HIGH ALTITUDE VEHICULAR EMISSION CONTROL PROGRAM
VOLUME VII. EXPERIMENTAL CHARACTERIZATION
OF VEHICULAR EMISSION AND
ENGINE DETERIORATION
PREPARED FOR:
STATE OF COLORADO
DEPARTMENT OF HEALTH
DENVER, COLORADO 80220
ENVIRONMENTAL PROTECTION AGENCY
REGION VIII
DENVER, COLORADO 80203
ATI /automotive
ATL/ testing TRW
/ LABORATORIES '
ENVIRONMENTAL
-------
STUDY TEAM
ATL
Douglas R. Liljedahl
Jerry L. Terry
TRW
Owen P. Hal 1, Ph.D.
Steven N. Crane
Paul S. Rosenbloom
- ATL, Project Manager
- Procedures Engineer
- TRW, Project Manager
- Systems Analyst
-------
4}
\ U^r\
oioH
HIGH ALTITUDE VEHICULAR EMISSION CONTROL PROGRAM
VOLUME VII. EXPERIMENTAL CHARACTERIZATION
OF VEHICULAR EMISSION AND
ENGINE DETERIORATION
PREPARED FOR:
STATE OF COLORADO
DEPARTMENT OF HEALTH
DENVER, COLORADO 80220
ENVIRONMENTAL PROTECTION AGENCY
REGION VIII
DENVER, COLORADO 80203
ATI /automotive „.../
AT L/testing TRW
/ LABORATORIES '
ENVIRONMENTAL
-------
DISCLAIMER
This report was developed for the Environmental Protection Agency
and the State of Colorado by TRW and ATL under contract numbers
68-02-1385 and C290526. The contents of this report are reproduced
herein as received from the contractor. The opinions, findings,
and conclusions are those of TRW and ATL and are not necessarily
those of the sponsoring agencies. Mention of company or product names
does not constitute endorsement by either the Environmental Protection
Agency or the State of Colorado.
The results and conclusions presented are based on the preliminary
data developed from the first six months of the ongoing deterioration
test program (conducted by Automotive Testing Laboratories). The extent
to which these data are not representative of the vehicle population
in the Denver area, however, could have a significant impact on the
resultant conclusions and recommendations.
-------
PREFACE
This report, "High Altitude Vehicular Emission Control Program,"
consists of seven volumes. Listed in the following are the subtitles
given for each volume:
• Volume I - Executive Summary, Final Report, January
1974.
• Volume II - Experimental Characterization of Idle
Inspection, Exhaust Control Retrofit and Mandatory
Engine Maintenance, Final Report, December 1973.
• Volume III - Impact of Altitude on Vehicular Exhaust
Emissions, Final Report, December, 1973.
• Volume IV - Analysis of Experimental Results, Final
Report, December 1973.
• Volume V - Development of Techniques, Criteria and
Standards to Implement a Vehicle Inspection,
Maintenance and Modification Program, Final Report,
December 1973.
• Volume VI - The Data Base, Final Report, January 1974.
e Volume VII - Experimental Characterization of Vehicular
Emission and Engine Deterioration, Interim Report, June
1974.
The first volume summarizes the general objectives, approach and
results of the study. The second volume presents a detailed description
of the experimental programs conducted to define the data base. Volume
III reports the methods and analysis used in developing the basic
relationships between mass emissions and altitude. A quantitative
analysis of the results from the experimental program is presented in
Volume IV. The fifth volume provides an analysis of the techniques
and criteria required in establishing a vehicle emission control program
for the Denver area. The actual data base developed from the experimental
-------
program is given in Volume VI. Lastly, Volume VII reports the results
of the six month deterioration program.
The work presented herein is the product of a joint effort by
several consulting firms. Automotive Testing Laboratories (ATL) was
responsible for the design and implementation of the basic experiments.
TRW provided the data management and analysis of the experimental
results.
-------
ACKNOWLEDGEMENTS
TRW would like to acknowledge the efforts extended by the Colorado
State Department of Health and Region VIII of the Environmental Pro-
tection Agency.
The contributions of Messrs. Don Sorrels and Robert Taylor of
the Colorado State Department of Health were of particular significance.
Mr. Dale M. Wells of Region VIII served as Project Officer.
-------
TABLE OF CONTENTS
Pa^e
1.0 SUMMARY AND CONCLUSIONS 1-1
2.0 INTRODUCTION 2-1
3.0 PROCEDURES DEVELOPMENT 3-1
3.1 Program Design 3-1
3.2 Test Vehicles 3-4
3.2.1 Vehicle Sample Composition 3-4
3.2.2 Vehicle Preparation and Handling 3-5
3.3 Laboratory Testing and Evaluation 3-6
3.3.1 Ehaust Emission Testing Procedures 3-6
3.3.2 Engine Diagnostic Procedures 3-7
3.4 Data Processing and Management 3-8
4.0 BASIC EXPERIMENTAL RESULTS 4-1
4.1 Unauthorized Maintenance 4-1
4.2 Comparison of Mandatory Maintenance Regime and
Idle Inspection Regime 4-3
5.0 IMPACT OF DETERIORATION ON ALTERNATIVE PROGRAMS 5-1
5.1 Idle Inspection and Maintenance 5-1
5.2 Mandatory Engine Maintenance 5-21
-------
LIST OF TABLES
Table Page
3-1 Distribution of Sample After Six Months 3-5
4-1 Summary of Unauthorized Maintenance 4-2
4-2 Estimated T-Scores for Maintenance Regimes for Total
Population 4-4
4-3 Estimated T-Scores for Maintenance Regimes With Un-
authorized Maintenance Eliminated 4-5
5-1 Idle Inspection Pass/Fail Criteria 5-2
5-2 Idle Inspection CVS Emissions Pre and Post Maintenance 5-3
5-3 Idle Inspection CVS Emissions Pre and Post Deterioration 5-5
5-4 Idle Inspection Mode Failure Rates for Total Population 5-13
5-5 Idle Inspection Mode Failure Rates With Unauthorized
Maintenance Eliminated 5-14
5-6 CVS Emissions for Idle Inspection Precontrol Vehicles .. 5-15
5-7 CVS Emissions for Idle Inspection Controlled Vehicles .. 5-16
5-8 Linear Least Squares Regressions for Idle Inspection
Deterioration CVS Emissions 5-18
5-9 Idle Inspection Emissions Levels to Achieve Significant
Deterioration 5-20
5-10 Analysis of Idle Inspection and Maintenance Regime for
Total Population 5-22
5-11 Analysis of Idle Inspection and Maintenance Regime With
Unauthorized Maintenance Eliminated 5-23
5-12 Idle Inspection CVS Emissions for Precontrol Vehicles
With Small Engines 5-24
5-13 Idle Inspection CVS Emissions for Precontrol Vehicles
With Large Engines 5-25
5-14 Idle Inspection CVS Emissions for Controlled Vehicles
With Small Engines 5-26
-------
LIST OF TABLES - (continued)
Table Page
5-15 Idle Inspection CVS Emissions for Controlled Vehicles
With Large Engines 5-27
5-16 Mandatory Maintenance CVS Emissions Pre and Post
Maintenance 5-28
5-17 Mandatory Maintenance CVS Emissions Pre and Post
Deterioration 5-29
5-18 CVS Emissions for Mandatory Maintenance Precontrol
Vehicles 5-37
5-19 CVS Emissions for Mandatory Maintenance Controlled ~
Vehicles 5-38
-------
Fi qure
LIST OF FIGURES
Page
5-1 HC Histogram for Idle Inspection Pre and Post
Deterioration for Total Population 5-6
5-2 HC Histogram for Idle Inspection Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-7
5-3 CO Histogram for Idle Inspection Pre and Post
Deterioration for Total Population 5-8
5-4 CO Histogram for Idle Inspection Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-9
5-5 NOx Histogram for Idle Inspection Pre and Post
Deterioration for Total Population 5-10
5-6 NOx Histogram for Idle Inspection Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-11
5-7 HC Histogram for Mandatory Maintenance Pre and Post
Deterioration for Total Population 5-31
5-8 HC Histogram for Mandatory Maintenance Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-32
5-9 CO Histogram for Mandatory Maintenance Pre and Post
Deterioration for Total Population 5-33
5-10 CO Histogram for Mandatory Maintenance Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-34
5-11 NOx Histogram for Mandatory Maintenance Pre and Post
Deterioration for Total Population 5-35
5-12 NOx Histogram for Mandatory Maintenance Pre and Post
Deterioration With Unauthorized Maintenance Eliminated . 5-36
-------
1.0 SUMMARY AND CONCLUSIONS
This report highlights the findings from the first phase (six
months) of the high altitude emission deterioration program. The deter-
oration data presented herein reflects the results from both the idle
inspection and mandatory maintenance programs.
9 Six months of emission data is insufficient to characterize the
overall effect of engine deterioration.
o Vehicle attrition and unauthorized maintenance has had a
significant impact on determining the actual effect of engine
deterioration.
a Approximately one-third of the vehicles initialized were
unavailable for recall at the end of six months.
o A substantial level of unauthorized maintenance, on the
order of 25 percent, occured for both the idle inspection
and mandatory maintenance fleets.
e The potential impact of increased vehicle attrition and
unauthorized maintenance may preclude developing statistically
significant estimates of engine deterioration during the next
six month period.
e An expanded experimental program may be necessary to fully
evaluate the actual effects. Another alternative would be to take
an emission surveillance "snapshot" of the vehicle population
on a periodic basis in order to observe general emission
deterioration trends.
o A program to better characterize the nature and extent of
unauthorized maintenance is required.
e In general, vehicular emission levels (as measured
in CVS units) did not show a statistically significant increase
over the six month period. One exception occured for the
mandatory maintenance fleet. Here, HC emissions increased
to a value larger than the original "as received" state. This
can be attributed to the small sample size and the high
incidence of ignition misfire within this population segment.
0 On the average, idle HC and CO rejection rates at six months
(using the 50 percent failure criteria) were lower than for
the pre-maintenance vehicle population.
-------
• Deteriorated emission levels need to reach approximately
80 percent of the "as received" values in order to show a
statistically significant difference from the post-tuned
emission state.
• Regression analysis indicated that accumulated mileage
over the six month period could not explain the observed
rate of engine deterioration. A larger sample size might
reveal better correlations between these two variables.
• The emission levels for the idle inspection and mandatory
maintenance fleets showed a statistical difference for
both the post-maintenance and post-deteriorated states.
This can be attributed primarily to the effect of maintenance
and the fact that the mandatory fleet contained somewhat
newer vehicles.
-------
2.0 INTRODUCTION
A basic unknown in evaluating the cost-effectiveness of vehicular
inspection/maintenance is the extent and characteristics of emission
deterioration. Studies conducted by TRW have shown that emission
deterioration can have a substantial effect on the effectiveness of
the selected procedure (e.g., idle inspection).* In an attempt to
assess the potential impact of emission deterioration at altitude,
an experimental test program was undertaken using approximately
250 vehicles. The project was designed to characterize emission
deterioration and engine degradation for vehicles operating in
the greater Denver area. The project, involving the re-test of
vehicles available at the conclusion of two consecutive six month
intervals recently progressed beyond mid-point. This interim report
presents a detailed analysis of the results from the six month engine
deterioration and degradation study.
The primary objective of this study was to develop emission
deterioration factors which are to be utilized to evaluate the
long term benefits of a mandatory idle inspection and maintenance
program and of a mandatory engine maintenance program. A secondary
object is to determine the extent of owner tampering, including its
impact upon the effectiveness of an inspection/maintenance program,
and to determine possible legislative requirements to maintain over-
all effectiveness of a mandatory program.
The preceeding section (Section 1.0) has provided a summary of
the conclusions and recommendations synthesized from the study. This
section (Section 2.0) focuses on the background of the test program
* TRW, CAPE-13-68 Research Program (1972).
-------
and presents a summary of the basic experimental objective. Section
3.0 outlines ATL's laboratory procedures and TRW's data management
system. Section 4.0 summarizes a comparison of the two programs as well
as discusses the extent and impact of unauthorized maintenance. Section
5.0 [highlights the results for the idle inspection and mandatory
maintenance programs]. This analysis includes frequency histogram plots
of pre and post deterioration CVS emissions, statistical regression
evaluation, and a variety of other statistical assessments.
-------
3.0 PROCEDURES DEVELOPMENT
3.1 PROGRAM DESIGN
The program was designed to utilize a preconditioned and pretested
sample of vehicles from which various data, pertinent to program
objectives, could be obtained.
A sample of three-hundred 1964 through 1973 model-year vehicles,
selected to represent that segment of the Colorado light-duty vehicle
(under 6000 lbs GVW) population, were used initially to evaluate
idle inspection and maintenance, emission control retrofit and mandatory
engine maintenance. This segment represented about 90% of the light-
duty vehicle population. All vehicles in the sample were-initially
subjected to inspection and maintenance. Idle emission inspections
were performed at ten selected state licensed motor vehicle safety
inspection stations. Station personnel were trained in advance and
were required to perform inspection and maintenance of vehicles in
accordance with specific procuedures. All vehicles were laboratory
tested in the as-received condition before delivery to the stations.
Vehicles which failed station inspection and were subsequently repaired
were re-tested by laboratory procedures to determine the effectiveness
of station performance. A segment of the vehicle sample was then
utilized to evaluate emission control retrofit and modified tune-up
specifications. *
* A more complete description of the basic test procedures can be
found in Volume II of this report.
-------
Although three-hundred vehicles comprised the initial sample,
the potential size of the sample available for re-test was reduced
to about two-hundred and fifty vehicles.* This reduction, numbering
about fifty vehicles, came about as a result of an initial loss of
several vehicles which had been tested and released to owners prior
to start-up of the deterioration study. A number of other vehicles
comprising the initial sample were determined to be unsuitable for
deterioration study purposes for various other reasons.
During the time interval following initial testing, vehicles
comprising the test sample were presumably operated in a typical
manner although several existing factors undoubtedly had some
impact on mid-point results. Weather conditions during the interval
(August, 1973 through May, 1974), for example, were for the most
part seasonably cold and presumably had an indirect effect on engine
warm-up characteristics, mileage accumulation and maintenance
requirements. Additionally, the fuel, crisis and attendant factors,
were predominant throughout much of the study interval and are
believed to have potentially biasing effects on study results by
altering mileage accumulation patterns, fuel preference and overall
vehicle useage. The effects of these variables are virtually im-
possible to evaluate. In any case, however, the project proceeded
according to design.
In planning for the interval which followed initial testing,
procedures for handling test vehicles exhibiting undesirable operating
characteristics or component failure were devised. A maintenance
* Of the original 250 vehicles, 150 were assigned to the idle
inspection fleet and 100 to the mandatory maintenance fleet.
These initial group sizes were reduced (the results of attrition)
to 87 and 78 vehicles, respectively.
-------
committment was established and communicated to the vehicle owner. A
vehicle prematurely returned to the laboratory for repair work was sub-
jected to an on-the-spot inspection to establish the validity of a re-
quest for repair. In certain situations, owner dissatisfaction with some
aspect of vehicle performance proved to be unfounded. In other situations,
a legitimate requirement for maintenance did in fact exist. In the
latter case, a loan car was issued and the test vehicle was retained
for further testing and maintenance. In this case, a series of tests,
identical to those performed initially, was conducted, repairs to the
vehicle were completed and the vehicle was returned to its owner.
Data developed as a result of these procedures were retained for sub-
sequent processing and reporting.
As reported earlier, two hundred and fifty vehicles of the
original sample were judged suitable for retest. It was originally
anticipated that a significant number of vehicles would be lost to
the program for various reasons including transfer of ownership,
owner relocation, accidents and negative owner reaction or loss of
interest. In this respect an attrition rate of 40 percent was allowed.
After the nominal interval of six months had elapsed, vehicles
were recalled for deterioration and degradation testing. An attempt
to maintain an initial tolerance of 180 + 5 days proved to be imprac-
tical. The tolerance was subsequently relaxed to 180 + 10 days to
maintain a high retest rate.
At the appropriate time, one or more attempts to contact owners
whose vehicles qualified for retest were made. As anticipated, a
significant number of owners had moved outside the area, had sold the
-------
test vehicles, had expressed disatisfaction with some aspect of the
program or had simply become disinterested. On the other hand, a
significant number of vehicles remained available. The remaining
vehicles, numbered at 165, were recalled and retested. Testing pro-
cedures, identical to those performed initially, were then applied.
At test completion seals applied to engine adjustments and identification
marks applied to emission related components were inspected and the
incoming status of each adjustment and component part was recorded.
Data were then processed and compiled.
Through the retest phase of the program laboratory instrumentation
and equipment calibration and operating procedures were maintained
in accordance with standards applied in the initial program. Quality
control tolerances were similarly maintained and procedures relating
to data auditing were applied.
3.2 TEST VEHICLES
Three hundred vehicles were initially selected and utilized to rep-
resent the 1964 through 1973 Colorado motor vehicle population. Approx-
imately two hundred and fifty were prepared for the deterioration study
phase. As anticipated, a significant level of attrition occured during
the six month period and one-hundred and sixty-five vehicles were
actually submitted for retest.
3.2.1 Vehicle Sample Composition
Table 1 shows the distribution of vehicles subjected to retest.
-------
TABLE 3-1
DISTRIBUTION OF SAMPLE AFTER 6 MONTHS
Model -
Year
73
72
71
70
69
68
67
66
65
64
Total
Make
Ammo
0
1
1
1
0
1
0
1
0
1
6
Buic
0
0
1
0
0
2
1
0
1
1
6
Cadi
0
0
0
1
0
0
0
0
2
0
4
Chev
6
4
3
3
2
2
5
5
5
5
40
Chry
0
1
1
1
0
0
1
1
0
0
5
Dodg
1
1
0
1
1
1
1
1
0
1
8
Ford
4
4
6
3
2
4
6
5
7
2
43
Merc
0
1
0
0
1
1
1
0
1
0
5
Olds
1
0
0
1
0
1
1
1
1
1
7
Plym
1
3
1
2
2
1
1
1
0
1
13
Pont
1
1
1
1
1
0
3
2
2
0
12
Volk
0
1
1
2
0
2
0
2
1
0
9
Toyo
1
0
1
1
1
0
0
0
0
0
4 ¦
Dats
0
0
1
0
0
0
0
0
0
0
1
Opel
1
1
0
0
0
0
_0
0
0
0
2
Total
16
18
17
17
10
15
20
18
20
14
165
In comparison of the initial sample (Table 2. Volume II) versus
the sample shown in Table 3-1 above, it can be seen that the retested
sample approximates the initial sample by a factor of 0.5 by both
make and model-year. From this comparison it can be concluded
that the retested sample approximates the distribution of light-duty
vehicles in Colorado.
3.2.2 Vehicle Preparation and Handling
Upon receipt of the vehicle for retest, an inspection of the
vehicle exterior, interior and exhaust system was performed to deter-
mine incoming status. A loan car was issued to replace the test
vehicle and the necessary vehicle agreement forms were completed.
-------
The vehicle was then moved to the laboratory for temperature soaking
prior to emission testing and engine inspection.
After a minimum soak period of twelve hours, the vehicle fuel
supply system was disconnected and reconnected to a laboratory fuel
supply system. A batch of summer-grade fuel, utilized for initial
testing had been retained and was used to perform the retests.
Emission tests were then performed, the vehicle was relocated
to another area in the laboratory and a inspection of engine com-
ponents and adjustments was completed. The vehicle was then returned
to its owner.
3.3 LABORATORY TESTING AND EVALUATION
Procedures employed for retest were identical to those applied
initially.
3.3.1 Exhaust Emission Testing Procedures
Laboratory standard exhaust emission tests were performed in
accordance with procedures outlined in Federal Register, Volume 3B,
Number 124, Part III, dated June 28, 1973. Standard tests were
preceeded by a minimum 12 hour temperature soak at laboratory
ambient conditions (68° F to 72° F).
Key mode tests were performed in accordance with procedures
outlined by Clayton Manufacturing Company of El Monte, California.
Key mode testing and resulting data have no direct bearing on the
objectives of this phase of the study but were included in the
test procedure merely to expand the data base.
Idle emission testing was performed in conjunction with key mode
testing. Emission samples were taken at no load conditions of curb
idle (Drive gear for automatic transmission equipped vehicles) and
-------
2500 engine rpm. Instrumentation and operating procedures were
identical to those employed during the initial phase of testing.
Both laboratory garage-type inspection equipment (listed in Table
5 of Volume II) were employed.
Instrument and equipment calibrations established in the initial
testing phase were maintained throughout the retest interval.
Analytical system calibrations were established using an inventory
of EPA named gases. Flow calibration of the CVS was verified
using the laminar flow element with calibration traceable to the
National Bureau of Standards. Dynamometer calibrations were estab-
lished and verified on a regular basis using the coast-down technique.
In addition, propane recovery tests, NOx converter efficiency checks
and analytical system leak checks were performed on a daily basis.
3.3.2 Engine Diagnostic Procedures
Diagnoses of engine conditions were performed at two points in
the overall vehicle procedure. During key mode operations on the
chassis dyanamometer the laboratory analytical system recorders were
operational for a period of about one minute. During this interval
and during periods of speed changes, emission traces were observed
for an indication of malfunction evidenced by abnormally high
hydrocarbon (HC) or carbon monoxide (CO) levels. The HC trace also
provided an indication of ignition system mis-fire. Observations
were recorded. Oxides of nitrogen (NOx) emission controls on applic-
able vehicles were also inspected for proper operation during key
mode testing. A fully operational system was indicated by the absence
of vacuum to the distributor at low cruise conditions and the pres-
-------
second point at which diagnostic procedures were applied was im-
mediately after dynamometer tests were completed. The vehicle was
removed from the dynamometer area and a more extensive diagnostic
procedure was applied. Concurrently, the inspection to determine
the extent of tampering and alteration or replacement of parts was
performed.
3.4 DATA PROCESSING AND MANAGEMENT
The combination of a large data base, and the necessity of a
series of complex operations involved in the analysis, necessitated
the use of a computerized data management system.
The data collected from the deterioration experiment has been
stored as a working file on the CDC 6500 disk pack. More permanent
copies of the data base are also stored on a magnetic tape and on
card files. Extensive computer software was utilized in managing
and processing the data. The following data management activities
were performed by TRW:
• retrieval and sorting of data
• development of graphic presentations
• statistical analysis
A brief description of each of these data management functions is
presented in the subsequent paragraphs.
Retrieval and Sorting of Test Data
The principal data handling program in the data management
system, DETER, serves as the basic interface between the data base
and other software. This program retrieves the selected data from
disk storage and sorts it by a number of classification systems. The
data can be culled in the following ways:
-------
e CVS emissions
9 Key modes
0 Engine parameters
e Vehicle characteristics
1) the total population
2) sort by age group
3) sort by manufacturer
4) sort by make, within a manufacturer
5) sort by engine size group
6) sort by weight group
7) sort by PASS/FAIL at idle inspection
Graphic Presentation
"One major element of the data analysts activity was to summarize in-
formation in the form of frequency histograms. Two programs are used
to preform the retrieval and plotting functions. DETER is used to
retrieve data records for the requested plot, sort the data, and write
a data input file for the plotting routine. HISTM provides histo-
gram frequency plots of the data. The plots are drawn on by a Cal-^
Comp plotter from a tape written by HISTM.
Statistical Analysis
Two main statistical packages were used in processing the data
relevant to the engine deterioration study.
The data handling program DETER had basic statistical capabilities
built into it. It computes means, standard deviations, T-scores,
and CVS emission deterioration factors.
The DETER program also is used to create an input file for the
other statistical applications program (TSTAGE). The TSTAGE program
performs ordinary least squares regressions. Equations were derived for
the deterioration of the three CVS mass emissions (HC, CO, NOx) as
functions of the odometer readings. Dummy variables were exploited
-------
to yield independent equations for the different sort groups (i.e.,
2
age group). T-scores, F-values, and R computations were all carried
out in an effort to determine statistical significance.
-------
4.0 BASIC EXPERIMENTAL RESULTS
This section presents the basic experimental results developed
from the first six months of the deterioration program. The analysis
i
focuses on the extent and frequency of unauthorized maintenance that
occured during the program. In addition, an assessment is made of the
potential differences between the two test fleets.
4.1 UNAUTHORIZED MAINTENANCE
During the course of this study it was necessary to return the
test vehicles to their owners for a deterioration period of six months.
Before this was done, key engine components were marked so that
unauthorized maintenance performed upon them could be detected during
the subsequent re-inspection. Once the vehicles were recalled an
inspection was performed to determine the state of the engine system.
This inspection provided a direct means for assessing the extent of
unauthorized maintenance. The results from this inspection are
summarized in Table 4-1. Fully twenty-f>ve percent of the vehicles
were subject to unauthorized maintenance during the six month
deterioration period. In general, the incidence of unauthorized main-
tenance appears uniformly consistent. Spark plugs and air filters
represented two of the more frequently changed components for both
vehicle fleets. This situation is somewhat surprising in the case of
the mandatory maintenance population because all vehicles in that group
underwent a thorough engine tune-up. The repair of other ignition
oriented components (e.g., wire, coil) may relate strongly to the high
level of ignition misfire detected in the vehicle population (15
percent versus 5 percent at sea level). Clearly, the extent of
-------
TABLE 4-1
SUMMARY OF UNAUTHORIZED MAINTENANCE
IDLE INSPECTION
MANDATORY MAINTENANCE
-p»
I
ro
TAMPERED ITEM
IGNITION WIRES
COIL
SPARK PLUGS
AIR FILTER
FUEL MIX. SCREWS
IDLE ADJUST. SCREWS
CHOKE SETTING
DIST. ADJ. SCREW
POINTS ADJ. SCREW
CONDENSER
ROTOR
DISTRIBUTOR CAP
PCV
NUMBER
5
0
7
12
1
4
4
4
4
4
4
5
3
PERCENT
6
0
9
14
2
5
5
5
5
5
5
6
4
NUMBER
3
2
6
7
2
3
4
3
5
4
3
6
2
PERCENT
4
3
8
9
3
4
6
4
7
6
4
8
3
TOTAL CARS TAMPERED
SAMPLE SIZE
23 27
(87)
19 25
-------
unauthorized maintenance observed in this experiment underscores the
need for the development of procedures for controlling this phenomena
in the general population. A study program to more fully evaluate the
possible effects of unauthorized maintenance on the operational ef-
fectiveness of the posed idle inspection program appears warranted.
Because of the possible bias of including vehicles which receive
unatuhorized maintenance, results were calculated for both the total
population and for those vehicles which had not received additional
maintenance.
4.2 COMPARISON OF MANDATORY MAINTENANCE REGIME AND IDLE INSPECTION
REGIME
To reasonably compare the data for the two maintenance programs
it must first be determined whether the vehicles in the two fleets
come from the same population or indeed were different. Tables 4-2
and 4-3 summarize T-scores for the two populations for the following
three states: pre-maintenance, post-maintenance, and post deteriora-
tion. These results are reported by control type. The analysis
that follows will be based upon Table 4-3 containing just those
vehicles not subjected to unauthorized maintenance. The T-scores
imply that HC and NOx emissions were comparable before maintenance,
while CO emissions were not.* Post-maintenance T-scores reveal
that at that time, emissions were significantly different between
the two regimes.
* A T-score greater than unity indicates that a statistical difference
exists between the two populations.
-------
TABLE 4-2
ESTIMATED T-SCORES FOR MAINTENANCE REGIMES FOR TOTAL POPULATION
MODE
PRE-MAINTENANCE
POST-MAINTENANCE
POST-
DETERIORATION
I
-P*
HC
CO
NO
X
SAMPLE SIZES
(l/M) (M/M)
HC
CO
NO
x
SAMPLE SIZES
(l/M) (M/M)
PRECONTROL VEHICLES
1.217 0.856
1.886 1.768
1.075 1.333
(49) (23)
(49) (23)
CONTROLLED VEHICLES
0.281 1.643
0.104 1.653
0.798 1.680
(38) (55)
(38) (55)
3.876
2.057
1.955
(49) (23)
0.509
1.332
1.027
-------
TABLE 4-3
ESTIMATED T-SCORES FOR MAINTENANCE REGIMES WITH UNAUTHORIZED MAINTENANCE ELIMINATED
MODE
HC
CO
NO
x
SAMPLE SIZES
(l/M) (M/M)
HC
CO
NO
x
SAMPLE SIZES
(l/M) (M/M)
PRE-MAINTENANCE POST-MAINTENANCE
PRECONTROL VEHICLES
0.466 1.001
1.664 1.812
0.909 0.687
(37) (14) (37) (14)
CONTROLLED VEHICLES
0.791 1.822
1.575 2.491
0.454 1.299
(27) (45)
(27) (45)
POST-
DETERIORATION
3.395
2.155
1.096
(37) (14)
0.447
2.289
0.998
-------
The important factor for this study is the post-deterioration
emissions differences.
For precontrol vehicles, HC emissions are significantly greater
for the mandatory maintenance fleet vis-a-vis the idle inspection fleet,
but the small sample size (14) must be taken into account in analyzing
this result. CO emissions cannot be reasonably compared because of
the two different populations. NOx emissions are slightly greater
for the idle inspection procedure.
For controlled vehicles HC and NOx emissions reveal themselves
to be not statistically different to a significant extent. CO emis-
sions cannot be reasonably compared. In summary, the emissions from
the two fleets for both the post-maintenance and post-deteriorated
states are different. Part of this difference can be attributed to
the higher attrition level of lower emitting vehicles in the mandatory
fleet and to the level of maintenance received.
-------
5.0 IMPACT OF DETERIORATION ON ALTERNATIVE PROGRAMS
5.1 IDLE INSPECTION AND MAINTENANCE
This section examines the impact of engine and emission deteriora-
tion on both the idle and mandatory maintenance programs. The
evaluation provides both histograms and statistical analysis of the
developed results.
The idle inspection and maintenance procedure, as detailed in
Volume IV, consisted of an initial examination of the key modes
idle HC and idle CO. If the vehicle conformed to the prescribed
standards (see Table 5-1) it was left untouched, however, if it
failed, a systematic program of engine maintenance was performed.
This discussion focuses on an evaluation of the potential impact
of emission deterioration on the idle inspection procedure. In par-
ticular, analyses have been performed in the following areas:
• pre and post maintenance vehicles remaining after six months
s pre and post deterioration emission frequency distributions
• idle mode failure rates
e factors influencing deterioration
• levels of statistical significance
t comparison of maintenance and non-maintained fleets
A brief review of the CVS mass emission results for the idle inspec-
tion and maintenance program is given in Table 5-2 for the 87 vehicles
remaining in the idle test fleet. These results do not appear sub-
stantially different than the ones developed during the initial test
phase for the fleet of 300 vehicles. The trends reflected in the
preceeding results also hold when vehicles that have received un-
authorized maintenance have been removed from the fleet.
-------
TABLE 5-1
IDLE INSPECTION PASS/FAIL CRITERIA
Measurement Cri teria
Pre-Control1ed Control 1ed
Idle HC 800 ppm 330 ppm
Idle CO
6%
-------
TABLE 5-2
IDLE INSPECTION CVS EMISSIONS PRE AND POST MAINTENANCE
MODE
PRE-MAINTENANCE
MEAN STD.
(GR/MI) DEV.
POST-MAINTENANCE
MEAN STD.
(GR/MI) DEV.
DIFFERENCE
(PERCENT REDUCTION)
T-SCORE
HC
CO
NO
8.18
111.80
2.64
5.38
51.57
1.42
6.99
101.12
2.60
3.57
47.37
1.41
14.5
9.6
1.5
1.79
1.56
-0.24
SAMPLE SIZE = 87
Ol
I
w UNAUTHORIZED MAINTENANCE ELIMINATED
HC
7.73
4.39
6.73
2.76
12.9
1.54
CO
103.95
43.51
96.51
44.95
7.2
1.15
NO
X
2.70
1.35
2.63
1.42
2.6
0.46
SAMPLE SIZE = 64
-------
Subsequent to a series of post-maintenance tests the vehicles
were returned to their owners for a six month deterioration period.
The cars were retested within a period equal to 180 - 10 days
of the post maintenance inspection. From Table 5-3 it can be seen
that HC and CO emissions increased over the deterioration period, but
the change is statistically insignificant, as witnessed by the low
T-scores.* NOx emissions decreased slightly over the six month period,
but this decrease is also statistically insignificant.
Figures 5-1 to 5-6 illustrates the shift in emissions over the
six month period for HC, CO and NOx, respectively. While both pre-
deterioration (unhatched bars) and post deterioration (hatched bars)
results show log-normal distributions (a distribution which is skewed
to the left with an extended tail to the right), the distribution for
the post deterioration emissions is shifted to the right with respect
to the distribution of the pre deterioration emissions, thus signifying
higher average emission levels. While these frequency plots do not
"prove" the significance of observed emission deterioration they do
present an indication of general trends. It should be further noted
that as a result of the log-normal distributions of the emissions
(as opposed to a normal or "bell-shaped" distribution) a majority
of the vehicles in any one population will emit at a lower level than
the mean level for that population. This results from the skewed
shape of the distribution.
' Typically, a T-score of less than one is not considered to yield
statistically significant results (i.e., the difference between the
two samples is not real).
-------
TABLE 5-3
IDLE INSPECTION CVS EMISSIONS PRE AND POST DETERIORATION
MODE PRE-DETERIORATION
MEAN STD.
(GI^MI) DEV.
POST-DETERIORATION
MEAN STD.
(GI^MI) DEV.
DIFFERENCE
(GR/MI/ (pR/MI/
1000 Ml) MONTH)
T-SCORE
cri
cn
HC
CO
NO
6.99
101.12
2.60
3.57
47.37
1.41
SAMPLE SIZE = 87
7.50
105.79
2.54
4.07
51.93
1.40
0.13
1.16
-.01
0.09
0.79
-0.01
0.83
0.61
-0.44
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
6.73
96.51
2.63
2.76
44.95
1.42
7.49
99.74
2.57
4.33
44.95
1.42
0.19
0.80
-0.02
0.13
0.55
-0.01
0.84
0.42
-0.40
-------
cn
a\
-------
Figure 5-2 HC Histogram for Idle Inspection Pre and Post Deterioration with
-------
LO
CO
in
CO
^21.00 84.20 147.40 210-60 273.50 337.00
CVS INSPECT. SM/ML
-------
en
<*o
^0.00 73.00 116-00 159-00 202.00 245.00
CVS INSPECT. CM/MI
Figure 5-4" CO Histogram for Idle Inspection Pre and Post Deterioration with
-------
cn
i
• 8:.odi ! lb.dm
tfr.lvr
-------
Figure 5-6 NOx Histogram for Idle Inspection Pre and Post Deterioration with
-------
A further analysis was performed on the total vehicle population
for the idle inspection regime to determine the impact of deterioration
on pass/fail rates. Vehicles which passed the idle mode criteria
prior to maintenance tended to remain in a good state of repair at the
end of six months (the rejection rate of 36.8 percent is somewhat mis-
leading because of the relatively small sample size). Vehicles which
failed the idle test prior to maintenance, on the other hand, showed
a decrease in the failure rates over the deterioration period. This
apparent anomaly (one would expect the failure rate to rise over time)
can not be fully explained with existing data. Potential explanation
could include climatic changes and test procedures. These results are
summarized in Tables 5-4 and 5-5 for the total population and un-
adjusted population, respectively. The impact of removing vehicles
which had undergone unauthorized repair does not appear to have an
impact in these trends, although again the sample size for "passed
vehicles" is unfortunately small.
The results just presented, however, do not reveal the whole
picture. Because of the great disparity between emission levels
for precontrol (1967 and earlier) and controlled vehicles (1968 and
later), it was decided to separate the vehicles into these two sub-
groups. Tables 5-6 and 5-7 present CVS mass emission data for this
particular matrix. Precontrol vehicles emit significantly more
pollutants than do the control vehicles, but the evidence implies
that controlled vehicles deteriorate at a more rapid pace. In fact,
HC emission increased significantly over the six month period. It
should be noted, however, the relative small sample sizes in these
cases may tend to bias the observed results.
-------
TABLE 5-4
IDLE INSPECTION MODE FAILURE RATES FOR TOTAL POPULATION
HC
(PPM)
CO
(%)
N°x
(PPM)
FAILED
(NUMBER) (%)
PRE-MAINTENANCE 502.32
POST-MAINTENANCE 597.47
POST-SIX MONTHS 651.53
SAMPLE SIZE = 19
PASSED PRE-MAINTENANCE
3.64
2.53
3.53
111.42
112.32
82.11
0
3
7
0.0
15.8
36.8
(
FAILED PRE-MAINTENANCE
PRE-MAINTENANCE 945.29
POST-MAINTENANCE 749.88
POST-SIX MONTHS 822.72
SAMPLE SIZE = 68
6.42
4.23
4.49
82.11
85.26
98.65
68
55
45
100.0
80.9
66.2
Xa) Reflects difference between laboratory and garage procedures (i.e.,
some vehicles passing the laboratory inspection were failed and maintained
-------
TABLE 5-5
IDLE INSPECTION MODE FAILURE RATES WITH UNAUTHORIZED MAINTENANCE ELIMINATED
ur
(PPM)
CO
(%)
NOx
(PPM)
FAILED
(NUMBER) (%)
PRE-MAINTE NANCE 484.36
POST-MAINTENANCE 578.36
POST-SIX MONTHS 561.43
SAMPLE SIZE = 14
PASSED PRE-MAINTENANCE
3.52
2.36
3.78
121.64
118.71
83.57
0
1
5
0.0
7.1
35.7
(
FAILED PRE-MAINTE NANCE
PRE-MAINTE NANCE 949.56
POSt-MAINTENANCE 755.74
POST- SIX MONTHS 815.74
SAMPLE SIZE = 50
6.10
4.25
4.25
63.82
81.66
109.98
50
41
31
100.0
82.0
62.0
Xa] Reflects difference between laboratory and garage procedures (i.e., some
vehicles passing the laboratory inspection were failed and maintained at
-------
TABLE 5-6
CVS EMISSIONS JOR IDLE INSPECTION PRECONTROL VEHICLES
MODE PRE-DETERIORATION
MEAN STD.
(GR/M\) DEV.
POST-DETERIORATION
MEAN STD.
(GK/MI) DEV.
DIFFERENCE
(GIVW (GR/MI
1000 Ml) MONTH)
T-SCORE
CJ1
I
HC
CO
NO
8.70
123.70
2.25
3.70
43.55
1.40
SAMPLE SIZE = 49
8.76
129.63
2.19
2.74
52.57
1.21
0.02
1.48
-0.01
UNAUTHORIZED MAINTENANCE ELIMINATED
0.01
1.00
-0.01
0.32
0.40
-0.24
HC
CO
NO
8.11
117.44
2.38
2.46
43.24
1.57
8.41
121.49
2.30
2.63
44.04
1.32
0.08
1.00
-0.02
0.05
0.69
-0.01
0.42
0.33
-0.18
-------
TABLE 5-7
CVS EMISSIONS FOR IDLE INSPECTION CONTROLLED VEHICLES
MODE PRE-DETERIORATION
MEAN STD.
(GI^MI) DEV.
POST-DETERIORATION
MEAN STD.
(GlVMI) DEV.
DIFFERENCE
(GF^/MI (GR/hM/
1000 Ml) MONTH)
T-SCORE
HC
CO
NO,
4.79
72.01
3.04
1.74
34.76
1.30
5.88
75.04
2.99
4.89
30.96
1.52
0.27
0.76
-0.01
0.18
0.51
-0.01
1.11
0.65
-0.45
SAMPLE SIZE = 38
en
i
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
4.83 1.90 6.23
67.82 28.79 69.93
2.98 1.11 2.94
SAMPLE SIZE = 27
5.74
24.77
1.49
0.34
0.52
-0.01
0.24
0.30
-0.01
0.94
0.40
-------
Regression and Statistical Analysis
One basic,, objective of this study was in identifing explanatory factors
that might contribute to the observed rate of emission deterioration.
Some of the potential forces influencing deterioration are listed
below:
o Previous driving history (e.g., mileage accumulation)
e Climate
e Fuel characteristics
9 Post-tune-up conditioning
o Unauthorized maintenance
In an attempt to evaluate these effects, a regression analysis was
performed on the experimental data. Mileage accumulation over the
period was selected as the primary variable of interest since the
characteristics of the other listed variables posed service technical
problems. The results of this preliminary regression analysis are
given in Table 5-8- A cursory examination of these results reveals
that mileage accumulation does not explain a significant component
of the observed emission deterioration (maximum of four percent).
Clearly, other forces were at work which will be examined at the end
of the twelve month period.
While these results are not incouraging, it must be kept in mind
that the measured emission deterioration rates, for the most part,
were not statistically significant. That is, the actual rates must
be assumed to be zero for lack of the necessary statistical support.
An important question emerging from this discussion is - what level
of emission deterioration will reveal statistically significant results
over a given time period? In an attempt to answer this question an
-------
TABLE 5-8
LINEAR LEAST SQUARES REGRESSIONS FOR IDLE INSPECTION DETERIORATION CVS EMISSIONS
VEHICLE TYPE
EMISSION
TYPE
CONSTANT
COEFFICIENT
PERCENT
DETERMINED
PRECONTROLLED
HC
-0.2906
+
0.00009664
X
OD
4.1
CONTROLLED
HC
= -1.5590
+
0.0005814
X
OD
4.1
PRECONTROLLED
CO
5.798
+
0.00003711
X
OD
0.0
CONTROLLED
CO
= -3.410
+
0.001413
X
OD
0.0
PRECONTROLLED
NO
x
-0.5230
-
0.00000128
X
OD
0.0
CONTROLLED
NO
= -0.4500
+
0.00008659
X
OD
0.0
UNAUTHORIZED MAINTENANCE ELIMINATED
PRECONTROLLED
HC
0.0951
+
0.00005626
X
OD
4.0
CONTROLLED
HC
= -1.6457
+
0.00064574
X
OD
4.0
PRECONTROLLED
CO
4.3479
-
9.00007935
X
OD
0.0
CONTROLLED
CO
= -5.3702
+
0.00158867
X
OD
0.0
PRECONTROLLED
NO
x
= -0.0897
+
0.00000295
X
OD
0.0
CONTROLLED
NO
X
= -0.3441
A
0.00006464
X
OD
-------
analysis was performed to determine what emission levels would be
necessary for a significant deterioration to have occured. Further-
more, on an assumption that the deterioration was linear with time,
an estimated length of deterioration was computed for the significant
emission levels. The results of this analysis are summarized in
Table 5-9 . They show that HC mass emissions require approximately
seven months for significant deterioration and CO mass emissions
between ten and fifteen months for significant deterioration. These
observations appear consistent with the data presented in Table 5-3
(i.e., the T-score for HC (0.83) is closest to unity). Furthermore,
they represent a return to approximately eighty percent of the "as
received" state. The NOx mass emissions tend to decrease with time
and will require nearly two years to achieve a statistically signifi-
cant difference from the post-tuned condition. These post deterioration
CVS emissions were then compared with the emissions before maintenance
by means of a T-test. For precontrol vehicles, the significant post
deterioration emissions were not significantly different from the
emissions before maintenance. For controlled vehicles the evidence
is not conclusive. HC emissions were not significantly different,
and once unauthorized maintenance was eliminated,C0 emissions were
not significantly different.
Effects of Maintenance and Vehicle Characteristics on Deterioration
One key issure concerning the impact of deterioration involves
the possible influences of maintenance and vehicle characteristics.
The analysis evaluated the relationship between maintenance and deter-
ioration by dividing the idle fleet into two subgroups, one
containing those cars which had maintenance performed upon them
-------
TABLE 5-9
IDLE INSPECTION EMISSION LEVELS TO ACHIEVE SIGNIFICANT DETERIORATION
MODE
PRE-MAINTENANCE
POST-DETERIORATION*
LENGTH OF
DETERIORATION*
(MONTHS)
T-SCORE
HC
CO
NO
8.18
111.80
2.64
SAMPLE SIZE = 87
7.57
108.66
2.39
6.8
9.7
21.3
0.843
0.400
0.795
cn
i
no
O
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
7.73
103.95
2.70
7.37
104.46
2.38
5.1
14.8
25.1
0.467
0.065
0.735
SAMPLE SIZE =64
-------
(failed garage inspection), the other group containing those cars
which passed the garage inspection and thus had no maintenance per-
formed. Tables 5-10 and 5-11 show how these two groups deteriorated.
Precontrol vehicles reveal very little deterioration in either sub-
group, however, those vehicles which received maintenance showed some
deterioration of CO mass emissions. Control vehicles did go through
significant deterioration during the six month period.
Tables 5-12 through 5-15 summarize deterioration results for
the idle inspection and maintenance vehicles divided into subgroups
by control type (pre-1968, post-1967) and by engine size (less than
200 cubic inches, greater than or equal to 200 cubic inches). Because
of the small sample sizes resulting from this partitioning scheme the
deterioration factors are not statistically significant except for
HC mass emissions for controlled vehicles with large engines.
5.2 MANDATORY ENGINE MAINTENANCE
An evaluation of mandatory engine maintenance was conducted by
comparing the emissions from a set of vehicles drawn.from the general
population with the same vehicles after they had received engine
maintenance (for detailed analysis see Volume IV, Section 6.0). The
results of this analysis are summarized in Table 5-16 for the 78 cars
which went through the deterioration study and for the 59 vehicles of
this fleet that did not receive unauthorized maintenance.
After the maintenance these cars were returned to their owners
for a period of six months. The procedure was identical to the idle
test program. The vehicles were then recalled and retested. The re-
sults of these tests for the CVS mass emissions are detailed in
Table 5-17. The HC emissions showed significant increases from
-------
TABLE 5-10
ANALYSIS OF IDLE INSPECTION AND MAINTENANCE REGIME FOR TOTAL POPULATION
MODE
RECEIVED MAINTENANCE
PRECONTROL CONTROLLED
MEAN
STD.
DEV.
NO MAINTENANCE
PRECONTROL CONTROLLED
MEAN
STD.
DEV.
MEAN
STD.
DEV.
MEAN
STD.
DEV.
HC
CO
NO
10.68
137.57
2.11
5.82
44.19
0.95
7.45
97.75
2.75
PRE-MAINTENANCE
6.68
57.79
1.55
7.99
117.23
2.69
2.50
41.66
1.77
4.48
74.51
3.47
0.89
39.63
0.97
ro
ro
HC
CO
NO
9.23
128.54
1.92
4.37
45.05
0.95
4.95
70.70
2.82
POST-MAINTENANCE
2.04
32.75
1.41
7.99
117.23
2.69
2.50
41.66
1.77
4.48
74.51
3.47
0.89
39.63
0.97
HC
CO
NO
9.48
139.01
1.96
2.74
57.34
1.15
5.26
75.80
2.82
POST-DETERIORATION
2.46
30.96
1.48
7.79
117.11
2.49
2.48
43.66
1.26
7.06
73.59
3.32
7.70
32.18
1.60
SAMPLE SIZES (28)
(25)
(21)
-------
TABLE 5-11
ANALYSIS OF IDLE INSPECTION AND MAINTENANCE REGIME WITH UNAUTHORIZED MAINTENANCE ELIMINATED
MODE
HC
CO
NO
RECEIVED MAINTENANCE
PRECONTROL CONTROLLED
MEAN
10.32
129.92
2.25
STD.
DEV.
5.97
38.33
1.01
NO MAINTENANCE
PRECONTROL CONTROLLED
MEAN
6.58
86.31
2.58
STD.
DEV.
MEAN
PRE-MAINTENANCE
2.98
31.60
1.10
7.95
114.48
2.86
STD.
DEV.
2.70
44.72
1.93
MEAN
4.39
66.13
3.43
STD.
DEV.
0.91
28.84
0.81
HC
CO
NO
8.25
119.95
1.97
2.31
42.94
1.08
5.14
68.98
2.68
POST-MAINTENANCE
2.33 7.95
29.65
1.21
114.48
2.86
2.70
44.72
1.93
4.39
66.13
3.43
0.91
28.84
0.81
HC
CO
NO
9.12
129.21
1.98
2.62
40.95
1.28
5.43
71.54
2.80
POST-DETERIORATION
2.90
23.73
1.56
7.58
112.42
2.68
2.48
47.02
1.30
7.39
67.58
3.14
8.39
27.20
1.44
SAMPLE SIZES (20)
(16)
(17)
-------
TABLE 5-12
IDLE INSPECTION CVS EMISSIONS FOR PRECONTROL VEHICLES WITH SMALL ENGINES
MODE
PRE-DETERIORATION
MEAN DEV.
(GR/MI)
POST-DETERIORATION
MEAN DEV.
(GR/MI)
DIFFERENCE
(GR/MI/ (GR/MI/
1000 Ml) MONTH)
T-SCORE
cn
i
ro
HC 8.30 3.02 8.43
CO 103.84 25.70 100.18
NOx 1.66 0.56 1.89
SAMPLE SIZE = 9
3.76
35.02
0.81
0.03
-0.93
0.06
0.02
-0.62
0.04
-0.13
-0.51
0.46
UNAUTHORIZED MAINTENANCE ELIMINATED
HC 7.63
CO 97.84
NO 1.77
x
SAMPLE SIZE =
2.30 7.79
21.47 93.67
0.60 2.09
4.09 0.04
36.56 -1.04
0.82 0.08
0.03 -0.23
-0.71 -0.56
-------
TABLE 5-13
IDLE INSPECTION CVS EMISSIONS FOR PRECONTROL VEHICLES WITH LARGE ENGINES
MODE
PRE-DETERIORATION
MEAN STD.
(GR/MI) DEV.
POST-DETERIORATION
MEAN STD.
(GR/MI) DEV.
DIFFERENCE
(GR/MI (G R/Ml/
1000 Ml) MONTH)
T-SCORE
HC 8.79 3.87 8.83
CO 128.15 45.70 136.25
NOx 2.38 1.50 2.26
SAMPLE SIZE = 40
2.51
53.89
1.28
0.01
2.02
-0.03
0.01
1.36
-0.02
0.47
0.68
-0.40
c_n
I
ro
c_n
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
8.22
122.00
2.52
2.52
45.95
1.70
8.56
127.99
2.35
2.24
43.59
1.42
0.08
1.48
-0.04
0.06.
1.01
-0.03
0.69
0.64
-0.37
-------
TABLE 5-14
IDLE INSPECTION CVS EMISSIONS FOR CONTROLLED VEHICLES WITH SMALL ENGINES
MODE
PRE-DETERIORATION
MEAN STD.
(Gfy/MI) DEV.
POST-DETERIORATION
MEAN STD.
(Gfy/MI) DEV.
DIFFERENCE
(GR/M\/ (GR/M\/
1000 Ml) MONTH)
T-SCORE
cn
i
ro
HC
CO
NO
4.88 2.42 4.40
68.97 36.66 68.69
2.35 1.10 2.25
SAMPLE SIZE = 15
1.40
33.00
1.14
-0.12
-0.07
-0.02
-0.08
-0.05
-0.02
-0.65
-0.07
-0.29
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
4.80
61.48
2.54
2.58
31.00
1.05
4.23
60.74
2.42
1.43
23.04
1.13
-0.14
-0.18
-0.03
-0.10
-0.12
-0.02
-0.73
-0.08
-0.37
-------
TABLE 5-15
IDLE INSPECTION CVS EMISSIONS FOR CONTROLLED VEHICLES WITH LARGE ENGINES
MODE
PRE-DETERIORATION
MEAN STD.
(GR/M1) DEV.
POST-DETERIORATION
MEAN STD.
(GIV'MI) DEV.
DIFFERENCE
(GR/MI/ (GR/MI/
1000 Ml) MONTH)
T-SCORE
CJ1
I
ro
HC
CO
NO
4.73 1.16 6.84
73.99 34.15 79.18
3.50 1.23 3.47
SAMPLE SIZE = 23
6.04
29.55
1.56
0.53
1.30
0.01
0.36
0.87
0.00
1.74
0.78
0.42
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO.
4.86
73.71
3.39
1.03
26.34
1.03
8.08
78.46
3.43
7.50
23.95
1.66
0.80
1.18
0.01
0.55
0.80
0.01
1.73
0.54
-0.36
-------
TABLE 5-16
MANDATORY MAINTENANCE CVS EMISSIONS PRE AND POST MAINTENANCE
MODE PRE-MAINTENANCE
MEAN STD.
(CR/M\) DEV.
POST-MAINTENANCE
MEAN STD.
(GR/MI) DEV.
DIFFERENCE
(% REDUCTION)
T-SCORE
cn
i
ro
00
HC
CO
NO.
7.59
108.65
2.53
4.49
49.99
1.40
SAMPLE SIZE = 78
6.68
101.97
2.34
2.96
46.76
1.31
UNAUTHORIZED MAINTENANCE ELIMINATED
12.0
6.1
7.5
1.30
0.89
0.96
HC
7.20
4.02
6.55
2.75
9.0
0.81
CO
104.85
46.40
101.77
45.83
2.9
0.33
NO
X
2.62
1.45
2.45
1.36
6.5
0.80
-------
TABLE 5-17
MANDATORY MAINTENANCE CVS EMISSIONS PRE AND POST DETERIORATION
MODE
PRE-DETERIORATION
MEAN STD.
(GIV'MI) DEV.
POST-DETERIORATION
MEAN STD.
(GR/MI) DEV.
DIFFERENCE
(GR/MI/ (GI^MI/
1000 Ml) MONTH)
T-SCORE
c_n
i
no
UD
HC 6.68 2.96
CO 101.97 46.76
NOx 2.34 1.31
SAMPLE SIZE = 78
9.24
107.39
2.35
9.10
57.31
1.43
0.64
1.35
0.00
0.43
0.91
0.00
1.65
0.19
-0.32
UNAUTHORIZED MAINTENANCE ELIMINATED
HC
CO
NO
6.55
101.77
2.45
2.75
45.83
1.36
9.38
107.04
2.42
10.00
54.29
1.36
0.70
1.30
-0.01
0.48
0.84
-0.00
1.48
0.35
-0.30
-------
6.68 gm/roi to 9.24 gm/mt. The T-score of 1.65 is extremely strong which
indicates the statistical significance of the deterioration factors
for HC. CO emissions showed no significant increases. Reported NOx
emissions were also not statistically significant. Again, the refer-
ence point for conductiong the statistical analysis was the post-
tuned state.
These results are illustrated within the histogram frequency
plots of Figures 5-7 through 5-12. These plots reveal that both pre-
deterioration (blank bars) and post deterioration results (hatched
bars) are log-normally distributed for HC, CO, and NOx emissions. There
is a general trend of a right shift of the post deterioration distribu-
tions with respect to the pre deterioration distributions for the
three emission types. Furthermore, the statistically significant in-
crease in HC emissions is explained by the small group of vehicles
for which the HC emissions have increased dramatically after six
months (the result of incipient misfire).
Analyzing the vehicle population as a whole assumes some degree
of homogeneity. This assumption is useful, but for policy analysis
it became necessary to classify the vehicle population by age group-
ings control type.* This partitioning reaps immediate benefits in the
form of Tables 5-18 and 5-19. Precontrol vehicles tended to deter-
iorate to a much greater extent than did the controlled vehicles. HC
emissions for precontrol vehicles showed an extremely significant
increase from 9.44 gm/mi to 15.90 gm/mi. For those cars which did
* Results presented in Volume IV confirm the fact that emission
levels from pre-1968 vehicles are indeed different from post-
1967 vehicles and must be treated separately, consequently these
two classes.
-------
o
r-
Figure 5-7 HC Histogram for Mandatory Maintenance Pre and Post Deterioration
-------
en
i
CO
ro
Figure 5-8 HC Histogram for Mandatory Maintenance Pre and Post Deterioration with
-------
CO
-------
Figure 5-10 CO Histogram for Mandatory Maintenance Pre and Post Deterioration with
-------
cn
i
CO
cn
Figure 5-11 NOx Histogram for Mandatory Maintenance Pre and Post Deterioration
-------
cn
i
CO
a\
1.40 2.80
CVS MRINT
4.20 5-60
GM/M I
7 .00
Figure 5-12 NOx Histogram for Mandatory Maintenance Pre and Post Deterioration with
-------
TABLE 5-18
CVS EMISSIONS FOR MANDATORY MAINTENANCE PRECONTROL VEHICLES
MODE PRE-DETERIORATION
MEAN STD.
(GM/MI) DEV.
POST-DETERIORATION
MEAN STD.
(GM/MI) DEV.
DIFFERENCE
(GM/MI/ (GM/MI/
1000 Ml) MONTH)
T-SCORE
HC
CO
NO.
9.44
143.70
1.80
2.53
42.38
1.12
15.90
156.97
1.59
12.07
50.31
1.17
1.61
3.42
-0.05
1.09
2.30
-0.03
2.53
0.65
-0.92
SAMPLE SIZE - 23
-------
TABLE 5-19
CVS EMISSIONS FOR MANDATORY MAINTENANCE CONTROLLED VEHICLES
MODE
PRE-DETERIORATION
MEAN STD.
(GM/MI) DEV.
POST-DETERIORATION
MEAN STD.
(GM/MI) DEV.
DIFFERENCE
(GM/MI (GM/MI/
1000 Ml) MONTH)
T-SCORE
CO
00
HC
CO
NO
5.52 2.29
84.70 36.82
2.57 1.32
SAMPLE SIZE = 55
6.46
86.66
2.67
5.64
46.56
1.42
0.23
0.49
0.03
UNAUTHORIZED MAINTENANCE ELIMINATED
0.16
0.33
0.02
0.69
-0.08
0.28
HC
CO
NO
5.81 2.33 6.89
88.85 37.06 92.66
2.57 1.37 2.60
SAMPLE SIZE = 45
6.11
47.13
1.31
0.27
0.94
0.01
0.18
0.64
0.01
0.65
0.27
-------
not have unauthorized maintenance performed on them the increase was
even more dramatic from 8.42 gm/mi to 17.38 gm/mi. The T-scores are
also extremely strong at 2.53 and 1.95, respectively. The primary
causes of this increase are the high incidence of ignition mis-
fire and the relatively small sample size. HC emissions for controlled
vehicles showed increases but these were not statistically significant.
CO emission for both groups increased insignificantly and NOx emissions
wandered insignificantly.
In general, the deterioration results for the mandatory main-
tenance fleet showed greater variability expecially in the case of
HC, than for the idle inspection fleet. These trends can be directly
associated with the detection of incipient misfire in the pre controlled
group, the attrition of lower emitting vehicles and the somewhat
newer car mix. The relatively high incidence of unauthorized maintenance
associated with the mandatory maintenance fleet can not be readily
explained. Typically, all vehicles in this fleet were tuned-up
to manufacturer's specifications. One possibility may involve the
climatic impact on engine driveability. The original post maintenance
measurements were made during autumn while the six month test occured
in the spring. The reported deterioration results may reflect the
impact of a "winter" maintenance tune-up.
------- |