IMS-001/FE-1
Effects of Inspection and Maintenance Programs on Fuel Economy
March 1979
(Revised June 1979)
Inspection and Maintenance Staff
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air, Noise, and Radiation
U.S. Environmental Protection Agency
-------
Introduction
The primary goal of inspection and maintenance (I/M) programs is to
improve air quality by reducing emissions from motor vehicles. Many
studies have indicated that I/M programs will achieve this goal. In
addition several studies have indicated that fuel economy improvement
can be expected to occur as a result of maintenance performed on vehicles
failing an I/M test. Most studies which have been performed have looked
at pre-1975 model year vehicles and have had expert mechanics performing
the maintenance work. It is the intention of the present report to
consider both the results from past studies and the results from more
recent studies in order to provide EPA's best estimates of fuel economy
benefit which can be attributed to I/M maintenance.
The results of this study are discussed in the body of the report by
model year groups, which reflect the level of emission control technology:
pre-1975, 1975-80, and 1981+. Attachment 1 provides a short tabular
summary of studies which relate to I/M's effect on fuel economy.
Attachment 2 provides a bibliography which gives references for most of
the studies summarized in Attachment 1 and alluded to in the body of the
report. The following section summarizes EPA's best estimates of I/M's
effect on fuel economy.
Summary
On the basis of past studies on pre-1975 model year cars, EPA has in-
dicated that the fuel economy of cars failing an I/M test will likely
improve following maintenance. Recently, with the availability of data
from Oregon's I/M program and further data and information on catalyst
and microprocessor controlled systems, it has become clear that I/M's
effect on fuel economy is not as straightforward as was previously
thought. All available data on the question of I/M's effect on fuel
economy have therefore been studied in order to better quantify what
this effect will be.
Many past I/M studies have indicated a fuel economy benefit due to such
maintenance, and more recent EPA studies have also indicated that a fuel
economy benefit is obtained following restorative maintenance performed
to manufacturers' specifications by skilled mechanics. However, data
from the Portland Study, covering 1972-77 model year cars, show no net
gain in fleet average fuel economy due to maintenance performed in
response to failing an Idle test in Oregon's I/M program. The apparent
contradiction between the field data from Portland and these other
studies is explainable. The past I/M studies have In most cases included
maintenance which was either performed by skilled mechanics or which was
(by the study's definition) more extensive than the typical I/M tune—up
performed in response to an I/M program. In addition, very few
-------
mechanics in the Portland area have been specifically trained in emis-
sion-oriented maintenance. This is in contrast to the emission-oriented
maintenance to manufacturers' specifications which has been performed in
EPA's studies of the effects of restorative maintenance. EPA concludes
that the failure to observe a net gain in fuel economy in Portland only
implies that any basic, idle test I/M program which is unsupported by an
emission-oriented mechanic training program should also be expected to
show no net gain in fuel economy. EPA also concludes on the basis of
the other studies that a net fuel economy gain due to I/M maintenance is
possible for pre-1980 model year cars, if the I/M program is supported
by mechanic training.
For 1980 model year cars technology will change somewhat from that
currently used outside of California. Data on 1975-77 model year
California cars have been used to estimate the effect of I/M on 1980
model year Federal cars. (The technologies are expected to be very
similar, as are the emission standards.) Even with skilled mechanics
performing repairs to manufacturer's specification, no significant fuel
economy improvement is seen. This is a somewhat unexpected result which
requires further investigation when data on 1980 model year Federal cars
are available.
For 1981 arid later models, fuel economy improvements due to emission-
related maintenance are expected. These improvements are expected to be
more substantial than the improvements possible for pre-1980 model year
cars, primarily because engine control system failures will often cause
the air/fuel mixture to enrichen over all modes of vehicle operation,
not just during idle operation, and thus degrade both fuel economy and
emissions.
The table below provides EPA's best estimates of fuel economy improvement
of failed cars following maintenance.
Average Fuel Economy Improvements for Failed Vehicles due to I/M
Model Year
Without Emission-Oriented With Emission-Oriented
Mechanic Training
Mechanic Training
pre-1975
1975-1979
1980
1981+
0%
0%
0%
9
9-30%
3-4%
3-4%
Discussion: Pre-1975 Model Year Cars
By the early 1980s when most I/M programs will be operational, pre-1975
models will account for less than 20% of the annual vehicle miles
traveled (VMT) for passenger cars and by 1987 this will be reduced to
-------
-3-
3%. Therefore, from the standpoint of l/M» pre-1.975 models are rela-
tively less important than 1975 and later models in terms of both
emission reductions achievable and fuel economy effects.
All past studies which have evaluated fuel economy benefit due to I/M
for pre-75 models have reported improvements in fuel economy resulting
from maintenance. Most of these studies have measured fuel economy
during city, as opposed to highway, driving cycles. The city cycle
includes more idle operation than the city/highway combination used by
EPA to measure typical passenger car fuel economy. In most of these
studies, mechanics performing the maintenance had a higher than average
level of expertise. As Attachment 1 shows, reported fuel economy im-
provements from past studies (studies 6-13) have ranged from 21 to 14%.
For past studies which best typify anticipated I/M situations (studies
6, 8, and 10), the range in fuel economy improvements is 2% to 4% for
pre-1975 models. In two of these three studies city driving cycles were
used to measure fuel economy. In all three studies the maintenance
performed*on failed cars was either more extensive than would usually
be necessary in a real I/M program or was performed by trained mechanics,
EPA's Portland Study, which is now in progress, provides befpre and
after maintenance fuel economy data from a real I/M program.— Analysis
of the data on the 1972-74 model year failed cars tested in this program
indicates that although maintenance performed on failed vehicles sig-
nificantly improves fuel economy during city driving,— the improvement
is relatively small. There is no statistically significant effect
during a typical city/highway combination of driving. City/highway
combined fuel economy changes from 16,8 miles per gallon (MPG) initially
to 16.9 MPG on failed cars after maintenance.
Although there is no net fuel economy gain or loss for all 1972-74 model
year cars in the Portland Study taken together, the worst idle HC and CO
emitters in the sample, as identified by Oregon's state inspection test,
obtained a substantial fuel economy benefit after maintenance. These 11
cars* fuel economy went from 16.9 MPG before maintenance to 17.9 MPG
after maintenance, an improvement of 6%.
The results for 1972-74 models are summarized in Table 1. To summarize
verbally, for a typical combination of city/highway driving, no net gain
or loss in fuel economy for pre-1975 models is observed in Portland.
However, on especially high idle emitters the possibility of improved
fuel economy after maintenance for failed cars is increased. Also,
based on past studies which were designed to evaluate I/M's effect on
fuel economy (notably studies 6, 8, and 10 in Attachment 1), it appears
that a fuel economy benefit of approximately 31 may occur if an emission-
oriented mechanic training program is in place. Very little emission-
oriented training of practicing mechanics has occurred in the Portland
area to date.
— Vehicles are passed or failed according to Oregon's state inspection
test (basically an idle test) and failed vehicles are maintained by
practicing mechanics in the Portland area,
~{n the statistical sense at the 0.05 level of significance.
-------
Table 1
Fuel Econoay Before and After Kaintfinance^
1972-74 HodeL Year Failed Vehicles in the Portland Study
N
Before
Maintenance
urban
After
Maintenance
Percent
Chanjte
Before
Maintenance
—Highway
After
Maintenance
Percent
Chan«e
Before
Malntenanee
-Combined^—
After
Maintenance
Percent
Change
All 1972-74 Models '
93
14.46
14.69
1.4^
20.89
20.75
-0.7
16.80
' 16.91
0.6
1972 Models
1973 Models
1974 Models 1
30
26
29
13.72
13.55
16.73
13.74
14,01
19.96
°-l2/
3.4 —
1.9
19.62
20.06
23.77
19.37
20.04
21.73
-1.3
-0.1
-0,2
15.87
15.87
19.30
15.81
16.21
19.46
°'4 2/
2.1-
0.8
Highest Idle HC Emitters
Highest Idle CO Emitters
Highest Idle HC and CO
20
20
13.63
15.50
13.89
15.56
1.9a/
0.4"
19.71
22.90
19.74
23.18
0.1
1.2
15.82
18.14
16,04
19.00
i:M/
Emitters
U
14.52
15.69
B.l
21.26
21.78
2.4
16.93
17.95
6.0
¦t-
i
—' Fuel economy la measured In miles per gallon.
2/
- This change was statistically significant at the 0,05 level through the
application of a paired t-teEt. Groups with leas than 20 vehicles were
not evaluated for statistical significance due to small sample size.
-------
-5-
1975-30 Model Year Cars
In the early 1980s when most I/M programs will be operational, 1975-80
model year cars will account for 65% of the VMT for passenger cars, and
by 1987 will still account for 27%. Therefore the effect of I/M on
these vehicles from the standpoint of fuel economy is an important
consideration. I/M's effect on fuel economy for 1975-80 model year
failed vehicles is estimated from studies of 1975-77 model year cars'
data. All studies are relatively recent.
Of the available studies, the study which best typifies anticipated I/M
situations is EFA's Portland Study. In the Portland Study the change in
fuel economy for failed vehicles after maintenance is small and is
statistically insignificant. Before and after maintenance combined
city/highway fuel economies were 17.9 miles per gallon and 17.8 miles
per gallon, respectively. Although 1975-77 model year failed vehicles
saw virtually no fuel economy change as a group, the 1977 model year
failed vehicles on the average received a statistically significant fuel
economy penalty of 3-4% following maintenance. The design of 1977
model year vehicles is very similar to the design of 1975-76 models,
except that the control of oxides of nitrogen is more stringent. Thus,
there is no obvious reason to expect I/M's effect on fuel economy to be
different for the 1977 models. A detailed review of the emissions data,
diagnostics, and maintenance reported for these vehicles is now underway
and indicates that improper adjustment of ignition, timing and improper
operation of vacuum advance units in their before maintenance condition
may explain some fuel economy penalties following maintenance.
As was the case for the 1972-74 models, it was anticipated that the
worst idle emitters among the 1975-1977 models might see fuel economy
improvements following maintenance. Therefore, the highest idle emitters
were identified so that fuel economy effects for this sample could be
analyzed. The results of this analysis indicated that those vehicles
with the highest idle HC and CO levels received a fuel economy penalty.
From a theoretical standpoint, there is no good explanation for this.
The Portland Study results discussed above for 1975-77 models are
summarized in Table 2. Results from other studies are described in the
following paragraphs.
In addition to the data from the Portland Study, EPA has analyzed data
from its Restorative Maintenance (EM) Programs in an I/H-oriented fashion.
The vehicles were screened for passage or failure of the idle test using
idle 1C and CO standards of 225 ppm and 1.0%, respectively, the standards
applied to the majority of 1975 and later model year cars in Oregon's
I/M program. FTP emission-oriented maintenance (as opposed to idle
emission-oriented maintenance) was performed to,manufacturers' specifications
-------
Table 2
Fuel Economy Before and After Maintenance-'
1975-?? Model tear Failed Vehicles in the Portland Study
--Urban-
-Highway
-Combined-
2 i
- Tills change was statistically significant at tins 0.05 luviU through the
application of a paired t-tatit. Croups with less than 20 vehicles were
nut evaluated for statistical al^uiffca.(u-.<± due to »11 uawple size.
N
Before
Maintenance
After
Maintenance
Percent
Change
Beforu
Maintenance
After
Maintenance
Percent
Change
Before
Maintenance
After
Maintenance
Percent
Clianae
All 1975-77 Models
110
15.55
15.55
0.0
21.90
21.51
-1.8
17.88
17.16
-0.7
1975 Models
1976 Models
1977 Modi;Is
36
36
3K
14.96
16.21
15.53
15.20
16.56
15.01
1,6 if
2.2 y
-3.3
21.27
22, 70
21.78
21.36
22.46
20.80
0.4
-1.1 lt
-4. 5
17.26
18.6t
17.83
17.47
18.78
17.16
1,2 1/
0,9
2!
-3.7
Highest Idle HC Emitters
Highest Idle CO Emitters
Highest Idle HC and
CO knitters
22
22
11
14.92
14, B5
15.40
14.15
14.98
15.09
-5.2
0.9
-2.0
20.88
20.81
21.54
19.70
20.23
20.36
-5. 7
-2.8
-5.5
17.12
17.05
17.67
16.20
16.96
17.00
-5.4
-0.5
-3.3
Fuel economy is measured
In in lies
per gallon.
0
1
-------
-7-
xn progressive steps by skilled mechanics. For the I/M analysis, the
"after maintenance" fuel economy was taken at the maintenance step where
the idle standards would be met regardless of whether further maintenance
was actually performed in the RM sequence. The observed fuel economy
improvement was significant in the statistical sense. Fuel economy
improved from 15.6 MPG to 16.2 MPG, or approximately 4%. These data,
which are summarized in Tables 3 and 4, suggest that up to a 4% im-
provement in fuel economy is possible with an emission-oriented mechanic
training program in place.
The California Air Resources Board (GARB) tested 150 1975-77 model year
cars in its Light Duty Vehicle Surveillance Testing Program (LDVSP-II) .
Because of more stringent emission standards for California vehicles,
these vehicles also closely represent non-California cars which will be
sold in the 49 states in model year 1980. Of the vehicles tested by
CARB, 31 failed the proposed California idle test standards of 30 ppm HC
and 0.5% CO, Following an X/M-type repair to manufacturer's specifications
by skilled CARB mechanics, fuel economy improved 1.8%, but this improvement
was not significant in the statistical sense.
Applying the 0.5% CO and 50 ppm HC idle test standards to the sample of
100 1975-76 model year California cars tested in EPA's Restorative
Maintenance Program in San Francisco, 14 vehicles were failed initially
and met the standards following maintenance. An average combined
city/highway fuel economy of 13.22 MPG before maintenance and 13.20 MPG
after maintenance was observed on these vehicles* This result corroborates
the CARB result of no fuel economy improvement for 1980 model year
Federal cars. Since emission control technology is not expected to change
dramatically in 1980, this result is somewhat puzzling and requires
further investigation once 1980 model year Federal car data become
available.
-------
-8-
Table 3
Fuel Economy Before and After Maintenance
1975-76 Model Year federal Failed Vehicles Tasted in EPA's
Restorative Maintenance Program—
(97 Cars)
Urban, Highway Combined
Before Maintenance 13.39 19-40 15.56
After Maintenance 14.10 19.69 16.17
Percent Change 5. 3 — 1.5 — 3.9 —
1/
— Harmonic means are used to calculate mean fuel economy, which
is presented in miles per gallon.
2 j
— This change was statistically significant at the 0.05 level through
the application of a paired t-test. Groups with less than 20 vehicles
were not evaluated for statistical significance due to small sample
size.
Table 4
Fuel Economy Before and After Maintenance
Eleven 1977 Model Year Failed Vehicles Tested in
EPA's Restorative Maintenance Program-
Urban Highway Combined
Before Maintenance 13.71 18.58 15.54
After Maintenance 14.39 19.05 16.17
Percent Change 5.0 —^ 2.5 4.1
~ Harmonic means are used to calculate mean fuel economy,
which is presented in miles per gallon.
If
— This change was statistically significant at the 0.05 level through
the application of a paired t-test. Groups with less than 20 vehicles
were not evaluated for statistical significance due to small sample
size.
-------
-9-
It appears that 1975-80 model year cars' fuel economy, on the average,
is unaffected by I/M if minimal maintenance is performed strictly in
response to an idle test failure. However, EPA concludes on the basis
of RM data that a strong emission-oriented mechanic training program
could result in fael economy improvements for 1975-79 model year failed
vehicles, estimated at up to 4%. EPA refrains from making a conclusion
with respect to 1980 models until data on 1980 model year Federal cars
become available,
Post-1980 Model Year Cars
The majority of I/M failures of current technology (1975-80) vehicles
results from the improper adjustment of the carburetor idle air-fuel
mixture. The proper adjustment reduces the fuel used at idle, however
idle fuel usage is only a small portion of the fuel used during typical
vehicle operation. Thus, maintenance specific to the idle system usually
results in a small improvement to fuel economy in everyday driving.
This is reflected in the fuel economy changes presented in the sections
above on pre-1981 model year cars in the Portland Study.
Beginning in 1981 there will be two types of changes in emission control
systems. Some smaller cars will retain essentially their current systems,
but their manufacturers will limit the adjustable range of the idle
air/fuel mixture. EPA hopes this will reduce the frequency of improper
carburetor adjustments among these vehicles. If it does, idle mixture
maintenance in I/M programs and the small improvement it gives in fuel
economy will also be less frequent for these cars.
The second type of change will affect the remaining smaller cars and
virtually all larger cars (excepting those which convert to Diesel
engines). These cars will also have reduced adjustability, but more
importantly they will have new emission control systems. In addition,
the new emission control systems will commonly incorporate three-way
catalysts with feedback carburetion which will, through the input from
an oxygen sensor in the exhaust, select the proper air-fuel mixture for
all operating conditions, including idle. Failure of the oxygen sensor,
the mini-computer, and many of its other input signals, will likely
cause the carburetor to supply much more fuel than is required for best
fuel economy and emissions at operating conditions such as acceleration
and cruise, as well as at idle. Maintenance performed on these systems
can thus be expected to increase fuel economy much more than with current
technology cars.
EPA has tested several vehicles which are equipped with feedback carburetion
systems. These data are summarized in Table 5. It is anticipated that
these vehicles will be representative of 1981 and later production
Federal vehicles (1980 and later in California). The results, although
based on only a few vehicles, indicate about a 9 to 30 percent potential
improvement in fuel economy "when repairs to the engine control system
are made. The fleetwide fuel savings cannot be estimated until exper-
ience with the frequency of failures and improper maintenance is obtained.
-------
Table 5
Repaired Vehicle Fuel Economy Benefit ,
Expected for 1981 and Later Model Year Vehicles —
Fuel Economy Fuel Economy ,
Vehicle Description before Maintenance after Maintenance- Percent: Change
1 1978 Pontiac Sunbird 21.6 MPG - 25.4 MPG 17.6%
2/
2 1978 Ford Pinto 21.9 MPG - 24.6 MPG 12,3%
3 1978 Pontiac Sunbird 19.9 MPG - 23.6 MPG 18.6%
2/
4 1978 Pontiac Sunbird 18.9 MPG - 20.6 MPG 9.0%
5 1978 Ford Pinto 21.3 MPG —^ 24.2 MPG 13.6%
6 1979 Mercury Marquis 13,4 MPG ~ 17.4 MPG 29.9%
7 1981 GM prototype - 16.2 MPG - 18.6 MPG 14.8%
— Each vehicle was tested first in tuned-up condition, and then with the
failure modes noted below. Testing waa performed by EPA, except as noted.
2/
— In tuned-up condition except with oxygen sensor disconnected; combined
city/highway fuel economy. The oxygen sensor is a maintainable item,
requiring periodic replacement.
3/
-r In tuned-up condition; combined city/highway fuel economy.
~ In the "before maintenance" test, the vehicle was in a tunedup condition
except that the "limited operating strategy," which can occur following
failure of the on-board computer, was in effect.
5/
— Testing was performed by GM, Data are preliminary.
— Tu the "before: maintenance" test, the vehicle was in a tuned-up condition
except that the carburetor waa operating as rich as possible. This mode
closely approximates that which would be caused by a disconnected oxygen sensor.
0
1
-------
Attachment I
Summary of Studies on Fuel Economy Benefit Bye to Inspection and Kalntenance
Study
Sample
1/
Screening Test Used
to Determine Failure
Level of Mechanic
Skill/Training ff
Relevance 11:
i/H y
Repaired Vehicles
Average Fuel Economy
Benefit »
I. EPA'b Portland Study
?• I/M-Ortented Analysts
of EPA'a KM 76 Program
3. I/H-Oriented Analysis
of EPA'b RHn Progm
KPA'a Analysis of Data
from CARB's LUVSP-II
5. EPA'b Analysla of Data
on 3-Hay CnCalynr
Equipped Vehicles
6. EPA'b Analyst* of Data
froa CAlB'e Riverside
Program
7. MITSA 322-Car Study
8. KSTSA 57-Car Study
9- Champion Spark Plug
Co, studies
93 1972-74 Models
11C 1975-77 Models
97 1975-76 Models
11 Vary Low
HIleage 1977 Models
31 1975-77 Models
7 Feedback Carburetlon
Vehicles
349 1955-74 Models
322 1968-7 3 Models
57 1969-72 Models
Oregon State Inspection Average
Test (MX failure rate)
(idle Test + Brief Phy.lnsp)
Idle Test (301 failure rate) KiKh
Idle Test (25!! failure rate) High
Idle Test (20* failure rate) High
Visual Inspect Ion of Oxygen High
Sensor or Spark Timing
Idle Teat <35% failure rate) Average-High
Loaded or Idle Test Average
Loaded Test High
310 pre-68 to 1975 Hodeis Extensive Diagnostic Inspection High
(mostly pre-75s)
10. CARB Degradation Study 59 196B-74 Models
11- Private Vehicle Fleet
I/H
12. EPA*s l/H Information
Document(EPA-400/2-78-
001}
13. J. Panzer"a Model of
Fuel Economy Benefit
U.S. Postal Service,
Baltimore Cty.Trana.,
& A.T.fc T. fleets
Mathematical Model
Mathematical Model
Idle Teat W02 failure rate) mgi,
Emission Teats High
Idle Test (varying fallucr rate)
High
HiRh
Med I ma
High
High
Low
Low
Low
Low
Low
Low
Low
Low
1972-74,; Q.fc*
5/
19/5-77:—0.11 V
3,M -V
4.OK V
5/
1.8% ~
9S-30X-'
Pre-71: 3. U hJ
1971-74: 2.02 §-'
4.72 V
3.5% 5'
ll.« 2'
i. ik y
M - 10X
« - 102
SZ - 1«
iy/
11/
It All vehicles described here have undergone maintenance due to failure of screening teat.
If "Average" mmant that naintenance was performed by mechanics who have not been specially trained in ernlaglun-nrlented maintenance tectmti|>ies.
~ "High" means that naintenance was performed, usually to manufacturer's specificatione, by skilled mechanics or i eclmir tuns.
3/ Relevance to l/H is determined on the basis ot how well the study's design represents anticipated I/M situations, and also on tmw well the technology of
fleet of vehicles studied represents the early 1980's T/K target populations.
4/ As measured Ismedlatelf after maintenance • except as noted.
5/ IPA rity/higliway combined fuel ecunumy (55X city driving, 455E highway driving).
1/ As measured ovef the 75 FTP (EPA1* city driving rycle).
7/ As reported by vehicle owners.
5/ After full tune-up to manufacturer's specif km Inn, plus mandatory replacement of spark plugs. Fuel eronmity Is measured tiver n transient
cycle representative of the Toledo metropolitan area's driving patterns.
9/ As measured over the 72 FTP (hag I and 2 of the 75 FTP) Tills assumes that nil vehlclp start-ups are from a rnld engine i ..ml H t.m
To/ No Standard test procedure was used to determine fuel economy benefit; not necessarily Hs measured turned la u-J y after imi Iill en.tm «.
11/ Kitlnatsd annual fuel economy benefit,
-------
Attachment 2
Bibliography of Studies Related to
the Effects of Inspection and Maintenance
Programs on Fuel Economy
Portland Study Interim Analysis: Observations on 51^ Months
of Vehicle Operation; l/M. Staff, Emission Control Technology
Division, Office of Mobile Source Air Pollution Control, U.S. EPA;
January 1979.
Ail Evaluation of Restorative Maintenance on Exhaust Emissions
from In-Use Automobiles; White, J.T.; SAE Paper 780082: March 1978.
Vehicle Inspection and Maintenance - The California Program;
Rubenstein, G., Ingels, R., Weis, R., Wong, A.; California Air
Resources Board; SAE Paper 760557; June 7-10, 1976.
Motor Vehicle Diagnostic Inspection Demonstration Program -
Summary Report; Innes, J. and Eder, L.; NHTSA Technical Report,
DOT HS-802 760; October 1977.
Fuel Economy Improvements through Diagnostic Inspection;
Bayler, T. and Eder, L.; NHTSA Technical Report DOT HS-802 284;
March 1977.
How Passenger Car Maintenance Affects Fuel Economy and Emissions:
A Nationwide Survey; Walker, D.L., Boord, J.O., Pigott, J.S., and
Sutton, E.R.; Champion Spark Plug Company; SAE Paper 780032;
February 27-March 3, 1978.
Car Maintenance Around the World; Champion Spark Plug Company;
August 1978.
Final Report - Degradation Effects on Motor Vehicle Exhaust Emission;
Olson Laboratories, Inc., Anaheim, California; March 1976.
Information Document on Automobile Emission Inspection and Maintenance
Programs; Kincannon, B. and Castaline, A., pp. 15-17; EPA-400/2-78-001,
February 1978.
Fuel Economy Improvements through Emissions Inspection/Maintenance;
Panzer, J., Exxon Research and Engineering Company; SAE Paper 760003;
February 23-27, 1976.
------- |