EPA/AA/CTAB-88-05
                        Technical  Report
          Emissions, Fuel Economy, and Performance of
             Light-Duty CNG and Dual-Fuel Vehicles
                               by
                       Robert I.  Bruetsch
                           June  1988
                             NOTICE

     Technical Reports  do not  necessarily represent  final  EPA
decisions or positions.   They are  intended  to  present technical
analysis of  issues  using  data which  are  currently  available.
The purpose in the release of such reports  is  to  facilitate the
exchange of  technical information and  to inform the  public of
technical developments which may form the basis for  a final EPA
decision, position or regulatory action.

              U.  S.  Environmental Protection Agency
                   Office of  Air and  Radiation
                    Office of Mobile Sources
              Emission Control Technology Division
           Control Technology and Applications Branch
                       2565 Plymouth Road
                   Ann Arbor, Michigan  48105

-------
        UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

                    ANN ARBOR: MICHIGAN  48105
                                                         OFFICE OF
                                                      AIR AND RADIATION
June 28, 1988

MEMORANDUM


SUBJECT:    Exemption  From Peer and Administrative Review
FROM:       Karl  H.  Hellman,  Chief
            Control  Technology and Applications Branch

TO:         Charles  L.  Gray,  Jr.,  Director
            Emission Control  Technology Division
     The  attached report  entitled "Emissions,  Fuel  Economy and
Performance   of   Light-Duty   CNG   and   Dual-Fuel   Vehicles,"
(EPA/AA/CTAB/88-05)   describes   MVEL   testing   of  AGA  and  Ford
compressed natural gas  vehicles and trucks.

     Since  this  report  is  concerned  only with  the  presentation
of  data  and  its  analysis and does  not  involve matters of policy
or   regulations,  your   concurrence   is   requested   to  waive
administrative  review according  to the policy  outlined  in your
directive of April 22,  1982.
   Concurrence ;   ' ,..--<^:-----      <*>7 i  _ Date :
                Cn~arles  L.  Gray,  j. ,/Dir . ,  ECTD
Nonconcur rence :	Date :
               Charles  L.  Gray,  Jr.,  Dir., ECTD

cc:  E. Burger,  ECTD

-------
                                             EPA/AA/CTAB-88-05
                        Technical  Report
          Emissions, Fuel Economy, and Performance of
             Light-Duty CNG and Dual-Fuel Vehicles
                               by
                       Robert I.  Bruetsch
                           June  1988
                             NOTICE

     Technical Reports  do not  necessarily represent  final  EPA
decisions or positions.   They are  intended  to  present technical
analysis of  issues  using  data which  are  currently  available.
The purpose in the release of such reports  is  to  facilitate  the
exchange of  technical  information and  to inform the  public  of
technical developments which may form the basis for  a final  EPA
decision, position or regulatory action.

              U.  S.  Environmental Protection Agency
                   Office of  Air and  Radiation
                    Office of Mobile Sources
              Emission Control Technology Division
           Control Technology and Applications Branch
                       2565 Plymouth Road
                   Ann Arbor, Michigan  48105

-------
Table of Contents


I .
II.



III.





IV.





V.

VI.
VII.
VIII
IX.


Background 	
Test Program 	
A. Test Vehicle Descriptions 	
B. Test Fuel Descriptions 	
C. Test Procedures 	
Exhaust Emissions and Fuel Economy Calculations
A. Net Heating Value 	
B. CNG, Hydrocarbon and Methane Densities . .
C. FID Correction Factor 	
D. Non-Methane Hydrocarbons 	
E. Fuel Economy 	
Test Results 	
A. Calculated Exhaust Emissions 	
B. Gasoline Equivalent Fuel Economy 	
C. Comparison to Standards 	
D. Comparison to Gasoline 	
E. Performance Data ... 	
Interpretation of Test Results 	
A. Variability and Assumptions 	
Conclusions 	
Acknowledgments 	
.References 	
APPENDIX 	
Page
Number
. , 1
. . 3
. . 4
. . 5
. . 6
. . 6
. . 7
. . 7
. . 8
. . 9
. , 11
. . 13
. . 13
. . 18
. . 18
. . 18
19
, , 19
19
22
23
24
. . A-l

-------
I.    Background


     Natural   gas   has   been   proposed   as   an   alternate
transportation  fuel   for  some  applications,   especially  in
metropolitan  areas requiring  additional  carbon monoxide  (CO)
emission control.

     Because  of  its very  low  energy density, natural  gas  must
be  compressed  (CNG)  or  liquified  (LNG)   to store  onboard  a
vehicle;  even  then  the   energy  density  does  not match  that
typically achieved with gasoline.  The  results achieved to  date
with  natural  gas  as  a  vehicle  fuel  place  constraints  on
operating   range,   fuel   storage   volume,   and  load-carrying
capacity.

     U.S.  natural  gas  reserves,  while  extensive,  are  being
depleted.   In  addition,  the  capacity  of   the  current   gas
pipeline  network  to  handle  the  distribution  of  significant
additional   natural   gas  for   vehicular   use   is   somewhat
limited.[1]*   Natural gas prices vary  widely across the United
States.   To  recover  the  cost of  vehicle  conversion  and the
added  cost  of compression  equipment  for refueling  stations,
natural  gas   must  be  very  favorably   priced  (compared  to
gasoline) to  achieve  enough  in fuel-cost  savings 'to recover the
capital  equipment  cost within a short payback  period.

     For  the  reasons described above,  natural  gas  tends  to be
used today  in niche markets  (such as certain  fleet operations),
where  vehicle  range  and/or  load-carrying   capacity  are  not
limiting factors,  where  fuel can be purchased at relatively low
cost   in commercial-level  quantities,  and   where  centralized
refueling can be utilized.

     CNG  fuel   system    technology   is   developed,   but   the
capability  of  CNG   vehicles  generally  lags  behind  that  of
gasoline-fuel  systems.   CNG  at its present stage of development
is  best for  centrally fueled  fleet vehicles that  have   ample
storage  volume and payload and that follow daily routes of  less
than 100 miles.

     Most   CNG  fuel   systems  are   completely  mechanical  in
operation  and  control,   and  are designed for  use  as  "second"
fuel  systems—in  addition to the  gasoline  systems.   Although
this  results  in  a vehicle capable  of  using   two separate  fuels
 (a  "dual-fuel"  vehicle as distinguished from a  "flexible-fuel"
vehicle),   operational  performance  with   either   fuel  may  be
compromised.
      Numbers  in  brackets  denote  references  at  end  of  paper.

-------
     Because  CNG is  a gaseous  fuel,  it  displaces  air,  which
otherwise could be used in the combustion process,  and this can
reduce  the engine's  maximum  power   output.   CNG  fuel  systems
also impose penalties on the vehicle  in weight  and fuel storage
volume.   All  of  this  reduces  vehicle  performance,  payload
capability, and overall fuel economy.

     The cost  to convert an  existing vehicle to  use CNG  as  a
fuel is  typically in the  range  of  $1,000  to $1,500, depending
on the number of storage tanks.  Dual-fuel  operation is usually
maintained.   Dedicated  CNG   vehicles  (single-fuel  vehicles)
could be produced by the automobile  manufacturers  for less of a
differential, but  some incremental  cost  increase is inevitable
because CNG storage tanks are  much more expensive than gasoline
storage  tanks.   The  cost  of  compressing  natural  gas  to 2,400
pounds  per  square  inch   is   significant.   The  operating  and
maintenance  cost of  compression are  in  the range  of  $0.10 to
$0.20  per  equivalent gallon  of  gasoline.   The capital  cost of
the  compressors must  be  added to  this  cost  to  arrive  at  a
total, delivered cost of CNG.[1]

     CNG-fueled  vehicles  have  not  been  tested  by  EPA  for
emissions,  fuel  economy  and performance  for  the  past seven
years.  EPA  representatives  attended a gas  industry meeting in
Indianapolis   in  September   1987   to  view  state-of-the-art
post-1981 dedicated  and dual-fueled  CNG light-duty vehicles and
trucks.  The  intent of this  visit was to  develop a  cooperative
test program to  acquire emission data on updated CNG  vehicles.

     Previous  EPA  data  [2]  have  shown  reductions  in   carbon
monoxide  and  non-methane  hydrocarbon  emissions  of  pre-1981
retrofit   CNG-fueled   vehicles   compared   to    operation   on
gasoline.   NOx emissions and  vehicle performance were somewhat
degraded  from  these same  vehicles  compared  to  the  gasoline
baseline.   The   significant   changes  in  gasoline   engine  and
vehicle  technology  brought  about by the  much  more stringent
passenger  car emission standards that took  effect  in 1980 and
1981  have  changed the context in which  natural  gas and other
alternative   fuels   need   to  be  considered.   First,   it  was
believed that changes  in vehicle manufacturer catalyst and  fuel
metering technology since  1981  might have a significant  effect
on   the   exhaust  emissions   of   in-use   CNG-fueled  vehicles.
Second,  the much lower gasoline-fueled vehicle emissions  give a
much  lower  baseline with which CNG  vehicles  must compete.  EPA
determined  that it  would  be valuable  to  develop emission  data
on  post-1981  gasoline-fueled  vehicles  that  had  been converted
to  dual-fuel  applications.   The  American  Gas Association  (AGA)
agreed to  assist  in this  endeavor  and  a  request  for vehicles
was  sent  to AGA. [3]   A  test program  for  both  dual-fuel and
dedicated  fuel  CNG vehicles was developed.

-------
                              -3-

II.   Test Program


     The intent  of  this test program  was to work  with the AGA
and Michigan Consolidated Gas Co., Inc.  (MichCon) to  obtain and
test a  range of late  model  CNG vehicles.   The  basic objective
was  a   comprehensive  characterization  of the  emissions,  fuel
economy  and  performance of  CNG vehicles  in order  to  permit  a
broad evaluation of the use of CNG as a transportation fuel.

     In  addition to obtaining  an emissions  database  of current
CNG  vehicle  technology,   this   test  program was   initiated  to
compare  the  test  results  to   those   of  similarly  equipped
gasoline vehicles as well  as to the applicable Federal emission
standards and other vehicles in  the  same equivalent test weight
class.[4]  An additional objective  was  to sample the exhaust of
CNG vehicles for formaldehyde (HCHO) emissions  to provide first
time   test   results   of   this   pollutant   (if  emitted)   from
CNG-fueled  vehicles.    HCHO  emissions  from CNG  vehicles  have
historically  been  assumed to  be low  or  zero  and therefore  a
benefit  of   CNG  utilization,  but these claims have  not  been
based on actual test results.

     The  CNG test  vehicles  are described  in  Table  1.   Four
vehicles  were  supplied   for  this  program.   Limited  vehicle
supplier  generated test   data  exist for  these  vehicles.[5,6]
Three  of the vehicles  supplied through AGA  are  dual-fueled
(operable on CNG or unleaded  gasoline) and only one of these,
the  1984 Oldsmobile  Delta 88,  is  a high  mileage vehicle.   All
dual-fuel vehicles  are equipped  with  a three-way  catalyst plus
closed-loop  air/fuel  ratio control.  Two of these vehicles,  a
1987 Ford LTD  Crown  Victoria  and  a  1987  Chevrolet  Celebrity,
were tested  under  two  different configurations as  noted in the
emissions results  in  a  later  section.   The Crown  Victoria was
tested  with  two different engine control calibrations  and the
Celebrity   was   tested   in  the   "as-received"   and  "after
maintenance" configurations.

     The 1987 LTD  Crown  Victoria  was  configured  as  a police
car, but has also  been suggested for taxicab application.  This
vehicle was  originally   calibrated  somewhat  rich  to  obtain
improved  NOx   control.     This  calibration   proved   to  be
unacceptably rich  when tested  over  the Federal Test  Procedure
(FTP).   High total  HC and  CO  emissions  were  measured, though
low   NOx   numbers   were   obtained.    The  vehicle   supplier
recalibrated the vehicle  and brought it back to be retested (on
CNG  only) with a leaner calibration.

     The   1987   Chevrolet    Celebrity    exhibited   vehicle
driveability problems   in  the  form  of  false  starts  and stalls
and  also had  to  be   recalibrated  after  the  first  round  of
testing.  There were  also repairs  to  the engine control system
which  resolved  the false  start/stall problem  that  existed with
the  gasoline-fueled testing  of  this  vehicle.

-------
                                    -4-

                                  Table 1

                       CNG Test Vehicle Description
Model Type
Mf r.
MY
Displacement
Carburetion
Comp. Ratio
Rated HP
Control
System
No. of Cyls.
Transmission
ETW
Axle Ratio
N/V
ADHP
Fuel Type
Delta 88
GM
1984
307
4
7.9
140
EGR/PMP/OXD/3CL
8
L3-1
4000
2.41
30.2
10.7
Dual-Fuel
Crown Victoria
Ford
1987
302
FI
8.9
160
EGR/PMP/OXD/3CL
8
L4-2
4250
3.27
30.0
13.1
Dual-Fuel
Celebrity
GM
1987
173
FI
8.9
125
EGR/3CL
6
L4-2
3250
3.33
32.4
7.3
Dual-Fuel
Ranger
Ford
1984
140
1
12.8
80
EGR/PMP/OXD
4
M4-1
3000
3.45
48.0
10.0
CNG-Only
Exhaust Emission Control System:

EGR  =     Exhaust gas recirculation
OXD  =     Oxidation catalyst
PMP  =     Air pump
3CL  =     Three-way catalyst + closed loop

-------
                              -5-

     None of  the dual-fuel vehicles were  tested with different
spark  timings  for  CNG than  gasoline  operation,  although  the
Delta  88 was  equipped with a switch for advanced  timing in the
CNG  test mode.   CNG  vehicles  are  routinely  calibrated  with
10-12° more  spark advance to account  for  the  slow  flame speed
and  reduced  power  output  of   the  fuel.   Higher   emissions,
particularly NOx,  are usually  the tradeoff for  advanced spark
timing on  these  vehicles.   An increase  of 55  percent  in  NOx
emissions  was  observed  in  the   vehicle  supplier  test  data,
though this  increase was not verified by EPA  as  part  of  this
test program.

     One  dedicated  CNG-fueled   Ford  Ranger  pickup  truck,
supplied by  Ford Motor Company,  was  also tested.   This truck
was  calibrated  to have the same engine dynamometer  rated power
and  similar  (only 2 percent slower 0  to 50 mph)  performance as
an  identically  equipped  gasoline-fueled  1984  Ford  Ranger.   A
primary  difference  is the higher  compression  ratio  of  the
CNG-fueled Ranger at  12.8:1  versus the gasoline-fueled Ranger
at 9.0:1.

     Test  fuel  for  CNG  vehicle   evaluations  was  provided  by
MichCon  in Melvindale, Michigan.[7-11] Vehicles were driven to
the  refueling station  (usually on  gasoline),  CNG   fuel tanks
were  filled  to 2500 psi,  and attempts  were made to  obtain fuel
analyses of  the  fuel  from  the  storage cascade  for  each batch
EPA  used.   MichCon was not  able  to supply  fuel analyses every
time  a vehicle was  brought  in  for  refueling.   Fuel parameters
of  the CNG used  to fuel  a specific vehicle were  assumed to be
the  same as  those from the  most  recent fuel  analysis  received
from  MichCon prior  to the time  that vehicle was tested.   The
standard EPA emissions  calculation program  is  not  set up  to
handle  emissions  from CNG-fueled  vehicles  so  certain  MichCon
fuel  parameters   (e.g.,  power heat   value,   density,  weight
percent  carbon,  etc.)  were  needed to determine estimates of
exhaust emissions and fuel economy.[12,13]

      Indolene  (HO)  test  fuel  was  used   for  all non-CNG  test
sequences  on  dual-fuel   vehicles   as   an  unleaded  gasoline  to
compare the emissions, fuel economy  and performance  obtained on
CNG  from the same  vehicles.   The Indolene used in  the EPA lab
is  supplied by Howell Hydrocarbons,  San Antonio, Texas.

      Each  vehicle was  tested  over  the Federal  Test  Procedure
(FTP)  and  the Highway Fuel Economy  Test (HFET).  If the vehicle
was  a dual-fueled  test  car,  it was  run  over  the FTP and HFET
cycles  at  least twice on each  fuel  (CNG  and  Indolene)  or until
repeatable  results  were  obtained.   Each  dual-fueled  car  was
also tested  for  5  to  60  MPH and 30  to 60  MPH performance.
These accelerations  were  repeated  five times  after each  vehicle
FTP  cycle.   The dedicated CNG  truck was tested  over the FTP and
HFET  cycles  on  CNG  only,  but  was   not  tested  by   EPA  for
acceleration performance.

-------
                              -6-

     All   cyclic   test   sequences   included   sampling   for
formaldehyde   emissions  as   well   as   emissions   of   total
hydrocarbons,  methane,  carbon  monoxide,  carbon  dioxide  and
oxides   of    nitrogen.     Methane   sampling    required   the
incorporation  of a  high  range  analytical  span gas.   Methane
emissions  from  gasoline-fueled  vehicles  are  relatively  low
(20-25  percent  of  a  low  total  HC) and can be  characterized
fairly  accurately  in the low  range of  the  instrument  using  a
nominal  50  ppm span  gas.   CNG  fuel  and  its  emissions  from
light-duty  vehicles  can be  from 75  to  95  percent  methane
requiring  an  instrument  with measuring  capacity well above a 50
ppm concentration.   Therefore, a 450 ppm span gas  was  procured
to  provide  more  accurate  measurement  of  high range  methane
analyzer   concentrations.   Actually,  several   span  gases  at
various  concentrations  should  be characterized  to develop  a
true  calibration curve for the  methane  analyzer,  but  complete
calibration  of the  methane  analyzer for CNG was  not  provided
for as  part of  this evaluation  program.   Measured  methane  is
thought  to   be  in  most   cases  higher  than  actual  methane
concentrations  since the measured  methane  subtracted  from the
relatively  well characterized total  hydrocarbon concentrations
yielded  non-methane  concentrations which were  generally either
zero  or negative.   Since negative nonmethane  hydrocarbons  do
not represent an actual physical  result, a variety of methods
for estimating non-methane hydrocarbons  were  developed as will
be discussed  in more detail in the next section.
Ill. Exhaust Emissions and Fuel Economy Calculations


     As  mentioned  above,  the  standard  EPA exhaust  emissions
calculation  computer  program  is  not set  up to  determine the
emissions  and  fuel  economy of  CNG-fueled vehicles.   The volume
of  CNG  testing over  the years  has  not  been  high  enough to
justify  developing subroutines  for  calculating  and correcting
for  factors affecting CNG exhaust emissions and  fuel economy.
Therefore,  the program was run  assuming  the test  fuel  used is
Indolene   and   a  separate  program  used  to  correct  for  the
particular  CNG  fuel  batch  used  in  the   test  vehicle.   The
exhaust  emission results  are  also  modified by  a factor which
accounts  for  the difference  in HC analyzer  response to methane
span gas  (used.as a surrogate for CNG) instead of using propane
span gas which is used for Indolene.

     Once  a fuel analysis was obtained  for  a given batch of CNG
used  in  a particular vehicle during testing, the values of  heat
content,  molecular  weight,   specific  gravity,  weight  percent
carbon,   and   weight  percent  "hydrocarbon"  of   the  fuel   were
determined.   These  fuel  analyses  upon  which our  calculations
were based were performed by MichCon and referenced to 60°F and
atmospheric pressure.

-------
                              -7-

     A.     Net Heat Value

     Vehicle testing was  performed under  laboratory conditions
of   75-77°F   and   atmospheric   pressure.    Variation   in  the
barometric pressure was  negligible.   The  Federal Register  and
Code  of  Federal   Regulations,  Title   40,   Part  86.   144-78,
"Calculations; exhaust emissions,"  list densities of  regulated
exhaust   emissions  determined  at    68°F   and  760   mm   Hg
pressure.[14]  Since the  allowable temperature  of certification
specification vehicle  testing is 68-86°F  and the CFR densities
are  68°F  values,  the   gas   analyses  parameters  were   also
converted  to 68°F  or  were  used on  a  mass  basis  (e.g.,  heat
content) in the calculations.

     The heating value reported on  the MichCon Gas  Analyses  is
the  gross  (high)  heat  value  (HHV)  of  the  natural  gas and  is
reported on  a  dry basis.  It  is standard practice, when  using
fuels  in engines  where  the  exhaust  water is not condensed  to
provide  energy,  to  use the net  (lower) heat value (LHV) of  the
fuel in  the  calculation  of fuel economy.  Since the MichCon gas
analyses provided  do  not  report LHV  of the  fuel,  several  gas
property  references  were examined   to  see  if  there  is  a
consistent  LHV/HHV  ratio.[15,16,17]    As  it  turns  out,   this
ratio  is consistently  very  close  to  0.90  in  each  reference
cited.    Since  a direct measure of  LHV  was  not  available,  and
the  gas  industry  could  not  suggest  a  better  value or  rule  of
thumb,  the  HHV numbers  supplied by  MichCon were multiplied  by
0.90   and  these   values  were  used   in  the   fuel   economy
calculations  as  the LHV.   This way,   a  reasonable approximation
of   the  LHV  of  the  actual   fuel  in  the  test  vehicle  was
determined,  we  avoided requiring  additional  fuel analysis  for
LHV  by  MichCon,   and  a  separate factor for   the  difference
between  the  fuel used  and the  "national  average" CNG  fuel  did
not  have to  be developed.  MichCon specification requires their
fuel energy  HHV to be  between  1000  BTU/SCF  and  1050  BTU/SCF.
As  a result, the  LHVs obtained in  this calculation methodology
are  between 900  BTU/SCF and 945  BTU/SCF.

     B.    Density

     Since   they  are  thought  not   to   affect   the  chemical
reactions  that  produce   photochemical  oxidation,  the  weight
percent  of  C02/  He,  and N2   are  deleted  from  the  density
when figuring  the "total  HC"  and   "non-methane  HC"  emissions
from vehicles  run on  natural  gas.    EPA calculated  this  "HC
density"  based  on  the weight  percent  and molecular  weight  of
all  hydrocarbon  components of the fuel  (e.g.,  methane, ethane,
propane, butane, pentane,  hexane, heptane, octane, etc.).

     The  entire  fuel   composition   (including   C02,   N2  and
trace  He) is  used in  the determination  of  the overall "CNG
density"  to  be  used in  the  fuel consumption calculation.   This
is  appropriate  since  the vehicle  operator  pays  for  the C02,
He,  and  N2  they  get  with  the rest  of the  fuel  and  it  is
"consumed" (i.e.,  used)  in the  operation of  the vehicle even if

-------
                              -8-

it is  not  consumed in  the  combustion process.  Therefore,  the
density used  in  the fuel economy  calculation will  be  slightly
higher than  that used  to  calculate  HC emissions  depending on
the amount of CO2, He,  and N2 in the fuel.

     The  actual  values  used  depend  on  which  MichCon  fuel
analysis  is   used.   The  weight  percent  carbon  is  reasonably
consistent in  the fuel analyses obtained  throughout this  test
program at  about 0.74,  so  this  value  is  used as  a constant.
Table  2  shows  a  summary  of  the  values  determined from  the
MichCon  analyses.   These  values   are  used  as  inputs  to  our
calculations.

                             Table  2

          Natural Gas Properties Used For Calculations


                               HC Density (g/ft3) at  68°F and
  Date of                  	One Atmosphere For	
  MichCon         LHV                  CNG Emissions
Gas Analysis     BTU/g     CNG MPG      (% Methane)      Methane

11/18/87         42.8       20.91          19.30          18.89
                                          (96.3)

02/26/88         42.3       21.61          20.43          18.89
                                          (91.3)

03/16/88         43.5       20.79          19.90          18.89
                                          (94.5)

04/29/88         43.2       21.35          20.26          18.89
                                          (93.5)

05/31/88         43.1       21.38          20.27          18.89
                                          (93.5)

     Note  that  although  the  variability of  the Table  2   fuel
properties  are low  (3  to 6 percent),  individual  fuel analyses
were  matched  up  with  the  vehicles.   Properties  for each  fuel
were   used  where   appropriate   to  obtain  the  most  accurate
emissions and  fuel  economy  estimates for each vehicle.

     C.    FID Correction Factor

     A cursory investigation was  made into the appropriateness
of  using  propane as  a span gas  for  the HC  analyzer,  a "cold"
Beckman   Model  400   flame  ionization  detector   (FID),   when
sampling  hydrocarbons  from CNG-fueled vehicles.   It  is  known
that    HC   analyzers    respond    differently   to   different
hydrocarbons.[18]   Also,  since CNG is largely  methane,  and  no
"standard"  CNG  fuel  is  currently used,  tests  were  run on  two
methane  span  gases  through  the   HC-FID  to   compare   analyzer
response  to  methane to the response obtained when spanning  with
propane.

-------
                              -9-

     The response to  a  450  ppm methane  span gas  is  10  percent
higher than with  propane,  and the response  to  a  50 ppm methane
span gas is 12  percent  higher.   Taking an average,  the  methane
response is 11'percent  higher than the  propane response on the
analyzer.  Therefore, the FID correction factor  (the  number to
multiply the  total  HC  concentration by  to  account for  use of
methane span gas)  is 0.90  or the  reciprical of  1.11.   The FID
correction was  applied  to the total HC g/mile value on each CNG
test to  account for  the use of methane  as  a  more appropriate
"surrogate" span gas for natural gas than propane.

     D.    Non-Methane HC

     As  a  result  of  problems  encountered  obtaining  accurate
methane   measurements,   different  ways   to    calculate   the
non-methane  HC  emissions   from   the   CNG-fueled  cars  were
developed,   because   non-methane   HC   are   very   important
contributors to oxidant formation.

     The  non-methane HC results  are determined  by subtracting
methane  measurements from  total HC measurements.   The  methane
instrument  used in  our laboratory  was  developed to measure the
methane  levels that are  typically seen from  gasoline-fueled
cars,  typically 0.1 grams per mile.  During the development of
the  instrument, it  was  not  evaluated with the  high levels that
are  typical from  the CNG-fueled vehicles.   For example, a total
HC value  for the  CNG-fueled cars of 3.5  grams per  mile is not
unusual  to measure.  This  implies  that  the methane levels the
methane analyzer sees are up  to 30  times  higher than the values
from  gasoline-fueled  cars.   EPA  purchased  a  special  higher
range  span gas  for the CNG car  testing.   Using  our existing
span gas,  the   analyzers all  read full  scale  when testing with
CNG-fueled  cars.    This  was  discovered  during  testing  of  the
first  CNG  vehicle,  the  Delta 88.   It is  fair  to say that it is
pushing these methane analyzers to  read higher levels than  seen
heretofore.   In  addition,  if  the  exhaust  HC  distribution is
anything  close to  the  fuel  HC distribution,  the  analyzer is
also seeing a  much different  HC distribution  than was  seen
during  the  process   of   developing   the   instrument.    The
instrument  may well  be measuring  other light  hydrocarbons as
methane.

     When  some  of  the preliminary  calculations of  the emissions
from the vehicle  tests  were made,  it was noted that some  of the
results  from  the  calculation of  non-methane  HC computed  were
negative.   This is a non-physical result.   It  would appear  that
the  appropriate value  to  use would be  zero when  the  computed
result  from   the   two  different  analyzers  spanned  on  two
different   calibration  gases  is   negative.   The  results  so
obtained   are   reported  as   0.00  g/mile  NMHC.   This   negative
result may be  attributable  to the  methane  analyzer  counting
other  light hydrocarbons as  methane.   Table  3  shows  a  summary
of the NMHC calculation approaches.

-------
                              -10-

                             Table 3
                 Various Methods to Compute NMHC
               Emissions from CNG-Fueled Vehicles
                 Method 1    Method 2    Method 3*    Method 4

HC Density        16.33       16.33       Various      Various
                                        see Table 2  see Table 2

Methane Density   16.33       16.33        18.89

FID Response       1.00        0.90         0.90        0.90

NMHC mass equals   Yes         Yes          Yes          No
HC mass minus
measured
CH4 mass?
*    For the  test  results with  gasoline  as the  fuel,  Method 3
     uses 16.33, 18.89, 1.00 and Yes.


     Method 1  is  the  data as it comes  from the computer.   This
assumes  a  density of HC  equivalent  to  the  value  used  for
testing with  Indolene.  This method  yields results  we believe
are  inaccurate since  they are unadjusted  for  CNG fuel, but may
be useful for  comparison  to other unadjusted  CNG data.  Method
2 is the  same as  Method 1,  except the  FID correction factor is
applied to  the HC  results.   Method  3  is  the  same  as  Method 2
except densities  of HC and methane have  been  adjusted as shown
in  Table  2.[19]   Method  4 calculates  non-methane  HC as  the
total  HC   values   (adjusted  for   the  FID  response  factor)
multiplied  by the  non-methane  fraction  of the  CNG  fuel.   In
other words, method 4  does not use the measured methane values.

     The  author   places   more   confidence   in   NMHC  numbers
generated using method 4  than the other methods  for CNG-fueled
vehicles since  it  includes  the  corrections for fuel density and
FID response and does   not rely on  methane measurements.  Method
4 NMHC  results may also  be  low  since  the  exhaust  may contain
non-methane HC from burned  lubricating  oil and therefore higher
NMHC emissions.  Gasoline-fueled vehicles  are  best described by
results from  NMHC  method 3,  which includes the  correction  for
the density of  methane.   NMHC emissions are compared in Section
IV.  "Test  Results" and Figure 2  by using  CNG results determined
from method 4  versus gasoline results determined  from method 3.

-------
                              -11-

     E.    Fuel Economy

     The  fuel  economy   calculation   for   natural  gas  fueled
vehicles  was  determined  on  a  gasoline  equivalent  basis  by
carbon  balance using  the weight  percent  carbon  and  net  heat
value of the fuel.[20,21]  The generalized  expression for miles
per  gallon  is obtained  by  dividing  the  net heat  value  of
gasoline  (BTU/gal)  by  the CNG  energy  expended  per  distance
traveled   (BTU/mile).    The   working  equation  and  a  sample
calculation are shown below.
       MPG =  BTU/qal = 	BTU/qal	
             BTU/mile   (gC/mile)(gCNG/gC)(BTU/gCNG)

Where:

   BTU/gal =  Net heat value of Indolene = 114,132 BTU/gal


  BTU/mile =  BTU of energy consumed on natural gas per mile


   gC/mile =  Grams carbon emitted per mile (g/mile)
           =  (wgt. fraction C)(HC) + 0.273 (C02) "+ 0.429 (CO)


   gCNG/gC =  Reciprical of weight percent carbon of CNG fuel
           =  1/0.74 = 1.35
  BTU/gCNG =  Net heat value of CNG (BTU/g).  See Table 2 for
              values
Example Gasoline Equivalent Fuel Economy Calculation

Assume:  HC =  1.62 g/mile   CO2 = 326 g/mile   CO = 0.1 g/mile
         LHV =42.8 BTU/gCNG

GEFE  = 	114,132  (BTU/qal) 	
        [(0.74MHC) +  (0.273)(C02) +  ( 0 . 429 ) (CO) ] ( 1. 35 ) ( 42 . 8)

      = 	i	114,132  (BTU/qal)	
        [(0.74)(1.62)  H-  (0.273X326)  +  ( 0 . 429 ) ( 0 . 1) ] (1 . 35) ( 42 . 8 )

      = 114,132  BTU/qal   =   21.9 miles
         5,214   BTU/mile            gal

      Generally  the   gasoline  equivalent   fuel   economy  of
CNG-fueled  vehicles   is  somewhat  higher  (about  7-12 percent)
than  the same vehicle  operating  on gasoline.    Therefore,  on
that  basis,  the  dual  fuel vehicle is more fuel  efficient  on CNG
than  on gasoline.

-------
                              -12-

     However,  the  results obtained  on CNG  were with  vehicles
that did  not  match the  performance  obtained on  gasoline.   Our
performance  tests   of  the  vehicles  operating  on  CNG  showed
substantially  reduced  performance,   measured  as  the  time  to
accelerate between two speeds on  the  chassis  dynamometer.   This
loss in performance was 25 to 35 percent.

     In  order  to  make  the  fuel  efficiency  comparisons  at
constant  performance,  two avenues are  possible.   One  would  be
to adjust the fuel  economy  data on CNG to account  for  the loss
in  fuel  economy that  would  be  expected  from  an   increase  in
performance to  match  the performance  on  gasoline.  The  other
approach  is to  adjust  the  fuel economy  result   obtained  on
gasoline  to the  gain in fuel economy that would  be expected to
result  from making  the  gasoline-fueled  configuration perform
the  same  as on  CNG.   We chose the  latter  since there is much
more  information   about  the  performance   and   fuel   economy
relationships for gasoline-fueled vehicles.

     Based on previous work [22], the sensitivity is 0.454 or,

     % A MPG =   (0.454) % A T

Where:

     % A MPG =   Percent change in fuel economy  •

       %  A T =   Percent  change  in  0  to  60  MPH   acceleration
                 time  (seconds)

     If,  for  example,  CNG gasoline  equivalent fuel  economy is
ten  percent higher than the  same vehicle on  gasoline  (say 17.6
MPG  on  CNG  and  16.0 MPG on gasoline),  but the CNG  5 to  60 MPH
test  took 14 seconds  while  the  same  test took only 11 seconds
using  gasoline  as  the  fuel,   what  would  the  fuel  economy
difference between CNG and gasoline be at constant performance?

     Making  the gasoline-fueled  vehicle  performance equivalent
to  the CNG  vehicle performance  is  a 27  percent  increase  in
acceleration time.   Assuming  that the sensitivity based on 5-60
accelerations   is   the   same   as  the  sensitivity  for   0-60
accelerations  the  0.454 value  can be  used.  Using  the  above
equation, this  translates into a 12.3 percent  increase in fuel
economy for  the gasoline-fueled vehicle,  or  18 MPG.  Therefore,
at  constant  performance  the  vehicle would  exhibit  about   2
percent   better  fuel  economy  (18/17.6  =  1.02)  when run  on
gasoline  than when  run on CNG.

-------
                              -13-

IV.  Test Results

     The  exhaust  emission   and   fuel  economy   test   results
obtained  in  this  test  program   are  displayed   in  Figures  1
through  6  and Table 4.   The  data in  Table  4 include  the fuel
analysis  information  used in  the calculations,  results  of  the
four NMHC  (g/mile)  methods calculated,  the  total HC,  CO, C02/
NOx  and  HCHO results,  and the gasoline  equivalent fuel economy
for  each test sequence.  Ford Ranger  CNG results  are  compared
to  Ford-generated  1984  gasoline-fueled Ranger  emissions  and
fuel economy.  Figures  1  to  6 show  the  FTP  emissions  and fuel
economy  results  of  CNG versus gasoline  operation.   Data points
are  displayed relative  to the line of equality between the  two
fuels.   In the  discussion  of these test results,  emphasis is
placed  on  the   four  data  points  that  represent  the   latest
vehicle  test configurations  though  the test  results  for  the
original  Crown   Victoria  and  Celebrity  calibrations   are  also
included.

     The  total  HC results shown  in  Figure  1  indicate that when
fueled with CNG the vehicles  emit between 4 to  10  times more
total  HC than when  fueled with gasoline.  All the tests on CNG
exceed the  level of the 0.41 gram per mile total HC standard.

     The  results for  non-methane  HC shown in Figure  2  show a
different   trend than   that   seen  for  total  HC.    The  NMHC
emissions when CNG is the fuel are lower than with gasoline.

     Carbon monoxide results  are  mixed as the data  in  Figure 3
show.  The initial  calibrations  on  the Crown Victoria- and the
Celebrity  were  worse  on CNG than on gasoline.  In fact,  it was
these  results that prompted  the  vehicle  developers  to  modify
the  vehicles.   After  the  vehicles  were   modified   their  CO
emissions  were  lower,  but the  initial  emission  tests  were the
only indication  that  vehicle  modification was  needed.   Vehicle
operation  did not signal a need  for adjustment or maintenance.
One  'vehicle, the Crown  Victoria  version  1,  exceeded  the  3.4
gram per mile  CO standard on CNG.   The Delta 88 exceeded the
3.4  CO standard  on gasoline.

     NOx emissions were also mixed,  but  in  this  case it was one
vehicle,  the Crown Victoria,  that  provided  results counter to
the  expected trend  of  higher  NOx  on  CNG.    The   Celebrity
exceeded 1.0 gram per  mile NOx standard on both calibrations on
CNG.   The  Delta 88 also exceeded  the  1.0  gram  per  mile  NOx
level  on  CNG.   The Crown  Victoria exceeded the  1.0  gram per
mile NOx standard when using  gasoline.   Since the  Ranger is  a
light-duty truck,  it  did not have to  meet the 1.0 gram per  mile
standard.

      In   reference   [23]   EPA   provided   some   guidance   for
estimating the   emissions of  CNG-fueled  vehicles.  The guidance
in  [23]   were prepared before  this test  program was  run.   To
compare  those  values  to the  ones  in this  report,  the  average
results  on CNG  were divided  by  the average  result on gasoline
fuel for  each  vehicle tested  on both  CNG  and  gasoline.  The
resulting  averages  are  shown   in   Table   5  along   with   the
guideline values.

-------
                    -14-
                  Figure 1
            HC  Emissions
                 (g/mile)
CNG Emissions
                         3.55
                         3.19
    GASOLINE BETTER THAN CNG
                             2.46
                              X
                  1.55     1.49
                   a   1.30 ^
       0.1


 x  Delta 6ft
 °  Celebrity 2
                           CNG BETTEB THAN GASOLINE
  0.2      0.3     0.4
Gasoline (HO) Emissions
   v Ranger
   » Crown Vic 1
    0.6

Crown Vic 2
Celebrity 1
Q.6
                  Figure 2
             NMHC Emissions
                  (g/mile)
  CNG Emissions
u.oo
0.3
0.25
0.2
01 £
.1 9

0.1
- 	 	
GASOLINE BETTER THAN CNG



-""^^
^^"^^
.. •'


^"".0.06
0.05- ^ ^ CNG BETTER
Q---~
1
0 0.05 0.1 0.15
Gasoline (HO)
Delta 88 " Ranger
* Celebrity 2 Crown Vic


01 a
.19 	
9 0.16

0.09 0.09
" '* 	 A
0.08
THAN GASOLINE
0.2 0.25 0.3 0.35
Emissions
- Crown Vic 2
1 * Celebrity 1

-------
                      -15-
                    Figure 3
              CO  Emissions
                   (g/mile)
  CNG Emissions
       GASOLINE BETTER THAN CNG
                      CNG BETTED THAN GASOLINE
   *  Delta 66

   a  Celebrity 2
                   4        6
              Gasoline (HO) Emissions
       Ranger         *  Crown Vic 2

       Crown Vic I      A  Celebrity 1
                    Figure 4
             NOx  Emissions
                   (g/mile)
2.5
  CNG Emissions
     GASOLINE BETTER THAN CNG
                        CNG BETTER THAN GASOLINE
 0 ~-

  0
0.5
              Gasoline (HO) Emissions
     Delta 88

     Celebrity 2
       Ranger

       Crown Vic I
1.5
•



Crown Vic 2

Celebrity 1

-------
                     -16-
                   Figure 5
           HCHO  Emissions
                 (mg/mile)
 CNG Emissions
                           5.87
                           ••• *	
     GASOLINE BETTER THAN CNG
                        CNG BETTER THAN GASOLINE
  *  Delta 86
  a  Celebrity 2
                234
             Gasoline (HO) Emissions
            *•  Ranger
            *  Crown Vic 1
*  Crown Vic 1
*  Celebrity 1
                   Figure 6
               Fuel  Economy
              Gasoline Equivalent
                  (miles/gallon)
25

20

15

10
  Natural Gas MPG
                             Ctlobrlty 2
                                 a
CNG BETTER THAN GASOLINE  Ctown vlctorla
                   Deltaa"
                       GASOLINE BETTER THAN CNG
 5r
                   10        15
                  Gasoline MPG
                                20
              25
             a  Vehicle Data
                       GAS • CNG

-------
                                                             Table 4
                                              Exhaust Emissions and Fuel Economy of
                                        Dedicated CNG and CNG/Gasoline Dual-Fuel Vehicles

Test
Number
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.
Avg.


Date
HO
HO
HO
HO
11/18/87
03/16/88
03/16/88
04/29/88
05/31/88
11/18/87
HO
HO
HO
HO
11/18/87
03/16/88
03/16/88
04/29/88
05/31/88
11/18/88

Test
"Vehicle Type BTU/g
Delta 88 FTP
Crn Vic FTP
Celebrity! FTP
Celebrity2 FTP
Delta 88 FTP 42.8
Crn Vic 1 FTP 43.5
Crn Vic 2 FTP 43.5
Celebrity! FTP 43.2
Celebrity2 FTP 43.1
Ranger FTP 42.8
Delta 88 HFET
Crn Vic HFET
Celebrity! HFET
Celebrity2 HFET
Delta 88 HFET 42.8
Crn Vic 1 HFET 43.5
Crn Vic 2 HFET 43.5
Celebrity! HFET 43.2
Celebrity2 HFET 43.1
Ranger HFET 42.8

HC Den
for MPG




20.9
20.8
20.8
21.3
21.4
20.9




20.9
20.8
20.8
21.3
21.4
20.9

HC Den
(emiss)




19.3
19.9
18.9
20.3
20.3
19.3




19.3
19.9
19.9
20.3
20.3
19.3

Methane
percent




96.3
94.5
94.5
93.5
93.5
96.3




96.3
94.5
94.5
93.5
93.5
96.3

Methane
Density




18.89
18.89
18.89
18.89
18.89
18.89




18.89
18.89
18.89
18.89
18.89
18.89
Total
HC
(g/mi 1
0.404
0.354
0.346
0.260
2.456
3.191
3.550
1.493
1.550
1.362
0.082
0.063
0.015
0.020
0.838
1.545
2.073
0.719
0.780
0.782
Nonmethane HC
Method
]
0.259
0.251
0.322
0.250
0.000
0.027
0.147
0.015
0.100
0.191
0.059
0.034
0.012
0.020
0.000
0.027
0.148
0.009
0.000
0.127
Method
2




0.000
0.000
0.000
0.000
0.050
0.082




0.000
0.000
0.000
0.000
0.000
0.049
Method
3
0.317
0.235
0.319
0.242
0.000
0.000
0.002
0.000
0.090
0.119
0.055
0.029
0.011
0.012
0.000
0.000
0.046
0.003
0.000
0.074
Method
4




0.082
0.158
0.176
0.087
0.090
0.057




0.028
0.076
0.103
0.042
0.050
0.026
                                                                                                      HCHO     CO    C02    NOx  GEFE
                                                                                                    (mg/mi)  (g/mi)  (g/itii) (g/mil  MPG
                                                                                                      4.03
                                                                                                      3.65
                                                                                                      4.41
                                                                                                      5.80
                                                                                                      4.82
                                                                                                        94
                                                                                                        87
                                                                                                        63
                                                                                                      4.31
                                                                                                      4.66
                                                                                                       .43
                                                                                                       ,66
                                                                                                      0.72
                                                                                                        01
                                                                                                        41
                                                                                                        87
                                                                                                      5.97
                                                                                                      0.90
                                                                                                      1.62
                                                                                                      1.38
9.84
  35
  01
  25
  69
4.34
0.45
2.55
0.00
0.04
1.89
0.10
0.02
0.10
1.19
5.55
0.00
2.62
0.00
0.01
632
582
435
434
464
429
417
354
324
328
349
456
335
342
267
339
365
252
250
251
0.40
1.07
0.60
0.60
1.18
0.65
0.93
1.63
1.19
1.98
0.29
0.67
0.63
0.30
1.13
0.40
1.54
2.08
2.20
2.47
13.67
15.15
20.30
20.30
15.30
16.02
16.66
19.83
21.90
21.83
25.15
19.45
26.46
25.90
26.69
20.23
19.23
27.78
28.50
28.64
"The Crown  Victoria  and  Celebrity  vehicles were  each  tested with two different calibrations.

-------
                              -18-

                            Table 5

                    CNG Emissions Comparison

                                     CNG/Gasoline
     Pollutant             This Report        Reference [23]

      NMHC                    0.46                0.60
      CO                      1.25                0.50
      NOx                     1.83                1.40

     Formaldehyde emissions were  also measured  during this test
program.  Formaldehyde, due to  its photochemical reactivity and
carcinogenicity,   is  of great  interest  for  both standard  and
alternate  fuel  vehicles.   The  results  in  Figure  5  show that
formaldehyde  is  emitted from CNG-fueled  vehicles and  that  the
results  are  about the  same  as the formaldehyde  emissions from
the tests using gasoline as a fuel.

     As  discussed  previously   the  gasoline  equivalent  fuel
economy  using CNG was slightly  better  than the results using
gasoline.    Gasoline   equivalent  fuel   economy   is  generally
improved  on  CNG  operation relative to gasoline  operation by 7
to  12  percent.   These  results,  though on  an energy equivalent
basis,    are   somewhat   misleading   relative    to   vehicle
performance.  The 5  to 60  MPH  performance  measured  on these
dual-fueled  vehicles  was significantly  degraded  by  about  29
percent  on  CNG   relative to  gasoline  operation.   Using  the
equation  for fuel  economy  as a function of  performance this
translates  into  4 percent higher  fuel  economy  on  gasoline  at
equivalent  performance.  On  the  road  in  actual use,  both CNG
and  gasoline MPG are  compromised on dual-fuel  vehicles  due  to
the  added  weight  of  having  two fuel  systems.   We did  not
account  for  this weight penalty in our tests.

     The  city,  highway and  composite gasoline  equivalent fuel
economy  for  each vehicle  are listed  in Table 6.  These data are
compared  to  measured  MPG  on gasoline for  the  dual-fueled cars
and  the  certification gasoline  fuel  economy  for  all vehicles
using  data  from  similarly equipped same model  year  vehicles  on
the  EPA  Test  Car List. [24,25]   The measured  CNG  and gasoline
fuel  economy are also compared to:   1) the fuel economy of all
vehicles  in the  same inertia  weight  class  (IWC)  from  the
appropriate  model year test car list;  and 2) the applicable MPG
standard for the  same particular gasoline  vehicle  (or  truck)
class  as  the CNG  or dual-fuel vehicle.

      As  mentioned above,   the measured CNG  fuel economy on  an
energy equivalent basis is 7 to  12 percent higher than the same
vehicle  operating  on  gasoline.   The  CNG  gasoline  equivalent
fuel   economy  is  7   to 15  percent  lower  than  the  comparable
certification   gasoline-fueled   vehicle   mileage   for   the
dual-fueled   vehicles,   and   essentially    the  same  as  the

-------
                              -19-

certification  light-duty truck  mileage for  the  dedicated  CNG
Ranger.    Measured  CNG  fuel   economy  is   low   compared  to
certification  gasoline-fueled  vehicles  in  the  same  inertia
weight  class  for  the  Delta  88,  Ranger,  and Celebrity,  and
higher  for  the Crown  Victoria.   Measured CNG  fuel  economy are
lower  than  the  applicable  model year  MPG  standards  for  all
dual-fueled  vehicles  and   20   percent   higher  for  the  1984
dedicated CNG Ranger.

     Table  7  shows  the  performance  data  measured  on  CNG  and
gasoline  for all  dual-fueled  vehicles  in  this  test  program.
The  dedicated CNG  Ranger  was   not  tested  for performance  on
natural gas, but judging from the rated power  of the CNG engine
and  estimates  provided by  Ford,  the  performance  of  this  truck
can  be   expected   to   be  nearly  equivalent   to   that   of   a
similarly-equipped   1984  gasoline-fueled   Ranger   truck.    The
reduced power output of  the  dual-fuel  vehicles  operating on CNG
is evident,  however,  with  23  to 33 percent slower  5  to 60 MPH
acceleration  times  and 29  to  36 percent  slower  30  to  60  MPH
acceleration times on CNG relative to gasoline operation.

     The  relationship   of   changes   in  MPG  with  changes  in
performance  is  shown in Figure  7.   These  data agree with  the
historical   gasoline   data,  but   show  higher   increases   in
performance  times  for  a  given  increase  in MPG  with  CNG
operation.   For  the  vehicles   fueled  on  CNG  to have  better
constant  performance  fuel  economy than on  gasoline  their  data
points would have to lie above the line on Figure 7.   None do.


V.   Interpretation of Test  Results


     One  must  be  careful   in   interpreting  the  test  results
obtained   in  this   CNG   evaluation   program.    Only   three
dual-fueled   vehicles    and   one  dedicated   CNG   truck   were
evaluated.  Two of the dual-fueled vehicles, the Crown Victoria
and  the Celebrity,  had to be recalibrated  for  better emissions
and  driveability  before being  retested.    The   other  dual-fuel
vehicle,  the Delta  88,  was not  tested with  advanced ignition
timing,   though  the  vehicle  was  equipped  with  a  switch  to
advance the  spark  10  to  12  degrees.   This vehicle may also have
not been  tested  in its optimum  form  since the  advanced timing
would be  expected to  compensate for  CNG's slower  flame  speed
and improve performance.   However, data supplied by  the  vehicle
supplier  indicate  higher  HC,  CO,  and NOx  emissions with  the
spark advanced.   The  Ranger  truck  appeared  to  be  calibrated
properly  as  received,  but  showed   markedly  different  fuel
economy than the results obtained by Ford.  Also,  it represents
only one data point for dedicated CNG  vehicles.

-------
                              -20-
                                 Table 6
                  Test Vehicle Fuel  Economy Comparisons
   Test Vehicle

CNG City (MPG)**
CNG Highway (MPG)**
CNG Composite  (MPG)**

HO City (MPG)
HO Highway (MPG)
HO Composite (MPG)

Cert City (MPG)*
Cert Highway (MPG)*
Cert Composite (MPG)*

IWC City (MPG)*
IWC Highway (MPG)*
IWC Composite  (MPG)*

Applicable MPG
Standard
                Delta 88  Ranger
                  15.3
                  26.7
                  18.9
                  27.0
21.8
28.6
24.4
13.7
25.2
17.2
17.3
25.9
20.3
17.3
26.8
20.6
15.0*
25. 6 +
18. 4*
21.5
29.5
24.5
23.5
31.9
26.7
20.3
                                       Crown Victoria
                                                    Celebrity
1
16
20
17










.0
.2
.7
15
19
17
17
26
20
13
19
15
2
16.7
19.2
17.7
.5
.4
.0
.2
.0
.3
.1
.8
.4
1
19.
27.
22.
20.
26.
22.

8
8
7
3
5
7














22


3
5
26






2
3
2
1
1
5







.0
.6
.6
.6
.9
.3
2
21.
28.
24.
20.
25.
22.







9
5
4
3
9
5






26.0
26.0
* *
From EPA Test Car List .
Gasoline equivalent fuel
Ford fuel economy data.
                              economy.
   Vehicle
Delta 88
Crown Victoria
Celebrity
                       Table  7

            Test Vehicle Performance Data

             Acceleration Times (Seconds)

                      Fuel           5-60 MPH
                   Gasoline
                     CNG
                  slower on
                                 CNG
                   Gasoline
                     CNG 1
                     CNG 2
                 slower on CNG 1
                 slower on CNG 2

                  Gasoline 1
                  Gasoline 2
                     CNG 1
                     CNG 2
                 slower on CNG 1
                 slower on CNG 2
            11.4
            14.8
             30

            10.8
            13.3
            14 .2
             23
             31

            10.6
            10.5
            13 .7
            14 .0
             29
             33
           30-60 MPH

               8.1
              10.8
               33

               7.6
               9.9
              10.3
               29
               36

               7.8
               7.7
              10.6
              10.5
               36
               36

-------
                      -21-
                     Figure 7.

   Fuel  Economy  Vs.  Performance
  % Change in Fuel Economy
20
15
     % delta MPG = (0.454) % delta T  xx


                            x   Delta 88
101- [[[ x

  !                   x               ^
  j                  X'              Celebrity 2
  >        •        X'^
  !               X
  i

 5 \- ............................................ X: .......................... Crow.n-VJoior-ia-2 ...............................................
   !          ."'
   i         X              f — I        I" I
   .

   i       xx                    Crown Victoria 1
    :-l                          Celebrity 1
    Ranger
-5L
             10         20         30         40


-------
                              -22-

     The emissions,  fuel  economy  and  performance  data  suggest
the  need for  further  work  in  the  optimization  of  dual-fuel
vehicle calibration.  Clearly,  CNG dual-fuel vehicles have the
potential  to provide  very  large CO  emission  reductions,  but
perhaps at the expense of  increased  NOx  emissions and decreased
performance  and   engine   power   output.    Vehicle  compression
ratios and  ignition  timings need to be chosen  carefully  or may
need to  be variable  in  order to  not  significantly degrade the
emissions,  efficiency,  and  performance  of  the  vehicles  on
either  fuel.   For   example,   compression  ratios  typical  of
current practice  using  gasoline  (generally about  9:1)  are much
lower than  the  optimum for CNG operation which may be closer to
13:1.   Conversely,  optimum  CNG   spark   timings  may  be  too
advanced for efficient  gasoline  combustion.   The development of
more advanced technology,  electronic fuel  metering systems, and
optimized,  dedicated,  CNG vehicles  would undoubtedly  enhance
the clean use of  CNG.[26]

     The measurement  and  analytical  procedures  for the accurate
determination   of   methane  and   formaldehyde   emissions  of
CNG-fueled vehicles  may need further development.  As mentioned
earlier,  the characterization of  the  methane analyzer response
to high concentrations of  exhaust  methane  needs to be performed
in  order  to place  more confidence  in the measured levels seen
from  CNG-fueled  vehicles.   Interpretation  of  these  methane
measurements  as  being  representative  of  CNG  vehicle  methane
levels yields NMHC estimates  which are very low.  Problems were
encountered  in  the  sampling  and  analysis   of  formaldehyde
emissions  throughout this  program which may have  added  to the
variability  of  measured  HCHO  results  by  unknown quantities.
The  consistency of  the  results  of  these  particular  HCHO  tests
and  their  levels,  i.e.,   below  5 mg/mile and  similar for both
CNG  and   gasoline,   indicate  that  these   HCHO  results  are
relatively   accurate,  particularly  for   comparison   of  CNG
formaldehyde to HCHO from  vehicles using other  fuels.


VI.  Conclusions
     This CNG vehicle test  evaluation program was useful in the
 characterization   of  late  model  CNG  vehicle  emissions,  fuel
 economy  and performance for comparison to  vehicle operation on
 other  fuels.   Some of  the more  significant  findings  of this
 study  are listed  below:

     1.     CNG  vehicle  calibration  techniques  are  critical  to
 low  emission performance.   For example, the Crown Victoria in
 its  as received  condition  was  calibrated  by  the supplier and
 yet  had  CO  emission levels much higher than  expected  on  CNG, so
 high that  they exceed  the current  CO  standard1.   The  vehicle
 exhibited   no   overt  driveability problems  while  operating  in
 this condition.   The  vehicle  was  recalibrated  on  an emission
 test chassis rolls and  in  its  second configuration  demonstrated
 a reduction in  CO over  the  gasoline  emission  level.

-------
                              -23-

     2.    Further  work   in  the  optimization   of   dual-fuel
vehicle  calibration  is  needed  for  the  efficient,  clean,  and
effective use of  both fuels in  light-duty  vehicles.   Effective
feedback fuel metering using CNG may be necessary.

     3.    Methane  analyzer  calibration using a  series of span
gases   could  improve  the  results.     Constructing   such   a
multipoint  calibration curve would  make methane analysis more
like the analysis used for the other gaseous pollutants.


VII. Acknowledgements


     The  author  would like  to  acknowledge Jeff  Seisler  (AGA),
Roberta  Nichols   (Ford),  Rich  Polich   (Consumers  Power),  James
Magan  (Total  Fuels),  Chris Bruch  (Garretson),  Tom  Minerick
(Wisconsin  Gas)  and Bill  Lampert (MichCon) for their assistance
and  cooperation   in  vehicle and  CNG  fuel  acquisition  required
for  this  test  program.   The  author  also wishes  to  recognize
Ernestine   Bulifant,   Bob  Moss  and  Ray  Ouillette  for  their
efforts  in  administering  the  vehicle   tests  and  assisting with
the  sampling and analysis  of  emissions test  data,  and Marilyn
Alff for  assisting  in  the report preparation.

-------
                              -24-

VIII. References
     1.    "Assessment  of Costs  and Benefits  of Flexible  and
Alternative  Fuel  Use  in  the U.S. Transportation Sector,"  U.S.
Department of Energy, Washington, DC, January 1988.

     2.    "Evaluation  of  the  Impact   on  Emissions  and  Fuel
Economy  of  Converting  Two Vehicles  to Compressed  Natural  Gas
Fuel,"  Penninga,  Thomas  J.,  EPA-AA-TEB-81-19,  U.S.  EPA,  Ann
Arbor, MI, June 1987.

     3.    "Vehicles  for  CNG  Emission Testing,"   letter  from
Wallace  D.  Tallent,  U.S.  EPA,  Ann  Arbor,  MI,  to  Jeffrey
Seisler, AGA, Arlington,  VA, November 6, 1987.

     4.    "Test  Plan:   Dedicated CNG  and  Dual-Fuel Vehicles,"
memorandum from Robert  I.  Bruetsch to Charles L.  Gray, Emission
Control  Technology  Division,  Office  of  Mobile  Sources,  U.S.
EPA, Ann Arbor, MI, October  19,  1987.

     5.    "The   Development  of  Ford's  Natural  Gas  Powered
Ranger,"  Tim Adams,  Ford Motor  Company, Society of Automotive
Engineers Paper Number  852277,  March 1985.

     6.    Letter  from  Stephen  A.  Carter,  CNG   Fuel  Systems,
Brampton,  Ontario,   CANADA   to  Richard  A.  Polich,  Consumers
Power, Jackson, MI, December  3,  1987.

     7.    "Gas   Analysis Report,"   Michigan  Consolidated  Gas
Co., Run No. 87-676,  November  18, 1987.

     8.    "Gas   Analysis Report,"   Michigan  Consolidated  Gas
Co., Run No  88-69, February  23,  1988.

     9.    "Gas   Analysis Report,"   Michigan  Consolidated  Gas
Co., Run No. 88-102,  March 16,  1988.

     10.   "Gas   Analysis Report,"   Michigan  Consolidated  Gas
Co., Run No. 88-207,  May  5,  1988.

      11.   "Gas   Analysis Report,"   Michigan  Consolidated  Gas
Co., Run No. 88-241,  June 1,  1988.

      12.   "Test   Data   Entry  Update,"  Laboratory  Emissions
Calculation  Computer  System,  U.S.   EPA,  Engineering Operations
Division, Ann Arbor,  MI,  1988.

      13.   "Michigan  Terminal Systems:  1200S," computer program
for emissions  calculation,   U.S.  EPA,  Certification Division,
Ann Arbor, MI,  1988.

-------
                              -25-

VIII. References (cont'd)
     14.   1987 Code of Federal Regulations, Title  40,  Part 87,
144-78, Federal Register, Washington, DC, 1987.

     15.   Internal   Combustion   Engines  and   Air  Pollution,
(Third  Edition),  Obert,  Edward  F.,  Harper  &  Row,  Inc.,  New
York, NY, 1973, pp. 235, 242.

     16.   The   Internal   Combustion   Engine   in  Theory  and
Practice, Vol. 2, Taylor, Charles F., MIT  Press,  Cambridge, MA,
1985, p. 121.

     17.   Gas  Engineers  Handbook,   American  Gas  Association,
Industrial Press, Inc., New York, NY, 1965, p.  2/48.

     18.   "Analysis  For Exhaust  Gas Hydrocarbons-Nondispersive
Infrared  Versus  Flame-Ionization,"   Marvin W.  Jackson,  Journal
of  the Air Pollution  Control  Association,  Volume  II,   No.  12,
December 1966.

     19.   46 Federal Register, No.  246, December 23,  1981.

     20.   "Calculation  of  Emissions   and  Fuel  Economy  When
Using  Alternate Fuels,"  EPA  460/3-83-009,  Charles  M.  Urban,
Southwest Research Institute, March 1983.

     21.   "Gasoline  Equivalent  Fuel Economy  Determination for
Alternate  Automotive   Fuels,"   Harvey,   Craig  A.,  SAE  Paper
820794, U.S. EPA, June  1982.

     22.   "Adjusting  MPG  for  Constant  Performance,"  memorandum
from  Karl H.  Hellman  to  Charles   L.   Gray,  Emission  Control
Technology  Division,  Office  of  Mobile  Sources,  U.S.  EPA,  Ann
Arbor,  MI, May 19, 1986.

     23.      "Guidance  on  Estimating   Motor  Vehicle   Emission
Reductions  from the  Use  of Alternative Fuels and  Fuel Blends,
EPA-AA-TSS-PA-87-4,   "Emission  Control  Technology   Division,
Office of Mobile Sources,  U.S.  EPA,  Ann Arbor,  MI, January 29,
1988.

     24.   "Control  of Air Pollution  from  New  Motor   Vehicles
and New  Motor  Vehicle  Engines:    Federal  Certification Test
results for  1984 Model  Year,"  Certification Division, Office of
Mobile Sources, U.S.  EPA, Ann Arbor,  MI, June  1984.

     25.   "Control  of  Air Pollution  from New  Motor   Vehicles
and New  Motor  Vehicle  Engines:    Federal  Certification  Test
Results for  1987 Model Year," Certification Division, Office of
Mobile Sources, U.S.  EPA, Ann Arbor,  MI, June  1987.

     26.   "Development of  Toyota Electronically Controlled CNG
Vehicles,"   Kimbara,  Yoshiro,   Touru  Ichimiya,  Masatake  Kan,
Shouji  Katsumata  and   Shunichi  Kondo,   Higashifuji  Technical
Center, Toyota Motor  Corporation, 1986.

-------
                              -26-

Additional References:
     27.   Letter from Robert I. Bruetsch, U.S.  EPA,  Ann Arbor,
MI,  to Jeffrey  Seisler,  American  Gas Association,  Arlington,
VA, February 2, 1988.

     28.   Letter from Robert I. Bruetsch, U.S.  EPA,  Ann Arbor,
MI, to  Roberta Nichols,  Ford Motor  Co., Dearborn, MI, March 10,
1988.

     29.   "The   Emission  Characteristics   of  Methanol   and
Compressed Natural  Gas  in Light Vehicles,"  Jeffrey  A.  Alson,
U.S.  EPA,  Ann Arbor,  MI,  Air Pollution  Control  Association
publication No. 88-99.3, June 1, 1988.

     30.   Handbook   of_   Compressed   Gases,    Compressed   Gas
Association, Van Nostrand Reinhold Co., New York, NY,  1966.

     31.   Compressed Natural Gas as a Motor Vehicle Fuel,  SAE
P-83/129,  Conference Proceedings of the Society of  Automotive
Engineers, Pittsburgh, PA, June 22-23,  1983.

     32.   Letter  from Karl  H. Hellman, U.S.  EPA,  Ann Arbor,
MI,  to John  M.  Arnold,  Institute   of  Gas  Technology, Chicago,
IL, May 2, 1988.

-------
IX.   APPENDIX

-------
Test GAS ANL Vehicle Test
Number DATE Type
881059 11-18-87 DELTA 88 FTP
881061 11-18-87 DELTA 88 FTP
881125 11-18-87 DELTA 88 FTP
AVG. 11-18-87 DELTA 88 FTP
881060 11-18-87 DELTA 88 HFET
881062 11-18-87 DELTA 88 HFET
AVG. 11-18-87 DELTA 88 HFET
881034 HO DELTA 88 FTP
881032 HO DELTA 88 FTP
881126 HO DELTA 88 FTP
AVG. HO DELTA 88 FTP
881032 HO DELTA 88 HFET
881033 HO DELTA 88 HFET
881147 HO DELTA 88 HFET
AVG. HO DELTA 88 HFET
881584 11-18-87 RANGER FTP
881585 11-13-87 RANGER FTP
881598 11-18-87 RANGER FTP
881730 11-18-87 RANGER FTP
881731 02-26-38 RANGER FTP
AVG. 11-18-87 RANGER FTP
881586 11-18-87 RANGER HFET
881632 11-18-87 RANGER HFET
AVG. 11-18-87 RANGER HFET
882636 03-16-88 CRN VIC FTP
882658 03-16-88 CRN VIC FTP
882660 03-16-88 CRN VIC FTP
AVG. 03-16-88 CRN VIC FTP
882637 03-16-88 CRN VIC HFET
882659 03-16-88 CRN VIC HFET
AVG. 03-16-88 CRN VIC HFET
882863 03-16-88 CRN VIC 2 FTP
882865 03-16-88 CRN VIC 2 FTP
882866 03-16-88 CRN VIC 2 FTP
883073 03-16-88 CRN VIC 2 FTP
AVG. 03-16-88 CRN VIC 2 FTP
882864 03-16-88 CRN VIC 2 HFET
883046 03-16-88 CRN VIC 2 HFET
AVG. 03-16-88 CRN VIC 2 HFET
BTU/g HC Den HC den Methane
for MPG (emiss) percent
42.8
42.8
42.8
42.8
42.8

42.8
42.8
42.8
42.8
42.3

42.8
42.8

43.5
43.5
43.5

43.5
43.5

43.5
43.5
43.5
43.5

43.5
43.5

20.9
20.9
20.9
20.9
20.9

20.9
20.9
20.9
20.9
21.6

20.9
20.9

20.8
20.8
20.8

20.8
20.8

20.8
20.8
20.8
20.8

20.8
20.8

19.3
19.3
19.3
19.3
19.3

19.3
19.3
19.3
19.3
20.4

19.3
19.3

19.9
19.9
19.9

19.9
19.9

19.9
19.9
19.9
19.9

19.9
19.9

96.3
96.3
96.3
96.3
96.3

96.3
96.3
96.3
96.3
91.3

96.3
96.3

94.5
94.5
94.5

94.5
94.5

94.5
94.5
94.5
94.5

94.5
94.5

Density
18.89
18.89
18.89
18.89
18.89

18.89
18.89
18.89
18.89
18.89

18.89
18.89

18.89
18.89
18.89

18.89
18.89

18.89
18.89
18.89
18.89

18.89
18.89

Total < — NON METHANE HC — >
HC Method 1 Method 2 Method 3 Method 4
2.41
3.07
1.89
2.46
0.64
1.03
0.84
0.45
0.24
0.52
0.40
0.12
0.05
0.08
0.08
1.38
1.36
1.36
1.36
1.34
1.36
0.78
0.79
0.78
3.68
2.68
3.22
3.19
1.71
1.38
1.55
3.52
3.77
3.53
3.38
3.55
2.08
2.07
2.07
0.38
0.16
0.44
0.33
0.09
0.03
0.05
0.06
0.25
0.23
0.23
0.24
0.00
0.19
0.13
0.13
0.13
0.04
0.00
0.04
0.03
0.05
0.00
0.03
0.12
0.20
0.15
0.11
0.15
0.16
0.14
0.15


0.12
0.10
0.09
0.10
0.00
0.08
0.05
0.05
0.05
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
" 0.00
0.00
0.37
0.15
0.43
0.32
0.09
0.03
0.05
0.06
0.16
0.14
. 0.14
0.15
0.00
0.12
0.07
0.07
0.07
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.06
0.03
0.05
0.08
0.10
0.06
0.08
0.02
0.03
0.03

0.05
0.05
0.05
0.05
0.11
0.06
0.03
0.03
0.03
0.18
0.13
0.16
0.16
0.08
0.07
0.08
0.17
0.19
0.17
0.17
0.18
0.10
0.10
0.10
HCHO
mg/mi
2.46
7.63
4.38
4.82
7.42
7.40
7.41
4.54
3.17
4.38
4.03
7.46
7.40
7.42
7.43
—
3.73
8.46
7.08
4.01
4.66
0.56
2.20
1.38
6.03
3.16
2.63
3.94
1.08
2.65
1.87
5.79
6.62
5.31
5.74
5.87
8.18
3.76
5.97
CO
g/m1
1.3
2.0
1.8
1.7
0.8
1.6
1.2
12.1
2.4
15.1
9.8
2.5
0.9
2.3
1.9
0.1
<0. 1
<0. 1
0.1
<0. 1
<0. 1
<0.1
<0. 1
<0.1
6.7
2.5
3.8
4.3
6.9
4.2
5.6
0.6
<0.1
1.1
0.1
0.4
<0.1


-------
Test GAS ANL Vehicle Test
Number DATE Type
882678 HO CRN VIC FTP
882676 HO CRN VIC FTP
883072 HO CRN VIC FTP
AVG. HO- CRN VIC FTP
882734 HO CRN VIC HFET
882677 HO CRN VIC HFET
AVG. HO CRN VIC HFET
883243 04-29-88 CELEBRTY1 FTP
883245 04-29-88 CELEBRTY1 FTP
883510 04-29-88 CELE8RTY1 FTP
883527 04-29-88 CELEBRTY1 FTP
AVG. 04-23-63 CELE8RTY1 FTP
883244 04-29-88 CELEERTY! HFET
883246 04-29-88 CELEb'RTYl HFET
883511 04-29-33 CELEERIYl HFET
•333528 04-29-88 CELEBRTY1 HFET
AVG. 04-29-88 CELE8RTY1 HFET
883247 HO CELE8RTY1 FTP
883249 HO CELE8RTY1 FTP
883541 HO CELE8RTY1 FTP
AVG. HO CELEBRTY1 FTP
883248 HO CELEBRTY1 HFET
883250 HO CELEBRTY1 HFET
883542 HO CELE8RTY1 HFET
AVG. HO CELE8RTY1 HFET
883848 HO CELEBRTY2 FTP
883852 HO CELEBRTY2 FTP
AVG HO CELEBRTY2 FTP
883849 HO CELEBRTY2 HFET
883853 HO CELEBRTY2 HFET
AVG. HO CELEBRTY2 HFET
883920 05-31-88 CELEBRTY2 FTP
883966 05-31-88 CELEBRTY2 FTP
883967 05-31-88 CELEBRTY2 FTP
AVG. CELEBRTY2 FTP
883921 05-31-88 CELEBRTY2 HFET
884051 05-31-88 CELEBRTY2 HFET
AVG. CELEBRTY2 HFET
BTU/g HC Den HC den Methane
for MPG (emiss) percent


43.2
43.2
43.2
43.2

43.2
43.2
43.2
43.2















43.1
43.1
43.1

43.1
43.1



21.
21.
21.
21.

21.
21.
21.
21.4















21.4
21.4
21.4

21.4
21.4



20.3
20.3
20.3
20.3

20.3
20.3
20.3
20.3















20.3
20.3
20.3

20.3
20.3



93.5
93.5
93.5
93.5

93.5
93.5
93.5
93.5















93.5
93.5
93.5

93.5
93.5

Density


18.89
18.89
18.89
18.89

18.89
18.89
18.89
18.89















18.89
18.89
18.89

18.89
18.89

Total < — NON METHANE HC — >
HC Method 1 Method 2 Method 3 Method 4
0.35
0.33
0.38
0.35
0.06
0.07
0.06
1.38
1.28
1.40
1.91
1.49
0.83
0.64
0.70
0.72
0.72
0.43
0.26
0.35
0.35
0.01
0.01
0.02
0.02
0.29
0.23
0.26
0.02
0.01
0.02
1.48
1.56
1.62
1.55
0.79
0.76
0.78
0.26
0.24
0.26
0.25
0.03
0.04
0.03
0.00
0.00
0.00
0.06
0.01
0.04
0.00
0.00
0.00
0.01
0.41
0.24
0.33
0.32
0.01
0.01
0.01
0.01
0.26
0.23
0.25
0.02
0.01
0.02
0.00
0.29
0.00
0.10
0.00
0.00
0.00


0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00














0.00
0.14
0.00
0.05
0.00
0.00
0.00
0.25
0.22
0.24
0.23
0.03
0.03
0.03
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.00
0.40
0.23
0.32
0.32
0.01
0.01
0.01
0.01
0.26
0.23
0.24
0.01
0.01
0.01
0.00
0.28
0.00
0.09
0.00
0.00
0.00


0.08
0.07
0.08
0.11
0.09
0.05
0.04
0.04
0.04
0.04














0.09
0.09
0.09
0.09
0.05
0.04
0.05
HCHO
mg/m1
2.78
2.80
5.36
3.65
1.69
1.62
1.66
3.30
1.74
3.86
2.85
2.94
0.00
0.61
2.60
2.08
1.32
4.85
3.96
5.44
4.41
0.38
1.04
0.74
0.72
5.54
6.06
5.80
1.20
0.82
1.01
3.66
4.62
4.65
4.31
0.90
2.34
1.62
CO
g/mi
1.2
1.2
1.7
1.4
0.1
0.1
0.1
8.0
2.0
0.1
0.1
2.6
9.3
1.1
<0. 1
<0.1
2.6
0.6
1.4
0.7
1.0
<0.1
<0.1
<0. 1
<0.1
1.4
1.1
1.25
0.1
0.1
0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
C02
g/mi
587
580
579
582
456
455
456
331
326
327
430
354
251
256
251
250
252
439
430
436
435
335
334
336
335
433
435
434
342
342
342
325
323
323
324
250
250
250
NOx
g/m1
1.1
1.1
1.1
1.1
0.7
0.6
0.7
0.7
1.1
1.5
3.2
1.6
0.6
2.4
2.7
2.6
2.1
0.6
0.6
0.7
0.6
0.7
0.5
0.7
0.6
0.6
0.6
0.6
0.3
0.3
0.3
1.2
1.2
1.2
1.2
2.1
2.2
2.2
GEFE
mpg
15.0
15.2
15.2
15.1
19.4
19.5
19.4
20.6
21.6
21.7
16.5
19.8
26.8
27.6
28.3
28.5
27.8
20.1
20.5
20.2
20.3
26.5
26.5
26.4
26.5
20.3
20.3
20.3
25.9
25.9
25.9
21.8
22.0
21.9
21.9
28.5
28.5
28.5
>

-------
                       COMPARISON OF PROPERTIES OF CNG, LNG, LPG, AND GASOLINE
Typical composition
State of fuel as
stored on vehicle
Pressure as stored
on vehicle, kPa
(psi)
Weight as stored
on vehicle, kg/liter
(Ib/gal)
Heat content as
stored:
mJ/liter (Btu/gal)
Specific gravity
of vapor at STP
(air = 1)
Self-ignition
temperature,
Stoichiometr ic A/F,
. by weight
Octane number (RON)
CNG
Methane, 90%
Ethane, 10%
Gas

13.8 (2,000)


0.1 (1.4)


8.4 (30,000)


0.55


705 (1,300)

17

100+
LNG
Methane, 90%
Ethane, 10%
Liquid*

0.21-0.41 (30-60)


0.43 (3.6)
'

21.5 (77,000)


0.55


705 (1,300)

17

110+
LPG
Propane, 95%
Propylene, 5%
Liquid

1.4 (200)


0.50 (4.2)


23.3 (83,500)


1.55


510 (950)

15.7

: 110+
Gasoline
C4 to Ci2
hydrocarbons
Liquid

Atmospheric


0.73-0.78 (6-6.5)


32.2 (115,800)


4.25


460 (860)

15

90-100
                                                                                                      U)
Temperature, -161°C    (-258°F)

-------
                                    A-.
                                   I S
                 MICHIGAN CONSOLIDATED GAS

DATE ANALYZED: 11-18-87
                                              RUN NO. 87-676
                         SAMPLE INFORMATION
LOCATION:
REQUESTER:
DEPARTMENT:
FIELD:
CITY,STATE:
PERMIT *
FORMATION:
SYSTEM:
OWNER:
PURCHASER:
RELATED TESTS:
               ALLEN RD.STA.
               P.PAI
               LAB.

               DETROIT MI
            CYLINDER   I.D.
            SAMPLE #
            SAMPLE POINT:
            SAMPLE DATEQTIME:
            SAMPLE RECEIVED:
            ATMOSPHERIC  TEMP.  :
            GAS TEMP.  (F):
            GAS PRESSURE (PSIG):
            WELLHEAD PRESSURE  (PSIG):
            FLOW  (MMCF/DAY):
            SAMPLED BY:
                                              S.LAB.

                                              AUTO FILL STA,
                                              9-15-87
                                              9-15-87
                                                              1500
                                                              R.LAYNG
       GAS ANALYSIS
                                        GROSS HEATING VALUE (BTU/SCF)
NITROGEN
CARBON DIOXIDE
HELIUM
METHANE
ETHANE
PROPANE
I-BUTANE
N-BUTANE
I-PENTANE
IM-PENTANE
HEXANES
HEPTANES
OCTANES
HYDROGEIM
C*n2_
MOL 7.

 0.41
 1.O2
 0.00
               WT. 7.

                0.68
                2.68
                0.00
 96.37 4- 93.08
  1.29 7  2.32
  0.21 Z^.^_O.55
~oTo5~  "  o. 17
        O.O2 -, 4 O.O8
        0.01 "_ 0.04
        0.02 r~J 0. 10
        0.02 i.-:, o. 11
        0.01 - -s"0.06
        0.03    0.00
                            H/C
                                               14.734 SAT / 14.650 DRY
                                        CALCULATED
                                        DETERMINED FIELD
                                        DETERMINED LAB

                                        	SPECIFIC GRAVITY..
                                        CALCULATED SP.  GR.
                                        DETERMINED FIELD
                                        DETERMINED LAB
                                        V	-.->-.
                                        x 0.577)
                                        	SULFUR (AS H2S)
                                        HYDROGEN SULFIDE
                                        MERCAPTANS
                                        SULFIDES
                                        RESIDUAL
                                        TOTAL SULFUR
                                      GR/CCF
                                        	OTHER.
                                        HYDROCARBON
                                         LIQUID (GAL/MCF)
                                        HYDROCARBON DEW
                                         POINT  (F @ PSIG)
                                        WATER DEW POINT
                                         (F 0 PSIG)
                                        Lbs.  WATER / MMCF
                                           0. 12
          TOTAL
                       1OO.OO   100.00
ANALYZED BY:
APPROVED BY:
DISTRIBUTION:
              N.MCEACHERN
              G.EVANINA^C
              P.PAI
REMARKS:
              SAMPLE TAKEN THROUGH CARBON  FILTER

-------
                                  A-5
                  MICHIGAN CONSOLIDATED GAS COMPANY

DATE ANALYZED:  2-23-B3                        RUN NO. 85-69

                         SAMPLE  INFORMATION

«•   '                           .             .
LOCATION:       C.N.G.OUTLET         CYLINDER  I.D.
REQUESTER:      K.CZERWINSKI         SAMPLE tt
DEPARTMENT:     TECH.DEVELOP.        SAMPLE POINT:              NORTH EAST FI.
FIELD:                               SAMPLE DATEGTIME:        .  2-22-SSS9200P
CITY,STATE:     MELVINDALE MI        SAMPLE RECEIVED:           2-22-S2
PERMIT  S                             ATMOSPHERIC TEMP.  (F):
FORMATION:                           GAS TEMP.  (F):
SYSTEM:                              GAS PRESSURE  CPSIG):       450
OWNER:                               WELLHEAD PRESSURE  (PS 12):
PURCHASER:                           FLOW CMMCF/DAY)':
RELATED TESTS:                '       SAMPLED BY:                 N.MCEACHERN


       GAS ANALYSIS                     GROSS HEATING VALUE (ETU/SCF)


                        MOL Y.    WT.  Y.           14.734 SAT  / 14.c50 DRY

NITROGEN                 0.94     1.54    CALCULATED                /I032
CARSON.DIOXIDE           0.31     1.31    DSTERMINED FI EL D
HELIUM                   0.02     0.00    DETERMINED LAB
       I                 94.20    SS.74
                         -r_ a/;     £_ -j-r    	SPECIF! 3 !3 r:.T-. '.•' I ."Y	
. .x~. .-..,_:                  O.::i     0.5'!    CALCULATED HP. GF:.        0.525
I-BUTANE                 0.09     0.27    DETERMINED FIELD
N-BUTANE                 0.07     0.23    DETERMINED LAB
I-PENTANE                0.03     0.12
N-PENTANE                0.02     O.OS    	SULFUR (AS H2S) GR/CC"	
HEXANE5                  0.02     0.10    HYDROGEN SL'LFIDE
HEPTANES                 0.03     0.17    MERCAPTANS
nnyfiMp:;                  A r>i     r\  1 -r    ct ;i rTncc
<-!*-• I f-«JXC.3                  IJ m -^ fc.     I./ . ^ ^f    ^J^Ji_i J. L-^.^
       IN                 0.01     0.00    RESIDUAL
                                         TOTAL SULFUR

                                         	OTHER	
                                         HYDROCARBON
                                          LIQUID  (GAL/MCF)
                                         HYDROCARBON DEW
                                          POINT (F Q PS IS;
                                         WATER DEW POINT
                                          
-------
                                 ' A-6
                       «=*IM<=*L_ YS I S
                           CONSOLIDATED GAS
DATE-- ANALYZED! ?3-ii-8S'
                        . SAMPLE. INFORMATION
LOCATION! • ' -. 'ft
REQUESTER:
DEPARTMENT:   '
FIELD:       :
CITY,STATE:
PERMIT #
FORMATION:
SYSTEM:
OWNER:
PURCHASER:
RELATED TESTS:
    ALLEN RD.
K.CZERWINSKI
TECH.DEVELOP.

DETROIT MI

    I
                                                         **;'-'$\'.'•'V-JT--.'..-• '•'•--;•• '-.''•' '•'':''-.'£,•_(. V <•'
                   •iy •- SAMPLE POINT: ^•.'•x-;^ .-"V•'"f <  LOW PRES.RUN
                    :'-SAMPLE DATEQf IME: "' •: ;     3-14-88Q145PM
                   "•SAMPLE RECEIVED:'--   ;  :   3-14-B8
                      ATMOSPHERIC TEMP.  (F):
                      GAS TEMP. (F):
                      GAS PRESSURE (PSIG):       300
                      WELLHEAD PRESSURE  (PSIG):
                      FLOW (MMCF/DAY):       .     ••"•'• ~=v '
                      SAMPLED BY:                 N.MCEACHERN
       GAS ANALYSIS
                         GROSS HEATING VALUE.(BTU/SCF)
                        MOL 7.
                 WT.  •/.
                                 14.734 SAT  /  14.650 DRY
NITROGEN
CARBON DIOXIDE
HELIUM
METHANE
ETHANE
PROPANE
I-BUTANE
N-BUTANE
I-PENTANE
N-PENTANE
HEXANES
HEPTANES
OCTANES
HYDROGEN
0.66
0.39
0. 10
96.24
2.05
0.36
0.07
0.06
0.02
0.01
0.02
O.01
0.01
0.00
1. 10
1.02
0.02
92.47
3.63
0.94
0.24
0.20
0.08
0.04
0. 10
O.O5
0.06
0.00
                         CALCULATED
                         DETERMINED FIELD
                         DETERMINED LAB
                         	SPECIFIC GRAVITY..
                         CALCULATED SP. GR.       /
                         DETERMINED FIELD
                         DETERMINED LAB
                         	SULFUR (AS H2S)  GR/CCF.
                         HYDROGEN SULFIDE
                         MERCAPTANS
                         SULFIDES
                         RESIDUAL
                         TOTAL SULFUR
                                        	OTHER.
                                        HYDROCARBON
                                         LIQUID  (GAL/MCF)
                                        HYDROCARBON  DEW
                                         POINT  (F @  PSIG)
                                        WATER DEW POINT
                                         (F © PSIG)
                                        Lbs. WATER / MMCF
          TOTAL
       100.OO   100.OO
ANALYZED.BY:  N.MCEACHERN
APPROVED BY:. . G.EVANINA/^C .
DISTRIBUTION: K.CZERWTNSKI+P.PAI
REMARKS:
FILLING HIGH 'PRESSURE TANKS
                                          t

-------
                                  A-7
                  MICHIGAN CONSOLIDATED GAS COMPANY


DATE ANALYZED:  15-4-88                        RUN NO. 88-207
- — —. — — •——-».— «-».«~.^—— — —- — •—————*- •-^-•.•••^•^•^^.—^ .— •—..^—.— ••.••.••:
    FLOW  (MMCF/DAY)»
    SAMPLED BY:
                      S.LAB.
                      1
                      WIS-ERG TRUCK
                      4-29-S8e900AM
                      4-29-eB
                                               1100
                                               H.WENZEL
        GAS ANALYSIS
                         GROSS HEATING VALUE (BTU/SCF)
                        HOL 7.   WT. X
 NITROGEN
 CARBON DIOXIDE
 HELIUM
 METHANE
 ETHANE
 PROPANE
 I-BUTANE
 N-BUTANE
 I-PENTANS
 N-PENTANE
 HEXANES
 HEPTANES
 OCTANES
 HYDROGEN
         0.73
         0/50
         0.01
        95. 5S
         2.6O
         0.41
         0.06
         0.03
         0.02
         0.01
         0.01
         O.01
         0.01
         o.oo
 1.21
 1.30
 0.00
91. 13
 4.64
 1.07
 0.20
 0.17
 0.00
 0.04
 0.05
 0.05
 0.06
 0.00
       14.734 SAT / 14.650 DRY

                         /1025
CALCULATED
DETERMINED FIELD
DETERMINED LAB
	SPECIFIC GRAVITY	
CALCULATED SP. 8R.       0.381
DETERMINED FIELD
DETERMINED LAB
	SULFUR (AS H2S) GR/CCF.
HYDROGEN SULFIDE
MERCAPTANS
SULFIDES
RESIDUAL
TOTAL SULFUR
                                         	OTHER.
                                         HYDROCARBON
                                         LIQUID  (3AL/MCF)
                                         HYDROCARBON  DEW
                                         POINT  (F  a  P3I8)
                                         WATER DEW  POINT
                                          (F d PSIG)
                                         Lbs.  WATER / MMCF
           TOTAL .
        100.00  100.00
 ANALYZED 3V:  N. MCEACRERJM
 APPROVED BY:  G. EVANINAxOf^
 DISTRIBUTION: E. LAMfV^RT-t-K. CZERWINSKI

 REMARKS:             '

-------
                                  A-8
                 MICHIGAN  CONSOLIDATED GAS COMPANY

DATE ANALYZED: 6-1-88                         RUN NO. 88-241

                         SAMPLE  INFORMATION
LOCATION: CNG TANK FARM
REQUESTER : B . LAMPORT
DEPARTMENT : SYS . FUELS
FIELD:
CITY, STATE: MELVINDALE MI
PERMIT tt
FORMATION:
SYSTEM:
OWNER:
PURCHASER:
RELATED TESTS:
CYLINDER I.D.
SAMPLE #
SAMPLE POINT:
SAMPLE DATEOTIME:
SAMPLE RECEIVED:
ATMOSPHERIC TEMP. (F) :
GAS TEMP. (F) :
GAS PRESSURE (PSIG):
WELLHEAD PRESSURE (PSIG):
FLOW (MMCF/DAY) :
SAMPLED BY:
S.LAB.
2
WIS-EPA TRUCK
5-31-88@200FM
6-1-88


1000


H.WENZEL
       GAS ANALYSIS
                          GROSS  HEATING VALUE (BTU/SCF)
NITROGEN
CARBON DIOXIDE
HELIUM
METHANE
ETHANE
PROPANE
I-BUTANE
N-BUTANE
I-PENTANE
N-FENTANE
HEXANES
HEPTANES
OCTANES
HYDROGEN
         MOL 7.

          0.59
          O.58
          0.01
         95.67
          n crc;
          *- • ^JwJ
          0.43
          O.06
          0.05
          0.02
          0.01
          0.01
          0.01
          O.O1
          0.00
WT. 7.

 0.93
 1.51
 0.00
91. 19
 4.55
 1. 12
 0.20
 0.17
 O.08
 0.04
 0.05
 0.05
 0.06
 0.00
       14.734 SAT  /  14.650  DRY

                          '1025
CALCULATED
DETERMINED FIELD
DETERMINED LAB
	SPECIFIC  GRAVITY	
CALCULATED SP. GR.        O.581
DETERMINED FIELD
DETERMINED LAB
	SULFUR  (AS H2S)  GR/CCF.
HYDROGEN SULFIDE
MERCAPTANS
SULFIDES
RESIDUAL
TOTAL SULFUR

• ••••••*••**• LJ I tl;—i~V »•••••••••
HYDROCARBON
 LIQUID (GAL/MCF)
HYDROCARBON DEW
 POINT  (F @ PSIG;
WATER DEW POINT
 (F 0 FSIG)
Lbs. WATER / MMCF
          TOTAL
        100.00   100.00
ANALYZED BY:
APPROVED BY:
DISTRIBUTION:
N.MCEACHERN
G.EVANINA   ££.
B.LAMPORT+K.C ZERWINSKI
REMARKS

-------
                                       A-9
    December  3,  1987

    Consumers Power
    212  West  Michigan Avenue
    Jackson,  Michigan
    49201   U.S.A.

    Attention:   Mr. Richard Polich
                                                         FUEL SYSTEMS
    Dear  Mr.  Polich;

    I   am writing with  regard to  the 1984 Oldsmobile Delta  88 Royale  that  CNG
    Fuel  Systems is providing for emissions testing at Ann Arbor.   As you  know,
    that   vehicle is equipped with our experimental  GEN II NGV fueling system.
    We hope  it will expand the knowledge of NGV technology.

    As you  know, CNG Fuel Systems' Board of Directors has elected  to close  the
    company.   Accordingly, we have  no vested interest in  the results of this
    test  work, other than altruism.

    We have performed  several FTP tests on the vehicle before releasing it  for
    these tests.  While the  tests were not exhaustive,  they are  informative,
    and are  summarized  below.
                                             Wtd. Emissions (g/mi)
Test

STD.


FTP
    HOT
    505
  Fuel



Gasoline

NGV

Gasoline

NGV
NGV
                        Ignition Timing
                         Gasoline
                         NGV +12
Est.
nmHc
.39
.10
.18
.09
.12
.17
HC
.41
.13
.74
.12
.48
.72
CO
3.4
4.8
1.2
8.1
2.2
3.9
ENERGY US!
NOU MJ/100 Km
1.0
.71
.58
.64
.69
1.07
-
504
529
463
469
463
         NOTE:   NGV  installations typically use the +12° timing

         NOTE:   Non-methane hydrocarbons (nmHC) are estimated values



    These results  verify  several points that we have made in our recent papers.

         1.    Current   aftermarket practice  of advancing  the NGV  mode timing
              significantly   increases NO   and may  cause  vehicles to  exceed
              the   NO   standard on NGV.   More development work  is required  to
              learn  now  to advance the timing  without sacrificing NO .  Like-
              ly  solutions include even better closed  loop control in the NGV
2150 Steeles Ave. East, Brampton, Ontario, Canada L6T 1A7 (416) 793-3560 Telex 06-988548

-------
                                  A-10
                                   -  2  -
          mode  to raise the  catalyst's  NO    conversion efficiency,  and a
          more sophisticated NGV  mode  ignition control system that provides
          less  aggressive spark  advance  in the sensitive driving modes for
          urban emissions.

          In  order to allow the  EPA   to  independently evaluate spark angle
          effects  on NGV mode emissions, we  have  equipped the car with an
          underhood  switch to select   either gasoline or  NGV mode timing.
          We hope the EPA will  evaluate both  modes.

     2.   The  HC emissions on NGV  may exceed the  standard for total hydro-
          carbons  yet be far  below a non-methane standard.   Ultimately,
          either  NGV vehicles would   need to be  evaluated against  a nmHC
          standard, or further development work must be performed to reduce
          methane exhaust emissions.

          Work  that we have done  indicates  that  very  precise closed loop
          control  about stoichiometry may be able to provide total hydro-
          carbon levels equal to  gasoline's  (i.e.  well below the standard).
          We  feel that achieving  that exceptionally tight  air fuel ratio
          control  is feasible for  an   aftermarket system.   At our current
          level of development, we  may need  refinements in the NGV metering
          device and closed loop  control  (i.e. the dynamic loop;  the vola-
          tile and keep alive adaptive memories).

     3.   We believe that the GEN II experimental  system currently can pro-
          vide NGV mode emissions compatible  with  the U.S. emissions stand-
          ards.

     4.   More NGV techology development  is  required.

The  GEN II system installed on the car is a  generic system aimed at after-
market  installation.  This  concept has  a  universal calibration  which is
further refined by the installer.  Pointedly, this system would not have an
engine  specific "factory calibration" developed   for new models.   We feel
that this universal concept is crucial if the NGV  industry is to flourish.

If  there are questions that arise  on the car or  our results, please feel
free to contact me.
Sincerely,
Stephen A. Carter
Vice President
Conversions

cc:  Mr. Jeff Allison (EPA)
     Mr. Jeffery Seisler (AGA)
     Mr. Peter Flynn (CNG Fuel  Systems
     Mr. Ulrich Oester (CNG  Fuel  Systems)
.._...                                                        FUEL SYSTEMS
SAC/dj

-------
                                                       A-ll




                                 Table 1 -  Ford  Ranger Exhaust Emissions

                                             (GraMS  Per Mile)

                                                          HC      CO      NO,

                    Federal Standards  < 8500 GVH        0.8    10.0     2.3

                    Gasoline Ranger  (1984)  I/           0.31*   3.2     1.1

                    Natural Gaa Ranger (1984) 2/        1.20**  0.03    1.9
                    *  Total HC emissions
                    ** Theae BC'a are  88k  CH«

                    I/ Bauaaion  levels  of  a  typical  1984, 49   atate low  Mileage
                    Ranger.

                    2/ Single vehicle  test at  low Mileage.
    The  replacement   of  a  dirty  air filter
    corrected one  and the  other was cured by
    the  replacement   of   the   fuel  priming
    circuit board.
    Power  steering pump  rattle -  One of the
    participants   reported   severe   engine
    knock  at  3500  rpn.    Investigation dis-
    closed  the  "knock"  was due to excessive
    end  play   in  the power  steering  pump
    shaft.   Replacement of the pump eliminated
    the noise.
      Proponents   of   natural   gas  powered
 vehicles  have  claimed  reduced  maintenance
 costs,   especially  in the  areas  of engine
 oil and  spark  plug  life.    While  there is
 no  evidence  that  a  natural  gas  powered
 vehicle  requires    more   maintenance,   there
 is also no evidence   that  it  requires less.
 Although natural  gas leaves  no lead,  carbon
 or varnish deposits  in the   oil,    the   oil
 will  still  gradually thicken and the addi-
 tives and  inhibitors  will  break  down even
 though  it  remains   clear.  Until  conclusive
 tests are  conducted,  it   is recommended that
 the oil be  changed   at  the  regular service
 intervals.
      The  absence  of  lead  salts and carbon
 build-up from  the combustion  of natural gas
•should  mean   that   spark  plugs  will  last
 longer  than  in  gasoline engines.   However,
 with  a  current  life of  48,000 km (30,000
 mi),  it  is too early-to  tell,  based on the
 lease fleet  experience  if  spark  plug life
 will be  noticeably   increased.    Either way,
 spark plugs  are not  seen as  a major factor
 in maintenance costs.

 INCREMENTAL COSTS

      The primary cost difference between the
 natural gas  powered Ranger  and the gasoline
 powered  Ranger  is   due   to  the  expense of
 the natural  gas  fuel cylinders.    The cost
 savings  from  the  elimination  of  the fuel
 tank  and  the  carburetor  from the gasoline
 powered  engine are  offset  by  the cost of
regulator, fuel mixer and natural gas cylinder
mounting brackets.
     In  mass  production,  the  cost  of  the
natural gas engine, including  further refine-
ments  for  natural  gas  operation,  would be
about the  same  as  a  gasoline  engine.  The
fuel tanks  are  significantly  more expensive
than gasoline tanks and  would  remain  one of
the  major  cost  issues  with  a  natural gas
powered vehicle.  The  present  estimated cost
for  an  after—market  conversion  of  an  in-
ternal   combustion   vehicle  to  operate  on
natural  gas  is  $1500.   It  is  anticipated
that  the  cost  premium  for  a  high  volume
factory-engineered  version  could be approxi-
mately  one-half  as  much.   Based on current
fuel prices,  the price  differential could be
amortized in about three years.

CONCLUSIONS

     Vehicle  operation  utilizing natural gas
as  a  fuel  offers the potential for substan-
tial  cost  savings.  Because public refueling
stations  for  natural  gas  virtually  do not
exist at present in the United States, natural
gas  operation  is  probably  confined  to the
fleet  operator  although  the  home refueling
option  utilizing  residential  gas  is under-
study.  Based  on  Ford's  experience  to date
with compressed  natural  gas  vehicles  it is
concluded that:
    There  are  no  unresolvable technological
    issues  that  would prevent motor vehicles
    from  operating  efficiently  and economi-
    cally on natural gas.
    Single  fueled   vehicles,  optimized  for
    compressed natural  gas operation, provide
    better  fuel  efficiency  and  performance
    than  dual  fuel vehicles, with acceptable
    range   for   most  fleet  operation.  The
    additional   tank   volume   required  for
    operating   range   reduces   the  vehicle
    carrying capacity and volume.
    In mass production, the cost of  a natural
    gas engine  would be  about the  same as a
                                                   777

-------