EPA-AA-SDSB-81-4
                          Technical Report
                An Evaluation of Three-Way Control
            Single and Dual Bed Catalysts As  Applied to
                     Heavy-Duty  Gasoline  Engines
                                 by

                            Thomas  Nugent

                             April 1981
                               NOTICE

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

-------
                          Table of Contents

                                                               Page

I.   Introduction 	  1

II.  Discussion	1

     A.    Equipment	1
     B.    Procedure	7
     C.    Engine History 	  7
     D.    Problems	8

III. Results	9

IV.  Conclusions	20

References	23

Appendix A	24

-------
                                  -1-
I.   Introduction

     A  test  program  to  evaluate  the  applicability of  three-way
control (TWC) and dual-bed catalysts  (TWC and Oxidation Catalysts)
in  combination  with  a closed-loop  control  stoichiometric  (feed-
back) carburetor on a  production  1978 IHC 404 CID heavy-duty gaso-
line engine  was  conducted  at  EPA's  Office  of  Mobile  Source  Air
Pollution control Laboratory  (OMSAPC) Ann  Arbor,  Michigan.   Test-
ing was conducted according  to the federal  test  procedure  for the
heavy-duty transient cycle.[1]

     The  test   program consisted  of  forty-six  cold-start  tests,
fifty-eight  confirmatory  hot-start  tests  run  to insure  accuracy
and  precision   of  the  cold-start  tests,  and  fourteen  hot-start
tests run to identify  the effects  of  air/fuel ratio  control points
on  NOx  emissions.  The effects of various combinations of  cata-
lysts,   EGR,  air/fuel  ratio  control  points,  carburetor  response
times and air pump  capacity  on emissions,  fuel economy and engine
power were examined.

     This project and  data developed  through this  project has been
presented previously  in an  SAE paper jointly developed by  EPA and
Engelhard  Industries.   The  paper was  formally  presented  at  the
1981 SAE international Congress and Exposition.[2]

     This  EPA   Technical  Report provides background  for  the  SAE
document.  All  individual test  results  are  presented  and  an  ex-
panded  review   is  done of  the  conduct   of  the  test   program  and
specific  problem  situations  encountered  in  conducting the  proj-
ect.  This  report will only  highlight results  of  this study.   A
more detailed analysis of results  can be  found  in the  SAE  publica-
tion.

II.  Discussion

     A.    Equipment

     Tests were conducted by EPA with   supportive  expertise  pro-
vided by  Engelhard  Industries, the  supplier of all catalysts  and
oxygen sensors  utilized in the  test  program.   The feedback  Carbur-
etor and  Logic  Control Box were purchased  by EPA from the Holley
Carburetor Company.

     The engine, an IHC 404  1978  production engine with a  Califor-
nia calibration representative of 1979  technology,  was loaned  to
EPA for  testing purposes, through  the  cooperative  efforts  of  the
International  Harvester  Company.    Technical  specifications  for
equipment utilized in  this program are described in Table l.[2]

     The Holley Model  2210  experimental  feedback  carburetor  was a
direct bolt  on  replacement  of the original Model  2210  on  the 1978
engine.  A 10 Hz  electric solenoid valve  controlled  both the idle

-------
                                -2-
                              Table  1

          Technical  Specifications:  TWC Catalyst Program

          Engine:  1978 Production IHC  404  cubic inch V-8
                   heavy-duty truck engine
Three-Way Catalysts;
Oxidation Catalysts:
Feedback Carburetor:
Logic Box;
Two,   151-CID  monolithic,   50   grams/ft3
loading,  Platinum-rhodium  ratio   of   5:1,
(manufactured    by    Engelhard),    Corning
substrate of 300 cells/in2.

Two,   113-CID  monolithic,   50   grams/ft3
loading,  platinum-palladium ratio  of  4:1,
(manufactured  by Engelhard),  Corning  sub-
strate of 300 cells/in2.

Holley  2210 modified  for  experimental  use
by Holley  Carburetor Co., Division  of  Colt
Industries,  for  stoichiometric  closed-loop
control  of  idle  and main-jets.   Engelhard
designed microswitch with  throttle  engage-
memt during  the last approximate  9  degrees
of  throttle  movement   prior  to  wide  open
throttle.   Microswitch activation  provided
a  fixed  4.6 percent CO (A/F  ratio  =  12.9)
enrichment  (in  place   of  the  conventional
power valve which provided  upto  6.0 percent
CO at  wide  open  throttle  on  the  1978  pro-
duction  carburetor).   Standard  accelerator
pump, choke and throttle  deceleration posi-
tioner were retained.

Holley  Model  8 experimental-type  designed
and built with adjustable,

- air/fuel ratio control point
  (350 mv to 800 mv)
- response time (CRT)
- AC  gain (a  gain   parameter  in  the  logic
  box)

- wide  open throttle  (WOT) open-loop  over-
  ride
- cold start open loop  override*
*    Provides an  open-loop  air/fuel  ratio  near stoichiometric from
the carburetor jets so that the choke may provide enrichment.

-------
                                 -3-

                          Table 1 (cont'd)

           Technical  Specifications;   TWO  Catalyst Program

Exhaust Gas Oxygen
Sensor;                Standard  production Robert  Bosch part  No.
                       0258001001.

Air Pump:              Each  pump  delivers 7.21-8.30  CFM  @  1000
                       pump  RPM and  1.6   inches  Hg  backpressure,
                       (IHC no. 446746-C92, 461369-C91).

EGR System;            Standard 1978 production.

Air Mixers;            Located   between   the   TWC  and   oxidation
                       catalysts.

-------
                                -4-

and main jet  circuits.    Choke  calibration  and  idle  speed  were
standard (1978).  The idle mixture  screws were  adjusted  to  provide
approximately 50 percent duty cycle on  the  solenoid  valve  in order
to achieve maximum control flexibility.

     A  throttle  actuated  microswitch  deactivated the  closed-loop
control  near  the wide open  throttle  (WOT)  position providing  en-
richment to about a 12.9 air/fuel ratio  (4.6  percent CO) as a com-
promise  for maximum  horsepower,   engine durability,  and  overall
emissions.  At  all  throttle  positions  between  idle  and the  near
WOT position  the closed-loop control system  was intended  to  pro-
vide  stoichiometric  air/fuel  ratio  conditions  so  that  the  TWC
catalysts could  function properly.   It  should  be noted that  when
the microswitch  was  activated near WOT,  resulting in an  air/fuel
ratio of 12.9, the TWC single-bed catalyst  provided  little  overall
conversion of HC, CO or  NOx.  However,  since  air was added  between
the two beds in  the TWC  dual-bed catalyst system the second bed or
oxidation catalyst removed a  large  fraction of  the HC and  CO emis-
sions.

     The engine  utilized a deceleration throttle  positioner which
opened the  trottle slightly  during  deceleration  (whenever  the  RPM
exceeded  1800  at high manifold  vacuums) in  order to reduce  mis-
firing and excessively rich mixtures.

     The Holley  Model 8  logic  box contained the adjustment  fea-
tures listed in  Table  1.  The millivolt control  point  established
the  operating  point  for  the  oxygen  sensor  output  voltage  and
thereby  determined  the  air/fuel  ratio  relative  to the  stoichio-
metric condition.  This  relationship is  represented by Figure 1.
The  characteristic  response  time   (CRT)  determined the  speed  at
which  the  air/fuel   ratio  cycled  around  the  millivolt   control
point.   A cold  start  override was controlled  by a  thermostatic
switch in the engine coolant  circuit which  was  set to  activate the
closed-loop system at 120°F.

     The  original  1978  production  engine  utilized  air  injection
from one air  pump into  the  exhaust ports  of each  cylinder bank.
This air path  was sealed  for the  conversion to the TWC  catalyst
system.  The air pump was  instead  routed to  the  air mixer  between
the dual-beds  as shown  in Figure  2.   The  air  pump  was not  used
when the TWC single-bed  catalysts were  used  alone.   No  attempt was
made to  divert air into  exhaust  ports upon cold  start in  order to
reduce cold start HC and CO  emissions  (where  the TWC catalyst ini-
tially acts as  an oxidation catalyst).   Doing  so  would  surely de-
crease emissions, as  demonstrated  in  the 1978  production  version
where the  air,  directed into  the  exhaust ports,  had  a very  sub-
stantial effect on HC emissions.[2]

     Tests  were  conducted  in a standard  EPA  heavy-duty  gasoline
transient  test  cell.  The cell utilized  a  double-ended  dynamom-
eter, water coolant system,  electronic  instrumentation  an  ambient

-------
                                 -5-
                               Figure
                             SrOICMIOMETRIC
Oxygen
Sensor
Output
Voltage
(millivolts)
                              Air/fuel  ratio
       Typical curve of oxygen  sensor  output voltage vs. air/fuel ratio

-------
                  TO AIR
                  MIXERS'
                                                            ELECTRIC
                                                         DYNAMOMETER
  MICRO
  SWITCH
LOGIC BOX
                      TWC CATALYSTS
                                         OXIDATION'
                                        CATALYSTS
                  FIG. 2.'-.SCHEMATIC OF EXHAUST SYSTEM IN TEST CELL CONFIGURATION
                  FOR DUAL BED CATALYST TESTING IN THIS PROGRAM

-------
                                -7-
air handling/humidity  conditioning  system  all of  which serve  as
independent units  for  each test cell.   Computer  and CFV/CVS  sys-
tems are  shared  between  cells.   A  more  complete  description  of
test cell  facilities can  be found in EPA's  1972-1973 baseline  en-
gine report.[3]

     Other  equipment  utilized  in  this  test  program  was  either
standard  1978  production  or  fabricated  to  meet  special  require-
ments of the program.

     B.    Procedure
     All testing  activities  conducted through this  program  follow
OMSAPC standard procedures for heavy-duty  gasoline  transient  test-
ing.   Specific  procedures  can be  found  in  the Code of  Federal
Regulations, 40 CFR, Part 86, Subpart N.

     Briefly, with  the  engine and emission  control system set  up
for  the configuration   to   be  tested,  one  heavy-duty  transient
cold-start  test and three confirmatory  hot-start  tests (individual
hot-start portions of the cold-start  transient tests)  were run.   A
twenty  minute soak  was  conducted  between  each  test.   As  each
sample  was  collected  it  was  analyzed  by  EPA  technicians  on
in-house analyzer A009.   Raw emission  figures were turned in for
computer processing upon completion  of  the test  sequence for  each
version.   Processing  was  usually complete  by the following  day
with the configuration not  being  changed  until confirmatory  emis-
sion  results and  satisfactory  cycle  performance  were  obtained.
Upon analysis of  emission results the  subsequent configuration  to
be tested was determined and testing scheduled.

     In the instances where  major version  changes,  such as  carbur-
etor substitution occured, or where major  engine  repairs  were  con-
ducted the engine was analyzed with  a Sun  2001 Diagnostic Computer
to insure proper tune prior to testing.

     C.    Engine History

     This  engine was  initially  utilized  in an  EPA program  to
determine engine emission sensitivity to the  1984 heavy-duty  tran-
sient  test  cycle.   Data  developed through the sensitivity  program
was  presented through  SAE  publication.[4]   The engine  was  also
utilized in  a 1979  test  program involving the application  of  var-
ious oxidation catalyst  and  air  pump configurations.   Data  devel-
oped in  this  program  was utilized in the  development  of the  1984
heavy-duty gaseous emission  regulations.[1]   The  data  was not  pub-
lished separately.  Testing  of the TWC/feedback  carburetor  system
begain on  January 17, 1980  at EPA's  MVEL.   Initial  baseline and
early  catalyst  testing  was  conducted in test cell  D103W.  Due  to
other  priority  testing  requirements the engine  was moved,  at ap-
proximately  mid-program  (6-26-80),  to  test  cell D104E  where the
test program  was completed  on December 19,  1980.   The engine was
removed from the test cell on December 23,  1980.

-------
                                -8-
     D.    Problems

     A  number  of  minor  problems  inherent  with  test  programs  of
this nature  were encountered.   These  included thermocouple  fail-
ure, air handling  limitations and resulting high  cell  temperature
problems, torque cell  failures,  computer downtime,  analyzer  down-
time,  logic  control box missettings  and various  miscalibrations.
These  did  not  seriously  disrupt  the  program.  A number of  more
serious situations however did delay program completion.

     During the  soak portion of  the  cold-start  test conducted  on
May  29,  1980  a fuel  regulator  failure (a  test  cell  component)
resulted in engine  flooding  and raw fuel collection in  the  cylin-
ders.  As  the  system was  dismantled  to dispell  the fuel four  of
the  eight  spark plugs  were  broken.   The  test  configuration  was
resumed with replacement of  the  fuel  regulator and all  eight  spark
plugs.  The problem of  engine modification from the baseline  con-
figuration (new spark plugs), and  potential  subsequent  data  varia-
tion was noted and testing resumed.

     A NOx correlation  testing  program on another  engine required
the IHC engine to  be move  to a new test cell  midway through  test-
ing (from  test  cell D103W to cell D104E).   The  NOx study was  de-
signed  to  determine the difference  in  bag  versus  continuous  NOx
measurement during  transient heavy— duty gasoline engine  emission
tests.  Cell D103W  is  physically located next to  diesel  cell  D102
which  is  equipped  with  a continuous  NOx measurement  capability.
Locating the  NOx correlation engine  in cell  D103W allowed,  with
minor  modification (creation of a passageway through the  common
wall separating  the two cells) ,  simultaneous  normal bag sampling
and continuous NOx sampling to occur.

     The transfer  of  the TWC test engine  to  another testing  cell
was  not  felt to have  been  a significant problem since the  same
physical parameters  or limits  apply  to all  test  cells  and  since
each  test  must  be compared  to  the  cycle  performance  specifica-
tions.  Any  parameter  exceeding  its  stated  limits  (temperature,
humidity etc.) or not passing the  cycle  performance specifications
is  susceptible  to  being voided  regardless of where the test  was
run.   It  of course,  would  have been  of  preference not to  have
moved the engine .
     In mid- July  a knock was  detected  in the engine  that
ponded to a noticable emission increase.   An engine  check discoved
a misgapped  spark plug.   The plug was  regapped and  testing
tinued with emissions returning to  near  expected levels.   However,
by mid-August the knock had intensified and  engine  analysis  indi-
cated a broken  piston.   Probably  the f orementioned  misgapped  plug
was closed by a piece of the  broken  piston that had  since  disin-
tegrated  with  continued  engine use.   It  was   further  determined
that arching was  occuring  from several plug wire  boots and  that
the oxygen sensor  had lost  some of  its  low temperature sensitivity

-------
                               -9-
through damage  from  either the piston  failure  or spark plug mis-
firing.  The piston, oxygen sensor  (aged at Engelhard), plug wires
and plug boots were replaced.

     In late  September while  running  consecutive hot-start tests
to determine  NOx emission levels at various  millivolt set points
(determinant of air/fuel ratio) and ignition  failure resulted in  a
complete burnout  of  the TWC catalysts.  The  ignition failure was
caused by  a short  (insulation melt)  in  the  primary  lead to the
coil.  EPA technician responsiveness resulted in damage being con-
fined to the TWC catalysts and oxygen  sensor.

     To insure  that  the oxidation catalysts  were  not damaged the
complete dual-bed system  and  oxygen  sensor  were  replaced.   The
replacement catalysts,  supplied by  Engelhard, were loaded  identi-
cally to those destroyed in the burn out.  The  catalysts were aged
by running a seris of  cold-start  tests  with forced  cool-downs over
a two  day  period.  A  total of approximately  eight break-in hours
of testing were run.

     All tests  determined  to  have been affected  by any of the
forementioned problems were voided and those versions  rerun.

III. Results

     A description  of  configurations  tested  as  well as  emission
and  fuel  economy  figures can  be  found  for  cold-start  tests in
Table  2.*   The most  favorable  emission results  were obtained in
the dual-bed configuration (Figure 2) with EGR  (version 45c).  The
HC level of 0.68 g/Bhp-hr, CO  level of  3.60 g/Bhp-hr  and NOx level
of 0.74 g/Bhp-hr are well  below the 1984 heavy-duty emission stan-
dards of 1.3 g/Bhp-hr  for  HC,  15.5  g/Bhp-hr  for CO and the statu-
tory NOx standard of  1.7 g/Bhp-hr.  The NOx  standard is  not part
of the 1984 regulation  but is  being developed as a separate rule-
making.  These  test  levels  represent  conversion efficiencies of
approximately 73 percent  for HC,  91 percent  for CO and 86 percent
for  NOx  as  compared  to  the  stock single  exhaust  configuration
(version 00).   Emission values  for  version 13  (version 45  without
EGR), version  21  (version 45  with a fast carburetor  response) and
version 23  (version  45 with a slow carburetor  response) also meet
these standards.

     The need  to  test  both single and  dual-bed configurations was
required since  before  testing  it could not  be determined if the
dual-bed configuration would  be  required to meet the HC and CO
emission standards.
     A  considerable  number of  confirmatory hot  start tests  were
run to  verify the cold^-s^art  data but the  test  results were  not
used  further  in the  a^TlaW- -    •"--- L-^  -----  ^---  --  --•	
Appendix A.           V__S
used further  in the a^ilaysis.   This hot  start data  is  given  in        \/

-------
                                          -10-
Version
Number

  00
  09
  10
  11
                                         Table 2

                                 IHC 404 Cold Start  Tests
                                                           Emissions
                             HC
                CO
 NOx
 BSFC
Engine Configuration
(g/BHP-hr)  (g/BHP-hr)   (g/BHP-hr)   (Ib.  fuel/BHP-hr)
1978 production engine,     2.55
1978 EGR, air pump with     2.48
air to exhaust ports,
standard carburetor, con-
ventional single exhaust.

1978 production engine,     2.65
1978 EGR, air pump with     2.65
air to exhaust ports,
standard carburetor,
fabricated single ex-
haust split into dual ex-
hausts, straight pipes
representing dummy cata-
lysts, no air mixers.

1978 production engine,     2.31
1978 EGR, air pump with     3.05
air to exhaust ports,       3.01
standard carburetor, fab-
ricated exhaust, single
exhaust split into dual
exhausts, two dummy TWC
catalysts, straight pipes
representing dummy oxida-
tion catalysts, no air
mixers.

1978 production engine,     3.58
no EGR, one air pump,       4.63
air to air mixers be-       4.57
tween the two dummy TWC     4.49
and two dummy oxidation
catalysts, feedback car-
buretor, open-loop at
wide open throttle and
under 120°F, medium car-
buretor response, 530 mil-
livolt control point, fab-
ricated exhaust.
               40.85
               42.13
               54.11
               50.65
               65.25
               72.82
               74.23
               50.96
               58.00
               57.70
               48.93
6..00
4.66
5.56
5.07
4.68
4.54
4.42
7.81
8.41
8.23
7.94
0.671*
0.674*
0.663*
0.689*
0.732
0.701*
0.713*
0.692
0.655**
0.642**
0.648**
*    Test conducted after installation of new piston and catalysts.

**   Test conducted after installation of new piston.

-------
                                          -11-
Version
Number

  12
  13
  15
  16
  17
  18
  19
                                     Table 2 (cont'd)

                                 IHC 404 Cold Start Tests
                                                           Emissions
                             HC
 CO
 NOx
 BSFC
Engine Configuration     (g/BHP-hr)"  (g/BHP-hr)"  (g/BHP-hr)   (Ib.  fuel/BHP-hr)
Version 11, except the      4.24
system is completely
closed-loop.

1978 production engine,     0.84
no EGR, one air pump,       0.49
air to air mixers be-       0.97
tween the two TWC and       0.77
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, medium carbur-
etor response, 530 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,     0.76
no EGR, no air pump, two
TWC catalysts, no oxida-
tion catalysts, feedback
carburetor, open-loop at
WOT and under 120°F, fast
carburetor response, 530
millivolt control point,
fabricated exhaust.

Version 15 except the       0.89
system is completely
closed-loop.

Version 15 with the car-    1.25
buretor response in the
slow position.

Version 15 except the       1.41
system is completely
closed-loop with the car-
buretor response in the
slow position.

1978 production engine,     1.27
no EGR, no air pump,        1.46
two TWC catalysts, no
oxidation catalysts, feed-
51.05
 5.74
 2.96
 3.11
 5.38
21.35
18.53
20.12
17.08
25.27
22.15
7.85
1.04
0.94
1.78
1.62
0.70
0.71
1.67
1.37
1.41
1.64
0.685
0.686
0.708
0.673
0.724
0.682
0.693
0.686
0.672
0.712
0.698

-------
                                          -12-


                                     Table  2  (cont'd)

                                 IHC  404  Cold Start Tests

                                                          Emissions
Version                                HC          CO          NOx           BSFC
Number    Engine Configuration     (g/BHP-hr)   (g/BHP-hr)   (g/BHP-hr)  (Ib. fuel/BHP-hr)

          back carburetor,  open-
          loop at WoT and under
          120°F, medium carburetor
          response,  530 millivolt
          control point, fabricated exhaust.

  20      Version 19 except the       1.30         17.07       1.53          0.699
          system is  completely
          closed-loop.

  21      1978 production engine,     0.71          4.11       2.07          0.698
          no EGR, one air pump,
          air to air mixers be-
          tween the  two TWC and
          two oxidation catalysts,
          feedback carburetor,
          open-loop  at WOT and un-
          der 120°F, fast carbur-
          etor response, 530 milli-
          volt control point, fab-
          ric ted exhaust.

  23      Version 21 with the car-    0.70          3.19       1.55          0.714
          buretor response time in
          the slow position.

  24      Version 21 except the       0.79          3.05       2.05          0.736
          system is  completely
          closed-loop with the car-
          buretor response in the
          slow position.

  26      1978 production engine,     0.87          3.22       1.93          0.738
          no EGR, one air pump,
          air to air mixers be-
          tween the  two TWC and
          two oxidation catalysts,
          feedback carburetor,
          completely closed-loop,
          medium carburetor re-
          sponse, 530 millivolt
          control point, fabri-
          cated exhaust.

-------
                                          -13-
Version
Number

  27
  35
  37
  41

  43
                                     Table 2  (cont'd)

                                 IHC 404 Cold Start  Tests
                                                          Emissions
                             HC
                CO
             NOx
               BSFC
Engine Configuration

1978 production engine,
no EGR, two air pumps,
air to air mixers be-
tween the two TWC cat-
alysts and two oxida-
tion catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, fast carburetor
response, 530 millivolt
control point, fabri-
cated exhaust.

1978 production engine,
EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, medium carbur-
etor resonse, 350 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,
EGR, no pump, two
TWC catalysts, no ox-
idation catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, medium carbure-
tor response, 350 milli-
volt control point, fab-
ricated exhaust.

Version 35 without EGR.

Version 19 at the 350
millivolt control point.
(g/BHP-hr)   (g/BHP-hr)   (g/BHP-hr)   (Ib.  fuel/BHP-hr)
   0.90
   0.81
   0.94
   0.75
   0.62

   1.20
   0.78
 4.00
 3.59
 9.72
17.98
 3.28

20.47
28.79
1.90
3.11
3.01
3.26
3.89

4.25
3.69
0.728
0.737
0.704
0.741*
0.691

0.643**
0.711*
*    Test conducted after installation of new piston and catalysts.

**   Test conducted after installation of new piston.

-------
                                          -14-
Version
Number

  45a.
  45c,
  45f
  45m.
  49
                                     Table 2  (cont'd)

                                 IHC 404 Cold Start  Tests
  53
Engine Configuration

1978 production engine,
EGR, one air pump, air
to air mixers between
the two TWC and two ox-
idation catalysts, feed
back carburetor, open-
loop at WOT and under
120°F, medium carbure-
tor response, 530 mil-
livolt control point,
fabricated exhaust.
Version 45 at the 710       0.68
millivolt control point.

Version 45 at the 630       0.56
millivolt control point.

Version 45 at the 630       0.65
millivolt control point.

1978 production engine,     4.35
EGR, one air pump,          5.29
air to air mixers be-
tween the two dummy TWC
and two dummy oxidation
catalysts, feedback
carburetor, open-loop
at WOT and under 120°F,
medium carburetor re-
sponse, 530 millivolt
control point, fabri-
cated exhaust, (version
11 with EGR).

Version 49 at the 350       4.47
millivolt control point.
Emissions
HC
(g/BHP-hr)
0.84
0.75
0.43
0.53
CO
(g/BHP-hr)
2.95
3.39
2.73
3.75
NOx
(g/BHP-hr)
1.13
1.94
1.42
1.44
BSFC
(Ib. fuel/BHP-hr)
0.713
0.649**
0.714*
0.750*
                                         3.60
                                         3.29
                                         3.33
                                        51.98
                                        46.06
0.74
0.71
3.78
5.36
4.69
0.704*


0.737*


0.748*
0.669
0.688*
                                        41.37
5.89
0.682
*    Test conducted after installation of new piston and  catalysts.

**   Test conducted after installation of new piston.

-------
                                           -15-
Version
Number

  55
                                     Table 2  (cont'd)

                                 IHC 404  Cold Start  Tests
Engine Configuration

Version 49 without EGR
at the 350 millivolt
control point.
Emissions
HC
(g/BHP-hr)
4.44
4.87
CO
(g/BHP-hr)
47.87
39.21
NOx
(g/BHP-hr)
8.33
7.74
BSFC
(Ib. fuel/BHP-hr)
0.760
0.659*
     Test conducted after installation of  new piston and  catalysts.

-------
                               -16-
     In  the  single  bed  catalyst  configurations  (TWC  catalysts
only) version  15-20,  37 and  43  approximate the  HC and NOx  stan-
dards but fall short of CO standards  by  some  25  percent.   For this
reason further discussion will be  confined  to  dual-bed  system con-
figurations.  The reader is referred  to  the SAE  document  as refer-
enced for further analysis of single bed configurations.[2]

     The effect of EGR on NOx emissions  can be seen in  the dualbed
configurations at  the  350 millivolt  (MV)  control  point   with  EGR
(version 35) as  compared  to  the similar configuration  without  EGR
(version 41.)   Here  an approximate  20  percent decrease in  NOx
emissions and  a  corresponding 6 percent decrease in fuel economy
can be seen.  The NOx  levels  however,  3.11  g/Bhp-hr for version 35
and  3.89 g/Bhp-hr  for version  41 both exceed  the proposed  NOx
standard.

     Configurations at  the  530 millivolt  control point that  meet
the proposed standard  indicated  a  9 percent increase in NOx  emis-
sions with  actual NOx  levels of  1.48  g/Bhp-hr  with EGR  (version
45, 4 test  average)  and 1.35 g/Bhp-hr  without EGR  (version  13,  4
test average).   These  averages are misleading however since  some
of the version 45 tests were  conducted  after  installation  of  a new
piston and  catalysts  (discussed in section D).   The only version
45 test  conducted before piston and catalysts  change yielded  a NOx
value of 1.13 g/Bhp-hr  or a 17  percent   reduction  in NOx  emis-
sions.   In general, the application of  EGR  resulted in  significant
NOx emissions reduction.

     The air/fuel ratio  (millivolt control  point) effect  on  emis-
sions is principally related  to NOx.   The  stoichiometric  point is
located  at   approximately  the   530   MV  control  point  with  rich
air/fuel ratios  at  higher MV  control  points  and  lean  air/fuel
ratios at  lower  MV  control  points.  Tests indicate a MV control
point independence  in  dual-bed  configurations  with regard  to HC
and  CO   control.   The  NOx  relationship  is   demonstrated by  the
series of hot-start  tests (version 45)  run at  various MV control
points (Table  3).   Here,  a requirement  for richer  than  stoichio-
metric air/fuel  ratios  to  meet  the proposed NOx standard  is  indi-
cated.   This engine would  have  to  run at approximately the 550 MV
control  point  to meet the NOx  standard of  1.7  g/Bhp-hr   (air/fuel
ration about 0.2 rich of stoichiometric.)

     The critical response   time  (CRT)  determines the   speed  at
which the air/fuel  ratio cycles around  the MV  control point from
full rich to  full lean.   A fast CRT  of 0.2 seconds, a medium CRT
of 2.8 seconds and a  slow CRT of  5.6 seconds  were  evaluated.  The
slow response  time,  according to  closed-loop  theory, produces the
tightest control under  steady state conditions but  not  necessarily
under transient  conditions.   This  is  represented by the  following
test  data.   Versions  21   and 23  represent individual  cold-start
tests while version 13 is the simple mean of four cold-start tests.

-------
                                     -17-






                                    Table 3




                            IHC 404  Hot  Start  Tests*

Version
Number
45a.










45b.

45c.

45d.

45e.

45f .

45g.

45h.

45i.

*5j.

45k.



Engine Configuration
1978 production engine,
EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WoT and
under 120°F, medium car-
buretor response, 530
millivolt control point,
fabricated exhaust .
Version 45 at the 740
millivolt control point.
Version 45 at the 710
millivolt control point.
Version 45 at the 685
millivolt control point.
Version 45 at the 655
millivolt control point.
Version 45 at the 630
millivolt control point.
Version 45 at the 605
millivolt control point.
Version 45 at the 575
millivolt control point.
Version 45 at the 550
millivolt control point.
Version 45 at the 495
millivolt control point .
Version 45 at the 440
millivolt control point.

HC
(g/BHP-hr)
0.30
0.30









0.43

0.42

0.40

0.36

0.33

0.39

0.31

0.38

0.38

0.32

Emissions
CO NOx
(g/BHP-hr) (g/BHP-hr)
0.61 2.02
0.81 1.97









0.89 0.93

0.94 0.92

1.07 1.18

0.84 1.00

0.74 1.21

0.82 1.63

0.64 1.54

0.69 1.88

1.02 2.28

0.77 3.01


BSFC
(Ib. fuel/BHP-hr)
0.693
0.665









0.689

0.689

0.687

0.658

0.678

0.608

0.675

0.684

0.647

0.653

All tests conducted after installation of new piston.

-------
                               -18-






                           Table 3 (cont'd)




                        IHC  404  Hot  Start Tests
                                                 Emissions
Version
Number
451.
45m.
Engine Configuration
Version 45 at the 385
millivolt control point.
Version 45 at the 350
HC
(g/BHP-hr)
0.27
0.24
CO
(g/BHP-hr)
0.56
0.55
NOx
(g/BHP-hr)
3.45
3.85
BSFC
(Ib. fuel/BHP-hr)
0.667
0.633
millivolt control point.

-------
                                -19-
                          Emission g/Bhp-hr     Ib.  fuel/Bhp-hr
Version   CRT (sec.)      HC     CO     NOx     	BSFC

  23         5.6         0.70   3.19    1.55         0.714
  13         2.8         0.77   4.30    1.29         0.698
  21         0.2         0.71   4.11    2.07         0.698

     Upon analysis  of  individual cold-start  tests  for version  13
(Table 2) and  the  Los  Angeles  Freeway  (LAF)  portion of  the  tran-
sient  test  as  presented in  the  referenced  SAE document  it was
determined that the best balance  of  emissions  and fuel  economy was
obtained with  the  2.8  second CRT.  This is further  exemplified  by
the  near  normal oxygen  sensor  voltage trace  as  indicated in the
same SAE document.[2]

     The effect of  open-loop at  WOT on emissions is presented  by
the comparisn of two dual-bed,  530 MV control  point  configurations
both without  EGR;  the  open-loop  at  WOT  version 13  and  the  same
configuration in the completely closed-loop mode version  26.   Ver-
sion 13 figures represent the simple mean of four cold-start tests.

                          Emission g/Bhp-hr     Ib.  fuel/Bhp-hr
Version     WOT-Mode      HC     CO     NOx     	BSFC

  13          open       0.77   4.30    1.29         0.690
  26         closed      0.87   3.22    1.93         0.738

Although open-loop  enrichment may result  in higher  emissions  (CO)
it was found  necessary  to  run open-loop at WOT in order to  main-
tain maximum engine horsepower  levels required  by heavy-duty  vehi-
cles.

     Maximum horsepower  levels  were  determined from WOT  map  tests
run  to establish  speed-load parameters  for  the  computer  driven
transient tests.  Results indicate  an average horsepower value  of
167  for  the  standard  1978   production engine  (version  10).  The
average in  the dual-bed, open-loop  at WOT configuration (version
45)  is  162  horsepower,  representing a 3  percent  loss in  maximum
horsepower.    The  corresponding  completely closed-loop  configura-
tion (version  46)  resulted  in an average  horsepower of  148  an  11
percent drop  from standard  1978  production.    Due  to  this  horse-
power loss subsequent testing was concentrated  in the  open-loop  at
WOT mode.[2]  Actual emission testing was  not  conducted on  version
46; mapping was done for horsepower determination only.

     The effect of  air  pump  capacity is demonstrated by a compari-
son  of  a dual-bed, 530 MV control  point, fast CRT configuration
without EGR and one air pump (version  21) and  the  same configura-
tion with  two  air  pumps (version  27).   Full  air  from  both air
pumps  was  directed to  the  air mixers.   Indicated  figures  repre-
sent one cold-start test for each version.

-------
                              -20-
                          Emission g/Bhp-hr     Ib.  fuel/Bhp-hr
Version    Airpump No.    HC_     CO     NOx     	BSFC	

  21           1         0.71   4.11    2.07         0.698
  27           2         0.90   4.00    1.90         0.728

     Since  a  reduction  in  fuel  economy was  shown  with two  air
pumps and  further since several  other  configurations produced  CO
emissions under  the  4 g/Bhp-hr level achieved with  two  air  pumps,
all subsequent testing was conducted in the single air pump mode.

     In analyzing the fuel consumption  figures, given in pounds  of
fuel/Bhp-hr or  Brake Specific  Fuel  Consumption  (BSFC),  consider-
able variation  for  any  given  version  is  indicated.   Thus  it  is
difficult to  determine fuel  economy  trends for  the various ver-
sions.   Table 4  indicates BSFC  for baseline configurations and  the
dual-bed closed-loop system with and without EGR.[2]

     Although slight fuel economy differences can be found between
individual  versions,  no significant  fuel  consumption  differences
can be  identified between baseline and  closed-loop  configurations
with and without EGR, between  single vs.  dual-bed   catalyst modes
or between open vs. closed-loop at WOT modes.

     Due to  the experimental nature of  this program  several test
runs were  accepted as valid that exceeded various   cycle perfor-
mance specifications.   High  engine  idle rates  that occur  during
the  early   portions  of  the   transient  cold-start   test  were
considered  the   primary  cause  of. this  situation.    This problem
could not be  corrected  by either software  or hardware  adjustments
due  to  time  and  equipment  constraints.   Since  the problem  was
limited to  the  first sample  bag overall emission effects  were  not
considered significant enough to void marginal  tests.

     A  problem  observed  with  the  preferred feedback  system  was
intermittent  engine  hesitation  on  strong accelerations.   This
occurred even with a warm  engine  and  intensified  with the addition
of  EGR.  The  problem,  which  occured  around  the   stoichiometric
air/fuel ratio,   is  generally  controlled  with the  rich  mixtures
typical of  current heavy-duty engines.   It is  theorized  that  the
problem  could  be  best  corrected  through manufacturing  design
changes.

     The SAE publication should be referred to  for a more complete
discussion of results of the test project.[2]

IV.  Conclusions

     The results  of  this test program indicated that  at low mile-
age  a  TWC dual-bed  catalyst system with  a closed-loop  stoichio-
metric  carburetor can yield  emissions substantially  below the 1984
HC and  CO  standards  and the statutory NOx  standard  for  heavy-duty
gasoline engines.

-------
                                -21-
                               Table  4

                      Heavy-Duty  Transient Test
                        Fuel Consumption  2/

                                            Fuel Consumption
Configuration                              (Lbs. Fuel/BHP-hr)

1978 Production Engine (See Table II)         0.671  1 run
Version 00                                    0.674  1 run

Version 9                                     0.663  1 run
                                              0.689  1 run

Version 10                                    0.732  1 run
                                              0.713  1 run
                                              0.701  1 run
                                Avg. of above 0.692 (7 runs)

Same Engine with Closed-Loop System           0.705 Avg. 17 runs
                                              (of various versions)
With EGR                                      St'd. Dev. = 0.032

Same Engine with Closed-Loop System           0.697 Avg. 16 runs
                                               (of various versions)
Without EGR                                   St'd Dev. = 0.032

-------
                                -22-
     Engine-out  HC  and  CO  emissions were  sufficiently high  how-
ever, that  the TWC single-bed  catalyst  alone could not be  utili-
zed.  Dual-bed catalysts were  required  and  were  able to  produce
high conversion efficiencies.  A number  of  key  improvements  to the
air/fuel preparation and  distribution as well as  improved  quality
of combustion  in  the  engine are required to  alleviate  the  problem
of  hesitation  upon acceleration.    These  improvements  are  con-
sidered feasible since they have been accomplished with light-duty
passenger car engines.

     Improved  air/fuel  distribution is  also  very  important  in
order to achieve maximum  engine horsepower  at the  leanest  possible
(but still net  rich) air/fuel ratios.  Maximum horsepower with the
closed-loop carburetor  and  catalysts was  approximately 3  percent
below the maximum  horsepower  of the  1978 production engine.   With
implementation of the above improvements this loss should  approach
zero.

     Brake  specific  fuel consumption  was   essentially  the  same
(within data scatter)  as obtained with the  1978 production engine.

     It would  appear   from  this test program,  even with  systems
that were not  optimized, that  TWC  closed-loop  systems could  pro-
vide efficient  emission  control of heavy-duty gasoline engines  as
well as  the same  overall  advantages as  currently achieved  with
similar  systems on light-duty  passenger  car  gasoline  engines.
Design  improvements to  the  heavy-duty  truck engine would  appear
justified as a result  of the feasibility study.[2]

-------
                               -23-
                             References

1.   "Gaseous Emission  Regulations  for 1984  and Later Model  Year
Heavy-Duty Engines,"  Federal Register,  Vol.  45,  No.  14,  Monday,
January 21, 1980.

2.   Hansel,   Dr.  James  G.;  Cox,   Timothy;  Nugent,  Thomas,  "The
Application of a  Three-Way  Conversion Catalyst System to  a Heavy-
Duty Gasoline Engine," SAE Paper 810086 (1981).

3.   Cox,  Timothy;  Diatchun, Zachary;  Nugent,   Thomas;  Passavant,
Glenn; Ragsdale,  Larry, "1972-73 Heavy-Duty  Engine Baseline  Pro-
gram and NOx  Emission Standard  Development,"  EPA  Technical Report
EPA-AA-SDSB-80-01, March 1981.

4.   Cox, Timothy P.,  "Heavy-Duty  Gasoline Engine  Emission Sensi-
tivity to  Variations in  the 1984   Federal  Test Cycle,"  SAE  Paper
801370 (1980).

-------
                                          -24-
Version
Number

  11
  12
  13
  15
  16
                                        Appendix A

                           IHC 404 Confirmatory Hot Start Tests
Engine Configuration
no EGR, one air pump,
air to air mixers be-
tween the two dummy TWC
and two dummy oxidation
catalysts, feedback car-
buretor, open-loop at
wide open throttle and
under 120°F, medium car-
buretor response, 530 mil-
livolt control point, fab-
ribated exhaust.

Version 11, except the
system is completely
closed-loop.

1978 production engine,
no EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, medium carbur—
ator response, 530 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,
no EGR, no air pump, two
TWC catalysts, no oxida-
tion catalysts, feedback
carburetor, open-loop at
WOT and under 120°F, fast
carburetor response, 530
millivolt control point,
fabricated exhaust.

Version 15 except the
system is completely
closed-loop.
:ion
mgine,
tump,
i be-

HC
(g/BHP-hr)
3.28
3.18
3.32

CO
(g/BHP-hr)
47.70
46.37
48.81
Emissions
NOx
(g/BHP-hr)
8.18
7.66
8.53

BSFC
(Ib. fuel/BHP-hr)
0.707
0.689
0.698
3.38
3.43
3.53
0.21
0.23
0.33
0.41
0.51
0.47
46.10
44.11
43.94
0.96
0.84
1.14
2.23
2.32
0.85
8.41
7.98
8.47
0.93
0.90
1.67
1.45
1.55
1.74
0.51
6.45
0.55
0.63
0.49
0.58
19.59
16.86
23.64
16.33
16.20
16.52
0.76
0.75
0.67
0.80
0.80
0.82
                                      0.698
                                      0.690
                                      0.704

                                      0.679
                                      0.692
                                      0.709
                                      0.717
                                      0.724
                                      0.664
0.700
0.765
0.707
0.706
0.686
0.712

-------
                                          -25-
Version
Number

  17
  19
  20
  21
  23
  24
                                   Appendix A (cont'd)

                           IHC 404 Confirmatory  Hot  Start  Tests
Engine Configuration

Version 15 with the
carburetor response in
the slow position.
          no EGR,  no air
          two TWC  catalysts,  no
          oxidation catalysts,
          feedback carburetor,
          open-loop at WOT and
          under 120°F, medium
          carburetor response,
          530 millivolt control
Version 19 except the
system is completely
closed-loop.
no EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, fast carbur-
etor response, 530 milli-
volt control'point, fab-
ricated exhaust.

Version 21 with the car-
buretor response time in
the slow position.

Version 21 except the
system is completely
closed-loop with the car-
buretor response in the
slow position.
Emissions
HC
:ion (g/BHP-hr)
:he 0.88
tse in 0.84
L. 0.85
ingine, 0.86
imp, 0.90
i, no 0.84
its, 0.78
:or, 0.85
and 0.75
.um
ise,
i i- m~\
LULU -L
I exhaust .
: the 1.04
:ely 0.76
0.86
ingine, 0.45
lump, 0.37
CO
(g/BHP-hr)
17.36
18.66
18.93
16.27
20.65
16.50
20.80
25.52
19.25


15.09
14.70
15.16
2.22
1.92
NOx
(g/BHP-hr)
1.77
1.76
1.65
1.51
1.17
1.44
1.41
1.12
1.59


1.51
1.36
1.37
2.11
2.02
BSFC
(Ib. fuel/BHP-hr)
0.686
0.699
0.683
0.693
0.686
0.676
0.696
0.740
0.700


0.691
0.695
0.697
0.698
0.695
0.34
0.40
0.71
0.39
0.40
0.85
0.96
0.57
0.72
0.73
2.17
2.14
 .91
 .95
                                                              1.97
0.710
0.712
0.735
0.729
0.726

-------
                                          -26-
Version
Number

  26
  27
  35
  37
                                   Appendix A (cont'd)

                           IHC 404 Confirmatory  Hot Start  Tests
Engine Configuration
no EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
completely closed-loop,
medium carburetor re-
sponse, 530 millivolt
control point, fabri-
cated exhaust.

1978 production engine,     0.64
no EGR, two air pumps,      0.58
air to air mixers be-
tween the two TWC cat-
alysts and two oxida-
tion catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, fast carburetor
response, 530 millivolt
control point, fabri-
cated exhaust.

1978 production engine,     0.28
EGR, one air pump,          0.33
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, medium carbur-
etor response, 350 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,     0.42
EGR, no air pumps, two      0.43
TWC catalysts, no oxi-
dation catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, medium carbure-
tor response, 350 mil-
livolt control point,
fabricated exhaust.
:ion
mgine ,
lump,
i be-

HC
(g/BHP-hr)
0.45
0.42
0.42

CO
(g/BHP-hr)
0.75
0.79
0.87
Emissions
NOx
(g/BHP-hr)
1.71
1.83
1.76

BSFC
(Ib. fuel/BHP-hr)
0.727
0.727
0.722
2.10
2.10
0.98
1.40
5.48
7.75
2.04
2.12
3.08
3.14
3.07
3.14
0.726
0.725
0.726
0.717
0.682
0.688

-------
                                          -27-
Version
Number

  41
  43
  45a.
  49
                                   Appendix A (cont'd)

                           IHC 404  Confirmatory  Hot  Start  Tests
Engine Configuration
no EGR, one air pump,
air to air mixers be-
tween the two TWC and
two oxidation catalysts,
feedback carburetor,
open-loop at WOT and un-
der 120°F, medium carbure-
tor response, 350 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,     0.92
no EGR, no air pump,        1.02
two TWC catalysts, no
oxidation catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, medium carburetor
response, 530 millivolt
control point, fabricated
exhaust.

1978 production engine,     0.38
EGR, one air pump, air      0.41
to air mixers between
the two TWC and two ox-
idation catalysts, feed-
back carburetor, open-
loop at WOT and under
120°F, medium carbure-
tor response, 530 milli-
volt control point, fab-
ricated exhaust.

1978 production engine,     3.70
EGR, one air pump,          3.51
air to air mixers be-
tween the two dummy
TWC and two dummy
oxidation
Emissions
HC
:ion (g/BHP-hr)
:ngine, 0.47
lump, 0.50
CO
(g/BHP-hr)
1.23
1.17
NOx
(g/BHP-hr)
3.93
3.89
BSFC
(Ib. fuel/BHP-hr)
0.689
0.688
19.46
23.82
 1.39
 0.85
48.65
48.31
3.03
2.63
0.99
1.13
5.51
5.45
0.649**
0.636**
0.697
0.705
0.722
0.642
**  Test conducted after installation of new piston.

-------
                                          -28-
Version
Number

  49
  53
  55
                                   Appendix A (cont'd)

                           IHC 404 Confirmatory  Hot  Start Tests
                                                          Emissions
                             HC
                CO
             NOx
               BSFC
Engine Configuration

catalysts, feedback
carburetor, open-loop
at WOT and under 120°F
medium carburetor re-
sponse, 530 millivolt
control point, fabri-
cated exhaust, (version
11 with EGR).

Version 49 at the 350
millivolt control point.

Version 49 without EGR
at the 350 millivolt
control point.
(g/BHP-hr)  (g/BHP-hr)   (g/BHP-hr)   (Ib.  fuel/BHP-hr)
   3.54
   3.42

   3.72
   3.71
33.16
33.89

38.78
39.34
5.85
5.95

8.68
8.74
0.672
0.662

0.645
0.649

-------