The Effects of Fuels on Engine Throttle Transients
by
Jean-Charles W. Bossert
Bachelor of Science in Mechanical Engineering
The University of Texas at Austin
(1992)
SUBMITTED TO THE DEPARTMENT OF
MECHANICAL ENGINEERING IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
MASTER OF SCIENCE IN MECHANICAL ENGINEERING
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
February 1994
© 1994 Massachusetts Institute of Technology
All rights reserved
Signature of Author
Department of Mechanical Engineering
January 14, 1994
Certified by
I
Wai K. Cheng
Associate Professor, Department of Mechanical Engineering
Thesis Supervisor
Accepted by
Ain A. Sonin
Chairman, Department Graduate Committee
MASSf, l4X'W'
MAR 99
t:q,9
4
(This page has been intentionally left blank.)
2
The Effects of Fuels on Engine Throttle Transients
by Jean-Charles Bossert
Submitted to the Department of Mechanical Engineering on January 14, 1994
in partial fulfillment of the requirements for the degree of
Master of science in Mechanical Engineering
ABSTRACT
Experiments to evaluate the effects of fuel composition on engine throttle
transients were performed. The goal of these experiments was to interpret and quantify
the difference 'between the amount of fuel injected and the amount of fuel burned as a
change in liquid fuel mass inside the intake port during accelerations. The fuels chosen
for this experiment were methanol-indolene blends: MO, M10, M25, M40, M60, M85,
and M100. The transient was simulated by simultaneously opening the throttle while
changing the amount of fuel injected to the engine to a predetermined fixed flow rate.
The engine speed was kept at a constant 1800 Rpm while the intake pressure increased
from 0.7 bar (part throttle) to 1.0 bar (wide open throttle). Once the engine reached
steady state at wide open throttle, the throttle was returned to its original position.
During the experiment, the ignition timing was kept constant, halfway between the MBT
points of the wide open throttle condition and the part throttle condition.
The
measurements included the in-cylinder pressure, the fuel flow rate, and the exhaust
relative air-fuel ratio. Each fuel was tested under six conditions: three levels of
enrichment at wide open throttle (=1.0, k=0.9, X=0.8), and two sets of engine coolant
temperatures (T=400C, and T=80°C).
Using a three dimensional engine map of the fuel conversion efficiency it was
possible to relate the Gross Indicated Mean Effective Pressure to the amount of fuel
burned during each cycle. The map included the effects of the relative air to fuel ratio,
the Gimep, and the ignition timing.
The data showed a uniform increase in the change of fuel film mass from MO to
M100 for the warmed up engine condition. Under stoichiometric condition the liquid
film mass increased from 0.10 to 0.22 gram. However, in the low engine coolant
temperature condition, there was no measurable change in fuel film mass from MO to
M60. A transition occurred between M60 and M85, where the change in fuel film mass
increased by 200% from 0.2 to 0.4 gram.
The data indicates that there is a good correlation between the mass of fuel
injected at WOT and the change of liquid film mass during the transient. Although the
correlation is not as good for low coolant temperatures, the low temperature trend can be
explained with a one dimensional heat transfer model of the intake port. The model
indicates that under low coolant temperature, the port wall remains close to the boiling
point of methanol under low intake pressure (part throttle). However, under high intake
pressure (wide open throttle) the average port wall temperature is below the boiling point
of methanol, causing a significant increase of liquid film mass inside the intake port. The
model also shows that the increase in volatility of methanol blends is offset by the
stoichiometric requirement of the fuel.
3
ACKNOWLEDGMENTS
The three semesters I have spent at the Sloan Automotive Lab. kept me busy.
This experience has been very rewarding and taught me a lot about research and graduate
school. All this could not have been possible without the help and contribution from so
many people. Where should I start'?
Without his
The first person that I should thank is professor Wai Cheng.
guidance, and knowledge I would probably still be taking data right now ..... I also want
to thank Professor John Heywood and Simone Hochgreb who have also greatly
contributed to this research with their ideas during the meetings. These meetings were
insightful and kept my research aimed in the right direction. However, all of this could
not have been possible without the funding provided by the Industrial Consortium on
engine/fuel Interactions. The consortium includes Chevron Research corp., the
Department of Energy, Exxon Research and Engineering Company, Nippon Oil
Company Ltd., and Shell Oil Company.
I want to use this opportunity to thank everybody who works at the Sloan Lab.
Brian Corkum has played an important role in helping me with the experimental set up.
Brian knows now "Everything You Always Wanted to Know about Dynamometers but
Were Afraid to Ask". I also want to thank Joan Kenney who kept my spirits up and
made sure that all the red tape was done on time....something that is easier said than
clone....But wait I am not finished....
All my friends at the Sloan Lab need to be thanked, specially my office mates:
Pete, Jon, Koji, and Tung-Ching. These are the people that had to put up with me for a
year and a half...I think they also deserve a diploma. But most of all I want to thank
ALL my friends here at the lab for their patience, help, and expertise: Pete for ALL the
questions he had to put up with, Jon for his Computer-Analytical-Math skills, Olivier the
only one who could really "understand" me, the California man-Ed for his expertise on
Kuo-Chiang for his expertise on fuel volatility
everyMicrosoft-computer-program,
modeling, Mike his expertise in turbulent flow and diffusion, K.D.M's knowledge of
everything, Christine's company during the night shift, and Gatis' help in the math
course.
Finally I want to thank my parents without whom all of this would not have been
possible...to whom I owe so much ..... and Gail who made it all worthwhile.
Jean-Charles Bossert
January 1994
4
Table of contents
ABSTRACT
......................................................................
3
ACKNOWLEDGMENTS ..................
......... .....................................
CHAPTER 1 -- INTRODUCTION ..............................................................................
1.1 Motivation .................................................................................................
1.2 Background.....................................................................
1.2.1 Transient Testing .........................................................................
1.2.2 Emissions Control ..................................................
1.2.3 Air/fuel Control ...........................................................................
1.2.4 Air/fuel Transport..................................................
4
8
8
9
9
10
11
12
1.3 Objective ...................................................
14
1.4 Methanol/Indolene Blends ..................................................
CHAPTER 2 -- EXPERIMENTAL APPARATUS ...................................................
2.1 Engine and Dynamometer ...................................................
14
16
16
2.2 Sensors and Measurements............................ ......................
18
2.3 Fuel and Air flow control ..............................................................
18
CHAPTER 3 -- EXPERIMENTAL PROCEDURE ...................................................
3.1 Introduction.........
.........
.........
21
....................................................................
21
3.2 Fuel Matrix ..................................................
3.3 Test Matrix ................................................................................................
3.3.1 Operating Conditions....................................................
3.3.2 Control Parameters ...............................................................
CHAPTER 4 -- DATA INTERPRETATION ..................................................
4.1 Introduction ..................................................
4.2 Air and Fuel Flow ............................................................
21
21
21
22
25
25
25
4.2.1 Air Flow ...........................................
.................................. 25
4.2.2 Fuel Flow ..................................................
26
4.3 Data Analysis ..................................................
29
4.3.1 Methodology ..................................................
29
4.3.2 Accuracy of the mass of fuel unaccounted for ............................. 31
4.4 Results ...................................................
4.5 Fuel effects on driveability ..................................................
34
38
CHAPTER 5 -- MODEL AND DISCUSSION ............................................................ 42
5.1 Discussion.........
..... .........
............................42.........
42
5.2 Indolene Model .................
47.....................................
47
5.3 Modeling the Intake Port..................................................
49
5.4 Modeling Results ..................................................
52
CHAPTER 6 -- CONCLUSION ......................
.......................................... 57
6.1 Review of the Methodology ..................................................
57
6.2 Model and Discussion ..............
................... 5........................
58
REFERENCES ..................................................
60
APPENDIX A ......... ......... ......... ......... ..........................................................................
62
APPENDIX B ...................................................
5
77
List of Figures
Figure 1.2.1
Speed versus time for the FTP 75 test(l) ..............................................
9
Figure 1.2.2
Efficiency of a three way catalytic Converter (2).................................. 10
Figure 1.2.3 Diagram of a typical control system for a modern engine ..................... 12
Figure 1.2.4 Typical air-fuel ratio excursions during the FTP cycle(l) ..................... 12
Figure 1.2.5
Schematic of a typical intake port in a modern engine ..........................
13
Figure 2.1
Schematic of the intake system.............................................................
17
Figure 2.2
Experimental set up ......................................................
21
Figure 3.1
Air and fuel flow rate during the experiment ........................................
24
Figure 4.1
Figure 4.2
Mapping of the air flow as a function of compression pressure ............. 26
Mapping of the gross indicated fuel conversion efficiency .................. 27
Figure 4.3
Gross indicated fuel conversion efficiency versus ignition timing......... 28
Figure 4.4
The first three steps in the Analysis for MO ..........................................
Figure 4.5
The cumulative mass of fuel unaccounted for for MO0
........................... 31
Figure 4.6
Adjustment of the mass of fuel injected at wide open throttle ............... 32
Effect of measurement error on the cumulative mass of fuel
Figure 4.7
unaccounted for.........
Figure 4.8
.........
30
..................................................................
33
Obtaining the mass of fuel injected at wide open throttle from the
cumulative mass of fuel burned ..................................................
34
Figure 4.9
Change in fuel film mass for the low temperature experiment ............. 35
Figure 4.10
Change in fuel film mass for the high temperature experiment ............ 35
Time constant t I for the cold coolant temperature experiment ............ 36
Figure 4.11
Figure 4.12
Figure 4.13
Figure 4.14
Figure 4.15
Figure 4.16
Figure 4.17
Figure 4.18
Figure 4.19
Time Constant I for the hot coolant temperature experiment ............. 36
Time Constant t2 for the cold coolant temperature experiment ........... 37
Time constant t2 for the hot coolant temperature experiment ............... 37
Sample Gimep curve with the 0-90% response time .............................
Effect of methanol content on the response time under cold coolant
condition (T=40o C ) ..................................................
Effect of methanol content on the response time under hot coolant
condition (T=80oC)..................................................
Effect of lambda target on the response time under cold coolant
condition (T=40oC) ..................................................
Effect of lambda target on the response time under hot coolant
condition (T=80oC) ..................................................
6
38
39
39
40
40
Figure 5.1
The relative effect of methanol content on the change in liquid
film mass under cold coolant conditions ............................................... 42
Figure 5.2
The relative effect of methanol content on the change in liquid
film mass under hot coolant conditions ...........................
Figure 5.3
Figure 5.4
Figure 5.5
.......... 43
Correlation between the change of liquid film mass and the change
of fuel mass injected (T=40oC) ...........................................................
45
Correlation between the change of liquid film mass and the change
of fuel mass injected ( T=80oC) .
.....................
......
Correlation between the change of liquid film mass and the mass
46
of fuel injected at wide open throttle
46
.............................................
Figure 5.6
Schematic of the Intake Port wall .........................................
Figure 5.7
Effects of Methanol Content on the change of liquid fuel film
mass for the model .........................................
50
53
Figure 5.8
Temperature profile along the port wall (Tc=390oK) ........................... 54
Figure 5.9
Figure 5.10
Temperature profile along the port wall (Tc=470oK) ........................... 54
Mass fraction of fuel evaporated along the port wall (Tc=390oK) ........ 55
Figure 5. 11
Mass fraction of fuel evaporated along the port wall (Tc=470oK) ...... 55
List of Tables
'Table 1. I Data on fuel Properties ....................................................
'Table 2.1 Engine Specifications ....................................................
15
17
Table 5.1 Distillation Parameters for Indolene and the Indolene Model ....................... 47
Table 5.2 Content of the Indolene Model ....................................................... ............ 47
7
CHAPTER 1 -- INTRODUCTION
1.1 Motivation
Since the early 1970's, the Environmental
Protection Agency has been forcing
automobile manufacturers for engines with greater fuel efficiency and lower emissions.
The clean Air Act of 1990 and the California Air Resource Board have set strict
emissions guidelines for the up coming years.
In order to meet these stringent requirements,
the automotive
industry has
developed complex engine control systems, sophisticated fuel metering strategies and the
associated sensors to improve emissions. Along with the hardware development, there is
substantial research using reformulated fuels to reduce emissions. For example,
methanol/gasoline blends offer promises in lowering emissions while decreasing the
United States' dependency on foreign oil imports.
Another important motivation for this research is the consumer's satisfaction. The
oil industry tries to satisfy its customers by providing fuels which give good driveability
while meeting the emission levels set by the EPA. Therefore it is important to understand
the phenomena that causes hesitation (uneven acceleration), and stumble (stepping during
acceleration).
The above discussions motivates the theme for this thesis: to study the fuel
effects on the transient behavior of spark ignition engines with a matrix of
methanol/gasoline mixtures.
8
1.2 Background
1.2. 1 Transient Testing
Since 1975 all EPA certified vehicles must pass the Federal Testing Program
(FTP). The test consists of a driving cycle simulating a typical trip. The vehicle is
installed on a chassis dynamometer while the speed and load are changed.
Figure 1.2.1
shows the speed versus time for the FTP 75 test (l). During the test, samples of the
exhaust gases are taken and analyzed. Under 1993 standards, each vehicle must not
generate more than an overall 0.25 gram per mile for Non-Methane Organic Gases, 3.4
gram per mile for Carbon Monoxide, and 0.4 gram per mile for NOx. Each vehicle must
go through the same transient test starting under cold and warmed up conditions.
70
....
Kso
50
E 40
'
30
> 20
10
0
0
300
600
900
1200
1500
100
Time, seconds
Figure 1.2.1
Speed versus time for the FTP 75 test#e.
Because of the various vehicle accelerations and decelerations in the FTP cycle, the test
9
procedure
is transient in nature in the sense that there are rapid throttle and speed
changes.
1.2.2 Emissions Control
Today's engines have a complex control system designed to optimize the engine's
performance and emissions level.
Emissions control hardware such as the catalytic
converter set very strict guidelines over the air fuel ratio range.
window of operation of a three way catalytic converter 2 .
Figure 1.2.2 shows the
Outside this window of
operation, the catalytic converter's efficiency drops dramatically. For example, during a
fuel rich situation, not enough oxygen is present in the exhaust gases, and the catalyst is
unable to oxidize carbon monoxide and the unburned hydrocarbons. The reverse happens
during lean conditions, too much oxygen is present, and the catalytic converter is unable
to reduce NOx. Therefore, it is crucial to keep the engine operating inside the specified
air/fuel control window to minimize the HC and NOx excursions.
100,
o-
U
-
U
14.3
14.4
14.5
14.6
Rich
Air/fuel ratio
14.7
Lean
14.8
Figure 1.2.2 Efficiencyof a threeway catalyticConverter(2).
10
14.9
1.2.3 Air/fuel Control
The mixture preparation system of a modern engine provides a mixture of precise
fuel air ratio delivered to the engine. Figure 1.2.3 shows a schematic of a typical control
system. The air mass flow meter (usually a hot wire or film anemometer) sends a signal
to the controller (the ECU).
To keep the engine stable during rapid accelerations or
decelerations, the throttle plate position sensor is used as a feed forward signal. Based on
the air mass flowing to the engine, the controller determines the amount of fuel that
needs to be injected in order to keep the mixture stoichiometric.
sensor in the exhaust senses the relative air/fuel ratio burned.
Finally, the lambda
This signal is fed to the
controller as an error signal. Therefore, the controller compensates partially when the
exhaust gas air fuel ratio falls outside the specified range. Although lead/lag
compensation schemes are used, the fuel metering system cannot always keep up with the
air/fuel excursions during transients
3-).
"li.
C
r,
P
S
S
wire
Anemometer
MnO
ThrottLe
Position
Sensor
Lambda
Sensor
Figure 1.2.3 Diagramof a typical control systemfor a modernengine
11
(
Figure 1.2.4 shows typical air-fuel ratio excursions during the FTP cycle test . There
are three reasons for these fluctuations:
the response time of the sensors, the controller's
algorithm, but most important of all, the physical processes inside the intake port.
18"
I
-
l!
-
17-
1.2
15i
i
14-
1$1
ji1ttT
13- .I
L
12.
._ I
,-
0
X
X
X
300
r
|
|
900600
w
1200
s
1500
1800
ime, sconda
Figure 1.2.4 Typicalair-fuelratio excursionsduring the FTP cycle(l).
1.2.4 Air/fuel Transport
The most important factor affecting air-fuel ratio control during transients
are the physical processes inside the intake port. Most engines today use a Multi-point
fuel injection system, the system provides fuel to each cylinder by an individual injector.
The injector is usually located approximately 10 cm up the intake port, aiming directly at
the back of the intake valve. Figure 1.2.5 shows a schematic of a typical intake port in a
modern engine.
12
Figure 1.2.5
The injector
Schematic of a typical intake port in a modern engine.
provides
fuel metering
and fuel atomization,
which facilitates
evaporation of the fuel before entering the combustion chamber.
the
After the fuel is
injected, the liquid breaks up into fuel droplets. Some of the droplets evaporate in the air
stream, others deposit on the port wall or the back of the intake valve. The droplets
depositing on the port wall form a liquid film. This liquid film can evaporate or partially
flow into the combustion chamber depending on operating conditions, such as speed,
load and temperature. During transients, the manifold pressure and air flow rate change.
These changes disrupt the dynamic equilibrium in the manifold.
Mass transfer
mechanisms such as diffusion and evaporation are altered, changing the size of the liquid
film in the intake port. Although several computer models of the mass transfer processes
inside the intake port have already presented( 4 ,5,6,7), no model has yet been able to predict
the effects of fuel composition on transient response. There are three major modes of
fuel mass transfer into the cylinder:
I. Approximately adiabatic evaporation of fuel droplets in mid-air, and the subsequent
13
flow of fuel vapor into the cylinder during the intake flow.
2. Approximately
isothermal evaporation from the liquid fuel film on the port wall and
the subsequent flow of fuel vapor.
3. The liquid fuel film flow into the cylinder. The flow is driven by shear forces created
by the air flow during the intake process.
1.3 Objective
The objective of this research is to relate the engine throttle transients behavior to
fuel properties.
A spark ignition engine has been tested with a set of 7 indolene/methanol
blends. The experimental results will be interpreted via the fuel transport processes inside
the intake manifold:
-The effects of fuel composition on the fuel film behavior.
-The effects of coolant temperature on the fuel film behavior.
-The physical processes governing fuel mass transfer through the intake port.
1.4 Methanol/Indolene Blends
Because methanol is an important choice as an alternative fuel, gasoline and
methanol blends are used as the fuel for the experiments.
Methanol has advantages over gasoline in emissions characteristics. For example,
methanol contains no sulfur, therefore does not create any sulfates during the combustion
process. Other advantages include lower emissions of CO and Hydrocarbons. Methanol
also has definite advantages from an engine designer's standpoint.
methanol
has an antiknock
index of 99, enabling
the designers
For example,
to increase
the
compression ratio, increasing the fuel conversion efficiency of the engine. These
14
advantages have boosted research in methanol, making it a feasible solution for meeting
the up coming fuel economy and emission standards.
Table 1.1 Data on fuel Properties
Fuel
Methanol
Indolene
CnH
Chemical Formula
CHIOH
Molecular Weight
32.04
-110
Antiknock Index:
99
87-84
6.47
14.6
20.0
44.0
1103
350
R7 n
(RON+MON)/2
Stoichiometric A/F Ratio
HV (MJ/kg)
QN
Heat of Vaporization: k/kg
However, methanol has certain disadvantages that engineers still haven't been
able to overcome. Methanol generates formaldehyde during combustion, a carcinogen
and an ozone forming gas. Other disadvantages include a lower energy density and a
lower stoichiometric air fuel ratio than gasoline. This means that an automobile would
require twice as much methanol to travel the same distance compared to an automobile
fueled with standard gasoline.
Other problems arise due to methanol's corrosive
properties. Fuel systems would need to be redesigned to be compatible with methanol.
Another issue with methanol is safety. During combustion, methanol emits very little
light.
It would be a major safety hazard during accidents. Finally,
due to its low
stoichiometric air fuel ratio and its high heat of vaporization, methanol exhibits poor cold
startability(8 ). Table 1.1 compares the properties of methanol with the properties of
indolene.
15
CHAPTER 2 -- EXPERIMENTAL APPARATUS
.2.1 Engine and Dynamometer
The engine used in this research consists of a production spark ignition four
stroke engine (Nissan SR20DE).
The four cylinder engine has been modified to run on
only one cylinder (cylinder #4). Figure 2.1 shows a schematic of the modifications done
to the intake system. The intake runners to cylinder number 1, 2 , and 3 were blocked
past the plenum chamber. Table 2.1 shows the engine specifications.
AIR
Figure 2.1
Schematic of the intake system
16
Table 2.1 Engine Specifications
Model
Type
Nissan SR20DE
4cylinders,4valves/
cylinder aluminum block
and head.
86
Bore (mm)
Stroke (mm)
Clearance
86
Volume
58.8
(cm 3 )
Displacement (cm3)
Compression Ratio
Valve Timing
1998
9.5:1
IVO: 13 BTC
IVC: 55 ABC
EVO: 57 BBC
EVC: 3 ATC
This engine has been in production since 1990 and is representative of today's state of
the art engine design. The engine uses a port fuel injection system. Each injector forms 4
sprays, two sprays are aimed at the back of each intake valve, while the other two collide,
enhancing mixing and evaporation.
To isolate the effects of fuel composition on
transients, the stock Electronic Control Unit has been replaced with a custom made open
loop controller.
The new controller gives us complete control over the injection pulse
timing, injection duration, and the ignition timing. The engine coolant circuit has also
been modified. The circuit is now connected to a heat exchanger running with city
water. A gate valve provides the different levels of cooling necessary.
The engine is coupled to a dynamometer. The dynamometer has a three phase
induction motor, a DC clutch and an eddy current brake. The motor is used to motor the
engine during start up. The eddy current brake is used to control the engine speed during
the experiment. The engine's crankshaft rotation is monitored via an inductive pick up.
The sensor is hooked up to the dynamometer controller, whose main purpose is to keep
the speed constant.
17
2.2 Sensors and Measurements
The air mass flow to the engine is monitored by a Kurz 505 laminar air flow
element.
The fuel flow to the engine is monitored by a digital oscilloscope, measuring
the injection pulse duration (the injected volume per pulse as a function of pulse width
has been calibrated separately by collecting the fuel injected in an ice-cooled cylinder).
The exhaust gas equivalence ratio is measured with an Horiba lambda 110 controller and
an oxygen sensor.
The controller may be programmed
to compensate
for the fuel
composition and the oxygen sensor sensitivity.
The cylinder pressure is measured with a Kistler 6051A piezoelectric pressure
transducer. The output of the pressure transducer and the Horiba controller are converted
to digital signals via a standard data acquisition card and stored on a hard disk.
The
clock of the acquisition card is triggered by the crank shaft encoder, giving one pulse
every crank angle degrees (720 degrees per cycle).
Therefore, one pressure point
measurement is recorded every crank angle degree.
Thermocouples
have been installed in the intake manifold, coolant system, oil
pan, and exhaust pipe. These thermocouples assure us that the test conditions remain
similar during each experiment.
2.3 Fuel and Air flow control
In order to simulate a throttle transient, both the air flow and the fuel flow to the
engine must both be changed simultaneously. Figure 2.2 shows a schematic of the
experimental set up. The throttle plate is connected to a pneumatic actuator, connected to
a 40 psig air tank. The actuator is controlled by a 12 Volts DC three way valve.
The
system provides opening and closing of the throttle plate in the order of 30 to 40 msec.
18
The fuel injector is connected to a custom made injector driver.
The injector
driver amplifies TTL pulses to open and close the injector. The switch that controls the
opening and closing of the throttle plate also controls which of the two pulse generators
is used to control the injector driver. During accelerations the pneumatic actuator opens
the throttle plate at the same time that the injection pulse is changed from T1 to T2.
Each of the two pulse generators are triggered at 20 degrees before top dead center of the
compression stroke (i.e. the injection is done once per cycle during closed valve).
19
r
Figure 2.2
Experimental set up.
20
CHAPTER 3 -- EXPERIMENTAL PROCEDURE
3.1 Introduction
The following chapter presents the experimental test matrix as well as the
complete data set recorded.
The description
will start with the fuel matrix chosen,
followed by the test matrix, and finally by the testing procedure.
3.2 Fuel Matrix
The parameter that we control over each experiment is the fuel composition.
We
chose 7 fuel blends ranging from indolene to pure methanol: MO, M10, M25, M40, M60,
M85, M 100. The number following the M describes the volume percentage of methanol
in an indolene fuel, i.e. M40 signifies 40% methanol, 60% indolene by volume.
The
blending of the fuels was done by refrigerating the fuels at OOCfirst to minimize vapor
losses.
3.3 Test Matrix
3.3.1 Operating Conditions
The experiments were designed to simulate a typical FTP driving cycle
transient described in section 1.2.1. We chose to recreate a sudden throttle opening from
a part throttle condition, followed by a sudden closing of the throttle back to the original
position.
Each experiment can be broken down into two sections: the up transient
21
(starting from part throttle at steady state finishing at steady state at Wide Open Throttle
(WOT)), and the down transient (starting from steady state at WOT finishing at steady
state at part throttle).
All the experiments were performed at a constant speed: 1800
Rpm. The intake pressures were 0.7 bar for the part throttle case and 1.0 bar for the wide
open throttle case. Other parameters that were kept constant throughout each experiment
include the coolant temperature, the ignition timing, and the injector timing.
3.3.2 Control Parameters
Each fuel was tested under 6 conditions. To simulate the typical response of the
engine's ECU three different levels of enrichment were used during wide open throttle: X
tl= 1.0, Xt2 =0.9, Xt3 =0.8. These levels of enrichment are referred to as Xtarget(Xt).. The
second parameter
changed was the coolant temperature.
All the experiments
were
performed under two coolant temperatures: T 1=400C, T2 =800C. The two temperatures
chosen simulate a warming up condition and a fully warmed up condition. Table 3.1
shows the test matrix.
Table 3.1 Test Matrix
The fuel flow to the engine at WOT was predetermined from the steady state fuel flow.
For example, in the case where Xt=0.9, the engine was ran at WOT with
=-0.9 and the
fuel injection pulse was recorded. This injection pulse became T2 in the experiment and
remained fixed during the rest of the experiment.
22
Figure 3.1 shows the air and fuel mass
flow to the engine during one of the experiments.
AA
U.U4.
o.w
I
-
,-
U.q4
0.4
I
0.035
-0.35
0.3
to
-25
I 0.025
.<-
co
-02t
I
i
I
- 0.15 o
C
0.02 -
- 0.1
_i
0.015
0
0.05
I
i
50
100
I~
1I0
200
250
Cycle Number
Figure 3.1
Air and filelflow rate duringthe experiment
3.4 Test Procedure
All the experiments were performed at ambient temperature (about 220 C). Prior
to the start of the experiment, the engine was warmed up, and the MBT timing was
recorded for the part throttle and WOT condition. The ignition timing was then set at the
average of the two MBT points.
In all the experiments the difference between the two
MBT points was less than 4 degrees. The engine was run stoichiometric at 1800 Rpm at
part throttle (intake pressure = 0.7 bar) for 15 minutes. The fuel injection pulse width
was recorded and entered as T 1. The engine was then run 3 minutes at WOT (Pi = 1 bar)
at the pre-selected enrichment level. The fuel injection pulse width was recorded and
entered as T2 . The intake pressure was returned to 0.7 bar and the coolant system
adjusted to maintain the selected temperature. Then the data acquisition system was
started. Ten seconds later, the throttle control switch was turned on for 1 minute. The
23
switch was then returned to the part throttle position.
turned off once the engine returned to steady state.
24
The data acquisition system was
CHAPTER 4 -- DATA INTERPRETATION
4.1 Introduction
The main goal of the data analysis is to account for the air and fuel flow during
each cycle.
Using the data recorded, such as the lambda meter output, and the in-
cylinder pressure measurement, expressions for the mass of fuel and air burned can be
calculated.
4.2 Air and Fuel Flow
4.2.1 Air Flow
During transients, the air mass flow to the engine changes rapidly. Because the
engine has been modified as a single cylinder engine,
difficult for conventional
the pulsating air flow makes it
instruments to accurately measure the air flow on a cycle by
cycle basis. To obtain an accurate measurement of the air flow, the air flow rate to the
engine was mapped under steady state conditions as a function of compression pressure
before spark (1 degree before ignition). Figure 4.1 shows the change of compression
pressure for different air flow rates to the engine under several loads and equivalence
ratios.
Figure 4.1 indicates that the mapping is linear, and independent of the
equivalence ratio. However, depending on engine temperature and ambient conditions,
the slope of the line shown on figure 4.1 can change. Therefore, by using the steady state
air flow measurements from the laminar air flow element at the steady-state part throttle
and the wide open throttle conditions, we can find two points determining the equation
of the line for each experiment.
Once the equation
25
for the line is known,
the
compression pressure before spark is used to find the mass of air inside the cylinder
during the transient.
4'"
1
-
11
.ck
u.
a 10
2
8-
a0
0.
7
5.
4.
3
3.5
4
5
4.5
6
5.5
6.5
Air Flow Rate (g/s)
*
Figure 4.1
lambda=lI
*
lambda=1.1
lambda=0.9 -
- -
Straight line
Mappingof the airflow as a function of compressionpressure.
4.2.2 Fuel Flow
In order to obtain a mass balance of the fuel flow, two fuel flows are needed: the
mass of fuel injected per cycle, and the mass of fuel burned during each cycle. The mass
of fuel burned per cycle could be directly calculated from the lambda sensor output.
However, it has been shown(9 ) that lambda sensors have response times on the order of
100 msec, which is inadequate for calculations on a cycle by cycle basis. Therefore a
new method was developed, using the Gross Indicated Mean Effective Pressure (Gimep)
and Tlig(the gross indicated fuel conversion efficiency). The Gimep is a measurement of
the work done by the engine during the compression and the expansion stroke divided by
26
the cylinder volume. The Gimep can be calculated by integrating the cylinder pressure
over the displacement volume.
i PdV
Gimep=
Vi-
(eq. 1)
The Gimep can also be related to the mass of fuel burned using 'ig.
Gimep = rfigx
Mburned
QHV
(eq.2)
The engine was mapped under several operating condition to measure Tlig. Figure 4.2
shows
ig as a function of
0.35 -
and GIMEP.
, _--
-
U 0.34
C
U
w
...-
13U ..---'
0.33
0
oo
.
3
//
C 0.32
C
-
0.31
0.3
GIMEP= 8.4 bar
o
GIMEP-6.8 bar
*
GIMEP= 5 bar
: /
U029
UL
, //
a*0 028
C
a{0
*
,/
0.27
co
(
Og 0.26
0.25
0.75
i
i
0.85
0.95
1.05
1.15
1.25
Lambda
Figure4.2
Mappingof the gross indicatedfuelconversionefficiency.
The gross indicated fuel conversion efficiency is linear inside two distinct regions: the
fuel rich region (<1.0),
and the lean region (
>1.0).
Therefore the efficiency was
mapped linearly as a function of X for these two conditions. By using the lambda meter
output and the Gimep measured, the fuel conversion efficiency can be calculated for each
cycle.
27
Another important aspect of the fuel conversion efficiency is its dependence on
the ignition timing. The timing must remain at MBT to maintain the fuel conversion
efficiency optimum. Figure 4.3 shows varying fuel conversion efficiencies for different
ignition timings. Although the MBT timing does not change by more than four degrees
between the two steady state conditions (at Wide Open Throttle and at part throttle), lean
and rich burning cycles during the transient significantly
change the flame speed,
changing the MBT timing, in turn changing the fuel conversion efficiency.
However, it
is possible to measure the amount of retard or advance based on the peak pressure angle.
Therefore the peak pressure angle (pp) was included to compensate for the change of
MBT timing during the experiment.
Clx+ C2xGimep +C3
2 X C4)
(I +(1pp-OMBT)
0.3 45
0 0.34
0S
= 0.335
o
, 0.325
0
0.32
2 0.315
0
0.31
5
10
15
20
25
30
35
40
45
Ignition Timing BTDC (degrees)
Figure 4.3
Grossindicatedfuel conversionefficiencyversus ignition timing.
28
Finally, by grouping these three engine maps together, the mass of fuel burned
during each cycle can be calculated from the following measurements:
-Gross Indicated Mean Effective Pressure
-Relative air fuel ratio (X)
-Peak Pressure Angle
The second fuel flow measurement necessary for the analysis is the amount of fuel
injected. This volume can be directly measured from the pulse width (T1,T2) driving the
injector.
The pulse width was recorded with a digital oscilloscope during each
experiment. Using the injector calibration curve the volume of fuel injected during each
cycle can be calculated.
4.3 Data Analysis
4.3.1 Methodology
From the data recorded it is possible to calculate the mass of fuel burned and the mass of
fuel injected during each cycle. With these measurements it is possible to calculate "the
mass of fuel unaccounted for", the difference between the mass of fuel injected and the
mass of fuel burned during each cycle. By summing up the mass of fuel unaccounted for
over each cycle, we obtain the cumulative mass of fuel unaccounted for (C,,.t). By
assuming that all of the fuel injected is either burned inside the combustion chamber or
stored inside the port in the form of a liquid film, we can interpret the cumulative mass
of fuel unaccounted as a measure of the change in liquid film mass in the port.
1i=n
Cmf = X[Minjecti - Mburned] (eq.4)
i=l
29
0.45
0.04
0.4
0.035
:
0.35
0.03
0.3
i,
025
S
0
a 0.025
0
0.15 V
I
i
0.1
0.02
0.05
0.015
0
10
9
t
8
a
7
6
5
33
31
-29
27
' 25
lo
CD23
23
21
19
17
15
0
50
100
150
Cycle Number
Figure 4.4
The first three steps in the Analysis for MO.
30
200
250
Figure 4.4 shows the different steps of the analysis for pure indolene.
The first graph
shows the fuel flow and the air flow to the engine obtained from the calibration showed
on figure 4. 1. The second graph shows the gross indicated mean effective pressure for
each cycle, obtained from equation . The third graph compares the fuel mass burned
and the fuel mass injected during each cycle, obtained from equation 2. Finally using
equation 3 the "cumulative mass of fuel unaccounted for" is obtained, shown in figure
4.5.
160
E
, 140
a10
0
o 100
' t8o
0
3
60
6
40
E
20
00
0
50
100
150
200
250
Cycle Number
Figure 4.5
The cumulative mass offuel unaccounted forfor MO.
4.3.2 Accuracy of the mass of fuel unaccounted for
Section 4.3 described the procedure
used in the analysis of the data .
It is
important to explain some of the difficulties encountered during the processing of the
data.
Using the engine map described in section 4.2, the analysis calculated the fuel
mass and the air mass inside the cylinder during each cycle. Problems arose when
31
redundant parameters were calculated and measured. For example, at the end of the up
transient (just before closing the throttle), the engine is assumed to be at steady state
since the lambda meter reading was stable. Therefore, by applying a conservation of
mass inside the port, the mass of fuel burned per cycle should be equal to the mass of
fuel injected per cycle.
However, due to errors generated by the inaccuracy of the
instruments, as well as errors in the engine map, the mass of fuel burned per cycle was
within +1% of the measured mass of fuel injected per cycle. When calculating the
cumulative mass of fuel unaccounted for, the error was cumulated over about 150 cycles,
amplifying the error. Figure 4.6 and 4.7 illustrate this example. Sample 3 plotted on
figure 4.7 corresponds to a negative one percent error, whereas sample 2 corresponds to a
I~~~~~~
positive one percent error. In order to avoid such an error, an algorithm was developed.
I33
+1%
-
-f
;
31
v
*
I
~~~~,
..;. I.....
;*
"I
q -
^
j;
'"
5
/,
..
,- ,
I
......
.. 'Wsm'
'. , .. .,,
' - . /N ;S 'Jl
'
r
....... ............--
,
I
i
'
29
II
E 27 25-
,.
,.
I
A*I .
;1'
i: I
o
o 23
t4
11
1
a
Me I
.
'. " -A
i
*
'I
f
~ ~~~
.
j-;
i
ml
19
17
15
0
I
I
50
100
I
150
200
250
Cycle Number
Figure 4.6
Adjustmentof the mass offuel injectedat wide open throttle.
By assuming that the engine is at steady state before closing the throttle, the mass
of fuel burned must be constant over the last few cycles at wide open throttle. Therefore,
the slope of the cumulative mass of fuel burned must be equal to the mass of fuel
injected at wide open throttle.
32
2.00E-01
cm 1.50E-01
0o
1.00E-021
0a
E00
E -5O.OOE+00
-5.00E-02
cycle number
sample 2.
Figure 4.7
......
Sample
----
sample3
Effect of measurementerror on the cumulativemass offuel unaccounted
for.
Figure 4.8 shows the straight line and the slope of the line before closing the throttle.
Since the calculated mass of fuel burned was within
1% of the measured fuel mass
injected, the fuel mass injected at wide open throttle was calculated from the slope of the
line for each experiment.
33
-
7
6
5
a
4
0
a
3
E
2
0
-1
cycle number
Figure4.8
Obtainingthe mass offuel injectedat wide open throttlefrom the
cumulative mass offuel burned.
4.4 Results
From the data it is possible to calculate the cumulative mass of fuel unaccounted
for during each cycle. Although it is not precisely known as to where does this fuel goes
( e.g. via the cylinder wall to the oil sump), a substantial amount of this fuel represents
the change in mass of the liquid film stored inside the intake port. Appendix A contains
the detailed data for each experiment (the air flow rate, fuel flow rate, Gimep, and
exhaust relative air-fuel ratio). Appendix B contains the graphs of the cumulative fuel
mass unaccounted for.
The cumulative mass of fuel unaccounted for is plotted with a
best fit curve using two time constants, one for the up transient and another for the down
transient.
For the up transient:
Cmfj =
a x 1- e-
34
For the down transient: Cnf = a x e '2
Constant a is a measure of the change in fuel film mass between the wide open throttle
condition and the part throttle condition. This constant was calculated for every
experiment.
V.OU
0.55
0.50
' 0.45
E
E 0.40
' 0.35
c 0.30
M 0.25
0.20
0.15
0.10
0
10
20
30
40
50
60
70
80
90
100
% Methanol by Volume
Figure 4.9
'Changein fuel film mass for the low temperature experiment.
U.35
0.30
0,
C
E
0.25
c
fa 0.20
o
0
0.10
0.10
0.05
0
10
20
30
40
50
60
70
80
90
% Methanol by Volume
Figure 4.10
Change in fuel film mass for the high temperature experiment.
35
100
Figure 4.9, and figure 4.10 show the value of constant a for the cold and the hot coolant
temperature experiments. For reference, the fuel mass injected per cycle at WOT at X
=1.0 for MO is 0.031 gram. Figure 4.11, 4.12, 4.13 and 4.14 show the value of time
-
constants 'r and Az for the cold and the hot coolant temperature experiments.
2.1
1.9
=
-
1.7
:
1.5
1.5
mO
m10
*
*3 1.3
m25
-0--
c
0
o, 1.1
E
m40
* rm60
0.9
&--
m85
*
m100
0.7
0.5
0.8
0.85
0.9
0.95
1
lambdatarget
Figure4.11
Timeconstant t ,for the cold coolanttemperatureexperiment.
1.5
mO
-*-
1.3
-I----
a
m10
1.1
m25
e-
-0 m40
0.9
0
o0
e
F
0.7
0.5
0.3
.1
0.8
-
I
0.85
!
t..'i
0.9
0.95
*,,
m60
----
m85
*
m100
1
LambdaTarget
Figure 4.12
Time Constant
for the hot coolanttemperatureexperiment.
36
,,
2.1
--
1.9
me
-0---
ml 10
1.7
cI
E 1.5
a
-&---m.
0' 1.3
E 1.1
100
F
-
0.9
- ml
me
mi
0.7
0.5
0.8
0.85
0.9
0.95
1
lambdatargt
Figure 4.13
Time Constantl 2Jbr the cold coolant temperatureexperiment.
1.7
1.5
l 1.3
c 1.1
C
o 0.9
0
E
F 0.7
0.5
0.3
0.8
0.85
0.9
0.95
1
Lambda Target
Figure 4.14
Time constant 2for the hot coolant temperatureexperiment.
37
4.5 Fuel effects on driveability
In section 1.1, we mentioned that one of the major motivations for starting this
research was to understand the effects of fuel composition on driveability ratings such as
hesitation and stumble.
Using the data from the throttle transient experiment, it is
possible to quantify the transient response time of each fuel under the different test
conditions.
These response times may be interpreted as a measure of stumble and
hesitation.
A standard method already used in the testing of fuels is the measurement of the
response time of the torque curve of the engine (in our case the Gimep curve)(' 0 ). Figure
4.15 shows a sample Gimep curve to illustrate the calculation. The response time is
defined here as the rise time between 0% and 90% of the final Gimep value at wide open
throttle.
10
9.5
9
8.5
2
8
E 7.5
7
6.5
6
5.5
0
50
100
150
Cycle Number
Figure 4.15
Sample Gimep curve with the 0-90% response time.
38
200
250
The response time was calculated for each fuel under the two coolant temperatures.
The
response times are showed in figures 4.16 and 4.17 as a function of methanol content.
Figures 4.18 and 4.19 show the effect of lambda target on the response time.
1.8
1.6
8 1.4
- -,.9
0 12
- 0.8
-'
0.6
1,=.8
0 . 0.8
0.2
0
0
20
40
60
80
100
% Methanol by volume
Figure 4.16 Effect of methanolcontent on the responsetime under cold coolant
condition (T=40, C).
1.8
1.6
8
1.4
0 1.2
E
0
cc 1
a'
Q,
0.8
-.---
I
0.6
I
I
0.4
02
i
i
0
0
20
40
60
80
100
% Methanol by volume
Figure 4.17
Effect of methanolcontent on the responsetime under hot coolant
condition (T=80(C).
39
11.0
.
1.6
----
1.4
0---
8 1.2
MO
M10
M25
aO
-0--
0.8
M40
M60
S 0.6
M85
0.4
M100
02
0
0.8
0.85
0.9
0.95
1
Lambda Target
Figure 4.18
(T=40C)
Effect of lambda target on the responsetime under cold coolantcondition
1.8
1.6
1.4
E 1.2
2
1
- 0.8
0.6
0.6
6 0.4
0.2
0
0.8
0.85
0.9
0.95
1
Lambda target
Figure 4.19
(T=80,C).
Effect of lambdatarget on the responsetime underhot coolantcondition
Figure 4.16 shows that under a coolant temperature of 400C, the response time increases
sharply between M60 and M100. This sharp increase of about 225% is not as sharp for
the richer case. For
Xtarget=0.9,
the increase is about 190% ,and for Xtarget=0.8,the
40
increase is down to 170%. Figure 4.18 shows that although the enrichment decreases the
response time by about 50% for MO through M60, it is more noticeable for M85 and
M 100 where the response time decreases by 58% and 63%.
Under a coolant temperature of 800C , the response time decreases between M25
and M40 by about 40%. The response time remains constant for the other fuels from
M40 through M 100.
41
CHAPTER 5 -- MODEL AND DISCUSSION
5.1 Discussion
Results :in section 4.3 showed the absolute effect of methanol content on the
change in liquid film mass during the transient. The relative effect of the methanol
content on the change of film mass is illustrated in figure 5.1 and 5.2
In these two
figures, the change in film mass for each fuel was normalized with the change in film
mass for indolene under the same test conditions.
enrichment used for
These two graphs show that the
did not change the relative increase of
and Xtarget=0.8
ktarget=0.9,
liquid film mass between the methanol blends and the indolene fuel. It is also important
to note that in the cold and the hot coolant conditions, the relative change in liquid film
mass between methanol (M 100) and indolene (MO) was about the same, in the order of
220 %.
2.6
2.4
2.2
2
-o 1.8
8
1.6
' 1.4
12
1
0.8
0
10
20
30
40
50
60
70
80
90
100
% MUhnuol by Volume
Figure 5.1
The relative effect of methanol content on the change in liquidfilm mass
undercold coolant conditions.
42
_A
2.4
22
1.8
1.6
1.4
12
1
0
10
20
30
40
50
60
70
80
90
100
% Methanol by Volume
Figure 5.2
Thzerelative eject of methanol content on the change in liquidfilm mass
under hot coolant conditions.
There are many physical parameters that affect the transient response of a spark
ignition engine.. The mixture preparation process of the liquid film is dependent on the
latent heat of vaporization,
the fuel's distillation curve, the temperature of the intake
manifold, the intake pressure, and the fuel's stoichiometric fuel-air ratio requirement.
This last parameter has two effects on the mixture preparation process.
As the
stoichiometric requirement increases, the total energy required to vaporize the fuel
increases accordingly ("). The second effect is the possible change of the liquid film
pattern inside the port.
Since the injector is aimed directly at the back of the intake
valves, the wetted surface area remains the same from cycle to cycle. Under steady state
conditions, as the fuel evaporates, the film's surface area decreases, and then increases
back to its original size at the time of the next injection. If the fuel used has a high
stoichiometric requirement, then there will be twice as much fuel over the same surface
area right after the injection process. Therefore, the liquid film thickness is likely to
43
scale with the stoichiometric
requirement of the fuel used.
As the film thickness
increases. the mass transfer processes inside the manifold are changed.
Because the
liquid film flow to the combustion chamber is driven by surface shear forces of the
intake charge, the film flow to the combustion
chamber scales with the fuel film
thickness :th. Equations 5 and 6 relate the shear stress to the film flow rate (assuming
that the liquid film covers half of the port diameter):
x dia x th x p x 'Vt (eq.5)
(eq.5
IdM ]
T=
th
= fJ'(p,V,,,,
U,,, R,,) (eq.6)
Subscripts:
Notation:
dia:
manifold diameter
f:
fuel film
th:
Fuel film thickness
m:
air-fuel mixture
p:
density
V:
Surface velocity of the film
g:
Dynamic Viscosity
v:
Kinematic viscosity
U:
mean velocity
R:
Specific gas constant
By grouping equation 5 and 6 we can see that the fuel film flow rate to the combustion
chamber scales with the square of the fuel film thickness: dM
th
In a first attempt to explain the change of liquid film mass inside the intake port,
we looked at the: correlation between the change of liquid film mass and the mass of fuel
injected per cycle.
44
Assuming that the absolute liquid film mass inside the intake port scaled with the
mass of fuel injected during each cycle: Mfilmoc Cx Minij, the change in liquid film to the
change in mass of fuel injected:C
CCl x[Minj]l.Obar
- C2 x [MinJ]0. 7 barr -
For these transient experiments the correlation coefficient was poor when the
coefficients
C and C2 were both set equal to one.
Figure 5.3 and 5.4 show the
correlation under hot and cold coolant conditions. However, much better correlation was
Figure 5.5 show the correlation between the
found when C, was neglected (C 2=O).
change of liquid film mass and the mass of fuel injected at wide open throttle.
In this
case correlation coefficients: r=0.94 (@ T=800C), and r=0.91 (@ T=400C) were found.
0.05
I=1.0 r=93
X
0.045
0.045
1--=0.9
r=0.94
0.04
1I=0.8r=0.94
~
0035
,
O
r=0.83
0.03
0.025
_
*
1.0
0
0.02
0.015
O
o
o
X
=O.9
0.01
0.005
Il
0
O.00E+00
I
I
1.00E-01
2.OOE-01
3.00E-01
4.00E-01
I
5.00E-01
6.00E-01
Change in liquid film mam (g)
Figure 5.3
Correlation between the change of liquidfilm mass and the change offuel
mass injected(T=40'C).
45
0.04
1=1.0r=96
A
0.035
1=0.9 r=0.92
1=0.8r=0.90
0.03
0
.
r=0.85
2 0.025
S
E
0.02
0.015
S
Ua
U
1=l.O
A
a
a3
0.01
O
_c
.C
0.005
-=0.9
X=).8
,
A
[!v
O.OOE+00
5.00E-02
__.
.
.
.
1.00E-01
1.50E-01
2.00E-01
_
2.50E-01
_1
,
3.00E-01
3.50E-01
Change In lquid film man (g)
Correlationbetween the change of liquidfilm mass and the change offuel
Figure 5.4
mass injected( T=80'C).
0.60
I
0.50 S
T=40 C
O
T=80 C
E 0.40
E
T--40 C
a
o3!cmU0.30
rI
C
.S
U
*
-
020
T=80 C
r=0.94 @ T=80 C
r=0.91 a T=40 C
0
0.10
_
0.00
0
0.01
__
0.02
0.04
0.03
Li
Figure 5.5
0.05
0.06
0.07
_·
0.08
·
0.09
0.1
of fuel Injected at WOT (g)
Correlationbetween the change of liquidfilm mass and the mass offuel
injected at wide open throttle.
46
There are several important preliminary results from the experiments. First, the
data indicate a sharp increase in the change of film mass between M60 and M85 for the
cold condition. The second important result is the relatively high change of film mass for
pure methanol in the hot coolant condition.
Although the coolant temperature (T=800C)
is above the boiling point of methanol (T=64.7 OC@ 1 ATM), the data indicates that
there is a liquid film build up.
There is good correlation between the mass of fuel
injected at wide open throttle and the change of liquid film mass during the transient.
From these preliminary results we can conclude that the latent heat of
vaporization of methanol plays an important role in the evaporation process. In order to
obtain a better understanding of the physical processes inside the intake port we
attempted to develop a simple model which would work with Multi-component fuels.
5.2 Indolene Model
Because indolene contains more than twenty components, a model was devised to
simplify the analysis. This model was developed by Kuo-Chiang Chen at MIT(12 )using
SUPERTRAPP.
a program written by the National Institute of Standards and
Technology (NIST)('3 ) containing a library of thermophysical properties of hydrocarbon
mixtures. The NIST program uses the Peng-Robinson equations of state to calculate the
vapor-liquid equilibrium compositions. Although the program takes into account the
non-ideal interactions between the components, it does not take into account the effects
of the polar structure of methanol.
The goal of this model was to simplify the composition while keeping the same
distillation parameters as the standard indolene: T10, T50, T90 and the Reid Vapor
Pressure . T10, T50, and T90 refer to the temperature at which 10%, 50%, and 90% of
the liquid evaporates under atmospheric pressure. The values obtained are shown in
47
table 5.1. Table 5.2 shows the content of the indolene model used.
Table 5.1 Distillation Parameters for Indolene and the Indolene Model.
Indolene
Indolene Model
(C)
53.3
53.5
T50 (C)
105
104.1
T90 (C)
151.6
152.1
Reid Vapor Pressure (Psi)
8.74
8.5
Tl
Table 5.2 Content of the Indolene Model
Component Name
Volume %
N-Butane
6.1
Isopentane
18.6
2-Methylpentane
8.17
2-Methylhexane
9.3
Toluene
14.5
2,2,4-Trimethylpentane
24.76
Meta-Xylene
N-Nonane
6
2.6
Cumene
7
Decane
2.9
Using the indolene model, it was then possible to perform adiabatic and isothermal flash
48
calculations using SUPERTRAPP. However, it is important to point out that the NIST
program performs steady state calculations, which might not always apply due to the
dynamic processes inside the intake port. However, the steady state value will give an
indication of the trend that the dynamic process will follow.
5.3 Modeling the Intake Port
The most crucial parameter for mixture preparation inside the intake port is the
port wall temperature. Several papers have been published emphasizing the effects of
cold manifold
Therefore,
walls and cold valves on start-up and transient
it is important
to model
the heat transfer
behavior
between
chamber/cooling jacket and the port wall/intake valves accurately.
(14.15.16,17)
the combustion
With an instrumented
and heated intake valve, Martins J.J.G. et al. have obtained important differences in
valve temperatures between gasoline and methanol under the same heating power ®
'8). For
example, under the constant heating power (Q=100OW), the methanol fuel lowered the
valve temperature
from 1800C to 650C, compared to the gasoline case.
In these
experiments, the fuel injector was aimed directly at the back of the intake valve,
preventing fuel interaction with the port wall. The heat transfer model presented in this
section calculates the temperature and the fractional mass of fuel evaporated along the
port wall, between the injector and the intake valve. To simplify the analysis several
assumptions were made:
-The fuel injector creates a uniform fuel film over the intake port and the valve.
-The port wall temperature close to the combustion chamber is constant .
-The heat is conducted along the length of the port wall.
-All of the injected fuel is deposited on the port wall.
-The fuel film temperature is uniform across its thickness.
49
-The fuel film has uniform thickness along the port wall.
-There is no film flow.
Although the last assumption does not apply to an actual engine, the purpose here is to
correlate the change in liquid film mass with the evaporation model of the fuels. Figure
5.6 shows a schematic of the intake port to illustrate the heat transfer analysis on the port
wall.
T;
d
Figure 5.6
Schematic of the Intake Port wall.
The port wall was divided into 10 elements of equal surface area. Referring to figure
5.6, the first element represents the valve and the immediate region around the valve
seat.
This first element receives heat from the combustion
chamber wall (Tc) by
conduction. The wall temperature of each element is obtained by the energy balance
between the heat flow into the element by conduction (Qcondn), and the energy leaving
the element. The energy leaving the element is defined as the energy required to heat up
the fuel from the injection temperature Tinj to the wall temperature Tn (Qs) and vaporize
it at the wall temperature (Qhfg), the energy leaving the element by conduction to the
next element (Qcondn+l), and the energy transferred at the surface due to forced
convection.
In the time
for one engine cycle, the energy balance for each wall
element becomes:
[·[ 1-. 1] _[
d
]
d
x
Kx[-TxxAcx=
Ac x = Minj Cpx [Tn- Tinj]+ Mevn x hfg+ h x Asx[Tn - Tair]X'
N
50
Notation:
Ac:
heat conduction cross sectional area along the port
As:
surface area of each element exposed to the intake air flow
Cp:
specific heat of liquid fuel
d:
Length of one segment
h:
heat transfer coefficient (set equal to Watt/m 2 .K) between the intake air and the
fuel film
hfg:
latent heat of vaporization of the fuel
K:
thermal conductivity of the port material (aluminum)
Minj:
mass of fuel injected per cycle
Mevn: mass of fuel evaporated per cycle at element n
Tn:
wall temperature at element n
Tc:
combustion chamber wall temperature
Tr:
time for one cycle
There are two boundary conditions for the model.
temperature is constant (Tc=constant).
constant (T 1 1=constant).
First, the combustion chamber
Second, the end of the port wall is also held
For practical purposes, Tl1 was held constant at 3000 K.
Assuming that the evaporation of the film reaches equilibrium before the end of the
cycle, the NIST program can be used to relate the wall temperature
(Tn) and the
fractional mass of fuel evaporated per cycle for each element.
[Mev]n= XnX[Mfilm] eq.(5)
Xn:
mass fraction of fuel evaporated at element n.
[Mfilm]:
liquid fuel film mass at any element n. Since the surface area of each
element is the same, and the film thickness is uniform.
X is a function of the temperature and pressure at the wall:
X=F(Tn,Pn)
Although the total fuel film mass is unknown, it is reasonable to assume that the sum of
the fuel mass evaporated over all the elements is equal to the mass of fuel injected
51
(assuming that there is no film flow present).
Since the pressure inside the manifold is
known, the model simplifies to two equations (eq.(4) and eq.(5)) and two unknowns (Tc
and mfin). The solution to the model is obtained by iterating the fuel film mass until the
fuel mass evaporated from the film equals to the fuel mass injected.
The solution
provides us with three sets of information: the fuel film mass necessary to generate the
required mass of fuel vapor, the temperature profile along the port, and the fractional
mass of fuel evaporated along the port wall.
5.4 Modeling Results
The port wall temperature model presented in section 5.3 solved for the absolute
liquid film mass inside the port for a constant pressure and a constant combustion
chamber wall temperature. The relative change of liquid fuel film mass between the part
throttle and the wide open throttle case is calculated from the model by subtracting the
liquid fuel film mass at part throttle from the liquid fuel film mass at wide open throttle.
Figure 5.7 shows the effects of methanol content on the change of liquid fuel film mass
for different boundary conditions. Represented here are three combustion chamber wall
temperatures. For T=470 0 K, the correlation coefficient between the model and the
experimental results under high coolant temperature (T=800 C) is r=0.97. Under low
coolant temperature, (T=400 C), the correlation works best for a combustion chamber
wall temperature equal to 3900 K, r=0.95.
Figure 5.7 indicates that several transitions take place when the temperature of the
combustion chamber wall increases.
52
0.7
-*
0.6
1 0.5
T=390K
W 0.4
-0---
0.3
T=430K
T=470K
.s
cr 0.2
0.1
0.0
0
20
40
60
80
100
% Methanol by volume
Figure 5.7
Effects of MethanolContenton the change of liquidfuelfilm massfor the
model.
1.
The first transition occurs between MO and M60. The change in fuel film mass
uniformly increases as the temperature decreases.
2.
The second transition occurs between M60 and M100.
As the combustion
chamber wall decreases, there is a rapid increase in fuel film mass. The transition first
occurs between M85 and M100 (between T=4300 K and 4700 K), then the transition
mostly affects M85 between T=4300 K and T=3900 K).
The details of the analysis are shown in figures 5.8, 5.9, 5.10, and 5.11. Figures
5.8, and 5.9 show two samples of the temperature profile along the port under different
combustion chamber wall temperatures (T=3900 K, and T=4300 K).
53
380
370
360
350
a- 340
IE
Boiling Point @ Pi=l.Oar
330
Boiling Point @ Pi=.7bar
320
310
300
2
1
3
4
5
6
7
8
9
10
Element number (n)
Temperature profile along the port wall (Tc=390°K)
Figure 5.8
460
440
-
MO
420
---
M10
-*-
400
0---
X 380
.
M25
M40
M60
360
340
---
320
-
300
1
2
3
4
5
6
7
8
9
10
Element number (n)
Figure5.9
Temperature profile along the port wall (Tc=4700 K)
54
M85
M100
1
0.9
-
MO
0.8
----
a 0.7
S
M25
-*
0.6
M10
0.5
0
0 0.4
M40
M60
c 0.3
A---
M85
0.2
-
0.1
--
M100
-
0
1
2
3
4
5
6
7
8
9
10
Element number (n)
Figure 5.10
Massfraction offuel evaporatedalong the port wall (Tc=3900 K)
Figures 5. 10 and 5.11 show the mass fraction of fuel evaporated along the port wall for
each element under the different temperature profiles.
1
0.9
-
MO
0.8
-E-l--
c 0.7
(
0.6
5
0.5
M10
*
M25
M40
a 0.4
co
c 0.3
-A
M60
----
M85
0.2
-
0.1
M100
0
1
2
3
4
5
6
7
8
9
10
Element number (n)
Figure5.1
Mass fraction offuel evaporatedalongthe port wall (Tc=470°K)
55
For methanol (M 100), figures 5.8 and 5.10 show that for low temperatures, 6 out of the
10 elements are below the boiling point of methanol at wide open throttle. However,
only 4 out of the 10 elements are below the boiling point of methanol at part throttle
(Pi=0.7 bar). The model indicates that this is the major reason for the sudden change in
liquid fuel film mass observed in the data under low coolant temperature.
Because a
significant portion of the liquid film area is below the boiling point of methanol at wide
open throttle, a much larger fuel film mass is necessary to generate the necessary fuel
vapor.
Figure 5.11 shows that M 10, M25, and M40 have a better volatility along the port
wall than indolene under the high combustion chamber wall temperature (T=4700 K).
However, the increase in volatility is offset by the increase in the stoichiometric air-fuel
ratio requirements of the methanol blends over indolene.
56
CHAPTER 6 -- CONCLUSION
6.1 Review of the Methodology
A method was developed to measure the effects of fuel composition on engine
throttle transients. A parameter used to measure the effect of fuel composition was the
change of the cumulative difference between the amount of fuel injected and the mass of
fuel burned in the cylinder (obtained from the Gimep) This difference is interpreted as
the change in liquid fuel mass inside the intake port (fuel film) between the initial (at
Pi=0.7 bar) and final state of the transient (at Pi=l.0 bar). The method uses a detailed
engine map of the gross indicated fuel conversion efficiency to calculate the mass of fuel
burned on a cycle by cycle basis.
The engine map includes parameters such as the Gross
Indicated Mean Effective Pressure (Gimep), the Peak cylinder pressure angle, and the
exhaust relative air to fuel ratio.
The mass of fuel injected was calculated from the
injector pulse width and the mass of fuel burned under steady state conditions.
The
change of liquid film mass over one cycle was calculated by subtracting the mass of fuel
burned from the mass of fuel injected during the cycle. By summing the change of liquid
film mass over the duration of the transient, we obtained the "cumulative mass of fuel
unaccounted
for" or, the change of liquid film mass between the two steady state
conditions.
The second parameter used to evaluate the effects of fuel composition on engine
throttle transients was the Gimep response time. The response time was defined as the 0
to 90% rise time of the Gimep curve.
57
6.2 Model and Discussion
The data indicates that there is good correlation between the mass of fuel injected
at wide open throttle and the change of liquid film mass during the transient. This
correlation holds true under hot and cold coolant temperatures. However, the correlation
coefficient is not as good for the cold case. The higher scattering obtained under cold
coolant conditions can be explained with the one dimensional model developed.
The model included a 10 component fuel model of indolene to simplify the
calculations.
The model explained the trends showed on figure 4.9 and 4.10.
The following main conclusions can be drawn from the data and the model presented
here:
1. Under warmed up engine conditions (T=800C) the correlations indicate that the
change of liquid film mass scales directly with the fuel mass injected at wide open
throttle. Although we anticipated the liquid film mass to scale with the change of fuel
mass injected per cycle, it is not yet clear as to why the fuel mass injected at part
throttle does not enter the equation.
2. Under warmed up and cold engine conditions there is no correlation between the
change in fuel film mass inside the intake port and the 0-90% Gimep response time.
3. Under warmed up engine condition, the 0-90% Gimep response times were fastest
and showed no change between M40 and M100.
4. Under low engine coolant temperature, the change of liquid film mass exhibits a
rapid transition between M60-M85 and M100. This transition is in the order of
220%.
5. Using the one dimensional model, it appears that the transition under low coolant
temperature is due to the intake port wall cooling below the boiling point of methanol
at wide open throttle. Under low intake pressure, the liquid film is above its boiling
58
point, when the intake manifold pressure increases (at WOT), the liquid film is below
its boiling point, causing a large increase of the response time, and a large increase in
the change of liquid film mass.
59
REFERENCES
I.
Gallopoulos N.E., "Bridging the Present to the Future in Personal Transportation
- The Role of Internal Combustion Engines", SAE paper 920721, 1992.
2.
Heywood, J.B., Internal Combustion Engine Fundamentals, McGraw-Hill Book
Company, New York, 1988.
3.
Benninger N.F., Plapp G., "Requirements and Performance of Engine
Management Systems under Transient Conditions", SAE paper 910083, 1991.
4.
Brown, C.N. and Ladommatros,N.,
"A Numerical Study of Fuel Evaporation and
Transportation in the Intake Manifold of a Port-Injected Spark-Ignition Engine",
Journal of Automobile Engineering, vol 205, 1991.
5.
Yoshikawa Y.,Nakada T., Itoh T., Takagi Y., Kawasaki T., and Kawabe R.,
"Numerical Simulation System for Analyzing Fuel Film Flow in Gasoline
Engine", SAE paper 930326, 1993.
6.
Nagaishi H., Miwa H., Kawamura Y., Saitoh M., "An Analysis of Wall Flow and
Behavior of Fuel in Induction Systems of Gasoline Engines", SAE paper 890837,
1989.
7.
Benyettou F.M., Thelliez M., "Modeling of Unsteady Multiphase Flow in the
Intake Manifold of Spark Ignition Engines", SAE paper 910392, 1991.
8.
DeWitte, K., "Fuel Effects on the Starting Behavior of Engines", M.S. Thesis,
Department of Mechanical Engineering, Massachusetts Institute of Technology,
1993.
9.
Almkvist G., and Eriksson S., "An Analysis of Air to Fuel Response in a Multi
Point Fuel Injected Engine under Transient Conditions", SAE paper 932753,
1993
10.
Ogawa T., Araga T.,Okada M., Kato M., and Nakada M., "Analysis of Poor
60
Engine Response Caused by MTBE-Blended Gasoline from the Standpoint of
Fuel Evaporation", SAE paper 920800, 1992.
11.
Harrison A.J., "Enthalpy Requirement -- A New Fundamentally-Based Gasoline
Volatility Parameter for Predicting Cold-Weather Driveability", SAE paper
881670, 1989.
12
Kuo-Chiang Chen, "Fuel Volatility Modeling", M.Sc. Thesis, Department of
Mechanical Engineering, Massachusetts Institute of Technology, 1994.
13
Fly J.J., and Huber M.L., "NIST Thermophysical Properties of Hydrocarbon
Mixtures Database (Supertrapp)", National Institute of Standards and Technology
Standard Reference Database 4 -- Version 1.0, Gaithersburg, MD 20899, (July
1992).
14
Servati, B.H. and Herman, E.H., "Spray/Wall Interactions Simulations", SAE
paper 890566, 1989.
15
Iwano H., Jaitoh M., Sawamoto K., and Nagaishi H., "An Analysis of Induction
Port Fuel Behavior", SAE paper 912348, 1991.
16
Aquino C., and Plensdorf W.D., "An Evaluation of Local Heating as a Means of
Fuel Evaporation for Gasoline Engines", SAE paper 860246, 1986.
17
Aquino C., "The Effects of Local Heating on A/F Ratio Control", SAE paper
820411, 1982.
18
Martins, J.J.G. and Finlay I.C., "Fuel Preparation in Port-Injected Engines", SAE
paper 920518, 1992.
61
APPENDIX A
62
i
Cumulative Mass of Fuel Unaccounted For
0.6
'
' '
'
I '
'
'
'
I '
'
'
'I
'
'
'
0.5
Co
' '
'
I '
'
0.4
U
O
0.3
(u
II
0.2
0.1
0
)
I
0.6
0.5
C)
II
0.4
-C
0.3
Cu
cm
en
<W
0.2
100
~50
',
150
I '
'
.
'
-
'
200
I '
-
,zf
/
I
' '
\~~~~~~~~~~~~~~
'
I '
250
'
'
'
I '
\~~~~~~~~~~~~~~~~~~~
I
O
I~~~~~~~~~~~~~~~~~~~~~
0.1
0
I
a
)
50
I
I
I
a
a
I
,
~
,
100
a
I
a a aj
150
I
200
a
a
I
250
0.6
-
0.5
i0
II
cm
(n
Cu
0.4
0.3
H
0.2
0.1
0
0
50
100
150
Cycle Number
63
200
250
Cumulative Mass of Fuel Unaccounted For
0.6
0
03
()
co
(n
2
I, I
0.5
_
0.4
_
0.3
0.2
I
I
I ,I
I
I,
I,
,
,
Q
0
ooo00
i
II
E-
oC
_
0.1
r)1
II, , I , I
I
w-I
I
I , . I
I
I
I
0
m ~
-
I
50
.
-
I
I
I
100
. .
Ih.-
150
. I
200
250
0.6
0.5
0.4
0.3
ooo
0.2
o
0
o)
II
U)
cn
0.1
I
0
,a
0
0.6
I
,,
50
''
I 'I ' '
,,, I,
100
'
'
150
i
'
I
200
I
I
250
I
0.5
0 0.4
o
0
C)
0.3
II
C,
co
H
0.2
0.1
_
0
0
IA
50
~
~
~~~~~~~~~~~~~~~~~~I
I
X
,,
I,
150
200
250
100
Cycle Number
64
Cumulative Mass of Fuel Unaccounted For
0.6
0
0.5
0
0.4
4r
C
0.3
0
en
0.2
0.1
0
D
I
50
X
100
I
I
0
I
50
I
100
150
i
I
150
,
0.6
i
0.5
i
,,,,
I
~ ~ ~~~
1
,,,
~
_
1
~
200
,
|
~
250
200
|
|
|
1
250~~~~~~~~~~~~
1
1~~~~~~~~
~
o
(3
0.4
C)
-
n
en
en
0.3
0.2
~~~~~~~~~~~~~~~~~~~~~~
!
0i
,
l
50
Ia I
>
8-
'
/
'I
100 '
II
'
'
'
150
I
'
I
'
'u
'
200
1-
1,,,,
'
'
'
250
I
1
1 2
0.1
*,
0
I
I,
,
I
I
I
2
0.6
0.5
0.4
II
[-
0.3
(0
cu
<a
0.2
0.1
0
0
50
100
150
Cycle Number
65
200
250
Cumulative Mass of Fuel Unaccounted For
0.6
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
0.5
0
0.4
Q
0
oC
I-
0.3
CU)
Ca
0.2
oII
_
- -
-
0.1
-/
/
.
.
.
.I.
I
.
.
.
...
. . .
I
n
v
.
.
.
.
.
.
50
0
.
.
.
.
100
.
.
.
150
.
. ..
.
.
.
.
.
200
.
I
.
.
I.
250
0.6
0.5
0)
,,
03
0.4
o
0.3
oC
00
II
C
0.2
,/iI,I,,,, I ,,,,
I
0.1
.,,, I,
0
50
0
0.6
I II
I
150
100
I
I
200
250
I
I
I- ~~~~~~
-1-III ~ ~ ~ ~ ~7'~~~~1~~~~~~~]
A-r I II F
-
I
I~~~
I
0.5
0.4
U
0
C)
0
10
00
0.3
cn
2r
I
0.2
I
I
f!f
<
o
I
I
I
''-'
0.1
IIII
I
0
0
50
100
150
Cycle Number
66
,
200
-
250
Cumulative Mass of Fuel Unaccounted For
0.6
0
0.5
-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
m
w
0.4
0.3
w
m
U
0
t
C
n
co
2
H1
0.2
0.1
nI
0
I
/
.
I
I
I
I
I
I
I
I
I
I
. I
I
I
I
v
50
100
150
200
250
0.6
0.5
II
1i
U
0.4
o
-
C
0.3
U)
n
C
0.2
0.1
,
i
0
0
50
100
150
200
25,
50
100
150
200
250
'T i I
0.6
'I I I
'
'
'I
0.5
0.4
0
0
0.3
C
U)
U)
co
ll
0.2
II
if
V)
0.1
0
0
50
150
100
Cycle Number
67
200
250
Cumulative Mass of Fuel Unaccounted For
0.6
o
-
0.5
. . .I .I .
. I
.I. .I .
I
.. .' .' .I
I
.' .' '. .I
'
. .. I . .I
.
.
'
m
0
0
0.4 _
0
C)
(I)
Cn
II
0.3
0.2
_
_
0.1
0
-
I
/1111
C
)
0.6
I
..I....I
1
50
100
' I
-i ' '
'
'
'
'
I
150
'I '
'
0.5
0)
I-I
i
,
'
200
I I '
' I
250
,' ,
I r
o
o0
00
II
0.4
0) 0.3
or)
(I
Mr
0.2
/
0.1
, ,
0
~~~.
,
. . .. . . .. .
,
I
50
0
I
I
I,
100
150
.,
I,
,
,
200
250
EcI
0.6
0.5
00
II
Co
0.4
00
0.3
U)
(n
c
0.2
0.1
I
II
I
I
0
. .I . ..1 . .I .. I. . ..
I
, ll
0
I
50
I
1
I
150
Cycle Number
68
I
1
I
t1,,
j
i
100
I
I
I
,,
200
I
I
1,
250
Cumulative Mass of Fuel Unaccounted For
0.6
o
0.5
0.4
0
0
0.3
U)
U)
VO
0.2
0.1
0
0
50
100
150
200
250
U
a
0.6
0.5
0)
)
co
In
Cd
0.4
Q
o
0.3
[-
't
0.2
0.1
0
0
50
100
150
200
250
U
0.6
0.5
co
II
-
0.4
0
II
0
[-
0.3
C)
0.2
0.1
0
0
50
100
150
Cycle Number
69
200
250
Cumulative Mass of Fuel Unaccounted For
i
0.6
0
II I I
''
' I 'I
0.5
0.4
co
0.2
0.1
)
0
,,
, II
0
n
U.u
I
I
100
.......................
- -
I
I
I
50
I I
I
I
I
, ,
,
150
I
U
i, i
0.3
(n
C'
I
I
I
I
I
I
I
o
I,
200
I
o
250
I
I
I
I I
-
CI
0.5
0)
Ii
C)~
C')
(Ti
co
2
0.4
_
0.3
M
o
II
Co
0.2
0.1
/
r%
U
' 'I I ' ' '
I
100
50
0
0.6
~T ~I
I
I
I
I
I
I
I
I 1
200
150
I
I
I
:
I
I I
I
250
I
I
I~1
I
-
0.5
II
0.4
0
0.3
0)
(n
II
(-,
0)
0.2
0.1
I.
r
0
0
50
100
150
Cycle Number
70
200
250
I- I
Cumulative Mass of Fuel Unaccounted For
0.6
o
. '
'
'
I '
'
'
I '
'
' '
'
I "
'
'
'
0.5
I
'
0.4
0.3
U
CD
u,
0d
au
2
0.2
Y
0.1
0
50
0
0.6
, ,I ,
100
I
I
I
I
I
fi |
150
200
I
I
250
I
I
I
0.5
CD
0)
0)
U)
U)
0
0.4
\0
0.3
2O
0
0.2
0.1
0
0
n
u.vr
50
-
i
I
0.5
-
I
-
I
-
-
I
100
-
.
I
.
I
150
I
I
200
250
IIII1LIII
-
I
-
-
-
I
.
/
0.4
0
II
i
................
0.3
ca
0.2
7
0.1
I
U
0
I
I
50
I
I -1
I
I
I
I
I
100
I
I
I
150
Cycle Number
71
I
I
200
I
I
250
I -
Cumulative Mass of Fuel Unaccounted For
0.6
o
1-
_
,
,
,
I
,
,
,
,
I,
,
,
I,
,
,
,
,
I
I
_
-~~~~~~
0.5
0.4
o
o0
00
oo
II
0.3
cn
U)
(
0.2
0.1
0
50
0
100
150
200
250
0.6
0.5
U)
0.4
H
0.3
U)
Cn
2
0.2
0.1
, , /
0
, J0 ,
, ,
,,
,i ,
,
I
00
o0II
[-
.00
250
0 , , , , 50
I , , , , 100
i , , , I 150
I , , , , 200
I I,
0.6
-,
I , , I,,,, I
0.5
oo0
II
0.4
C-
U)
(n
CZ
0.3
0.2
0.1
)
0
0
, I,
I,
50
100
I,
I
150
Cycle Number
72
, , 250
i , 1
I, I,, I,
I '
200
0
0
II
0:
'
I '
250
Cumulative Mass of Fuel Unaccounted For
0.6
0.5
so
C)
0.4
Q
o
o
0.3
(I,,
II
0.2
11
0.1
0
0
50
100
150
200
250
0.6
0.5
a)
iI
0.4
U
0)
0.3
so
(n
()
CZ
Eo
0
0.2
0.1
0
0
50
100
150
200
250
II
0.6
0.5
co
ifI
0.4
CD
U)
0.3
[..
or00
Cn
cn
U)
CZ,
2n
0.2
0.1
0
0
50
100
150
Cycle Number
73
200
250
Cumulative Mass of Fuel Unaccounted For
0.6 _
0o
,
I
,,
I
, ,
,
I
,,,
,
I
, ,
I
,I
0.5
0.4
0
0.3
00
II
o
z
03
0.2
0.1
0
0
0.6
)
W)
00
~50 ~~..... 100
.- ..
X, 50
° I
, I,, 100
, I , , I , ' '
'
I X '
150
. ...
, 150
'
'
200
,,5 200
I '
'
' '
I '
250
25
I,
250
'
'
'
I ' :
0.5
a)
cnII
()
Cd
2
m~~~~~~~~~~~~~
0
0.4
00
0.3
0.2
0.1
0
-'I'' I|''' ' ' I'I'I'I'''
' ' I':
0
50
0.6
0.5
100
150
200
250
co
II
0)
II
n
(n
(:
C)
0.4
oIr
00oo
0.3
E
b'o
0.2
0.1
,
0
0
,
50
I
I
I
I,,,,
150
100
Cycle Number
74
I
200
250
Cumulative Mass of Fuel Unaccounted For
0.6
0.5
C
Q
0.4
CD
CD
Zn
,oZ
2,
CD
"It
0.3
C
C)
0.2
0.1
0
0
50
100
150
200
250
0.6
0.5
U
0.4
cr)i
0)
II
Cn
0.3
CZ
0.2
C)
C)
I
0.1
0
0
50
100
150
200
250
0.6
0.5
co
e Ii
Q
0.4
C)
H
C
J) 0.3
(n
en
co
C
0.2
0.1
0
0
50
100
150
Cycle Number
75
200
250
I I
Cumulative Mass of Fuel Unaccounted For
0.6 -' - ' ' ' I '
0
0.5
m
' '
II
'
I '
' '
'
'I
'
'
I
I
U
o
0.4
I)
oo
0.3
o0
00
II
_
(n
H
vc
__
0
0)
_
0.2
0.1
- I JI
1
A
U
. ..
0
I
i i
, I I I I I I
1
';
. . . . I . . . . . . . . . . . .-
50
100
150
- - -Ii - - -
200
250
0.6
0.5
C-II
U
o
0.4
000
_
(-
0.3
Eo
oC
II
U)
CZ
u,
0.2
0.1
0
0
50
100
150
200
250
o
0.6
0.5
ICI
0.4
c,
U)
0.3
cn
Ca
0.2
CD
o
o
II
H
0
0l
0.1
0
0
50
100
150
Cycle Number
76
200
250
APPENDIX
B
77
INDOLENE =1.0 @ 40°C
0.08
'-
7D
0.6
0.062
)
0.4
oa)C.)
0.04
a,
> 0.2
0.02 U)°
0r
U)
0
0
50
0
100
150
250
200
12
12
10
10
a-0
8
8
6
6
LL
:E
-
4
I
I
I
I
-,
1.6
I
I
I
I
I
-- r,
I
I
I
I I
I
I
100
50
0
I
I
I
I
I
I
I I
I
I
I
I
I
I
I
- A
-
200
150
I
I
I
I
I
250
I
I
I
I_
1.6
0
1.4
1.4
LL
1.2
1.2
<
1.0
1.0
X
0.8 ID
03
-
l-
0
0.8
-I
I
I
I
50
I
I
I
I
I
I
I
I
I
I
150
100
cycle number
78
I
I
I
I
I
200
I
I
I
I
250
INDOLENE X=0.9 @ 40°C
0.08
L
a,
0a
(D
0
a,
0.6
0
0.06
0.4
0.04
3
0.02 °
CO
C)
Cn
0.2
U)
M
0n
co
2t
0
0
LLI
._
w
50
100
150
200
250
12
12
10
10
8
8
6
6
4
0
50
100
150
200
4
250
1.6
1.U
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
co
EC
a,
a:
a)
IL
co
a)
.>
a)
:
a,
OC
0
50
100
150
cycle number
79
200
250
INDOLENE X=0.8 @ 40°C
cm
0.4
0
()
U)
a,
7>
0.06
0
0.04
-'
0.6
a,
C.)
0.08
0.2
0.02 °n
0o
0
0
50
100
150
200
250
12
12
10
10
co_
U0
w
8
8
6
6
4
4
0
50
100
150
200
250
1.6
1.0
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
o
Cr
a,
rr,
0
50
100
150
cycle number
80
200
250
=1.0 @ 80°C
INDOLENE
-T' '
'
'
I '
'
'
I' 'I '
'
'
I
'
'
'
' I'
'
0.08
'I '
06
0) 0.6
C)
a)
O
0.4
a(
--
1
-
o.
I
0.04
,,_
o
0
*-
0.2
0.02 CD
°
~~I
,I
0
0
I
IJ
I
I
I
50
I
I
I
I
I
o
I
I
I
I
I
150
100
I
I
i
I
I
200
I,
I
I -
250
12
12
10
10
8
8
6
6
.-
c:E
4
4
0
50
150
100
200
250
1 -
1.6
1.t
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
or
C:
L
a,
a)
a)
-J V
0
50
150
100
cycle number
81
200
250
INDOLENE k=0.9 @ 80°C
0.08
0.6
O
0.06
-a)
0.04
-
C,
o 0.4
.O
:3
(n
en
0.2
0.02 °
cn
CO
O
0
0
0
CL
Q
w
50
100
150
200
250
12
12
10
10
8
8
6
6
2
4
A
0
50
100
150
200
250
1.6
1.6
1.4
1.4
4
1.0
1.2
1.0
1.0
0.8
0.8
o
1-
-
0
a:
.>
Oa
Er
0
50
100
150
cycle number
82
200
250
INDOLENE X=0.8 @ 80°C
0.08
L.
Q)
0.6
a)
0.4
0
.-
0.06
(C)
8
0.04
-
o
'.-
(n 0.2
0.02 °
Cr
n
2
0
50
0
100
150
250
200
12
12
10
10
8
8
6
6
Q.
CL
w
4
250
4
50
0
Tl-r II
1.6
0o
a:
LL
I
100
I
I
I
I
I
150
I
I
I
i
200
I
I
I
IfI
I
I
-
1.6
1.4
1.4
1.2
1.2
_
I_
1.0
1.0
0.8
0.8
a)
.>
a)
f:
0
50
150
100
cycle number
83
200
250
(U
ru)
(T
M10
=1.0 @ 40°C
0.08
.)
0.6
CD
O
0
a>
0.4
0.06 ~C)
0.04-
o)
C
0.2
(d
(1)
0
0
0
50
100
150
200
250
12
12
10
10
U-
C"
.0
w
Q
8
8
6
6
L11
A
4
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
Or
-
1.0
1.0
0
0.8
0.8
o
(a
0
50
100
150
cycle number
84
200
250
M10 X=0.9 @ 40°C
0.08
.-
cm
0.6
0.06
a)
:ia)
a)
O
0.4
0.04
r)
O)
0n
(n
0.2
0.02 °
cn
C,
0
0
50
100
150
200
n
v
250
12
12
10
10
0
w
8
8
6
6
2
4
0
50
100
150
200
4
250
1.6
o
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
a
LL.
a)
a:
0
50
100
150
cycle number
85
200
250
0
M10 ;=0.8 @ 40°C
n
-
06
O
R
0.6
0.06
0
O 0.4
0.04
:o
n
0.2
0.02 °
co
co
0
0
0
50
L0
150
In
10
10
14
_II 1I I 1-I II -I
I
I 1
I -I
I
I
Il
I
I
I 1
-I _~~~~
8
8
6
6
4
0
-,
250
200
12
C'
w
100
50
150
100
200
4
250
1.6
1.6
(z 1.4
1.4
LL
1.2
1.2
<
1.0
1.0
-6 0.8
0.8
0o
I.
0
zC
0
50
100
150
cycle number
86
200
250
M10 =1.0 @ 80°C
0.08
7>
^ 0.6
0.06 a
a)
.
0.4
C"
) 0.2
0.04
-
0.02
'
Ct
()
n
250
0
50
0
100
150
200
12
12
10
10
L-
LL
8
8
6
6
2
4
A
50
0
1.6
_ i
I
II , -I
-
100
lr X-
150
I
-
Il
I
I
250
200
I
F
I
I
I
I
I
.1
1.1
I
0o
Cr
1.4
1.4
1.2
1.2
1.0
1.0
-6
L:
U-
0
a)
.iL
CZ
a)
I
._
a:
(.,
0.8
~
0
0.8
I
I
I
50
III
I
I
100
I
I
150
cycle number
87
I
I
I
I
I
200
,
I
I
I :
250
M10 =0.9 @ 80°C
0.08
0)
0.6
0.06 -~
a)
.)
0
a)
0.4
0.04
o
0n
rn
a>
0.2
0.02
'
()
U)
0
0
50
0
12
A ,I
I
100
II
I
I
150
I
I
I
250
200
I
I
-1
12
-I
10
10
0
.0
w
8
8
6
6
2
4
_
0
I
--I
II
I
50
I
I ,, I
I
I
I
I
100
I
I
150
I
I
I I
I
200
I
I
I
4
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
.t
o
a)
r_
a)
a:
irr
0
50
100
150
cycle number
88
200
250
M10 =0.8 @ 80°C
0.08
·-
c)
>
0.6
0.06 -a
a)
C.)
O
0 0.4
0.04
CZ
o
i
0.2
0.02 °
co
CD
n
0
250
200
150
100
50
0
12
12
10
10
8
8
6
6
n
CZl
.0
2
4
50
0
1.6
-'
' '
'
I '
CaO
U-
1.4
||
100
'
'
II
4
250
200
150
4e
'
'i
'
'
'I '
I
--
I.0
1.4
a)
3
a)
cal
w
0
1.2
1.2
1.0
1.0
0.8
a:
13E
0.8
o
-
0
50
150
100
cycle number
89
200
250
2
M25
=1.0 @ 40°C
0.08
o
0.6
)_
0
:>,
a)
0.4
0.04
'
a)
or
CO
(n
ca
0.2
0.02 °
(n
0
0
0
I,
50
100
150
200
250
12
12
10
10
8
8
6
6
w
4
A
'"'11'
(0
50
100
150
200
250
1.6
I.b
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
CE
LL
0
a)
C)
ar
0
50
100
150
cycle number
90
200
250
00
M25
=0.9 @ 40°C
0 08
-
0.6
C)
0.06
a)
0
O
a)
O
0 0.4
0.04
a)
0
M
=3
-4'I-
) 0.2
0.02 (E°
CZ
U)
CO
0
0
50
0
12
_
-I
I
I
I
100
I
I
I
I
I
150
I
I
I
I
250
200
I
I
I
I
I
I
I
12
I
10
10
Q.
w
8
8
6
6
I
A-
7
4
0
1.6
100
50
-
I
I
I
I
I
o
I
I
I
I
I
I
I
I
I
I
4
250
200
150
I
I
I
I
I
I
I
I
I
1.6
CZ
CE
1.4
1.4
LL
1.2
1.2
1.0
1.0
z
0
a)
Ca
a)
.M
-- I
0.8
0
IIIII
50
I
150
100
cycle number
91
_- 1
II11
200
0.8
250
M25 ;=0.8 @ 40°C
0.6
,
0.06 D
a)
a)
0
0
0.4
0.04
0
(d 0.2
-.
0.02 o0C,
(/
C,
0
0
0
50
100
150
200
250
12
12
10
10
Us
8
8
6
6
4
0
50
100
150
200
4
250
1.6
O
0
C
.
zaC
LL.
Cl
.,
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
a)
Cc
a)
0
50
100
150
cycle number
92
200
250
M25 2t=1.0 @ 80°C
0.08
7>
0.6
0.06 B
L-
0a)
a)
0.4
0.04
o)
O
n
o)
Ct
0.2
(0
O
0
0
0
50
100
150
200
250
12
12
10
10
8
8
6
6
L-
-0
w
:E
4
A
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
CE
3:
LL
0
o
CD
a)
a:
0
50
100
150
cycle number
93
200
250
M25 k=0.9 @ 80°C
0.08
Lc-
0.6
>
0.06 c
O
au
o
CD
a)
0.4
0.04
o
a)
.-
0
o~
0.2
0.02 (,°
(0
n
250
0
50
0
100
150
I
r
200
I,
12
-
I'
I
I
'I
I
I
I
II
1I
-
10
10
I
0-
8
8
6
6
LU
- I
4
0
.
I
I
50
I
I
I
I
I
I
100
I
I
I
150
I I I
I
II
200
I
-
A
-r
250
1.6
41
1.4
1.4
LLu
1.2
1.2
<
1.0
1.0
0.8
0.8
-
1.0
0
._
3c
0
50
100
150
cycle number
94
200
250
M25 ;=0.8
@ 80°C
0.08
0.6
-'
0.06
C)
0.4
-
0.04.
0.04
5
(n 0.2
2
a,
0
250
0
50
0
12
100
150
I
I
I
200
I
-
Io-mQ
10
CZ
-
8
12
10
8
2
6
6
I
4
250
4
50
0
1.6
-I
I
I
I
I
I
I
I
I
200
150
100
'
I
I
I
I
I
I
I
I
I
I
I -
1.0
0
(G
1.4
1.4 I
an
LL
0
o
.-
<
C
U)
C"
"
0
a)
0
1.2
'
I
1.0
08
-I , ,
0
I , ,,,
50
I ,
,
,
100
1.2
1.0
0.8
,' I
150
cycle number
95
I I I I I
200
,
,
250
M40 k=1 .0 @ 40°C
0.08
^- 0.6
0
O>
a)
s
C)
a)
0.4
0.04
0,_
-
0.2
Or(
o3
0.2
0.02
0
(/)
CE
n
250
0
50
0
12
'T I
I
100
I
I
I
I
150
I
I
I
200
I
I
I I '
I
I
-
12
10
10
8
8
6
6
aLu
,
A
0
I
I
I
50
I
I
I
I
I I
I
100
I I
150
I
I I
I
200
I
I
I
4
250
1.6
1.6
CZ
1.4
1.4
LL
1.2
1.2
ai)
o
a)
1.0
1.0
0.8
0.8
C-
0
C:
'5
0
.
_
.
a)
0
50
100
150
cycle number
96
200
250
M40 k=0.9 @ 40°C
o)08
>
0
0.6
0.06
a)
C)
0
a)
a,
0.4
0.04
L.
C)
z
0)
.O
4-
co
0.2
0.02 (n°
n
(I,
0>
0
0
50
100
150
200
250
12
12
10
10
,_
.0-
w
8
8
6
6
ll
4
4
0
50
100
150
200
250
1.6
J.
C)
1.4
1.4
LL.
1.2
1.2
1.0
1.0
0.8
0.8
Co
a:
0
a)
co
a,
a)
0
50
100
150
cycle number
97
200
250
ca
M40 =0.8 @ 40°C
0.08
O
c, 0.6
0.06
O
a)
D
a)
a)
0 0.4
0.04 -:
(0
Cn
0.2
0.02 °
TC
:2
en
(/
0
0
0
50
100
150
200
250
12
12
10
10
0.Q
8
8
6
6
4
0
a
co
CE
50
100
150
200
4
250
1.6
1.b
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
IL
cO
0
50
100
150
cycle number
98
200
250
M40
=1.0 @ 80°C
0.08
7>
0
0.6
.
a)
>
0
0a
0.4
0.04
a)
oen
(n
o3
~5
0.2
0.02 °
(n
0
0
0
50
100
150
200
250
12
12
10
10
Q
LL
8
8
6
6
4
0
50
100
150
200
4
250
1.6
1.6
n- 1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0o
r.>
o3
a)
nC
C:
0
50
100
150
cycle number
99
200
250
M40 L=0.9 @ 80°C
0.08
o
0.6
-
a
0.06
O
;
a)
a)
o 0.4
0.04
._
a)
03
o
n
or
0.2
0.02
°
(n
CO
n'
0
50
0
12
_
I
I
100
I
I
I
I
150
I
I
I
I
I
250
200
I
I
I
I
I
I
I
-
10
10
n
w
12
8
8
6
6
-I
A
r
I
I
I
I
I
I
'-
I
I
' I
I
I
I
I
I
100
50
0
1.6
I
I
I
I
I
I
I
I
I
I
II
I
I
I
200
150
I
I
I
I
I
I
I I
I
4
250
-
1.6
0
U-
0
C)
CE
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
50
100
150
cycle number
100
200
250
E
M40 ,=0.8 @ 80°C
0.08
-c 0.6
0
6
a)
0>1
a)
C.)
.o
0.4 -
0
Or
r.
0.04
6a)
-
-W-0.2
0.02 C/)
°
cn
C)
0
50
0
0
200
150
100
250
I
12
I
I
I
I
I
I
I
I
1
I I I
I
I
II
I
,'
1'
-
10
10
co
CL
8
8
6
6
LU
:E
I
A
r
I
I
I
I I
I
1.6
-_-
I
I
I
I
I
I
100
50
0
I
I
I
'I'
'I
I
I
I
I
I
I
I
II
4
250
200
150
I
I
I
I
I
-
1.6
0U.4,
1.4
o
a
1.2
m
_
1.4
1.2
1.0
1.0
0.8
'I
Cr
0
I
I
I
50
I
I
UL.IYLIYLILY
II
I~LI~ILLI~uW~~.
I
I
I
du., Ll.-ur
I-'-Y---'
I
100
I
I
150
cycle number
101
0.8
I
I
I I
200
I
I
I
-
250
M60 =1.0 @ 40°C
0.08
a,
a,
O
0.6
0.06 D
Q)
a>
a)
0
.> 0.4
0.04
Io 0.2
-q
a,
0.02 (n
°
(I)
CT
:E
0
0o
0
50
100
150
250
200
12
12
I
I
I
I
,I
I
I
I
I
I
I
I
I
10
I
I
IL
10
-o
LL
8
8
6
6
4
4
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
:Z 0.8
0.8
0
n'-
0o
a,
0
50
100
150
cycle number
102
200
250
M60 X=0.9 @ 40°C
' 'I
I
'rT' I
I '
'
'
I' '
'
'
I' '
'
'
' _ 0.08
0.6
-
0.06
o
o
(d
CZ
0.04
I
0.2
0.02 °cn
c)
I.
v
0
I
I
I
I
50
I
I
I
I
II
I
I
100
I I
150
I I
200
I
I
I
0
250
12
12
10
10
.0
-0
8
8
6
6
4
0
0
50
100
150
200
4
250
64 -
1.6
1 .t
so
CZ
a: 1.4
1.4
z
=3
LL
0
1.2
1.2
1.0
1.0
0.8
0.8
a)
._
co
a:
Q)
n
w
0
-a
O
a)
0.4
0
I
0)
n:
0
50
150
100
cycle number
103
200
250
M60 ,=0.8 @ 40°C
0.08
·'
a,
->
0
0.6
0.06
(a,
0
_0
a,
p 0.4
0.04
o
n
4--
0.2
0.02 °
Ct
0n
u,
ca
:E
0
50
0
100
150
250
200
- - - - - - - -
-
12
L,
,,
1a
10
10
Cu
-Q
w
2I_
8
8
6
6
4
-
0
I
I
I I
50
I
I
I
I
I
I
I
I
150
100
I
I
I
I
I
200
I
I
I
I
-
A
It
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
.2
CZ 0.8
a)
0.8
C,-<
0o
CC
4-
0
._=
:
:
a)
a:
0
50
100
150
cycle number
104
200
250
M60
=1.0 @ 80°C
0.08
-'
C)
0.6
C)
0.06
aa)
0.04
i
0.4
C0
3
4o
(I
0.2
0.02 °
0n
as
ed
0
50
0
Ias
-Q
0L
w
100
150
250
200
12
12
10
10
8
8
6
6
4
A
50
0
1.6
-
I
I
I
I
I
0
o
CC
LL
o,
100
I
I
I
I
I
150
I
I
I
I
I
250
200
I
I
I
I
I
I
I
I
I -
1.6
1.4
1.4
1.2
1.2
1.0
1.0
. _
0.8
-
0
I
I
I
I
50
I
I
I
I
0.8
I I
I
100
I
I
150
cycle number
105
I
I
I
I
I
200
I
I i
i
250
M60 X=0.9 @ 80°C
0.08
-
a)
7>
0.6
()
0
0.06
a)
a
a)
C)
a)
0 0.4
0.04
-
0
0.2
0.02 °
o,
CZ
III II_
0
0
50
0
12
'
' '
m
100
'
'
10
'
'
150
''
'
250
200
'
'
'
'
'
12
'
10
._
Q
LPU
I
8
I
I
I
I
I
I
I
I
II
II
I
I
I
I
I
I
I
I
I -
8
2
6
6
4 to
0
50
100
150
200
4
250
1.6
1.6
1.4
1.4
LL
1.2
1.2
o
1.0
1.0
a)
0.8
0.8
0
a
Cr
0
50
100
150
cycle number
106
200
250
M60 =0.8 @ 80°C
0.08
-
a)
0.6
aD)
0.06
O
0
a,)
0
M
(n 0.2
0.02
03
0O
0
50
0
12
-
100
150
250
200
- I- - I -I -I I- I- - - I - I - I - I - I - II - I - I - -I -III - - - -
I
12
I
10
10
8
8
6
6
03
w
2
I
4
-
1.6
0
I
I
I
50
0
I
- I- - - I - - I- 150
100
.
I
200
II
I
4
250
_,,
, i,,, I I,I,,,,,, i, , ,-
1.4
_
/'
1 .t
1.4
a,
LL
o
1.2
1.2
m,
1.0
1.0
0.8
0.8
._
a0
CD
o
(n
0
50
150
100
cycle number
107
200
250
M85 =1.0 @ 40°C
0.08
-
0)
0.6
>
0.06
0
-,
O
()
0.4
c
0.2
0.02 C,°
CZ
(/
Cn
0
0
0
LU
50
100
150
200
250
12
12
10
10
8
8
6
6
2
4
4
0
0o
C:
LL
50
100
150
200
250
1.6
1.4
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
1)
0
50
100
150
cycle number
108
200
250
M85 k=0.9 @ 40°C
0.08
L,
0.6
O
o
0 0.4
Cr
0.06
0a)
0.04
-
0c
(n
0)
0.2
0.02 co
°
2
0
0
0
(LI
.0
I--,
0-
w
50
100
150
200
250
12
12
10
10
8
8
6
6
2
4
0
,?
0
50
100
150
200
4
250
1.6
J
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
co
CE
IL
0
o
c)
U-:
a0)
CD
CD
a)
z
0
50
100
150
cycle number
109
200
250
M85 .=0.8 @ 40°C
0.08
0
0.6
0.06-o
a)
(5
0C)
0.4
0.04
CZ
U)
0.2
0.02
CZ
0
O
0
50
100
150
200
250
12
12
10
10
-.
Q_
8
8
6
6
L
4
4
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
o
cc
o:
IL
0
0a
-C
0a
0
Cn
Q,
cc:
0
50
100
150
cycle number
110
200
250
0
(n
U)
=1.0 @ 80°C
M85
-;z
0.08
a)
0.06
a
0.04
-_
0.6
a)
0 0.4
'U
a
o
a) 0.2
0.02
2n
CD
(n
0
50
0
100
150
250
200
X A\
12
12
10
10
8
8
6
6
-Q
w
:O..
E
I..
-I
A
'"1"
0
I
I
I
I
50
I
I
I
I I
I
100
I
I
150
I
I
I
I
I
200
I
I
I
I -
4
250
1.6
1.0
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
o
.-
Ca
a)
o
'
,-
0
50
100
150
cycle number
111
200
250
M85 k=0.9 @ 80°C
-
I
I I
I
I
I
I
I
III
II
I
I
I
I
I
I
0.08
I
,- 0.6
o
>1
0.06
a)
o 0.4
:3
a)
0.04
,--
C
ao
0
.-
<,
.-
0.2
0.02 °
I
C
0n
I
I
-
0
I
I
-I - I - I -
.
-
'
'
'
'
I
I
100
50
0
12
I I I
I
I
I
I
I
I
I I
I
I
I
.
I
I
.
I
.
I
O
250
200
150
. .
I
I
. .
.
I
J=
1I
I
I
10
10
cU
-0
8
8
6
6
,_
I
4
I
I
, I
I
I
I
_1
- I
I
I
I
I,
100
50
0
1.6
I
I
I
I
I
I
I
I
I
I
I
I
I
150
200
I
I
I
I
I
I
I
,
4
250
I
I
I
I,
1.6
0
._
Ca
o
1.4
1.4
I~~~~~~~~~~~~~
1.2
1.2
1.0
1.0
U-
.-
(:
Ca
ao)
-
0.8
0.8
-
0
I
I
50
I
100
I
150
cycle number
112
I
200
I
I
I
250
M85 k=0.8 @ 80°C
0.08
' 0.6
,)
a)
a)
0
' 0.4
0.04
U--
o0
n
0.2
0
0
12
-I
- .
.
.
.
.
.
co
n
50
100
. .
I '
.
150
. . . .
I
I '
.
I
.
.
.
250
200
.
I I'
.
'
. .
' '
.
I
12
I I
10
8
8
6
6
LL
A
-,
I
I
I
I
I
I
I
I I
I
I
I
I I
'1
0
50
100
150
I
I
I
I
200
I
I
,'
4
250
1.6
1.6
o
1.4
1.4
,U-
1.2
1.2
1.0
1.0
0.8
0.8
0
C<
0
Ia
z
0.02 °
10
TZ
'-
CO
L
w~
'a0
CD
r:
0
50
100
150
cycle number
113
200
250
M100
k=1.0 @
40°C
0.08
a)
a,
0.6
0
0.06 ":
o0
0
(n
,n
CZ
a)
0.4
0.04
:3
0.2
0.02 c°
2
co
0
0
0
50
100
150
200
250
12
1
10
10
8
8
6
6
I"
oo
0-
w
Z>
4
A
"If
0
50
100
150
200
250
1.6
_11.0-
[r
0 1.4
1.4
O
._
a)
:3
LL
1.2
1.2
1.0
1.0
0.8
0.8
O
<t
U0
a)
a)cc
-
0
50
100
150
cycle number
114
200
250
M100 @=0.9 @ 40°C
0.08
-
0.6
C.)
0.06 aa)
a)
O
C.)
.
0.4
0.04
._
0
I,
C"
03
03
:3
0.2
0.02 Cn
°
cu,
o
CO
0
50
0
12
-I
I
I
100
I
I
I
I
150
11
I
I
200
25iO
12
I
I
I
I
I
I
I
I
I
I
10
10
8
8
6
6
a
LL
._
- I
A
I
I
I
I
I
I
I
I
I
I
I
I
I
'"l
0
50
100
150
I
I
I
I
I
200
I
I
I
4
250
1.6
.1
1.
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
cU-
0
an
LL
cc
0c-
-'
a)
0
50
100
150
cycle number
115
200
250
M100
A=0.8 @ 40°C
0.08
CD
0.6
0.06 Da)
0
a,
0.4
0.04
.4-
0(n 0.2
0.02 °
(n
Ca
on
(/
n
0
0
50
100
150
200
250
12
12
10
10
8
8
6
6
1
wD
A
4
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0
LL
o
a)
Ir:3co
0
50
100
150
cycle number
116
200
250
M100 =1.0 @ 80°C
0.08
-
0.6
0.06
U)
(>
0
'a)
O
-,
o
o
0.04
0.4
O.
n() 0.2
Ct
(d
)0.02
0
0
50
0
12
_ r= I
I
I
100
I
I
I
I
I
I(
150
I
I
I
I
I
250O
200
I
I
I
I
I
I
-I
I
I
-
10
12
10
c.
wL
8
8
6
6
A -
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
"1'
0
0
U)
CO
:>
50
100
150
200
I
I
I
I -
4
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
CE
0
50
100
150
cycle number
117
200
250
<d
M100
k=0.9 @
80°C
0.08
0.6
0.06-c
a)
a,
7>
0
a)
C
U)
0.4
0.04
o0
Un
0.2
-
0.02 C)°
n)
0
0
50
100
150
200
n
v
250
12
12
10
10
8
8
6
6
.U
Q0
EL
4
0
50
100
150
200
A
-r
250
1.6
1.6
CU
Cr
1.4
1.4
LL
1.2
1.2
1.0
1.0
0.8
0.8
0
0
r1
Co
C
0
50
100
150
cycle number
118
200
250
M100 .=0.8 @ 80°C
0.08
L -
0.6
0
a)
.5
O
C)
>3
0.4
0.04
:
._
co
0-cn
(n 0.2
0.02 °
CS
en
co
0
0
Q_
LL
50
100
150
200
250
12
12
10
10
8
8
6
6
2
4
4
0
50
100
150
200
250
1.6
1.6
1.4
1.4
1.2
1.2
1.0
1.0
a) 0.8
0.8
0
o
a)
CE
0
50
100
150
cycle number
119
200
250