EPA-AA-IMS-ST-80-1
          Analysis of In-House I/M Testing of a Three-Way Chevrolet
                    Citation and a Three-Way Dodge Aspen
                                January, 1980
                                   NOTICE

Technical Reports  do  not necessarily represent final EPA decisions or posi-
tions.  They are intended to present technical analysis of issues using data
which are currently  available.   The purpose  in  the  release  of such reports
is  to  facilitate  the  exchange  of  technical  information and  to  inform the
public  of  technical developments  which  may form the basis  for a final EPA
decision, position or regulatory action.
                        Inspection/Maintenance Staff
                    Emission Control Technology Division
                Office of Mobile Source Air Pollution Control
                      Office of Air, Noise and Radiation
                    U.S. Environmental Protection Agency

-------
                                  -2-
I.  INTRODUCTION

This report is an analysis of the effectiveness of I/M-type "short  tests"  in
identifying grossly  emitting vehicles.   Specifically,  this report examines
the question  for  two vehicles equipped with the technology to be  introduced
nationwide  in 1981.   This  technology incorporates  a  high degree of engine
control  through  the  use of sensors, actuators and an on-board computer, and
allows  the  simultaneous conversion  of  HC, CO and NOx  in a three-way  cata-
lyst.  .'•'••

In a testing  program performed in-house at the Motor Vehicle Emission Labor-
atory, Ann Arbor, Michigan, a 1980 Chevrolet Citation and  a 1979 Dodge  Aspen
were selectively disabled to simulate possible in-use vehicle conditions.   A
wide range  of testing was performed at each condition  including FTP testing
and I/M short tests.   In  this  report,  the FTP results will  be compared  to
the results of  two .I/M tests to  examine  their effectiveness in identifying
vehicles  with gross  FTP  emissions.   Each vehicle  will  first  be  discussed
separately and then an overall comparison will be given.   Performance of the
two tests using  the  recently proposed  207(b)  cutpoints  will  be  evaluated
separately  in Section  IV.   Full  data sets for the  two vehicles  can be ob-
tained  from the reports referenced at the end of this  report.

II.  1980 CHEVROLET CITATION

     A.   Vehicle Description

     The  Citation  was equipped with  a  2.5  liter  L-4 engine and  the General
     Motors "C-4" emission control  system.  The  C-4 system incorporates  an
     oxygen   sensor,   digital  on-board   computer,  and  a  solenoid operated
     feedback carburetor.   The Citation  is  equipped  with EGR and a single
     three-way catalyst for the conversion of HC, CO,  and NOx.   It is  cali-
     brated to meet  the 1980 California standards of Q.41/9.0/1.0.

     B.   Analysis

     For  the  Citation,  the emissions from seven distinct  vehicle  conditions
     were examined.   The  I/M tests  were  performed twice  for each  condition.
     Two  baseline  test  sequences were performed at the  beginning  and  the end
     of  the  test  program.  Three of  the  seven vehicle conditions  had  gross
     FTP  HC and CO emissions.  These  were:
                                                                  I
          -   Mixture  control solenoid disconnected
          -   Mixture  control solenoid disconnected and  EGR disconnected.
          -   Oxygen  sensor  disconnected and lead short-circuited.

     The  other vehicle conditions  resulted in low  FTP HC and CO  emissions
     and  involved  disconnecting  such   things  as   the  coolant  temperature
     switch,  the  EGR valve,  the closed  throttle  switch,  and  the oxygen
     sensor (without  short-circuiting the  lead).

-------
     To examine how well  the I/M tests could  identify the grossly emitting
     vehicle conditions,  the  FTP emissions  were  plotted against  the I/M
     test's emissions.   For the Citation,  these plots can be seen in Figures
     1-4.   To  be  a successful  screen in identifying  gross  emitters,  while
     not falsely  identifying clean  cars,  an  I/M  test should  yield values
     which   roughly correlate  with  the  FTP   emissions.   Therefore,  for  a
     successful test,  plots  such as  Figures  1-4  should  have a  cloud of
     points in the lower  left  hand corner from the "clean" cars and a cloud
     of points in the upper  right  hand corner  from  the  grossly emitting
     cars.   Conversely,  any point in  the upper left  hand  corner  (above an
     I/M test's  cutpoint  and  below  the FTP  standard) would  represent an
     error  of  commission  (i.e.  a falsely identified vehicle)  and  any point
     in the  lower right hand  corner (below the I/M cutpoint  and  above the
     line   representing gross  FTP emissions)  would  represent an  error of
     omission  (i.e.  a  vehicle  with  high  enough  FTP  emissions  to warrant
     identification in an  I/M  program).   Vehicles would routinely be tested
     for both  HC  and  CO,  and  failure for  either pollutant (i.e.  emissions
     above  a selected  cutpoint)  would  be sufficient to identify the vehicle
     as needing maintenance.

     As can  be seen  in Figures  1-4, both the Loaded  Two  Mode test and the
     non-loaded Two Speed Idle test were successful in  identifying the gross
     emitters.   The  scatter of  points for  the two  tests  are  very similar,
     the chief  difference being that the emissions from the 2500 rpm mode of
     the Two Speed Idle  test were somewhat lower  than all the other modes.
     These   emissions   from  the  2500  rpm mode are  not normally used  as  a
     criterion for evaluating whether or not a car would pass or fail an I/M
     test.

     It is  also clearly  evident that no errors of  commission or omission
     would  result  from either test.  This is based on cutpoints^of 1% CO and
     100 ppm HC and  assumes the emissions from the 2500 rpm mode of the Two
     Speed  Idle test are  not essential in  determining failure as discussed
     above.

III.  1979  DODGE ASPEN

     A.  Vehicle Description

     The Aspen was equipped with a  225 CID engine,  EGR,  a vacuum  operated
     feedback  carburetor,   an   oxygen  sensor,  an  on-board  computer  and  a
     three-way  catalyst  for the  conversion of HC,  CO, and NOx.  The Aspen
     was also equipped with an oxidation catalyst and an air pump to provide
     additional HC and CO  control.  It was designed to meet  the 1979 Cali-
     fornia standards of 0.41/9.0/1.5.

     B.  Analysis

     For the Aspen, the emissions from five distinct vehicle conditions were
     evaluated as well as two baseline tests at the beginning  of the program
     and one baseline  test at  the  end of  the program.   The five vehicle
     conditions were:

-------
                             -4-
   - oxygen  sensor disconnected
   - coolant  temperature  sensor  disconnected
   - mixture  control vacuum  solenoid  disconnected
   - air  injection bypassed
   - EGR  disconnected

As with  the Citation,  two sets  of I/M tests were  run for each condi-
tion.   Unlike  the  Citation,  however,  none of  the  conditions had truly
gross FTP  HC  and CO emissions.  Even those conditions which  involved a
loss of   feedback  control did not result  in  high HC/CO  emissions on
either the FTP or the I/M  tests.   This is due  primarily  to the addi-
tional  oxidation  capacity provided by  the  extra oxidation catalyst in
series with  the three-way  catalyst.   This oxidatioi catalyst is  sup-
plied with oxygen  by  an  air pump which leans out the- exhaust stream in
the oxidation catalyst even  when a rich mixture  is  bein,? provided by
the carburetor.   Thus, the  catalyst  is better  able  to  .leal with the
higher levels  of  HC  and  CO which  would normally result from a loss of
engine  feedback control.   The  only  vehicle  condition  with emissions
significantly above the standard is that with the air pump disconnected
(FTP HC = 1.03 gm/mi,  FTP CO =  29.8 gm/mi).  This underlines the issue
discussed  above,  that  is, with  the air pump disabled  and much of the
additional  oxidation  capacity lost,  the Aspen's  emission performance
begins to  resemble the Citation's more  closely.

As with the Citation,  the I/M test's emissions were plotted against the
FTP emissions.   It is  important to note the  different scales used on
these plots as opposed to the Citation's plots.  Both the  I/M emissions
and the FTP  emissions  have different  scales.   These  plot 5 can be  seen
in Figures 5-8.

As can be  seen in Figures 5-8, the  single vehicle condition with higher
FTP emission  levels was  essentially  separated  from  the other points.
This is especially true  for CO  (Figures  5  and  7).   It  is important  to
note,   however,  that  a relatively low  I/M cutpoint  would  have   been
needed  to identify this car as  needing maintenance  (approximately  .5%
CO).  For HC,  the FTP emissions  for  the  one  higher condition were not
as grossly above  standards  as for  the  FTP CO emissions, and  the corre-
sponding  I/M  tests'  emission levels were also  not  as high.  As can be
seen  in  Figures  6 and 8,  however,  the  emission levels  at  idle  were
essentially separate for the one higher point.

As  can also  be  seen  in Figures  6 and  8,  one  point  (representing a
disconnected  EGR with FTP HC emissions  at 0.4 gm/mi and with  I/M tests'
emissions  of 70 ppm  and  90  ppm for  the  Loaded Two  Mode and the Two
Speed Idle respectively),  could  possibly have  caused  an error of  com-
mission.   This would depend of course on the I/M cutpoint  used.

One data  point  for  the  Aspen was dropped as  it  was an apparent  data
collection error.

-------
                                 -5-


IV.  PERFORMANCE OF THE TWO I/M TESTS USING 207(b) CUTPOINTS

     A.  Description of 207(b) Outpoints

     Section 207(b) of  the Clean Air Act  provides  for the establishment  of
     I/M cutpoints which  could  be used in warranty  claims against the auto
     manufacturers.   In  this  portion of the  report,  the currently proposed
     cutpoints  (Reference  3)  will be applied to the data from the Aspen and
     the Citation.

     For the Two  Speed  Idle test, the  207(b)  cutpoints are 1.0% CO and 200
     ppm HC.   The  lower  of  the  emission levels from the two  idle modes
     and/or the  emission  levels  from  the  2500 rpm mode  can  be selected  as
     the criteria by  a  state  or  municipality  to  determine whether or not a
     vehicle passed the  test.   Thus, the idle emissions alone, the 2500 rpm
     emissions alone,  or  the  two combined can be  used as a basis for deci-
     sion.   It is also important  to note that if a vehicle fails any section
     of the test and for either CO or HC, it fails the whole test.

     For the Loaded Two  Mode  test, the cutpoints are  1.2% CO and 220 ppm  HC
     for both  modes  of  the test.   Here  too, a  state or  municipality can
     choose the emissions  from  either or both of the modes as the basis for
     decision,  and if a  vehicle  fails any section of the test for either  CO
     or HC, it fails the whole test.

     B.  1980 Citation

     For the Citation, both the Loaded Two Mode test and the Two Speed Idle
     test demonstrated perfect  performance (i.e.  no errors of commission  or
     errors of  omission)  using  the 207(b) cutpoints in all possible config-
     urations of  the  test  criteria.   All  of the failures  were  for high  CO
     levels.

     C.  1979 Dodge Aspen

     For the Aspen, none of the  I/M  tests'  emissions were above the 207(b)
     cutpoints,  so  there were  therefore  no  errors  of commission possible.
     The failure  rate for correctly  identified vehicles was also  therefore
     zero.

     In looking at  errors of  omission, points  to the right of (i.e. above)
     the FTP standard  in Figures  5-8 must be identified.  For the Aspen, the
     applicable  standards are  0.41/9.0/1.5  for  1979  California vehicles.
     Using  these  standards, three of the  eight vehicle conditions could  be
     called "errors of omission", however  in terms of  significance, only the
     vehicle condition  with  the  disconnected  air  pump  could be  termed a
     significant  error of  omission.   The other two conditions had FTP emis-
     sions only marginally over the standard.

     In summary  then, analysis  of the Aspen's emissions using 207(b) cut-
     points is  inconclusive,  since to properly evaluate  the  issue, a wider
     spread of data would be needed.  There were no errors of commission and
     there was one significant error of omission.

-------
                                 -6-
V.  CONCLUSIONS

Both  the  Loaded Two  Mode and  the  Two  Speed  Idle tests demonstrated  their
ability to  identify the grossly emitting  vehicle conditions.  This depends
of  course  on  the  cutpoints  used and what  is  defined as grossly emitting.
For  CO,  the  issue  is  clear,  and  especially  so for  the  grossly  emitting
conditions  from  the Citation.   For HC,  the  issue is  clear  for  the  Citation
and somewhat  clouded  for  the Aspen.  In general,  the  results  from the  Aspen
are of  limited value,  since the central question involves whether I/M  tests
can identify  gross emitters  or not,  and  the Aspen had  generally very  clean
emissions.  One  thing  that can be  said  for  both  the  Citation and the  Aspen
is  that the vehicle conditions with  low FTP emissions also  had  low  I/M test
emissions,  i»e.,  there were  no significant errors  of commission.  This  is
especially  true  for CO and  generally true for HC except for the one  ques-
tionable point from the Aspen discussed  above.

-------
                                 -7-
References:

1.   Evaluation  of  Applicability of Inspection/Maintenance Tests  on  a Dodge
Aspen Prototype.  August 1979, Thomas J. Penninga, Technology Assessment and
Evaluation  Branch,  Emission  Control Technology  Division,  OANR, U.S.  EPA.

2.  Evaluation of Applicability of Inspection/Maintenance Tests on a Chevro-
let  Citation.   October  1979,  Thomas J.  Penninga.   Technology Assessment and
Evaluation  Branch,  Emission  Control Technology  Division,  OANR, U.S.  EPA.

3.   Light  Duty  Vehicle  and Light Duty  Truck  Emission Performance Warranty;
Short  Tests and  Standards.   December  1979,   Inspection/Maintenance Staff,
Emission Control Technology Division, OANR, U.S. EPA.

-------
                                           -8-




                                     Figure 1


                                     CITATION


                           FTP CO vs. Loaded Two Mode CO
01
13
O
S

o
"$.
H
cd
O
       8.0
        7.0
        6.0
        5.0
4.0
        3.C.
        2.C.
        l.C.
                        25
                              50
75
                                                                +


                                                                D
                                                                  D
                                                                  D
                                            FTP  CO

                                           (gm/mi)
                                                 +  =
    100


30 mph.

Idle"
                                                                               B
                                                                                 125

-------
     240
     220
     200
     180
     160
     140
*••
  0.
  CL,

•g~ 120

T3
cfl
O
     100
      80
      60
      40
      20
                                           -9-




                                        Figure 2


                                        CITATION


                              FTP HC vs. Loaded Two Mode  HC
                                                           B
                                 B
                                                                      D

                                                                      D
                      0.5
1.0
1.5
                                    FTP HC

                                    (gm/mi)
  2.0



30 mph.

Idle
2.5
3.0

-------
           -10-
           Figure 3




           CITATION




FTP CO vs. Two Speed Idle CO
7.0
6.0
8 5.0
M
^N
0) S-'
01
w* 4.0
o
H
3.0

2.0
1.0
0
D


A
A


A
A A
A
•
0 25 50 75 100 125
FTP CO + = 1st Idle
(gm/mi) A = 2500 rpm
              O =  2nd Idle

-------
            -11-
          Figure 4




          CITATION




FTP HC vs. Two Speed Idle HC
240
220
200
180
160
140
0)
H
/-N
•0 S
aj p<
0) P.
w ^ 100
80
60
40
20
0
C
-
-
Ifa
D Q
m
A
A

A
A
•
' -f
•^
) 0.5 1.0 1.5 2.0 2.5 3.0
FTP HC + = 1st Idle
(gm/mi) A = 2500 rpm
O = 2nd Idle

-------
             -12-





          Figure 5




            ASPEN




•FTP CO vs. Loaded Two Mode CO
J..U
0;9
0.8
0.7
o
u
a) 0.6
-a
s
o
H &5
•a ^ 0.5
cu
n)
O
0.4

0.3

0.2
0.1
0







r +

•
n

a


a
0 5 10 15 20 25 30 35
FTP CO Q = 30 mph
(gm/mi) + = Idle

-------
            -13-
          Figure 6




            ASPEN




FTP HC vs. Loaded Two Mode HC
J.OU
140
120

o 100
a
0)
1
^~\
o S
££
3~ 8°
n)
0
60
40

20
0

-
+





+ +

•
D 4H- +
° D D
+
Ifc D
i i -i -1 - j i i i i • i i
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
FTP HC O = 30 mph
(gm/mi) + = Idle

-------
            -14-
          Figure 7




            ASPEN




FTP CO vs. Two Speed Idle CO
I.U
0.9
0.8
0.7
0.6
o
o
0)

-------
                                              -15-






                                           Figure 8




                                             ASPEN




                                FTP HC vs. Two  Speed Idle HC
J.VJVS
140
120
K 100
0)
'S ^
"O CX
Q) CX
Q) •• —
£ 8°
1
60

40

20
0
D



a
a

EB
D
at—*
Bu
+ la 1
A n A
/\
A /\
• | i | i ( | | 11 i
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
FTP HC + = 1st Idle
(gin/mi) A = 2500 rpm
                                                       O =  2nd Idle
* VS. GOVERNMENT POINTING OFFICE: 1980- 651-112/0168

-------