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 systemsin 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 aftermarket 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
------- |