EPA-AA-IMS-81-11
Technical Report
A Study of the Effectiveness of Mechanic Training
For Vehicle Emissions Inspection and Maintenance Programs
R. Bruce Michael
April, 1981
Notice
Technical Reports do not necessarily represent final EPA decisions or
positions. They are intended to present technical analysis of issues using
data which are currently available. The purpose in the release of such
reports is to facilitate the exchange of technical information and to inform
the public of technical developments which may form the basis for a final EPA
decision, position or regulatory action.
Inspection and Maintenance Staff
Emission Control Technology Division
Office of Air, Noise and Radiation
U.S. Environmental Protection Agency
Ann Arbor, Michigan
-------
2
TABLE OF CONTENTS
/
Page
1.0 SUMMARY AND CONCLUSIONS . 1
2.0 BACKGROUND 2
3.0 TRAINING COURSE USED 3
4.0 FACILITY TRAINING 3
5.0 STUDY DESIGN 5
5.1 Vehicle Selection and Recruitment 5
5.2 Vehicle Testing 6
5.3 Repairs by Participating Facilities 7
6.0 RESULTS 7
6.1 Emissions and Fuel Economy 7
6.2 Cost and Amount of Repairs by Commercial 15
Facilities
6.3 Quality of Repairs 16
6.4 Repair Effectiveness by Facility Type 24
7.0 DISCUSSION AND CONSLUSIONS: POSSIBLE REASONS FOR THE
OBSERVATIONS IN PORTLAND AND GENERAL IMPLICATIONS FOR
MECHANIC TRAINING BENEFITS
7.1 Possible expectations of training, actual 25
results and possible explanations
7.2 General implications for mechanic training 26
benefits and limitations of this study
8.0 MECHANIC TRAINING EMISSION REDUCTION BENEFITS
CONTAINED IN MOBILE2
APPENDIX - Vehicle Test Fleet 33
-------
1.0 SUMMARY .
This report describes a study to determine if a short, practical training
course for emissions repairs has a supplementary emission reduction and fuel
economy benefit to an Inspection and Maintenance programi The study was
performed in the early part of 1980 in Portland, Oregon.
Two matched samples of 21 vehicles each were sent to ten repair facilities
before and after the facilities received training. The experiment was blind
in that the ten repair facilities did not know that the study was in progress
or that the 42 vehicles were in any way connected with EPA. The training
course used was a 16 hour course in repairing vehicles which fail an idle
emissions test.
Vehicles were tested before and after repairs on the Federal Test Procedure
and various short tests for emissions and fuel economy. Contractor laboratory
personnel then attempted to restore the vehicles to their specifications and
vehicles were retested.
Results indicate that:
1. Training in Portland did not produce greater emission reduction
benefits from repairs than the emission reduction benefits without
training. Similar emission reductions of approximately 41% HC and 53% CO
on the Federal Test Procedure occurred from repairs both before and after
training.
2. A 0.8% improvement in fuel economy from repairs (average of city and
highway driving) was seen after training. This improvement was not
statistically significant, however.
3. The average cost of repairs was nearly identical before and after
training, about $19.50.
4. Several measures of the quality of repairs indicate a slight improve-
ment with training. The quality of carburetor adjustments, which is
highly important, did not improve with training, but spark timing adjust-
ments and driveability improved slightly.
5. Emission reductions from repairs performed by different types of
facilities (gas stations, independent garages and chain stores) did not
differ significantly in emission reductions either before or after
training.
6. Contractor repairs to the vehicles show that the potential exists for
significant further emissions reductions beyond the reductions achieved by
commercial facility repairs. This potential may not be realizeable in
practice. Significant potential for further fuel economy improvement also
exists.
-------
There are two major reasons why this study may not have shown significant
emission reduction benefits from training. First, the Portland idle emission
standards are more stringent than almost any other I/M program will adopt,
thus there is not as much "room" for improvement that training could
accomplish in Portland, compared to other I/M programs in which the mechanics
do not have to adjust the vehicles' emissions.so low. For example, the idle
CO standard for most late model vehicles in Portland is 1.0% (including the
allowed tolerance) and the average after repair idle CO level is 0.1%; in New
Jersey the standard for those vehicles is 3.0% and the average after repair
idle CO level is 0.8%. Second, the I/M program had been operational in
Portland for over four years prior to the study. It is expected that during
that time mechanics would have gained some on-the-job training and "grapevine"
information which would lessen the effect of a training course. Also, the
mechanics would have formed habits for repairing the emission failure vehicles
which a short training course could not greatly change; for example, many
facilities in Portland always perform, for a set fee, a "DEQ adjustment",
meaning standard adjustments to pass the state idle emissions test.
Additional discussion and conclusions are contained in Section 7.0.
2.0 BACKGROUND
Numerous studies in the past have indicated that mechanic training is a very
important part of an Inspection and Maintenance (I/M) program. In 1973 the
National Academy of Sciences[1]* stated its concern that the service industry
may not be able to adequately service cars from an emission control stand-
point. Since that time many other studies have agreed that mechanic training
is essential to an I/M program in order to achieve the full benefits, both in
terms of emissions reduction and fuel economy improvement. This need can be
seen by observing the effort the auto manufacturers put into training dealer-
ship mechanics. A service training manager from one of the major manufactur-
ers summarized the general need: "Engineers continually make refinements and
improvements so technicians need continual refreshers to keep up with the
changes in existing technology; it's a never-ending process."(2]
Twenty-nine states will have operating I/M programs in 1987. States will
receive "credits", representing expected emission reductions due to I/M, which
they will use in 1982 to demonstrate compliance with the National Ambient Air
Quality Standards. EPA estimated in 1978 that additional emission reductions
could be achieved from a mechanic training program; many states are planning
to institute training partly in order to claim more credits. Since training
is a relatively low cost strategy, it is very attractive. The 1978 EPA credit
estimates were based (in MOBILEl) on a mathematical model using emission
results from a large number of in-use vehicles. The credits associated with
mechanic training were based on the assumption that trained mechanics would be
able to repair failed I/M vehicles to the Federal standards for which they
were designed. This assumption had not been tested in a field experiment.
Therefore, in 1979 EPA made plans for testing the effectiveness of a training
course on emissions in an area which currently had I/M, but whose mechanics
had little or no formal training in the area of emissions. Results from that
study were to be used in a revision of MOBILEl, called MOBILE2.
* Numbers in brackets designate references at the end of the report.
-------
The Portland, Oregon area was chosen as the site to test the effectiveness of
training. This was a logical choice for EPA since the area had an established
I/M program with very little previous formal training, and EPA already had a
contractor-operated emissions testing laboratory present. The contractor had
experience in recruiting vehicles and testing them using a variety of tests
including the Federal Test Procedure (FTP), the same procedure EPA uses to
certify new cars.
3.0 TRAINING COURSE USED
In order for emissions training to be widespread and accepted by a large
number of auto technicians, it would have to be relatively short in length.
EPA had a short, practical course designed by the National Center for Vehicle
Emissions Control and Safety of Colorado State University (CSU) over the
1978-79 period. Design for this course began more than one year prior to its
use in Portland and it had already been field tested, although CSU felt that
minor refinements may still have been needed.
The course was designed to be a maximum of 16 hours in length, which could,
for example, be taught over several evening sessions. In this way mechanics
would not have to take leave from work, a design which was considered
essential to the course becoming widespread.
The course mainly teaches the proper diagnosis of I/M failures; in other
words, hov the mechanic should proceed in order to repair a vehicle which has
failed the I/M idle test. The titles of each of the six units taught in the
course follow.
Unit Topics
1. I/M Programs and Vehicle Emissions; Short Tests.
2. Equipment Used (Infrared Analyzer).
3. Types of Emission Failures; General Troubleshooting
Information.
4. Correction Procedures for Excessive HC Emissions.
5. Correction Procedures for Excessive CO Emissions.
6. Proper Carburetor Adjustment Procedures.
4.0 FACILITY TRAINING
Five categories of repair facilities which perform work for emission repairs
were designated. The original design called for the recruitment of three of
each type, for a total of 15 facilities to be used in the study. Fifteen
facilities were recruited, but five later had to be dropped due to low or no
participation in the training, including the only new car dealership which had
been signed up. (Dealership mechanics were generally not agreeable to the
training, because they already attended many training sessions provided by the
manufacturers.) The facility categories and the number of facilities of each
that were actually used throughout the study are listed below.
-------
Facilities Used No.
1. Gasoline-service stations s 2
2. Independent repair garages 2
3. Chain stores (e.g., Wards, Goodyear) 2
4. New car dealerships 0
5. Facilities advertising I/M emissions 4
repairs (one was a gas station, the
other three independent garages).
10
Approximately 100 facilities were originally contacted. Facilities were told
that EPA was sponsoring a pilot offering in the Portland area of a short
course in emissions repair that had been developed by Colorado State
University (CSU). The course was offered at no charge to mechanics.
Incentives to the mechanics were free training, a certificate of completion,
and a monetary incentive of $50 for completing the course. They were also
told that we wanted to include only those facilities in which all tune-up
mechanics could take the course. (The reason for this was to assure that test
vehicles sent to the facilities would be repaired by participating
mechanics.) Although this was not strictly adhered to for one facility,
results from that facility were not used in the final analysis. All mechanics
in the 10 facilities listed above received the training.
Many facilities contacted were either not interested or could not get all the
tune-up mechanics there to take the training. The EPA Project Officer also
asked questions concerning previous training in order to identify facilities
which had an unusually high amount of training and would therefore be
unrepresentative of the norm. Nearly all facilities reported that their
mechanics had at least some previous training, but none were judged as being
outside the norm. The mechanics from the facilities used in the final
analysis had varied experience and prior training. The median amount of
tune-up experience was five years and the median amount of training was one
prior course. The range of experience was from one to twenty-nine years and
the range of the amount of courses taken was from none to seven (nearly half
of the mechanics had previously taken one prior course).
The mechanics were trained in February and March of 1980. Five classes at
three community colleges were held, averaging 7 mechanics per class. Four of
the classes were held in five evening sessions over two-week periods, and the
fifth class was given on two consecutive Saturdays in all-day meetings. All
but three of the mechanics who started attended all sessions. A total of 23
mechanics were trained from the 10 facilities used in the study. Their
average test scores on the same written test at the beginning and end of the
training were 71 and 90 respectively. CSU felt that a minimum final score of
80 should be attained by anyone who had successfully learned material in the
course. All of the mechanics achieved scores of at least 80 at the end of the
training.
-------
Prior to the training, a two-day workshop was given by CSU for the instructors
who would be teaching the classes. Orientation was given to familiarize the
teachers with the course and the CSU instructional materials and to give an
idea of how much time to spend on each section. EPA and CSU felt that this
was necessary, because the instructors had not seen the course material and at
that time there was also no mention of time to be spent on each section of the
course - since the course covers a lot of material over a relatively short
period of time, mention of the time to be spent on each section was more
important than with most courses. Because teaching the course would be a new
experience for the teachers, we felt that the workshop was necessary and would
not give them more knowledge or expertise than teachers in an actual I/M
mechanic training program would have after teaching the course once or twice.
Some experience in teaching the course would be the normal situation to
evaluate anyway, since it can be expected that in I/M areas which offer a lot
of training, each teacher would hold several classes.
5.0 STUDY DESIGN
5.1 VEHICLE SELECTION AND RECRUITMENT
Two groups of 1974-77 model year light duty passenger vehicles were used. The
Scope of Work called for the two groups to match each other by vehicle type
and emission failure type. Vehicle type includes model year, make, engine
size, fuel system (1, 2, or 4 venturi carburetor or fuel injection) and
transmission type. Emission failure type means the failure of HC, CO or both
HC and CO, on the state idle emissions test. The first group of vehicles was
sent to the facilities for repairs before training had occurred, the second
group after training. Each facility was sent the same types of vehicles after
training as had been sent before training.
The pre-training group was selected in order to be representative of a failing
fleet of vehicles. For the 1975-77 model years this was a simple matter.
Another EPA study in Portland had already obtained a representative sample of
failing vehicles which were used for this training study. Another approach
was needed for the 1974 model year cars. First, a group of vehicles
representing high sales was chosen. From this group were recruited enough
willing owners of vehicles such that the assumed failure rate of 50 percent
for that model year would yield the required number of failing vehicles, and
they would be representative of a failing fleet. The post-training group of
vehicles was recruited in order to match the first group.
The pre-training group consisted of 43 vehicles. Due to several factors
including four facilities dropping out of the training program and miscellan-
eous vehicle problems, only 21 were used in the final match-up. Matching
these vehicles exactly, in order to produce the post-training group, was
somewhat difficult. In total, 55 vehicles were recruited for the second
group. This number was large mainly for two reasons. First, EPA planned on
there being 31 vehicle pairs and was recruiting vehicles accordingly. Second,
matching exactly for emission failure type caused more vehicles to be
recruited than could be used in the study.
-------
8
One of the 21 pairs used was not an exact match for emission failure type, but
was considered acceptable. In that one case-the pre-training vehicle failed
the state idle test for CO only, and the postftraining vehicle failed for both
HC and CO. The HC level was not extremely high and no misfire was present,
however, so that maintenance to reduce the CO level was all that was required
in order to also reduce the HC level, the same type of maintenance the
pre-training vehicle needed. Nine of the 21 pairs did not match for model
year, but matched in everything else.
5.2 Vehicle Testing
Owners brought their vehicles to the EPA contractor test laboratory, completed
a questionnaire, and had no further involvement. Contractor personnel drove
the vehicles to the state test lane for inspection. If a vehicle failed the
state inspection test (SIT) for emissions, it entered the program. Otherwise
it was returned to the owner.
A series of tests was performed on each study vehicle under a minimum of three
conditions: as received, after maintenance by the commercial repair facility
and after restorative maintenance (RM) by the contractor. Each series of
tests consisted of the following:
0 Diagnostic Inspection
0 Federal Test Procedure
0 50 MPH Cruise Test
0 Highway Fuel Economy Test
0 Four-Mode Idle
0 Loaded Two-Mode
0 State Idle Test at Laboratory
Separate from the purpose of this study, tire pressures and spark timing at
curb idle conditions were recorded during one or more of the series of tests.
The purpose of the RM sequence was to return the vehicles to their specifica-
tions concerning tune-up parameters and emission controls, within reasonable
limits. The sequence was designed to follow diagnosis and repair information
that was given in the training course, so that the trained mechanics in the
study should have been able to perform them. No extremely expensive repairs
were made, but repairs were more extensive than mechanics might usually
perform. For example, no major engine work was performed even if diagnosed as
probably being needed, however, carburetor repairs and replacements were made
if diagnosed as being needed.
In some cases, more than one RM sequence was performed. For the final matched
list this occurred four times in the pre-training sample and once in the
post-training. All of the additional RM sequences in the pre-training sample
were performed due to a concern that carburetor idle mixture adjustments may
not have been performed according to the specifications of the manufacturers.
There had been a misunderstanding on the part of the contractor concerning
this adjustment, which was easily cleared up. The one instance of additional
repairs to a post-training vehicle was done because of driveability problems.
-------
The problem was not serious, but was noticeable, and was solved by further
adjustment and repairs. These additional repairs were acceptable to the study
design since repair procedures do take driveability into consideration.
5.3 Repairs by Participating Facilities
Contractor personnel, rather than the owners, drove the vehicles to the repair
facilities. This was done so that the best documentation of vehicle repairs
could be achieved, consistent instructions to mechanics could be given, and
laboratory scheduling would be smoothest. The drivers acted as if they were
the owners, simply requesting that the cars be repaired in order to pass the
state emissions test. The state inspection sheet which gives the failure
condition(s) was always in the car. No mention of a price limit was made.
However, if asked, the driver would instruct the mechanic to perform the least
expensive adjustments or repairs in order to simply make the car pass the
emissions test. Due to the low average repair costs seen for I/M repairs in
other studies, it is assumed that this low-cost instruction is commonly given
by owners.
6.0 RESULTS
Average emission changes due to repairs before and after training are an
obvious highly important result, but is not the only important result. Also
of importance in determining the degree of success of the training is the type
and quality of repairs, cost of repairs, fuel economy changes, the emission
reductions versus the potential reductions of the vehicles, individual
facility performances, and more. The motivation of the type and amount of
repairs must also be considered when determining the success of training.
What may appear as lack of success by the training program may actually be due
to a failure to change economic motivations, rather than a failure to impart
new skills. For example, if low cost is always the top priority, repairs
which are diagnosed as needed to restore the vehicle's emission control system
to full function, but not necessary in order to pass the state emissions test,
will not be done.
6.1 Emissions and Fuel Economy
Average emissions and fuel economies are presented for six categories in
Tables 1 and 2. The emissions are from the Federal Test Procedure (FTP)
driving cycle and from the idle emissions test (performed at the laboratory).
The fuel economies are from the FTP and the Highway Fuel Economy Test (HFET).
Table 1 is for the before-training group of vehicles, Table 2 is for the
after-training group. The FTP HC, CO and fuel economy levels are shown
graphically in Figures 1-3. The percent and absolute changes in FTP emissions
and fuel economies are presented in Tables 3 and 4. All changes are from the
as-received condition.
-------
As-Received
After Repairs
by Facilities
After Repairs
by Laboratory
10
Table 1
Before Training
Emissions and Fuel Economy Levels
(21 Vehicles)
Idle Emissions**
Federal Test Procedure (Using Garage-
Emissions (grams per mile) Type Analyzer)
HC CO NOx HC -(ppm) CO (%)
2.94
1.77
1.73
47.4
22.7
16.9
2.57 252
2.39 107
2.53 106
3.02
0.38
0.20
Fuel Economy***
(miles per gallon)
FTP HFET
15.15
15.16
20.95
20.59
15.82 21.69
Federal/State*
Standards
(Average for all
vehicles)
1.78
18.6
2.84 258
1.2
15
21
Table 2
After Training
Emissions and Fuel Economy Levels
(21 Vehicles)
Idle Emissions**
Federal Test Procedure (Using Garage-
Emissions (grams per mile) Type Analyzer)
HC CO NOx HC (ppm) CO (%)
As-Received 3.33
After Repairs 1.92
by Facilities
After Repairs . 1.64
by Laboratory
45.1 2.93 307 4.2
20.2 2.93 64 0.15
13.9 2.65 60 0.07
Fuel Economy***
(miles per gallon)
FTP HFET
14.91
15.23
20.58
20.32
15.41 21.03
* Federal emission standards are based on the 1975 FTP - the 1974 Federal emission
standards were converted in order to average them with the 1975-77 standards. The
fuel economies are based on the new car certification values; fuel economy stan-
dards for a few vehicles were estimated. "State standards" refers to the state
idle test cutpoints.
** From the second idle portion of the SIT taken at the contractor's laboratory.
*** The conclusion stated in Section 1.0 that training produced a 0.8% improvement
in fuel economy is based on changes in combined FTP (city) and HFET (highway) fuel
economy.
-------
11
u.o
3.5
-3.0
a.
2.5
DC
UJ
en
ac
cl.O
0.5
0.0
Figure 1
HTDROCflRBONS
MEflN LEVELS BEFORE flND flFTER TRfllNING
(21 Vehicles)
2.9^
3.33
fl/R fl/M R/M
BEFORE TRAINING
fl/R fl/M R/M
flFTER TRfllNING
fl/R - flS RECEIVED
fl/M - flFTER MflINT,
R/M - RESTORflTIVE
-------
12
60
50
Figure 2
CflRBON MONOXIDE
HERN VALUES BEFORE RNO RFTER TRfllNING
(21 Vehicles)
= 30
oc
IU
ti-
er, 20
ac
ec
cc
O
10
47.
22.7
16.9
US. 1
fl/R fl/M R/M
BEFORE TRAINING
fl/R fl/M R/M
flFTER TRfllNING
fl/R - flS RECEIVED
fl/M - flFTER MfllNT.
R/M - RESTORflTIVE
-------
13
21.0
Figure 3
CITY FUEL ECONOMY
HERN VflLUES BEFORE flND flFTER TRfUNING
(21 Vehicles)
18.0 .
15.0 ;
£12.0
t-
u.
o 9.0
e_
6.0
3.0
0.0
15.1 15.2
15.8
14.
fl/R fl/M R/H fl/R fl/M R/H
BEFORE TRfllNING flFTER TRRINING
fl/R - RS RECEIVED
fl/M - flFTER MfllNT.
R/M - RESTORflTIVE
-------
BEFORE TRAINING
After Repairs
by Facilities
After Repairs
by Laboratory
AFTER TRAINING
After Repairs
by Facilities
After Repairs
by Laboratory
BEFORE TRAINING
After Repairs
by Facilities
After Repairs
by Laboratory
AFTER TRAINING
After Repairs
by Facilities
After Repairs
by Laboratory
14
Table 3
Percent Changes In Emissions and
Fuel Economy From Repairs*.
(21 Vehicles)
Federal Test Procedure
HC
CO
NOx
Fuel Economy**
FTP HFET
-40% -52%
-41%
-2%
+0.1%
+4.4%
-1.7%
+3.5%
-42% -55%
-51% -69%
-10%
+2.1%
+3.4
-1.3%
+2.2%
Table 4
Absolute Changes In Emissions and
Fuel Economy From Repairs*
(21 Vehicles)
Federal Test Procedure
(grams per mile)
HC
CO
NOx
-1.17 -24.7 -0.18
-1.21 -30.5 -0.04
Fuel Economy**
(miles per gallon)
FTP HFET
+0.01 -0.36
+0.67 +0.74
-1.41 -24.9 0.00
-1.69 -31.2 -0.28
+0.32 -0.26
+0.50 +0.45
* All changes are relative to the as-received levels.
** The conclusion stated in Section 1.0 that training produced a 0.8% improve-
ment in fuel economy is based on changes in combined FTP (city) and HFET
(highway) fuel economy.
-------
15
Several statistical tests were used to analyze the data in Tables 1-4. The
standard two-sample t-test and the paired t-test are the two tests most
appropriate (the latter pairs the matching vehicles of the two groups before
comparison of the groups). These tests all indicate that there is no
significant difference (at even the 90% confidence level) between the emission
and fuel economy levels of the before and after-training groups at any of the
repair stages. In other words, the two groups were so similar in both the
as-received condition and the after maintenance condition, whether from the
trained or the untrained facilities, that the differences which appear in the
tables could easily be due to random variations. The latter statement holds
true whether the comparison is made using the after-maintenance levels, or the
changes due to maintenance (the ideas of Tables 2 and 4).
The as-received emission levels compare favorably to samples of vehicles in
other test programs, thus giving credibility to the representativeness of the
vehicles used in the training study. Comparisons of 1975-1977 model year
vehicles from this training study and the EPA Portland Study (which was an
evaluation of the Portland I/M program) are given in Table 5. Both the
aatched, paired sample of 34 vehicles (17 vehicles each in the before and
after-training samples were 1975-77 model years), and all 75 vehicles from the
mechanic training evaluation which are 1975-1977 model years are shown. (The
Portland Element III program used only 1975-77 model year vehicles, which is
the reason for selecting only these years from the training study for compar-
ison.) The slightly lower HC and CO emissions of the Portland Element II and
III vehicles is expected, since these vehicles were tested when they were
younger and had less mileage than the vehicles in the Mechanic Training study.
-------
16
Table 5
/
As Received Status of Mechanic Training
Evaluation Vehicles and Portland Study Vehicles
1975-1977 Model Years Only
Federal Test Procedure
Study
Portland
Element II & III
Mechanic
Training
Evaluation (Matched)
Mechanic 75 44,931 3.13 44.0 2.67
Training
Evaluation (All)
N
207
34
Odometer
31,100
41,722
HC
2.69
3.17
CO
39.7
47.3
NOx
2.72
2.53
-------
17
6.2 Cost and Amount of Repairs by Commercial Facilities
The average maintenance costs were similar for the two fleets and are listed
in Table 6. The types and amounts of repairs for the two fleets were also
very similar. The repairs performed are listed in Table 7.
Table 6
Average Cost of Repairs by Facilities
Before Training $19.14
After Training $19.81
Table 7
Type and Amount of Repairs by Facilities
Item
Spark Plugs
Carburetor
Carburetor
Idle Speed
Timing
Air Filter
Oil
EGR System
Choke
Vacuum Hoses
Type of
Repair
Amount
Before
Training
Amount
After
Training
Amount By
Laboratory*
Replace 2
Adjust 20
Overhaul 1
Adjust 20
Adjust 2
Replace 0
Change 0
Repair/Replace 0
Repair/Adjust 0
Reroute/
Replace/
Repair ?
1
20
1
20
3
1
1
0
0
5
17
4
18
5
6
0
8
2
* Average for the before and after training groups.
-------
18
6*3 Quality of Repairs
>
The contractor performed various inspections ,.of the vehicles after each test
sequence. These inspections gave an evaluation of the maintenance primarily
in the areas of the carburetor and ignition timing, which are the most
frequent maintenance items performed.
Three measurements were taken which reflect the quality of maintenance to the
carburetion system: (1) the position of the carburetor idle mixture screw(s)
relative to the laboratory setting, (2) the maximum amount of engine speed
increase at idle due to the addition of propane gas, called "propane gain",
and (3) the idle emission levels, particularly idle CO. For commercial
repairs, each of these measurements can be compared to the laboratory measure-
ments for determining maintenance quality. (However, this method assumes that
the laboratory adjusted the carburetors perfectly which may not always have
been the case.) The carburetor idle mixture screw positions were measured in
terms of the number of 1/4 turns necessary to reach the seated position.
Propane gain correlates with the leanness of the air-fuel mixture; the greater
the engine speed increase with propane added, the leaner the setting. The
idle CO is important, but less so than the two measurements previously stated;
this is because nearly all of the vehicles have catalytic converters which
substantially decrease HC and CO resulting in readings at the tailpipe that do
not reflect well the emissions from the engine itself. Results of these three
measurements, shown in Tables 8-10 do not show any trend in the quality of
carburetor adjustments after training.
Concerning Table 10, training had no noticeable effect on the distribution of
idle CO levels. The mean levels and standard deviations are very similar,
when leaving out the one noted vehicle in Table 10. Also, 17 of 21 vehicles
in each group were adjusted to between 0.0 and 0.3% CO, confirming the
similarity in distribution.
-------
19
Table 8
Evaluation of Carburetor Idle Mixture
Screw Settings by Commercial Facilities
A. Number of 1/4 Turns Different from Contractor Setting*
(Positive numbers mean the commercial settings were richer than the contractor
settings.)
BEFORE TRAINING
AFTER TRAINING
Mean
.45
.24
Standard
Deviation
1.7
1.5
B. Absolute Value of Number of 1/4 Turns Different from Contractor Setting**
BEFORE TRAINING
AFTER TRAINING
Mean
0.9
1.3
Standard
Deviation
1.5
1.4
C. Frequency of Commercial Settings In Directions Relative to Contractor
Settings.
BEFORE TRAINING
AFTER TRAINING
Richer
11
11
Same
15
12
Leaner
7
10
* Example: If one screw were 2 1/2 turns out (from the seated position) after
commerical repair and only 2 turns out after contractor repair, the number of
1/4 turns different would be +2.
** Example: If the number of 1/4 turns different for one screw was +2 and
another was -2 the simple mean would be zero; the absolute mean would be 2,
though, because only the numbers are considered in computing absolute values,
not the signs. Using absolute values results in a higher mean for this reason.
-------
20
Table 9
v
Evaluation of Carburetor Idle Mixture Settings
Using Propane Gain Measurements
(Idle Speed Increase Due to Addition of Propane Gas)
Standard
Deviation
Mean* Mean of the
Increase Difference Differences
(rpm)
BEFORE TRAINING
Commercial Facilities 144
Contractor 130 -14 81
AFTER TRAINING
Commercial Facilities 104
Contractor 112 8 82
* Note that although the mean increases from the commercial facilities and the
contractor are very similar, the standard deviation is large, showing that for
individual vehicles the propane gains from the commercial facilities were
often quite different from the contractor setting. The differences were
positive one-half of the time and negative the other half, both before and
after training.
Table 10
Idle CO Settings (%}
Standard
Mean Deviation
BEFORE TRAINING
Commercial Facilities 0.38 (.18)* 0.92 (.29)*
Contractor 0.20 0.39
AFTER TRAINING
Commercial Facilities 0.15 0.24
Contractor 0.07 0.17
* Excluding one vehicle in the group which had a very high CO reading results
in the amounts shown in parentheses. The facility apparently adjusted this
vehicle to pass the state I/M test, got the certificate of compliance at the
state I/M lane, and then readjusted the vehicle before returning it to the
"owner". The assumed reason was that the vehicle had poor driveability when
adjusted to pass the DEQ. With more thorough repairs the contractor achieved
acceptable driveability and low emissions, however.
-------
21
The contractor measured three types of ignition timing which are useful in
evaluating the quality of repairs to this system by the commercial facili-
ties. The three types are basic, mechanical, and total advance. Basic timing
is the initial setting, usually made at idle. Mechanical (or centrifugal)
advance is the increase in advance when the engine is operated at higher
speed. Total advance measures the first two types plus engine vacuum advance,
and is also measured at a moderately high engine speed. Since the basic
timing is the only one which affects idle emissions, that will be the main
type of timing noted here; it also affects fuel economy significantly and, is
an important measure of the quality of repairs for that reason. Table 11
shows the number of vehicles with basic timing retarded before and after
repair.
Table 11
Vehicles With Basic Timing Retarded By More Than 2°
BEFORE TRAINING AFTER TRAINING
As-Received 2 1
After Repair by
Commercial Facilities 5 1
In the before training group, the two vehicles in Table 11 with retarded
timing as-received were unchanged by repair. Therefore, a total of three
vehicles were retarded by repair. In the after training group, the one
vehicle with retarded timing as-received was corrected by repair and one other
vehicle which had timing properly set as-received was retarded by the repair
facility.
Considering both mechanical and total advance together, two vehicles less in
each group were retarded (by more than 4°) after repair, thus some correct
repairs were mada in each group. It was not apparent what repairs were made
to the before-training cars to correct the mechanical and total retard, but in
the after-training group it was evident that two cars had their vacuum advance
mechanisms fixed.
A fourth measure of the quality of repairs can be made by looking at the type
of repairs performed for the different types of failures. A previous concern
with I/M repairs was that mechanics did not know what the possible causes of
HC and CO failures were and would perform some needless repairs. One
particular problem cited was that of performing repairs to the ignition
system, such as spark plugs and wires, distributor cap, rotor, etc., on a
vehicle which failed for CO only. The repairs just mentioned would be
unnecessary because they do not have an effect on CO. Earlier EPA studies of
the Portland I/M program showed that there was a significant amount of the'se
unnecessary repairs performed.
In this study almost no unnecessary repairs were noted. In the pre^training
group, one vehicle which failed for CO-only received a timing adjustment.
This was unnecessary, but is a reasonable and often recommended adjustment to
make when doing carburetor adjustments, and therefore is of almost no
-------
22
concern. No other adjustments or repairs to any part of the ignition system
were performed on any of the other vehicles which failed for CO-only, showing
good diagnosis even without training. Similar results were seen in the
post-training sample. Only one unnecessary, repair item was performed, the
replacement of spark plugs to a vehicle which failed for CO-only. The
majority of the vehicles failed for both HC and CO and therefore no judgements
could be made as to the types of repairs performed except to say that they
appeared to be correct.
The above refers to only the paired vehicles used in the analysis. Inspecting
the repairs of all other vehicles in the study, 'there were two instances in
the pre-training sample of expensive misdiagnoses, but none in the
post-training sample. One pre-training vehicle was diagnosed by a mechanic as
needing a new timing chain and gears with an estimated repair cost of $220.
This vehicle was sent to another facility and received a carburetor overhaul
which apparently was what it needed, although this cost $111. A second
vehicle was adjusted to pass the idle test, but the mechanic thought it might
have a bad catalytic converter. The laboratory found that the real problem
was leaking intake manifold gaskets; in all fairness, however, this was a
difficult repair case with the laboratory reportedly spending 6 hours in
repairs and noting that one cylinder had low power which could have been a
valve problem.
One expensive repair that the faciities often did not diagnose as necessary,
but which the laboratory found necessary, was carburetor overhauling. Of all
98 cars in the study, 25 received carburetor overhauling or replacements. Of
these 25, only 8 were diagnosed as being needed by the facilities (in 3 cases
outside of the matched, paired groups, EPA did not let them do the work even
though the laboratory later agreed that the carburetor needed to be over-
hauled). Apparently, repair facilities do not generally diagnose carburetor
overhauling unless they cannot get the vehicle to pass the idle test. This is
expected with the assumption that owners usually try to obtain the lowest cost
repairs which will allow their vehicles to pass the test.
As expected, vehicles which need carburetor repairs, but do not receive them,
still have high FTP emissions after repairs even though they pass the idle
test. Table 12 shows the FTP and idle emission levels at the different repair
stages of the 17 vehicles which needed carburetor work (as determined by the
contractor), but did not receive work by the repair facilities. Only 8 of the
17 were from the final matched samples (4 from each group in A and B). As can
be seen, FTP HC and CO remain quite high after repairs by the facilities. One
vehicle (not from the matched pairs) also had very high emissions after
laboratory repairs. The carburetor was overhauled, but the repair had no
effect on FTP HC or CO emissions. The contractor should have had further
repairs performed, because CO emissions remained high at the high speed
portion of the state idle test, which is an indication that there is still a
carburetor problem. This is a diagnostic check that was used in other cases,
but was not used on this vehicle after the carburetor overhaul, for
unexplained reasons. Without this one vehicle the mean FTP HC and CO for the
"combined" group (after laboratory repairs) would have been similar to the
mean for all the other vehicles after repairs by the laboratory (see Tables 1
and 2).
-------
23
Table 12
Emissions and Fuel Economy of Vehicles Needing
Carburetor Repairs But Not Receiving Them By The Commercial Facilities*
Idle Emissions**
Federal Test Procedure (Using Garage- Fuel Economy
Emissions (grams per mile) Type Analyzer) (miles per gallon)
BEFORE TRAINING (n=7)
As-Received
After Repairs
by Facilities
After Repairs
by Laboratory
AFTER TRAINING (n=10)
As -Received
After Repairs
by Facilities
After Repairs
by Laboratory
COMBINED (n=17)
As-Received
After Repairs
by Facilities
After Repairs
by Laboratory
HC
4.08
(3.88)
2.97
(2.44)
2.26
(1.75)
3.78
2.80
1.72
3.91
(3.83)
2.87
(2.71)
1.94
(1.73)
CO
72.1
(65.0)
57.5
(45.4)
32.7
(19.0)
63.5
46.4
17.5
67.0
(64.0)
51.0
(46.1)
23.8
(18.1)
NOx
2.37
(2.55)
2.11
(2.29)
2.46
(2.63)
2.37
2.29
3.22
2.37
(2.44)
2.22
(2.29)
2.91
(3.00)
HC (ppm) CO (%)
334
(366)
169
(162)
120
(118)
365
73
75
352
(365)
112
(106)
94
(92)
2.8
(3.1)
0.70
(0.72)
0.33
(0.24)
2.7
0.20
0.09
2.7
(2.8)
0.41
(0.40)
0.19
(0.15)
FTP
15.58
(16.46)
15.66
(16.89)
15.98
(16.98)
14.96
13.90
15.53
15.21
(15.49)
14.57
(14.88)
15.71
(16.04)
HFET
20.69
(21.52)
20.66
(22.05)
21.55
(22.62)
20.30
18.63
21.64
20.46
(20.74)
19.42
(19.78)
21.60
(21.99)
* Emissions and fuel economies without the one vehicle that still had a carburetor
problem are shown in parentheses.
** From the second idle portion of the SIT taken at the contractor's laboratory.
-------
24
A final measurement which would reflect the quality of repairs was a drive-
ability evaluation at each repair stage. The main problems looked for were
engine surging, stumbling, backfiring, stretchiness, misfiring and run-on
("dieseling"). The number of vehicles having at least one problem was greater
before training than after training, as shown in Table 13. The reader should
be aware of the fact that this evaluation was quite subjective and that only a
trend can be deduced from the results, i.e., that driveability remained good
after repairs by trained mechanics, whereas it was worse in some cases after
repairs by untrained mechanics. The after repair frequencies of problems are
not statistically different.
Table 13
Number of Vehicles With Driveability Problems
After Repairs After Repairs
As-Received by Facilities by Laboratory
BEFORE TRAINING (n=21) 2 63
AFTER TRAINING (n=21) 2 1 2
-------
25
6.4 Repair Effectiveness by Facility Type
As was mentioned in Section III,, four types of repair facilities were used:
gasoline-service stations, independent repair garages, chain stores, and
facilities advertising I/M repairs. In order to increase the number of
vehicles in some of the categories for statistical comparison purposes, the
facilities in the "advertising" category will be put into the appropriate
other three sections. This will particularly help the "independent" category
which formerly had only two facilities and three vehicles sent to it, because
three of the four facilities which advertised were independents (the fourth
advertiser was a gas station). The FTP HC and CO emissions, and fuel
economies are shown for each of the categories at the different repair stages
in Table 14. The two-sample t-test and the paired t-test show that there is
no significant difference between any of the different facility types' repairs
before and after training. The tests were performed for both the after
maintenance levels and the changes due to maintenance.
Table 14
Effectiveness of Different Repair Facility Types
BEFORE TRAINING
As After %
Received Repair Change
AFTER TRAINING
As After %
Received Repair Change
Gas Stations (n=7)
FTP HC
CO
mpg*
Independents (n=8)
FTP HC
CO
mpg*
Chain Stores (n=6)
FTP HC
CO
mpg*
2.68
47.0
15.30
3.17
44.1
14.65
2.93
52.4
15.68
1.65
24.5
15.35
2.06
20.1
14.65
1.51
24.1
15.67
-38%
-48%
0.3%
-35%
-54%
0
-48%
-54%
0
2.85
35.7
15.72
3.79
47.2
14.32
3.26
53.3
14.83
1.25
9.38
15.88
2.16
22.1
14.76
2.38
30.2
15.14
-56%
-74%
+ 1.0%
-43%
-53%
+3.1%
-27%
-43%
+2.1%
* City fuel economy. Highway fuel economy is not shown by facility type in this
table but is shown in Tables 1-4. The conclusion in Section 1,0 that training
produced a 0.8% improvement in fuel economy is based on combined city/highway fuel
economy.
-------
26
7.0 DISCUSSION AND CONCLUSIONS: POSSIBLE REASONS FOR THE OBSERVATIONS
IN PORTLAND AND GENERAL IMPLICATIONS FOR MECHANIC TRAINING BENEFITS
The results of this study showed small and statistically insignificant
incremental benefits due to training in the important areas of emissions and
fuel economy. This conflicts with the higher expectations of EPA, as well as
the expectations of many people in state agencies, industry, and the
commercial repair industry. This section will attempt to resolve this
apparent conflict.
Part A will discuss the optimum results of training, the actual results, and
the possible explanations for the differences between the expectations and the
results. These three topics will cover several important areas in the
training, e.g., diagnosis of the cause of the I/M failure.
Part B will discuss the implications that this study has for full-scale
mechanic training programs conducted as part of an operating I/M program and
the limitations of this study in determining the benefits of training.
7.1 Possible expectations of training, actual results and possible
explanations
7.1.1 Maximum expectations of what the training should accomplish
The ideal situation following an I/M mechanic training program would be
that all diagnoses and repairs which affect emissions and fuel economy are
completely accurate, no vehicles are left unrepaired to such an extent
that they are unable to pass the I/M test, and vehicle driveability is
good. The only repairs which are performed are ones which are in fact
necessary to correct an emissions or fuel economy problem; wasteful
replacement of parts that are still good would be eliminated. All
repairable emissions problems would be corrected, not just those which are
directly causing the I/M test failure. The entire vehicle emission
control system would ideally be working at its full function after
repair. The emission levels of individual repaired vehicles would
therefore be close to the Federal Test Procedure certification standards,
and fuel economy would be optimal.
Most I/M areas will have cost waivers. In areas which do allow cost
waivers, the number of waivers would ideally be less with trained
mechanics due to proper diagnoses and repairs resulting in lower repair
costs (assuming that the repairs necessary to pass the I/M test are
performed first). Also, fewer I/M-avoidance measures (registering a
vehicle in a non-I/M area, fraud, cheating) would occur, contributing to
lower fleet emission levels. The acceptable driveability of vehicles
after repair would result in fewer vehicles receiving readjustments after
repair, thus further contributing to low fleet emission levels.
-------
27
7.1.2 Actual results
The repair facilities appeared to perform nearly identical proper
diagnoses both before and after training. In the pre-training sample
there were only two cases of mis-diagnoses and none in the post-training
sample; however since problems which are difficult to diagnosis are rare,
the mechanics were not tested rigorously on their diagnosis abilities.
Repairs were generally correct, but not entirely accurate, resulting in
similar emissions and fuel economies before and after training. Judging
from the emissions and fuel economies achieved by the contractor
personnel, significant improvement could be made in the accuracy and/or
quantity of repairs by commercial facilities. The commercial facilities
did not correct all of the emissions problems that were in fact repairable.
No vehicles were left unrepaired to such an extent that they would not
pass the I/M retest, either before or after training. However, one
instance of a facility readjustment occurred in the before-training
sample. The readjustment was apparently due to poor driveability. It is
not possible to determine if this would have recurred after training if
the same circumstances had appeared.
7.1.3 Possible explanations
Mechanics in the Portland area had been familiar with I/M repairs for a
few years. Through experience on-the-job they seem to have learned to
diagnose the common problems of failing vehicles well.. Unusual problems
continue to give them difficulty, however, which result in some incorrect
diagnoses.
Concerning repair procedures, mechanics learned the correct procedures
(particularly carburetor adjustment procedures) well in the training, but
there is no evidence in the FTP or other data that they took the time to
do them precisely at the repair facilities. This is partly due to the
mechanics being satisfied with adjustments which enabled the vehicles to
pass the I/M test with a margin of safety and fairly good driveability.
Although the training course did try to teach mechanics to repair other
emissions problems not necessary to pass the I/M test, not enough
incentive existed for them to do so. The overriding incentives from
owners seem to be low cost and acceptable driveability. Also, due to
competition, repair shops want to have low average repair charges to their
customers and hope to have a reputation for inexpensive repairs.
7.2 General implications for mechanic training benefits and limitations of
this study
Little Likelihood of Complete Repair - In order to get the vehicles'
emission control system as close as possible to a fully functional
condition, the contractor's laboratory personnel often spent more than one
hour in diagnoses and repairs. Because of the substantial extra cost
necessary for these repairs, it is doubtful that a training program would
-------
28
ever result in commercial mechanics being as thorough. A legal require-
ment plus a large enforcement effort might be successful in mandating very
thorough repairs of all repairable problems, but this would likely be an
unacceptable option to all I/M areas.
Learning Curve - There may exist a learning curve such that mechanics will
have learned as much after a period of time of on-the-job training as they
would have with a formal training course. Consider the hypothetical
curves in Figure 4. The top curve may represent the knowledge gained when
formal training is given to mechanics in the first year of an I/M
program. That curve continues to rise due to continued knowledge gained
through experience, i.e., on-the-job type training. The bottom curve may
be the learning associated with on-the-job training (and no formal
training). In this hypothetical case, the difference between the curves
is greatest during the first year and then decreases steadily until the
third or fourth year at which time they are approximately equal. The
value of formal training is the area between the two curves and may
translate, to some extent, into emissions benefits during the first few
years. Since the formal training course in Portland occurred after I/M
had been in effect for about four years, the value of early training would
not have been detected.
The implication is that in any I/M program, in the early years formal
training will (1) reduce emissions and repair costs, (2) reduce the number
of cars which are not successfully repaired at first attempt and must
therefore get re-repaired (the ping-pong problem) or obtain waivers, and
(3) prevent cases of poor driveability after repairs with subsequent
readjustments. The results of (2) and (3) would lead to further emissions
benefits during the early years.
The effect of training in the later years of an I/M program appears to
depend on the stringency of the cutpoints used in the inspections.
Effect of Stringent Cutpoints - In mature I/M programs with cutpoints like
Portland's, there appears to be no emissions, cost or driveability
benefits from training. The mechanics may perform slightly more accurate
repairs, but if so, the difference in terms of emission levels after
repair appears to be minimal.
Effect of Lenient Cutpoints - In I/M programs with more lenient cutpoints
than in Portland the implications are less clear.
One possibility is that the learning curve mentioned above will occur,
with associated temporary benefits from formal training. Other than that,
mechanics may not perform any better than they would have after a few
years of on-the-job training, and at that point in time, trained mechanics
will not do any better than untrained mechanics. Mechanics may not adjust
cars to a point far below the cutpoints, but will stop at a margin of
safety point with which they are comfortable.
-------
29
Another possibility is that mechanics can be convinced to adjust the
vehicles to far below the cutpoints, to about the same levels as Portland
mechanics achieve (with the same FTP emissions). No extra time or
equipment is needed, making this idea plausible. Evidence from Portland
shows it is easily possible and can be done with low repair costs. If
this possibility occurs, there will be a permanent benefit from formal
training.
EPA believes that the second possibility is likely to occur. However,
only a training experiment in an I/M area with cutpoints much higher than
in Portland would verify this.
-------
30
EMISSION REPfUR KNOWLEDGE VERSUS TIME
(Conceptual Diagram)
UJ
o
o
UJ
o
z
ae
UNTRflJNED
TRfllNED
TIME (YEflRS)
-------
31
8.0 MECHANIC TRAINING EMISSION REDUCTION BENEFITS CONTAINED IN MOBILE2
Based on this study of mechanic training, I/M emission reduction benefits
associated with mechanic training were derived and are available in MOBILE2..
As would be expected from the above analyses, the incremental benefits due to
mechanic training are substantially smaller than those appearing in MOBILE1.
Generally, the assumption used in the derivation of MOBILE2 was that the
after-maintenance mean FTP levels are approximately equal to those seen in
this mechanic training study if Portland short test cutpoints are assumed. If
the cutpoints differ from Portland's, an adjustment is made. For a full
explanation of the derivation, see the documentation for derivation of
pre-1981 I/M credits[3].
Mechanic training only effects I/M benefits for pre-1981 model year cars in
MOBILE2. In brief, it is felt that 1981 and later model year cars will tend
to experience component failure instead of the maladjustments typical of
pre-1981 model year cars. Thus, a car of advanced technology experiences
quantun instead of continuous emission reductions. For more detail see the
documentation for derivation of 1981 and later I/M credits[4].
Table 15 gives examples of the incremental fifth-year benefits due to mechanic
training as seen in MOBILE2. These examples show typical results of 0-2
percent HC and 0-6 percent CO reductions associated with emissions from
pre-1981 model year cars. After averaging these effects with the portion of
the fleet experiencing no effect from mechanic training, incremental benefits
due to mechanic training are typically 0-1 percent for HC and 0-4 percent for
CO.
Table 15
MOBILE2
Incremental Benefits Due to Mechanic Training
Start I/M 1/1/83
Evaluate 1/1/88
Pre-1981 Pre-1981 Total
Stringency Fleet* Fleet**
HC CO HC CO
20 05 03
30 2614
40 26 14
* Incremental benefits in percent reduction to pre-1981 model year emission
factors.
** Incremental benefits in percent reduction to total fleet emission factors
assuming a 50% Identification Rate for 1981 and later model year cars.
-------
32
References
1. National Academy of Sciences, Report by the Committee on Motor Vehicle
Emissions, February 12, 1973, pp 74-75.
2. Automotive News, December 19, 1978, pp 18-19.
3. "Derivation of I/M Benefits for Pre-1981 Light Duty Vehicles for Low
Altitude, Non-California Areas", EPA-AA-IMS/81-4 (In preparation at the
time of this report).
4. "Derivation of I/M Benefits for Post-1980 Light Duty Vehicles for Low
Altitude, Non-California Areas", EPA-AA-IMS/80-8, January 1981.
-------
33
APPENDIX
Vehicle Test Fleet
-------
Vehicle List
Pre-Training Vehicles Used For The Final Matched Analysis
Vehicle #
Make
CID
CYL
CARS TRANS. AIR
ODOMETER
I/M FAIL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
60506
40602
42001
61057
42002
61002
61006
62002
62102
60151
61000
41501
71015
71105
52057
51016
52018
52019
72111
72005
71120
Chev
Olds
Ford
Ford
Ford
Chev
Chrys
Ford
Chrys
Chrys
Chev
Chrys
Pont
Ford
Chrys
Chev
Chev
Chrys
Chrys
Chrys
Ford
250
350
302
250
302
305
318
140
318
318
305
225
301
140
225
350
140
225
318
225
250
6.
8.
8.
6.
8.
8.
8.
4.
8.
8.
8.
6.
8.
4.
6.
8.
4.
6.
8.
6.
6.
1-V
4-V
2-V
1-V
2-V
2-V
2-V
2-V
2-V
2-V
2-V
1-V
2-V
2-V
1-V
2-V
2-V
1-V
2-V
1-V
2-V
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
MANUAL
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
MANUAL
AUTO
AUTO
AUTO
AUTO
NO
NO
NO
YES
NO
NO
NO
YES
NO
NO
NO
NO
NO
NO
NO
NO
NO
NO
YES
YES
NO
34615
42503
74020
25541
75391
47364
30329
75065
51893
25304
93687
63158
41184
38866
42182
51366
49444
75571
33306
24192
20676
HC,
CO
CO
HC,
CO
CO
HC,
HC,
HC,
CO
HC,
CO
HC,
CO
HC,
HC,
... -- JIG,
"HC,
CO
HC,
HC,
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
>
X)
•a
0>
3
a
X
-------
Vehicle List
Post-Training Vehicles Used For The Final Matched Analysis
Vehicle #
Make
CID
CYL
CARB TRANS.
ATR?
ODOMETER
I/M FAIL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
52001
40605
42003
60568
42005
60416
62201
70398
60302
60507
76012
41503
70262
61101
60584
60308
60385
60388
51055
70287
70508
Chev
Olds
Ford
Ford
Ford
Chev
Chrys
Ford
Chrys
Chrys
Chev
Chrys
Pont
Ford
Chrys
Chev
Chev
Chrys
Chrys
Chrys
Ford
250
350
302
250
302
305
318
140
318
318
305
225
301
140
225
350
140
225
318
225
250
6.
8.
8.
6.
8.
8.
8.
4.
8.
8.
8.
6.
8.
4.
6.
8.
4.
6.
8.
6.
6.
1-V
4-V
2-V
1-V
2-V
2-V
2-V
2-V
2-V
2-V
2-V
2-V
2-V
2-V
1-V
2-V
2-V
2-V
2-V
2-V
1-V
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
MANUAL
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
AUTO
NO
NO
NO
YES
NO
NO
NO
YES
NO
NO
NO
NO
NO
NO
NO
NO
NO
NO
NO
YES
YES
58393
56942
32763
39789
95156
46938
26254
58393
37182
44303
36401
35147
42703
23668
32107
29508
61747
30424
27652
45990
16517
HC,
CO
CO
HC,
CO
HC,
HC,
HC,
HC,
CO
HC,
CO
HC,
CO
HC,
HC,
HC,
HC,
CO
HC,
HC,
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
•§•
*o
m
«
a.
H»
X
------- |