EPA-460/3-77-021
   December 1977
            AN EVALUATION OF
RESTORATIVE MAINTENANCE
       ON EXHAUST EMISSIONS
      OF 1975-1976 MODEL YEAR
           IN-USE AUTOMOBILES
   U.S. ENVIRONMENTAL PROTECTION AGENCY
       Office of Air and Waste Management
       Mobile Source Air Pollution Control
       Emission Control Technology Division
          Ann Arbor, Michigan 48105

-------
                        EPA-460/3-77-021
      AN EVALUATION OF
RESTORATIVE MAINTENANCE

  ON EXHAUST  EMISSIONS

   OF 1975-1976 MODEL YEAR

     IN-USE AUTOMOBILES
                  by

               Jeffrey C. Bernard
                Jane F.Pratt

               Calspan Corporation
               4455 Genesee Street
               Buffalo, N.Y. 14221
             Contract No. 68-03-2386
           EPA Project Officer: Lois A. Platte
                Prepared for

         ENVIRONMENTAL PROTECTION AGENCY
           Office of Air and Waste Management
         Office of Mobile Source Air Pollution Control
           Emission Control Technology Division
             Ann Arbor, Michigan 48105

               December 1977

-------
This report is issued by the Environmental Protection Agency to report technical data of
interest to  a limited  number  of readers. Copies are available free of charge to Federal
employees, current contractors and grantees, and  nonprofit organizations-in  limited
quantities-from the Library Services Office (MD-35), Research Triangle Park, North
Carolina 27711; or, for a fee, from the National Technical Information Service, 5285 Port
Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by Calspan Corporation,
4455 Genesee St., Buffalo, N.Y. 14221  in fulfillment of Contract No. 68-03-2386. The
contents of this report are reproduced herein as received from Calspan Corporation. The
opinions, findings, and conclusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency. Mention of company or product names is
not to be considered as an endorsement by the Environmental Protection Agency.
                        Publication No. EPA-460/3-77-021

-------
                              TABLE OF CONTENTS
Section                                                               Page


 1.0       INTRODUCTION	   1-1

           1.1   Background	   1-1

           1.2   Purpose and Design of RM Program	   1-2

 2.0       SUMMARY AND CONCLUSIONS	   2-1

 3.0       ANALYSIS OF THE PERFORMANCE OF THE EMISSION
           RELATED SYSTEMS   	   3-1

           3.1   A Discussion of Malperformance of All Vehicles
                 Taking the Initial Test by City and Manufacturer   .   3-6

           3.2   A Discussion of Malperformances for Vehicles
                 Passing the Initial Restorative Maintenance
                 Test by City and Manufacturer   	   3-10

           3.3   A Discussion of Malperforraance of Vehicles
                 Failing the Initial Restorative Maintenance
                 Test by City and Manufacturer   	   3-13

           3.4   A Comparison of Malperformance for Passed and
                 Failed Vehicles   	   3-15

           3.5   Examination of Malperformances of Passed and
                 Failed Vehicles Whose Emissions are
                 Extrapolated to 50,000 Miles  	   3-17

  4.0      EFFECT OF ADJUSTMENTS OUTSIDE OF SPECIFICATION
           TOLERANCES FOR TIMING, IDLE RPM, AND IDLE CO	   4-1

           4.1   Percentage of Vehicles Within and Outside of
                 Specification Tolerances      	   4-2

           4.2-   Correlation Between Vehicles Within and Outside
                 of Specification Tolerances and the Failure of a
                 Vehicle to Pass the FTP   	   4-3

           4.3   Degree to Which Maladjustments Affect Emissions
                 and Fuel Economy    	   4^-7

  5.0      EFFECT OF THE RESTORATIVE MAINTENANCE TESTS  1-4 ON
           EMISSIONS AND FUEL ECONOMY    	   5-1
           5.1   Empirical  Distribution  of Emissions  .  ,	5-3

           5.2   Vehicle Mean Emissions  for  Test Sequences  1-4   .  .   5-11

           5.3   Effect of  Specific Malperformances on  Emissions
                 and Fuel Economy  (Tests 5-10)    	5-13

           5.4   Vehicles Failing  the  Emissions Standards  for
                 Test  Sequences  1-4     	5-14


                                      iii

-------
                            TABLE OF CONTENTS   (cont.)
Section
Page
  6.0      VEHICLE DRIVEABILITY 	  6-1
           6.1   Driveability and Deviation from the Specification   .  6-1
           6.2   Owner-Perceived Driveability Problems  	  6-7
           6.3   Contractor-Perceived Driveability Problems   ....  6-8
           6.4   A Comparison Between Pairs of Test Sequences   . .   .  6-12
           6.5   A Comparison of Idle CO and Idle RPM Before
                 and After Adjustment   	  6-14
  7.0      A COMPARISON OF CERTIFICATION AND RESTORATIVE MAINTENANCE
           FUEL ECONOMY       	7-1
  8.0      REGRESSION ANALYSIS AND CONTINGENCY TABLES   	  8-1
           8.1   Linear Regression Analysis   	  8-1
           8.2   Contingency Table Analysis   	  8-3
APPENDIX A  -    TABLES A-l through A-103     	  A-l
APPENDIX B  -    TABLES B-l through B-35      	B-l
APPENDIX C  -    TABLES C-l through C-68      	C-l
                                      IV

-------
1.0      INTRODUCTION

         The purpose of the Restorative Maintenance (RM)  program is twofold:
to determine the apparent reasons for the poor emission performance of the
1975-1976 model year vehicles and to examine and quantify the individual
and combined effects of malperforming emission components on emissions and
fuel economy.  To this end, the analysis is performed individually for
hydrocarbons (HC),  carbon monoxide (CO) , nitrous oxides (NO ),  and for urban
                                                           A.
and highway fuel economies.  The data are analyzed separately by manufacturer
and by city as well as for all vehicles combined.

1,1      Background

         The Environmental Protection Agency (EPA) conducts annual vehicle
emission test programs, the Emission Factor Programs (EFP), for the purpose
of estimating the average emissions from a nationally representative sample
of in-use vehicles.  The emissions data are used by various Federal, State,
and local agencies  for the purpose of estimating the impact of light duty
vehicle emissions on air quality.  Results of 2 recently completed EFP indi-
cated that a large  percentage, approximately 60 percent, of the 1975 model
year vehicles in as-received condition have emissions above the 1975 Federal
Standards after only one year of use.  The 1975 model year was the first model
year with large numbers of catalyst equipped vehicles.   Similar results from
the most recent EFP indicate that approximately 55% of the 1976 model year
vehicles fail the Federal Standards after only one year of use.

         Attempts were made to determine the probable reasons for the high
failure rate of 1975 and 1976 cars using existing data for investigation.
However, the purpose and design of the EFP do not include the needed meas-
urements, emission component checks, and emission tests to precisely deter-
mine the causes of high emissions.  The RM program was specifically designed
to address the concerns about the high failure rate of the 1975 and 1976
model year vehicles.
                                     1-1

-------
 1.2      Purpose and Design of the RM Program

         There are two purposes for the Restorative Maintenance Study:

         1.   To go beyond the basic Emission Factor testing in deter-
              mination of apparent reasons for emission malperformance
              of in-use vehicles.

         2.   To investigate and quantify the individual and combined
              effects of defects, disablement or maladjustment actions
              on exhaust emissions and fuel economy.

         As a result of this program, EPA will:

         1.   Be able to assess the effectiveness of the present
              Light Duty Vehicle Certification Process in relation-
              ship to the performance of defect-free, properly tuned,
              in-use vehicles.

         2.   Provide background for planning which could result in
              further requirements for refinement of powerplants and
              emission control devices.   An example of this may be a
              mandated restriction on the adjustment of sensitive
              engine parameters such as idle mixture and basic
              ignition timing.

         3.   Generate information which can be used in planning for
              Inspection and Maintenance (I/M) programs, Selective
              Enforcement Audit (SEA) and Recall.

         This program is not expected to be able to assess who is responsible
for any maladjustments or disablements.   However, since vehicles were tested
for driveability and owners were questioned as to the maintenance practices,
the program may begin to give some insight into why a large percentage of
1975/76 vehicles are maladjusted or have emission components disabled.
                                     1-2

-------
        Three hundred vehicles were tested in the RM Program, 100 vehicles
from each of three metropolitan areas;  Chicago,  Detroit,  and Washington.
Independent testing laboratories under contract to the EPA performed the
testing.  Three major domestic automobile manufacturers were represented
equally at each city location.  Sales-weighting techniques were  used to
specify the models and engines to be evaluated.   Vehicles  from the 1975
and 1976 model years were selected from the general public at random with
the requirement that they were less than twelve months old and had accumulated
fewer than 15,000 miles.  In addition, the owners were asked questions to
preclude vehicles which had been abused or extensively modified  and to
ascertain how the vehicle had been used and maintained and how well the
vehicle performed.

         Once accepted into the program, a varying number of tests were
performed on each vehicle according to the test plan.   Each of the tests was
separated from the following test by a decision point and an appropriate
action.  Individual test sequences consisted of a 1975 Federal Test Procedure
(FTP) followed by a Highway Fuel Economy Test (HFET) and five short cycle
tests.  This 1975 FTP was modified to exclude the evaporative emissions
and the extensive preconditioning procedures used in certification of the
vehicle.  The short cycles were ones which are currently being employed or
considered for I/M programs by a number of state and local agencies.  The
contractor also evaluated the driveability of each vehicle as part of each
test sequence.

         A varying number of test sequences were performed on each vehicle,
depending upon whether the vehicle failed the FTP on the preceding sequence
and whether it required correction of a malperforming emission control item
or scheduled maintenance.  The full test sequence consisted of four steps:
an initial test sequence, a sequence following correction of maladjustments
and disablements other than idle mixture and idle speed, a test sequence
after these idle settings were readjusted, and a fourth sequence after the
restoration of all emission control components in conjunction with a
complete tune-up.
                                      1-3

-------
         Certain test vehicles were then subjected to further test sequences,
Each vehicle selected at this point in the program had met the FTP standards
Most had received a complete tune-up, although some were accepted for addi-
tional testing after a successful emission component inspection.   The
vehicles were then subjected to "selective maladjustments" where a single
engine parameter, e.g., ignition timing, or a specified combination of
parameters was maladjusted or disabled.   Table A-102 provides a flow chart
and narrative of the Restorative Maintenance Program test plan.
                                     1-4

-------
2.0      SUMMARY AND CONCLUSIONS

         The following results have been obtained by analysis  of the
Restorative Maintenace (RM)  program data:

         1.   For the 300 vehicles tested,  74% have at least one
              malperformance of an emissions related component
              or system.

         2.   Chrysler vehicles have the largest percentage, 96%,
              of at least one malperformance and 94% of all
              Chrysler vehicles have a malperformance of the
              carburetor/fuel system.

         3.   Of the nine emission related systems investigated,
              the carburetor/fuel system contributes the largest
              percentage, 66%, of malperformances.

         4.   The emissions  related components of the carburetor/fuel
              system with the largest percentages of malperformance
              are:  disabled limiter caps,  maladjusted idle mixture
              screws, maladjusted idle speed settings, and maladjusted
              choke assemblies.

         5.   Certain combinations of malperforming components,
              particularly within the carburetor/fuel and ignition
              systems, correlate with vehicles failing the standards,
              although the exact relationship between combinations
              of malperforming emission components and their additive
              or multiplicative effect upon emissions is not yet known.

         6.   Seventy-two percent of the 300 vehicles were outside
              at least one specification tolerance for either idle RPM,
              idle CO or timing, and 93% of all Chrysler vehicles were
              outside of at  least one specification tolerance.
                                     2-1

-------
      Seventy-six percent of all Chrysler vehicles were
      outside of the idle CO specification (that is,
      had tailpipe idle CO greater than .5%).

 7.   General Motors vehicles with tailpipe idle CO
      greater than .5% correlate with the failure of a
      GM vehicle to meet the CO standards 90% of the time.
      The same is true for Chrysler vehicles  74%  of the
      time and for Ford vehicles only 44% of the time.

 8.   It appears that disablement of the EGR valve or lines
      strongly correlates with the failure to pass NOX
      standards.

 9.   A significant change in emissions levels due to adjust-
      ment or maladjustment of emission components outside
      their specification tolerances is not necessarily
      accompanied by a significant change in fuel economy.

10.   Adjustment of the vehicle within accepted specification
      tolerances does not imply acceptable driveability quality.

11.   Disablement and maladjustment of any emission components
      thought to be typical for a certain type of vehicle
      almost always resulted in the failure of a vehicle to
      meet the standards.

12.   The overall ability of the short cycle  tests to pass
      or fail a vehicle as compared to the FTP is best for
      the Federal Short Cycle Test.

13.   Investigation of the distribution of emissions shows that
      they are log-normally distributed as in Figure 2-1 ,
      following:
                            2-2

-------
                            emission  standard
                        emissions

             Figure 2-1  LOG-NORMAL EMISSIONS
The median measurement (the 50th percent!le) of a log-normal
distribution is equal to the geometric mean, exp C-—^") , of the
measurements.  A set of measurements whose distribution follows
the log-normal will have an arithmetic mean that is greater than
the median (or geometric mean).  The arithmetic mean emission
value is used in air-quality projections.  The log normal distri-
bution is used in the prediction of percent of vehicles failing
standards.
(14)
Investigation of the distribution of emissions for
vehicles with tailpipe idle CO less than or equal to
.5% and for vehicles with tailpipe idle CO greater
than .5% shows that, for the most part, vehicles with
high tailpipe idle CO correlate with vehicles failing
the standards, and the vehicles with less than or equal
to .5% correlate with vehicles passing the standards as
demonstrated in Figure 2-2 below.
         (^-emission standard
                    vehicles within specifications
                            vehicles outside of specifications
                        emissions
  Figure  2-2    EMISSIONS  WITHIN AND OUTSIDE OF SPECIFICATIONS
                             2-3

-------
      There is little doubt that vehicles with high tailpipe idle
      CO (or vehicles  outside  any of  the  specifications  for  idle
      RPM and timing)  contribute to the  log-normality  of the
      distribution  of emissions  for all vehicles,  although it
      cannot be ascertained if vehicles  outside of specifica-
      tions contribute exclusively  to the log-normality  of
      the entire distribution.  Whereas,  the effect of being
      outside of just the idle CO specification on emissions
      was determined,  the interrelationships between idle CO,
      idle RPM and  timing, and their  combined effect upon
      emissions as  the three vary,  cannot be determined,
      although the  implication is that they vary multiplicatively.

CIS)   The interrelationships between  malperforming emission
      components and  their effect on  emissions was strikingly
      highlighted by  investigation  of the emission behavior
      of one vehicle:  a 1976 GM Seville.   When the Seville
      was tuned to  manufacturer's specifications,  it passed
      all FTP standards.   When several components  were inten-
      tionally maladjusted (i.e., plugging the EGR line,
      disabling the air pump,  supplying  full vacuum advance
      to the distributor and advancing the timing  to +15  degrees),
      the Seville failed the FTP only because of high  NOX
      emissions.  After the Seville was restored to manufacturers'
      specifications  and again passed the FTP standards,  only
      the EGR valve was disabled.  The result was  that the
      Seville failed  the FTP because  of high NOX and high CO
      emissions.  Although this  is the result for  investigation
      of one vehicle,  it  does  demonstrate the tendency noted
      throughout  the RM program;  that  combinations  of  mal-
      perforraances, whether disablements,  defects  of maladjust-
      ments,  and  combinations  of varying  degrees or deviations
      from all  specifications, can result  in increases in
      emissions  that may  be different  than the additive  effects
      of individual malperformances or deviations  from specifications,
                            2-4

-------
3.0      ANALYSIS OF THE PERFORMANCE OF THE EMISSION RELATED SYSTEMS

         The percent of emission component and/or system malperformances
given in this report are slightly inaccurate due to a number of minor changes
made to the data.  These changes were made after the analysis given in this
report was completed.  The changes that have occurred usually were the result
of a manufacturer representative's suggestion or clarification concerning
emission component functions.  For example, the manufacturer representative
may have pointed out that a particular component was able to function but did
not function when the vehicle was tested due to a malperformance in a distinct
although associated component.  Most cases such as this were caught early in
the program but some further problems were found during more extensive review.
In no case will the rate of malperformance given in this report deviate from
the correct rate of malperformance by more than two percentage points.  There-
fore, the conclusions given in this report regarding emission component mal-
performance are still valid.  The report was not redone because the small
error involved did not warrant the amount of work, cost, and time that would
be required to update the rates of malperformances given throughout the report.
The emission measurements were not affected by these changes and are accurate
as given.

         The focus of the following analysis will be the performance of each
emission related system and each system component or subsystem.  The purpose
is to investigate the emission systems and subsystems which do not perform
properly, to determine the frequencies or rates of malperformance for these
systems, to define types of malperformances and to delineate the specific
reasons for malperformance.

         This analysis is conducted on all three hundred vehicles after they
complete their first test sequence in the Restorative Maintenance  (RM) program.
The results of this section of the analysis are embodied in Tables A-l through
A-100 in Appendix A as well as in summary tables in the text.  Included are
analyses by each major emission related system, by each component or subsystem,
by city, by manufacturer, by vehicles passing the initial test, by vehicles
                                     3-1

-------
failing the initial test,  and by vehicles'whose emissions are extrapolated
to 50,000 miles.  Possible relationships between malper-formances ,  vehicle
mileage, and cubic inch displacement were investigated and are reported
wherever significant.

         Nine major emissions related systems were examined for malperform-
ance of their subsystems or components.   The following list displays the
nine systems and the components that were investigated in each system.
  Emission Related Systems

  Induction System
Emission Components for
Given System

Heated Air Inlet Door
Heated Air Inlet Diaphragm
Temperature Sensors, Switches,
  Modulators
Delay Valve
Air Filter Element
Hoses, Tubes, Lines, Wires
  Carburetor/Choke/Exhaust Heat Control
    Valve System
       Carburetor Subsystem
       Choke Subsystem
       Exhaust Heat Control
         Valve Subsystem 	
Carburetor Assembly
Limiter Caps
Tailpipe ICO
Idle Speed
External Idle Enrichment
Idle Stop Solenoid
Dashpot and Other Throttle
  Modulators
Fuel Filter Element
Hoses, Lines, Wires

Choke Adjustment
Vacuum Diaphragm
Electrical Controls
Hoses, Lines, Wires

Exhaust Heat Control Valve
  Assembly
Actuating Diaphragm
Coolant Temperature Sensing
  Switches
Check Valve
Hoses, Lines, Wires
  Ignition System
                                  3-2
Distributor Assembly
Initial Timing
Spark Plugs and Their Wires
Vacuum Advance Diaphragm
Spark Delay Devices
Coolant Temperature Sensing
  Switches
Hoses, Lines, Wires
Dwell

-------
  EGR System
EGR Valve Assembly
EGR Valve Backpressure
  Transducer
EGR Time Delay Solenoid
Venturi Vacuum Amplifier
High Speed Modulator
Vacuum Reservoir
Coolant Temperature Sensing
  Switches
Hoses, Lines, Wires
  Air Pump  System
Air Pump Assembly
Bypass and/or Dump Valves
Check Valve
Electrical PVS
Solenoid Vacuum Valve
Floor Pan Switch
Vacuum Differential Control
Drive Belt, Attaching Hardware
Hoses, Lines, Wires
  PCV System
PCV Valve Assembly
Filters
Hoses, Lines, Wires
   Exhaust System
Exhaust Manifold,  Tailpipe,
Muffler Catalyst
   Evap Control System
 Evap  Canister
 Canister  Filter
 Hoses,  Lines,  Wires
   Engine Assembly/Miscellaneous
 Engine  Assembly
 Engine  Oil and Filter
 Cooling System
 Mechanical Valve Adjustment
 Carburetor and Intake
   Manifold Mounting Bolts
 Belt Tensions
 Hoses,  Lines, Wires
         Tables A-l through A-100 present the percent of vehicles  with  each
type of performance for each subsystem of each major emission related system
by city and manufacturer.   The performance of each system or component  in its
as-received condition is defined by one of 8 performance codes which are
defined on each of the Tables A-l through A-100.   The performance  codes are
as follows:
                                     3-3

-------
              1  -  no malperformance
              3  -  not applicable to particular vehicle
              4  -  maladjusted
              5  -  disabled
              6  -  defective
              7  "  inadequate or improper maintenance
              8  -  improper part - misbuild
              9  -  failure of non-OEM part

         The performance codes used for components and systems in this program
were determined in accordance with the following reasoning:

No Malperformance:  The component or system was present, inspected and found
         to be operating properly.  This code was also used in cases where
         the component or system was not able to be inspected, but where there
         was no evidence that it was not operating properly.   An example
         of this is mechanical valve adjustment on a vehicle  which passed
         early in the sequence and was released without an actual
         inspection.

Maladjusted:        This refers to an adjustable component or system which
         was found to be outside of the tolerance band around the nominal
         specification.   Examples are idle speed, basic timing, and choke
         settings.   Acceptable ranges for the idle speed were ±100 rpm
         while +2° was used for basic timing.  Allowable ranges for choke
         adjustments  were the production tolerances as provided by the
         manufacturer's  representative.
                    Solely for the purpose of coding and analysis in this
         program,  as-received idle mixture adjustment was judged on the
         basis of a 0.5% tailpipe idle CO cutpoint.  This treatment had
         no impact on the actual vehicle testing which was performed
         according to manufacturers'  specifications but is useful in making
         comparisons  among the various vehicles and in the evaluation of a
         basic idle mode short test.
                                     3-4

-------
Disabled:    A component or system which is found not  to be  functioning
       properly due to some person's willful or inadvertent  action.  Examples
       are plugged, disconnected,  or rerouted vacuum lines,  carefully
       damaged EGR valves, and broken or missing limiter  caps.

Defective:   A component which is  found not to be functioning properly due to
       a manufacturing fault or normal deterioration prior to any service
       interval.  Examples of these are leaking vacuum diaphragms,  coolant
       temperature sensing vacuum  switches which do not open or close at
       appropriate temperatures, timing devices which stay on or off too
       long or too short, and broken EGR backpressure transducers.
             This code is also used when the condition of the component  or
       system cannot be absolutely determined by the basic functional checks
       prescribed in the program but a replacement and a subsequent emission
       test reveals a significant  difference in emission levels.  This was
       the case where carburetor replacements corrected a high CO problem.

Failure Due to Inadequate or Improper Maintenance:     A component or system
       which is not functioning properly due to the owner's neglect.
       Examples of this are a dirty air cleaner, or lack of spark plug
       change at a specified time.  This code is only used in those cases
       where the condition was determined to have a significant effect on
       exhaust emission levels.

Improper Part Due to Misbuild:    Lacking any firm evidence of replacement
       after production, this is the determination that the component present
       was not the correct one for the engine family/emission control system
       applicable to the test vehicle.  An example of this is an instance
       in which the test vehicle was equipped with non-resistor spark plugs
       when resistor type are specified.

Failure of Non-OEM Part:    A failed component which is not an exact replace-
       ment of original equipment.  An example of this  is an after-market
       brand of spark plug which has fouled.  Normally, however, such
       components which were found  to be operating properly received a "pass"
       rating.
                                      3-5

-------
 3.1      A DISCUSSION OF MALPERFORMANCE OF ALL  VEHICLES TAKING THE
         INITIAL TEST BY CITY AND MANUFACTURER
         The  rates of malperformances  given  in  this section are expressed
 as  a percent  of  the total number of vehicles being considered, not as a
 percent of the total number of vehicles that are equipped with a given
 component.  The  rates of malperformance will be expressed in this way
 throughout this  report unless it is stated otherwise.

         Of the  nine emission related  systems investigated, the carburetor/
 fuel system contributes the largest percentage  of malperformances, 66%, of
 any major system, followed by the ignition system, 26%, the exhaust gas
 recirculation system, 15%, and the induction system, 6%.  All remaining
 systems have  less than a 2% level of malperformance as indicated by Tables
 III-l and III-2.  For all three hundred vehicles tested, 74% have at least
 one malperformance.

         Analysis of malperformance by city indicates no relationship between
 the two, but analysis of malperformance by manufacturer, Table III-2, indi-
 cates that Chrysler vehicles have the  largest percentage of malperformance
 as  compared to General Motors and Ford.  For the carburetor/fuel system,
 Chrysler has a 94% rate of malperformance as compared to 56% for Ford and 49%
 for General Motors.  For the ignition  system, Chrysler has a 32% rate of
 malperformance as compared to 25% for  Ford and 21% for General Motors.   For
 the exhaust gas  recirculation system, Chrysler and Ford have about the same
 rate of malperformance,  19% and 18% respectively, with General Motors at 9%.
 Ford and Chrysler vehicles have the largest rates of malperformance with 9%
 for the induction system,  followed by General Motors with about 2%.  Overall,
Chrysler has a 96% rate  of at least one malperformance followed by Ford with
 69% and General Motors with 59%.   Whereas the carburetor/fuel system is
undoubtedly the biggest  contributor to malperformance for all vehicles,
this system's  malperformance is especially significant for Chrysler vehicles.
                                      3-6

-------
         Tables A-l through A-18 present the percent  of vehicles  with  each
type of performance for each component or subsystem of each major emission
related system.  In each table,  all the codes for each type of performance
are presented for completeness whereas only codes 4 through 9, inclusive,
are considered a malperformance.  The reader should be informed that not
every manufacturer employs every subsystem or component indicated in the
tables.  Therefore, in assessing the percentage of vehicles with  a particular
malperformance for a component or subsystem, one must check to see if  all
vehicles are equipped with the component.  Code 3 of the performance codes
in each table indicates that the vehicle is not equipped with the subsystem
or component indicated.  For instance, the external idle enrichment listed
in Table A-4 for the carburetor/fuel system does not  apply to any of the
General Motors or Ford vehicles and does not apply to 41.4% of the 99  Chrysler
vehicles.  That is, only 58 of the 99 Chrysler vehicles employ external idle
enrichment and 55 of the 58 have no malperformance.  One of the 58 vehicles
has a disabled idle enrichment and two of the 58 vehicles have a defective
idle enrichment.

         Analysis of the induction system, Tables A-l through A-2, indicates
that most of the malperformances, 4%, were due to disablement of hoses,
tubes, and wires.  Table III-3 is a summary of the significant systems and
subsystems contributing to malperformance.

         Analysis of the carburetor/fuel system, Tables A-3 and A-4,  indi-
cates that the components with the largest percentage of malperformances
are the limiter caps, the idle mixture adjustment, the idle speed, and the
choke adjustment.  The limiter caps were disabled on 45% of the vehicles,
the idle mixture was maladjusted on 38% of the vehicles, the idle speed was
maladjusted on 25% of the vehicles, and the choke was maladjusted on  10%  of
the vehicles.  There were very few defective components in the carburetor/
fuel system and these were scattered over 6 of the 16 remaining subsystems.

         Further analysis by manufacturer, Table A-4, reveals that limiter
caps were disabled on  70% of all Chrysler vehicles as compared to 36% for
Ford and 30% for General Motors.  The idle mixture adjustment was maladjusted
                                      3-7

-------
 on  71%  of  all  Chrysler vehicles  as  compared  to  15%  for  Ford and  27%  for
 General Motors.  The  idle  speed  was maladjusted on  31%  of  all Chryslers
 as  compared  to 24%  for Ford and  19% for General  Motors.  There seems  little
 doubt that the high malperformance  rate for  Chrysler is a  result of  the
 large number of maladjusted idle mixtures  and idle  speeds.

         Tables  A-5 and A-6 present results  of  the  ignition system by city
 and manufacturer.   These tables  indicate that the initial  timing was mal-
 adjusted on  19%  of  all vehicles.  Washington had a  slightly higher rate
 with 26% of  Washington vehicles  having maladjusted  timing.  Approximately
 19% of  the General  Motors, Ford, and Chrysler vehicles had maladjusted timing.

         Tables  A-7 and A-8 present results  of  the  exhaust gas recirculation
 (EGR) system and indicate  that 14 of the 68  vehicles, or 21% (mostly Fords),
 equipped with  an EGR valve backpressure transducer  were defective.  Also 2
 of  the  40 Chrysler  vehicles, or  5%, equipped with an EGR time delay solenoid
 were defective.  Approximately 8% of the Chrysler vehicles were found with a
 disabled EGR valve.  There were  no General Motors vehicles equipped with
 disabled or  defective EGR valves and only  2  Chryslers and  one Ford were
 equipped with  a  defective EGR valve.

         Analysis of combinations of malperforming  emissions related systems
 was performed  and the results may be noted in Tables A-19  and A-20.  Note
 that the analysis determines how many vehicles have a malperformance in two
 different systems simultaneously.  The largest frequency of malperformance
 for combinations of systems occurs between the carburetor/fuel system and
 the ignition system with 65 of 300 vehicles  having both malperforming
 carburetor/fuel  systems and ignition systems.  The next largest frequency
 of malperformance,  35 out of 300 vehicles, for a combination of systems occurs
between  the carburetor/fuel system and the exhaust gas recirculation system.
 The ignition and exhaust gas recirculation systems and the induction and
 carburetor/fuel  systems have 16 of 300 vehicles  and 15 of  300 vehicles with
malperformances  in both system combinations, respectively.   The implication
 of this  analysis is that almost all, 65 of the 79, vehicles with ignition
                                     3-8

-------
system malperforraances also have carburetor/fuel malperformances.   Also,
35 of the 46 vehicles with exhaust gas recirculation malperformances also
have carburetor/fuel malperformances.   Only 16 of the 46 vehicles  with
exhaust gas recirculation malperformances also have ignition malperformances,
but 15 of the 19 vehicles with induction system malperformances also have
carburetor/fuel system malperformances.   The conclusion is that the vehicles
with either ignition, exhaust gas recirculation, or induction system mal-
performances, most probably also have carburetor/fuel system malperforraances.

         To further clarify which combinations of components or sub-
systems result in malperformances, Table III-4 is offered only for the
significant combinations of components or subsystems for maladjusted and
disabled components for all vehicles.

         Table III-4 indicates that 93 of the 300 vehicles or 31% have both
disabled limiter caps and maladjusted idle mixtures.  Presented another way,
93 of the 113 vehicles (82%) with maladjusted idle mixtures also have dis-
abled limiter caps.  Thirteen percent or 39 of 300 vehicles have both
disabled limiter caps and maladjusted idle speeds.  Ten percent of all
vehicles or 31 of 300 vehicles have both maladjusted idle mixtures and
maladjusted idle speeds.  Also, 8% or 23 of 300 vehicles have both mal-
adjusted chokes and idle mixtures.  Six percent or 19 of 300 vehicles have
both disabled limiter caps and maladjusted chokes.

         Comparisons of malperformance for both the carburetor/fuel and
ignition systems show that 33 of 300 vehicles, or 11%, of all vehicles have
both disabled limiter caps and maladjusted timing.  Nine percent or 26 of
300 vehicles have both maladjusted idle mixtures and timing, and 6% or 18
of 300 vehicles have both maladjusted idle speed and timing.

         Comparisons of malperformances for both the carburetor/fuel and
EGR systems show that 9 of 300 vehicles have both disabled  limiter caps
and disabled or defective EGR valves.  Also, 9 of 300 vehicles have both
maladjusted idle mixtures and disabled or defective EGR valves.
                                     3-9

-------
         Comparisons  of malperformances  for both the  carburetor/fuel and
 induction  systems  show that  8  of  300 vehicles have both  disabled hoses,
 tubes  and  wires, and disabled  limiter caps.

         The above results confirm the  interdependency of the subsystem of
 the  ignition, EGR  and induction systems with the subsystems or components
 of the carburetor/fuel system.  Thus, not only have  the  major emission
 related systems producing malperformances been reduced to the carburetor/
 fuel,  ignition, EGR  and induction systems, but the components or subsystems
 within each major  system that  produce the majority of the malperformances
 have been  defined.

         The specific reasons  for component/subsystem malperformance are
 listed in  Table A-101.  The  table also  indicates the frequency of occurrence
 of the  various causes of the component  or subsystem malperformances.

 3.2      A DISCUSSION OF MALPERFORMANCES FOR VEHICLES PASSING THE INITIAL
         RESTORATIVE MAINTENANCE TEST BY CITY AND MANUFACTURER

         Of the 300 vehicles that took  test 1, the as-received test, of the
 RM program, only 125  or 41.7%  passed all three emissions standards. Any  emission
 values  less than or equal to 1.5 gin/mi. HC, 15 gm/mi. CO, and 3.1 gm/mi.  NOX
were called passing vehicles in this report.  When certified, the 1975 and
 1976 model year vehicles were  determined to pass if their emissions were
 less than  1.55 gm/mi. HC, 15.5 gm/mi. CO, and 3.15 gm/mi. NOX.  Therefore,
the passing rates given in this report may be slightly lower than those that
would result from using the  cutpoints as used in the certification procedure.
The small  difference  in passing rates will not alter the conclusions of this
report.  It is the purpose of  this section to explore the relationship
between vehicles with emission component malperformances and vehicles that
passed the emissions standards.  Vehicles that passed the standards are not
necessarily free of emission component malperformances.

         The effect of an individual emission component  or system malperform-
 ance on emission levels and FTP failure rates cannot be estimated from the
                                    3-10

-------
results given in this section.   Further on,  in Section 5.3,  there is  some
discussion of individual malperformances on  emission levels.   The vehicles
that have malperformances in a particular component or system may also have
malperformances in other systems.   Because of the multiple system and/or
component malperformances, it is not possible to estimate the effect  of an
individual system malperformance on emissions with the results of this section.
The results given here are an estimate of the combined effect of malperform-
ances on emissions and failure rate.

         Tables III-5 and III-6 present the  percent of malperformance by city
and manufacturer, respectively, for vehicles that passed the initial  test.
The carburetor/fuel system has the largest rate of malperformance, 41%, for
passed vehicles, followed by the ignition system with 13%, the induction
system with 6% and the exhaust gas recirculation system with 4%.  All
remaining systems have a malperformance rate less than 1%.  For all 125
vehicles that passed the initial test, 50% have at least one ma1performance.
For three of the four systems accounting for the majority of malperformances,
the percentage of vehicles passing the initial test with a malperformance is
significantly less as compared to the percentage of vehicles with a malper-
formance for all vehicles.  Only for the induction system does the percentage
of vehicles with a malperformance remain the same at 6%.

         Table III-5 reveals that about the same number of vehicles pass  the
initial test in each of the three cities.  Also, about the same percentage
of vehicles have the same rate of malperformance in each emission related
system for each city.  Table III-6, however, reveals that the number of
vehicles passing the initial test by manufacturer is greatly different for
Chrysler vehicles with  17 passing than for either General Motors or Ford,
each with 51 and 57 passing, respectively.  Of significant importance is
that, although only 41% of all passed vehicles have a malperformance for  the
carburetor/fuel system, 88% of Chrysler vehicles have a carburetor/fuel system
malperformance.  Only 44% and 22% of Ford and General Motors vehicles
respectively have a carburetor/fuel system malperformance.
                                    3-11

-------
          A comparison of these carburetor/fuel system malperformance
 percentages for vehicles that passed test 1 with the percentages for all
 vehicles taking the initial test reveals that Chrysler vehicles have about
 the same rate of malperformance, with Ford and General Motors vehicles
 having a much smaller rate of malperformance for passed vehicles.
 Examination of passed vehicles, with at least one malperformance by manu-
 facturer, reveals that Chrysler vehicles have an 88% rate of at least one
 malperformance as compared to 56% and 31% for Ford and General Motors,
 respectively.

          Review of the individual subsystems within each of four major
emission  related  systems producing malperformances, see Tables A-21 through
A-32, shows that  the  following  subsystems or components contribute the
following rates of malperformance for the 125 passed vehicles:  19% with
disabled  limiter  caps, 9% with  maladjusted  idle mixtures, 18% with mal-
adjusted  idle speeds,  7% with maladjusted chokes,  10% with maladjusted
timing, 0% with disabled or defective EGR valves,  3% with a defective EGR
valve transducer, and 4% with disabled hoses, tubes and wires related to
the induction system.  The rates of malperformances for subsystems shows
that these rates  are  less for passed vehicles in their as-received condition
as compared to the rates for all vehicles in their as-received condition.
Tables A-21 through A-38 present the performance codes for all subsystems of
the major systems for passed vehicles.

         An investigation of which combinations of systems result in mal-
performance is displayed in Tables A-39 by  city and A-40 by manufacturer.
Nine of the 125 passed vehicles, or 7%, have malperformances in both the
ignition and carburetor/fuel systems.  Six  of the  125 passed vehicles, or 5%,
have malperformances  in both the induction  and carburetor/fuel systems.
Only 2 of the 125 passed vehicles, or 2%, have malperformances in both the
exhaust gas recirculation and carburetor/fuel systems.  The result is that
there is a very small correlation between major emission systems for passed
vehicles with malperformances.  Before making too general a statement, the
rates of malperformance for vehicles failing the initial RM test must be
examined. Rates of malperformance for failed vehicles will be discussed in
the next section.
                                     3-12

-------
         Table III-7 is a summary of the significant  systems  and  subsystems
contributing to malperformances for vehicles passing  the initial  test  by
manufacturer.

3.3      A DISCUSSION OF MALPERFORMANCE OF VEHICLES FAILING THE  INITIAL
         RESTORATIVE MAINTENANCE TEST BY CITY AND MANUFACTURER

         Of the 300 vehicles that took test 1, the as-received test,  of the  RM
program, 175 or 58.3% of all vehicles failed one or more of the  emissions
standards for hydrocarbons,  carbon monoxide, and nitrous oxides.  This  section
will investigate the rate of emission component malperformance for vehicles
failing the initial test to  determine if vehicles failing the initial  test
necessarily have a high rate of malperformance.  Tables II1-8 and III-9 present
the rate of malperformance for all failed vehicles by city and manufacturer.
The carburetor/fuel system has the largest rate of malperformance with 84%,
followed by the ignition system with 36%, the exhaust gas recirculation
system with 23%, and the induction system with 6%.  For the 175 vehicles
failing the initial test, 91% have at least one malperformance.

         Table III-8 indicates that about the same number of vehicles fail the
initial test in each city location.  Also, for each particular emission related
system, the rate of malperformance is approximately the same from city to city.
Examination of the rates of malperformance by manufacturer, Table III-9, shows
that Chrysler has the largest rate of at least one malperformance, with 98%,
followed by General Motors and Ford, each with approximately 86%.  Chrysler
vehicles also have the highest rate, 95%, of malperformance in the carburetor/
fuel system as compared to General Motors with 76% and Ford with 71%.   There
is little difference among manufacturers in the rate of malperformance for each
of the remaining emission related systems examined individually.

         A comparison of the rates of malperformance for the 175 failed
vehicles with the rates of malperformance for all 300 vehicles indicates
higher rates of malperformance for failed vehicles for the carburetor/fue1,
the ignition and the exhaust gas recirculation systems.  There is no differ-
ence in the rate of malperformance for failed vehicles as compared to all
vehicles for the induction system.  There are other emission related systems
which show higher rates of malperformance for failed vehicles. However, the
                                    3-13

-------
 rates of malperformance for these systems, the air pump, positive crankcase
 ventilation, exhaust, evaporative and engine assembly systems are 2% or less.
          Examination of the malperformances for the significant subsystems for
 failed vehicles, Tables A-41 through A-58, reveals the following rates and
 types of malperformance:   64% with disabled limiter caps, 58% with maladjusted
 idle mixtures, 30% with maladjusted idle speeds, 12% with maladjusted chokes,
 26% with maladjusted timing, 6% with either a defective or disabled EGR valve,
 6% with a defective EGR valve transducer, 11% with disabled EGR system hoses,
 lines and wires, and 4% with disabled induction system hoses, lines and wires.
 The rates and types of subsystem malperformance are greater for failed vehicles
 as compared to the rates and types of malperformances for all vehicles taking
 test 1.
          Tables  A-59  and A-60 present the  frequencies  of  malperformance  for
 combinations  of  emission related  systems by  city and manufacturer,  respectively.
 Fifty-six of  the 175  failed  vehicles, or  32%, have both  carburetor/fuel  and
 ignition  system  malperformances.  Thirty-three  of the  175  failed vehicles, or
 19%,  have  both carburetor/fuel and exhaust gas  recirculation  system malperform-
 ances.  Sixteen of the  175 failed  vehicles, or 9%, have both ignition and exhaust
 gas  recirculation  system malperformances   and 9 of  175, or 5%, have both
 induction  system and  carburetor/fuel system malperformances.  The result is
 that  56 of the 63  vehicles with ignition system malperformances also have
 carburetor/fuel  system malperformances.  Thirty-three  of  the  41 vehicles with
 exhaust gas recirculation malperformances also have carburetor/fuel system
 malperformances.   Nine of the 11  vehicles with  induction  system malperformances
 also  have  carburetor/fuel system  malperformances.  Only  16 of the 41 vehicles
 with  exhaust gas  recirculation malperformances also have  ignition system
 malperformances.   One conclusion  that may be made is that a failed vehicle
with  a malperformance in any or all of the following systems:  the ignition,
 exhaust gas recirculation or induction systems, has at least an 80% chance of
 a malperformance in the carburetor/fuel system.  Another, more obvious
 conclusion is that the carburetor/fuel system, either alone or in combination
with  other systems, contributes the largest rate of malperformance of any major
 emission related system for all manufacturers in all cities for vehicles
 failing the initial test.

         Table 111-10 presents a summary of the significant systems and sub-
 systems contributing to malperformances for vehicles failing the initial test
by manufacturer.
                                     3-14

-------
3.4      A COMPARISON OF MALPERFORMANCE FOR PASSED AND FAILED VEHICLES

         The section investigates the relationship between the rate of raal-
performance and whether a vehicle will pass or fail the emissions standards.
One-hundred and twenty-five vehicles pass the initial test and 175 vehicles
fail the initial test.   Let us define, m ,  as the number of emission component
malperformances for vehicles passing the initial test, and, mc, as the
                                                             r
number of such malperformances for vehicles failing the initial test.   Then,
(m /125) times 100% and (m /175)  times 100% would be the percentages or
rates of malperformance for passed and failed vehicles, respectively.

         Consider the case where   m  = 0 and  m^ = 175.  This case would
                                   P           F
 imply that (m /125) times 100% equals 0% and (m /175) times 100% equals 100%.
 This would mean that all vehicles passing  the emissions standards would be
 free of malperformances and that all vehicles failing the emissions standards
 would all have malperformances.   Thus, the statistic defined by

              (mF/175 - m /125) times 100%

 would equal 100% and all malperforming vehicles could be said to positively
 correlate with all vehicles failing the initial test.

         Next, consider the situation where  m  = 125  and  mp = 0.  Then
 the statistic  (m /175 - m /125) times 100% would equal -100% and all mal-
 performing vehicles could be said to negatively correlate with all vehicles
 passing the initial test.

         If no correlation existed between malperforming vehicles and vehicles
 that passed or failed the test,  then  m /175 would equal  m /125 and the
 statistic  (m /175 - m /125) times 100% would be zero.  Table III-ll presents
 a summary of the statistic (mp/175 - m /125) times 100% for a selected number
 of important systems and subsystems which have been shown in Sections 3.1,
 3.2, and 3.3 to contribute to malperformance.  The table also presents a
 breakdown by manufacturer since differences between malperformance by manu-
 facturer were shown to exist in previous sections.  Reporting of the
                                     3-15

-------
correlations between malperformances and passed and failed vehicles by manu-
facturer cause the statistic reported to be generalized to

                   (m  ,  - m  , )  times 100%
                    r/F    P'p
where  F  and  P  are the number of vehicles failing and passing the test,
respectively, for each particular manufacturer.

        The interpretation of Table III-ll is that most malperformances posi-
tively correlate with vehicles that failed the initial test, although some of
the correlations are very weak.  Malperformances of the induction system do
not correlate with either a passed or failed vehicle.   The carburetor/fuel
system has the strongest correlation between vehicles  with a malperformance
and vehicles failing the initial test. Of the  individual  components,
maladjustment of the idle mixture correlates the best  with vehicles
failing the test as compared to other components.   Maladjustment of the
idle mixture for any vehicle implies that the vehicle will also fail the
emissions standards about 50% of the time.  Of course, this failure rate for
maladjusted idle mixtures varies from manufacturer to  manufacturer.  Mal-
adjusted idle mixture on Chrysler vehicles implies that the same vehicles
will also fail the emissions standards 57% of the time, while maladjusted
idle mixtures for Fords will also fail the emissions standards only about
11% of the time.

        Interpretation of these correlations for individual subsystems or
components is not advised,  however.   Previous sections have shown the inter-
relationships between malperformances and combinations of emission systems
and components.   For instance,  a maladjustment  of the  idle mixture might be
accompanied by maladjustment of ignition  timing and/or idle speed.   The
combined effect  may result  in emissions levels  which may still pass the
standards.   The   effects  of  changing or maladjusting certain components or
combinations of components  will be explored in  later sections.
                                    3-16

-------
3.5     EXAMINATION OF MALPERFORMANCES OF PASSED AND FAILED VEHICLES
        WHOSE EMISSIONS ARE EXTRAPOLATED TO 50,000 MILES

        The malperformance and emissions levels examined in the RM program
are for a sample of 300 vehicles with mileages between 0 and 15,000 miles.
Since 1972, new vehicles have been required to have emissions below the level
of the applicable standard in order to be certified by the Federal government.
Because of depreciation of a vehicle's engine and accompanying control equip-
ment (carburetor/fuel system, ignition system, EGR system, etc.) with time
                                                            123
and mileage, emissions are expected to change.  Many studies ' '  have been
conducted on various groups of vehicles as a function of mileage and age
to determine the rate at which emissions deteriorate.  Generally, the results
of these studies indicate that hydrocarbons and carbon monoxides increase
with increasing mileage.  While NOX emissions decreased or remained constant
with time, prior to the introduction of NOX control, trends for NOX controlled
vehicles are not clear.

        Results of deterioration studies show that linear regressions of
emissions with mileage are adequate to define the deterioration factors for
groups of vehicles.  These deterioration factors were determined from
certification durability data and are expressed as the ratio of the 50,000
mile emissions levels to the emissions levels at the 4,000 mile or break-in
point.  The 50,000 mile figure is used since in order to be certified, vehicles
must comply with the standards at 50,000 miles.  Thus, the predicted emissions
levels for each RM vehicle at 50,000 miles can be calculated, through
interpolation, using the certification deterioration factor and the RM
vehicle emissions at the known test mileage.  Deterioration factors less than
1.0 were set equal to 1.0 for this analysis since it was assumed that all
emissions increased or remained constant over the 4,000 to 50,000 mile range.

        Since deterioration increases the emissions for the vehicle sample
under consideration in the RM program, more vehicles will fail the initial
RM test if deterioration is taken into account.  Tables 111-12 through 111-15
present the malperformance rate for those vehicles projected to pass and  fail
standards at 50,000 miles by city and manufacturer.  Of  course,  the  percent  of
                                      3-17

-------
 malperforraances  that would  occur on  the  RM  vehicles when  they  are  at  50,000
 miles  is  unknown.  The  tables  in this  section merely  isolate the percent mal-
 performance  (at  the RM  test point) for that group of  vehicles  projected to
 pass and/or  fail standards  at  50,000 miles.  The rate of  malperformance (at
 the RM test  point) for  these vehicles  is  investigated to  determine whether
 the distribution of malperformances  is different for  these vehicles than for
 vehicles  that pass or fail  at  the time of the RM test.  Section 3.4 demon-
 strated a positive correlation between vehicles with  malperformances  and
 vehicles  failing the initial test.  Discussions in this section will  determine
 if the  correlation between malperformances  and vehicles that are projected to
 pass or fail at  50,000  miles is different as compared to  the relationships
 determined in 3.4.

          Tables  111-12  through 111-15  show  that only  102  vehicles pass the
 initial RM test assuming deterioration to 50,000 miles.   Therefore, 198
 vehicles  are projected  to fail the initial  test at 50,000 miles.  The rate
 of malperformance for the carburetor/fuel system for  those vehicles projected
 to pass at 50,000 miles is 38% as compared  to 11% for the ignition system,
 7% for  the induction system, and 3% for the EGR system.   The rate of  mal-
 performance for the carburetor/fuel system  for those  vehicles  projected to
 fail at 50,000 miles is 80% as compared to  34% for the ignition system,
 22% for the EGR system, and 6% for the induction system.

         The malperformance rate is higher  for those  vehicles  projected to
 fail than for those vehicles projected to pass at 50,000 miles for the
 carburetor/fuel system, the ignition system and the exhaust gas recirculation
 system.  There is no significant difference in the rate of malperformance for
 the induction system between the projected passed and failed vehicles at
 50,000 miles.

         Tables A-61 through A.-100 present the performance rates for each sub-
 system of each emission related system by city and manufacturer for vehicles
projected to pass and for vehicles projected to fail at 50,000 miles.   Table
 III-16 presents the correlation between malperformances and the projected
passed or failed vehicles at 50,000 miles by manufacturer.  Tabulated in

                                      3-18

-------
Table 111-16 are the differences in performance rates between failed and
passed vehicles as in Section 3.4.   Comparison of the percents of correlation
between the vehicles in Table III-11 in Section 3.4 with mileages between
0 and 15,000 and the vehicles in Table 111-16 all with mileages of 50,000
show little change in the correlations between malperformances and failed
vehicles for the subsystems investigated.
                                      3-19

-------
                                       TABLE III-l PERCENT OF MALPERFORMANCE

                                       BY CITY GROUP FOR EACH EMISSION  SYSTEM
CITY
CHICAGO
DETROIT
WASHINGTON
TOTAL

*
CARS
100
100
100
300
EMISSION RELATED SYSTEM
INDUCTION
8.00
7.00
4.00
6.33
CARBURETOR
FUEL
63.00
66.00
69.00
66.00
IGNITION
20.00
25.00
34.00
26.33
EXHAUST
GAS
RECIRCULA-
TION
16.00
12.00
6.00
15.33
AIR
PUMP
0.00
2.00
0.00
0.67
POSITIVE
CRANKCASE
VENTILATION
1.00
1.00
0.00
0.67
EXHAUST
0.00
0.00
0.00
0.00
EVAPORA-
TIVE
2.00
2.00
0.00
1.33
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.00
3.00
0.00
1.00
AT LEAST
ONE MAL-
PERFORMANCE
71.00
76.00
76.00
74.33
I
NJ
O

-------
 TABLE  II1-2  PERCENT OF MALPERFORMANCE
BY MANUFACTURER FOR EACH EMISSION SYSTEM
MANUFAC-
TURER
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL
f
CARS
102
99
99
300
EMISSION RELATED SYSTEM
INDUCTION
1.96
9.09
2.67
6.33
CARBURETOR
FUEL
49.02
55.56
93.94
66.00
IGNITION
21.57
25.25
32.32
26.33
EXHAUST
GAS
RECIRCULA-
TION
8.82
18.18
19.19
15.33
AIR
PUMP
0.00
2.02
0.00
0.67
POSITIVE
CRANKCASE
VENTILATION
0.00
1.01
1.01
0.67
EXHAUST
0.00
0.00
0.00
0.00
EVAPORA-
TIVE
0.98
1.01
2.02
1.33
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.00
2.02
1.01
1.00
AT LEAST
ONE MAL-
PERFORMANCE
58.82
68.69
95.96
74.33

-------
                         TABLES III-3    FREQUENCY OF DISABLED, MALADJUSTED,  AND DEFECTIVE
                                     COMPONENTS  OR SUBSYSTEMS  BY  MANUFACTURER

Manufacturer
General
Motors
Ford
Chrysler
Total
Subsystems of Carburetor/ Fuel System
Disabled Maladjusted Maladjusted Maladjusted
Limiter Idle Idle Choke
Caps Mixture Speed
31/102 28/102 19/102 13/101
36/99 15/99 24/99 6/99
69/99 70/99 31/99 11/99
136/300 113/300 74/300 30/299
Ignition
System
Maladjusted
Timing
19/102
20/99
18/99
57/300
EGR
System
Defective
or Dis-
abled EGR
Valve
0/102
1/99
9/97
10/298
Induction
System
Disabled
Hoses ,
Tubes
and Wires
1/102
6/99
5/99
12/300
NJ

-------
TABLE II1-4    FREQUENCY OF COMBINATIONS OF DISABLED OR MALADJUSTED COMPONENTS OR
                                 SUBSYSTEMS FOR ALL VEHICLES
Maladjusted
Idle
Mixture
n- ui j 93/300
Disabled
Limiter Caps
Maladjusted
Idle Mixture
Maladjusted
Idle Speed
Maladjusted
Choke
Maladjusted
Timing
Disabled or
Defective EGR
Valve
Disabled Hoses,
Tubes, Wires
of Induction
Systems
Disabled
or Defec-
Maladjusted Maladjusted Maladjusted tive EGR
Idle Choke Timing Valve
Speed
39/300 19/300 33/300 9/300
31/300 23/300 26/300 9/300
7/300 18/300 2/300
10/300 1/300
3/300
_
_ -
Disabled
Hoses,
Tubes.
Wires of
Induction
System
8/300
4/300
5/300
1/300
1/300
0/300
-
Total
Frequency
of Disabled
or Maladjusted
Component
Taken by Itself
136/300
113/300
74/300
30/300
57/300
10/300
12/300

-------
                         TABLE III-5  PERCENT OF MALPERFORMANCE BY CITY GROUP


                    FOR EACH EMISSION SYSTEM FOR VEHICLES THAT PASSED THE INITIAL TEST
CITY
CHICAGO
DETROIT

WASHINGTON

TOTAL

*
CARS
44

49

32

125
EMISSION RELATED SYSTEM
INDUCTION
6.82

10.20

0.00

6.40
CARBURETOR
FUEL
43.18

42.86

34.38

40.80
IGNITION
11.36

16.33

9.38

12.80
EXHAUST
GAS
RECIRCULA-
TION
2.27

6.12

3.13

4.00
AIR
PUMP
0

0

0

0
POSITIVE
CRANK CASE
VENTILATION
0.00

2.04

0.00

0.80
EXHAUST
0

0

0

0
EVAPORA-
TIVE
0

0

0

0
ENGINE
ASSEMBLY &
MISCELLA
NEOUS
0

0

0

0
AT LEAST
ONE MAL-
PERFORMANCE
47.73

57.14

43.75

50.40
trl
I

NJ

-------
TABLE III-6 PERCENT OF MALPERFORMANCE BY MANUFACTURER FOR ALL CITIES FOR
             EACH EMISSION SYSTEM FOR VEHICLES THAT PASSED THE INITIAL TEST
MANUFAC-
TURER
GENERAL
MOTORS
FORD

CHRYSLER

TOTAL

t
CARS
51

57

17

125
EMISSION RELATED SYSTEM
INDUCTION
0.00

10.53

11.76

6.40
CARBURETOR
FUEL
21.57

43.86

88.24

40.80
IGNITION
11.76

15.79

5.88

12.00
EXHAUST
GAS
RECIRCULA
TION
3.92

5.26

0.00

4.00
AIR
PUMP
0

0

0

0
POSITIVE
CRANKCASE
VENTILATION
0.00

1.75

0.00

0.80
EXHAUST
0

0

0

0
EVAPORA-
TIVE
0

0

0

0
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0

0

0

0
AT LEAST
ONE MAL-
PERFORMANCE
31.37

56.14

88.24

50.40

-------
TABLE III-7   FREQUENCY OF DISABLED, MALADJUSTED, AND DEFECTIVE
              COMPONENTS OR SYSTEMS BY MANUFACTURER, FOR VEHICLES
              PASSING THE INITIAL TEST

Manufacturer
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL
Carburetor/Fue 1
Dis-
abled
Limiter
Caps
2/51
14/57
8/17
24/125
Mal-
adjusted
Idle
Mixture
1/51
6/57
4/17
11/125
System
Mai- Mal-
adjusted adjusted
Idle Choke
Speed
6/51
11/57
5/17
22/125
2/50
5/57
2/17
9/124
Ignition
System
Mal-
adjusted
Timing
4/51
8/57
0/17
12/125
EGR
System
Defective
or Dis-
abled EGR
Valve
0/51
0/57
0/16
0/124
Induction
System
Dis-
abled
Hoses,
Tubes
5 Wires
0/51
	 	
4/57
1/17
_
5/125
                                 3-26

-------
                  TABLE II1-8  PERCENT OF MALPERFORMANCE
BY CITY GROUP FOR EACH EMISSION SYSTEM FOR VEHICLES FAILING THE INITIAL TEST
CITY
CHICAGO
DETROIT

WASHINGTON

TOTAL

f
CARS
56

51

68

175
EMISSION RELATED SYSTEM
INDUCTION
8.9

3.9

5.9

6.3
CARBURETOR
FUEL
78.6

88.2

85.3

84.0
IGNITION
26.8

33.3

45.6

36.0
EXHAUST
GAS
RECIRCULA-
TION
26.8

17.6

25.0

23.4
AIR
PUMP
0.0

3.9

0.0

1.1
POSITIVE
CRANKCASE
VENTILATION
1.8

0.0

0.0

0.6
EXHAUST
0

0

0

0
EVAPORA-
TIVE
3.6

3.9

0.0

2.3
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.0

5.9

0.0

1.7
AT LEAST
ONE MAL-
PERFORMANCE
89.3

94.1

91.2

91.4

-------
                                 TABLE III-9  PERCENT OF MALPERFORMANCE

                BY MANUFACTURER FOR EACH EMISSION SYSTEM FOR VEHICLES FAILING THE INITIAL TEST
MANUFAC-
TURER
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL

*
CARS
51
42
82
175
EMISSION RELATED SYSTEM
INDUCTION
3.9
7.1
7.3
6.3
CARBURETOR
FUEL
76.5
71.4
95.1
84.0
IGNITION
31.4
38.1
37.8
36.0
EXHAUST
GAS
RECIRCULA-
TION
13.7
35.7
23.2
23.4
AIR
PUMP
0.0
4.8
0.0
1.1
POSITIVE
CRANKCASE
VENTILATION
0.0
0.0
1.2
0.6
EXHAUST
0
0
0
0
EVAPORA-
TIVE
2.0
2.4
2.4
2.3
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.0
4.8
1.2
1.7
AT LEAST
ONE MAL-
PERFORMANCE
86.3
85.7
97.6
91.4
I
N)
GO

-------
TABLE 111-10   FREQUENCY OF DISABLED,  MALADJUSTED,  AND DEFECTIVE COMPONENTS OR
              SYSTEMS BY MANUFACTURER FOR VEHICLES FAILING THE INITIAL TEST

Manufacturer
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL
Carburetor/Fuel System
Mai- Mai- Mai-
Disabled adjusted adjusted adjusted
Limiter Idle Idle Choke
Caps Mixture Speed
29/51 27/51 13/51 11/51
22/42 9/42 13/42 1/42
61/82 66/82 26/82 9/82
112/175 102/175 52/175 21/175
Ignition
System
Mal-
adjusted
Timing
15/51
12/42
18/82
45/175
EGR System
Defective Disabled
or Dis- Defective EGR
abled EGR EGR Hoses,
Valve Transducer Lines
0/51 1/3 7/51
1/42 9/30 4/42
9/81 0/0 9/82
10/174 10/33 20/175
Induction
System
Disabled
Hoses,
Lines ,
Wires
1/51
2/42
4/82
7/175

-------
             TABLE III-11    PERCENT CORRELATION  BETWEEN  EMISSION  COMPONENT MALPERFORMANCES
                           AND VEHICLES THAT PASSED  AND FAILED  INITIAL TEST, BY MANUFACTURER



Manufacturer
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL
ANY MAL-
PERFORMANCE

Carburetor/ Fuel System
Disabled Maladjusted Maladjusted Maladjusted
Limiter Idle Idle Choke
Caps Mixture Speed

+53.0 +50.9 +13.7 +17.6
+27.8 +10.9 +11.6 -6.4
+27.4 +57.0 +2.3 -0.8
+44.8 +49.5 +11.1 +4.7

+ 43.2
Ignition
System
Maladjusted
Timing


+ 21.6
+ 14.6
+ 22.0
+ 16.1

+ 23.2
EGR
System
Defective
or
Disabled
EGR Valve


0.0
-2.4
+ 11.1
+5.7

+ 19.4
Induction
System
Disabled
Hoses ,
Tubes ,
Wires


-2.0
-2.2
-1.0
0.0

-0.1
Difference between the ma1performance rates of failed minus passed vehicles.   A + sign denotes
a positive correlation between a ma 1performance and a failed vehicle.   A - sign denotes a negative
correlation or a correlation between a ma1performance and a passed vehicle.   Zero represents no
correlation between malperformance and passed or failed vehicles.

-------
TABLK 111-12     PERCENT OF MALPERFORMANCE BY CITY GROUP FOR EACH EMISSION
  SYSTEM FOR VEHICLES PROJECTED TO PASS THE AS-RECEIVED TEST AT 50,000 MILES
CITY
CHICAGO
DETROIT
WASHINGTON
TOTAL
f
CARS
33
42
27
102
EMISSION RELATED SYSTEM
INDUCTION
9.09
9.52
0.00
6.86
CARBURETOR
FUEL
39.39
40.48
33.33
38.24
IGNITION
6.06
19.05
3.70
10.78
EXHAUST
GAS
RECIRCULA-
TION
0.00
4.76
3.70
2.94
AIR
PUMP
0
0
0
0
POSITIVE
CRANK CASE
VENTILATION
0.00
2.38
0.00
0.98
EXHAUST
0
0
0
0
EVAPORA-
TIVE
0
0
0
0
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0
0
0
0
AT LEAST
ONE MAL-
PERFORMANCE
39.39
57.14
37.04
46.08

-------
TABLE III-13  PERCENT OF MALPERFORMANCE BY MANUFACTURER FOR EACH EMISSION SYSTEM
              FOR VEHICLES PROJECTED TO PASS THE AS-RECEIVED TEST AT 50,000 MILES



MANUFAC-
TURER
GENERAL
MOTORS
FORD
CHRYSLER
TOTAL



*
CARS

43
47
12
102
EMISSION RELATED SYSTEM



INDUCTION

0.00
10.64
16.67
6.86


CARBURETOR
FUEL

16.28
44.68
91.67
38.24



IGNITION

11.63
12.77
0.00
10.78
EXHAUST
GAS
RECIRCULA
TION

2.33
4.26
0.00
2.94


AIR
PUMP

0
0
0
0

POSITIVE
CRANKCASE
VENTILATION

0.00
2.13
0.00
0.98



EXHAUST

0
0
0
0


EVAPORA-
TIVE

0
0
0
0
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS

0
0
0
0


AT LEAST
ONE MAL-
PERFORMANCE

25.58
53.19
91.67
46.08

-------
TABLE 111-14   PERCENT OF MALPERFORMANCE FOR VEHICLES PROJECTED TO FAIL THE
               AS-RECEIVED TEST AT 50,000 MILES FOR EACH EMISSION SYSTEM BY CITY
CITY
CHICAGO
DETROIT
WASHINGTON
TOTAL
i
CARS
67
58
73
198
EMISSION RELATED SYSTEM
INDUCTION
7.46
5.17
5.48
6.06
CARBURETOR
FUEL
74.63
84.48
82.19
80.30
IGNITION
26.87
29.31
45.21
34.34
EXHAUST
GAS
RECIRCULA
TION
23.88
17.24
23.29
21.72
AIR
PUMP
0.00
3.45
0.00
1.01
POSITIVE
CRANK CASE
VENTILATION
1.49
0.00
0.00
0.51
EXHAUST
0
0
0
0
EVAPORA-
TIVE
2.99
3.45
0.00
2.02
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.00
5.17
0.00
1.52
AT LEAST
ONE MAL-
PERFORMANCE
86.57
89.66
90.41
88.89

-------
                     TABLE 111-15   PERCENT OF MALPERFORMANCB  FOR  VEHICLES PROJECTED TO
                                    FAIL THE AS-RECEIVED TEST  AT 50,000  MILES  FOR EACH
                                    EMISSION SYSTEM BY MANUFACTURER
MANUFAC-
TURER
GENERAL
MOTORS
FORD

CHRYSLER

TOTAL

*
CARS
59

52

87

198
EMISSION RELATED SYSTEM
INDUCTION
3.39

7.69

6.90

6.06
CARBURETOR
FUEL
72.88

65.38

94.25

80.30
IGNITION
28.81

36.54

36.78

34.34
EXHAUST
GAS
RECIRCULA
TION
13.56

30.77

21.84

21.72
AIR
PUMP
0.00

3.85

0.00

1.01
POSITIVE
CRANKCASE
VENTILATION
0.00

0.00

1.15

0.51
EXHAUST
0

0

0

0
EVAPORA-
TIVE
1.69

1.92

2.30

2.02
ENGINE
ASSEMBLY &
MISCELLA-
NEOUS
0.00

3.85

1.15

1.52
AT LEAST
ONE MAL-
PERFORMANCE
83.05

82.69

96.55

88.89
CM
-£>.

-------
         TABLE 111-16
PERCENT CORRELATION  BETWEEN EMISSION COMPONENT MALPERFORMANCES
AND VEHICLES PROJECTED TO PASS OR PAIL AN AS-RECEIVED TEST AT
50,000 MILES BY MANUFACTURER
Carburetor/ Fuel System
Mai-
Disabled adjusted
Limiter Idle
Manufacturer Caps Mixture
GENERAL +48.5 +43.5
MOTORS
C^J
£ FORD +28.7 +4.5
CHRYSLER +13.0 +61.5
TOTAL +42.0 +43.7
ANY MAL-
PERFORMANCE +42.1
Mai- Mal-
adjusted adjusted
Idle Choke
Speed
+16.1 +13.8
+9.6 -8.7
-2.0 -6.4
+12.1 +1.7

Ignition
System
Mal-
adjusted
Timing
+ 16. 1
+ 18.2
+ 20.7
+ 16.4
+ 23.5
EGR
System
Defective
or Dis-
abled EGR
Valve
0.0
+ 1.9
+ 2.1
+4.1
+ 18.8
Induction
System
Disabled
Hoses ,
Tubes.,
Wires
+ 1.7
-0.6
-3.7
+0.1
-0.8
Difference between the malperformance rates of failed minus passed vehicles.
A + sign denotes a correlation between a malperformance and a failed vehicle.
A - sign denotes a correlation between a malperformance and a passed vehicle.
Zero represents no correlation between malperformance and passed or failed vehicles.

-------
4.0      EFFECT OF ADJUSTMENTS OUTSIDE OF SPECIFICATION TOLERANCES
         FOR TIMING,  IDLE RPM AND IDLE CO

         The degree to which a vehicle subsystem or component  is  out  of
adjustment is as important as the frequency or rate of malperforraance of
that component.  A particular component may have a high rate of maladjust-
ment, but the degree  to which it is maladjusted may have a very small effect
on emissions.  On the other hand, it is possible for a component  to have a
very small rate of malperformance, but the degree to which it  malperforms
may be large (i.e., it may be totally disabled) and the result may  be a
large increase of the level of emissions.

         The most prevalent emission component or subsystem malperformances
found on the RM test  vehicles are high idle CO, maladjusted idle speed,
and maladjusted timing.  The analysis of this section examines these three
types of malperformances and their effect on emission levels and FTP failure
rates.  It is emphasized that the effects of these malperformances  as given
in this section are not independent of one another (nor are the effects
independent of other malperformances).  For example, a vehicle with high
idle CO may also have maladjusted idle speed and perhaps other malperformances.

         New vehicles are tested and certified with their vehicle parameters,
i.e., timing, at the  mean of their allowable tolerance levels.  That is,
every vehicle is tested when certified at the manufacturer's specification
for timing and idle RPM with tolerances of ±2° for timing and ±100  RPM for
idle RPM.  Prior testing programs conducted by EPA have indicated a correlation
between excessive tailpipe idle CO rates and the failure of a vehicle to pass
the standards.  Since most vehicles do not have idle CO specifications, an
idle CO value was selected to define the difference between adjusted and mal-
adjusted idle CO.  A value of 0.5% was selected for the idle CO specification,
where values greater than 0.5% are considered outside of tolerances.

         Investigations of the effect of maladjustments (adjustments outside
of the allowed tolerances) on emissions are considered for the 300 vehicles
for the initial test of the RM program.  The effect of maladjustments on fuel

                                     4-1

-------
 economy  (both  the  Federal  Test  Procedure,  FTP,  fuel  economy and  the  Highway
 Fuel  Economy Test,  HFET) will be  explored.   The  FTP  fuel  economy is  repre-
 sentative  of urban  or  city driving,  and  the  HFET is  representative of  high
 speed, non-urban driving.   Differences between  cities  and manufacturers
 are also explored.

 4.1      PERCENTAGE OF VEHICLES WITHIN AND OUTSIDE OF  SPECIFICATION  TOLERANCES

         Tables IV-1 and IV-2 show the percent  of vehicles  outside of  the
 defined specifications for timing, tailpipe  idle CO  and idle RPM by  city and
 manufacturer.  For  instance, 35%  of  all  vehicles were  outside of the defined
 specification  tolerances for timing,  39% were greater  than  the idle  CO speci-
 fication of 0.5%, and  35%  were outside of the defined  specification  tolerances
 for idle RPM.  Seventy-two percent of all vehicles were outside  of at  least
 one of these specifications.

        The largest differences between  cities occur for  timing  with 24% of
 Chicago vehicles out of specification tolerances and 45%  of Washington vehicles
 out of specification tolerances and  for  idle RPM with  27%  of Chicago vehicles
 out of specification tolerances and  46%  of Detroit vehicles out  of specifica-
 tion  tolerances.  Chicago,  thus,  has  the lowest  percentage  of vehicles outside
 of specification tolerances for timing and idle  RPM.  There are  no city
 differences for idle CO.

        There are no differences between manufacturers for  timing.   Chrysler
has the largest percent of vehicles outside of the specification  tolerances
for idle CO with 76% and  for idle RPM with 46%.   Ninety-three percent of all
Chrysler vehicles  are outside of at  least one specification as compared with
64% for Ford and 61% for  General Motors.
                                     4-2

-------
4.2     CORRELATION BETWEEN VEHICLES WITHIN OR OUTSIDE OF SPECIFICATION
        TOLERANCES AND THE FAILURE OF A VEHICLE TO PASS THE FTP

        The purpose of this section is to determine if there is a correlation
between vehicles outside of specification tolerances for idle CO, timing
and/or idle RPM and vehicles failing the emissions standards.  It has been
shown in Section 3 that 175 of the 500 vehicles in test 1 fail one or more
of the emissions standards for HC, CO, and NOX.  If we assume that the
emissions are normally distributed (this will be discussed in more detail in
the next section), then the distribution for all 300 vehicles taking test 1
might be as postulated in Figure 4-1 for any of the three emissions.
                                          emission  standard
            C/3
                              emissions  level
            Figure  t-1   HYPOTHETICAL DISTRIBUTION OF EMISSIONS ASSUMING
                        NORMAL DISTRIBUTION FOR 300 VEHICLES IN TEST 1

        Figure 4-1 is only a qualitative example of an assumed normal distri-
bution whose mean is greater than a standard.   If the distribution of vehicles
in Figure 4-1 was partitioned into two distributions, those vehicles within
specification tolerances, and those vehicles outside of specification tolerances
for a particular component (i.e., timing), then each distribution (also
assuming each is normally distributed) might be as portrayed  in Figure 4-2.
                                      4-5

-------
                                     I  emissions standard
                                     r
                                       vehicles within specification tolerances
                                                 vehicles outside of specifica-
                                                     tion tolerances
                               emission level
            Figure 4-2  HYPOTHETICAL DISTRIBUTION OF EMISSIONS FOR
                        VEHICLES WITHIN AND OUTSIDE OF SPECIFICATIONS
          Quite  clearly,  if the  distribution  of  vehicles  represented  in  Figure
 4-1  is normally distributed,  then  the  partitioned  distribution  represented  in
 Figure 4-2  cannot  also both be  normally  distributed.  However,  Figure 4-2 does
 demonstrate qualitatively  the distribution obtained when  the emissions  from
 the  300 vehicles are partitioned into  vehicles  within and vehicles outside of
 specifications.  In fact,  the means  for  the  HC  and CO FTP emissions  and bag
 values are  always  larger for  the vehicles outside  of specifications  than for
 vehicles  within specifications, although the differences  in the means between
 within and  outside of specifications is not  always statistically significant.
 Table IV-3  presents those  FTP emissions  and  bag values whose differences in
 means between within and outside of  tolerances  are statistically significant
 at the 0.05  level for each component (timing, idle RPM and idle CO)  by manu-
 facturer.   The  group defined as "At  Least One"  is  that group with vehicles
 that have at least one of  the three  items (timing, idle RPM, and/or  idle CO)
within tolerances or outside of tolerances.   "At Least One" for vehicles
within specifications would be that group of vehicles within all three speci-
 fications for timing, idle RPM and idle CO simultaneously.

         Table  IV-3 indicates that the differences between means of  the within
and outside of specifications groups are significant primarily for idle CO for
hydrocarbons and carbon monoxide.   In the few places where the differences are
                                     4-4

-------
significant for nitrous oxides,  the means for vehicles  within  specifications
are greater than the means for vehicles outside of specifications  at  the
0.05 level of significance.

         If all the vehicles outside of a specification tolerance  (i.e.,  idle
CO) failed the FTP standards, and if all the vehicles within a specification
tolerance passed the FTP standards, then all the vehicles  less than the
standard (passing the standard)  in Figure 4-1 would be within  specification
tolerances and all vehicles  greater than the standard (failing the standard)
would all be outside of specification tolerances.   For this case,  there would
be a positive correlation between vehicles outside of specification tolerances
and vehicles failing the FTP standards.

         Next consider the situation where all the vehicles outside of speci-
fication tolerances also fail the FTP standards but the vehicles within
specification tolerances also fail the FTP standards for most  vehicles.
The situation would be such that all of the vehicles outside of specifica-
tions would be greater than the standard (fail the standard) and most of the
vehicles within specifications would also be greater than the standard. Thus
no correlation could be said to exist between vehicles outside of tolerances
and vehicles failing the standard since most vehicles within specifications
also fail the standards.

         The percent of vehicles failing the FTP standards for vehicles out-
side of tolerances minus the percent of vehicles failing the FTP standards
for vehicles within specification tolerances would be one statistic that
would classify the degree of correlation.  The closer this situation is to
100%, the greater the degree of correlation between vehicles outside of
specification and vehicles  failing the  FTP standards.  Also, by definition,
the closer this statistic is to  100%,  the greater the degree of correlation
between vehicles within specifications  and vehicles passing the FTP standards.
If the statistic is zero, there  is no  correlation.   If there are no statis-
tically significant differences  in the  means between vehicles within and
vehicles outside of specifications,  then the statistic is  apt to be zero
and no correlation will exist.
                                     4-5

-------
          Tables IV-4 through IV-7 show the percent of vehicles failing each
 standard and at least one standard for vehicles within and outside of
 specification tolerances.  The category "At Least One" for vehicles within
 specification tolerances  delineates those vehicles within specifications
 for all  specifications of timing, idle CO, and idle RPM simultaneously.
 There are a total  of 83 vehicles  in this group and 36% or 30  vehicles fail
 at  least one of the  FTP standards.   Eleven of 40 General  Motors  vehicles
 within all three specification groups  fail the FTP, 15 of 36  Ford vehicles
 within all specification  groups fail the FTP,  and 4 of 7  Chrysler vehicles
 within all three specification groups  fail the FTP.   The  group of vehicles
 within all the  specification groups simultaneously is an  important  one.
 Since 36% of these vehicles  fail  the FTP,  the  other component  malperformances
 listed in Section  3  account  for a number of the vehicles  that  fail  to meet
 standards.   But  even after all  emission  components  and subsystems  have been
 adjusted and/or  repaired,about  19%  of  the  vehicles  still  fail  standards.
 These  19%  will be  discussed  further in Section 5.

         The  remaining discussion will investigate  each specification group
 individually.  Note,  however,  that  these groups  are  not independent  from one
 another  (nor  are these malperformances independent  from EGR, air  pump, etc.
malperformances),  and some  linear combination  of timing,  idle  RPM  and idle CO
values might  be  a  better  discriminator to  determine  whether a  vehicle will
pass or  fail  the FTP  standards  than  any  individual  specification  group.

         Tables  IV-4  through  IV-7 show that  idle  CO  is the best indicator of
a pass or  fail of  the  FTP standards.  The  correlation  statistic previously
defined  is 52% for idle CO for  all vehicles.   The correlation  statistic is
64% for  General Motors, 21% for Ford, and  43%  for Chrysler for idle CO.
Whereas  the overall  correlation is best for  idle CO  as compared to  idle RPM
and timing, idle CO is a much better discriminator  for General Motors  vehicles
than for Ford or Chrysler.

         Idle CO is an even better indicator of a pass or fail of the HC and
CO standards.  The correlation statistic for all vehicles failing the HC
standard is 62%, 71% for General Motors, 20% for Ford, and 61% for Chrysler.
                                     4-6

-------
The correlation statistic for all vehicles failing the CO standard is 76%,
90% for General Motors, 44% for Ford,  and 74% for Chrysler.   There is a very
low or negative correlation statistic for NOX.

         Figures 4-3 through 4-8 substantiate the previous correlation
statistics and lend support to the hypothesized distributions of within and
outside of specification groups in Figure 4-2.   Figures 4-3 through 4-8 plot
the vehicle number by emissions levels for HC,  CO and NOX for vehicles with
tailpipe idle CO less than or equal to .5% and vehicles with tailpipe idle
CO greater than .5%.  Most all of the vehicles with idle CO less than/equal
to .5% pass the HC and CO emissions standards which is indicated by the
vertical dashed line in each plot.  A smaller but significant number of
vehicles with tailpipe idle CO greater than .5% fail the HC and CO standards.
Examination of the vehicles with low idle CO for NOX shows almost an equivalent
number of vehicles failing the MOX standard as the number of vehicles failing
the NOX standard for vehicles with the high idle CO values.

4.5      DEGREE TO WHICH IDLE CO, IDLE RPM, AND TIMING MALADJUSTMENTS
         EFFECT EMISSIONS AND FUEL ECONOMY

         Discussions in this section are divided into two parts:  first, the
effect of the degree of idle CO, idle RPM, and timing maladjustments on
emissions, and, second, the effect of the degree of these maladjustments on
fuel economy.  The results of Section 4.2 have demonstrated that  there is a
high degree of correlation between GM and Chrysler vehicles that  have idle
CO greater than .5% and GM and Chrysler vehicles that  fail the  HC and CO
standards.  Because of the results of Section 4.2, discussions  in this section
will focus primarily on the idle CO specification tolerances.   The idle RPM
and timing specification tolerances will be discussed  but to a  lesser degree.

         Tables IV-8 through  IV-10 show that the magnitude of the mean emis-
sions  increases as the positive  deviation  from idle CO of  .5% increases.
Caution is advised in  interpreting these  tables  since  the mean  emissions of
the vehicles  in the groups with  idle CO -1 to  -2 deviations  from  0.5%  (or
means  of all  vehicles  with tailpipe idle  CO between 0  and 0.25%)  is  derived
                                      4-7

-------
  VEHIO  I
  360    +
  320
  280
o
o
to
 , 240
a:
U)
   200
O
   160
6
n
UH
   120
U
    so
    40
     0   +
                                                   PLOT OF   VEHICLES  PASSING  IDLE-CO

                                           PLOT OF Vf.HID*FTPHC     SYMBOL  USED  IS  *
 *     <•*   *
 *   *   * *
     *  *   »* *
e   * ** *
      **           *
             ** *  *
                               *  *
                                  *«*   *
      •  *      *
   *     *        **
 *  *  *«      *         *
     *  *  **    *
 M *tt  *  lit   * *
     * *     *      «    *   «   *«r
     *    * *                    «   *
     ft        *         *     *r *
           *ft           * «      *  *
   .«*  »*     * *
    *«     * *               *
      *  *  *         *       * ft
         *  * *              *   *
            *      *       *   *
  It      **  *   *  *               *
    *   *           *
                                                                          11C  STANDARD
           0.0    0.2    0.4    0.6    0.8     1.0     1.2     1.4     1.6     1.8    2.0    2.2    2.4     2.6
                                                                                     	1--
                                                                                       2.8
                                                                                              3.0    3.2    3.4
   NOTE:     17 OBS HIDDEN
                                               FTPHC    HYDROCARBONS  (gin/mi)


                   Figure 4-3   VEHICLES WITHIN IDLE CO SPECIFICATION FOR UC

-------
  VEHID  I
  360    «•
  320
  280
  '240
  200
§
   160
u,
  120
(U
u
I— I
X
   40
                                                   PLOT OF VEHICLES FAILING   IOLE_CO

                                           PLOT OF VEHID*FTPHC    SYMBOL USED IS *
                           ft    tt
                               *   HI
                                   * *
                                R*
                                **   It
                                      it
                                      * *
                                      *
                                                 *     * *
                                                  *
                                                      * *
                                               UC STANDARIJ
                                        *   *    *M
          0.0    0.4    0.8     1.2     1.6     2.0     2.4    2.8    3.2    3.6    4.0    4.4    4.8    5.2
5.6    6.0    S. .1    (i. 8
                                                                    FTPHO     HYDROCARBONS (gm/rai)
  NOTE:      2 OBS HIDDEN
                             Figure  4-4   VEHICLES OUTSIDE OF THE  IDLE CO  SPECIFICATION FOR UC EMISSIONS

-------
                                               PLOT OF   VEHICLES PASSING IDLE-CO



                                        PLOT OF VEH1D*FTPCO    SYMBOL USED IS «
VEHID 1
360 +
320


^ 280
1
i
d 24°

PS
W

200
^
§
U 160
1-1
PL,
|-~
«
Q 120

3
i— <
3 80



40




0




* * * *
* * * it
* *» **
** * * ** * *
* * ** *
** * ** *
** * *
* *** *
* * * *
* *
*** *

* * *
* * *
* it * *
* * **
** *****
* * * * * *
* * * it *
* * * *
*** * * * *
*
*
*
*
* * * * *
* * * * *
* * it * * *
* ** **
* * M * * It
* * * * *
* * * * *
** It * It **
* * *
*
**
*
•f *
0 3 6 9 12 1



*


*
* *


*
*

*
*
*
it
* it
*




CO STANDARD
*


* *




*

5 18 21 24 27 30 33 36 3D 42 45
NOTE:     14 OBS HIDDEN
                                                                 FTPCO   CARBON  MONOXIUli (gn/mi)
                           Figure 4-5   VEHICLES WITHIN THE IDLE CO SPECIFICATION  FOR  CO EMISSIONS

-------
                                                      PLOT OF VEHICLES  FAILING    IDLE-CO

                                              PLOT  OF  VEHID*FTPCO     SYMBOL  USED  IS  *
     VEHID I
     360   +
     320
     280
   o
   o
   to
     240
   oc
   w
     200
     160
*>•  I-H
1   U,
   U)
   a 120
   u
   I—I
   X
      80
      40
                  *
                 *   *
       0   +
                   *  *     «
                        *     **
                          *
                     *
                     *
                                .CO STANDARD
                                *     *   *
                * *         *
                 * *   «    *          *     «
                     *        *
                                     *    *
                                               • * *
                              *                *            •
                              a        *  *          *  *
                                  *         *  *         * *
                                              *
                                                                   *    *
                     10      20      30      40      50      60      70       80       90       100      110

                                                                       FTPCO     CARBON MONOXIDl  (gra/mi)
                                                                                                             120      130      MO
     NOTE:
3 OBS HIDDEN
                                     Figure 4-6    VEH1OLRS OUTSIDE OF THE IDLIi CO  SPECIFICATION FOR CO

-------
   VEHIO  I
   360    +
   320
   280
o
o
ro
-i  240


CQ

§  200

O
I—I

U  160
U4
U4

I—(
OC
   120
    BO
    40
     0   *
                                                    PLOT OF   VEHICLES PASSING  IDLE.-CO

                                           PLOT OF VEHID*FTPNOXC    SYMBOL USED  IS  *
         * * *   *
     *  *   *
         *          * *
     •A * * *    *
          **    *    *
*         III      Ik* *
                                       *  *   *
                             *-«  **
                                       *  *
                                       * Ik
                                 ft*  Ik       *
                            * * *       * * **
                             *       *   *
                                         w
                           *             *   *ft
                             *          **
                           *     ** *     *
          *            **
    *        *****
          *   * *   * *
           *   **   *
  *    *  *    *  *
              *      ** *
       *tt  **  *      *
                              *     *
                                *  * **
                             *     *
                                                      -NOX STANDARD
                                                       *   *
                                                       *
                                                       *   *
           0.5    1.0    1.5    2.0    2.5    3.0     3.5     4.0     4.5     5.0     5.5    6.0
                                                                         6.5    7.0
                                                                                                           	^—
                                                                                                              7.5
8.0    8.5
                                                                     FTPNOKC    (gm/nii)
   NOTE:      7 OBS HIDDEN
                                      Figure 4-7     VEHICLES WITHIN THE  IDLE CO SPECIFICATION FOR NOX

-------
VEHID I
360   +
320
280
240
200
160
120
 80
 40
                                                 PLOT OF  VEHICLES  FAILING    IDLE-CO
                                        PLOT OF  VEHID*FTPNOXC    SYMBOL  USED IS  *
                              *
                           «r   *
                     ft   *  *      *
                    *        *   n
                    «         ft   *     *
                     *   *    *     *
                             ft*   *
               ft      ft       *  ft
                         *      **  *
                    *