 7 presents the estimated effect on emissions as a
gran-per-mile increase. As with catalyst removal, the
increase expressed in grams per mile is assumed not to change
with mileage.
-------
-32-
Table 7
Increase in
Emissions Due to Misfuelinq
Technology
Increase in
HC Emissions
(qm/mi)
Increase in
CO Emissions
(qm/mi)
Increase in
NOx Emissions
qm/mi
Oxidation Catalyst
3-way Catalyst
2.47
1.57
20.96
11.30
0.00
0.76
The average cost of replacing a misfueled catalyst will be
less than replacing a removed catalyst since in some
instances, only the catalytic material within the catalyst
need be replaced. Some manufacturers' catalysts have a
removable plug for this purpose and provide kits with
replacement catalytic material. In this analysis, the
average cost for replacing misfueled catalysts is assumed to
be $150.
If repair of the fuel inlet restrictor is required,
replacement cost of the restrictor will vary substantially.
Some vehicles' filler necks can be easily replaced with a new
OEM part, while others would require replacement of the
entire fuel tank. It is possible to repair the fuel inlet by
simply gluing in a metal washer using a gasoline resistant
epoxy, however, no credit will be achieved by programs which
allow epoxy/washer fixes, because there is no assurance that
misfueling would not continue since these washers could be
easily inserted and removed. In this analysis the average
repair cost for tampered fuel inlet restrictors is assumed to
be $80.
As will be discussed in a later section of this report, EPA
is concerned that many owners of vehicles with tampered
inlets may repair those inlets before submitting to inlet
inspection in the first year of a new inlet inspection
program, to avoid the costly penalty of catalyst
replacement. A State must require that inlets be repaired
only with OEM parts and also require the vehicle owner to
show proof of purchase and installation of a new or certified
catalyst since the purpose of such a program is to require
catalyst replacements for misfueled vehicles.
-------
3.4 Positive Crankcase Ventilation System
The positive crankcase ventilation (PCV) system in
automobiles provides a means to purge the crankcase of gases
escaping from the cylinders by the piston rings. These gases
are detrimental to engine life since they dilute, and break
down engine oil and are corrosive. Originally these gases
were vented to the atmosphere, but with the advent of
pollution control, these gases have been diverted to the
vehicle's intake system for recombustion. The value of the
PCV system is well known and established; therefore, its
deliberate disablement is relatively rare. Only a small
percentage of the vehicles in EPA's surveys had their PCV
vacuum hoses disconnected resulting in the blowby gases being
released to the atmosphere. Other PCV problems, such as
disconnected "fresh air" hoses, also occur but are not
believed to cause a significant increase in emissions from
the automobile.
Disablement of the PCV system usually takes the form of a
disconnected vacuum line or missing components. These
disablements are easily identified either, visually or by a
simple check for vacuum at the fresh air:*hose. Since all of
the components are relatively inexpensively. and since many
disablements are simply disconnections, average repair coats,
are assumed to be $10.
The primary effect of a disabled PCV system is the increase
in non-exhaust HC emissions. There are not enough data from
recent testing programs on the effects of PCV disablement on
current vehicles to determine with complete certainty how
much HC emissions would increase. However, it is estimated
in MOBILE2 that the average crankcase HC emissions from early
I960's vehicles without PCV systems were about 4.1 gm/mi[5].
At the time, most engines had eight cylinders. it is
reasonable to assume that uncontrolled crankcase emissions
are proportional to the number of cylinders, so current and
future vehicles, which will on average have fewer than eight
cylinders, will have proportionately less of an increase when
their PCV systems are disabled. Based on this assumption,
6-cylinder engines should have a 3.08 gm/mi effect and
4-cylinder engines a 2.05 gm/mi effect.
To estimate the average effect of PCV disablements, the mix
of four, six, and eight cylinder engines in the various model
year groups must be determined. Using information on the
past and predicted production of vehicles produced in the
U.S.[6] and assuming that nearly all imported vehicles are
equipped with four cylinder engines, the percent mix of
engine sizes can be estimated for each model year group.
These values were used to combine the estimates for crankcase
-------
-34-
HC emissions from each engine size to determine an overall
figure for each model year group. These overall figures are
presented in Table 8.
Table 8
Increase in HC Emissions
Due to PCV Disablement
Increase in HC Emissions
(gm/mi)
Model Years LDV & LDTl LDT2
Pre-1963
1963-1968 3.80
1968-1970 3.74 5.20
1971-1974 3.51 4.88
1975-1977 3.44 4.78
1978-1979 3.29 4.57
1980 2.83 3.93
1981-1982 2.68 3.73
1983 and Later 2.49 3.46
As shown in Table 15 at the end of this section, all 1968 and
later vehicles are assumed to be equipped with PCV systems.
Between 1963 and 1968 another system was used to vent
crankcase emissions into the air cleaner of most vehicles.
These systems have no PCV valves or vacuum hoses, yet they
were fairly effective in controlling crankcase emissions. It
is therefore likely that, contrary to the assumption used in
this report, a disablement of a PCV system which leaves the
hose connection to the air cleaner intact, will not return
the vehicle's crankcase emission levels to those of an
uncontrolled vehicle. Also, the emission estimates of an
uncontrolled vehicle assume a continuous evacuation of the
crankcase using a vent. tube. Even a disablement of a PCV
system that interferes with both the vacuum hose to the
carbureter and the fresh air hose to the air cleaner will not
cause a continuous evacuation of the crankcase to the
atmosphere. For these reasons, the estimates in the report
for excess emissions from PCV disablements may be
overstated. Since the PCV tampering rates are fairly low,
this is not a major concern to EPA. EPA is planning a test
program to better determine the effect of PCV system
tampering.
-------
-35-
3.5 Evaporative Emission Control System
The evaporative control system is intended to capture the
gasoline fumes which are naturally given off whenever
gasoline is stored and used. These fumes are made up of pure
hydrocarbon (HC) emissions and represent a significant
portion of a vehicle's total HC emissions. The evaporative
control system captures the fumes given off by both gasoline
in the fuel tank and the gasoline in the carburetor (early
systems dealt only with evaporative losses from the fuel
tank). These fumes are stored in a charcoal canister,
usually mounted in the engine compartment, and then routed to
the engine for burning at appropriate times.
Disablement can take the form of disconnected or cut hoses,
missing canisters, or removal of the entire system. Once
again, these forms of disablement are identifiable by trained
inspectors. A quick visual check can usually determine
whether the canister is still intact and if all the hoses are
attached to it. An average repair cost of $10 has been
assumed since most repairs will involve simply reconnection
of hoses.
The emission increases assigned to each grouping to represent
a tampered system come from M0BILE2. The passenger car model
year groupings used in M0BILE2 are: pre-1970, 1970-1971,
1972-1974, 1975-1977, 1978-1980, 1981 and later. The
assumption used to determine the increase in emissions due to
evaporative system disablement for pre-1977 vehicles was that
any disablement would return the vehicle to uncontrolled
levels (pre-1970) of evaporative HC. This assumption is
necessary since there has been no disablement testing done
for evaporative control systems on these older vehicles.
These vehicles are similar, however, in size and design to
the pre-controlled vehicles so that the error should be
small. Newer vehicles have smaller carburetors and gas tanks
and, therefore, should emit less evaporative emissions even
if tampered.
Two 1981 model year vehicles have been tested with and
without disabled evaporative canisters. These vehicles are
listed in the Appendix. As expected the average evaporative
emissions with the evaporative canister disconnected were
less than for pre-controlled vehicles. Since downsizing for
passenger cars began with the 1977 model year and leveled off
after the 1980 model year, the uncontrolled emission levels
for those model years were interpolated between the
evaporative emission levels of pre-1970 vehicles and the test
results from the 1981 vehicles. The resultant increases in
evaporative HC emissions due to disablement of the
evaporative control system are tabulated in Table 9. These
-------
-36-
increases were derived from vehicle testing with Indolene,
the fuel used in EP&'s certification program. The effect of
canister disconnection where commerical fuel with higher
volatility is sed is not known at this time.
Table 9
Increase in HC Emissions Due To
Increase in
Evaporative HC
Emissions (gm/mi)
Model
Passenger
Light-Duty Trucks
Years
Cars
{0-6000 lbs)
(6000-8500 lbs)
1971
0.69
0.81
1972-1976
1.18
1.39
-
1977
1.01
1.39
-
1978
1.70
2.41
-
1979
1.53
2.41
1.88
1980
1.36
2 .41
1.88
1981 and Later
1.50
2.58
2.01
Because of different assumptions for average mileage traveled
per day and uncontrolled evaporative emission levels for
light-duty trucks, the increases in evaporative emissions for
these vehicles are somewhat different than for passenger
cars. Most light-duty trucks over 6000 pounds built before
the 1979 model year were not equipped with evaporative
control systems other than the PCV system. The increase in
evaporative emissions for light-duty trucks also assumes no
downsizing.
3.6 Light-Duty Truck Effects
In MOBILE2 light-duty vehicles (passenger cars) are treated
separately from light-duty trucks. In fact, MOBILE2 divides
light-duty trucks into two groups, those less than 6,000 lbs
gross vehicle weight (LDTl) and those between 6,000 and 8,500
lbs GVW (LDT2). Since light-duty trucks make up a
significantly smaller portion of the vehicle fleet than
passenger cars, less is known about the occurrence and
effects of tampering on these vehicles than on passenger cars.
Since the emission standards applicable to light-duty trucks
(LDTs) in a given calendar year are often quite different
from passenger cars, it can be expected that emission control
devices used on LDTs, such as air pumps and catalysts, will
differ in a given calendar year from those on passenger
cars. However, as the emission and fuel economy standards
-------
-37-
for light-duty trucks become more and more stringent, these
vehicles will closely resemble passenger cars with similar
emission standards. Tables 12 through 15 present the
assumptions used in this report regarding the number of
light-duty trucks equipped with various emission control
components. These estimates were taken from EPA's emission
factor samples where adequate samples were available.
In general, the per-vehicle emission benefits estimated for
passenger cars have been used for light-duty trucks using the
same emission control components. The primary differences
will be in the model years using a particular estimated
benefit. For example, only the 1979 and later LDT2s are
assumed to have been equipped with catalysts and therefore
would receive emission benefits from a catalyst inspection
program.
3.7 Oxygen Sensor Check
Many passenger vehicles beginning with the 1981 model year
utilize a computer to control vehicle engine parameters and
to optimize performance, economy and emissions under variable
operating conditions. The key to the proper operation of
these systems is the ability of the computer to accurately
sense the engine performance and operating conditions and
then alter the engine parameters to improve engine
performance. These systems are often referred to as
"closed-loop" systems. Early versions of these systems were
used in California vehicles as early as 1978. Table 10
estimates the number of these closed-loop systems in the 1981
and 1982 federal fleet from the EPA Emission Factor Sample.
Since these systems depend upon accurate sensing of engine
performance and operating conditions, malfunction of sensing
devices can cause the computer to incorrectly set engine
parameters or revert to default operating modes. Both of
these conditions can cause large increases in all three
regulated pollutants and substantial loss of fuel economy.
In particular, closed-loop vehicles require precise control
of the air/fuel mixture to optimize the operation of the
three-way catalyst used on these cars. Too much oxygen in
the exhaust impedes NOx control and not enough oxygen leads
Table 10
Model Year
Percent of Passenger Vehicles
Equipped With Closed-Loop Systems
1981
1982
65.3%
57.0%
-------
-38-
to loss of HC and CO control. Therefore, an important
sensing device for controlling HC, CO and NOx emissions is
the oxygen sensor, which can sense the amount of unreacted
oxygen in the engine exhaust. Malfunction or disconnection
of the oxygen sensor will cause partial or complete loss of
computer control of the air/fuel mixture. In some vehicles,
this causes very high emissions.
EPA has conducted tests on 20 closed-loop vehicles which have
had the oxygen sensor disabled or with the oxygen sensor
leads grounded in the laboratory, to determine the effect of
this disablement on emissions. The average increase in HC
and CO and NOx emissions is presented in Table 11.
Table 11
Increase in HC, CO and NOx Emissions
Due to Oxygen Sensor Disablement
Model Years
Increase in
HC Emissions
(gm/mi)
Increase in
CO Emissions
(gm/mi)
Increase in
NOx Emissions
(gm/mi)
1981 and Later
Closed-Loop
Vehicles
1.38
45.08
0.01
Many vehicles with disabled or malfunctioning oxygen sensors
causing high FTP emissions of HC and CO will fail the idle
test portion of an i/M inspection. In the EPA study, 45% of
closed-loop vehicles with deliberately disabled oxygen
sensors were correctly identified by a simple idle test.
However, these vehicles accounted for over 80% of the excess
HC and CO emissions from those tests. The other 55% of
disabled vehicles did not respond to the disablement with
particularly high emissions, so identifying them is less
important. A two-speed idle test or a loaded-mode test does
even better. Thus it may be appropriate to take credit for
only 20% of the excess emissions as calculated in Table 11
when an oxygen sensor check is added to an I/M program. This
percentage would vary depending on the individual program's
cutpoints.
Disablement of the oxygen sensor directly affects the
vehicle's air/fuel mixture and can readily be detected in
most important cases by an idle test. For this reason the
oxygen sensor check will be most useful for an in-use
emission control program which checks for tampering as part
of an established saftey inspection or one which does not use
-------
-39-
an idle or other exhaust test. For example, a program
surveying large commercial fleets for tampering might find an
oxygen sensor check worthwhile. Also, an oxygen sensor check
would be a useful element in a tampering check to determine
eligibility for a repair cost waiver in I/M programs which
give waivers only to untampered vehicles.
Since these closed-loop systems are relatively new to the
vehicle owning public and since they have only been in
widespread use for a few years, it is difficult to judge with
any certainty how vehicles owners will react to these
systems. Because disablement of the oxygen sensor will often
cause a loss in fuel economy, knowledgeable car owners would
be unlikely to deliberately disconnect it. However,
disconnections may occur inadvertently or because of
mi sperceptions.
Replacement costs for new oxygen sensors are high enough that
some vehicle owners may disconnect a faulty sensor rather
than incur the cost of replacement. Any faulty oxygen
sensors which were left properly connected would easily pass
a visual inspection. In most of the important cases, the
idle test portion of the I/M inspection will identify the
vehicle as needing repair.
Vehicles made by General Motors which have on-board
self-diagnostic routines within the computer system will
cause a dashboard trouble light to come on if there are
serious engine problems such as a defective oxygen sensor.
This trouble light could be checked on all GM vehicles as
part of the underhood inspection to detect properly
connected, but defective, oxygen sensors. Because the
trouble light on the dashboard can turn back off at times
even with a defective oxygen sensor, a more reliable check
would be to test the GM computer system for "trouble codes",
which are more permanently stored. This check is more
complicated than a visual inspection, but may still be
practical depending on local circumstances.
There is little knowledge as to the rate at which oxygen
sensors will be disabled. Past tampering surveys did not
check this component. The FY83 survey did look for oxygen
sensor disablements, but the results are not yet available.
It is not expected that enough observations could have been
made anyway to determine an appropriate rate of occurrence.
In the EPA Emission Factor database, only one of nearly 800
closed-loop vehicles tested had a disabled oxygen sensor and
none had dashboard "check-engine" lights indicating trouble,
although some vehicles had stored self-diagnostic trouble
codes indicating problems with the oxygen sensor. In the
absence of survey data, then, one must speculate that the
rate of disablement will be small, probably less than 1%.
-------
-40-
Replacement of a defective or missing oxygen sensor can be
fairly expensive. An oxygen sensor for a typical General
Motors closed-loop vehicle will cost about $50. Most
disablements, however, will involve simple disconnection of
the wiring to the sensor. Repair will simply mean
reconnection of the sensor. Faulty oxygen sensors will
either not be detected by the visual inspection or will be
covered by the vehicle's emission warranties for vehicles
with less than 50,000 miles.
-------
Model
Year
Percent
Air Pump &
No Catalyst
Table 12
Estimates of Technology Mix for Passenger Cars
Percent
Catalyst &
Air Pump
Percent
Catalyst &
No Air Pump
Percent EGR
& No 3-Way
Catalyst
Percent EGR
& 3-Way
Catalyst
Percent 3-Way
Catalyst
& No EGR
Pre-1968
-
-
-
-
-
-
1968
5
-
-
-
-
-
1969
5
-
-
-
-
-
1970
5
-
-
-
-
-
1971
5
-
-
-
-
-
1972
10
-
-
-
-
-
1973
30
-
-
80
-
-
1974
30
-
-
90
-
-
1975
15
30
50
90
-
-
1976
10
30
55
90
-
-
1977
10
20
65
90
-
-
1978
5
25
65
90
-
-
1979
5
25
65
90
-
-
1980
-
65
30
90
7
-
1981
-
85
15
5
85
2
1982
-
70
30
5
85
2
1983
-
60
40
5
85
2
1984
-
60
40
-
93
7
1985
-
40
60
-
93
7
1986
-
40
60
-
93
7
1987&Later
-
30
70
-
90
10
-------
Table 13
Estimates of Technology Mix for Light-Duty
Trucks Less Than 6/000 lbs. GVW
Percent
Percent
Percent
Percent EGR
Percent EGR
Percent 3-U
Model
Air Pump &
Catalyst &
Catalyst &
& No 3-Way
& 3-Way
Catalyst
Year
No Catalyst
Air Pump
No Air Pump
Catalyst
Catalyst
& No EGR
Pre-1968
.
1968
5
-
-
-
-
-
1969
5
-
-
-
-
-
1970
5
-
-
-
-
-
1971
5
-
-
-
-
-
1972
10
-
-
-
-
-
1973
30
-
-
80
-
-
1974
30
-
-
90
-
-
1975
10
30
30
90
-
-
1976
10
30
30
90
-
-
1977
10
20
55
90
-
-
1978
10
20
55
90
-
-
1979
10
40
45
100
-
-
1980
10
40
45
100
-
-
1981
-
50
50
100
-
-
1982
-
50
50
90
10
-
1983
-
50
50
80
20
-
1984
—
50
50
60
40
-
1985
-
50
50
60
40
-
1986
-
50
50
60
40
-
1987&Later
50
50
15
85
-
-------
Table 14
Estimates of Technology Mix for Light-Duty
Trucks Between 6,000 and 8,500 lbs. GVW
Percent
Model Air Pump &
Year No Catalyst
Pre-1968
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987&Later
Percent
Catalyst &
Air Pump
50
50
50
50
50
50
50
50
50
Percent
Catalyst &
No Air Pump
50
50
50
50
50
50
50
50
50
Percent EGR
& No 3-Way
Catalyst
30
30
30
30
30
30
100
100
100
100
100
60
60
60
15
Percent EGR
& 3-Way
Catalyst
40
40
40
85
Percent 3-Way
Catalyst
& No EGR
i
U)
I
-------
-44-
Table 15
Estimate of Percentages of Vehicles
Equipped with PCV and Evaporative Control Systems
Model Year Percent LDV Percent LDTl Percent LPT2
Group PCV Evap PCV Evap PCV Evap
Pre-1968 - - - -
1968-1970
1971-1978
1978 & Later
100
100 100
100 100
100
100 100
100 100
100
100
100 100
-------
-45-
4.0 METHOD FOR CALCULATION OF HC AND CO BENEFITS OF
ANTI-TAMgE&lNG AND ANTI-MISFUELING PROGRAMS
This section describes in general how benefits presented in
Section 5.0 are calculated. Specific assumptions made in
calculating the benefits are stated and explained in Section
5.0. However, the explanations there depend on the framework
presented in this section.
This section also presents the additional, or excess,
emissions caused by tampering and misfueling in the absence
of any special programs to reduce them. The purpose of doing
so is to illustrate the size of the problem to be addressed
by an anti-tampering or anti-misfueling program. This
section also illustrates the relative importance of different
forms of tampering. Section 5.0 presents estimates of how
much emission reduction is possible from different types of
programs.
4.1 Discussion of Method
The approach used in this report begins with a single model
year's vehicles. The calculation described below is
performed for each of the last 19 model years, resulting in a
total emissions impact for each from all forms of tampering
combined. These 19 model year-specific impacts are then
added using age-based vehicle miles traveled (VMT) fractions
as weighting factors to arrive at the impact on the composite
emissions of, for example, passenger cars of all ages.
The description below assumes passenger cars, but the same
procedure is used for light-duty trucks.
The calculation consists of the following steps for each
model year:
(A) Separate the model year into subgroups with distinct
combinations of equipment, such that all cars in a
subgroup are susceptible to the same types of
tampering. Specifically, cars with air pumps and
catalysts must be separated from cars with only air
pumps and cars with only catalysts, since simultaneous
air pump and catalyt tampering is possible for one
subgroup but not the others. The sales fraction for
each of these subgroups must be known; the necessary
fractions were given in Section 3.0. Because in a
single model year all cars either have or do not have
PCV and evaporative controls, and because the impacts
of PCV and evaporative tampering are strictly additive
to the impacts of misfueling, catalyst removal, and
air pump disablement, there is no need to define
subgroups based on PCV and evaporative equipment.
-------
-46-
(B) Identify all the unique combinations of tampering that
can occur on cars in each subgroup. These are as
follows:
Catalyst Air
Air Pump/Catalyst only Pump Only
1) Air Pump/Catalyst
2) Air Pump/Misfueling(lnlet)
3) Air Pump/MisfuelingjOther)
4) Air Pump/Catalyst/Misfueling(inlet)
5) Air Pump/Catalyst/Misfueling(Other)
6) Catalyst/Misfueling(inlet) X
7) Catalyst/Misfueling(Other) X
8) Air Pump Only X
9) Catalyst Only X
10) Misfueling(lnlet) Only X
11) Misfueling(Other) only X
In the above list, "(Inlet)" designates habitual
misfueling accompanied by tampering of the inlet
restrictor. "(Other)" designates habitual misfueling
accomplished by other means, such as a small pump
nozzle or a funnel. As before, PCV and evaporative
tampering can be kept separate.
(C) Find the percentage of vehicles with each of the above
unique combinations of tampering on the evaluation
date assuming no special program to reduce tampering
and misfueling. Since the tampering rates derived in
Section 2.0 depend on mileage, the odometer of the
model year on the evaluation date (always a January 1)
must be known. This report uses the same values for
this odometer variable as does the draft version of
M0BILE3. Anti-tampering programs will be evaluated by
EPA using the final version of M0BILE3. Mileage
accumulation rates and other parameters of the model
used in this report are subject to change. Given an
odometer value, the equations from Section 2.0 can be
used to calculate the overall air pump rate (AIR),
catalyst removal rate (CAT), the rate of misfueling
via inlet tampering (INLET), and the rate of
misfueling via other means (OTHER). These overall
tampering rates are the sum of the rates for two or
more of the above unique combinations of tampering.
To calculate the individual rate for each unique
combination, additional assumptions are necessary. To
fill this need, EPA has had to assume that the rate
for a given overlap combination is always proportional
to the overall rate of one or the other of the forms
of tampering that make up the overlap combination.
-------
-47-
For example, EPA has had to assume that the rate of
simultaneous air pump and catalyst tampering is 6.6%
of the overall air pump tampering rate, regardless of
any local variation in overall air pump tampering rate
or overall catalyst tampering rate. (The figure of
6.6% was determined from the 1982 Tampering Survey
data.) An exception is made if necessary to prevent a
logical contradiction? in the example given, the rate
of simultaneous air pump and catalyst tampering is
never assumed to be larger than the overall rate of
catalyst or air pump tampering determined from Section
2.0. Similar assumptions are made for other overlap
combinations. The full set of assumptions is as
follows:
Rate (1) = .066 x AIR
Rate (2) = .111 x AIR
Rate (3) = .105 x AIR
Rate (4) = .238 x CAT
Rate (5) = .032 x CAT
Rate (6) = .441 x CAT
Rate (7) = .050 x CAT
Rate (8) = AIR - [Rate( 1 )+Rate(2 )+Rate(3 )-H?ate(4)+Rate(5 ) ]
Rate (9) = CAT - [Rate(1)+Rate(4)+Rate(5)+Rate(6)+Rate(7)]
Rate (10) = INLET - [Rate(2)+Rate(4)+Rate(6)]
Rate (11) = OTHER - [Rate(3)+Rate(5)+Rate(7)]
As mentioned, alterations are made as necessary to
prevent logical contradictions that would otherwise
result in one or more of the last four rates being
negative. PCV and evaporative tampering rates come
directly from the equations in Section 2.0.
(D) Find the percentage of vehicles which had each of the
unique combinations of tampering on the day the
anti-tampering/anti-misfueling program being analyzed
began. This information is needed for step E below.
It is calculated in the same way as in the previous
step, but with appropriately lower odometer and hence
lower tampering rates.
(E) Determine the tampering rate for each unique
combination of tampering, this time assuming a
specific anti-tampering/anti-misfueling program is in
operation. This requires the information developed in
steps C and D, and assumptions regarding the
effectiveness of the specific program being analyzed.
The assumptions used in this step are stated and
explained in Section 5.0. A simple hypothetical
assumption might be that in a program that inspects
-------
-48-
only for catalyst presence, 95% of further catalyst
removal ceases, 95% of missing catalysts are replaced,
but no other tampering is corrected. Mathematically,
this can be expressed as follows:
(Tampering Rate With Program) =
f(Tampering Rates Without Program)
or,
With Program Without Program
Rate
(1)
0.05 Rate
(1)
Rate
(2)
=
Rate(2)+ 0
.95
Rate(4)
Rate
(3)
=
Rate(3)+ 0
.95
Rate( 5)
Rate
(4)
=
0.05 Rate
(4)
Rate
(5)
=
0.05 Rate
(5)
Rate
(6)
=
0.05 Rate
(6)
Rate
(7)
—
0.05 Rate
(7)
Rate
(8)
=
Rate(8)+ 0
.95
Rate(1)
Rate
(9)
=
0.05 Rate
(9)
Rate
(10)
=
Rate(10)+
0.95
Rate(6)
Rate
(11)
=
Rate(11)+
0.95
Rate(7)
Where the new rates on the left are with the catalyst
check program and the old rates on the right are from
step C. The equations express the fact that most
vehicles which formerly suffered from catalyst removal
and another form of tampering now suffer only from the
other form since the missing catalyst is replaced.
In this example, the tampering rates with the catalyst
inspection program depend only on the "without" rates
on the evaluation date, i.e., the results of step C.
For some program types, the tampering rate with a
program depends on the results of both steps C and D.
In the hypothetical example, deterrence of catalyst
removal might have been assumed more effective than
replacement of already missing catalysts. The
catalyst removal rate with the program would then
depend on the level of removal when the program
started, step D.
As before, PCV and evaporative tampering are kept
separate.
Assign each unique combination of tampering an
emissions impact per vehicle. The impacts are taken
from Section 3.0, with the following further
assumptions regarding cases of simultaneous tampering.
-------
-49-
o The impact of simultaneous catalyst removal and
either or both of misfueling or air pump tampering
is the same as stated in Section 3.0 for catalyst
removal alone.
o The impact of simultaneous misfueling and air pump
tampering is the same as stated in section 3.0 for
misfueling above.
(G) Multiply tampering rate by tampering impact for each
unique combination, and add the result for all
combinations taking into account the sales split
between the air pump only subgroup, the air
pump/catalyst-equipped subgroup and the catalyst-only
subgroup. Add to this the rate-times-irapact result
for PCV and evaporative tampering. The sum is the
excess emissions due to the tampering and misfueling
that remain despite the operation of the
anti-tampering/anti-misfueling program being analyzed.
(H) Repeat the calculation skipping step E, i.e., assuming
no program in operation. This gives the excess
emissions in the absence of any specia-L program.
(I) Subtract the result of step G from that of step H.
The difference is the benefit of the program for the
one model year in question.
Composite excess emissions ahd the composite benefit can be
calculated by weighting each model year by its age based VMT
fraction, also known as its travel fraction. This can be
done at steps G and H or at step I, depending on the desired
outcome.
The method described above assumes that the user of the
result of the calculation is interested in a situation in
which vehicles are driven under standard conditions of
temperature, speed, etc. All of the benefits shown in this
document assume such a situation as well. It is possible to
analyze other situations if correction factors for
non-standard conditions are applied at an appropriate step in
the calculation. EPA's forthcoming computer program MOBILE3
will have this capability, although in the standard version
of MOBILE3 step E above will be bypassed. Users interested
in performing calculations with MOBILES which would require
step E should contact EPA.
The method described above applies fox both I/M and non-l/M
areas, provided either I/M or non-l/M tampering rates are
used consistently.
-------
-50-
The reader should note that in contrast to M0BILE3, its
predecessor MOBILE2 does not internally calculate excess
emissions from tampering and misfueling, with or without a
program to control them. MOBILE2 calculates only a single
estimate of in-use emissions and does not identify which
portion is attributable to tampering and misfueling. The
analysis behind M0BILE2 did explicitly account for and
identify some but not all of what is now recognized as the
effect of tampering and misfueling, and the data base used to
develop MOBILE2 implicitly accounts for some further but
unknown and unidentifiable portion of the actual effect.
Consequently, consistent and reliable analysis of the effects
of tampering and misfueling and programs to reduce them are
only possible with MOBILE3. Until MOBILE3 is available, one
may assume that M0BILE2's predictions of in-use emissions is
correct, and may subtract from the MOBILE2 estimate the
appropriate program benefits shown in this report to arrive
at an estimate of in-use emissions with a
tampering/misfueling control program. Once MOBILE3 is
available, it must be used (by itself with EPA assistance or
in conjunction with the pre-calculated benefits in this
report) unless it can be shown that use of MOBILE3 would
cause unreasonable delays in analysis and adoption of air
quality improvement measures.
4.2 Example Calculation
4.2.1 One Model Year
This example will calculate the excess emissions due to
tampering and misfueling for the 1977 model year. We will
assume that the vehicles are located in a non-l/M area, and
we will use the national average tampering and misfueling
rates described in Section 2.0. In order to show how an
anti-tampering program affects the excess emissions, we will
assume that an annual catalyst presence inspection began on
January 1, 1984. We will evaluate all excess emissions for
January 1, 1988.
On average, the 1977 model year is estimated to have
accumulated 76,998 miles by January 1, 1984 and 105,156 miles
by January 1, 1988. Using these mileages and the rate
equations from Section 2.0 the overall rates of tampering and
misfueling can be estimated. These are presented in Table 16.
-------
Table 16
Example Calculation
of Tampering and Misefulinq Rates
Non-I/M Area LDV Rate at Rate at
Rate Equations (Percent) program Start Evaluation
System Zero-Mile LeveT Increase/IOK miles (76,998 miles) (105,156 miles)
Air Pump -2.71 2.652 .177 .252
Catalyst -1.75 1.611 .105 .150
Fuel Inlet -1.43 2.022 .141 .198
Other
Misfueling 1.65 0.559 .060 .075
PCV 0.02 0.248 .019 .026
Evaporative -0.48 0.335 .021 .030 £
I
-------
-52-
As outlined in Section 4.1, these overall rates are used to
estimate the size of the 11 overlap categories. Category 12
represents untampered vehicles. These categories do not
include PCV and evaporative canister tampering, which are
addressed later in this Subsection. For HC and CO excess
emissions there are three technology types of interest; air
pump only, catalyst only and air pump with catalyst.
Vehicles with air pumps but without catalysts have no
overlaps and, therefore, have the air pump tampering rate in
air pump only (tampering category 8) , and zero rates in all
other categories. This could also be done by adding
categories 1 through 5 to category 8 and categories 6, 7, 9,
10 and 11 are added to category 12 (no tampering or
misfueling). The categories are eliminated and their rates
shifted to other categories in order to remove categories
which cannot occur with the given technology type. For
example, category 1 represents vehicles with disabled air
pumps and missing catalysts. If we are analyzing vehicles
equipped with air pumps and without catalysts, this category
is meaningless. To correctly account for all air pump
disablement, however, the vehicle rate assigned to category 1
must be added to category 8 (air pump disablement only) .
Vehicles with catalysts, but without air pumps, have zero
rates in categories 1 through 5 and category 8 since these
require the possibility of air pump tampering. The rates in
categories 1, 2, 3, 4, 5, and 8 are added to categories 9,
10, 11, 6, 7, and 12 respectively. Vehicles with air pumps
and catalysts have rates in all 11 categories. Using the
equations described in Section 4.1(C) and (E) and rates
estimated in Table 16 the 11 category sizes can be
determined. These are presented in Table 17. The category
sizes for air pump only and catalyst only vehicles can be
derived from the rates in Table 17.
-------
Table 17
Example Calculation of
Overlap Categories
(Catalyst vehicle equipped with air pumps only)
Overlap
Category
At Program
Category
Descr iption*
Equation
Start
(1)
AIR/CAT
. 066*AIR
.0117
(2)
AIR/NCK
. 111*AIR
.0196
(3)
AIR/OTHR
. 105*AIR
.0186
(4)
AIR/CAT/NCK
. 238*CAT
.0250
(5)
AIR/CAT/OTHR
. 032*CAT
.0034
(6)
CAT/NCK
. 441*CAT
.0463
(7)
CAT/OTHR
.050*CAT
.0053
(8)
AIR
AIR-(1-5)
.0987
(9)
CAT
CAT-(1,4,5,6,7)
.0133
(10)
NCK
NCK-(2,4,6)
.0501
(11)
OTHR
OTHR-(3,5,7)
.0327
Category Size
At Evaluation
.0166
.0280
.0265
.0357
.0048
.0662
.0075
.1404
.0192
.0681
.0362
k AIR: Air Pump Disabled
CAT: Catalyst Removed
NCK: Misfueling by Enlarging Fuel Filler Inlet
OTHR: Other Misfueling
-------
-54-
The excess emissions from this model year without the proqram
can be estimated from the evaluation date estimates of
tampering and misfueling rates from Table 17. First, the
emission impact of each of the categories must be
determined. Since all of the 1977 model year uses oxidation
catalyst technology, the emission impact of air pump
disablement, catalyst removal and misfueling can be taken
directly from the appropriate tables in Section 3.0. For
simplicity only total HC emissions will be addressed in this
example. It is assumed that the effect of catalyst removal
supercedes all other tampering and misfueling effects,
therefore the overlap categories 1, 4, 5, 6, 7, and 9 which
all contain catayst removal would experience the emission
impact of catalyst removal. The effect of misfueling is
assumed to supercede the emission impact of air pump
disablement. The overlap categories 2, 3, 10, and 11 which
all contain misfueling but without catalyst removal would
experience the emission impact of misfueling. Only category
8, which contains only air pump disablements experiences the
air pump disablement emission impact. These emission impact
groups are summed in Table 18.
The excess emissions due to tampering and misfueling are
determined by multiplying the size of each emission impact
group times the appropriate excess emission estimate from
Section 3.0. The three technology types are then weighted by
their fleet fractions also from Section 3.0 and summed for
the combined excess emissions from air pump, catalyst, and
misfueling. This calculation is presented in Table 18.
-------
Table 18
Example Calculation of Emission Impact
Technology
Type
Air Pump
With
Catalyst
Air Pump
Only
Catalyst
Only
Emission
impact Groups
Air Pump Disabled
Catalyst Removed
Misfueled
Air Pump Disabled
Catalyst Removed
Misfueled
Overlap
Categories
(8)
(1,4,5,6,7,9)
(2,3,10,11)
(1-5,8)
(1,4-7,9)
(2,3,10,11)
(A)
Emission
Impact
Group Size
.1404
.1500
.1588
.2520
.1500
.1588
(B)
Excess
Total HC
Emissions
(gm/mi)
1.37
3.05
2.47
1.37
3.05
2.47
(C)
Technology
Fleet
Fraction
.20
.20
.20
.10
.55
.55
Composite
Emission
Impact
(A) * (B) * (C)
.038
.092
.078
.035
.252
.216
Total Emission Impact
0.71 gm/mi
-------
-56-
PCV and evaporative canister tampering effects are assumed
not to overlap with any of the other tampering and
misfueling. As a result the excess emissions due to these
types of tampering can be determined by simply multiplying
together the evaluation date rate estimated from Table 16 and
the appropriate excess emissions and technology type fleet
fraction from Section 3.0. This calculation is presented in
Table 19.
Table 19
Example Calculation of Excess Emission
from PCV and Evaporative Canister Disablements
(A)
Tamper ing
System Rate Fraction
PCV .026
Evap .030
(B)
Excess Total
HC Emissions
(gm/mi)
3.44
1.01
(C)
Technology
Fleet
Size
Factor
1.0
1.0
Composite
Emission Impact
(A) * (B) * (C)
.089
.030
Total Emissions Impact
.012 gm/mi
The effect of an annual ' catalyst presence inspection is
estimated by reducing the size of the appropriate overlap
category sizes to reflect assumptions about how effective the
program would be in reducing catalyst removal. In the
hypothetical case of Section 4.1, 95% of catalyst removals
which occurred previous and would occur subsequent to the
start of the inspection program are assumed to be replaced or
deterred from occurring. Later sections will examine the
issue of program effectiveness in more detail. The
effectiveness of a catalyst inspection program used in the
example will be the 95% assumed in later sections.
In some cases the effectiveness of an inspection program will
differ depending on whether the tampering and misfuelinq
occurs before or after the program begins. This is because,
in some cases, it can be assumed that a program will be more
effective in deterring tampering and misfueling which has not
yet occurred, than it can be in detecting and permanently
correcting tampered or misfueled vehicles. As a result two
sets of overlap categories representing tampering and
misfueling which occurred before and after the program start
must be calculated. The "previous" category sizes are the
cateogry sizes at the program start. The "subsequent"
category sizes can be calculated by subtracting the
-------
-57-
"previous" category sizes from the category sizes at the
evaluation date. The previous and subsequent overlap
category sizes are presented in Table 20.
Table 20
Example Calculation of
Previous and Subsequent Tampering
and Misfueling Category Sizes
(A)
(B)
Overlap
Previous Rates
Rates at
Subsequent Rates
Category*
(At Program Start)
Evaluation
[(A) - (B)1
(1)
.0117
.0166
.0049
(2)
.0196
.0280
.0084
(3)
.0186
.0265
.0079
(4)
.0250
.0357
.0107
(5)
.0034
.0048
.0014
(6)
.0463
.0662
.0199
(7)
.0053
.0075
.0022
(8)
.0987
.1404
.0417
(9)
.0133
.0192
.0059
(10)
.0501
.0681
.0180
(11)
.0327
.0362
.0035
*See Table 17 for description.
Since in this example the effect of the inspection program on
previous and subsequent tampering and misfueling rates is
identical, for simplicity the example will apply the
effectiveness factors to the sum of the previous and
subsequent rates. Normally two separate calculations are
required.
As described in Section 4.1(E), if the catalyst inspection
program is assumed to be 95% effective, then categories 1, 4,
5, 6, 7, and 9 must all be reduced to 0.05 of their previous
sizes. However, since no other tampering or misfueling is
affected by the inspection program, 95% of the rates in
categories 1, 4, 5, 6, and 7 are added to categories 8, 2, 3,
10, and 11 respectively. This process is shown in Table 21.
-------
-58-
Table 21
Example Calculation of Resultant
Category Sizes With an Annual Inspection Program
Resultant
Overlap Category
Resultant
Category*
Adjustment
Category Size
(1)
.05 * (1)
.0008
(2)
(2) + .95 * (4)
.0619
(3)
(3) + .95 * (5)
.0311
(4)
.05 * (4)
.0018
(5)
.05 * (5)
.0002
(6)
.05 * (6)
.0033
(7)
.05 * (7)
.0004
(8)
(8)
.1404
(9)
.05 * (9)
.0010
(10)
(10) + .95 * (6)
.1310
(11)
(11) + .95 * (7)
.0433
*See Table 17 for descriptions.
Once the category sizes have been adjusted to reflect the
inspection program, the previous and subsequent categories
can be added together, grouped by emission impact, and
multiplied by the excess emissions estimated for each group,
as was done to estimate the excess emissions without the
program in Table 18. This process is shown in Table 22.
-------
Table 22
Technology
Type
Air Pump
With
Catalyst
Air Pump
Only
Catalyst
Only
Example Calculation of Emission impact
With an Annual Catalyst Inspection Program
Emission
Impact Groups
Air Pump Disabled
Catalyst Removed
Misfueled
Overlap
Categor ies
(8)
(1,4-7,9)
(2,3,10,11)
(A)
Emission
Impact
Group Size
.1404
.0075
.2673
(B)
Excess
Total HC
Emissions
(gm/mi)
1.37
3.05
2.47
(C)
Technology
Fleet
Fraction
.20
.20
.20
Composite
Emission
Impact
(A) * (B) * (C)
.038
.005
.132
Air Pump Disabled
(1-5,8)
.2520
1.37
10
035
Catalyst Removed (1,4-7,9) .0075 3.05 .55 .013
Misfueled (2,3,10,11) .2673 2.47 .55 .363
Total Emission Impact 0.59 g/mi ^
<£>
I
-------
-60-
Table 23 compares the excess emissions from this single model
year with and without the catalyst presence inspection
program. This process is done for each model year and
weighted together by the vehicle mileage fraction contributed
by each model year to the fleet mileage accumulation to give
a composite excess emission estimate. This is shown in the
next Subsection.
Table 23
Comparison of Excess Emissions With and
Without an Annual Catalyst Inspection Program
Excess Emissions (gm/mi)
Without with
Program Program
Excess Emissions from Air
Pump Disablement, Catalyst
Removal and Misfueling 0.71 0.59
Excess Emissions from PCV
and Evaporative Canister
Disablements 0.12 0.12
Total Excess Emissions 0.83 gm/mi 0.71 gm/mi
4.2.2 Composite of All Model Years
Once an estimate of the excess emission due to tampering and
misfueling has been made for each of the last 19 model years,
these estimates are weighted together by the vehicle mileaae
fraction contribution of each model year to the fleet mileaqe
accumulation. The sum of the weighted excess estimates then
the composite vehicle excess emissions with and without an
inspection program. The difference between these sums the
emission reductions which can be attributed to the program.
Table 24 is an example of the calculation of the proaram
benefit of inspection for catalyst presence on all 1975 and
later vehicles.
4.3 Emissions Due to Tampering and Misfueling: All Types
Tables 25 and 26 present the estimates of excess emissions on
January 1, 1988, due to all forms of tampering and habitual
misfueling using the estimates of tampering and misfueling
rates as discussed in Section 2.0 and the increases in
emissions due to tampering and misfueling from Section 3.0.
-------
-61-
Table 24
Example Calculation of Composite
Emissions Benefit From All Model Years*
(a)
(b)
(c)
(a)[(b)-(c)]
VMT Fraction
Excess
Emissions (gm/mi)
Emission Reduction
Model
in Evaluation
Wi thout
With Catalyst
Due to Inspection
Year
Year (1/1/88)
Program
Inspection Program
Program
1970
.001
1.33
1.13
.00020
1971
.001
1.26
1.07
.00019
1972
.003
1.19
1.01
.00054
1973
.007
1.12
0.95
.00119
1974
.011
1.05
0.89
.00176
1975
.018
0.98
0.83
.00270
1976
.025
0.90
0.77
.00325
1977
.031
0.83
0.71
.00372
1978
.045
0.75
0.64
.00495
1979
.057
0.67
0.58
.00513
1980
.067
0.60
0.51
.00603
1981
.075
0.52
0.45
.00525
1982
.095
0.44
0 .38
.00570
1983
.113
0.37
0.32
.00565
1984
.104
0.30
0.26
.00416
1985
.083
0.22
0.19
.00249
1986
.109
0.15
0.13
.00218
1987
.120
0.08
0.07
.00120
1988
.028
0.00
0 .00
.00000
.993**
Total Benefit
.056 gm/mi
* This table is for illustration purposes only and should not be used to
determine program benefits.
**VMT fractions do not sum to 1.000 because model years older than 19
years account for the remaining VMT. No excess or benefit is assumes for
these model years for calculational simplicity.
-------
-62-
As discussed earlier these results have not been adjusted for
non-standard conditions. Table 25 assumes that there is no
I/M program in the area of interest, while Table 26 assumes
the existence of an I/M program. For comparison, M0BILE2
predicts that without I/M on January 1, 1988 the total
composite emissions from these vehicles to be:
MOBILE2
HC
CO
Passenger Cars
2.4 2 gm/mi
27.47 gm/mi
Light-Dutv Trucks:
( 6000*lbs)
(6000-8500 lbs)*
2.59 gm/mi
1.57 gm/mi
24.80 gm/mi
14.11 gm/mi
~These heavier trucks emit more HC and CO emissions than
passenger cars or the lighter trucks of the same model year,
however, MOBILE2 assumes that the majority of the vmt
accumulated by these trucks is accumulated by the new (and
cleaner) model years so that this composite number shows a
lower contribution than would occur if the distribution of
VMT were similar to the passenger cars.
These MOBILE2 emission levels, however, assume only an 8%
rate of misfueling and contain much smaller rates of
tampering than observed in the tampering surveys.
Section 5.0 will discuss how anti-tampering and anti-
misfueling programs can reduce the excess emissions and will
estimate the benefits of these programs.
The method described in subsection 4.1 aggregates the effects
of all forms of tampering for one model year and then
combines model years. For Tables 25 and 26, the method was
modified to provide the reader not only with the all-forms,
all-model years composite excess emission results, but with
additional detail on the contributions of different types of
tampering to the total.
The excess emissions from each of the 11 unique combinations
of tampering and misfueling, plus the excess from PCV and
evaporative tampering, were aggregated separately across
model years before being combined together with each other.
Tables 25 and 26 do not show 13 separate categories of
tampering, however, in the interest of comprehension. All of
the excess emissions from the six tampering combinations
which involve catalyst removal alone or in conjunction with
other forms have been added together and appear with the
-------
-63-
heading of "catalyst removal" in Tables 25 and 26. The
remaining four categories involving misfueling are shown
together under the "misfueling" heading. The category of air
pump tampering only appears under the "air pump tampering"
heading.
Table 25
Per Vehicle Excess
Emissions Due to Tampering and
Misfueling in Non-l/M Areas (January 1, 1988)
Emi ssion
Control
Passenger Car
Composite Per Vehicle
Increase in Emissions (mg/mi)
Light-Duty Truck
(0-6000 lbs)
(6000-8500 lbs)
Component
HC
CO
HC
CO
HC
CO
Air Pump*
30.20
765.89
88.19
1998.97
56.26
1280.56
Catalyst
133.70
1297.91
627.29
5846.52
562.84
5246.07
Misfueling**
151.37
1184.71
364.68
3044.70
325.74
2719.70
PCV System
38.06
0.00
157.84
0.00
229.49
0.00
Evaporative
Cani ster
18.27
0.00
150.89
0.00
96.29
0.00
Totals(mg/mi)
371.59
3248.51
1388.88
10890.21
1270.62
9246.33
Totals(gm/mi)
0.37
3.25
1.39
10.89
1.27
9.25
Tons***
301.80
2638.40
237.00
1858.29
152.95
1113.01
~Because some of the vehicles with disabled air pumps also had
catalysts removed or had been misfueled, the excess emissions due
to the overlap has been removed from the air pump category to
avoid double counting.
**Because of the overlap between catalyst removal and misfueling,
the excess emissions due to the overlap have been removed from the
misfueling category to avoid double counting.
***An estimate of annualized tons calculated assuming a fleet of
100,000 vehicles of all types and using M0BILE3 estimates of
passenger car and light-duty truck vehicle miles traveled (LDV:
0.614, LDTl: 0.129, LDT2: 0.901). Each vehicle is assumed to have
traveled 12,000 miles in the last year.
-------
-64-
Table 26
Emission
Control
Per Vehicle Excess
Emissions Due to Tampering and
Misfueling in I/M Areas (January 1, 1988)
Composite Per Vehicle
Increase in Emissions (mg/mi)
Passenger Car
Light-Duty Truck
(0-6000 lbs) (6000-8500 lbs)
Component
HC
CO
HC
CO
HC
CO
Air Pump*
13.50
342.74
30.49
688.09
19.21
434.16
Catalyst
46.19
451.26
166.72
1553.10
149.33
1391.17
Mi sfueling**
92.95
717.55
242.26
2013.11
219.57
1824.29
PCV System
38.06
0.00
157.84
0.00
229.49
0.00
Evaporative
Canister
18.27
0.00
150.89
0.00
96.29
0.00
Totals(mg/mi)
208.97
1511.54
748.21
4254.30
713.88
3649.61
Totals(gm/mi)
0.21
1.51
0.75
4.25
0.71
3.65
Tons***
169.72
1227.65
127.67
725.95
85.93
439.31
*Because some of the vehicles with disabled air pumps also
had catalysts removed or had been misfueled, the excess
emissions due to the overlap has been removed from the air
pump category to avoid double counting.
~~Because of the overlap between catalyst removal and
misfueling, the excess emissions due to the overlap have been
removed from the misfueling category to avoid double counting.
***An estimate of annualized tons calculated assuming a fleet
of 100,000 vehicles of all types and using M0BILE3 estimates
of passenger car and light-duty truck vehicle miles traveled
(LDV: 0.614, LDT1: 0.129, LDT2: 0.091). Each vehicle is
assumed to have traveled 12,000 miles in the last year.
-------
-65-
5.0 BENEFITS OF ANTI-TAMPERING AND ANTI-MISFUELING PROGRAMS
This section estimates the benefits of anti-tampering and
anti-misfueling programs using the data and method described
in previous sections. As discussed in the previous sections,
the benefits of anti-tampering and anti-misfueling programs
will depend on three major factors.
These are:
o The rate of tampering and misfueling in the area.
o The amount of excess emissions caused by tampering and
misfueling.
o The effectiveness of the program in eliminating
tampering and misfueling.
The rate of tampering and misfueling was addressed in Section
2.0. The amount of excess emissions caused by tampering was
discussed in Sections 3.0 and 4.0. This section will discuss
the effectiveness of specific anti-tampering and
anti-misfueling programs and estimate their benefits in both
I/M and non-I/M areas.
There are several factors which influence the effectiveness
of anti-tampering and anti-misfueling programs:
o The perceived incentives for tampering and misfueling.
o The ability of the program to detect tampering and
misfueling.
o The size of the penalty for tampering and misfueling.
o The extent of enforcement action to assure that the
program operates as designed.
o The number of vehicle owners who continue to tamper or
misfuel after the program begins.
o The rate of inadvertent disablements.
Each of the following sections will address these issues and
decide on an appropriate level of effectiveness for each type
of disablement and each program design in both I/M and
non-I/M areas.
In order to claim the full benefits estimated in the tables
in this section the program must require the following
elements to assure operation as designed. Programs lacking
-------
-66-
some of these elements are feasible but would require
individual evaluation.
o inspector training.
o A method to assure vehicle owner compliance with the
program requirements.
o A method to determine which vehicles require which
emission control components.
o Data collection to monitor the program and identify
bad actors among inspectors, inspection stations, and
repair facilities.
o Periodic audits of inspection stations in
decentralized programs to verify inspector proficiency
and compliance with other program requirements.
o Enforcement actions such as using an "unmarked" test
car in decentralized programs to assure inspector
compliance with program rules.
o A fraud-resistant referee system for decentralized
programs to resolve disputes.
o A public awareness program.
The above design requirements are intended to prevent
deliberate cheating. Centralized programs, by their design,
should be able to prevent cheating more easily than
decentralized programs. The credits calculated in this
report assume that there will be no significant amount of
cheating in the inspections. EPA will evaluate
anti-tampering programs for their ability to prevent cheatina
before agreeing to allow credits for the program. If EPA
review of the program design suggests that significant but
unquantifiable cheating could still occur, reduced or no
credits would be qiven.
Public acceptance of a vehicle inspection program which
requires catalyst replacement where misfueling is indicated
will be improved if there is a visible program to also
require compliance with fuel regulations on the part of
retail gasoline outlets. The Plumbtesmo test may fail a
vehicle whose only use of leaded fuel was inadvertent due to
contamination or mislabeling at the pump. EPA does not have
enough information to predict the conditions under which this
might occur. It is important that these occurrences be
minimized for equity reasons. Therefore, EPA advises that if
a State or local area intends to use the Plumbtesmo test to
-------
-67-
detect misfueling, there should also be a program of
unscheduled periodic inspections of retail gasoline outlets.
This program should inspect the diameter of fuel pump
nozzles, determine that the pumps are properly labeled, and
analyze the lead content of the fuel being sold.
Benefits from anti-tampering and anti-misfueling programs are
obtained by addressing two problems: (1) existing tampering
and misfueling and (2) the tampering and misfueling which has
not yet occurred. Existing tampering and misfueling can only
be addressed by identifying tampered and misfueled vehicles
and requiring their repair. Tampering and misfueling that
have not yet occurred can be detected when they do occur or
can be prevented from occurring by the assurance of detection
and penalty in the program. Tampering and misfueling which
have already occurred are calculated as the rate of
occurrence at the start date of the tampering inspection
program (assumed to be January 1, 1984 for the benefits
presented here). The tampering and misfueling which will
occur between the program start date and the evaluation year
without the intervention of the inspection program is the
difference in the rates calculated for the start date of the
program and the evaluation date, assuming no program.
5.1 I/M Programs
I/M programs offer a good opportunity to address the
tampering issue. Although the available data show that I/M
programs affect the incidence of tampering and misfueling to
some extent without any special activity, substantial amounts
of tampering can continue. Fortunately, the fact that large
segments of the fleet are periodically inspected provides a
low cost, convenient opportunity to specifically check for
tampering and misfueling. Some I/M programs have seen the
advantages in expanding the inspection and already include a
check for tampering.
Section 2.0 discussed the effect of I/M on tampering rates.
The I/M rates discussed in that Section are the rates used
for all calculations in this Section, except that overlap
among tampering types is accounted for as described in
Section 4.0. The individual vehicle benefits and costs of
repairs of tampering and misfueling are those discussed in
Section 3.0. The methodology explained in Section 4.0 was
used to calculate excess emissions due to tampering and
misfueling. Only annual and biennial programs are considered
in this section.
-------
-68-
5.1.1 Program Effectiveness
For periodic inspection programs, such as I/M programs, it is
assumed that the program will require repair or replacement
of the disabled emission control components once they are
discovered, followed by reinspection of the vehicle and/or
the repair receipts to verify compliance.
The assumptions used to calculate benefits for inspection of
individual components and combinations of components are
explained and justified below. Section 5.1.2 then presents
the results of the calculation of benefits. The details of
the calculation are not presented. For all components,
benefits are shown separately for 1984 and later vehicles,
for 1980 to 1983 vehicles, and for older vehicles, for the
convenience of jurisdictions which plan to inspect only 1980
and later or 1984 and later vehicles. Jurisdictions
interested in other model year cutoffs may interpolate for a
rough approximation and may consult with EPA for an exact
calculation.
The only site in the 1982 EPA tampering survey which has an
anti-tampering inspection is the Portland, Oregon site.
Portland has also had a biennial I/M program since 1975. The
fact that Portland has an anti-tampering program is
presumably most of the reason why Portland has a lower
tampering rate than any of the other I/M sites in the 1982
survey. However, other factors, such as local behavior, the
stringency and age of the I/M program, and the age of the
anti-tampering program itself probably all contribute to the
effectiveness observed in Portland. Also, the survey in
Portland was conducted at the I/M inspection site. Vehicle
owners presenting their vehicles for inspection knew
beforehand that their vehicle would be inspected for
tampering and that they would be required to repair any
tampering before they could register their vehicle. It is
possible, therefore, that a few vehicle owners repaired their
vehiclestampering just before presenting their vehicles for
inspection. This would cause the survey to underestimate the
actual rate of tampering and misfueling in Portland if some
retampering occurs between inspections. In addition, some of
the vehicles surveyed probably were at the I/M station for
reinspection following repair, of tampering discovered in a
previous inspection. Comparison to Portland is, therefore,
used only as a guide to estimate the effectiveness of
anti-tampering programs in other areas. All comparisons to
Portland made in the following subsections compare the mixed
group of cars and trucks in Portland to the mixed group of
cars and trucks in other specified areas.
-------
-69-
5.1.1.1 Positive Crankcase Ventilation (PCV) and
Evaporative System Inspection
The inspection for the PCV system is quite simple. The
inspector need only assure that the PCV valve and connecting
hose to the carburetor are both present and connected. The
evaporative control system is more complicated. The canister
may be located somewhere other than in the engine
compartment, misleading an inspector into thinking it has
been removed or encouraging the inspector not to check hose
connections at the canister. . Often there are spaces for
extra connections on the canister which are unused even when
it is properly connected. A false failure can be avoided by
checking the hose routing diagram attached underneath the
hood. it is advisable for programs which check the
evaporative canister to also require an OEM gas cap to be
present. Althouqh the rate of missing gas caps is small, the
evaporative control system does not work properly without
it. Generic gas caps are often used to replace lost OEM
caps, but they may fail to make the tight seal required for
effective control of evaporative emissions. Similiar
problems may exist with replacement locking type gas caps
which are popular in some areas.
In Portland, the rate of disabled PCV systems is 56% less
than in the other nine site's in the survey. The rate of
evaporative canister tampering is 57% less. This difference
is assumed to be entirely due to the tampering check
performed in Portland as part of the biennial I/M program.
An annual inspection is expected to reduce the number of
disablements even more, so an annual PCV check is assumed to
be 70% effective and an annual evaporative canister check is
assumed to be 70% effective.
The rather low effectiveness values (56% for PCV and 57% for
evaporative) observed in Portland are somewhat surprising but
can be explained. In the case of the evaporative canister
and the PCV system, it can be speculated that many
disablements are inadvertent, since there is virtually no
incentive for vehicle owners to deliberately disconnect these
devices. Moreover, the penalty, reconnection or replacement,
is so inexpensive that there is little incentive to repair
the systems between inspections even if the owner is aware of
the disablements. Consequently, deterrence of these two
forms of tampering is probably low. The Portland inspectors
also may not be 100% accurate in the inspections for PCV and
evaporative systems.
Benefits from a PCV or evaporative canister inspection are
reductions in non-exhaust emissions and, therefore, can be
added to any of the other inspections which offset exhaust
-------
-70-
emissions. The benefits from these inspections are
unaffected by the presence or absence of the other
inspections discussed below.
5.1.1.2 Catalyst-Only Inspection
Inadvertent removal of catalysts does not occur. Therefore,
if the public is well informed that failure of the catalyst
check will require catalyst replacement, one can expect that
there will be few new instances of catalyst removal. Such
public awareness should be nearly automatic in an annual
program. The exception, if any, will be a small group of
owners convinced beyond persuasion that their catalysts
should be removed. Such owners may reinstall the catalyst as
necessary in order to pass the inspection, or may remove the
active material from the catalyst container making visual
detection of the disabled catalyst nearly impossible.
In addition to some catalysts being successfully removed or
disabled in a way that escapes detection, inadvertent
inspector errors may result in failure to replace all
catalysts missing at the start of the program. Inspector
errors can be minimized by proper training and access to good
reference material. Materials are available which list the
emission control equipment required on vehicles (Colorado
State University, 1968-1982 Automotive Emissions Systems
Applications Guide [4]) .
It is true that in the 1982 tampering survey, no catalyst
removals were observed in over 300 inspections at the
Portland site. Since the Portland program has been in
operation since the advent of catalyst equipped cars, this
indicates that the catalyst inspection can effectively
prevent vehicle owners from removing catalysts, except
perhaps for a few owners who reinstall the catalyst each time
to pass inspection or remove the active material. This
deterrence can be achieved with a program which provides a
reasonably high probability of detection. The Portland
observation is not inconsistent with an assumption that
inspections will not be quite 100% accurate.
For the reasons discussed above, an inspection for removal of
the catalyst will be assumed to be 95% effective in detecting
and forcing replacement of catalysts. The 95% value allows
for some inspection errors and some concealed tampering and
retampering by owners. A biennial inspection program is
assumed to be as effective as an annual inspection.
The effectiveness of 95% applies to the group of vehicles
whose only form of tampering is catalyst removal, and it
applies both to removals which had occurred before program
-------
-71-
start-up and to those which would otherwise have occurred
between start-up and the evaluation date. Vehicles which
have both their catalyst removed and one or more other forms
of tampering require special discussion, which follows.
There are five groups of multiple tampered vehicles of
interest here: catalyst removed/misfueled(inlet); catalyst
removed/misfueled(other): catalyst removed/misfueled
(inlet)/air pump disabled: catalyst removed/misfueled
(other)/air pump disabled, and catalyst removed/air pump
disabled. EPA assumes that 5% of the vehicles in each of
these groups will remain there despite an inspection for
catalyst presence, due to errors and concealed catalyst
tampering or retampering. Of the catalyst removed/misfueled
(inlet) and the catalyst removed/misfueled(other) groups, the
other 95% are assumed to enter the corresponding
misfueled-only group. In other words, owners of these
vehicles are forced to replace their catalysts (or are
deterred from removing them after inspections begin), but
they continue to misfuel thereby poisoning the new
catalysts. A poisoned catalyst is only slightly more
effective than no catalyst, according to Section 3.0.
Of the catalyst removed/misfueled (inlet)/aj.r
and the catalyst removed/misfueled(other)/air
groups, 95% are assumed to enter the
misfueled/air pump disabled group. Of
pump disabled
pump disabled
corresponding
the catalyst
removed/air pump disabled group, 95% are assumed to enter the
air pump disabled-only group. Except for these new entrants,
the rates for other forms of tampering are unaffected.
The following table presents these assumptions in concise
form. The format of this table is used for later inspection
types also.
Overlap Category
1) Air Pump/Catalyst
2) Air Pump/Misfueling
(Inlet)
3) Air Pump/Misfueling
(Other)
4) Air Pump/Catalyst/
Misfueling (Inlet)
Effect of Catalyst Check
95% become air pump only.
5% remain air pump/catalyst.
100% remain air pump/misfueling
(Inlet).
100% remain air pump/misfueling
(Other).
95% become air pump/misfueling
(Inlet).
5% remain air pump/catalyst/
misfueling (Inlet) .
-------
-72-
5) Air Pump/Catalyst/ 95%
Misfueling/(Other)
5%
6) Catalyst/Misfueling 95%
(Inlet)
5%
7) Catalyst/Misfueling 95%
(Other) 5%
8) Air Pump Only 100%
9) Catalyst Only 95%
5%
10) Misfueling (inlet) 100%
Only
become air pump/misfueling
(Other).
remain air pump/catalyst/
misfueling (Other) .
become misfueling (inlet)
only.
remain catalyst/misfueling
(inlet) .
become misfueling (other) only,
remain catalyst/misfueling
(other).
remain air pump only,
become OK.
remain catalyst only.
remain misfueling (inlet) only.
11) Misfueling (Other) 100% remain misfueling (other) only.
Only
Some owners who have removed their catalysts have probably
done so thinking it would harm their vehicles to misfuel
while the catalyst was still present. It is assumed above,
however, that essentially all vehicle owners, who remove
their catalysts and also misfuel, will misfuel even if
prevented from removing the catalyst by the program. This
assumption is supported by the fact that in the latest
tampering survey, 62% of the habitually misfueled vehicles
had not removed the catalyst, indicating that most misfuelers
believe it is safe to misfuel even if the catalyst is left on
the vehicle. Given the perceived incentives for misfueling,
owners who are forced to replace catalysts will probably come
to believe the same, or will find a way to defeat the
catalyst check entirely.
5.1.1.3 Air Pump-Only Inspections
With air pumps, removal or failure of the drive belt is the
most likely disablement. Air pump belts may eventually break
if they are not periodically replaced. This may account for
some portion of observed disablements. Since this
disablement is relatively easy to identify and replacement is
inexpensive, some deliberate tampering with the air pump may
occur even with a vigorous anti-tampering program. Some
vehicle owners may be willing to risk detection and the
subsequent penalty, replacement of the belt, in order to
-------
-73-
achieve perceived benefits in fuel economy and performance.
Some vehicle owners may even replace and remove their air
pump belt before and after their periodic inspection to avoid
detection by the program. This would be expected only on the
part of die-hard tamperers, however, since such a repair may
take a considerable amount of time. To address this problem,
the public awareness effort should attempt to eliminate the
misperceptions associated with tampering.
In Portland the rate of air pump disablement is about 74%
less than in the other I/M sites. However, since the survey
was performed at an I/M station where a tampering check is
performed, some vehicle owners may have reconnected the air
pump for the inspection with the intention of disabling it
immediately after meeting the legal requirements. It may be
speculated that the number of vehicle owners who do this is
only a small portion of the fleet. However, we will assume
that an annual inspection program will have an 80%
effectiveness, and a biennial program will have a 70%
effectiveness. This applies to both existing and subsequent
tampering. The calculation of benefits for most programs
will not be very sensitive to the choice of effectiveness
values for air pump inspections, as the air pump-only
category is small and has a relatively small emissions excess.
The above effectiveness values apply only to the group of
vehicles which have air pump disablement as the only form of
tampering. In other words, 70% or 80% of these vehicles are
restored to a tamper-free state with reduced emissions and
the remainder stay as they were (or as they would have
otherwise become). Similarly, 70% or 80% of the vehicles
with disabled air pumps which were also misfueled revert to a
misfueled-only status, but under the assumptions given in
Section 4.0 their emissions are not reduced. The same holds
for cars with disabled air pumps and missing catalysts, with
or without misfueling. The table of effects is omitted here,
since it is very simple and not required for understanding of
the effectiveness assumptions.
5.1.1.4 Fuel Inlet Restrictor-Only and Inlet/Air Pump
Inspections
Two types of inspection requirements for the fuel inlet
restrictor are discussed here and defined as "fuel inlet
restrictor-only" inspections. One would require repair only
of the restrictor when it was found to be tampered. The
other would require replacement of the catalyst and repair of
the restrictor but would not inspect for catalyst presence.
This document does not contain benefits for either of these
approaches, and EPA does not recommend either approach for
new anti-tampering programs.
-------
-74-
Th e first approach, repair of the restrictor only, would have
no benefit for the vehicles, since poisoned catalysts would
remain and cannot be expected to recover their function. it
would also have very limited deterrence for subsequent
misfueling, since the penalty of a restrictor repair is
relatively low, since no catalyst replacements would take
place, and since other means to misfuel could be used.
The second approach, repair of both catalyst and restrictor,
would in principle have benefits. However, the inequity of
not also requiring new catalysts on vehicles with missing
catalysts (regardless of inlet status) would in EPA's opinion
be contrary to the objectives of an anti-tampering program.
The above discussion also applies to an inlet inspection
combined with an air pump inspection, so no benefits for it
are given in this document.
Inlet restrictor inspections are most useful in combination
with catalyst presence inspections, which are discussed in
subsection 5.1.1.7 below.
5.1.1.5 Plumbtesmo-Only inspections
Just as with inlet-only inspection, it is in concept possible
to perform only a Plumbtesmo inspection, with the penalty
being a mandatory catalyst replacement. However, it is
inequitable to force catalyst replacement for a vehicle which
fails Plumbtesmo without also forcing it for a vehicle with a
tampered fuel inlet. In addition, equity requires that
vehicles with missing catalysts also be repaired in any
program that requires misfueled vehicles to have the
catatlyst replaced.
Consequently, this report does not contain benefits for
programs that perform Plumbtesmo tests without also doing
inlet checks, or for programs that do inspect for both of
these but not for catalyst presence as well. The same holds
for Plumbtesmo/air pump, Plumbtesmo/inlet/air pump,
Plumbtesmo/catalyst, and Plumbtesmo/air pump/catalyst
inspections.
5.1.1.6 Catalyst and Air Pump inspections
An inspection for catalyst presence and air pump is a
feasible approach. The benefits are based on assumptions
consistent with those stated above in Sections 5.1.1.2 and
5.1.1.3 for catalysts and air pump separately. Some minor
-------
-75-
additional assumptions for overlap categories affected by
both inspections are necessary, however. Specifically, the
catalyst/air pump inspection has the following effects on the
different overlap categories:
Overlap Category
1) Air Pump/Catalyst
2,3) Air Pump/Misfueling
4,5) Air Pump/Catalyst/
Misfueling
6,7) Catalyst/Misfueling
8) Air Pump Only
9) Catalyst Only
10,11) Misfueling Only
Effect of Catalyst/Air Pump Check
80% (70% if biennial) become OK.
15% (25% if biennial) become air
pump only.
5% remain air pump/catalyst.
80% (70% if biennial) become
misfueling only.
20% (30% if biennial) remain air
pump/mi sfueling.
80% (70% if biennial) become
misfueling only.
15% (25% if biennial) become air
pump/misfueling.-
5% remain air pump/catalyst/
misfueling.
95% become misfueling only.
5% remain catalyst/misfueling.
80% (70% if biennial) become OK.
20% (30% if biennial) remain air
pump only.
95% become OK.
5% remain catalyst only.
100% remain misfueling only.
The above description applies for both tampering which is
present in the fleet when the catalyst/air pump check is
initiated and tampering which would otherwise have occurred
subsequently.
5.1.1.7 Catalyst and Fuel Inlet Restrictor Inspections
Inspection of the fuel inlet restrictor is straightforward,
and there is no reason to expect errors if inspectors are
diligent. However, the real issue is the effectiveness of
-------
-76-
the inlet inspection in causing poisoned catalysts to be
replaced and not re-misfueled, and in deterring misfueling
which would otherwise occur after the start of the
catalyst/inlet check. Clearly, an inlet check will not
reduce misfueling that occurs by means other than inlet
tampering.
EPA believes that an inlet check alone will be far from
completely effective for misfueling accomplished by inlet
tampering, for reasons explained below. However the Agency
is willing to evaluate innovative inspection programs which
attempt to enhance the effectiveness of simple inlet checks.
EPA Regional Offices should be contacted to discuss
alternative approaches.
It is assumed that any fuel inlet restrictor which allows
entry of a legal size leaded fuel nozzle is an indication of
habitual misfueling and, therefore, that the catalyst has
been rendered inoperative. Therefore, if the fuel inlet
restrictor has been enlarged, the vehicle owner must be
required to replace the catalyst. In addition, the vehicle
owner will have to repair or replace the restrictor so that a
leaded fuel nozzle will not fit.
Misfuelers are divided into four categories for anaylsis: 1)
existing misfuelers with tampered inlets only, 2) existing
misfuelers with both tampered inlet and missing catalyst, 3)
existing misfuelers without tampered inlet, 4) future
mi sfuelers.
The category one misfuelers are a problem group. It is
unclear how many of the misfuelers will pre-fix their inlets
or will continue to misfuel even after detection and catalyst
replacement. Because of this uncertainty an assumption is
made that many will continue to misfuel or will pre-fix.
An owner may defeat the inspection by simply repairing the
inlet prior to the first inspection, if he or she is aware of
the start of the catalyst/inlet check and of the consequences
of failure. An owner who has submitted his or her vehicle
for inspection, and who has failed the inlet check and then
learned of catalyst replacement consequences can still defeat
the program by repairing the inlet and seeking another
inspection as though the vehicle had not already been
inspected, unless additional safeguards are imposed. This
can be prevented, for example, by punching the vehicle
registration card at the first inspection, or a similar
measure, but this would not solve the problem of misfuelers
who have advance warning of the consequences of inlet
failure. Even if an owner replaces the catalyst after
failing the inlet check, there is the possibility of
re-misfueling using means other than inlet tampering.
-------
-77-
Category 2 misfuelers will be much easier to catch and
deter. They will be caught because if they pre-fix their
inlet they will still faii the catalyst check. Also, it is
unlikely that they will continue to fuel switch after
detection because besides the large cost of catalyst and
inlet replacement which they will incur, most if not all of
this group believe that they must remove their catalyst in
order to fuel switch.
It is assumed that category three misfuelers are not deterred
at all.
With respect to misfueling which would have occurred
subsequent to the start of the catalyst/inlet check,
(category 4 misfueling) , EPA believes the effectiveness of
the inlet check will be limited only by the opportunity for
potential misfuelers to be able to misfuel by other means and
by whether they perceive the benefits of misfueling as
outweighing the potential of being caught by the inspection
program. Public awareness programs should be targeted
directly at this group. This fact plus the existence of an
inspection check which will prevent them from tampering with
their inlet or removing their catalyst, should result in
significant levels of deterrence of future misfueling.
There is an assumption that the mere existence of an idle
test will increase the effectiveness of an inlet check,
because vehicle owners will be concerned that misfueling will
cause failure of the idle test.
In light of the above considerations, EPA has made the
following assumptions regarding the effect of a
catalyst/inlet check in areas with an I/M program.
Overlap Category
1) Air Pump/Catalyst
Effect of Catalyst/Inlet Check
95% become air pump only.
5% remain air pump/catalyst.
2) Air Pump/Misfueling
(Inlet)
33% of vehicles initially in this
category become air pump only.
67% of vehicles initially in this
category remain air pump/
misfueling.
70% of vehicles which would otherwise
enter this category subsequent
to the start of the program
become air pump only instead.
-------
-78-
3) Air Pump/Misfueling
(Other)
4) Air pump/Catalyst/
Misfueling (Inlet)
5) Air Pump/Catalyst/
Misfueling (Other)
6) Catalyst/Misfueling
(Inlet)
7) Catalyst/Misfueling
(Other)
8) Air Pump Only
9) Catalyst Only
10) Misfueling (Inlet)
Only
30% of vehicles which would
otherwise enter this category
subsequent to the start of the
program remain air pump/
misfueling.
100% remain air pump/misfueling.
59% of "initial" vehicles become air
pump only.
36% of "initial" vehicles become air
pump/misfueling.
5% of "initial" vehicles remain air
pump/catalyst/misfueling.
70% of "subsequent" vehicles become
air pump only.
25% of "subsequent" vehicles become
air pump/misfueling only.
5% of "subsequent" vehicles remain
air pump/catalyst/misfueling.
95% become air pump/misfueling.
5% remain air pump/catalyst/
misfueling.
36% of "initial" vehicles become
misfueling only.
59% of "initial" vehicles become OK.
5% of "initial" vehicles remain
catalyst/misfueling.
25% of "subsequent" vehicles become
misfueling only.
70% of "subsequent" vehicles become
OK.
5% of "subsequent" vehicles remain
catalyst/misfueling.
95% become misfueling only.
5% remain catalyst/misfueling.
100% remain air pump only.
95% become OK.
5% remain catalyst only.
33% of "initial" vehicles become OK.
67% of "initial" vehicles remain
misfueling only.
70% of "subsequent" vehicles become
OK.
-------
-79-
30% of "subsequent" vehicles remain
misfueling only.
11) Misfueling (Other) 100% remain misfueling only.
Only
5.1.1.8 Catalyst, Fuel Inlet Restrictor, and Air Pump
Inspections
Adding an air pump inspection to a catalyst/inlet check has
the effect of correcting most air pump disablements, without
altering the effectiveness for catalyst removal and
misfueling. The following overlap categories are affected
differently than stated in Section 5.1.1.7 for a
catalyst/inlet check. Those not listed are affected the same
as described in Section 5.1.1.7.
Overlap Category
1) Air Pump/Catalyst
Effect of Catalyst/Inlet/Air Pump Check
80% (70% if biennial) become OK.
15% (25% if biennial) become air pump
only.
5% remain air pump/catalyst.
2) Air Pump/Misfueling
(Inlet)
33%
47%
20%
70%
10%
20%
of "initial" vehicles become OK.
(37% if biennial) of "initial"
vehicles become misfueling only.
(30% if biennial) of "initial"
vehicles remain air pump/
misfueling.
of "subsequent" vehicles become
OK.
(0% if biennial) of "subsequent"
vehicles become misfueling only.
(30% if biennial) of "subsequent"
vehicles remain air pump/
misfueling.
3) Air Pump/Misfueling
(Other)
4) Air Pump/Catalyst/
Misfueling (Inlet)
80%
20%
(70% if biennial)
misfueling only.
(30% of biennial)
pump/misfueling.
become
remain air
59% of "initial" vehicles become OK,
21% (11% if biennial) of "initial"
vehicles become misfueling only,
15% (25% if biennial) of "initial"
vehicles become air pump/
misfueling only.
-------
-80-
5% of "initial" vehicles remain air
pump/catalyst/misfueling.
70% of "subsequent" vehicles become
OK.
10% (0% if biennial) of "subsequent"
vehicles become misfueling only.
15% (25% if biennial) of "subsequent"
vehicles become air pump/
misfueling.
5% of "subsequent" vehicles remain
air pump/catalyst/misfueling.
5) Air Pump/Catalyst/
Misfueling (Other)
80% (70% if biennial) become
misfueling only.
15% (25% if biennial) become air pump/
misfueling.
5% remain air pump/catalyst/
misfueling.
8) Air Pump Only
80% (70% if biennial) become
20% (30% if biennial) remain
only.
OK.
air pump
5.1.1.9 Catalyst, Fuel Inlet Restrictor, and Plumbtesmo
Inspections
As was pointed out in Section 2.0 EPA has been using a
lead-sensitive chemically coated paper, whose trade name is
Plumbtesmo, to detect tell-tale lead deposits in the
tailpipes of vehicles in the latest tampering surveys.[2]
This test is a powerful tool in detecting previous use of
leaded fuel when there is no leaded fuel in the tank or
damage to the fuel inlet restrictor. Its primary fault lies
in its inability to determine the extent of catalyst damage
due to misfueling. EPA does not know how long lead deposits
from misfueling remain in the tailpipe, therefore the
possibility cannot now be ruled out that a single tankful of
leaded fuel used during an emergency or bought from an
unscrupulous gasoline dealer as unleaded may cause a
Plumbtesmo test failure months later, even though unleaded
fuel has been used at all other fuelings.
If some simple, reliable test to determine the extent of
damage to the catalyst by lead deposits could be developed,
then such a test could be used to allow vehicle owners whose
vehicles fail the Plumbtesmo test to prove that their
catalysts were still active and did not need to be replaced.
Without such a test, persons who do not deliberately misfuel
but accidentally buy leaded gas might be caught by the
Plumbtesmo test. Although EPA is currently assessing the
-------
-81-
feasibility of such a catalyst diagnostic test, no simple
test is as yet available. Inequities will be reduced by an
aggressive program of sampling fuel from retail gas stations.
Since the required catalyst replacement cost associated with
a Plumbtesmo failure would be expensive, some vehicle owner
dissatisfaction with the Plumbtesmo test will result unless a
back-up catalyst efficiency test is available as a second
opinion. Thus EPA recommends the use of a back-up test when
requested by a vehicle owner whose vehicle's fuel inlet
restrictor has not been defeated and who contends that leaded
fuel has not been used.
A less serious, but equally complicating factor is the fact
that in EPA tests some vehicles which have obviously been
misfueled pass the Plumbtesmo test. As yet no full
explanation has been determined for those cases.* As a
result, some grossly misfueled vehicles may escape detection
by a Plumbtesmo test.
The main attraction of the Plumbtesmo test is its potential
effectiveness in deterring misfueling. With a Plumbtesmo
inspection, vehicle owners could never be sure that they
could avoid detection if they misfuel, even though the EPA
experience is that only about one-half of habitually
misfueled vehicles will fail the test. Besides this high
false negative rate, there is also the possibility that
vehicle owners will attempt to circumvent the Plumbtesmo test
by cleaning or replacing their tailpipes prior to
inspection. However, these acts would make misfueling much
less attractive.
A program which would require replacement of the catalyst
whenever a vehicle fails the Plumbtesmo test or has a
tampered inlet is assumed to prevent 85% of misfueling which
would otherwise have occurred. Since the tailpipe would be
contaminated with lead, replacement or cleaning of ' the
tailpipe as well as replacement of the catalyst would be
required to avoid a Plumbtesmo test failure at the next
inspection.
The 85% value pertains only to deterrence of continued or new
misfueling, and is based on the psychological effect of the
Plumbtesmo test on misfueling bahavior. For vehicles which
*One possible explanation is that the unstable lead-detecting
compounds in the test paper became inadvertently deactivated,
or a defective lot was used during testing. An inspection
program forewarned of these problems could easily avoid using
inactive test paper.
-------
-82-
were misfueled' before the inspection program started, the
actual effectiveness will be less. Even if previous
misfuelers stop misfueling in an attempt to avoid failure of
the piumbtesmo test, there will be emission reductions only
in the cases in which the previous misfueling is detected and
the catalyst is replaced. In the EPA survey, about 80% of
habitually misfueled vehicles failed the combination inlet
and Piumbtesmo check. In a real program, owners of misfueled
vehicles which fail only the inlet check (one-fourth of the
80%) might successfully defeat the inlet check and avoid
catalyst replacement, as described in Section 5.1.1.7 above.
However, some vehicles will fail Piumbtesmo which in the EPA
survey would not have counted as habitually misfueled but
which likely have suffered some catalyst damage. EPA assumes
that these two influences approximately balance each other,
and has assumed a 75% effectiveness for previous misfueling,
regardless of the means used to accomplish it. For
subsequent misfueling, the 85% value is used directly.
As noted earlier, because a Piumbtesmo test may fail a
vehicle whose only use of leaded fuel was inadvertent due to
contamination or mislabeling at the pump, it is important
that these occurrences be minimized. This can be done
establishing the program of fuel pump inspections described
at the beginning of Section 5.0.
In detail, the effect of a catalyst/inlet/Plumbtesmo
inspection is assumed to be as follows:
Overlap Category
1) Air Pump/Catalyst
2,3) Air Pump/Misfueling
Effect of Catalyst/
Inlet/Plumbtesmo Check
5) Air Pump/Catalyst/
Misfueling
95% become air pump only.
5% remain air pump/catalyst.
75% of "initial" vehicles become air
pump only.
25% of "initial" vehicles remain air
pump/mi sfueli ng.
85% of "subsequent" vehicles become
air pump only.
15% of "subsequent" vehicles remain
air pump/misfueling.
75% of "initial"vehicles become air
pump only.
20% of "initial" vehicles become air
pump/mi sfueli ng.
5% of "initial" vehicles remain air
pump/catalyst/mi sfueli ng.
-------
-83-
85% of "subsequent" vehicles become air
pump only.
10% of "subsequent" vehicles become air
pump/misfueling.
5% of "subsequent" vehicles remain air
pump/catalyst/misfueling.
6,7) Catalyst/Misfueling 75% of "initial" vehicles become OK.
20% of "initial" vehicles become
misfueling only.
5% of "initial" vehicles remain
catalyst/misfueling.
85% of "subsequent" vehicles become OK.
10% of "subsequent" vehicles become
misfueling only.
5% of "subsequent" vehicles remain
catalyst/misfueling.
8) Air pump Only 100% remain air pump only.
9) Catalyst Only 95% become OK.
5% remain catalyst only.
10,11) Misfueling Only 75% of "initial" vehicles become OK.
25% of "initial" vehicles remain
misfueling only.
85% of "subsequent" vehicles become OK.
15% of "subsequent" vehicles remain
misfueled only.
5.1.1.10 Catalyst, Fuel Inlet Restrictor, Plumbtesmo, and
Air Pump Inspections
Adding an air pump inspection to the previous combination of
catalyst/inlet/Plumbtesmo results in correction or deterrence
of most air pump disablements. Correction of misfueling and
catalyst removal is not changed. In detail, the effect of a
catalyst/inlet/Plumbtesmo/air pump inspection is assumed to be
as follows:
Overlap Cataqory
1) Air Pump/Catalyst
Effect of Catalyst/Inlet/
Plumbtesmo/Air Pump Check
80% (70% if biennial) become OK.
15% (25% if biennial) become air
pump only.
5% remain air pump/catalyst.
2,3) Air Pump/Misfueling
75% (70% if biennial) of "initial"
vehicles become OK.
-------
-84-
5% (0% if biennial) of "initial"
vehicles become misfueling only.
0% (5% if biennial) of "initial"
vehicles become air pump only.
20% (25% if biennial) of "initial"
vehicles remain air pump/
misfueling.
80% (70% if biennial) of "subsequent"
vehicles become OK.
5% (15% if biennial) of "subsequent"
vehicles become air pump only.
15% of "subsequent" vehicles remain
air pump/misfueling.
4,5) Air Pump/Catalyst 75%
Misfueling
5%
0%
15%
5%
80%
5%
10%
5%
6,7) Catalyst/Misfueling 75%
20%
5%
85%
10%
5%
8) Air Pump Only 80%
20%
(70% if biennial) of "initial"
vehicles become OK.
(0% if biennial) of "initial"
vehicles become misfueling only.
(5% if biennial) of "initial"
vehicles become air pump only.
(20% if biennial) of "initial"
vehicles become air pump/
misfueling only.
of "initial" vehicles remain air
pump/catalyst/misfueling.
(70% if biennial) of "subsequent"
vehicles become OK.
(15% if biennial) of "subsequent"
vehicles become air pump only.
(10% if biennial) of "subsequent"
vehicles become air pump/
misfueling only.
of "subsequent" vehicles remain
air pump/catalyst/misfueling.
of "initial" vehicles become OK.
of "initial" vehicles become
misfueling only,
of "initial" vehicles remain
catalyst/misfueling.
of "subsequent" vehicles become
OK.
of "subsequent" vehicles become
misfueling only,
of "subsequent" vehicles remain
remain catalyst/misfueling.
(70% if biennial) become OK.
(30% if biennial) remain air
pump only.
-------
-85-
9) Catalyst Only 95% become OK.
5% remain catalyst only.
10,11) Misfueling Only 75% of "initial" vehicles become OK,
25% of "initial" vehicles remain
misfueling only.
85% of "subsequent" vehicles become
OK.
15% of "subsequent" vehicles remain
misfueled only.
5.1.2 Results: Benefits for I/M Programs
Table 27 presents the benefits of inclusion of a tampering
inspection with an annual I/M program. There are separate
results for pre-1980, 1980 through 1983, and 1984 and later
vehicles, so that the benefits from programs which exempt
pre-1980 or pre-1984 vehicles can be estimated. Table 28
presents a biennial version for each of the benefits in Table
27. Tables 27 and 28 were prepared using assumptions
in Section 5.1.1.
5.1.3 Program Costs
This subsection states assumptions necessary to calculate the
cost of a tampering inspection program when added to an
existing I/M program.
Repairs - The obvious cost of anti-tampering and
anti-misfueling programs is the cost to vehicle owners for
repairs of disablements, whether they were deliberate or
inadvertent. In terms of all cars being inspected, the per
vehicle cost for repairs will be relatively small, since
usually only some small fraction of vehicles will require
repairs. Also, if the program continues to operate beyond
December 31, 1987, the cost-effectiveness of the repairs will
improve until essentially the only cost incurred by the
program will be the cost of inspection. Section 3.0
discusses the repair costs which we have assumed for this
analysis.
Using the rate of tampering at the start of the program, the
number of vehicles which require repairs at the start of the
program can be estimated. By assuming an average repair
cost, the initial year repair cost can be estimated.
After the program begins,, some tampering will continue to
occur and subsequently be detected and repaired. The number
of vehicles tampered after the program begins will depend on
the effectiveness of the program in deterring tampering and
the rate at which new vehicles move into the area. The
-------
-86-
effeetiveness will depend on which emission control
components are inspected. For catalyst and fuel inlet
restrictor tampering, it is assumed that only those vehicles
identified in the first year of the program will require
repairs. Vehicles not identified are assumed to continue to
avoid detection in subsequent years. Also, no significant
amount of new tampering is expected to be discovered in
subsequent years since vehicle owners will be aware of the
program and its penalties. Air pump, PCV, and evaporative
canister disablements occur at moderate rates even in an
inspection program which checks for such disablements. In
these cases all disablements are assumed to be repaired in
the first year, and in each subsequent year repairs will be
done on all disablements which reappear.
Inspections - In addition to the cost vehicle owners must pay
in repairs, a tampering inspection program will result in
increased inspection costs and extra administrative costs
related to the tampering inspection. A rough estimate of the
additional costs can be made by estimating the increase in
personnel time, both inspector and administrative, necessary
to include the tampering check.
In decentralized programs, only the additional time an
inspector will need to perform the tampering check should be
attributed to the anti-tampering program. As with
centralized programs, administrative costs can probably be
estimated by the need to hire additional personnel.
It is expected that most of the duties required by the
addition of a tampering inspection can be integrated into the
operation of the I/M program without any substantial increase
in program costs. Although this cost will likely vary
substantially from program to program depending on many
factors, EPA assumes an overall increase in program
administrative and inspection costs to be 34 cents in
centralized and $1.00 in decentralized inspection programs
per inspection, as an example. This added cost would include
not only additional costs to perform the inspections but also
include additional administrative duties to oversee the
additional program elements.
The cost has been estimated by assuming that a single
inspector in a centralized program could complete the
necessary inspection and additional paperwork for a check of
all the components in about one minute. If the inspector is
a mechanic costing $20 per hour including fringe benefits and
overhead, this works out to be about 34 cents per
inspection. In a decentralized program, the inspector will
be less specialized and will likely take longer to
satisfactorily complete the inspection, we have assumed the
-------
-87-
decentralized program inspector will take three minutes to
complete the inspection, which at $20 per hour, will be $1.00
per inspection. These estimates are for an inspection of all
items discussed in this report. An inspection of fewer items
would be shorter and, therefore, cheaper.
If a Plumbtesmo inspection is performed, the cost of the test
paper may be significant when calculating overall program
costs. EPA has purchased Plumbtesmo paper at a price of 43
cents for a piece of paper which is large enough to cut in
half for testing two vehicles. A per vehicle cost of 7 cents
is possible in inspection proqrams because large quantities
of test paper can be purchased at one time.
5.1.4 Cost-Effectiveness
Because local variations in program design and costs affect
the cost-effectiveness calculation, this report does not
attempt to calculate cost-effectiveness for all of the
numerous possible program types.
5.2 Non-I/M Periodic Inspection Programs
Non-I/M periodic inspection programs offer another
opportunity to address the tampering issue. A tampering
program can be added to a periodic safety inspection, or an
entirely new inspection requirement can be established.
Costs will obviously be higher in the latter approach.
Section 2.0 discussed tampering rates in non-l/M areas. The
rates discussed in that section are the rates used for all
calculations in this section, except that overlap is
accounted for among tampering types. The individual vehicle
benefits and costs of repairs of tampering and misfuelinq are
those discussed in Section 3.0. The methodology explained in
Section 4.0 was used to calculate excess emissions due to
tampering and misfueling. Only annual and biennial
inspection programs are considered in this section.
5.2.1 Program Effectiveness
For periodic inspection programs as in I/M programs, it is
assumed that the program will require repair or replacement
of the disabled emission control components once they are
discovered, followed by reinspection of the vehicle and/or
the repair receipts to verify compliance. In addition, to
claim the benefits estimated in this section, the inspection
program would have the same requirements as anti-tampering
and anti-misfueling programs in I/M programs described at the
beginning of Section 5.0. All of the effectiveness
assumptions used for I/M programs will be assumed to apply to
-------
-88-
periodic inspections which are not part of I/M programs with
one exception which is explained in more detail in the
following paragraph. The reader should refer to Section
5.1.1 for the discussion of inspection effectiveness for all
other components and inspection types.
The exception that distinguishes tampering inspection
programs in I/M areas and non-I/M areas concerns the
effectiveness of fuel inlet restrictor inspections without
accompanying Plumbtesmo inspections. As stated in 5.1.1.7,
EPA believes that in non-I/M areas owners are more likely to
feel that defeating the inlet check is an easy, safe, and
reliable way to permanently avoid the requirement for a
catalyst replacement. Therefore, EPA assumes a lower
effectiveness for inlet inspections in non-I/M areas. The
specific assumptions are as follows:
Overlap Category
Effect of Catalyst/Inlet Checks
1) Air Pump/Catalyst
2) Air Pump/Misfueling
(Inlet)
3) Air Pump/Misfueling
(Other)
4) Air Pump/Catalyst/
Misfueling (Inlet)
95% become air pump only.
5% remain air pump/catalyst.
17% of "initial" vehicles become air
pump only.
83% of "initial" vehicles remain
air pump/misfueling.
30% of "subsequent" vehicles become air
pump only.
70% of "subsequent" vehicles remain
air pump/misfueling.
100% remain air pump/misfuelinq.
33% of "initial" vehicles become air
pump only.
62% of "initial" vehicles become
air pump/misfueling only.
5% of "initial" vehicles remain
air pump/catalyst/misfueling.
30% of "subsequent" vehicles become
air pump only.
65% of "subsequent" vehicles become
air pump/misfueling only.
5% of "subsequent" vehicles remain
air pump/catalyst/misfueling.
-------
-89-
5) Air Pump/Catalyst/
Misfueling (Other)
6) Catalyst/Misfueling
(Inlet)
7) Catalyst/Misfueling
(Other)
8) Air Pump Only
9) Catalyst Only
10) Misfueling Only
(Inlet)
11) Misfueling Only
(Other)
95% become air pump/misfueling only.
5% remain air pump/catalyst/misfueling.
33% of "initial" vehicles become OK.
62% of "initial" vehicles become
misfueling only.
5% of "initial" vehicles remain
catalyst/misfueling.
30% of "subsequent" vehicles become OK.
65% of "subsequent" vehicles become
misfueling only.
5% of "subsequent" vehicles remain
catalyst/misfueling.
95% become misfueling only.
5% remain catalyst/misfueling.
100% remain air pump only.
95% become OK.
5% remain catalyst only.
17% of "initial" vehicles become OK.
83% of "initial" vehicles remain
misfueling only.
30% of "subsequent" vehicles become OK.
70% of "subsequent" vehicles remain
misfueling only.
100% remain misfueling only.
Overlap Category
1) Air Pump/Catalyst
Effect of Catalyst/Inlet/Air Pump Checks
80% (70% if biennial) become
15% (25% if biennial) become
only.
5% remain air pump/catalyst.
OK.
air
OUTTtO
2) Air Pump/Misfueling
(Inlet)
17% of "initial" vehicles become OK.
63% (53% if biennial) of "initial"
vehicles become misfueling only.
20% (30% if biennial) of "initial"
vehicles remain air pump/misfueling.
30% of "subsequent" vehicles become OK.
50% (40% if biennial) of "subsequent"
vehicles become misfueling only.
20% (30% if biennial) of "subsequent"
vehicles remain air pump/mis fuelinq,
-------
-90-
3) Air Pump/Misfueling
(Other)
4) Air Pump/Catalyst/
Misfueling (Inlet)
5) Air Pump/Catalyst/
Misfueling (Other)
6) Catalyst/Misfueling
(Inlet)
7) Catalyst/Misfueling
(Other)
8) Air Pump Only
9) Catalyst Only
10) Misfueling Only
(Inlet)
80% (70% if biennial) become misfueling
only.
20% (30% if biennial) remain air pump/
misfueling.
33%
47%
15%
5%
30%
50%
15%
5%
of "initial" vehicles become OK.
(37% if biennial) of "initial"
vehicles become misfueling only.
(25% if biennial) of "initial"
vehicles become air pump/misfueling
only.
of "initial" vehicles remain
air pump/catalyst/misfueling.
of "subsequent" vehicles become OK.
(40% if biennial) of "subsequent
vehicles become misfueling only.
(25% if biennial) of "subsequent"
vehicles become air pump/misfueling
only.
of "subsequent vehicles remain air
pump/catalyst/misfueling.
80% (70% if biennial) become misfueling
only.
15% (25% if biennial) become air pump/
misfueling only.
5% remain air pump/catalyst/misfueling
33% of "initial" vehicles become
62% of "initial" vehicles become
misfueling only.
5% of "initial" vehicles remain
catalyst/misfueling.
30% of "subsequent" vehicles
65% of "subsequent" vehicles
misfueling only.
5% of "subsequent" vehicles
catalyst/misfueling.
OK
become
become
rema in
OK.
95% become misfueling only.
5% remain catalyst/misfueling.
80% (70%
20% (30%
only
if biennial)
if biennial)
become
remain
OK.
air
pump
95% become OK.
5% remain catalyst only.
17% of "initial" vehicles
83% of "initial" vehicles
misfueling only.
become
remain
OK
-------
-91-
30% of "subsequent" vehicles become OK.
70% of "subsequent" vehicles remain
misfueling only.
11) Misfueling Only 100% remain misfueling only.
(Other)
As pointed out in Section 2.0f areas without I/M programs
tend to have higher tampering and misfueling rates than I/M
areas. In this section, all benefits are calculated using
tampering and misfueling rates predicted for non-I/M areas.
5.2.2 Results: Benefits for Non-I/M Periodic Inspection
Programs
Table 29 presents the benefits of an annual tampering
inspection program. There are separate results for pre-1980,
1980 through 1983, and 1984 and later vehicles so that the
benefits from programs which exempt pre-1980 or pre-1984
vehicles can be estimated. Table 30 presents a biennial
version for each of the benefits in Table 29.
5.2.3 Program Costs
This subsection states assumptions necessary to calculate the
cost of a tampering inspection program when added to an
existing safety inspection program or when initiated
independently.
Repairs - The obvious cost of anti-tampering and
anti-misfueling programs is the cost to vehicle owners for
repairs of disablements, whether they were deliberate or
inadvertent. In terms of all cars being inspected, the per
vehicle cost for repairs will be relatively small, since
usually only some small fraction of vehicles will require
repairs. Also, if the program continues to operate beyond
December 31, 1987, the cost-effectiveness of the repairs will
improve until essentially the only costs incurred by the
program will be the cost of inspection. Section 3.0
discusses the repair costs which we have assumed for this
analysis.
Using the rate of tampering at the start of the program, the
number of vehicles which require repairs at the start of the
program can be estimated. By assuming an average repair
cost, the initial year repair cost can be estimated.
After the program begins, some tampering will continue to
occur and subsequently be detected and repaired. The number
of vehicles tampered after the program begins will depend on
-------
-92-
the effectiveness of the program in deterring tampering and
the rate at which new vehicles enter the program area. The
effectiveness will depend on the emission control component.
For catalyst and fuel inlet restrictor tampering, it is
assumed that only those vehicles identified in the first year
of the program will require repairs. Vehicles not identified
are assumed to continue to avoid detection in subsequent
years. Also, no significant amount of new tampering is
expected to be discovered in subsequent years since vehicle
owners will be aware of the program and its penalties. Air
pump, PCV, and evaporative canister disablements occur at
moderate rates even in an inspection program which checks for
such disablements. in these cases all disablements are
assumed to be repaired in the first year and in each
subsequent year repairs will be done on all disablements
which reappear.
Tampering Inspections Added to Safety - In addition to the
cost vehicle owners must pay in repairs, a safety inspection
program which adds a tampering check will incur additional
expenses from the added tampering inspections at individual
inspection stations and additional administrative costs
related to adding the tampering inspection to the I/M
requirements. A rough estimate of the additional costs can
be made by estimating the increase in personnel time, both
inspector and administrative, necessary to include the
tampering check.
In centralized inspection programs the tampering inspection
might be added to the inspection procedure without any needed
increase in personnel. This would be the case if personnel
and operating hours did not require expansion; better
scheduling of inspections or simply tolerating longer waitinq
lines could be used to allow tampering inspections with the
existing facility and personnel time constraints. It is more
likely that additional inspectors, administrative personnel,
or possibly inspection stations would be required. In such
cases the added salaries of the additional personnel and
other costs would be attributed to the tampering inspection.
In decentralized programs, only the additional time an
inspector will need to perform the tampering check should be
attributed to the anti-tampering program. As with
centralized programs, administrative costs can probably be
estimated by the need to hire additional personnel.
It is expected that most of the duties required by the
addition of a tampering inspection can be integrated into the
operation of the safety program without any substantial
increase in program costs. Although this cost will likely
vary substantially from program to program depending on many
factors, EPA assumes an overall increase in program
-------
-93-
administrative and inspection costs to be 34 cents for
centralized and $1.00 for decentralized programs per
inspection, as an example. This added cost would include not
only additional costs to perform the inspections, but also
include additional administrative duties to oversee the
additional program elements. Section 5.1.3 discusses how
these costs were estimated.
If a Plumbtesmo test is conducted, the cost of the test paper
must be considered. A per vehicle cost of 7 cents should be
possible in inspection programs where large quantities of
test paper can be purhcased at one time.
Tampering Inspections Without Safety - In this case, the
tampering check is responsible for the full cost of the
inspection program, including the cost of facilities and
personnel that in existing safety programs can be attributed
to the safety element. Costs in such a program would
probably range from $5 to $10.
5.3 Other Anti-Tampering and Anti-Misfueling Programs
The anti-tampering and anti-misfueling programs in this
subsection do not involve periodic inspection of vehicles
and, therefore, must rely more heavily on the possibility of
detection to deter misfueling and tampering. Correction of
tampering already present at the start of the program will be
less complete than in a periodic inspection program, since
only a fraction of the fleet is ever directly affected by the
enforcement actions. (Owners of already tampered vehicles
will wait until caught before repairing their vehicles since
it is assumed that there is no fine in addition to repairs.)
As a result, the uncertainty inherent in the benefits from
these programs is larger than in programs where every vehicle
is inspected periodically.
Although there are numerous ways in which tampering and
misfueling might be reduced without periodic inspection, this
report will focus only on a few approaches which seem to
provide the best probability of large emission benefits and
low uncertainty. Other approaches not considered in this
report may also provide similar benefits. If an area wishes
to investigate programs other than those analyzed in this
report, the EPA Regional Office should be contacted for an
evaluation of the potential of the specific approach proposed.
To claim all of the benefits estimated in the tables in this
section, the anti-tampering and anti-misfueling program must
meet all of the requirements outlined at the beginning of
-------
-94-
Section 5.0. These include such design features as periodic
audits and inspector training.
5.3.1 Change-of-Ownership Inspection Programs
A change-of-ownership anti-tampering inspection program would
require an inspection of the vehicle to assure proper
connection of the emission control devices every time the
vehicle changed ownership or moved into the area for the
first time. Title and registration in the new owner's name
would be withheld until the vehicle was in compliance.
Although nearly all vehicles change hands at least once in
their lives, the time between sales can vary and will often
be many years. This time period would* allow vehicle owners
an opportunity to operate tampered vehicles for long periods
of time before any penalty. Some vehicle owners could avoid
any penalty by selling the vehicle outside the area covered
by the program or simply retaining or junking the car. Also,
within-family transfers are often exempt since any
requirements could be easily circumvented by simply leaving
the title in the original owner's name. States may also be
reluctant to intrude into family transactions. These
problems will cause the effectiveness of such programs to be
less than for periodic inspection programs.
Vehicle owners of cars with the catalyst removed or misfueled
will probably not replace the catalyst until forced to in
order to complete the sale. Therefore, the number of
catalysts that are replaced will depend on the fraction of
vehicles which change ownership each year. The same will be
true of vehicle owners who have removed or disabled their air
pump. Since evaporative and PCV tampering is assumed to be
inadvertant and undeterrable, and to recur after repair, no
significant benefit for those systems can be expected in a
change-of-ownership program. No benefits for PCV or
evaporative system inspections have therefore been estimated.
Benefits from a change-of-ownership inspection program assume
that ownership will change in a random fashion, that is older
cars will change owners with the same probability as newer
cars. For this analysis, it is assumed that 15% of the fleet
changes owners each year. This is considered a normal rate.
Some areas may differ. Over the initial four years of the
program (1984 through 1987) about 48% of the fleet will have
changed owners. The benefits therefore assume that 48% of
tampering which occurred before the program began will be
affected by the program. The effectiveness of the inspection
for this 48% will be assumed to be the same as for biennial
inspections. This assumes that the efficiency of the
-------
-95-
inspection will not be significantly less in a
change-of-ownership program than in a biennial program. The
biennial effectiveness values will also be applied to all of
the excess emissions due to tampering that would have
occurred after the program began. This assumes that few
vehicle owners will tamper knowing that the tampering must be
fixed before selling the vehicle.
Tables 31 and 32 show the benefits of a change-of-ownership
inspection program. Both I/M and non-I/M cases are given.
The I/M area case assumes an annual I/M program is in
operation during the change-of-ownership program.
5.3.2 Random Audit for Tampering and Misfuelinq
A random audit inspection program would commit to inspecting
some percentage of the areawide fleet each year, randomly
chosen from all vehicles in operation. Steps would of course
have to be taken by the program to assure that vehicle owners
cannot avoid inspection. Each vehicle would be checked for
tampering and failed if tampering were discovered. The
vehicle owner would then repair or replace the tampered
emission control component and resubmit his or her vehicle
for inspection.
There are at least two ways to implement an audit program.
The more familiar way would be to select vehicles for audit
while they are actually operating on a roadway, and perform
the audit inspection immediately on the roadside. Vehicles
that failed would be issued a "fix it" ticket and be required
to be submitted for reinspection (and examination of repair
receipts in the case of catalyst replacement) at a designated
location within a certain time period. The cost of repair
and the inconvenience of having to be reinspected would serve
as the deterrent to tampering. To ensure compliance with the
reinspection requirement, a fine would be added for late
reinspection, registration denial at next renewal date would
be a back-up enforcement strategy, and court proceedings
would begin for seriously deliquent vehicle owners. The
random roadside approach has the advantage of surprise?
vehicle owners are given no opportunity to conceal tampering
or misfueling before inspection as they would be in a
periodic inspection program. The random roadside approach
has the disadvantages that alert motorists can often avoid an
inspection trap, inspections would be at times that will
often be inconvenient to drivers, and poor weather can make
it difficult to achieve desired inspection volumes.
The second approach is to use the vehicle registration system
to select cars for audits and to enforce the
repair/reinspection requirement. In this approach, owners of
-------
-96-
randomly selected vehicles would receive along with their
license renewal form a notice that their vehicles had been
I^T^cted for a tampering/misfueling audit and that before
registration renewal can be completed the vehicle must pass
inspection at a designated location. This approach would
have a back-up enforcement mechanism to ensure that within a
certain period after the notice the vehicle either passes
inspection or is disposed of outside the program boundaries.
The approach has the advantages that true randomness in
selection is possible, owners cannot evade inspection by
avoiding roadside traps, and inspections are more at the
owner's convenience. A disadvantage is that owners have a
better opportunity to defeat the inspection by reinstalling a
non-functioning catalyst, getting an OEM filler neck
restrictor repair, or replacing or cleaning the tailpipe to
pass the Plumbtesmo inspection. In this regard, the audit
program could be similar to a periodic inspection program.
EPA assumes that the advantage of evasion-proof audits in
this approach counteracts its lack of surprise relative to
the roadisde approach, so that at equal audit rates the two
approaches give equal credit, all else being the same. As
noted below, the registration system lends itself to greater
publicity efforts which may make it more effective, however.
The effectiveness of a roadside pullover program will depend
on the number of vehicles actually inspected and the risk
perceived by vehicle owners that their vehicle will be
selected for audit. Obviously, a program that inspects only
a small percentage of the fleet will present only a small
risk to vehicle owners who tamper.
As with change-of-ownership programs, vehicle owners cannot
be expected to repair previous tampering until they are
inspected. The following is an estimate of the percentage of
the previously tampered vehicles in the fleet which would
have been inspected at least once in the initial four years
of the program depending on the audit rate. Audit rates
greater than 5% are not considered feasible if audits are
performed by random roadside pullover. Higher audit rates
are feasible if the registration system is used to issue
audit orders.
Pullover Percent of Previous Tampering
Rate Detected by January 1, 1988
r% 4%
2% 8%
5% 19%
For the previously tampered vehicles which are inspected, we
will assume the same inspection effectiveness as for a
biennial inspection.
-------
-97-
In addition, it is assumed that some percentage of vehicle
owners will not tamper after the program begins. The number
of vehicle owners who do not tamper will depend on the
visibility of the random audit inspection program, since it
determines the perceived risk of detection. Visibility in
turn will depend on the percentage of vehicles inspected each
year. In this analysis we assume that if 5% of the fleet is
inspected each year, the program will be 50% as effective as
a biennial periodic inspection in deterring new tampering and
misfueling. A 2% pullover program is assumed to be 35% as
effective and a 1% program is assumed to be 25% as effective.
Visibility can be greatly enhanced by intense publicity and
direct notification of vehicle owners. For example, if the
registration system were used to issue audit orders, every
vehicle owner could be notified by direct mail each year
along with his or her renewal form that the audit program
exists and that although his or her vehicle may not have been
selected this year it may be next year, so tampering and
misfueling in the coming year is ill-advised. This notice,
personal accounts of friends and acquaintances who have been
audited, and press coverage would make would-be tamperers
well aware of the consequences. States using this approach;
or a similarly effective publicity approach can be expected
to achieve more credit than indicated by the above paragraph,
and should consult EPA for specific estimates for their
unique program designs.
Some of the new tampering that does occur will be detected
and corrected, as with tampering that occurred prior to the
start of the program. However, the percentage of the
tampered fleet affected will be small due to the low audit
rate, and deterrence will by far have the larger effect. For
calculational simplicity, EPA has not accounted for the small
benefit of detecting a few percent of the undeterred
subsequent tampering and misfueling.
Tables 33 through 38 show the benefits of a random audit
inspection program for these audit rates. Benefits are given
for non-I/M areas and I/M areas separately. Benefits are
shown for the several different inspection types covered in
previous tables. The benefits are smaller than any of the
programs presented earlier, due to less complete coverage and
less effective deterrence. Although costs have not been
calculated for this program, the cost of a roadside
inspection including owner inconvenience is likely to be
higher than an inspection at a licensed garage or state-run
inspection station. Tending to counteract this is the fact
that fewer inspections are performed and much of the benefit
is from deterrence rather than correction. The cost per
vehicle of centralized audits in response to registration
-------
-98-
based audit orders would be the same as in a periodic I/M
program, but far fewer inspection stations would be needed.
The overall result is that a registration based approach is
likely to be highly cost-effective.
5.3.3 Fueling Station Enforcement Program
In this program, plain-clothes State or local enforcement
officers would visit each fuel station unannounced to observe
the fuelings that occur and to discover instances of the
introduction of leaded fuel into catalyst equipped vehicles.
Responsibility for misfueling could be placed with the
vehicle owner. The penalty could be mandatory replacement of
the catalyst on that vehicle. A fine could be imposed if,
within a reasonable period (i.e., one month) after the ticket
had been issued, the catalyst had not been replaced. Court
action to collect the fine could be started after a certain
period. New license plates or renewal tags for that vehicle
could also be denied until the catalyst had been replaced, as
a back-up enforcement tool.
Federal law and regulation also prohibit station owners from
introducing or allowing the introduction of leaded fuel into
catalyst equipped vehicles. It would be possible to enact
similar State prohibitions and enforce them to the extent
practicable. Station personnel observed to allow the
misfueling could also, then, be subject to a penalty.
Fuel station enforcement programs would most likely require
new State or local regulations to be adopted to provide the
implementing agency with the necessary authority. EPA
believes that these approaches or similar ones would be a
good deterrent to misfueling, if well publicized. However,
no benefits are estimated in this report, since such benefits
are highly dependent on the nature and aggressiveness of the
specific State or local program. Areas interested in
establishing fuel station enforcement programs are invited to
discuss with EPA design criteria and emission reduction
potential.
5.3.4 Other Approaches to Remove Incentives for and Aids to
Tampering and Misfueling
Some portion of tampering and misfueling occurs in response
to incentives that a jurisdiction can control. Limiting
those incentives would reduce tampering and fuel switching
without resort to a more negative enforcement/penalty
approach. Even in cases in which a jurisdiction cannot
effectively limit an incentive, it may be able to restrict
the access to the means of tampering and misfueling, and
-------
-99-
thereby reduce their incidence without actually taking
enforcement action against would-be tamperers or misfuelers.
An example of the first approach would be equalization of the
prices of leaded and unleaded gasoline, which would remove
the immediate financial incentive to misfueling. An example
of the second approach would be a ban on the sale of
ready-made devices used to replace a removed catalyst;
removal of a catalyst would then require more expensive
custom fabrication of replacement exhaust piping, of which
not every vehicle owner and small garage is capable. Other
examples of the two approaches may also be possible.
EPA invites jurisdictions who are contemplating approaches in
either or both categories to consult with EPA about the best
design for the approach and the credits which would be
available.
-------
-100-
Table 27
Inspection
Program
Air Pump
Only
Catalyst
Only
Air Pump &
Catalyst
Fuel Inlet
& Catalyst
Fuel Inlet
& Catalyst
& Air Pump
Plumbtesmo
Fuel Inlet
Catalyst
Fuel Inlet
Catalyst &
Air Pump
PCV*
Canister
All Items**
Benefit of Annual Tampering
Inspections in I/M Areas (January 1,
1988)
Af fected
Model
All Items** All Years
(gm/mi)
Per Vehicle
in Emissions
Reduction
(mg/mi)'
Light-Duty Trucks
(0-6000 lbs) (6000-8500 lbs
Years
HC
CO
HC
CO
HC
CO
Pre-1980
4
.74
106
.10
13.
04
291.
65
3
.07
68,
.74
1980-1983
4
.91
131
.10
9.
82
221.
36
10
.43
233,
.24
1984+
1
.15
36
.99
1.
54
37.
46
1
.87
45,
.36
Pre-1980
- 6
.34
61
.04
17.
95
173.
77
7
.43
70,
.83
1980-1983
5
.11
65
.37
26.
09
257.
99
29
.53
283,
.99
1984+
2
.50
45
.29
12.
84
160.
66
13
.82
172,
.38
Pre-1980
11
.37
173
.50
31.
72
481.
77
10
.88
148.
.03
1980-1983
10
.57
211
.04
37.
02
504.
55
41
.30
547.
.14
1984+
3
.78
86
.54
14.
65
204.
90
16
.01
225.
.55
Pre-1980
21
.25
178
.97
55.
19
461.
68
23
.11
188.
.88
1980-1983
23
.10
177
.68
83.
98
692.
15
95
.06
778,
.04
1984+
13
.06
110
. 66
49.
23
424.
73
53
.35
458.
.77
Pre-1980
26
.90
305
.37
71.
00
815.
33
27
.65
290.
.44
1980-1983
29
.93
360
.39
98.
73
1025.
11
111
.32
1141,
.70
1984+
14
.76
165
.41
52.
65
508.
36
57
.31
555.
.60
Pre-1980
32
.98
267
.82
87.
75
707.
38
36
.50
286.
.71
1980-1983
37
.36
259
.92
136.
59
1087.
71
151
.31
1199.
.59
1984+
37
.08
274
.44
91.
38
742.
81
97
.45
790.
.36
Pre-1980
39
.38
410
.91
105.
70
1108.
90
42
.14
412.
.82
1980-1983
45
.55
478
.81
154.
52
1492.
38
171
.33
1647.
,36
1984+
39
.09
339
.01
95 .
66
847.
37
102
.38
910.
.82
Pre-1980
7
.09
0
.00
32.
73
0.
00
43
.02
0.
.00
1980-1983
5
.80
0
.00
22.
17
0.
00
34
.69
0.
.00
1984+
3
.86
0
.00
14.
55
0.
00
23
.26
0.
.00
Pre-1980
3
.02
0
.00
20.
26
0.
00
4
.95
0.
,00
1980-1983
3
.28
0
.00
26.
15
0.
00
21
.10
0.
,00
1984+
1
.55
0
.00
18.
47
0.
00
15
.35
0.
,00
Pre-1980
49
.49
410
.91
158.
69
1108.
90
90
. 11
412 .
,82
1980-1983
54
.63
478
.81
202.
84
1492.
38
227
.12
1647.
,36
1984+
44
.50
339
.01
128.
68
847.
37
140
.99
910.
,82
148.62 1228.73 490.21 3448.65 458.22 2971.00
(0.15) (1.23) (0.49) (3.45) (0.46) (2.97)*
*PCV or evaporative canister benefits can be added directly to any of
the above programs.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
Table 28
-101-
Benefit of Biennial Tampering
Inspections in i/m Areas (January 1, 1988)
Inspection
program
Air pump
Only
Catalyst
Only
Air Pump &
Catalyst
Fuel Inlet
& Catalyst
Fuel inlet
& Catalyst
& Air Pump
Plumbtesmo
Fuel Inlet
Catalyst
Plumbtesmo
Fuel Inlet
Catalyst &
Air Pump
Affected
Model
Per Vehicle Reduction
in Emissions (mg/mi)
Light-Duty Trucks
Years
HC
CO
HC
CO
HC
CO
Pre-1980
4.
15
92
.84
11.
41
255.
19
2.
69
60.
.14
1980-1983
4.
29
114
.71
8.
59
193.
69
9.
12
204,
.08
1984+
1.
01
32
.37
1.
35
32.
78
1.
63
39.
,69
Pre-1980
6.
34
61
.04
17.
95
173.
77
7.
43
70.
.83
1980-1983
5.
11
65
.37
26.
09
257.
99
29.
53
283.
,99
1984 +
2.
50
45
.29
12.
84
160.
66
13.
82
172.
.38
Pre-1980
10.
74
159
.44
30.
00
443.
27
10.
45
13a.
.38
1980-1983
9.
89
192
.83
35.
66
473.
73
39.
83
514.
.25
1984+
3.
62
81
.38
14.
43
199.
37
15.
74
218.
,90
Pre-1980
21.
25
178
.97
55.
19
461.
68
23.
11
188.
.88
1980-1983
23.
10
177
.68
83.
98
692.
15
95.
06
778.
.04
1984+
13.
06
110
.66
49.
23
424.
73
53.
35
458,
.77
Pre-1980
26.
27
291
.32
69.
28
776.
83
27.
22
280.
.79
1980-1983
29.
25
342
.19
97.
37
994.
29
109.
84
1108.
.81
1984+
14.
60
160
.25
52.
42
502.
83
57.
04
548.
,96
Pre-1980
32.
98
267
.82
87.
75
707.
38
36.
50
286,
.71
1980-1983
37.
36
259
.92
136.
59
1087.
71
151.
.31.
1199,
.59
1984 +
37.
08
274
.44
91.
38
742.
81
97.
45
790,
.36
Pre-1980
38.
63
394
.18
103.
60
1062.
38
41.
51
398,
.84
1980-1983
44.
58
452
.96
152.
44
1445.
78
169.
01
1595,
.94
1984+
38.
84
330
.94
95.
12
834.
30
101.
76
895,
.76
PCV'
Canister
All Items**
Pre-1980
4
.83
0
.00
22
.32
0
.00
29
.33
0
.00
1980-1983
3
.95
0
.00
15
.12
0
.00
23
.65
0
.00
1984 +
2
.63
0
.00
9
.92
0
.00
15
.86
0
.00
Pre-1980
2
.11
0
.00
14
.14
0
.00
3
.46
0
.00
1980-1983
2
.29
0
.00
18
.25
0
.00
14
.72
0
.00
1984 +
1
.08
0
.00
12
.88
0
.00
10
.71
0
.00
Pre-1980
45
.57
394
.22
140
.06
1062
.56
74
.30
398
.93
1980-1983
50
.82
453
.02
185
.80
1446
.03
207
.38
1596
.21
1984 +
42
.55
330
.94
117
.92
834
.30
128
.33
895
.76
All Years
138
.94
1178
.18
443
.78
3342
.89
410
.01
2890
.90
(gm/mi)
(0.
14)
(1.
18)
(0.
44)
(3.
34)
(0.
41)
(2.
89)
*PCV or evaporative canister benefits can be added directly to any of
the above programs.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
Table 29
-102-
Benefit of Annual Tampering
inspections in Non-I/M Areas (January 1, 1988)
Inspection
Program
Air Pump
Only
Catalyst
Only
Air Pump &
Catalyst
Fuel Inlet
& Catalyst
Fuel Inlet
& Catalyst
& Air Pump
Plumbtesmo
Fuel Inlet
Catalyst
Plumbtesmo
Fuel Inlet
Catalyst &
Air Pump
PCV*
Affected
Model
Years
Per Vehicle Reduction
in Emissions (mg/mi)
Passenger Car
HC CO
Pre-1980
10.
.75
240.
.50
1980-1983
10.
.82
289.
.07
1984+
2.
.59
83.
.14
Pre-1980
20.
.25
198.
.29
1980-1983
15.
.78
206.
.64
1984 +
5.
.54
100.
.38
Pre-1980
31.
.67
453.
.73
1980-1983
27.
,87
529.
.83
1984+
8.
.42
193,
.06
Pre-1980
35.
.83
318.
.50
1980-1983
32.
.21
300.
.76
1984+
14.
.93
157.
.44
Pre-1980
48,
.12
593.
.43
1980-1983
46.
.03
670.
.25
1984+
18.
.23
263,
.50
Pre-1980
83.
.50
683,
.63
1980-1983
84.
.50
597,
.15
1984 +
53.
.02
397.
.17
pre-1980
98.
.57
1020.
.73
1980-1983
103.
.69
1109.
.97
1984 +
57,
.58
543,
.48
Pre-1980
7,
.09
0,
.00
1980-1983
5,
.80
0,
.00
1984 +
3,
.86
0,
.00
(0-6000
Light-Duty Trucks
lbs)
HC
CO
(6000-8500 lbs)
HC CO
35.47 793.58
28.03 632.34
7.05 173.27
8.05
28.98
7.98
180.18
648.20
196.08
67.12 658.90 27.88 270.51
98.60 985.86 111.46 1085.66
49.50 617.57 53.23 662.24
104.70
130.09
57.80
110.58
162.53
84.02
151.69
200.43
95.28
231.63
341.38
193.09
282.75
396.61
211.94
32.73
22.17
14.55
1499.50
1696.40
821.84
978.90
1434.32
843.04
1898.54
2289.87
1121.21
1864.93
2689.29
1574.45
3008.50
3936.47
2041.96
0.00
0.00
0.00
37.03
144.51
62.59
45.78
183.05
90.61
56.80
223.56
103.17
95.57
382.67
207.99
111.83
443.43
228.83
43.02
34.69
23.26
475.20
1825.13
892.55
396.61
1590.13
906.54
643.04
2496.33
1216.88
744.73
2993.91
1692.70
1108.59
4353.06
2209.07
0.00
0.00
0 .00
Evaporative* Pre-1980
Canister 1980-1983
1984 +
3.02
3.28
1.55
0.00
0.00
0.00
20.26
26.15
18.47
0.00
0.00
0.00
4.95
21.10
15.35
0 .00
0 .00
0.00
All Items**
Pre-1980
1980-1983
1984 +
108.68 1020.73
112.77 1109.97
62.99 543.48
335.74 3008.50 159.80 1108.59
444.93 3936.47 499.22 4353.06
244.96 2041.96 267.44 2209.07
All Items** All Years
(gm/mi)
284.44 2674.18 1025.63 8986.93 926.46 7670.72
(0.28) (2.67) (1.03) (8.99) (0.93) (7.67)
*PCV or evaporative canister benefits can be added directly to any of
the above programs.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
Inspection
Program
Air Pump
Only
Catalyst
Only
Air Pump &
Catalyst
Fuel inlet
& Catalyst
Fuel Inlet
& Catalyst
& Air Pump
Plumbtesmo
Fuel Inlet
Catalyst
Plumbtesmo
Fuel Inlet
Catalyst &
Air Pump
PCV*
Taole 30
Benefit of Biennial Tampering
Inspections in Non-I/M Areas (January 1, 1988)
-103-
Pre-1980
1980-1983
1984 +
Per Vehicle Reduction
in Emissions (mg/mi)
Affected
Light
-Duty Trucks
Model
Passenger Car
(0-
¦6000 lbs)
(6000
-8500 lbs)
Years
HC
CO
HC
CO
HC
CO
Pre-1980
9
.41
210.
44
31,
.04
694.
39
7.
05
157
.66
1980-1983
9
.47
252.
94
24.
.52
553.
30
25.
35
567
.18
1984 +
2
.26
72.
75
6.
.17
151.
61
6.
98
171
.57
Pre-1980
20
.25
198.
29
67,
.12
" 658.
90
27.
88
270
.51
1980-1983
15
.78
206.
64
98.
.60
985.
86
111.
46
1085
.66
1984 +
5
.54
100.
38
49.
.50
617.
57
53.
23
662
.24
Pre-1980
30
.24
421.
80
100.
.00
1394.
42
35.
89
449
.61
1980-1983
26
.36
489.
43
126,
.15
1607.
58
140.
38
1732
.69
1984+
8
.06
181.
47
56.
.76
796.
30
61.
42
863
.76
Pre-1980
35
.83
318.
50
110.
.58
978.
90
45.
78
396
.61
1980-1983
32
.21
300.
76
162.
.53
1434.
32
183.
05
1590
.13
1984 +
14
.93
157.
44
84,
.02
843.
04
90.
61
906
.54
Pre-1980
46
.35
558.
62
145,
.13
1777.
61
54.
71
609
.43
1980-1983
44
.11
626.
58
193,
.54
2176.
18
216.
16
2376
.13
1984+
17
.87
251.
91
93,
.85
1092.
55
101.
60
1184
.79
Pre-1980
83
.50
683.
63
231.
.63
1864.
93
95.
57
744
.73
1980-1983
84
.50
597.
15
341.
.38
2689.
29
382.
67
2993
.91
1984+
53
.02
397.
17
193,
.09
1574.
45
207.
99
1692
.70
Pre-1980
96
.83
981.
68
276,
.95
2878.
79
110.
10
1069
.69
1980-1983
101
.43
1049.
46
390,
.52
3798.
95
436.
75
4203
.84
1984+
57
.01
525.
19
209,
.66
1985.
38
226.
30
2146
.50
4.83
3.95
2.63
0.00
0.00
0.00
22.32
15.11
9.92
0.00
0.00
0.00
29.33
23.65
15.86
0.00
0.00
0.00
Evaporative* Pre-1980
Canister 1980-1983
1984 +
2.11 0.00 14.14 0.00 3.46 0.00
2.29 0.00 18.25 0.00 14-72 0 .00
1.08 0.00 12.88 0.00 10.71 0.00
All Items**
Pre-1980
1980-1983
1984+
All Items** All Years
(gm/mi)
103.77 981.68
107.67 1049.46
60.72 525.19
313.41 2878.79 142.89 1069.69
423.88 3798.95 475.12 4203.84
232.46 1985.38 252.87 2146.50
272.16 2556.33 969.75 8663.12 870.88 7420.03
(0.27) (2.56) (0.97) (8.66) (0.87) (7.42)
*PCV or evaporative canister benefits can be added directly to any of
the above programs.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-104-
Table 31
Benefit of Tampering Inspections At Change
of Ownership in Non-l/M Areas* (January 1, 1988)
Inspection
Program
Air Pump
Only-
Catalyst
Only
Air Pump &
Catalyst
Fuel Inlet
& Catalyst
Fuel Inlet
& Catalyst
& Air Pump
Plumbtesmo &
Fuel Inlet &
Catalyst
Plumbtesmo &
Fuel Inlet &
Catalyst &
Air Pump
All Items**
Affected
Model
Per Vehicle Reduction
in Emissions (mg/mi)
Passenger Car
Light-Duty Trucks
(0-6000 lbs) (6000-8500 lbs,
Years
HC
CO
HC
CO
HC
CO
Pre-1980
6
.05
135
.24
19
.34
432
.64
4
.78
106
.97
1980-1983
7
.67
206
.95
18
.98
428
.62
20
.10
449
.57
1984+
2
.26
72
.75
6
.14
150
.93
6
.95
170
.79
Pre-1980
13
.44
131
.63
39
.32
384
.71
16
.76
161
.79
1980-1983
13
.12
175
.56
63
.43
630
.24
72
.64
702
.14
1984+
5
.54
100
.38
45
.33
567
.74
48
.83
609
.56
Pre-1980
19
.87
275
.47
59
.82
843
.28
22
.17
282
.80
1980-1983
21
.70
407
.10
84
.64
1109
.42
95
.40
1211
.23
1984+
8
.06
181
.47
52
.52
744
. 66
56
.94
809
.09
Pre-1980
23
.91
212
.40
65
.83
579
.96
28
.03
241
.21
1980-1983
26
.90
253
.78
107
.21
937
.48
122
.29
1052
.29
1984+
14
.93
157
.44
77
.46
777
.39
83
.67
837
.12
Pre-1980
30
.76
367
.95
87
.54
1078
.63
34
.13
384
.39
1980-1983
36
.72
522
.67
131
.32
1503
.11
148
.55
1662
.58
1984+
17
.87
251
.91
87
.20
1021
.34
94
.55
1109
.42
Pre-1980
54
.53
445
.83
138
.23
1106
.85
58
.68
453
.58
1980-1983
67
.08
472
.64
226
.25
1760
.98
256
.81
1985
.57
1984+
51
.92
389
.22
179
.04
1455
.67
193
. 14
1567
.10
Pre-1980
63
.15
638
.84
166
.30
1734
.83
68
.19
666
.28
1980-1983
80
.85
844
.08
262
.34
2576
.89
297
. 13
2887
.48
1984+
55
.90
517
.24
194
.91
1849
.47
210
.70
2002
.70
All Years
199
.90
2000
.16
623
.55
6161
.19
576
.02
5556
.46
(gm/mi)
(0.
20)
(2.
00)
(0.
62)
(6.
16)
(0.
58)
(5.
56)
~Assumes a random 15% changeover of the fleet each year with prograi
beginning January 1, 1984.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-105-
Table 32
Inspection
Program
Air Pump
Only
Catalyst
Only
Air Pump &
Catalyst
Fuel Inlet
& Catalyst
Fuel Inlet
& Catalyst
& Air Pump
Plumbtesmo
Fuel Inlet
Catalyst
Plumbtesmo
Fuel Inlet
Catalyst &
Air Pump
Benefit of Tampering Inspections At Change
of Ownership in I/M Areas* (January 1, 1988)
Per Vehicle Reduction
~ in Emissions (mg/mi)
All Items** All Years
(gm/mi)
Affected
Model
Years
Passenger Car
Light-Duty Trucks
(0-6000 lbs)
(6000-8500 lbs
HC
CO
HC
CO
HC
CO
Pre-1980
2.
,67
59.
.78
7,
.42
165.
,88
1.
,93
43.
.15
1980-1983
3.
,50
94.
.27
7.
,28
164.
,26
7.
,87
176.
,11
1984+
1.
,01
32,
.37
1.
.35
32.
,78
1.
,63
39.
,69
Pre-1980
4.
,09
39.
.31
10.
.59
101.
,77
4.
,50
42.
.41
1980-1983
3.
,98
51.
.50
16.
.96
165.
.20
19.
.45
183.
.86
1984+
2.
,50
45.
.29
11.
.79
148.
,00
12.
,72
158.
.98
Pre-1980
3.
,53
73,
.91
7,
.91
187.
.78
2.
,61
56.
.37
1980-1983
6.
,06
141.
.96
12.
.24
239.
,55
14.
,37
263.
.37
1984+
3.
.62
81,
.38
11.
.52
171.
,61
12.
.66
189.
,55
Pre-1980
14.
,76
123.
.90
35.
.06
291.
,10
15.
,18
122.
,86
1980-1983
19,
.64
149,
.48
60,
.10
487.
.15
68,
.85
554.
.80
1984+
13,
,06
110.
. 66
46.
.49
398.
,30
50.
.45
430.
.81
Pre-1980
18,
.03
197,
.04
44,
.23
496.
.22
18.
.07
187.
.64
1980-1983
24.
,67
285.
.35
71.
.12
736.
,17
81.
.18
830.
.69
1984+
14,
.60
160,
.25
49,
.68
476,
.41
54,
.14
521,
.00
Pre-1980
20.
.12
162,
.65
50.
.08
399.
,36
21,
.27
164.
.23
1980-1983
25.
.72
174,
.44
83,
.72
649,
.35
. 93,
.89
724,
.18
1984+
34.
.70
257.
.35
83.
.59
675.
.51
89.
,22
719.
.23
Pre-1980
23 ,
.77
244,
.26
60,
.33
628,
.65
24,
.73
241,
.72
1980-1983
31.
.56
331,
.80
96.
.55
939,
.41
108,
.39
1048,
.47
1984+
36,
.46
313,
.85
87,
.33
767,
.00
93.
.53
824,
.63
91.79 889.91 244.21 2335.06 226.65 2114.82
(0.09) (0.89) (0.24) (2.34) (0.23) (2.11)
*Assumes a random 15% changeover of the fleet each year with program
beginning January 1, 1984.
**Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-106-
Table 33
Benefit of Anti-Tampering Inspections
During 5% Random Audit Non-l/M Areas
Inspection
Program
Air Pump
Only
Affected
Model
Years
Pre-1980
1980-1983
1984+
Per Vehicle Reduction
in Emissions (mg/mi)
Passenger Car
HC CO
2.65
3.64
1.13
59.27
98.37
36.38
Light-Duty Trucks
(0-6000 lbs) (6000-8500 lbs)
HC
8.37
8.87
3.07
CO
187.24
200.46
75.39
HC
2.14
9.46
3.47
CO
47.86
211.72
85.31
Catalyst
Only
Pre-1980
1980-1983
1984+
6.06 59.38 16.89 165.05 7.28 70.12
6.34 85.38 28.28 280.40 32.55 313.74
2.80 50.72 22.46 281.53 24.20 302.35
Air Pump &
Pre-1980
8,
.88
122,
.45
25,
.77
363.
.56
9,
.70
124
.17
Catalyst
1980-1983
10.
.41
195.
.45
38.
.18
504.
.15
43.
,25
552,
.95
1984+
4.
.06
91,
.27
26.
.04
369.
.79
28,
.25
401,
.90
Fuel Inlet
Pre-1980
10,
.71
95.
.22
28.
.18
248,
.19
12.
.14
104,
.41
& Catalyst
1980-1983
12.
.93
122.
.66
47,
.83
417.
.57
54.
.82
470,
.85
1984+
7.
.49
79.
.25
38.
.25
384.
,51
41.
,33
414.
,18
Fuel Inlet
Pre-1980
15,
.56
197,
.43
42,
.80
560.
.74
16.
,94
202,
.46
& Catalyst
1980-1983
22.
.97
345.
.64
69.
.50
857.
,36
78.
.98
952.
.42
& Air Pump
1984+
11,
.22
168.
.51
49,
.10
600.
.20
53,
.46
655,
.80
Plumbtesmo & Pre-1980
24.
.45
199.
.85
59.
.84
478.
,11
25.
, 70
198,
.03
Fuel Inlet & 1980-1983
32,
.05
225,
.58
102.
.02
790.
.57
116 ,
.32
895,
.52
Catalyst
1984+
26.
.14
195.
.98
89.
,04
723.
.27
96.
,08
778,
.88
Plumbtesmo & Pre-1980 28.01 282.69 70.68 738.30 29.25 286.38
Fuel Inlet & 1980-1983 38.30 400.05 116.58 1148.74 132.69 1293.12
Catalyst & 1984+ 28.13 259.99 96.62 916.11 104.49 992.38
Air Pump
All Items* All Years 94.44 942.73 283.88 2803.15 266.43 2571.88
(gm/mi) (0.09) (0.94) (0.28) (2.80) (0.27) (2.57)
~Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-107-
Table 34
Benefit of Anti-Tampering Inspections
During 2% Random Audit in Non-l/M Areas
Affected
Inspection
Model
Passenger Car
Program
Years
HC
CO
Air Pump
Pre-1980
1.49
33.38
Only
1980-1983
2.38
64.79
1984+
0.81
25.98
Catalyst
Pre-1980
3.42
33.51
Only
1980-1983
4.07
55.29
1984+
1.92
34.87
Air Pump &
Pre-1980
5.01
69.08
Catalyst
1980-1983
6.74
127.81
1984+
2.82
63.83
Fuel Inlet
Pre-1980
6.31
55.82
& Catalyst
1980-1983
8.62
80.94
1984+
5.37
55.79
Fuel Inlet
Pre-1980
8.02
94.69
& Catalyst
1980-1983
11.72
166.02
& Air Pump
1984+
6.42
89.66
Plumbtesmo
St
Pre-1980
14.11
114.98
Fuel Inlet
&
1980-1983
20.58
144.08
Catalyst
1984+
18.10
135.45
Plumbtesmo
&
Pre-1980
16.00
160.82
Fuel Inlet
&
1980-1983
24.55
258.13
Catalyst &
1984+
19.53
181.17
Air Pump
All Items*
All Years
60.08
600.12
(gm/mi)
(0.06)
(0.60)
Per Vehicle Reduction
in Emissions (mg/mi")
Light-Duty Trucks
(0-6000 lbs) (6000-8500
TFsT
HC
CO
HC
CO
4.59
5.68
2.19
8.57
15.59
14.98
13.43
21.90
17.54
15.25
28.07
26.35
20.42
35.27
29.82
32.19
59.30
60.48
37.34
67.30
65.70
102.59
128.34
53.77
83.40
153.77
188.09
192.22
296.61
250.91
132.63
241.39
262.21
251.55
409.27
348.88
255.07
454.38
489.62
390.39
668.05
624.42
1.26
6.13
2.48
3.78
18.12
16.15
5.20
25.03
19.03
6.73
32.44
28.49
8.33
40.41
32.37
14.15
68.16
65.29
15.81
77.27
71.10
28.15
137.23
60.84
36.28
173.63
202.09
67.97
328.07
272.96
57.05
274.65
282.63
94.56
458.72
379.39
107.89
519.16
527.50
155.32
758.46
676.88
(1.68) (0.16) (1.59)
~Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-108-
Table 35
Benefit of Anti-Tampering Inspections
During 1% Random Roadside Audit in Non-I/M Areas
Inspection
Af fected
Model
Per Vehicle Reduction
in Emissions (mg/miT
Light-Duty Trucks
Air Pump
All Items*
All Years
(gm/mi)
41.36
(0.04)
417 .38
(0.42)
Program
Years
HC
CO
HC
CO
HC
CO
Air Pump
Pre-1980
1.02
22.78
3.11
69.50
0.87
19.42
Only
1980-1983
1.69
45.88
3.99
90.-3 3
4.33
96.84
1984+
0.58
18.71
1.57
38.71
1.78
43.79
Catalyst
Pre-1980
2.33
22.89
5.61
54.55
2.51
23.96
Only
1980-1983
2.90
39.52
10.55
103.91
12.32
117.73
1984+
1.40
25.36
10.81
135.68
11.65
145.80
Air Pump &
Pre-1980
3 .42
47.17
8.91
128.29
3 .48
45.82
Catalyst
1980-1983
4.79
90.88
14.99
204.39
17.18
226.64
1984+
2.05
46.21
12.64
180.88
13.72
196.79
Fuel Inlet
Pre-1980
4.26
37.78
9.67
84.54
4.33
36.87
& Catalyst
1980-1983
6.14
57.81
18.60
160.45
21.60
183 .38
1984+
3.90
40.58
18.93
188.66
20.47
203.39
Fuel Inlet
Pre-1980
5.46
64.47
13.30
165.55
5.50
62.92
& Catalyst
1980-1983
8.37
118.32
23.84
278.93
27.43
313.62
& Air Pump
1984+
4.66
65.00
21.46
251.17
23 .30
273.18
Plumbtesmo
&
Pre-1980
9.14
74.65
20.13
159.83
8.97
68.48
Fuel Inlet
&
1980-1983
14.00
98.80
38.69
296.72
44.66
340.39
Catalyst
1984+
12.69
95.52
42.35
344.63
45.72
371.35
Plumbtesmo
&
Pre-1980
10.61
107.53
24.60
259.75
10.61
105 .27
Fuel Inlet
&
1980-1983
17.04
181.41
45.89
459.67
52.83
523.17
Catalyst &
1984+
13.71
128.44
46.31
443.16
50.12
480.46
116.80 1162.58
(0.12) (1.16)
113.56
(0. 11)
1108.90
(1.11)
*Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-109-
Table 36
Benefit of Anti-Tampering Inspections
During 5% Random Audit in I/M Areas
Per Vehicle Reduction
Inspection
Af fected
Model
in Emissions (mg/mif
Passenger Car (0-6000 lbs) (
Light-Duty Trucks
6000-
8500 lbs)
Program
Years
HC
CO
HC
CO
HC
CO
Air Pump
Pre-1980
1.17
26.22
3.26
73.02
0.88
19.69
Only
1980-1983
1.66
44.86
3.49
78.86
3.80
84.95
1984+
0.50
16.18
0.67
16.39
0.82
19.84
Catalyst
Pre-1980
1.83
17.54
4.56
43.72
1.96
18.39
Only
1980-1983
1.89
24.54
7.59
73.55
8.75
82.18
1984+
1.26
22.88
5.85
73.43
6.31
78.89
Air Pump &
Pre-1980
3.07
45.37
8.01
120.91
2.94
40.38
Catalyst
1980-1983
3.73
74.33
11.46
160.99
13.00
177.36
1984+
1.82
40.93
6.64
92.78
7.27
102.16
Fuel Inlet
Pre-1980
6.71
56.23
15.43
127.76
6.76
54.56
& Catalyst
1980-1983
9.47
71.99
27.59
222.48
31.73
254.45
1984+
6.54
55.57
23.02
197.14
24.99
213 .28
Fuel Inlet
Pre-1980
8.15
88.48
19.47
218.24
8.08
84.00
& Catalyst
1980-1983
11.86
136.78
32.85
341.30
37.63
386.53
& Air Pump
1984+
7.31
80.36
24.61
236.19
26.83
258.38
Plumbtesmo
&
Pre-1980
8.79
70.92
21.27
168.78
9.11
69.86
Fuel Inlet
&
1980-1983
11.76
78.99
36.64
281.15
41.29
315.00
Catalyst
1984+
17.30
128.32
41.44
334.38
44.24
356.08
Plumbtesmo
&
Pre-1980
10.31
106.09
25 .42
266.52
10.50
103 .28
Fuel Inlet
&
1980-1983
14.39
152.73
42.24
414.57
47.65
464.59
Catalyst &
1984+
18.18
156.56
43.31
380.13
46.39
408.78
Air Pump
All Items*
All Years
(gm/mi)
42.88
(0.04)
415.38
(0.42)
110.97
(0.11)
1061.22
(1.06)
104.54
(0.10)
976.65
(0.98)
~Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-110-
Table 37
Benefit of Anti-Tampering Inspections
During 2% Random Audit in i/M Areas
Per Vehicle Reduction
in Emissions (mg/mi)
Affected Light-Duty Trucks
Inspection
Model
Passeng
er Car
(0-5000 lbs)
(6000-8500 lbs.
Program
Years
HC
CO
HC
CO
HC
CO
Air Pump
Only
Pre-1980
1980-1983
1984+
0.66
1.09
0.36
14.79
29.61
11.56
1.86
2.34
0.48
41.53
52.83
11. 71
0.54
2.56
0.58
12.04
57.35
14.17
Catalyst
Only
Pre-1980
1980-1983
1984+
1.01
1.17
0.87
9.67
15.35
15.73
2.33
4.22
3.91
22.16
40.38
49.09
1.02
4.91
4.22
9.53
45.52
52.77
Air Pump &
Catalyst
Pre-1980
1980-1983
1984+
1.71
2.38
1.27
25.39
48.20
28.62
4.29
6.80
4.47
66.09
98.86
62.91
1.62
7.77
4.90
22.94
109.62
69.39
Fuel Inlet
& Catalyst
Pre-1980
1980-1983
1984+
4.14
6.39
4.64
34.53
48.04
39.08
9.02
17.26
16.02
73.96
137.36
136.28
4.04
19.96
17.40
32.23
158.25
147.51
Fuel Inlet
& Catalyst
& Air Pump
Pre-1980
1980-1983
1984+
4.97
7.98
5.19
53.04
91.11
56.79
11.35
20.76
17.16
125.98
216.49
164.17
4.84
23 .92
18.72
50.27
246.72
179.72
Plumbtesmo
Fuel Inlet
Catalyst
&
&
Pre-1980
1980-1983
1984+
4.84
7.04
11.83
38.86
46 .42
87.64
11.06
20.26
28.09
86.61
151.84
225.88
4.83
23 .04
29.99
36.35
171.69
240.58
Plumbtesmo
Fuel Inlet
Catalyst &
Air Pump
&
&
Pre-1980
1980-1983
1984+
5.66
8.71
12.45
58.37
94.43
107.81
13 .21
23.70
29.42
140.22
237.32
258.55
5.56
26.97
31.53
55.47
268.04
278.22
All Items*
All Years
(gm/mi)
26.82
(0.03)
260.61
(0.26)
66.33
(0.07)
636.09
(0.64)
64.06
(0.06)
601.73
(0.60)
~Plumbtesmo, fuel inlet,
canister checks.
catalyst, air
pump,
PCV i
and evaporative
-------
-Ill-
Table 38
Benefit of Anti-Tampering Inspections
During 1% Random Audit in I/M Areas
Per Vehicle Reduction
Inspection
Affected
Model
in Emissions (mg/miT
Passenger Car
Light-Duty Trucks
(0-6U00 Ibs)(6000-8500
lbs)
Air Pump
All Items*
All Years
(gm/mi)
18.23
(0.02)
179.20 44.51
(0.18) (0.04)
430.65 43.37
(0.43) (0.04)
Program
Years
HC
CO
HC
CO
HC
CO
Air Pump
Pre-1980
0 .45
10.10
1.27
28.41
0.37
8.39
Only
1980-1983
0.77
20.98
1.66
37.55
1.82
40.82
1984+
0.26
8.32
0.35
8.43
0.42
10.21
Catalyst
Pre-1980
0.68
6.55
1.53
14.51
0.68
6.30
Only
1980-1983
0.83
10.85
2.86
27.30
3.34
30.88
1984+
0.63
11.44
2.82
35.42
3.04
38.08
Air Pump &
Pre-1980
1.16
17.29
2.87
44.57
1.10
15.63
Catalyst
1980-1983
1.68
34.13
4.70
68.86
5.38
76.48
1984+
0.92
20.72
3.23
45.38
3.53
50.05
Fuel Inlet
Pre-1980
2 .68
22.43
5.58
45.86
2.55
20.36
& Catalyst
1980-1983
4.45
33.53
11.38
90.40
13 .23
104.65
1984+
3.35
28.25
11.44
97.40
12.43
105.45
Fuel Inlet
Pre-1980
3.24
34.93
7.15
80.87
3 .10
32.60
& Catalyst
1980-1983
5.56
63.82
13.82
145.75
15.99
166.60
& Air Pump
1984+
3.74
41.00
12.27
117.48
13.38
128.64
Plumbtesmo
&
Pre-1980
3.04
24.42
6.71
52.37
2.96
22.18
Fuel Inlet
&
1980-1983
4.61
30.40
12.75
94.69
14.58
107.65
Catalyst
1984+
8.23
61 .25
19. 54
157.49
20.86
167 .77
Plumbtesmo
&
Pre-1980
3.66
38.25
8.45
91.25
3.61
36.56
Fuel Inlet
&
1980-1983
5.89
65.17
15.56
158.39
17.79
179.50
Catalyst &
1984+
8.68
75.78
20.50
181.01
21.97
194.87
410.93
(0.41)
~Plumbtesmo, fuel inlet, catalyst, air pump, PCV and evaporative
canister checks.
-------
-112-
6.0 ADJUSTMENT TO LOCAL CONDITIONS: USE OF MOBILE3 TO
CALCULATE PROGRAM BENEFITS
All of the tables in this report assume standard FTP ambient
conditions and use national average fleet descriptors. As a
result, the emission reductions estimated must only be
compared to emission levels for cars and trucks which were
also estimated for standard conditions using national
averages.
Although this report has extensive tables which provide the
emission reductions estimated for various anti-tampering and
anti-misfueling programs, it is likely that, in many cases,
the particular inspection scenario preferred by a local area
may differ from any of the scenarios provided in the tables.
For example, all of the tables assume an inspection program
begins on January 1, 1984. If an area has already begun the
inspection program before that date or will not have started
inspections until later, then the tables provided in this
report will not accurately match the needs of the user.
Also, local conditions such as vehicle model year
registration distributions or tampering and misfueling rates
which differ significantly from the national averages may
make the estimates in the table inappropriate for a local
area. Care should therefore be used when using the tables in
this report to estimate potential emission reduction benefits
from inspection programs.
The upcoming MOBILE3 emission factor model, used by EPA and
most areas to predict emission levels from highway mobile
sources, includes the effects of tampering and misfueling
using the assumptions and estimates developed for this
report. The model also incorporates the capability to adjust
these emission estimates for local conditions. As a result,
MOBILE3 is ideally suited to predict the potential emission
reduction benefits for areas which have local conditions
which differ from national averages.
M0BILE3 also allows users to obtain the emission reductions
due to inspection programs by selecting the appropriate
inspection program options entering the inspection program
parameters, and providing a set of emission reduction
effectiveness factors. MOBILE3, as released, will not
contain these emission reduction effectiveness factors, since
they will depend on so many variables that M0BILE3 could not
reasonably store them all. EPA is prepared, however, to
provide these factors to M0BILE3 users for any specific
inspection program scenario specified by a local area. If
preferable, a local area may request that EPA estimate the
emission reductions. Local areas should not attempt to
estimate the emission reduction factors for themselves.
-------
-113-
Once the inspection program parameters, local conditions, and
emission reduction factors have been input into M0BILE3, users
will be able to use the M0BILE3 output directly in SIP
calculations. If M0BILE3 is not available, local areas should
consult with EPA on the appropriate method to adjust mobile
source emission estimates to account for an inspection program.
Once M0BILE3 output reflecting an anti-tampering/anti-misfueling
program has been obtained, it can be used in SIP inventory
calculations in exactly the same manner as normal M0BILE3
output. The incremental effect of the
anti-tampering/anti-misfueling program can be determined by
using M0BILE3 to calculate two inventories, one with and one
without the program. By subtracting one from the other, the
user will be able to identify the credit attributable to the
program.
For example, an area using M0BILE3 predicts a 25.00 gm/mi
highway mobile source composite CO emission factor in a
particular evaluation year without a tampering inspection
program. It predicts a 23.00 gm/mi CO emission factor using the
EPA supplied emission reduction factors specific to the proposed
tampering inspection program. The incremental benefit of the
inspection program is therefore 2.00 gm/mi CO. In this area,
the highway mobile source CO inventory is 1000 tons per day at
25 gm/mi without the inspection program. The incremental
benefit of the inspection program in tons is therefore:
(2.00/25.00) * 1000 tons/day = 80 tons/day
All of the tables in this report contain the incremental
benefits of the various programs in milligrams per mile. If the
incremental benefit of an inspection program is needed in tons
and M0BILE3 is not available, the tables in this report can be
used to estimate the benefit, if the program parameters (i.e.,
start date) do not differ significantly from the parameters
assumed for the tables.
First the separate mg/mi benefits for LDV, LDT1, LDT2 must be
combined into a single mg/mi benefit from mobile sources. For
example, if the tables show a 2330 mg/mi CO benefit for LDV, a
3000 mg/mi CO benefit for LDT1 and a 2500 mg/mi CO benefit for
LDT2 for a proposed tampering inspection program, the three
benefits must first be combined. This is done by weighting the
benefit from each vehicle type by the fraction of the total VMT
(vehicle mileage traveled) contributed by each of the vehicle
types to the composite VMT for all highway mobile sources. In
this example, let's assume that LDV accounts for 70% of all VMT
by highway mobile sources and LDT1 and LDT2 each contribute 8%
a nd 5 % .
-------
-114-
The composite benefit would therefore be:
(2330 * 0.70) + (3000 * 0.08) + (2500 * 0.05) - 2000 mg/mi CO.
This 2000 mg/mi is then converted (by dividing by 1000) to give
2.00 grams per mile to be compared to MOBILE2 output which is
reported in grams per mile.
Second the benefits in the tables in this report must be
adjusted for local conditions. This can be roughly estimated by
comparing the mobile source composite emission factor used to
compile the inventory to the composite emission factor that
would result if FTP ambient conditions and national average
fleet descriptors had been used. For example, if MOBILE2 were
used to compile a January 1, 1988 CO inventory, the local
composite emission factor would be compared to the 27.84 gm/mi
predicted by M0BILE2 using FTP conditions and national averages.
If, for example, the local area used M0BILE2 with alternate
vehicle registration distributions and non-FTP conditions to
estimate the mobile source inventory on January 1, 1988, with a
resultant 25.00 gm/mi composite CO emission factor, all the
benefits in the tables in this report must be adjusted by the
factor 25.00/27.84 to roughly correct for the effect of local
conditions on the benefits.
2000 * (25.00/27.84) = 1.80 gm/mi.
Next, as with the incremental benefit estimated by MOBILE3, the
adjusted benefit from the tables is divided by the composite
emission factor used to estimate the inventory and then
multiplied by the inventory in tons. The result is an
approximation of the incremental benefit in tons. A better
estimate is provided by using M0BILE3 using the method outlined
above.
(1.80/25.00) * 1000 tons/day = 72 tons/day.
Benefits in tons can also be estimated by multiplying the
adjusted mobile source composite benefit (200 mg/mi) by an
average VMT figure. The VMT figure must be for all highway
vehicles in the appropriate geographic area for the appropriate
time interval. The VMT figure must also be consistent with the
VMT figures used to prepare other portions of the SIP inventory.
Again, if there are significant differences between local
conditions, tampering rates and vehicles and the national
averages and assumptions used in this report, then the tables
provided will not be useful to accurately estimate the benefits
of tampering inspection programs. In these cases, EPA should be
consulted either to provide the emission reduction factors for
use in MOBILE3 or to make the estimates.
-------
-115-
References
1. Motor Vehicle Tampering Surveys. National Enforcement
Investigation Center, Denver, Colorado for EPA Field
Operations and Support Division of the Office of Mobile
Sources. 1982 Survey. 1981 Survey, March 1982
(EPA-330/1-82-001). 1979 Survey, May 1980
(EPA-330/1-80-001). 1978 Survey, November 1978.
2. "Evaluation of the Applicability of a Lead-Sensitive Test
Paper as a Diagnostic Tool for Detecting Habitual
Misfueling of Catalyst-Equipped Motor Vehicles."
Technical Report. Bill Smuda, U.S. EPA, I/M Staff.
July 1980.
3. "Assessment of Current and Projected Future Trends in
Light-Duty Vehicle Fuel Switching," Energy and
Environmental Analysis, Inc., Arlington, Virginia. June
1982.
4. "1968-1982 Automotive Emission Systems Application
Guide," Department of Industrial Sciences, Colorado State
University, Ft. Collins, Colorado.
5. "Compilation of Air Pollutant Emission Factors: Highway
Mobile Sources," U.S. EPA Emission Control Technology
Division, March 1981. (EPA460/3-81-005).
6. 1981 Wards Automotive Yearbook, Ward's Communications,
Inc. Detroit, Michigan. Library of Congress Number
40-33639.
-------
APPENDIX
Disablement Data Base
(EPA-AA-TSS-83-10)
Table
A-l:
Air Pump (Oxidation Catalyst)
Table
A-2:
Air Pump {Three-way Catalyst)
Table
A-3:
Catalyst (Oxidation Catalyst)
Table
A-4:
Catalyst (Three-way Catalyst)
Table
A-5 :
Misfueling (Oxidation Catalyst)
Table
A-6:
Misfueling (Three-way Catalyst)
Table
A-7:
EGR (pre-Catalyst)
Table
A-8:
EGR (3.1 Std.) Oxidation Catalyst
Table
A -9:
EGR (2.0 Std.) Oxidation Catalyst
Table
A-10:
EGR (1.0 Std.) Three-Way Catalyst
Table
A-ll:
Oxygen Sensor
Table
A-l 2;
Evaporative Canister
References
-------
Table A-l
Air Pump Disablement Testing
of Oxidation Catalyst Vehicles
Vehicle
Model
FTP
(gm/mi)
Site*
Number
Year
Manufacturer
Test
Condition
HC
CO
Chicago
6034
1976
Ford
4
Ok
0.61
2.72
7
Di sabled
2.72
34.74
Chicago
5035
1975
Ford
2
Ok
0.75
4 .43
7
Disabled
2.23
46.54
Chicago
6041
1976
Ford
1
Ok
0.71
1.44
7
Di sabled
2.03
21.30
Chicago
6050
1976
Ford
4
Ok
1.31
8.66
9
Di sabled
1.63
16.67
Chicago
6062
1976
Ford
4
Ok
0.67
4.44
6
Di sabled
0.94
19.89
Washington
6052
1976
Ford
4
Ok
0.64
11.05
7
Di sabled
0.69
15.25
Washi ngton
6060
1976
Ford
4
Ok
1.09
2 .99
7
Di sabled
2.72
30.77
Washington
6073
1976
GM
4
Ok
0.55
7.74
8
Di sabled
0.87
13.66
San Francisco
6003
1976
Chrysler
1
Ok
0.68
8.65
8
Di sabled
4.89
83.18
San Francisco
6010
1976
Chrysler
1
Ok
0 .46
5.83
7
Di sabled
2.37
40.3 5
San Francisco
6012
1976
Chrysler
2
Ok
0 .48
7.27
9
Di sabled
2.90
78. 17
San Francisco
6036
1976
Ford
2
Ok
0.70
3.83
7
Di sabled
1.99
41.52
San Francisco
6039
1976
Ford
1
Ok
0.45
1.52
9
Di sabled
0.99
26.42
Averages
Ok
0.70
5.43
Di sabled
2.07
36.04
Average Excess
1.37
30.61
~Vehicles from Restorative Maintenance (RM76) Programs (see References).
-------
Table A-2
Air Pump Disablement Testing
of Three-Way Catalyst Vehicles
Model Vehicle ID FTP (gm/mi)
Program*
Year
Manufacturer
(if a ny)
Condition
HC
CO
LA80
1980
Ford
#46
Ok
Di sabled
0.24
0.90
1.41
24.56
MVEL
1979
Ford
-
Ok
Di sabled
0.24
0.69
1.33
10.70
MVEL
1979
Chrysler
-
Ok
Di sabled
0.41
1.03
4.10
29.80
GM
(Prototype)
GM
-
Ok
Di sabled
0.23
0.52
1.40
8.32
Averages
Ok
Disabled
0.28
0.79
2 .06
18.3 5
Average Excess
0. 51
16.29
~Programs listed in references.
-------
Table A-3
Catalyst Removal Disablement
Testing of Oxidation Catalyst Vehicles
Vehicle Model FTP (gm/mi)
Program*
Number
Year
Manufacturer
Condition
HC
CO
STL
9401
1979
Ford
Ok
Bypassed
0.67
3 .48
6.77
32.50
STL
9402
1979
Ford
Ok
Bypassed
0.91
3.54
3 .67
13.10
STL
9403
1979
GM
Ok
Bypassed
0.54
1.99
3.57
7 .20
STL
9404
1979
GM
Ok
Bypassed
0.87
2.55
12 .47
14.35
Averages
Ok
Bypassed
0.75
2 .89
6.62
16.79
Average Excess
2 .14
10 .17
~Programs listed in references.
Excess Calculation:
((2.89/0.75)-1.0)* 1.07 = 3.05 gm/mi HC
((16.79/6.62)-1.0)* 18.23 = 28.01 gm/mi CO
This calculation increases the MOBILE3 estimate
for oxidation catalyst technology vehicles by the
emissions from catalyst removal to give the excess
report.
of zero-mile emissions
percentage increase in
emissions used in this
-------
Data Point Selection Criteria
1. Vehicles which were designed to California standards
were considered acceptable for estimating the effects of
disablement on similar federal certified vehicles.
2. Whenever possible, only testing of 1981 and later
closed-loop vehicles was used to estimate the effects of
disablements on closed-loop vehicles. Existing data on
prototype vehicles were not used, except in the case of air
pump disablements.
3. Both deliberate disablement testing program results and
repair of in-use disablements were considered acceptable for
estimating the effects of disablements. Multiple repair or
disablement results were not used.
4. Whenever more than one test was conducted in a
particular mode, the average of all tests was used to
establish the emission levels of that mode.
5. Engine-out emissions and
catalyst bypass were considered
effects of catalyst removal.
emissions measured with a
acceptable for estimating the
6. The last test run using leaded fuel was used to
establish the misfueled case for emission levels.
7. When data sample size allowed, only deliberate
disablement results were used to determine the excess
emissions of a particular disablement.
8. When possible, all data were taken from computer stored
files. In some cases these data varied slightly from
published reports.
-------
Table A-4
Catalyst Removal Disablement Testing
of Three-Way Catalyst Equipped Vehicles
Vehicle
Model
FTP (gm/mi)
Program*
Number
Year
Manufacturer
Condition
HC
CO
NOx
OHIO
1
1981
GM
Ok
Removed
0.42
2 .88
2.76
14.08
0.79
2.59
OHIO
2
1981
Ford
Ok
Removed
0.54
2 .20
10.55
55.49
0.63
1 .49
OHIO
4
1981
VW
Ok
Removed
0.35
1.70
2.01
10.67
0.98
3.45
OHIO
5
1982
GM
Ok
Removed
0.13
1.39
2 .09
8.35
0.43
3.96
Averages
Ok
Removed
0.36
2.04
4.35
22 .15
0. 71
2.87
Average
Excess
1.68
17.80
2.16
~Programs 1i sted in references.
-------
Table a-5
Habitual Misfueling Testing
of Oxidation Catalyst Vehicles
Vehicle Model FTP (gm/mi)
Program*
Number
Year
Manufacturer Condition
HC
CO
STL
9401
1979
Ford
Ok
Mi sfueled
0.67
3 .20
6.77
21.70
STL
9402
1979
Ford
Ok
Misfueled
0.91
3 .04
3 .67
13 .00
STL
9403
1979
GM
Ok
Mi sfueled
0.54
1.61
3 .57
6.90
STL
9404
1979
GM
Ok
Misfueled
0.87
2.06
12 .47
15.30
Averages
Ok
Misfueled
Average Excess
0.75
2 .48
1.73
6.62
14.2 3
7.61
~Programs listed in references.
Excess Calculation:
((2.48/0.75)-1.0)* 1.07 = 2.47 gm/mi HC
((14.23/6.62)-1.0)* 18.23 = 20.96 gm/mi CO
This calculation increases the MOBILE3 estimate of zero-mile emissions
for oxidation catalyst technology vehicles by the percentage increase in
emissions due to habitual misfueling to give the excess emissions use in
this report.
-------
Table A-6
Habitual Misfueling
of Three-Way Catalyst
Te s t i ng
Vehi cles
Vehicle
Model
FTP (gm/m
i)
Program*
Number
Year
Manufacturer Condition
HC
CO
NOx
FOSD
1
1982
GM
Ok
Mi sfueled
0.28
2.81
3 .36
23.64
0.47
1.02
FOSD
2
1982
Toyota
Ok
Mi sfueled
0.22
1.70
3 .12
11.76
0.39
1.78
FOSD
3
1983
Ford
Ok
Misfueled
0.24
1.90
3.60
10.00
0.61
1.49
OHIO
1
1981
GM
Ok
Mi sfueled
0.42
2. 18
2 . 76
11.20
0.79
1.21
OHIO
2
1981
Ford
Ok
Mi sfueled
0.54
2. 14
10.55
39.64
0.63
0.82
OHIO
4
1981
VW
Ok
Mi sfueled
0.35
1.40
2 .01
2 .24
0.98
1.98
OHIO
5
1982
GM
Ok
Mi sfueled
0.13
1.05
2 .09
8. 12
0.43
1.27
Averages
Ok
Mi sfueled
0.31
1.88
3.93
15.23
0.61
1.37
Average Excess
1.57
11.30
0.76
~Programs listed in references.
-------
Table A-7
EGR Disablements Among Pre-1975
Model Year Non-Catalyst Vehicles
FTP NOx (gm/mi)
Vehicle
Model
EGR
Condition
Number*
Year
Manufacturer
OK
Di sabled
Excess
5030
1973
Ford
1.40
2.65
1.25
5032
1973
GM
1.86
3 .03
1.17
Average NOx
1.63
2.84
1.21
*Both vehicles were
See references.
repaired as
part of
Task 5 in
Portland
-------
Table A-8
EGR Disablements Among 1975-1976
Model Year Oxidation Catalyst
Vehicles Built to a 3.1 gm/mi NOx Standard
FTP NOx (gm/mi)
Vehicle
Model
EGR
Condition
Site*
Number
Year
Manufacturer
Test
OK
Disabled
Excess
Chicago
6004
1976
Chrysler
4/8
1.88
6.10
4.22
Chicago
6013
1976
Chrysler
3/7
2.93
4.42
1.49
Chicago
6014
1976
Chrysler
3/6
2.45
5.04
2.59
Chicago
5026
1975
Chrysler
4/7
2.98
4.77
1.79
Chicago
6033
1976
Chrysler
4/7
2.65
3.90
1.25
Chicago
6034
1976
Ford
4/6
2.93
4.25
1.32
Chicago
5035
1975
Ford
2/6
2.62
4.06
1.44
Chicago
6041
1976
Ford
1/6
2.16
5.20
3.04
Chicago
5047
1975
Ford
4/6
2.89
4.98
2.09
Chicago
6050
1976
Ford
4/6
3 .34
7.21
3.87
Chicago
6059
1976
Ford
2/6
1.88
7.21
5.33
Chicago
6062
1976
Ford
4/9
2.79
8. 11
5.32
Chicago
6068
1976
GM
1/8
2.62
7.45
4.83
Chicago
6071
1976
GM
2/8
3.09
7.62
4.53
Chicago
6081
1976
GM
1/8
2 .43
4.87
2.44
Chicago
6083
1976
GM
1/8
2.42
3.71
1.29
Chicago
6094
1976
GM
2/8
2 .67
2.64
-0.05
Chicago
6099
1976
GM
2/8
2.96
6.19
3.23
Wash., D.C.
6011
1976
Chrysler
4/9
2 .94
10.99
8.05
Wash., D.C.
6029
1976
Chrysler
4/7
2.66
7.42
4.76
Wash., D.C.
6031
1976
Chrysler
4/6
3 .01
5.07
2.06
Wash., D.C.
6049
1976
Ford
4/8
2.99
6.38
3.39
Wash., D.C.
6052
1976
Ford
4/6
1.82
7.10
5 .28
Wash., D.C.
6060
1976
Ford
4/8
3 .07
7.66
4.59
Wash., D.C.
6073
1976
GM
4/7
2 .39
5.50
3.11
Wash., D.C.
6086
1976
GM
4/9
2.18
3.84
1.66
Wash., D.C.
6096
1976
GM
3/8
2.65
3.68
1.03
Detroi t
6002
1976
Chrysler
3/6
2.95
9.82
6.87
Detroit
6021
1976
Chrysler
4/6
1.83
5 .69
3.86
Detroi t
6024
1976
Chrysler
2/6
3 .09
4.09
1.00
Detroit
6025
1976
Chrysler
3/6
3.05
4.45
1.40
Detroit
6035
1976
Ford
1/7
2.20
4.30
2.10
Detroit
6040
1976
Ford
1/7
1.69
5.83
4.14
Detroi t
6046
1976
Ford
4/7
1.71
6.43
4.72
Detroit
6064
1976
Ford
2/7
1.37
5.47
4.10
Detroit
6066
1976
Ford
1/7
2.07
6.72
4.65
Detroit
6083
1976
GM
1/7
1.91
2.85
0.94
Detroit
6090
1976
GM
1/7
2.16
7.68
5.52
Detroit
6091
1976
GM
2/7
2.03
7.75
5.72
Average NOx
2.50
5.81
3.31
~All vehicles deliberately disabled as part of the FY76 Restorative
Maintenance Program. See References.
-------
Table A-9
EGR Disablements Among Oxidation Catalyst
Vehicles Built to a 2.0 gm/mi NOx Standard
FTP NOx (gm/mi)
Vehicle
Model
EGR
Condition
Proqram*
Site
Number
Year
Manufacturer
OK
Disabled
Excess
DRIVE
NA
1
1977
GM
1.80
4.92
3.12
DRIVE
NA
2
1977
GM
2.72
3.71
0.99
DRIVE
NA
4
1978
GM
1.66
5.21
3.55
DRIVE
NA
5
1977
GM
1.66
4.94
3.28
DRIVE
NA
6
1978
GM
1.26
4.12
2.86
DRIVE
NA
7
1977
GM
1.67
5.11
3.44
DRIVE
NA
8
1977
GM
2.00
7.60
5.60
DRIVE
NA
9
1977
GM
2.56
8.30
5.74
DRIVE
NA
10
1977
Ford
1.45
4.14
2.69
DRIVE
NA
11
1977
Ford
1.45
7.68
6.23
DRIVE
NA
13
1977
Ford
2.03
7.15
5.12
DRIVE
NA
15
1977
Ford
2.42
8.42
6.00
DRIVE
NA
16
1977
Ford
1.80
6.34
4.54
DRIVE
NA
17
1977
Chrysler
1.43
2.65
1.22
DRIVE
NA
20
1978
AMC
1.11
4.23
3.12
RM76
SF
6003
1976
Chrysler
1.42
5.18
3.76
RM76
SF
6010
1976
Chrysler
1.41
4.22
2.81
RM76
SF
6012
1976
Chrysler
1.17
5.59
4.42
RM76
SF
6023
1976
Chrysler
1.59
4.54
2.95
RM76
SF
6036
1976
Ford
1.50
3.72
2.22
RM76
SF
6039
1976
Ford
2.02
5.48
3.46
RM76
SF
6065
1976
Ford
1.96
5.27
3.31
RM76
SF
6123
1976
Chrysler
1.53
3.58
2.05
REG
NA
10
1978
GM
1.26
3. 54
2.28
REG
NA
12
1978
GM
1.20
1.79
0.59
REG
NA
15
1978
Ford
1.61
6.68
5.07
Average NOx
1.68
5.16
3.48
*See References.
-------
Table A-10
EGR Disablements Among Three-Way Catalyst
Vehicles Built to a 1.0 qm/mi NOx Standard
FTP NOx (qm/mi)
Vehicle
Model
EGR Condition
Proqram*
Site
Number
Year
Manufacturer
OK
Disabled
Excess
FY80
LA
344
1981
GM
0.37
1.68
1.31
FY80
SA
19
1981
GM
0.63
1.19
0.56
FY81
SA
617
1981
VW
2.45
3.27
0.82
FY82
ARB
8197
1981
Ford
1.37
2.38
1.01
FY82
ARB
8314
1981
GM
1.06
2.81
1.75
CL82**
ARB
705
1981
Ford
0.88
1.92
1.04
MVEL**
ARB
-
1982
GM
0.62
2.74
2.12
Average NOx
1.05
2.28
1.23
*See references.
~~Deliberate disablements.
-------
Table A-ll
Oxygen sensor Disablement Testing
Vehicle
Model
FTP (gm/mi)
Program*
Number
Year
Manufacturer
Condition
HC
CO
NOX
CL82
701
1981
AMC
Ok
Disabled
0 .25
0 .27
2.05
2 .57
0 .96
0 .88
CL82
702
1981
Chrysler
Ok
Disabled
0 .24
0 .25
1.72
3 .28
1.12
0 .64
CL82
703
1981
Chrysler
Ok
Disabled
0 .50
0 . 87
9.21
32.90
0.80
0.31
CL82
704
1981
Ford
Ok
Disabled
0.63
0.74
6.58
9 .95
1.24
0 .41
CL82
705
1981
Ford
Ok
Disabled
0 .30
0 .27
3 .40
3 .71
0 .88
1.26
CL82
707
1982
GM
Ok
Disabled
0.09
1.58
1.35
59.71
0.50
0 .39
CL82
708
1982
GM
Ok
Disabled**
0 .15
3 .19
3.56
157.66
0.54
0 .22
CL82
711
1981
Ford
Ok
Disabled
0.31
0.25
2.00
1.62
0.89
1.59
CL82
713
1981
VW
Ok
Disabled
0.11
2.41
1.38
109.59
0.32
0 .07
CL82
714
1981
Toyota
Ok
Disabled
0 .10
1.68
0.96
32.90
0.58
0 .20
CL82
715
1982
GM
Ok
Disabled**
0 .14
4 .17
4.03
175 .08
0 .37
0 .11
CL82
716
1981
Chrysler
Ok
Disabled
0 .72
1.52
9.73
31.25
1.00
0 .37
-------
Table a-11 (cont'd)
Oxygen Sensor Disablement Testing
Vehicle Model FTP (gm/mi)
Program*
Number
Year
Manufacturer
Condition
HC
CO
NOx
CL82
718
1981
Ford
Ok
Disabled
0 .97
1.00
8.12
14.67
0.67
0.45
CL82
720
1981
GM
Ok
Disabled
0 .40
0.31
1.34
1.46
0.99
3 .02
CL82
721
1981
GM
Ok
Disabled
0 .26
0.31
2.74
2 .88
0.79
1.83
CL82
722
1981
VW
Ok
Disabled**
0 .33
10 .70
5.17
186 .06
1.30
0 .11
MVEL
-
1982
GM
Ok
Disabled
0 .13
0.42
1.85
11.63
0.62
0 .46
MVEL
-
1982
GM
Ok
Disabled**
0 .13
4 .83
1.85
171.99
0.62
0 .29
POSD
1
1982
GM
Ok
Disabled
0 .28
0 .24
3.36
2.19
0.47
1.82
FOSD
2
1982
Toyota
Ok
Disabled
0.21
0 .33
2.29
7.09
0.37
0 .75
FOSD
3
1983
Ford***
Ok
Disabled
0 .24
0.23
3 .60
4.70
0.61
0.51
Averages
Ok
Disabled
0.31
1.69
3.63
48.71
0.74
0 .75
Average Excess
1.38
45.08
0.01
* Programs listed in references.
** Oxygen Sensor grounded.
***Light-Duty Truck.
-------
Table A-12
Model
Year
1981
Evaporative Control
System Disablement Testing
Manufacturer
GM
Model
Cutlass
Condition
Ok
Disabled
Evaporative HC
Diurnal Hot Soak
(gm/day) (qm/trip)
0.67
18.81
0.97
7.34
1981
GM
Malibu
Ok
Disabled
2.24
19.51
12.70
12.73
Average Disabled
19.16
10.04
Trips per day = 3.05
Miles per day = 31.10
((3.05 * 10.04) + 19.16)/31.10 = 1.60 gm/mi.
-------
Appendix References
1. "Evaluation of Applicability of Inspection/Maintenance
Tests on a Dodge Aspen Prototype," (MVEL) Thomas J.
Penninga, TAEB, U.S. EPA, August 1979,
(EPA-AA-TEB-79-11).
2. Evaluation of Applicability of Inspection/Maintenance
Tests on a Ford EEC-II Prototype," (MVEL), Thomas J.
Penninga, TAEB, U.S. EPA, June 1979, (EPA-AA-TEB-79-9) .
3. "Testing of New Technology Three-Way Catalyst Equipped
Vehicles in Los Angeles," (LA80), Automotive
Environmental Systems, California, August 1980,
(EPA-460/3-80-27) .
4. Letter from T.M. Fisher, Director, Automotive Emission
Control, General Motors Corporation to Charles L. Gray,
Director, Emission Control Technology Division, U.S.
EPA, (GM), Test results of prototype C-4 emission
control system equipped on a Chevrolet Caprice with and
without the air pump disabled. Dated March 27, 1979.
5. "Evaluation of Restorative Maintenance on 1975 and 1976
Light-Duty Vehicles in Chicago, Illinois," Automotive
Testing Laboratories, Colorado, January 1977,
(EPA-460/3-76-030).
6. "Evaluation of Restorative Maintenance on 1975 and 1976
Liqht-Duty Vehicles in Washington, D.C.," General
Environments Corporation, Virgina, March 1977,
(EPA-460/3-76-031).
7. "Evaluation of Restorative Maintenance on 1975 and 1976
Light-Duty Vehicles in Detroit, Michigan," Olson
Laboratories Inc., Michigan, May 1977,
(EPA-460/3-76-032).
8. "Evaluation of Restorative Maintenance on 1975 and 1976
Light-Duty Vehicles in San Francisco, California,"
Automotive Environmental Systems, California, October
1977, (EPA-460/3-76-033).
9. "The Emission Effects of Misfueling Five 1981-82 Model
Year Automobiles with 10 Continuous Tankfuls of Leaded
Gasoline," (OHIO), R. Bruce Michael, TSS, U.S. EPA,
August 1983, (EPA-AA-TSS-83-2).
10. Projects A-ll and A-14, "Tampering Program Conducted at
MVEL," (FOSD), Larry Oeler, MOD, U.S. EPA, (Unpublished).
-------
Appendix References (cont'd)
11. "Effects of Lead Poisoning on Catalyst Efficiency:
Project 2R7904," California Air Resources Board (CARB),
July 1979, (MS-79-028).
12. "A Study of the Effects of Fuel Switching on Catalyst
Equipped Vehicles," (STL), Automotive Testing
Laboratories, Colorado, (EPA-460/3-80-23).
13. "Catalyst Poisoning and Catalyst Recovery Due to
Misfueling," (CALIF), California Air Resources Board,
Tasks #2 and #3, (EPA-460/3-80-22).
14. "Disablement Testing of 1981-1982 Model Year Vehicles
with Closed-Loop Emission Control Systems," (CL82), R.
Bruce Michael, TSS, U.S. EPA, September 1982,
(EPA-AA-TSS-82-5).
15. Memo from Thomas Penninga, TEB, to Charles L. Gray,
Director, ECTD, (EVAP). Test results from evaporative
testing of two 1981 vehicles with and without the
evaporative control system disconnected. Dated May 10,
1983.
16. "Testing Support for Evaluation of
Inspection/Maintenance Issues," Test Group #5, (Task 5),
Hamilton Test Systems, Arizona, (EPA-460/3-80-016).
17. "Light-Duty Vehicle Driveability Investigation,"
(DRIVE), Suntech Group, Pennsylvania, (EPA-460/3-78-012).
18. "Regulated and Unregulated Exhaust Emissions from
Malfunctioning Non-Catalyst and Oxidation Catalyst
Gasoline Automobiles," (REG), Southwest Research
Institute, Texas, November 1978, (EPA-460/3-80-003).
19. "A Study of Emissions from Light-Duty Vehicles in Los
Angeles," Automotive Environmental Systems, California,
(FY80), September 1981, (EPA-460/3-81-018).
20. "A Study of Emissions from Light-Dutv Vehicles in San
Antonio," EG&G Automotive Research, Texas, (FY80),
September 1981, (EPA-460/3-81-019).
21. "A Study of Emissions from Light-Duty Vehicles in San
Antonio," EG&G Automotive Research, Texas, (FY81) , Julv
1982, (EPA-460/3-82-009).
-------
Appendix References (cont'd)
22. "Evaluation of Applicability of Inspection/Maintenance
Tests on a Chevrolet Citation," Thomas J. Penninga,
TAEB, U.S. EPA, (MVEL), November 1979,
(EPA-AA-TAEB-80-6).
23. "Evaluation of Applicability of Inspection/Maintenance
Tests on a Toyota Celica Supra," Bill Smuda, IMS, U.S.
EPA, (MVEL), December 1980, (EPA-AA-IMS-80-9).
24. Memo from Thomas Penninga, TEB, to Bruce Michael, IMS
(MVEL). "Phoenix I/M Disablement Testing Summary,"
Disablement testing on a throttle body injection GM
x-body. Dated May 4, 1982.
------- |