Addsis Alem Thesis.

Abstract
In order to reduce emissions of carbon dioxide (CO2) to minimize global warming

and climate change, and pollutants emissions from vehicles for better air quality, it is
necessary to improve the efficiency and minimize pollutant emissions from internal
combustion engines by means of better engine technologies and low/zero carbon
fuels, as well as the electrification of the powertrain system in vehicles.
Renewable ethanol and methanol are the two alcohol fuels which can significantly
reduce CO2 emissions. In this project, a state-of-art single cylinder spark ignition (SI)
engine and testing facility were set up and used to study the combustion
characteristics, fuel efficiency as well as gaseous and particle emissions from the
engine operations with ethanol or methanol. The engine performance and the
combustion process of ethanol and methanol were measured and compared with
E10 RON95 gasoline.
The first part of the study was carried out at three engine speeds of 2000rpm,
3000rpm and 4000rpm under different load conditions with the same engine
parameter settings (cam timings, injection parameters) for gasoline, ethanol and
methanol to have a direct comparison between alcohol fuels and gasoline. Particular
emphasis was on the high load operations at and above 16bar IMEP when gasoline
engine operation was found to be prone to knocking combustion and over-fuelling
was introduced to keep the exhaust gas temperature below 780°C. In comparison,
engine could be operated at stoichiometric conditions without exceeding the exhaust
gas temperature limit due to their higher enthalpy of evaporation and the use of MBT
timings without knocking combustion for ethanol and methanol. The result shows
that methanol leads to 3.6% higher brake thermal efficiency than gasoline, and
ethanol by 3.3% when operated at higher load operations at around 18bar IMEP.
Both ethanol and methanol fuels lead to substantial reductions in the emission of
particles, with the particle numbers reduced by up to 90%.
The second series of experiments were then carried out to find the best injection
strategies for ethanol and methanol. The result shows that both ethanol and
methanol have larger best efficiency island than gasoline because of their greater
i
evaporation and charge cooling effect. In addition, there’s also a reduction in
emission across the fuel matrix testing.
Finally, a low speed & load spark timing sweep was carried out to assess the
performance of the engine fuelled by ethanol and methanol under catalyst light off
condition. The result shows that both ethanol and methanol could operate with more
retarded spark timings than gasoline to enable by faster catalyst-light-off. In addition,
particulate emissions released by alcohol fuels are much less and of smaller sizes.
The novelty of the work is enabling detailed systematic analysis of alcohol fuelled
combustion, fuel consumption, performance and emissions under advanced, highly
boosted SI engine conditions. The load sweep part fills in the gap of high load test
conditions of pure ethanol/methanol fuelled SI engine. The fuel matrix part presents
the effect of injection timing and injection pressure variation on fuel efficiency,
combustion characteristic and emissions which gives a deeper understanding of air-
fuel mixture of ethanol and methanol benefit from their greater evaporation and
charge cooling effect. The cold start spark timing sweep shows that both ethanol and
methanol could operate with more retarded spark timings than gasoline to enable by
faster catalyst-light-off.
ii
Acknowledgements
First of all, I would like to express my deep gratitude to my supervisor, Professor Hua
Zhao, for granting me the opportunity to study in the great engine research group at
Brunel University London, and also for his invaluable guidance on my research and
learning, and for all the help given to make my life easier during these years.
I would like to acknowledge MAHLE Powertrain UK for all the support provided to this
work. Particularly I am grateful to Mr Anthony Harrington, Mr Jonathan Hall, Mr Adrian
Cooper and Dr Mike Bassett for share of experience and knowledge and the warm help;
and to Mr Tony Cains and Mr Justin Mape for their excellent technical supports on EMS
system and engine rebuild.
I am thankful to my friends and colleagues in the group. Thanks to Dr Xinyan Wang for
lots of valuable help on both my study and life. Thanks to Dr Reza Golzari for his
teaching and company in engine testing. Thanks to Mr Andy Selway, Mr Eamon Wyse
for their expertise and assistance to support the research work.
Thanks to Mr Ziyan Fu and Mr Chris Nichols for their friendship and help.
I also would like to thank all other friends met here for their company, help and
encouragement. The life is much more joyful here thanks to all of you.
Finally, my most sincere gratitude goes to my family - my mother, my father and my

sister for their endless and selfless love, encourage and support, which are the most
important resources and powerful backing for me to pursue better life.
At the end of writing the thesis, particularly and most important, I must express my
gratitude to my wife Ms Jiawen Si - without your unconditional love and support and
your endeavour, it would be not possible for me to overcome the difficulties and
complete the study. Also thank my new born Son Zebang Lu - you did so well and made
so less trouble during my writing and your smile is the best gift spiriting me up.
iii
List of Abbreviations
ABDC After Bottom Dead Centre

AFR Air Fuel Ratio
APV Alternatively-Powered Vehicles
ATDC After Top Dead Centre
BBDC Before Bottom Dead Centre
BDC Bottom Dead Centre
BLD Borderline Detonation
BMEP Brake Mean Effective Pressure
BSFC Brake Specific Fuel Consumption
BTDC Before Top Dead Centre
BTDCF Before Top Dead Centre Firing
BTDCNF Before Top Dead Centre Non-firing
CA Crank Angle
Crank Angle at 10% Mass Fraction of
CA10
Mixture Burned
CA50
Mixture Burned
CA90
Mixture Burned
CAD Crank Angle Degrees
CAFE Corporate Average Fuel Economy
CAI Controlled Autoignition
CFD Computational Fluid Dynamics
CI Compression Ignition
COV Coefficient of Variation
CR Compression Ratio
DAQ Data Acquisition
DI Direct Injection
eBoost Electrical Boosting
ECU Electronic Control Unit
EGR Exhaust Gas Recirculation
iv
EMOP Exhaust Maximum Opening Point
EMS Engine Management System
EOI End of Injection
EPA Environmental Protection Agency
EVC Exhaust Valve Closing
FID Flame Ionization Detector
GDI Gasoline Direct Injection
Homogeneous Charge Compression
HCCI
Ignition
HEV Hybrid Electric Vehicle
ICE Internal Combustion Engine
Gross Indicated Fuel Conversion
IEffg
Efficiency
Net Indicated Fuel Conversion
IEffn
Efficiency
Pumping Indicated Fuel Conversion
IEffp
Efficiency
IMEP Indicated Mean Effective Pressure
IMOP Intake Maximum Opening Point
ISFC Indicated Specific Fuel Consumption
IVC Intake Valve Closing
IVO Intake Valve Opening
KI Knock Intensity
LIVC Late Intake Valve Closing
LIVO Late Intake Valve Open
Minimum spark advance for Best
MBT
Torque
MFB Mass Fraction burn
NA Naturally Aspirated
NI National Instruments
NIMEP Net Indicated Mean Effective Pressure
Net Indicated Specific Fuel
NISFC
Consumption
v
NOx Nitrogen Oxides
PFI Port Fuel Injection
PID Proportional Integral Derivative
PM Particulate Matter
PMEP Pumping Mean Effective Pressure
PN Particulate Number
PRT Platinum Resistance Thermometer
RON Research Octane Number
rpm revolutions per minute
SI Spark Ignition
SOI Start of Injection
TDC Top Dead Centre
ULG Unleaded Gasoline
VVA Variable Valve Actuation
VVT Variable Valve Timing
WOT Wide Open Throttle
Chemical Abbreviations
CO Carbon Monoxide
CO2 Carbon dioxide
H Hydrogen atom
N Nitrogen atom
N2 Nitrogen
NO Nitric oxide
NO2 Nitrogen dioxide
O Oxygen atom
O2 Oxygen
THC Total hydrocarbons
vi
List of Figures
Figure 2-1 Toxic Emissions Standards for Passenger Cars [8] .................................................... 6
Figure 2-2 Reduction of HC+ NOx on Passenger Cars from Euro 0 to Euro 4 [10] .................... 8
Figure 2-3 The Fraction of CO2 Emissions by Source [11] ......................................................... 9
Figure 2-4 Historical Fleet CO2 Emissions Performance and Current Standards (gCO2/km
normalized to NEDC) for Passenger Cars [12] ........................................................................... 9
Figure 2-5 WardsAuto Annual Survey: Technologies to Help Meet 2025 CAFE Standards [13]... 11
Figure 2-6 Fuel Energy Conversion Process and Efficiencies in Internal Combustion Engines
[22,26-31]................................................................................................................................ 15
Figure 2-7 BSFC Contour Mao of a Turbocharged Gasoline DI Engine .................................... 18
Figure 2-8 Gasoline direct injection with boosting [45] .......................................................... 24
Figure 2-9 The global production of ethanol from 2007 to 2015, by country [69] ................. 31
Figure 3-1 Schematic of Single Cylinder Engine Testbed ......................................................... 38
Figure 3-2 Brunel-MAHLE Single Cylinder DISI Engine on the testbed .................................... 39
Figure 3-3 Combustion chamber with the injector and the sparkplug [91] ............................ 41
Figure 3-4 Intake and exhaust cam profile and phasing .......................................................... 42
Figure 3-5 Coriolis mass flow meter by Endress+Hauser ........................................................ 47
Figure 3-6 Kistler 4005B type Piezoresistive Absolute Pressure Sensor ................................. 51
Figure 3-7 Comparison of the dynamometer torque curve and the engine torque curve ..... 52
Figure 3-8 MAHLE Flexible engine control unit ....................................................................... 53
Figure 3-9 High and low-speed DAQ cards .............................................................................. 56
Figure 3-10 Transient Combustion Analyser user interface [97] ............................................ 58
Figure 3-11 P-V diagrams of a four-stroke SI engine at full load and part load [98] ............... 59
Figure 3-12 Indicated work and fuel conversion efficiencies .................................................. 60
Figure 3-13 Band-Pass filtering calculation [99] ...................................................................... 61
Figure 3-14 Kistler TDC sensor toolkit...................................................................................... 63
Figure 3-15 TDC determination in the combustion analyser .................................................. 63
Figure 3-16 Log P-Log V diagram of engine motoring at 1200 RPM ....................................... 64
Figure 3-17 In-cylinder pressure pegging example.................................................................. 65
Figure 3-18 Horiba MEXA-554JE for CO, CO2, and O2 measurements ................................... 66
vii
Figure 3-19 Signal Rotork Analysis model 523 FID analyser used for HC measurements (first
unit at the top) and Signal Ambitech model 443 Chemiluminescent analysers used for
NO/NOx measurements (bottom 3 units) ............................................................................... 67
Figure 3-20 Cambustion DMS 500 fast response particulate analyzer ................................... 68
Figure 3-21 Measured error of Endress+Hauser Promass 83A01 flow meter ........................ 75
Figure 3-22 Peak cylinder pressure and its angle recorded for daily motoring checks (the x-
axis shows the test days) ......................................................................................................... 78
Figure 4-1 Combustion characteristics of 2000rpm load sweep ............................................. 80
Figure 4-2 Combustion phasing, Pmax and Rmax ................................................................... 81
Figure 4-3 Lambda, TExh and injection parameters ................................................................ 82
Figure 4-4 Emissions of 2000rpm load sweep ......................................................................... 83
Figure 4-5 Engine conditioning pressure and temperature .................................................... 85
Figure 4-6 Particle numbers of methanol and Ethanol............................................................ 86
Figure 4-7 2000rpm load sweep particle size spectral size density ........................................ 87
Figure 4-8 Combustion characteristic of 3000rpm load sweep............................................... 88
Figure 4-9 Combustion phasing, Pmax and Rmax ................................................................... 90
Figure 4-10 Lambda, TExh and injection parameters .............................................................. 92
Figure 4-11 Emissions of 3000rpm load sweep ....................................................................... 94
Figure 4-12 Engine conditioning temperature and pressure .................................................. 97
Figure 4-13 Particle numbers of gasoline, ethanol and methanol .......................................... 98
Figure 4-14 2-10bar IMEP Size spectral density ...................................................................... 99
Figure 4-15 10-20bar IMEP size spectral density................................................................... 100
Figure 4-16 20-28bar IMEP size spectral density ................................................................... 101
Figure 4-17 Combustion characteristic of 4000rpm load sweep........................................... 102
Figure 4-18 Combustion phasing, Pmax and Rmax ............................................................... 104
Figure 4-19 Lambda, TExh and injection parameters ............................................................ 106
Figure 4-20 Emissions of 4000rpm load sweep ..................................................................... 107
Figure 4-21 Engine conditioning temperature and pressure ................................................ 110
Figure 4-22 Combustion characteristics of 4000rpm load sweep ......................................... 111
Figure 4-23 Particulate number of gasolines, ethanol and methanol ................................... 112
Figure 4-24 Size spectral density of 4000rpm load sweep .................................................... 114
Figure 5-1 BSFC of different inject timing and rail pressure .................................................. 118
viii
Figure 5-2 Brake Thermal Efficiency of gasoline, ethanol and methanol .............................. 120
Figure 5-3 Spark Timing of gasoline, ethanol and methanol ................................................. 121
Figure 5-4 Mass fraction burn (MFB) and burn duration of gasoline, ethanol and methanol
................................................................................................................................................ 122
Figure 5-5 THC and NOx emission of fuel matrix ................................................................... 124
Figure 5-6 Particle numbers released by gasoline of different regulations .......................... 126
Figure 5-7 Particle numbers released by ethanol of different regulations ........................... 126
Figure 5-8 Particle numbers released by methanol of different regulations ........................ 126
Figure 5-9 Size spectral density of gasoline ethanol and methanol injection start at 275CA
BTDC ....................................................................................................................................... 128
BTDC ....................................................................................................................................... 129
BTDC ....................................................................................................................................... 130
BTDC ....................................................................................................................................... 131
Figure 5-13 BSFC_cor and CA50 of gasoline, ethanol and methanol .................................... 132
Figure 5-14 Indicate specific fuel consumption of gasoline, ethanol and methanol ............ 134
Figure 5-15 Burn duration of gasoline, ethanol and methanol ............................................. 135
Figure 5-16 Exhaust gas temperature and CO2 ..................................................................... 137
Figure 5-17 THC and NOx released by gasoline, ethanol and methanol ............................... 139
Figure 5-18 O2 and CO released by gasoline, ethanol and methanol ................................... 142
Figure 5-19 Throttle position and spark timing ..................................................................... 144
Figure 5-20 LNV and combustion stability of gasoline, ethanol and methanol .................... 145
Figure 5-21 Peak in-cylinder pressure and maximum pressure rise rate .............................. 146
Figure 5-22 Total particle number of gasolines, ethanol and methanol ............................... 147
Figure 6-1 Combustion characteristic of cold spark sweep ................................................... 150
Figure 6-2 NV, maximum in-cylinder pressure and maximum pressure rise rate ................. 152
Figure 6-3 Emissions released by gasoline, ethanol and methanol....................................... 153
Figure 6-4 Particle numbers data of gasoline, ethanol and methanol .................................. 155
Figure 6-5 Size spectral density of gasoline, ethanol and methanol ..................................... 156
ix
List of Tables
Table 2-1 EU Emission Standards for Passenger Cars (Category M1*) [9] ................................ 7
Table 2-2 Key points for improving Gasoline Engine Efficiency .............................................. 16
Table 2-3 Reasons for BSFC Increases in a Gasoline Engine Operation Map .......................... 18
Table 3-1 Engine Specification ................................................................................................. 40
Table 3-2 Intake and Exhaust Cam Specification and Phasing ................................................ 43
Table 3-3 The Main ECU inputs and outputs used during the experiments ........................... 53
Table 3-4 Rotork Analysis model 523 FID analyser (HC) specifications ................................... 67
Table 3-5 Signal Ambitech model 443 Chemiluminescent analyser (NOx) specifications ....... 68
Table 3-6 Specifications of the Emission analysers ................................................................. 68
Table 3-7 Raw gas molar mass fraction of the exhaust gases for gasoline ............................. 71
Table 3-8 Properties of the fuels used for engine experiment................................................ 72
Table 3-9 Engine operation parameters and boundary conditions......................................... 73
Table 3-10 Motoring and firing daily checks settings .............................................................. 77
Table 5-1 Fuel matrix test point ............................................................................................. 117
Table 6-1 Fuel matrix test point ............................................................................................. 149
x
Contents
Abstract ................................................................................................................................................. i
Acknowledgements .......................................................................................................................... iii
List of Abbreviations ........................................................................................................................ iv
List of Figures ................................................................................................................................... vii
List of Tables ....................................................................................................................................... x
1. Chapter One: Introduction ....................................................................................................... 1
1.1 Introduction............................................................................................................................ 1
1.2 Alcohol Fuels for IC Engines .............................................................................................. 2
1.3 Project Objectives ................................................................................................................ 3
1.4 Outline of Thesis................................................................................................................... 4
2. Chapter Two: Literature review .............................................................................................. 6
2.1 Introduction............................................................................................................................ 6
2.2 Main Component of Exhaust Gas ...................................................................................... 8
2.2.1 Carbon Dioxide (CO2) .................................................................................................. 8
2.2.1.1 Technologies to Reduce CO2 Released by Vehicle.......................................... 10
2.2.2 Carbon Monoxide (CO) ............................................................................................. 11
2.2.3 Unburned Hydrocarbon (THC) ................................................................................. 12
2.2.4 Nitrogen Oxides (NOx) ............................................................................................... 13
2.2.5 Particle Matter (PM) ................................................................................................... 14
2.3 Internal Combustion Engine Efficiency ........................................................................... 14
2.4 Technologies for Improving Gasoline Spark Ignition Engine Fuel Economy ............ 20
2.4.1 Gasoline Direct Injection ........................................................................................... 20
2.4.2 Stratified Lean Combustion ...................................................................................... 21
2.4.3 CAI and HCCI ............................................................................................................. 21
2.4.4 Gasoline DI Engines and Boosting (Downsizing) .................................................. 22
2.4.5 Variable Valve Actuation and Air Intake ................................................................. 24
2.4.6 Variable Compression Ratio ..................................................................................... 25
2.4.7 Water Injection ............................................................................................................ 26
2.4.8 Cooled Exhaust Gas Recirculation .......................................................................... 27
2.4.9 Miller Cycle .................................................................................................................. 27
2.5 Spark Ignition Engines with Alternative Fuels ................................................................ 28
2.5.1 Natural Gas Engine .................................................................................................... 28
2.5.2 Alcohol as Alternative Fuel in Spark Ignition Engine ................................................ 29
2.5.2.1 Overview of Alcohol Fuels .................................................................................... 29
2.5.2.2 Application of Alcohol Fuels in Spark Ignition Engines .................................... 33
3. Chapter Three: Experimental Facility and Methodology ............................................... 37
3.1 Introduction.......................................................................................................................... 37
3.2 Experimental Set-up .......................................................................................................... 37
3.2.1 The Single Cylinder Engine ...................................................................................... 39
3.2.2 Details of Cam Profiles and Valve Timing .............................................................. 41
3.2.3 Engine Oil System ...................................................................................................... 44
3.2.4 Coolant system ........................................................................................................... 45
3.2.5 Fuelling System .......................................................................................................... 46
3.2.6 Intake System ............................................................................................................. 48
3.2.7 Exhaust System .......................................................................................................... 50
3.2.8 Dynamometer ............................................................................................................. 51
3.3 Engine Control Unit and Management System ............................................................. 52
3.4 Data Acquisition (DAQ) System and Instrumentation................................................... 55
3.4.1 Data Acquisition Hardware ....................................................................................... 55
3.4.2 Data Acquisition Software (Combustion Analysis) ................................................ 57
3.4.3 Finding the Top Dead Centre ................................................................................... 62
3.4.4 In-cylinder Pressure Pegging ................................................................................... 64
3.5 Exhaust Emission Measurements ................................................................................... 65
3.5.1 Emission Analysers .................................................................................................... 65
3.5.2 Calculation of specific emission ............................................................................... 69
3.6 Testing and Data accuracy ............................................................................................... 73
3.6.1 Cam Timings Validation ............................................................................................ 74
3.6.2 In-Cylinder Pressure and Fuel Flow Measurements Validation .......................... 75
3.6.3 Daily Engine Check Points........................................................................................ 76
3.7 Summary ............................................................................................................................. 78
4. Chapter Four: Spark Ignition Engine Operations with Ethanol and Methanol ........ 79
4.1 Introduction.......................................................................................................................... 79
4.2 Results and Discussion ..................................................................................................... 79
4.2.1 2000rpm Load Sweep................................................................................................ 79
4.2.2 3000rpm Load Sweep................................................................................................ 88
4.2.3 4000rpm Load Sweep.............................................................................................. 101
4.3 Summary ........................................................................................................................... 114
5. Chapter Five: Effects of Injection Timing, Rail Pressure on Ethanol and Methanol
Engine’s Performance and Emissions ..................................................................................... 116
5.1 Introduction........................................................................................................................ 116
5.2 Experimental Setup and Test Conditions ..................................................................... 116
5.3 Results and Discussion ................................................................................................... 118
5.3.1 2000rpm 4bar IMEP Fuel Matrix Analysis ............................................................ 118
5.3.2 3000rpm 16bar IMEP Fuel Matrix Analysis .......................................................... 131
5.4 Summary ........................................................................................................................... 147
6. Chapter Six: Analysis of the Spark Retard Capability of Alcohol Fuels for Fast
Catalyst Light-off............................................................................................................................ 149
6.1 Introduction........................................................................................................................ 149
6.2 Experimental Set-up ........................................................................................................ 149
6.3 Result and discussion ...................................................................................................... 150
6.4 Summary ........................................................................................................................... 157
7. Chapter Seven: Conclusion ................................................................................................ 158
7.1 Introduction........................................................................................................................ 158
7.2 Conclusions ....................................................................................................................... 159
7.2.1 Engine Performance, Combustion and Emissions of Alcohol Fuels ................ 159
7.2.2 Effect of fuel injection timing and pressure........................................................... 160
7.2.3 Cold Start Spark Timing Sweep ............................................................................. 161
7.3 Recommendations for Future Work .............................................................................. 161
Reference ......................................................................................................................................... 163
1. Chapter One: Introduction
1.1 Introduction
In the modern world, the invention of the Internal Combustion Engine (ICE)
represents an important development, both in terms of mobility and power
generation. There is a long history of ICE engines which dates back to 1876 when
Otto developed the spark ignition (SI) engine which was followed by Diesel in 1892
with the demonstration of a single cylinder compression ignition (CI) engine. The IC
engines are used in a range of applications and they are the most popular powertrain
for land and water vehicles all over the world [1].
In the early years, fuel economy and power output were the chief motivations for the
development and optimization of these engines. To meet both the requirements of
power and fuel efficiency, technologies such as turbocharging, direct injection have
been developed.
In the last few decades, the rapid increase in the application of internal combustion
engines in vehicles has resulted in increased concerns regarding the environmental
and health effects of their exhaust emissions and their legislations. It was in the early
1960s that the United States began regulating air emissions from automobiles.
Beginning in California at first, then nationwide with the United States Environmental
Protection Agency (EPA), emission regulations for automobiles were developed and
enacted. Emissions legislations have then been adopted by other countries, such as
Europe, Japan, Australia and China.
As a result of the popularity of internal combustion engines in the automotive sector,

the demand for fossil fuels has also increased significantly. To achieve a substantial
reduction in fuel consumption, greater fuel efficiency must be achieved from those
engines. In particular, as the main greenhouse gas, legislations on fuel economy and
CO2 emissions have been introduced for light duty vehicles in the last 10 years.
These factors have led to intensive research into more environmentally friendly and
fuel saving engines.
1
1.2 Alcohol Fuels for IC Engines
In terms of decarbonizing fleet CO2 for internal combustion engines, renewable

alcohol fuels have been increasing used as neat fuel or a mixture with gasoline.
Series of research have already been conducted on ethanol and methanol blend
fuels, and the idea of blending greater amounts of alcohol into fossil gasoline is not
new [2,3]. The use of ethanol blended fuel is already prevalent around the world
today, and it contributes to reducing embedded CO2 with a content of up to 10% in
most gasoline supplies of these countries. The Brazilian government has taken a
step further and has begun making ethanol blends up to neat form available in its
publicly available fuel supplies [4]. The availability of methanol as a fuel source is
less prevalent. Nevertheless, it was historically used as a fuel for racing cars and
more recent research has examined its use in IC engine [3,5]. It would therefore be
necessary for the production of pure ethanol and pure methanol to be sourced from
renewable sources from 2nd generation feedstocks in order to justify the use of such
fuels.
According to previous research, both ethanol and methanol can increase the thermal
efficiency of a SI engine as well as reducing particle emissions. The efficiency
benefit is attributed to faster laminar flame speed and charge cooling effects of
alcohol fuels [3,6]. In addition, alcohol fuels have a higher knocking resistance than
gasoline fuels. Last but not the least, due to their lower boiling temperature and high
oxygen content, ethanol and methanol fuels will produce less particles.
In the last few years, electric vehicles are moving rapidly into the light passenger
vehicle market in the form of battery electric vehicles (BEV) and hybrid engine and
electric vehicle. BEV is well suited for the customers who can have easy access to
the electric charging facility and regions where electricity is produced from renewable
sources. Hybrid vehicles can overcome the shortcomings of the BEV by combined
use of an efficient engine preferably running on renewable or zero carbon fuels for
extended range and the zero exhaust pipe emissions of an electrical powertrain.
Furthermore, for countries and regions where the electricity is generated from fossil
2
fuels or limited electric supply infrastructure, and applications where the battery will
be considered to be too bulky, expensive, taking too long charging time or of limited
range, IC engines will remain the most viable powertrain system.
On 25 August 2022, California announced that the sale of new gasoline-powered

cars and light trucks will be banned by 2035. Standard hybrid vehicles which have no
plug and rely much more on gasoline than plug-in models are treated the same as
gasoline models due to the emissions[7].
However, the current legislation only considers the emissions emitted by vehicles at
the tailpipe without considering the emissions produced during production (e.g.
batteries) and energy sources: hydrogen and electricity, respectively. The research
done by Burton et al shows that It is unclear whether widespread adoption of BEVs
in the United States will reduce greenhouse gas emissions substantially over the
lifespan of the vehicle in the near term compared to FHEV and PHEV alternatives,
particularly when the required infrastructure upgrades are considered [8].
Therefore, the application and optimisation of renewable ethanol and methanol fuels
in IC engines provide an alternative solution to achieve low-carbon and zero-carbon
transport in the future.
1.3 Project Objectives
The aim of this study is to investigate the effect of methanol and ethanol on the
engine’s performance, combustion and emissions by carrying out an extensive and
systematic experiments on a single cylinder direct injection (DI) spark ignition engine
which has been updated to operate with pure ethanol and methanol. The engine
efficiency, as well as emissions data, will be analysed and compared with RON95 E10
gasoline.
The specific objectives of the study include quantification of:
3
(1.) The effect of methanol and ethanol on the in-cylinder heat release and
combustion process and the resulting impact on the SI engine’s
performance, efficiency and emissions at different engine speeds and load
conditions. In particular, how the high-load engine operations can be
improved by the use of methanol and ethanol fuels in terms of knocking
combustion and exhaust gas temperature control.
(2.) The effect of injection strategies of different fuels on the engine’s
performance, efficiency and emissions when gasoline, ethanol and
methanol fuels, by adjusting the injection timing and rail pressure at different
engine speeds and loads.
(3.) The capability and effectiveness of methanol and ethanol fuels to operate
with retarded spark ignition and combustion for fast catalyst light off.
1.4 Outline of Thesis
Chapter 2 provides the literature review which includes some basic knowledge
related to direct injection spark ignition engine, combustion efficiency and emissions.
The advanced engine technologies are introduced and reviewed and both their
advantage and disadvantage are discussed. the current and future emission
regulation are shown which includes CO, CO2, NOx, THC as well as particulate
emission.
In chapter 3, all the equipment on the engine testbed are described which include
the engine itself and its conditioning system. All temperature sensor and pressure
transducers used as well as all emission analysers are described, together with the
engine management system (EMS) and data acquisition system. The equations of
some important calculation of combustion parameters are explained such as
indicated mean effective pressure (IMEP), indicated power, net indicated specific
fuel consumption (NISFC). Finally, the procedures for improving the data accuracy
are explained.
4
Chapter 4 presents of the results of load sweep at 2000, 3000 and 4000rpm load
sweep between 2bar IMEP to 28 bar IMEP. The difference between alcohol fuel
(ethanol& methanol) and baseline (E10 gasoline) of combustion characteristics, fuel
efficiency as well as emissions are analysed.
Chapter 5 focus on the different injection timings and rail pressures on the engine
performance, efficiency, combustion characteristics, gaseous and particle emissions.
Two groups of fuel matrix test which are 2000rpm 4.6bar IMEP and 3000rpm 16bar
IMEP were used to investigate the difference between alcohol fuel and gasoline.
Chapter 6 shows the result of 2000rpm 2bar cold start sweep in which the 50% mass
fraction burn was retarded from 8 Deg ATDC to 45 Deg ATDC. This test is aimed to
find the most retarded spark timing and the maximum exhaust gas temperature
during the spark sweep for the catalyst light off.
The main conclusions of this experimental study on the difference between pure
alcohol fuel and E10 gasoline are summarised in Chapter 7.
5
2. Chapter Two: Literature review
2.1 Introduction
In the last 50 years, substantial progress has been made on internal combustion
engines in terms of their performance, fuel efficiency and emissions.
The history of emission legislation started in the 1960s when the air quality became
worse day by day, particularly noted in California US. Ever since, government
agencies were set up and legislation introduced to limit the tailpipe emission [9].
Figure 2.1 shows the toxic emissions standards in different countries and regions
[10]. The development of emission regulation is mainly about decreasing the toxic
component in the exhaust as well as making improvement of test cycles.
Figure 2-1 Toxic Emissions Standards for Passenger Cars [10]
The first stage of Europe emission regulation came out in July 1992 and it has been
upgraded to 6th stage in September 2014. All the listed pollutants have been
reduced substantially stage by stage which is shown in Table 2.1 [11]. It can also be
found that the limit on particle mass (PM) has been applied for gasoline engine from
Euro 5 and the limit of particle numbers (PN) has been added to Euro 6.
6
Table 2-1 EU Emission Standards for Passenger Cars (Category M1*) [11]
CO HC HC+NOx NOx PM PN
Stage Date
g/km #/km
Positive Ignition (Gasoline)
Euro 1† 1992.07 2.72 (3.16) - 0.97 (1.13) - - -
Euro 2 1996.01 2.2 - 0.5 - - -
Euro 3 2000.01 2.30 0.20 - 0.15 - -
Euro 4 2005.01 1.0 0.10 - 0.08 - -
Euro 5 2009.09b 1.0 0.10d - 0.06 0.005e,f -
d e,f
Euro 6 2014.09 1.0 0.10 - 0.06 0.005 6.0×1011 e,g
Compression Ignition (Diesel)
Euro 1† 1992.07 2.72 (3.16) - 0.97 (1.13) - 0.14 (0.18) -
Euro 2, IDI 1996.01 1.0 - 0.7 - 0.08 -
Euro 2, DI 1996.01a 1.0 - 0.9 - 0.10 -
Euro 3 2000.01 0.64 - 0.56 0.50 0.05 -
Euro 4 2005.01 0.50 - 0.30 0.25 0.025 -
b f
Euro 5a 2009.09 0.50 - 0.23 0.18 0.005 -
Euro 5b 2011.09c 0.50 - 0.23 0.18 0.005f 6.0×1011
Euro 6 2014.09 0.50 - 0.17 0.08 0.005f 6.0×1011
* At the Euro 1..4 stages, passenger vehicles > 2,500 kg were type approved as Category N1 vehicles
† Values in brackets are conformity of production (COP) limits
a. until 1999.09.30 (after that date DI engines must meet the IDI limits)
b. 2011.01 for all models
c. 2013.01 for all models
d. and NMHC = 0.068 g/km
e. applicable only to vehicles using DI engines
f. 0.0045 g/km using the PMP measurement procedure
g. 6.0×1012 1/km within first three years from Euro 6 effective dates
7
Figure 2.2 shows that the limits of tailpipe emissions has been reduced significantly
from Euro 0 to Euro 4 [12]. What’s more, since the regulation become more and
more stringent, new technologies such as electronic control, fuel injection
technology, as well as aftertreatment devices have been developed to limit the
tailpipe emissions.
Figure 2-2 Reduction of HC+ NOx on Passenger Cars from Euro 0 to Euro 4 [12]
2.2 Main Component of Exhaust Gas
2.2.1 Carbon Dioxide (CO2)
Recently, CO2 caught more attention than other toxic component in the tailpipe
emission since CO2 is the main greenhouse gas responsible for the increase of
global temperature. Figure 2.3 shows the CO2 release proportion by sources. The
CO2 released by cars running on fossil fuels takes the second position [13]. The most
effective way to reduce CO2 release by cars is to reduce the fuel consumption by
improving the engine efficiency and the use of low carbon and zero carbon fuels.
8
Figure 2-3 The Fraction of CO2 Emissions by Source [13]
Governments (China, United States, The Europe Union, Japan, India, Mexico,
Canada, Saudi Arabia, Brazil, South Korea) which are the core members among the
top 15 vehicle markets have set out mandatory CO2 emission standard for light duty
vehicles to reduce “greenhouse” gas emission of CO2 [14]. Figure 2.4 shows that the
standard of CO2 emissions becomes more and more stringent so that all the car
makers should apply new technology to reduce the fuel consumption also guarantee
the engine efficiency of cars.
Figure 2-4 Historical Fleet CO2 Emissions Performance and Current Standards (gCO2/km
normalized to NEDC) for Passenger Cars [14]
9
2.2.1.1 Technologies to Reduce CO2 Released by Vehicle
Due to the standard of fuel economy become more stringent, advanced engine
technologies and designs have been developed and applied to optimise the fuel
consumption of cars, including
• Improved combustion and engine control methods
• Improved transmission efficiency and powertrain integration
• Intelligent thermal management
• Brake/ kinetic energy recovery
• Exhaust/heat energy recovery
• Alternative fuels
• Propulsion electrification
• Improved aerodynamics and rolling resistance
• Light-weighting design and materials
Recently, an annual survey done by WardsAuto shows that the important things
carmaker focus on are light-weighting as well as optimising fuel consumption to meet
the CO2 standard of 2025, as shown in Figure 2.5 [15].
10
Figure 2-5 WardsAuto Annual Survey: Technologies to Help Meet 2025 CAFE Standards [15]
2.2.2 Carbon Monoxide (CO)
CO is a kind of toxic gas which has big effect on human’s health since it can
combine with haemoglobin in the blood to form CO2 so that the capacity of oxygen
carried by blood is reduced. Small quantities of CO may cause light headedness
while large quantities will lead to loss of consciousness and asphyxia [16].
The amount of CO exist in the tailpipe gas is mainly affected by the Air Fuel Ratio
(AFR) [17]. It was found that the production of CO increases rapidly when
equivalence ratio is greater than 0.95. What’s more, the production rate of CO is also
influenced by the inhomogeneity of the mixture as well as the burned gas
temperature and temperature gradients in the expansion and exhaust strokes [18].
The kinetic chemical formation process of CO is shown below, in which R is a

hydrocarbon radical [19].
11
𝑅𝐻 → 𝑅 → 𝑅𝐻2 → 𝑅𝐶𝐻𝑂 → 𝑅𝐶𝑂 → 𝐶𝑂
The CO is oxidised into CO2 in the burned gas in a slow speed and the reaction
process is shown below.
𝐶𝑂 + 𝑂𝐻 = 𝐶𝑂2 + 𝐻
It has also been found that the concentration of CO after the exhaust valve open is
lower than when in cylinder but it’s still higher than at equilibrium conditions [20].
2.2.3 Unburned Hydrocarbon (THC)
Unburned Hydrocarbons are also an important component in emission and they can
lead to worsening human health such as coughing, dry throat and wheezing. What’s
more, HC can react with NOx to produce O3 and photochemical smog which is quite
harmful to humans as well as the environment [16].
Hydrocarbon compounds emitted from vehicles are represented with the general
term of uHCs which include paraffins, olefins, acetylene and aromatics, etcall of
them can be found in exhaust gas. Different kinds of HC emissions can lead cause
the phenomenon call photochemical smog which is quite harmful to human as well
as environment.
There are so many influencing factors which have effects on the production of HC, of
which incomplete combustion is the main factor. What’s more, fuel impingement on
the cylinder surface in a DI gasoline engine also leads to the production of HC and it
is mainly affected by the injection strategies (injection timing and rail pressure) and
injector orientation relative to the cylinder axis and the piston crown [21]. It has been
found that flame quenching has big effect on the production of THC on the surface of
cylinder [22]. Moreover, the fuel which cannot be oxidised by the flame front
propagate may cause the production of THC and lubricating oil can also be oxidised
with the charge to form THC [23].
12
2.2.4 Nitrogen Oxides (NOx)
There are two specific emissions referred to NOx, which are nitrogen dioxide (NO2)
and nitric oxide (NO). Researchers have proved that the reaction between NO and
haemoglobin is harmful to humans. NO2 is a kind of irritant chemical which become
harmful when it reaches a high concentration [16].
An equilibrium analysis by Stone [24,25] shows that the content of NO is greater than
the content of NO2 in the cylinder of the gasoline engine. There are two ways in
which NO can be produced in the engine. The first one is because of the oxidation of
the nitrogen in the fuel, while the other and most common one is the oxidation of
atmospheric nitrogen [25,26]. Nitrogen is present in negligible quantities in gasoline.
The process to form NO from atmospheric nitrogen when the combustion is
stoichiometric is as followed.
𝑂 + 𝑁2 = 𝑁𝑂 + 𝑁
𝑁 + 𝑂2 = 𝑁𝑂 + 𝑂
𝑁 + 𝑂𝐻 = 𝐻 + 𝑁𝑂
The formation of NOx is mainly affected by the temperature of burned gas. With the
increase of the gas temperature, the concentration of NO2 also increases. What’s
more, since spark timing also has a big effect on the temperature of burned gas so
the production of NOx can be optimised by controlling of spark timing. The peak in-
cylinder pressure increases with the advanced of spark timing since more fuel is
burned before top dead centre and the peak pressure moves close to top dead
centre when the volume is smaller. An increase in peak in-cylinder pressure is
accompanied with an increase in peak burned gas temperature, and as a
consequence, there is a higher rate of NOx formation.
The burned gas fraction or exhaust gas recirculation (EGR) is another variable which
would affect NOx emission. Thanks to the presence of CO2 and water vapour in the
EGR, the combustion temperature can be significantly decreased with the increase
of burned gas mass fraction, because of their dilution and increased heat capacity of
13
the in-cylinder charge. It has been found that the production of NOx can be
decreased by more than 50% with the employment of 15-25% EGR [27].
2.2.5 Particle Matter (PM)
PM comprises carbon particles (soot) and condensates of liquid (sulphate, unburned

fuel and oil). PM has a bad effect on human health and the extent of its potential
danger mainly depends on the type (nucleation mode and accumulation mode) and
size of PM. The very fine particles whose size is very small (≤10nm) would
contaminate the lung as the lung itself cannot clear it.
Because the production of organic particles (soot) by diesel engine is far more than
the gasoline engine, diesel engine requires Diesel Particulate Filters (DPFs) to
reduce the quantity of PM to meet emission legislation and protect human and
environment. Similar technology is also being applied to DI gasoline engines for the
EU 6.0 vehicles.
2.3 Internal Combustion Engine Efficiency
As a means of reducing the fuel consumption and CO2 emissions of automobiles, it

is essential to improve the fuel conversion efficiency of the IC engine. The process
that which chemical energy contained in the fuel is converted to mechanical power in
an internal combustion engine is very complicated since this process involves
combustion, thermodynamic, fluid dynamics, mechanic movement as well as energy
losses such as friction. The fuel conversion efficiency of fuel is defined as how much
chemical energy of the fuel is converted into mechanical power output by
percentage. The summary of the energy conversion process and efficiency principles
are shown in Figure 2-6 [24,28-33].
14
Figure 2-6 Fuel Energy Conversion Process and Efficiencies in Internal Combustion
Engines [24,28-33]
Gasoline Spark ignition gasoline engine is still the first choice of power plant for
passenger cars all over the world since nearly 80% of light-duty vehicles use
gasoline engine according to the data shown by IEA. This number is a little bit lower
in developed countries but still close to 70% [34]. Since the fuel consumption of the
diesel engine is lower than that of the gasoline engine, diesel engine vehicles are
very popular in some European countries. Diesel is produced as part of the
distillation process of petroleum and can also be produced from renewable sources.
The diesel CI engine operates with higher thermal efficiency and much lower fuel
consumption. But they emit more gaseous and pollutant emissions and would need
expensive and costly aftertreatment devices. The key points for gasoline engines are
to improve its fuel consumption and fuel efficiency. Table 2.2 shows the main
methods to improve the engine fuel economy and the associated technical
challenges.
15
Table 2-2 Key points for improving Gasoline Engine Efficiency
Objectives Limitation of gasoline engine operation
1) Abnormal combustion: knocking, low speed

pre-ignition (LSPI)
Increase compression ratio 2) The design and reduction of friction result of

the limit of in-cylinder peak pressure.
1) The operation of a spark ignition (SI) engine is

stoichiometric combustion (Lambda=1) which
is richer than a diesel engine due to the fellow
reasons:
a) The air-fuel ratio (AFR) has a big effect

on the spark ignition and the flame
propagation of the gasoline engine. The
AFR of some typical SI engines should
be less than 20.
b) In order to decrease the CO, HC and

NOx in the emission of exhaust, a 3-
way catalyst is used on the SI engine.
The fraction of the air and fuel should
be stoichiometric of the combustion.
2) Because the combustion type is stoichiometric

combustion which can decrease the specific
heat ratio so that the combustion temperature
of gasoline is higher than that of a diesel
engine.
16
Increase specific heat ratio 3) When the engine is running at high load and
high speed, enrich mixture should be used to
cool down the temperature of the exhaust so
that to protect the exhaust component. But this
will cause a decrease in the combustion
efficiency as well as the specific heat ratio.
At part load conditions, the throttle is used to reduce

the mass of intake air so that to control the output as
Reduce pumping loss well as keep the AFR stoichiometric. High pumping
loss and the pressure of intake decrease as a result
of this.
1) Combustion time losses are caused by the

finite combustion speed in the gasoline
engine. Furthermore, both the increase of
residual gas fraction and EGR will decrease
the speed of combustion.
Optimise combustion time
2) Knocking at high load condition pushes the

retarded of the spark timing as well as CA50
which causes the combustion phase away
from the optimum point.
17
Figure 2-7 BSFC Contour Map of a Turbocharged Gasoline DI Engine
The specific fuel consumption of a gasoline engine varies with load and speed.
Figure 2-7 shows that the minimum BSFC is achieved in the middle-speed high load
area. As the speed and load go lower or higher from this area, the BSFC gets worst
due to factors listed in Table 2.3.
Table 2-3 Reasons for BSFC Increases in a Gasoline Engine Operation Map
BSFC
Main reasons
increase
i. Throttling at a lower load leads to the intake pressure decrease

which would cause higher pumping loss.
(1) ii. High residual gas fraction is also caused by low intake pressure.
What’s more, the in-cylinder temperature is low at the condition
of low load. Both of these results in combustion speed slowing
down and combustion time loss increasing.
18
i. Heat loss and gas leakage increase as a result of the time of
each cycle increase.
(2) ii. The decrease of engine speed would lead to in-cylinder charge
motion and turbulence becoming weak so that the burning rate
decrease and combustion duration extend.
i. For the gasoline engine, pumping loss decrease with the

increase of engine load.
ii. However, knocking combustion happens when the engine run at

high BMEP since the thermal load is higher. The spark timing
should be retarded which results in later CA50 and increase in
combustion time loss to deal with knocking combustion.
(3) & (4) iii. At low speed and high load conditions, the knock tendency
increase since the low combustion speed and high gas
temperature. Spark timing and CA50 should be retarded to
decrease the knocking tendency.
• The low-speed pre-ignition (LSPI) may occur at a low-

speed high load area in a highly downsized gasoline
engine.
i. Knocking tendency reduces at high-speed high load.
ii. Enrich AFR is adopted at high load high-speed condition to

reduce the exhaust temperature so that to avoid destroying the
(5) turbine blades in a turbocharged engine. Naturally Aspirated
engines employ enrich AFR is also used for improving the peak
power. Combustion efficiency and heat ratio get worse as a result
of over-fuelling.
The mechanical losses increase with the increase of engine speed

(6)
since friction and pumping loss increase.
19
2.4 Technologies for Improving Gasoline Spark Ignition Engine Fuel Economy
Recently, a series of new technologies have been researched and adopted on

gasoline spark ignition engines to improve the fuel economy and engine efficiency as
well as reducing the emissions. Some new technologies and will be reviewed are
introduced in this section.
2.4.1 Gasoline Direct Injection
In 1996, mass production of direct injection gasoline engines was introduced to the
market to achieve stratified lean combustion by Mitsubishi and Toyota. There are
some major advantages of the direct injection technology in the development of
gasoline spark ignition engines [35].
(1.) The injector injects the fuel into the combustion chamber directly at high
pressure and most of the fuel evaporates in the air which can absorb heat
from the air and lead to charge cooling effect. The in-cylinder temperature
decrease also causes the reduction of the heat loss, knocking tendency
and improved engine volumetric efficiency so that torque and power can
be increased.
(2.) Direct injection is necessary to achieve the stratified lean combustion for
higher thermal efficiency at part-load operations and can be employed to
improve the cold start performance and emissions.
(3.) Direct injection combined with boosting enables the engine downsizing to
improve fuel economy by operating the gasoline engine in the best BSFC
region at higher load.
(4.) Direct injection can be used to facilitate the fast catalyst light off during the
cold-start operation by retarding injection timing to form a stratified fuel rich
mixture in the central region of the combustion chamber, so that stable
ignition and combustion can be achieved after TDC to enable retarded
combustion and hence higher exhaust gas temperature.
20
2.4.2 Stratified Lean Combustion
The reason why the fuel economy of gasoline engines is worse than diesel engine is
the larger pumping loss and lower specific heat ratio of a stoichiometric mixture.
Research on lean combustion to reduce fuel consumption of gasoline engines has
been conducted for a long time [36-38]. It is essential to improve the ignition process
to ensure reliable ignition of homogenous lean combustion. In the stratified lean
combustion, the reliable ignition is achieved by generating a slight rich or near
stoichiometric air/fuel mixture around the spark plug at the time of spark ignition and
the rest of the chamber is filled with a leaner mixture.
The first generation of the DI gasoline engine with stratified lean combustion systems
[31,39], was achieved by wall-guide as well as air-guided direct injection. In 2006 and
2007, a new generation of gasoline direct injection engines with spray-guided
stratified combustion systems was developed by Mercedes-Benz and BMW [31]. It is
also important to note that Ford developed a spray-guided combustion system that
uses a solenoid-actuated multi-hole injector but it was not as robust [40].
The stratified lean burn DI gasoline engine requires an additional NOx after-treatment
system [41,42], which is costly and sensitive to fuel sulphur since NOx storage
catalysts can be destroyed by a high concentration of sulphur in the fuel of sulphur
poisoning. Particle matter emission is another problem of these stratified combustion
gasoline engines. Because of these challenges, most the manufacturers gave up on
the stratified lean-burn combustion. However, there are still some producers such as
Mecedes-Benz still have interests in this technology since it is likely to meet future
CO2 regulations [43].
2.4.3 CAI and HCCI
Controlled Auto-ignition (CAI) which is also called Homogeneous Charge

compression Ignition (HCCI) combustion is different from traditional spark ignition
21
combustion in a gasoline engine and compression ignition diffusion combustion in a
diesel engine [44]. Premixed fuel and air mixture which in a condition of highly
diluted and lean is used together with multiple auto-ignition sites in the combustion
chamber. The high temperature zones can be eliminated and the production of
particle matters and NOx decreased as a result of this technology. The gasoline
engine with CAI /HCCI enables unthrottled operation by using the lean mixture of air
and fuel with recycled burned gas so that higher engine efficiency and better fuel
economy can be obtained than traditional spark ignition engine.
According to previous research, CAI combustion can be achieved using the NVO
(Negative Valve Overlap) method by closing exhaust valves well before top dead
center (TDC), and retarding the opening of intake valves after TDC during the intake
stroke. As a result of early closure of the exhaust valves, a large amount of hot
burned gas is trapped inside the cylinder. The big amount of residual gases which
controls the engine’s load is determined by the exhaust valve close timing. What’s
more, the residual gas also plays an important role to increase the charge
temperature to achieve auto-ignition point. For a longer operating range of diluted
HCCI combustion, intake valve timing was considered as a potential method of
controlling the distribution of residual gases and charge temperature since intake
valve timing would have a direct impact on the intake flow and its interaction with
trapped residual gases [45].
There are also some challenging for CAI/HCCI such as the control of combustion
timing and switching between CAI/HCCI and spark ignition combustion mode, which
have hampered the production of this kind of engines. However, Mazda have just
recently announced that introduced the first commercial HCCI gasoline engine,
Skyactive X will be used in their new vehicles in 2019 based on the so-called Spark
Controlled Compression Ignition (SPCCI) [46], which is similar to what is proposed in
the project.
2.4.4 Gasoline DI Engines and Boosting (Downsizing)
22
Gasoline direct injection engine with homogenous stoichiometric combustion process
was researched and produced in order to avoid the complex exhaust aftertreatment
system on the stratified combustion engine. Such engines are equipped with the
standard three-way catalyst for emission control. However, there is little
improvement on the fuel economy level.
Over the last decade, recently, engine downsizing has been adopted by major
European OEMs to improve the fuel economy. It involves the replacement of a NA
engine by a boosted engine with smaller displacement volume and the same even
more power [47]. Downsized engine will run at higher BMEP more in the area of
good fuel consumption with better efficiency due to less pumping loss. Moreover,
engine downsizing also reduces CO2 emission by reducing the numbers of cylinder
so that the total friction and total weight of the engine decrease [48].
It is essential necessary to decrease the compression ratio of boosted downsizing

spark ignition engine in order to prevent knocking combustion which has bad effect
on fuel economy as well as engine performance. Because direct injection can result
in charge cooling effect to reduce charge temperature and knock tendency, there is
the synergy between GDI and boosting. Boost/ downsized direct injection spark
ignition engine is so success that the majority of new gasoline engine would be
turbocharged DI engines by 2018 in Europe as the Figure 2.8 shows [49].
Recently, the development of internal combustion engine has been changed from
downsizing towards rightsizing. In this trend, the focus is not on reducing
displacement, but instead on selecting the appropriate size to achieve a balance
between customer expectations for operating comfort and the manufacturer's ability
to reduce fuel consumption and greenhouse gas emissions [50].
23
Figure 2-8 Gasoline direct injection with boosting [49]
2.4.5 Variable Valve Actuation and Air Intake
A lot of work has been done by researchers to improve the air intake system to
optimise the gas exchange process since this process has a big effect on pumping
loss as well as combustion process engine performance on gasoline engine.
Different devices have been shown to make an improvement of fuel consumption
between 3%-12%: Variable Intake Manifold, Variable Cam Timing / Phaser (VCT /
VCP), 2-Step or 3-Step Cam Profile Switching (CPS) / Variable Valve Lift (VVL),
Continuous Variable Valve Lift (CVVL), camless Valvetrain and Variable Charge
Motion [51].
Variable intake manifolds can be realised by step adjusting or continuous adjusting

the manifold length to optimise intake turning effect for different engine speeds. The
volumetric efficiency of the engine increase as a result of variable intake manifold.
Therefore, engine torque can be optimised in the whole speed range [52].
24
It is very common to use VCT and VCP devices on modern spark ignition engine.
During the operation, the valve open and close timing can be changed by crank
timing which related to camshaft timing and it can be advanced or retarded by VCT
or VCP devices. Both hydraulically and electrically driven continuously adjustable
VCT are widely used in gasoline engines on both intake and exhaust valves to
optimise the valves open/close timings so that to reduce pumping loss as well as
improve low speed peak torque through scavenging behaviour [53,54].
One limit of these devices is that both the valves opening and closing timings have to
be changed simultaneously since VCT devices move the camshaft timing. The fixed
valve lift profiles of intake and exhaust valves cannot be changed is are another limit
of this type of systems. By switching the cam profile, some devices can provide 2-
step or 3-step lift adjustment of intake and exhaust valve such as Audi Valvelift
System (AVS) and Honda VTEC. CVVL systems can provide continuous valve lift
adjustment in a big range such as BMW Valvetronic, Nissan VVEL, Toyota
Valvematic and Mitsubishi MIVEC. VVL especially CVVL systems can reduce
pumping loss significantly by reducing throttling even provide un-throttling operating
on gasoline engine. These devices can be used with other technologies together,
such as Miller cycle, CAI/HCCI combustion to optimise engine performance and
emissions [55]. What’s more, Electronic and electrohydraulic camless valvetrains can
provide even more flexible control of valve actions. However, because of the limited
durability, cost and complexity, these devices haven’t been used on mass production
engines.
There are still some other devices and systems which been developed to adjust the
intake charge motion. A tumble flap fitted in the intake manifold is used by some VW
/ Audi gasoline DI engines which are able to increase tumble motion at part loads by
blocking the bottom half of intake ports [56,57]. These systems work on enhance air
charge motion and turbulence so that to beneficial air fuel mixing as well as
combustion process hence improve gasoline engine fuel consumption and efficiency.
2.4.6 Variable Compression Ratio
25
Higher thermal efficiency can be achieved by increasing the compression ratio. The
main reason why the gasoline engine has worse fuel conversion efficiency than the
diesel engine is its lower CR since the detonation limit at high load operation. With a
Variable Compression Ratio mechanism, fuel conversation efficiency of gasoline
engine can be improved since it allows gasoline engine use high CR when it runs at
low load and while reduced CR at high load operation. Different VCR systems have
been shown to improve the fuel economy of the gasoline engine by 5% to 12 %. If
VCR can cooperate with other technology such as lean burn, CAI/HCCI, VVL, Miller
cycle, there would be more benefit [58]. Nissan has introduced the VCR technology
in their premium brand vehicles. VCR hasn’t been more widerly applied on mass
production engine since there are still quite challenge such as durability, complexity
and high cost. However, VCR technology is considered with great potential in
reducing CO2 emission. Nissan has introduced the VCR technology in their premium
brand vehicles in 2016.
2.4.7 Water Injection
A big problem in the gasoline engine is the knocking combustion at high load which
is normally solved by retarding spark timing resulting in poor performance and fuel
economy. For boost downsized engines, this problem becomes more intensive and
limits their performance.
The research on water injection can date back to the 1920s when gasoline piston
engines were used in aircraft. Recently, this concept attracts the sight of researchers
again since its potential to reduce knocking combustion and decrease heat transfer
loss and consequently improve the fuel conversion efficiency of downsized SI
engines [59]. The main challenge is the requirement of additional storage and supply
of distilled water.
26
2.4.8 Cooled Exhaust Gas Recirculation
Exhaust Gas Recirculation (EGR) technology has traditionally been used to reduce
NOx emission and fuel consumption on gasoline engines. EGR is used to dilute the
in-cylinder charge and maintain stoichiometric AFR at the same time so that to lower
the pumping loss at part load and in-cylinder temperature in the combustion
chamber. More recently, cooled EGR has been exploited to reduce knocking
tendency at high load operation. Toyota produce their latest hybrid vehicles with Si
engine which use EGR as its key measure and it achieves the highest thermal
efficiency of 41% [36,60].
2.4.9 Miller Cycle
The name Miller Cycle came from an American engineer, Ralph Miller who proposed
and applied the early intake valve closure as an effective means to improve the
boost SI engine’s performance in the 1950s. The main potential benefits are listed
below:
• For a supercharged SI engine, thermal load and knocking/ pre-ignition

tendency should be decreased by the earlier IVC.
• The charge temperature would decrease by reducing effective
compression ratio (ECR) associated with earlier IVC while retain the
thermodynamic benefit of a high expansion ratio at the same time.
• The AFR should be correct to control auto-ignition and enable
operating at higher load and high expansion ratio / geometric
compression ratio.
• Load control by intake valve can help to reduce the pumping loss by
wider open throttle.
• Scavenging should be improved since high boost pressure and
subsequent increased differential between the boost pressure and the
exhaust back pressure.
27
• Different load operation should apply different valve lift.
• A supercharged/ turbocharged engine should be a preferred
application, but the process can also be used on a NA engine as well.
• The work can apply on gases, diesel as well as gasoline engine.
Related to the Miller cycle, the Atkinson cycle should also be mentioned. It is named
by the British engineer James Atkinson who was the first person to describe a four-
stroke internal combustion engine process with unequal compression stroke and
expansion stroke. This concept was first presented in 1886 for an opposed piston
engine [61]. After that, a four-stroke engine with special crank mechanism was used
to realize the Atkinson process in 1887 [62]. However, the name of Atkinson cycle
engine has recently been adopted to describe the technology based on the retarded
intake valve closure by means of a longer intake valve lift profile. Similar to the early
intake valve close (EIVC), the late intake valve closure (LIVC)in such engines is
employed to lower the effective compression ratio to enable higher geometric
compression ratio (hence longer expansion stroke) being used for better engine
efficiency, such as the engines used in the Toyota Prius Hybrid Vehicles.
2.5 Spark Ignition Engines with Alternative Fuels
2.5.1 Natural Gas Engine
Natural gas, whose main composition is methane, is an alternative fuel to gasoline

for the spark ignition engine with a higher H/C ratio and greater research octane
number (about 130). When changing the fuel from gasoline to CNG, the H/C ratio is
changed from 1.85 to 3.7-4.0. Compared to the stoichiometric SI gasoline engine
operation, the natural gas engine can be operated at a higher compression ratio. If
operated with a lean burn mixture and a high rate of EGR, a significant reduction in
pollutant emissions and an improvement in thermal efficiency can be achieved by a
natural gas engine [63]. It has been shown that the natural gas engine can achieve a
28
CO2 emission reduction by 20% compared with the gasoline engine at equal torque
and power [64].
Moreover, since natural gas doesn’t contain aromatic compounds such as benzene
and contains less dissolved impurities like sulphur compounds than petroleum so that
natural gas engine produces less little PM than gasoline engine particle emissions.
Natural gas is also reliable when consider its safety. Natural gas has the characteristic
of low density as well as high dispersal and it is difficult to form explosive mixture when
there is a leak event.
With the significant increase in the production of natural gas and bio-methane from
renewable sources, there is increased demand for the research and development of
high-efficiency and ultra-low emission gas engines for light-duty and heavy-duty
vehicles.
2.5.2 Alcohol as Alternative Fuel in Spark Ignition Engine
2.5.2.1 Overview of Alcohol Fuels
The chemical structure of alcohol can be expressed by CnH2n+1OH. When they are
produced from renewable sources, alcohol fuels can lead to the reduction of fossil
fuel consumption and greenhouse gas emissions as well as toxic emissions [65]. In a
spark ignition engine fuelled by alcohol, the air-fuel mixture has some advantages
such as reducing knock tendency and toxic emissions [66].
As a fuel for high-performance engines, methanol which is the chemically simplest

alcohol was used during the 1930s as an alternative to gasoline [67]. The liquid form
of methanol is toxic, tasteless, colourless, and is commonly referred to as wood
alcohol and can be produced from a range of raw materials and sources [68]. What’s
more, methanol has a lower boiling point compared with gasoline, which allows it to
evaporate more quickly, which has a positive effect on engine combustion and
reduces hydrocarbon emissions. In addition, as a result of the high oxygen content of
29
methanol and its simple structure, its combustion can be more efficient and less
polluting in spark-ignition engines [69,70].
Ethanol is derived from fermentation of biological material and is a renewable and

efficient fuel [71,72]. In the United States, ethanol was first proposed for using as a
fuel for internal combustion engines in the 1930s, and has been widely spread since
1970. Brazil established the national alcohol program (NAP) in response to the first
oil crisis in 1973, which led to a reduction in its dependency on fossil fuels [73].
Approximately 60% of global ethanol production was produced in 2014 by the United
States, while the Brazilian government produced approximately 23.47 billion litres
representing 25% of global production. From 2007 to 2015, the global ethanol
production by country is represented in Figure 2-9 [74]. The production rose sharply
between 2007 and 2010, however, the largest production occurred in 2015, after that
it went down in 2011 and 2012. With gasoline prices increasingly high and emission
regulations becoming stricter, ethanol could become a more important renewable
fuel. There have been extensive studies undertaken to ensure high ignition
temperatures, higher research octane numbers, lower freezing points, higher heats
of vaporization for ethanol in comparison with gasoline [75,76].
30
Figure 2-9 The global production of ethanol from 2007 to 2015, by country [74]
Regular gasoline can have an auto-ignition point ranging from 221°C to 257°C, and
that of ethanol is at 329°C, and methanol is 470°C [77].
When a substance is in the process of changing from liquid to vapor, additional heat
must be absorbed in order to bring about that change, and this is how heat of
vaporization come from [78]. To changing from liquid state to vapor, ethanol needs to
absorb 918.7KJ/Kg (25°C, 1ATM), while gasoline needs 341.9KJ/Kg. Since ethanol
absorb more heat during evaporation, it can create a cooler operating condition
during injection in an IC engine. Because it has a high latent heat of evaporation,
ethanol contributes to the reduction of NOx emissions [79]. It is typically thought that
high latent heat of evaporation of alcohol fuels has indirectly affect on knocking by
cooling the air-fuel mixture evaporating during the intake stroke while the fuel is
being injected into a DISI engine and thereby postponing the ignition of the fuel
under knock conditions [80]. This is also called charge cooling effect.
As alcohol fuels contain a higher percentage of oxygen compared with gasoline, their
combustion could be more complete and cleaner, and the in-cylinder temperature
could be reduced because of the oxygen content and high latent heat of vaporization
[81]. Ethanol’s oxygen content (34.8 by weight) is lower than that of methanol (49.9
31
by weight). Methanol will therefore function as a less diluent than other oxygenates.
The engine fuelled by alcohol fuel has higher combustion efficiency compared with
gasoline. According to some studies, this can be attributed to the relative oxygen
content in fuel, despite the fact that oxygen limits fuel consumption performance
because it does not permit any additional energy conversion. In addition, the oxygen
content in alcohol fuel contributes to the completeness of combustible [82]. What’s
more, since alcohol fuels have higher oxygen/carbon ratios, its lower heating value is
lower than gasoline which causes more fuel is consumed to achieve the same power
as gasoline [83].
A higher RON (research octane number) is generally associated with alcohol than
the MON (motor octane number). The RON of ethanol and methanol in this research
are 107 and 108, their MON are 89.5 and 88.7. The sensitivity (RON-MON) of
gasoline is lower than that of ethanol and methanol.
Similar to ethanol, butanol can be produced from corn grain and another biomass
[84]. Biobutanol has the advantages of being immiscible in water, having a higher
energy content, and having a lower Reid vapor pressure when compared with
ethanol. What’s more, butanol meets the Renewable Fuel Standard's threshold for
reducing greenhouse gas emissions by 20%.
Isobutanol attracts currently the most active commercialization work among the four
isomers of butanol. It is permitted to blend up to 12.5% biobutanol into gasoline
under two provisions of the Clean Air Act. In addition, the U.S. Environmental
Protection Agency (EPA) recently granted a waiver that allows a 16% biobutanol
blend to be considered the same as E10 as a legal fuel. The EPA has approved
blends containing up to 16% biobutanol in June 2018.
With the following advantages which make biobutanol an alternative to conventional

transportation fuels:
• In comparison with gasoline alternatives, biobutanol has a relatively high

energy content.
• The vapor pressure of biobutanol is lower than that of ethanol, which results in
less volatility and evaporative emissions.
32
• A variety of feedstocks can be used to produce biobutanol.
• Butanol releases less greenhouse emissions compared with other fossil fuels.
What’s more, growing feedstocks capture carbon dioxide, which is offset by
carbon dioxide released by the combustion of biobutanol, thereby reducing
greenhouse gas emissions overall.
• Due to its immiscibility with water, butanol might be capable of being
transported in pipelines, thereby reducing transportation costs.
2.5.2.2 Application of Alcohol Fuels in Spark Ignition Engines
Ethanol and methanol have been in used as a blended component with gasoline [85].
The most widely used one is E10 which consist of 10% ethanol and 90% gasoline.
Brazilians have long used ethanol as a neat fuel. Approximately five million cars on
the road are fuelled by 190-proof ethanol (95% ethanol, 5% water) and hence neat
(pure) ethanol are well suited for use as a primary fuel for vehicles.
Several studies have indicated that ethanol and pure methanol increase engine
thermal efficiency as a result of lower in-cylinder temperatures and peak in-cylinder
pressures. What’s more, pure ethanol and methanol release less NOx and CO2 than
gasoline [86]. When change gasoline to pure methanol, the output power and CO
decrease while the brake specific fuel consumption increase since the lower heating
value of methanol is lower than that of gasoline [87].
Koichi et al. [88] studied the effect of pure ethanol on spark ignition engine compared
with gasoline. The experiment was carried out on a 1.5L 1NZ-FE natural aspirated
Toyota engine whose compression ratio has been increased from 10.5:1 to 13:1 by
changing the piston. During testing, the engine’s speed and load were 2800rpm,
2bar BMEP. It has been concluded that ethanol fuelled engine’s thermal efficiency
and torque have been increased since ethanol has the better anti-knock quality and
lower cooling heat loss. What’s more, the NOx released by ethanol is lower because
of the lower in-cylinder temperature. In addition, ethanol has lower THC since
gasoline has components that boil at high temperatures, but ethanol does not.
33
Senthil et al. [89] did simulation research about the possibility of using early intake
valve close (EIVC) Miller cycle to increase the efficiency of an ethanol, methanol
fuelled turbocharged heavy-duty spark ignition engine by GT-Power. The test was
done at 1200rpm 25bar IMEP. The result shows that even if the turbocharger
efficiency is only 49%, the brake thermal efficiency could still be increased by 2-3%
by increasing the geometric compression at stoichiometric combustion. At the
condition of lean burn, minimal turbocharger efficiency of 55% is needed to promise
the BTE increase for both ethanol and methanol with Miller cycle since the pumping
loss increase.
An experimental study about compression ratio on methanol’s combustion

characteristic, performance and emissions was carried out on a 0.25L variable
compression ratio engine. The engine speed during testing was between 1500rpm to
3500rpm. The result shows that with the compression ratio increased from 6:1 to
10:1, the engine power and brake thermal efficiency increased by up to 14% and
36%. In addition, CO, CO2 and NOx emissions decreased by 7%, 30% and 22%
respectively. HC emissions increased by approximately 12% with an increase in CR
from 6 to 10 since the surface to volume ratio increases as CR increases. As a result
of this, the flame cools in the places near to surface hence misfire [90].
As a result of the better anti-knock characteristic of alcohol fuel, a higher

compression ratio (CR) can be achieved (to levels of 12:1 and above) without spark
retarding in order to avoid knocking. In 1981, Ford achieved a 20% increase in
power and a 15% increase in efficiency for the M85 Escort model due to the increase
in compression ratio [91].
Due to the faster flame speed and wide flammability ranges of alcohols, many
alternative options are available for load control, particularly in the case of methanol.
An investigation has been done on an experimental turbocharged lean-burn 0.61L
methanol engine. The engine speed was controlled at 1200rpm. The result shows
that the BTE increased by 14% compared with stoichiometrically fuelled engines with
throttled load control. What’s more, A reduction of over 50% CO was achieved. On
the other hand, the lean burning strategy resulted in NOx emissions increased by
150%, which brings into question the practicality of such a strategy [92].
34
Another research also worked on methanol fuelled spark ignition engine load control
strategies. In this research, three strategies were compared including the throttled
stoichiometric operation, wide open throttle (WOT) lean burn operation and WOT
stoichiometric operation using EGR to control the load. Compared with the throttled
operation, the other two strategies increased the BTE by up to 5% because of
cooling and dissociation losses. What’s more, there was also a sharp decrease of
NOx by WOT EGR since dilution caused decreasing in in-cylinder temperature. But
there wasn’t a significant reduction of NOx by lean burn. EGR plays a more important
role here compared with lean burn since exhaust gases have a greater heat capacity
than air. Above all, WOT EGR appears to be an appropriate strategy for methanol
engines [93].
There are still some challenges for using pure methanol on vehicles due to its cold
start problem when the temperature is under 15°C as well as safety concern [94].
Under cold conditions, vaporizing fuel and producing a combustible mixture can be
achieved in several ways, such as heating the fuel, heating the intake port, and
raising the temperature of compressed air. Heating the fuel can be achieved by
adding a heater to the injector. Research did by Daniel et al. [95] focused on the
heated injector and fuel rail to deal with ethanol’s cold start problem. With prototype
systems, the E100 cold starts were robust and effective at temperatures as low as -
5°C without gasoline assistance. What’s more, it has been also found that the heated
injector can decrease THC and CO by 40%, but there is a tiny increase in NO x.
As an alternative to directly heating intake air, increasing the compression ratio and
optimizing valve timing can be used to increase the compressed air-fuel mixture’s
temperature. In practical use, raising compression ratio and optimizing valve timing
are thought to be most effective, since they do not require preheating the fuel as well
as intake air [88].
Although the potential benefits of methanol and alcohol fuels in a spark ignition
engine had been demonstrated by the previous publications, there are limited
studies to quantify the improvements that could be achieved by methanol or ethanol
fuel in a modern direct injection highly downsized spark ignition gasoline engine.
35
Therefore, the current study is aimed to fill the gap by performing a systematic
experimental study of the alcohol-fuelled engine’s performance, combustion and
emission characteristics across the whole range of engine operations. It is also a
direct comparison between E10gasoline, ethanol and methanol in the same boost
high compression ratio spark ignition engine. Some high speed& load testing were
also carried out with the IMEP up to 28bar. In addition, the effect of injection timing
and rail pressure on the engine performance and emissions was investigated to find
the best injection strategies for ethanol and methanol at different speeds & loads.
Particle emission data were collected for every test condition which includes load
sweep as well as injection timing & pressure sweep since there is limited research
done for ethanol and methanol’s particle emission. Finally, the spark sweep study
was carried out to evaluate how alcohol fuels can be used to facilitate the fast
catalyst light off for effective cold start emission controls.
36
3. Chapter Three: Experimental Facility and Methodology
3.1 Introduction
This research is based on alcohol fuel (ethanol & methanol) running in a highly
downsizing, single cylinder, direct injection engine with a flexible Engine Control Unit
(ECU) to improve the fuel economy and decrease all gaseous emission as well as
Particulate Matter (PM). All the experiment facilities and experimental methodology
which used to collect the data in this project are described in this chapter.
3.2 Experimental Set-up
In this project, the most important experimental facility is the single cylinder engine
test bed which will be shown in this section. The main part of the engine test bed
which includes a single cylinder engine, conditioning and supply system (external
boost rig, oil, coolant, fuelling system), data acquisition system, engine control unit
(ECU) as well as the emission analysers (HC, NOx, CO2, CO, O2, particulate matter).
The engine is coupled to a transient electrical motor dynamometer which can both
control the engine speed & throttle position and measure the engine’s torque output
during engine testing. All the above are shown in Figure 3-1.
37
Figure 3-1 Schematic of Single Cylinder Engine Testbed
38
3.2.1 The Single Cylinder Engine
All the data are collected from the experiments conducted on a single-cylinder GDI
engine. The GDI engine is built with a high-speed single cylinder Ricardo Hydra
crankcase. The cylinder head was provided by MAHLE which is used for their 1.2L
3-cylinder downsizing demonstrator engine [96-102] and a special cylinder block was
designed to match the single cylinder crankcase. All the test bed facilities, as well as
the engine main specification, are shown in Figure 3-2 and Table 3.1.
Figure 3-2 Brunel-MAHLE Single Cylinder DISI Engine on the testbed
39
Table 3-1 Engine Specification
Displacement Volume 400cm3
Stroke 73.9mm
Bore 83mm
Connecting rod length 123mm
Compression ratio 11.43:1
Number of valves 4
Maximum In-cylinder pressure 120bar
Maximum pressure rise rate 6 bar/CAD
Intake cam timing (maximum open 80-120 CAD ATDC

point)
Intake cam duration 240 CAD
Exhaust cam timing (maximum open 100-120 CAD BTDC

point)
Exhaust cam duration 278 CAD
Direction injection system Bosch Gasoline Direct Injector
Intake port fuel injection system Bosch EV 12 with 4 narrow conical sprays
and maximum injection pressure of 8 bar
The engine is built with a pent- proof combustion chamber which has two intake
valves and two exhaust valves. The Bosch direct injector is fitted on the middle of the
cylinder head and the spark plug is mounted next to the direct (Figure3-3) injector to
achieve the spray charge stratification. For this direct injector, three times of injection
which include one early injection (main injection) and two late injection could be
40
achieved in one cycle. The injection timing and pressure for the main injection as
well as the injection timing and duration for the late injection could be adjusted on
ECU (INCA software). The maximum injection pressure for this injector is 200bar
and it was adjusted in line with the engine speed and load. For each injection event,
the minimum injection duration is 0.15ms and the minimum gap between the end of
one injection and the start of the next is 0.2ms.
Figure 3-3 Combustion chamber with the injector and the sparkplug [97]
3.2.2 Details of Cam Profiles and Valve Timing
There are two hydraulic variable cam phasers situated in the cylinder head, one of
them is for the intake cam and the other one is for exhaust cam. As a result of this,
the intake and exhaust cam timings can be adjusted up to 40crank angle degrees
(CAD). The design details and phasing of the three intake cams and the exhaust
cam for the testing engine are shown in Table 3.2 and Figure 3-4. In this research,
the testing engine is fitted with the standard intake camshaft with 240CAD opening
41
duration and an exhaust camshaft with 278CAD opening duration, both of the intake
and exhaust valves have a maximum 11mm lift.
Figure 3-4 Intake and exhaust cam profile and phasing
42
Table 3-2 Intake and Exhaust Cam Specification and Phasing
Minimum Overlap Phasing [dCA ATDCNF] Maximum Overlap Phasing [dCA ATDCNF]
Opening
Cam Duration IVO/EVO IVC/EVC IVO/EVO IVC/EVC
(TOR)[dCA] MOP MOP
(0.5mmlift)
(0.5mm lift) (0.5mmlift) (0.5mm lift)
EIVC
152 75 16 134 35 -24 94
Intake
Standard
240 120 13 227 80 -27 187
Intake
LIVC
292 144 16 272 104 -24 232
Intake
Exhaust 278 -140 -262 -17 -100 -222 23
43
3.2.3 Engine Oil System
The engine oil system is responsible for supplying the engine with lubrication oil at
sufficient pressure and maintaining the specific oil temperature. For this testbed, the
oil system consists of an oil tank, oil filter, oil heater, oil pump and an oil heat
exchanger.
A wet sump lubrication system which stores the standard Mobil 1 0W-40 oil is
featured in this single cylinder engine for all testing work. The oil pump in this system
is an externally located single speed three-phase electronic oil pump which can
circulate the oil at a nominal flow rate of 9.1L/min. In order to pre-heat the oil before
running the engine or heat the oil to the requested temperature at low speed& load,
two 1kW electronic oil heaters are immersed in the oil sump. An oil heat exchanger
which is used to cooling the oil with the maximum heat rejection rate of 4kW situated
upstream of crankcase oil gallery. The flow rate of raw cooling water run through the
heat exchanger is controlled by a Spirax Sarco capillary actuator. The oil
temperature is controlled by switching on/off the oil heater as well as adjusting the
cooling capacity of the heat exchanger which is achieved by turning on/off tap of
cooling water. For low load testing, the oil temperature can be controlled by
switching on/off directly but for high load, the oil heater is off all the time and the oil
temperature is controlled by adjusting the cooling capacity of the heat exchanger. An
AC Delco X19 equivalent oil filter which is mounted next to the oil heat exchanger is
used to filter the oil.
In order to keep the oil temperature and pressure at a safe range (especially high
speed& load test point) during testing, a set of sensors are fitted in the oil circuit. There
are three pressure sensors situated upstream of the crankcase oil gallery and the
minimum oil pressure is set to 4.2bar at this point. A pressure gauge for quick reading
during the start-up is located after the crankcase oil gallery. A pressure switch for the
testbed emergency shutdown mechanism and a Druck PTX1400 (0-10bar, 4.2mA)
pressure transducer provides oil pressure information to the low-speed data
acquisition system. The oil temperature inside the crankcase oil gallery is measured
by two Platinum Resistance Thermometers (PRT), one of them is for testbed high oil
temperature emergency stop which is set to 100°C and connected to the engine
44
dynamometer controller, the other one is connected to the low-speed data acquisition
system. The last PRT is located at the outlet of the oil sump which is used for oil
temperature acquisition.
3.2.4 Coolant system
The coolant system is responsible for supplying coolant to the engine with a desired
flow rate and temperature. The main part of the coolant system includes a coolant
tank, coolant pump, ball valves, coolant flow meters and coolant heat exchanger.
The coolant which is used for all testing work is a mixture of half de-ionised water
and half of the ethylene glycol. The coolant tank which is used to store the additional
coolant in the circuit is situated 120mm higher than the top of the engine cylinder
head coolant jacket so that the cooling jacket is fully submerged in coolant. What’s
more, the coolant tank can also allow the expansion of coolant when it starts to
become hot. An internal Pierburg 12V DC Coolant pump which speed is controlled
by engine control system (ECU) is used to circulate the coolant in the circuit with
different flow rate depends on engine speed and load (30L/min at low load, 50L/min
at high load). The entire coolant flow rate into the engine is varied by a bypass ball
valve which is installed upstream of the engine inlet, and the flow rate is set to
13L/min so that the delta coolant temperature can be controlled within 6°C between
engine inlet and outlet. The coolant run through the cylinder head and cylinder block
was split by two ball valves situated outlets of cylinder head and blocks so that both
of their coolant flow rates can be controlled independently.
During testing, the way to control coolant temperature is similar to oil temperature. A
3kW immersion heater is connected to the coolant system in order to heat up the
coolant at the warm up stage or the load is low. What’s more, a Bowman heat
exchanger whose maximum heat rejection rate is 53kW is located after the Pierburg
pump, and it works as the oil heater. When the engine is running, the coolant
temperature is controlled to 90°C by adjusting the effect of coolant heating and
cooling.
45
The coolant flow rate, temperature as well as pressure are measured by a few
sensors situated in the coolant circuit. Two Apollo turbine type flow meters are used
to measure the cylinder block and total engine coolant flow rate, one is located
between the outlet of the cylinder block, and another one is located before the
Pierburg pump. The coolant pressure is measured by a Druck PTX1400 0-4bar
4.2mA pressure transducer which is before the engine coolant inlet where a PRT is
at the same location to measure the coolant temperature into the engine and another
PRT is used to measure the coolant outlet temperature. Another PRT which located
into the coolant tank and connected to the dynamometer is for the testbed
emergency shutdown. What’s more, an automotive type coolant temperature sensor
which fitted at the cylinder block outlet side provides the input to ECU.
3.2.5 Fuelling System
A low-pressure loop and a high-pressure system are the main parts of the fuel
conditioning and supply system. The low-pressure part is to supply the fuel with a
specific temperature (30°C ±3) and pressure. A fuel flow meter is also included for
fuel consumption measurement. The high-pressure fuel system consists of a high-
pressure fuel pump, high-pressure rail and direct injector.
The fuel used for baseline testing is EURO 6 E10 with a Research Octane Number
(RON) of 95, pure Ethanol and Methanol. Two 50L stainless steel fuel tanks located
higher than all other components are used to store fuel, one is for normal gasoline,
and the other is for alcohol fuel. Two Bosch automotive 12V low-pressure fuel pumps
are installed under the fuel tanks. Each one is used for its matched fuel tank, and the
low-pressure pumps are controlled by the fuel circuit of the testbed depending on the
test fuel. Two groups of fuel systems run independently and cannot run at the same
time. The fuel runs through a heater exchanger (fuel cooler), which controls the fuel
temperature by adjusting the flow rate of raw cooling water run through it as a result
of sucking by the fuel pump. After that, the fuel is running through a filter, then a
mechanical pressure regulator which can promise the fuel supply pressure is 5bar,
and it is shown on a gauge and should be checked every time before running the
46
engine. This loop is connected with the entrance of the fuel tank at atmospheric
pressure by a pipe that goes upwards, and its top ends are higher than the rest of
the system so that the air bubbles can be removed from the low-pressure fuel
system.
After the regulator, a Druck PTX 1400(1-10bar, 4.2mA) pressure sensor and a PRT
temperature sensor aim to measure the pressure and temperature of the supply fuel,
both of them are connected to the low-speed data acquisition system. During all
engine testing, the fuel consumption is measured by a Coriolis type mass flow meter
which is manufactured by Endress+Hauser. It has a DN01 1/24’’ sensor size, which
is more suitable for very small mass flow rate measurement. For this project, the
range of the fuel flow meter is 0-20kg/h, which is determined by considering the
lower heating value (LHV) of ethanol and methanol.
Figure 3-5 Coriolis mass flow meter by Endress+Hauser
The fuel line is split into two lines after the fuel flow meter. One of them connects to
the low-pressure fuel rail which supplies fuel to the PFI injector directly (not used in
this project). The other fuel line supplies fuel to the high-pressure cam-driven type
47
fuel pump which is driven by the intake camshaft. The required amount of fuel is
supplied to the high-pressure common rail, and the fuel pressure can be increased
up to 200bar by the high-pressure fuel pump. The rail pressure is controlled by the
ECU through a proportional–integral–derivative (PID) control. To achieve this, a
high-pressure automotive type pressure sensor is put on the common rail. The
physical rail pressure is detected by the pressure sensor which gives feedback to
ECU, after that it compare the actual rail pressure with the set pressure and change
the rail pressure by increasing or decreasing the fuel quantity fed into the common
rail. The high-pressure fuel is supplied from the common rail to the injector through a
short stainless-steel pipe.
3.2.6 Intake System
The boosted downsized GDI engine requires the supply of compressed air above
ambient pressure. This is achieved by an external boost rig.
The dried compressed air with pre-set pressure and temperature is supplied to the
engine by the external boost rig. The boost rig includes a compressor (CompAir
HV22RS AERD hydrovane type compressor) which is driven by a 22kW electric
motor, a refrigerator (dryer unit), a five mivron oil filter and a 272 litres receiver.
The minimum pressure which provided by the compressor is 6bar, the air has a
nominal flow rate of 3.53m3/min at this pressure. After the compressor, there is a
refrigerator which can provide air with humidity less than 3% (according to the
manufacturer manual) and it can also cool down the air to approximately to 3°C. A
receiver which volume is 272 litres situates downstream of the refrigerator, it can
store and keep the compressed dried air’s pressure stable. The rig controller turns
the compressor on/off in order to maintain the pressure of the air in receiver between
6.5 to 7bar. After that, two Parker Hannifin EPDN4 type pressure regulators are used
to regulator the intake pressure to the required pressure for engine air consumption
and has a precision of ±0.15bar. The first is only used when the load is low and is
controlled manually. It will be bypassed or fully opened when the engine load is high
48
since more compressed air is required. The second Parker regulator is close-loop
controlled by the ECU and is fitted after the first Parker pressure regulator.
The intake temperature is set to 40°C for all experiment in this thesis. In order to
increase the dried compressed air temperature which cooled by the dryer from 3°C
to 40°C, an electrical 3 kW Secomark 632 type heater is used after the second
Parker pressure regulator. This intake heater is closed- loop controlled by ECU with
the precision of ±1°C.
A large volume plenum(accumulator) with a k-type thermocouple which volume is

40L and made of stainless steel is included in the intake system. The thermocouple
is used to measure the air temperature inside the plenum and then as a feedback to
a Eurotherm PID controller to control the intake heater. Another function of the large
plenum is to minimize pressure fluctuations. A Bosch DV-E5 40mm automotive type
electronic throttle is situated downstream of the plenum which is controlled by dyno.
The temperature and pressure between the plenum and throttle body are measured
by a Bosch automotive type boost pressure and temperature sensor and gives
feedback to ECU. After receiving the feedbacks, the ECU will increase or decrease
the boost pressure before the throttle body by adjusting the second pressure
regulator. A large diameter and long intake pipe are used between the throttle body
and the engine intake port which is for achieving the desired and stable intake port
pressure required at different speeds and loads. The pressure wave after the throttle
body could be more stable as a result of this. The air mass flow after the throttle
body is measured by a Bosch 1-way hot wire automotive type mass air flow meter
which gives signal to ECU. The pressure and temperature are also measured after
the throttle has been applied using pressure and temperature sensors which
communicate with the ECU.
There is also a Kistler 4005B piezoresistive absolute pressure sensor (Figure 3-7)
installed in the intake port just before the intake valves for the purpose of measuring
the transient intake port pressure for the combustion analyser software and for
subsequent analysis. The intake port air temperature is also measured by a PRT in
the same location.
49
3.2.7 Exhaust System
The main component of the exhaust system is made of austenitic stainless steel
pipes that have no sudden changes in diameter, thereby ensuring that the exhaust
gas flow is as smooth as possible. The pipe is designed to withstand temperatures of
up to 780°C and pressures of up to 4 bars in the exhaust gas. In order to minimize
the back pressure associated with exhaust, a large automotive-type exhaust muffler
is installed at the exit of the exhaust system. Additionally, the exhaust back pressure
is controlled by a servo motor actuated butterfly valve fitted upstream of the muffler.
When the dynamometer computer is required to adjust the valve, the valve can be
remotely adjusted to achieve the required back pressure on the testbed so that the
presence of the turbocharger turbine can be simulated.
It is necessary to install a K-type thermocouple close to the exhaust valves in order

to measure the exhaust port temperature. In spite of its higher accuracy than
thermocouples, PRT is not used as a measurement point for exhaust port
temperatures because it cannot withstand high exhaust temperatures. Two different
kinds of pressure sensors are used in the exhaust system. The first one is fitted
100mm after the exhaust port which is Kistler 4005B type piezoresistive absolute
pressure sensor (Figure3-6) with a cooling water adaptor, the cooling water is
controlled by a valve and should be on before running the engine. The transient
exhaust pressure is measured by the Kistler pressure sensor which is necessary for
gas exchange analysis. Kistler pressure sensor is connected to the high-speed NI
card. The specification of the Kistler sensor is also included in Figure3-6.
50
Figure 3-6 Kistler 4005B type Piezoresistive Absolute Pressure Sensor
A Druck PTX 1400 (0-10bar) type low-speed pressure sensor is used to measure the
mean exhaust pressure. This pressure sensor is located downstream from the first
and is connected to the low-speed NI card. What’s more, two automotive-type
Universal Exhaust Gas Oxygen (UEGO) Lambda sensors are installed on the
exhaust pipe. The lambda is controlled by ECU while the signal is from the first
lambda sensor, the second lambda sensor is only used for confirming the reading of
the first sensor is right. What’s more, there are also some emission analysers such
as gaseous and particulate analyser connect to the exhaust pipe. The NOx is
measured by a Signal Ambitech Model 443 Chemiluminescent Analyser, the HC is
measured by a Signal Rotork Model 523 FID HC analyser, and the CO, CO2 and O2
are measured by a Horiba MEXA-554JE. In addition, a fast response DMS 500
particle analyser is used for measuring particle size and particle numbers.
3.2.8 Dynamometer
The engine’s torque and load are controlled by an electrical motor dynamometer
which model is CPEngineering 48kW AC motor with a 4 quadrant AC regenerative
inverter derive. For this dynamometer, the maximum speed is limited to 6000rpm
51
and the maximum operating torque is 140 Nm. The dynamometer is controlled by the
CP CADET V14 dynamometer control system in order to adjust the engine’s speed,
load as well as exhaust back pressure valve position when the test is required. The
operating envelope of the dynamometer is shown in Figure 3-7. The highly
downsized boosted GDI engine used in this experimental study was capable of 120
kW/litre. This is equal to 48 kW of power for the single cylinder engine which has the
capacity of 0.4 litres.
Figure 3-7 Comparison of the dynamometer torque curve and the engine torque curve
3.3 Engine Control Unit and Management System
An advanced engine management system (EMS) is used in order to control the

engine more flexible and smoother. This system is consisting of the engine control
unit (ECU), some sensors, actuators, a wiring harness which includes two looms as
52
well as a computer used to control the ETAS INCA V7.0 software for the ECU
communication.
The ECU (shown in Figure 3-9) which is the main part of the engine management
system is provided by MAHLE Powertrain. (the ECU is based on the AFT PROtroniC
platform). In order to control the engine running, ECU needs to deliver enough
processing power and supports sufficient input and output channels. What’s more,
this ECU has all the functions to operate a modern engine. Also, it has the ability of
programming extra code for additional control if needed.
Figure 3-8 MAHLE Flexible engine control unit
The main channels of ECU to control the main engine parameter is shown in Table
3.3. All these data are also logged by ETAS INCA software for further data
processing and analysis.
Table 3-3 The Main ECU inputs and outputs used during the experiments
Inputs(sensors) Outputs(actuators)
Throttle position Electronic Throttle body
Manifold pressure and temperature
53
Intake cam position Intake cam phaser
Exhaust cam position Exhaust cam phaser
Exhaust lambda PFI Injector
Boost pressure and temperature Boost pressure regulator
Air mass flow meter DI injector
High pressure fuel rail pressure High pressure fuel pump
Crank angle sensor
Knock sensor Ignition coil
Coolant temperature
Battery voltage
The gasoline direct injection (GDI) injector is controlled by a Vemac driver which is
also coupled to the ECU so that to control different injection strategies (rail pressure
& injection timing). The ECU is controlled remotely by a Controller Area Network
(CAN). The CAN is connected with an ETAS 571.3 interface card and the card is
connected to the testbed PC. ETAS INCA V7.0 is used to communicate between the
testbed computer and the ECU. Some important inputs are controlled by the
dynamometer not ECU in order to prevent at the time testbed PC or ECU not
working properly, the engine could still be controlled. These inputs consist of ignition,
throttle position pedal, engine speed as well as the low-pressure fuel pump.
In this project, the exhaust lambda is close-loop controlled by MAHLE’s flexible

ECU. Boost pressure and exhaust back pressure (EBP) can also set to close loop
control. A map of intake manifold pressure (MAP) and engine speed dictate the
spark timing of the engine. The knocking indication is given by a knocking sensor
which is couple with an oscilloscope and the knocking indication has no effect on the
54
control of spark timing. Moreover, the knocking sensor is situated at the end of the
cylinder head so that its feedback cannot be accurate all the time.
3.4 Data Acquisition (DAQ) System and Instrumentation
A data acquisition system was built in order to display and record the important
parameter received from several sensors and devices on the engine testbed. In a
DAQ system there are two major components - DAQ hardware that collects
information from the sensors, actuators, and devices on the testbed, and DAQ
software that processes the signal, performs combustion analysis in real time, and
also logs the raw combustion data as necessary.
3.4.1 Data Acquisition Hardware
The in-cylinder pressure, intake absolute pressure, exhaust absolute pressure, fuel
flow rate, encoder clock, reference and torque channels are logged by a National
Instruments (NI) USB-6353 card. This card is used for high-speed data logging, it
has a maximum sampling speed of 1MS/s and can record up to 32 channels. The
DAQ hardware is shown in Figure 3-10.
55
Figure 3-9 High and low-speed DAQ cards
An Encoder Technology EB58204040 shaft encoder with a 0.25 CAD resolution is

used for the in-cylinder pressure analysis.
The signal from thermocouples, low-speed pressure sensors, coolant flow meters,
coolant pressure, exhaust port temperature, exhaust manifold temperature, oil
pressure, low-pressure fuel pump and average exhaust pressure are logged by
another NI USB 6210 card for low-speed data recording. This card’s sampling
frequency is 0.5 Hz and it has 16 channels for data recording in this project.
All the PRT’s data are recorded by an eDam-0915 acquisition card which can
support maximum of 7 PRT inputs. A serial connection is used between this
acquisition card and the testbed computer.
56
3.4.2 Data Acquisition Software (Combustion Analysis)
A Bespoke software called “Transient Combustion Analyser” receive all the data
transferred from the DAQ cards, this software was developed at Brunel University by
Dr. Yan Zhang [103]. The interface of this software is shown in Figure 3-11. All the
signal received from the DAQ cards are processed by this software and it also shows
information such as the temperature, pressure, flow rates, torque and engine speed.
What’s more, a lot of real-time calculations are also done and displayed by this
software such as engine power output, specific fuel consumption, combustion
characteristics and heat release rate are performed based on the engine
specifications. These data could also be recorded for hundreds of cycles for further
analysis. In this project, 300 consecutive cycles’ raw data was recorded for
combustion analysis. Heat release and combustion characteristics are analysed by
one-zone heat release analysis. In the one-zone model the cylinder contents are
considered as a single fluid - burned and unburned regions of gas in the combustion
chamber were treated as one and modelled as homogeneous mixture, and in-
cylinder pressure variations were related to energy released from the combustion of
the fuel. This model allows to include heat transfer and gas flow phenomena in a
simplified way [29,104].
57
Figure 3-10 Transient Combustion Analyser user interface [103]
The P-V diagram of a four-stroke spark ignition engine at both fully open throttle and
the part open throttle is shown in Figure 3-11. Equation 3-1 is used to calculate Net
Indicated Mean Effective Pressure (NIMEP) which is the work received by the piston
over one four-stroke cycle. (IMEP= area A- area B in Figure 3-11).
540
𝑃 ̇
𝑁𝐼𝑀𝐸𝑃 = ∫ 𝑉(𝜑)𝑑𝜑 (4 − 𝑠𝑡𝑟𝑜𝑘𝑒)
−180 𝑉𝑠
Equation 3-1
58
Figure 3-11 P-V diagrams of a four-stroke SI engine at full load [104]
What’s more, Gross Indicated Mean Effective Pressure (GIMEP) is calculated by

equation 3-2. The GIMEP correlates to the work delivered to the piston over the
compression and combustion strokes only (area A + area C in Figure 3-16).
180 𝑃 ̇
𝐺𝐼𝑀𝐸𝑃 = ∫−180 𝑉𝑠 𝑉(𝜑)𝑑𝜑
Equation 3-2
In those two equations, 𝑉𝑠 is the displacement volume of the engine, 𝑃 is real-time

in-cylinder pressure, 𝜑 is the crank angle degree and 𝑉̇ (𝜑) is the cylinder volume
correspond to that crank angle degree.
The calculation of Pumping Mean Effective Pressure is shown in equation 3-3. In

Figure 3-11, this parameter equal to the area of – (area B + area C).
59
𝑃𝑀𝐸𝑃 = 𝑁𝐼𝑀𝐸𝑃 − 𝐺𝐼𝑀𝐸𝑃
Equation 3-3
Net Indicated Fuel Conversion efficiency (IEffn), Gross Indicated Fuel Conversion
Efficiency (IEffg) and Pumping Indicated Fuel Conversion Efficiency (IEffp) are
calculated from NIMEP, GIMEP, PMEP and fuel mass flow (measured by fuel flow
meter). The calculation processes are shown in Figure 3-12.
Figure 3-12 Indicated work and fuel conversion efficiencies
According to equation 3-3 and Figure 3-12, the relationship between IEffn, IEffg and IEffp
can be written as equation 3-4, and it can be used for efficiency breakdown analysis.
𝐼𝐸𝑓𝑓𝑛 = 𝐼𝐸𝑓𝑓𝑔 + 𝐼𝐸𝑓𝑓𝑝
Equation 3-4
60
The in-cylinder pressure and knocking intensity which are updated on a cycle-to-
cycle basis are also shown on the Yantech software. The high-speed DAQ card
receives the in-cylinder pressure signal from the cylinder pressure transducer and
then transfers it to Yantech directly. The definition of Knocking intensity (KI) is the
difference between the real-time pressure measured by the in-cylinder pressure
transducer and the predicted pressure. Equation 3-5 shows the way how predicted
pressure is calculated. The predicted pressure of one point is the average pressure
of the 10 points before and 10 points after this point.
∑𝑛+2.5
𝑛−2.5 𝑃
𝑃𝑛 = 21
Equation 3-5
Figure 3-13 shows how the predicted pressure is calculated. The predicted pressure
Pn at crank angle n is calculated by averaging the pressure from crank angle n-2.5
to n+2.5 with a gap of 0.25 CAD.
Figure 3-13 Band-Pass filtering calculation [105]
61
The knocking intensity can be calculated from the difference between real-time
actual pressure (Pf, pressure feedback from the in-cylinder pressure transducer) and
the predicted pressure (Pn), this process is shown in equation 3-6.
𝐾𝐼 = 𝑃𝑓 − 𝑃𝑛
Equation 3-6
3.4.3 Finding the Top Dead Centre
In this project, a lot of work has been done in order to increase the accuracy of the
data. Since the in-cylinder pressure raw data is used for the calculation of indicated
efficiency and combustion analysis, precise determination of the Top Dead centre
(TDC) is required to achieve the accurate combustion analysis. A Kistler 2629C
capacitive type TDC sensor is used to find the correct TDC position. This sensor’s
accuracy is 0.1 CAD but because the crank encoder’s resolution is 0.25 CAD, the
actual accuracy of the TDC sensor is 0.125 CAD. The toolbox of the TDC sensor is
shown in Figure 3-15.
62
Figure 3-14 Kistler TDC sensor toolkit
Figure 3-17 shows the graphs generated by the TDC sensor. The mechanical TDC
position is identified by the peak signal output in the figure. In Figure 3-16 a, there’s
a 20.1 CAD offset between the peak signal (TDC position) and zero crank angle
degree. Figure 3-16 b shows that after inputting 20 CAD offset into the combustion
analyser, the 0 CAD is now matched to the mechanical TDC position.
Figure 3-15 TDC determination in the combustion analyser
A motoring test is done after finding the correct mechanical TDC position. The
engine speed is 1200 rpm and the log P- log V diagram is shown in Figure 3-17 a.
From Figure 3-1 b, it is shown that there’s a less than1 CAD offset between the peak
in-cylinder pressure and the TDC points due to the heat loss [106].
63
Figure 3-16 Log P-Log V diagram of engine motoring at 1200 RPM
3.4.4 In-cylinder Pressure Pegging
The piezoelectric pressure sensor is used to measure the in-cylinder pressure

cannot provide absolute pressure. This sensor outputs a charge proportional to the
pressure change. As a result of this, the transducer’s output in the cycle needs to be
referenced or pegged to a known pressure to calculate absolute pressure [106].
Cylinder pressure pegging can be achieved by several methods and are summarized
by Randolph [107].
In this project, a revised method is used to pegging the in-cylinder pressure. The
intake and exhaust pressure measured by piezoresistive absolute pressure sensors
are used as reference pressure and it is shown in Figure 3-18.
64
Figure 3-17 In-cylinder pressure pegging example
The Yantech combustion analyser allows the user to input the “pegging” value which
means manually selecting a point on the exhaust pressure curve for in-cylinder
pressure to be pegged to. By changing the “pegging” value, the in-cylinder
pressure’s curve is moved upwards or downwards. After changing the pegging value
at each test point, the in-cylinder pressure should match the exhaust pressure after
exhaust valve close (EVC) and also match intake pressure before intake valve close
(IVC). If the in-cylinder pressure cannot match either the intake pressure or exhaust
pressure, intake pressure has a higher priority than exhaust.
3.5 Exhaust Emission Measurements
3.5.1 Emission Analysers
Several emission analysers are used to measure gaseous emissions in this project.
A Horiba MEXA-554JE analyser is used to measure CO, CO2 and O2, a Signal
Ambitech model 443 Chemiluminescent analyser is used to measure the NOx and
another Signal Rotork Analysis model 523 FID analyser is used to measure the HC.
A combustion DMS 500 fast response particulate analyser is used to measure the
particulate number and spectral size density. All the emission analysers are shown in
Figure 3-19, Figure 3-20 and Figure 3-21. The specification of those emission
analysers are shown in Tables 3.4, 3.5 and 3.6.
65
Figure 3-18 Horiba MEXA-554JE for CO, CO2, and O2 measurements
66
Figure 3-19 Signal Rotork Analysis model 523 FID analyser used for HC measurements
(first unit at the top) and Signal Ambitech model 443 Chemiluminescent analysers used
for NO/NOx measurements (bottom 3 units)
Table 3-4 Rotork Analysis model 523 FID analyser (HC) specifications
67
Table 3-5 Signal Ambitech model 443 Chemiluminescent analyser (NOx) specifications
Figure 3-20 Cambustion DMS 500 fast response particulate analyzer
Table 3-6 Specifications of the Emission analysers
Min
detectable
Device Repeatability Linearity Sensitivity
concentration/
range
NOx analyser ±1% FS ±1% FS 0.2 ppm
HC analyser ±1% FS ±1% FS 0.1 ppm
1.0 ×103
DMS 500 5 nm- 1µm
(dN/dlogDp/cc)
68
3.5.2 Calculation of specific emission
The CO, HC and NOx are recorded from the analyser as volume concentration in
parts per million(ppm) and they should be converted to indicate specific emissions
following the UN regulation number 49. The air humidity and water-in-fuel content
are taken into account in order to obtain wet basis. But the NOx’s humidity and
temperature correction are not applied in this project.
The main toxic exhaust gases CO, NOx and HC are transferred from ppm to g/kWh
by Equations 3-7, 3-8 and 3-9.
𝑢𝑐𝑜 [𝐶𝑂]𝑘𝑤 𝑚̇𝑒𝑥ℎ

𝐼𝑆𝐶𝑂 = 𝑃𝑖
Equation 3-7
𝑢𝑁𝑂𝑥 [𝑁𝑂𝑥 ]𝑘𝑤 𝑚̇𝑒𝑥ℎ

𝐼𝑆𝑁𝑂𝑥 = 𝑃𝑖
Equation 3-8
𝑢𝐻𝐶 [𝐻𝐶]𝑘𝑤 𝑚̇𝑒𝑥ℎ

𝑆𝐻𝐶 =
𝑃𝑖
Equation 3-9
Where 𝑢𝑖 is the molar mass fraction of each gas, [𝑖] is the gas concentration in ppm,
𝑘𝑤 is the dry-to-wet correction factor, 𝑚̇𝑒𝑥ℎ is the exhaust mass flow rate and 𝑃𝑖 is the
indicated power. These three parameters can be written as follows
𝑚̇𝑒𝑥ℎ = 𝑚̇𝑎𝑖𝑟 + 𝑚̇𝑓𝑢𝑒𝑙
Equation 3-10
𝑚̇𝑎𝑖𝑟 = 𝑚̇𝑑𝑟𝑦 𝑎𝑖𝑟 + 𝑚̇ℎ𝑢𝑚𝑖𝑑𝑖𝑡𝑦
Equation 3-11
69
𝑚̇𝑓𝑢𝑒𝑙 = 𝑚̇𝑎𝑐𝑡𝑢𝑎𝑙 𝑓𝑢𝑒𝑙 + 𝑚̇𝑤𝑎𝑡𝑒𝑟
Equation 3-12
The water saturation pressure polynomial estimation proposed by [108] is used to

calculate the dry air and water (present in air)
𝑆𝑃 = 604.8346 + 45.9058(𝑇𝑎 − 273.15) + 1.2444(𝑇𝑎 − 273.15)2

+ 0.03522481(𝑇𝑎 − 273.15)3 + 0.00009322091(𝑇𝑎 − 273.15)4
+ 0.000004181281(𝑇𝑎 − 273.15)5
Equation 3-13
In which 𝑇𝑎 is the ambient temperature. Air humidity 𝐻𝑎 can be calculated with RH

(relative humidity) and 𝑝𝑎 (ambient pressure).
6.211 𝑅𝐻 𝑆𝑃
𝐻𝑎 =
(𝑅𝐻 𝑆𝑃)
𝑝𝑎 − 100
Equation 3-14
𝑚̇𝑎𝑖𝑟
𝑚̇𝑑𝑟𝑦 𝑎𝑖𝑟 =
1 + 𝐻𝑎
Equation 3-15
𝑚̇ℎ𝑢𝑚𝑖𝑑𝑖𝑡𝑦 = 𝑚̇𝑑𝑟𝑦 𝑎𝑖𝑟 𝐻𝑎
Equation 3-16
The term 𝑢i is the tabulated ratio between the component and exhaust gas density,
which varies according to the fuel used as depicted in Table 3.7.
70
Table 3-7 Raw gas molar mass fraction of the exhaust gases for gasoline
Exhaust Gas 𝝁𝒊 (Gasoline)
CO 0.000966
NOx 0.001587
THC 0.000499
The ambient condition, as well as the added water content from the fuel, may affect
the dry-to-wet correction factor 𝑘𝑤 which applied to CO and NOx, the calculation is
shown in equation 3-17.
𝑚̇𝑓𝑢𝑒𝑙
1.2442𝐻𝑡 + 111.19𝑊𝐴𝐿𝐹 (𝑚̇ )
𝑑𝑟𝑦 𝑎𝑖𝑟
𝑘𝑤 = 1.008(1 − )
𝑚̇𝑓𝑢𝑒𝑙
773.4 + 1.2442𝐻𝑡 + 1000(𝑚̇ )𝑘𝑓
𝑑𝑟𝑦 𝑎𝑖𝑟
Equation3-17
𝑘𝑓 = 0.05559𝑊𝐴𝐿𝐹 + 0.0070046𝑊𝐸𝑃𝑆
Equation3-18
𝑊𝐴𝐿𝐹 stands for the hydrogen content in the fuel while 𝑊𝐸𝑃𝑆 stands for the oxygen
content. The total humidity factor 𝐻𝑡 is used to replace the original air humidity factor
to take into account the water-in-fuel added additional content.
𝐻𝑡 = 𝐻𝑎 + 𝐻𝑓
Equation3-19
𝐻𝑓 = 𝑚̇𝑤𝑎𝑡𝑒𝑟 /𝑚̇𝑑𝑟𝑦 𝑎𝑖𝑟
Equation3-20
71
Table 3-8 Properties of the fuels used for engine experiment
Eu6 Gasoline
Ethanol Methanol
RON95 E10
Chemical Formula C5-10H12-22 C2H5OH CH3OH
R.O.N 95.7 107 108
M.ON 86 89.5 88.7
Purity [%] n/a 99.7 99.8
LHV (MJ/kg) 41.25 26.8 19.8
Hydrogen/Carbon Ratio 1.9 3 4
Oxygen/Carbon Ratio 0.0317 0.5 1
Stoichiometric AFR 13.9 9 6.5
Density [kg/L] 0.75 0.79 0.79
72
3.6 Testing and Data accuracy
The operating parameters and boundary conditions of the engine test are shown in
Table 3.9.
Table 3-9 Engine operation parameters and boundary conditions
Variables Control criteria
Engine load (NIMEP) Controlled by throttle position

(dynamometer) and boost pressure
(ECU)
Engine speed Controlled by the dynamometer (0-5400

rpm)
Intake cam timing (MOP) Controlled by ECU
Exhaust cam timing (MOP) Controlled by ECU
Exhaust lambda Controlled by ECU (set as 1 unless

lambda sweep or over-fuelling)
High pressure fuel rail Controlled by ECU
Low pressure fuel rail 5 bar, controlled by pressure regulator
Injection strategy Controlled by ECU
Spark timing Controlled by ECU (Optimized by heat

release rate at low and mid load,
optimized by knock limited at high load)
Air humidity Dry air, humidity<3%
Boost air temperature 40± 3°C
73
Coolant temperature 90± 3°C
Oil Temperature 90± 3°C
Low pressure fuel temperature 20± 3°C
Combustion stability NIMEP_COV ≤ 3% (Except for cold

spark sweep, NIMEP_COV ≤ 15%)
Fuel type E10 Gasoline(baseline), Ethanol,

Methanol
Exhaust back pressure A butterfly valve is used for adjustment,

Full opening represents the ambient
back pressure
The quality of the recorded data has been carefully scrutinized in order to obtain the
most accurate and consistent results possible. Before each testing, the following
steps were taken to ensure accuracy and consistency. Before beginning the
experimental work, the sensors, analysers, and other measurement devices were
calibrated thoroughly, including flow meters, pressure sensors, temperature sensors,
shaft encoders, and emission analysers. The calibration of these devices was also
checked regularly so that adjustments could be made as necessary.
3.6.1 Cam Timings Validation
According to Figure 3-17, the pressure curves are used to validate the cam timings
and valve timings as the precise control of cam timings are important for the
experiments. Following the identification of valve timing on intake, exhaust and
cylinder pressure curves, the controls are validated. Every time after the engine was
rebuilt, this validation was performed.
74
3.6.2 In-Cylinder Pressure and Fuel Flow Measurements Validation
Since precise measurements of fuel flow and in-cylinder pressure are critical to the
success of this project, this section describes the steps taken to ensure high levels of
accuracy. DN01 1/24" sensor size was carefully selected for the Coriolis mass flow
meter on this single cylinder engine. It measures a maximum flow rate of 20 kg/h and
has a very small error of less than 0.1% in the range of flow rates of 1 kg/h to 20 kg/h
(Figure 3-22). Generally, the error increases as the flow rate decrease especially
when the flow rate is lower than 1kg/h, for example, when the flow rate is as low as
0.2 kg/h, the error might be less than 0.5 %. Before starting the engine testing, the
flow meter was calibrated in the factory with the manufacturer and checked again in
the university's laboratory.
Figure 3-21 Measured error of Endress+Hauser Promass 83A01 flow meter
The Kistler pressure transducers were checked and calibrated, as necessary, on a

dead weight tester in the laboratory, in order to minimize the error in measuring the
indicated pressure. What’s more, the Kistler intake and exhaust absolute pressure
transducers and their amplifiers are checked and do zero adjustment every time
before testing according to each day’s local ambient pressure. In order to ensure
75
accurate readings, the exhaust pressure transducer was cleaned regularly in an
ultrasonic bath in order to prevent the accumulation of carbon deposits. In addition, a
motoring test is also conducted every time before starting the main experiments in
order to check the health of both the engine and the measurement system (Table
3.10). With this test, the peak in-cylinder pressure (Pmax) and crank angle of Pmax
(heat loss angle) can be determined in order to check the consistency. Besides the
above measurements, the accuracy and consistency of all the other measurements
are examined as well. Prior to each test, the emissions analysers were calibrated
using pure air as the zero gas and the specific span gas.
3.6.3 Daily Engine Check Points
Daily motoring and firing tests were conducted and the results were recorded for
further analysis before the experimental data were taken. As a first step, a zero log is
recorded after daily calibrations are completed to ensure the baseline measurements
in the correct range. After that, a daily motoring test is conducted and the results are
recorded. Motoring tests are conducted after the engine has warmed up to the desired
temperature before an analysis of the indicated measurements is undertaken. Finally,
a firing test is conducted under a fixed operating condition for checking the engine's
health and ensuring that the overall testing system is consistent. Table 3.10 shows the
operating details for the motoring check and daily firing check.
76
Table 3-10 Motoring and firing daily checks settings
Variables Motoring Check Daily Firing Check
Engine Speed 1200 rpm 2000 rpm
Engine Load Throttle fully open NIMEP=4.6 bar
Intake and exhaust cam

Minimum valve overlap Minimum valve overlap
timing (MOP)
Exhaust Lambda 1
Rail pressure 100 bar
Single early injection 320

Injection Timing
CAD BTDCf
Spark timing CA50= 8 CAD ATDCf
Air humidity Dried air (humidity≤ 3%) Dried air (humidity≤ 3%)
Boost air temperature 40±3°C (pre-throttle) 40±3°C (pre-throttle)
Coolant temperature 90±3°C 90±3°C
Oil temperature 90±3°C 90±3°C
Low pressure fuel

20±3°C
temperature
Exhaust back pressure Ambient EBP (0%)
Figure3-23 illustrates an example of the results of a daily motoring test. In

preparation for the main tests, the peak in-cylinder pressure during engine operation
as well as the angle of peak in-cylinder pressure are monitored every time.
77
Figure 3-22 Peak cylinder pressure and its angle recorded for daily motoring checks (the
x-axis shows the test days)
3.7 Summary
The single cylinder engine and all other testbed facilities which include DI fuelling
system, intake/ exhaust system, coolant system, oil system, conditioning system as
well as emission analysers are presented in this chapter. Also, the research
methodology and the main equations used for combustion analyser software and
emission are shown and discussed. What’s more, the way how to increase the
accuracy and consistency of the data are also discussed in this chapter.
78
4. Chapter Four: Spark Ignition Engine Operations with Ethanol and Methanol
4.1 Introduction
This chapter presents the spark ignition engine operations with ethanol and
methanol and explains the difference between alcohol fuel (ethanol& methanol) and
baseline (E10 gasoline) on the combustion performance, fuel efficiency and
emissions. The test was carried out on a single-cylinder direct injection spark ignition
engine at selected load points at 2000rpm, 3000rpm and 4000rpm. The overall air/
fuel ratio was maintained at the stoichiometric condition with the help of closed-loop
exhaust lambda control at low and medium loads. At high loads, over-fuelling was
used to keep the exhaust gas temperature under the maximum limit and avoid
knocking combustion.
4.2 Results and Discussion
4.2.1 2000rpm Load Sweep
At 2000rpm, the engine load was varied from 2bar to 11bar IMEPs in order to
explore the difference between these three fuels at low speed. For these three fuels,
the same injection timing (i.e. the start of injection) was used for each load. It was
fixed at 320 Deg BTDCf from 2bar to 6.9bar, then it was retarded to 318 Deg BTDCf
at 8.9bar and 312 Deg BTDCf at 11bar. The rail pressure was also kept the same for
these fuels and the rail pressure increase from 55bar to 145bar with the increase of
load. With the load increase, more fuel was injected into the combustion chamber
since the combustion is stoichiometric. Gasoline’s fuel flow rate was increased from
0.44kg/h to 1.75kg/h, the flow rate of ethanol was increased from 0.62kg/h to
2.49kg/h and that of methanol was increased from 0.82kg/h to 3.26kg/h. The
gasoline baseline boundary conditions (injection timing, rail pressure) were kept the
same with the data from MAHLE’s DI3 engine. Due to the lower heating value of
79
methanol is lower than ethanol than gasoline, that’s the reason the fuel flow rate is
the highest of them.
The spark timing, brake thermal efficiency, CA50, combustion duration as well as
knocking intensity are shown in Figure 4-1.
Threshold: (y = 10)
Threshold: (y = 8)
Threshold: (y = 8):Polygon
baseline_2000rpm load swipe.nc (NIMEP,CA50)
Brunel Single Cylinder - 2000rpm Load Sweep - Gasoline_Ethanol_Methanol methanol_2000rpm load sweep_Sheet1.nc (NIMEP,CA50)
ethanol_2000rpm load sweep_Sheet1.nc (NIMEP,CA50)
2100 10.0
50MBFav0 [°CA ATDCF]

9.5
2050
Speed [rpm]
9.0
2000 8.5
8.0
1950
7.5
1900 7.0
24 22
22 21
Spark Timing [°BTDC]
10-90 Burn Duration
20 20
19
18
18
16
17
14 16
12 15
10 14
40 0.45
0.40
KnockingIntensity [bar]
35
Eng_EffTherma [%]
0.35
KPmaxav0 [bar]
30 0.30
0.25
25 0.20
0.15
20
0.10
15 0.05
0 5 10 15 0 5 10 15
NMEPav0 [bar] NMEPav0 [bar]
Figure 4-1 Combustion characteristics of 2000rpm load sweep
For all three fuels, the MBT (Minimum ignition advance for Best Torque) spark
timings were achieved and retarded with increasing load. As shown in Figure 4-1, all
test point’s CA 50s are around 8 Deg ATDCf at MBT timing. Since the CA50 is
controlled at 8 Deg, the burn duration becomes shorter with the increasing load as
the flame speed increases. The spark timing of these fuels advanced from 18.2 Deg
BTDC (methanol), 19.7 Deg BTDC (ethanol), 22.3 Deg BTDC (gasoline) to 11.8 Deg
BTDC as a result of this. The laminar flame speed of methanol is faster than ethanol
and gasoline and the CA50 is controlled at 8 Deg so that the spark timing of
80
methanol is more advanced than ethanol and gasoline at 2bar and 4bar. This
phenomenon can also be found in CA10, and CA10-50 (shown in Figure 4-2). The
brake thermal efficiency (BTE) of the three fuels increases with the load. The
maximum BTE of them are all at 11bar (gasoline 34.33%, ethanol 35.86%, methanol
36.17%). The BTE of methanol and ethanol was increased by 4.5% and 5.5%
compared to gasoline. The knocking intensity becomes higher with load increase
since the in-cylinder pressure increases result in bigger difference between real time
in-cylinder pressure and the predicted in-cylinder pressure, but there wasn’t any
knocking happen during the 2000rpm load sweep. The highest KI for these three
fuels are at 11bar IMEP but all of them are lower than 1 which is the KI limit.
baseline_2000rpm load swipe.nc (NIMEP,APmax)

Brunel Single Cylinder - 2000rpm Load Sweep - Gasoline_Ethanol_Methanol methanol_2000rpm load sweep_Sheet1.nc (NIMEP,APmax)
ethanol_2000rpm load sweep_Sheet1.nc (NIMEP,APmax)}
20 15.6
19 15.4
Spk-10% Burn Duration
APmax [°CA ATDCF]
18 15.2
17 15.0
16 14.8
15 14.6
14 14.4
13 14.2
12 14.0
11 13.8
11.0 140
10.5 120
10-50% Burn Duration
10.0 100
Pmax [bar]
9.5 80
9.0 60
8.5 40
8.0 20
7.5 0
10.5 4.0
10.0 3.5
Rmax [bar/dCA]
9.5 3.0
9.0 2.5
8.5 2.0
8.0 1.5
7.5 1.0
7.0 0.5
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Figure 4-2 Combustion phasing, Pmax and Rmax
The combustion phasing results are shown in Figure 4-2. CA10, CA10-50, and
CA50-90 all decrease with the increase of load because of the faster flame speed at
elevated temperature. The higher laminar flame speed of methanol causes shorter
CA10 and CA10-50 than methanol and gasoline. At 11bar IMEP, the CA10-50 of
81
gasoline is shorter as a result of the CA50 of gasoline is more advanced than
methanol and ethanol. The peak in-cylinder pressure crank angle (Pmax) for these
fuels are around 14.6 Deg ATDCf. The peak cylinder pressure of gasoline, methanol
and ethanol increases from 15bar to 60 bar linearly with the load. The maximum rate
of pressure rise (Rmax) of these fuels increase similarly from 0.76bar/dCA to
2.5bar/dCA with the load increase from 2bar to 8.9bar. At 11.2bar, the Rmax of
gasoline is 3.82bar/dCA which is higher than methanol’s Rmax (3.56bar/dCA) and
ethanol's (3.23bar/dCA) because the burn duration of gasoline at 11.2bar is shorter
than methanol and ethanol, with the same in-cylinder peak pressure.

1.0002 320
Fuel Injection Start [°BTDCf]
1.0000 310
0.9998 300
Lambda [#]
0.9996 290
0.9994 280
0.9992 270
0.9990 260
1000
150
140
950 980°C Temp Limit 130
Conventional Multi Cyl Limit
120
pRail[bar]
900 110
100
850 90
80
ExhaustTemp [degC]
800 70
60
750 780°C Temp Limit 50
Revised Single Cylinder Limit 3.5
700 3.0
Fuel Flow kgh [kg/h]
2.5
650
2.0
600 1.5
1.0
550
0.5
500 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
NIMEP [bar] NMEPav0 [bar]
Figure 4-3 Lambda, TExh and injection parameters
From Figure 4-3, it’s easy to find that all the test points’ lambda which was controlled
by the engine’s ECU is kept very close to 1. The exhaust temperature increases with
load from 515°C (gasoline), 508°C (methanol), 508°C (ethanol) to 589°C (gasoline),
575°C (methanol) 602°C (ethanol), respectively. Since all these test points knocking
intensity is under 1 so spark retarded is not applied. From 2bar to 8.9bar, the
82
exhaust gas temperature of ethanol and methanol is lower than gasoline since their
evaporation enthalpy is higher than gasoline, charge cooling effect causes the in-
cylinder temperature decrease thus the exhaust gas temperature of ethanol and
methanol are lower than gasoline. At 11.2bar, the exhaust gas temperature of
ethanol is higher than the other two fuels since its burn duration is longer than
gasoline and methanol. The injection start, rail pressure and fuel flow rate of
gasoline, ethanol and methanol are also shown in Figure 4-3. Since Methanol’s
lower heating value is lower than ethanol and gasoline, methanol has the most
retarded injection end timing and the highest fuel flow rate. The dashed line shown in
Figure 4-3 shows the injection end timing of gasoline, ethanol and methanol.

14.4 1200
14.2 1100
14.0 1000
13.8 900
CO2 [%Vol]
THC [ppm]
CO2 [%]
800
13.6
700
13.4
600
13.2 500
13.0 400
12.8 300
12.6 200
0.90 1.4
0.85 1.3
0.80 1.2
CO (High) [%]
CO [%Vol]
O2 [%]
0.75 1.1
0.70 1.0
0.65 0.9
0.60 0.8
0.55 0.7
2200
2000
1800
1600
NOx [ppm]
1400
1200
1000
800
600
400
200
1 2 3 4 5 6 7 8 9 10 11 12
NMEPav0 [bar]
Figure 4-4 Emissions of 2000rpm load sweep
The exhaust data are shown in Figure 4-4. The CO2 released by gasoline increases
from 14.02%Vol to 14.22%vol with the load increase from 2bar to 4.6bar then keep
the same at 6.9bar. It drops from 14.2%vol to 13.84%vol with the load increase to
11bar. The CO2 released by ethanol increases from 13.12%vol to 13.48% from 2bar
to 4.6bar then decreases to 13.32%vol at 6bar and increases to 13.66%vol from
6.9bar to 11bar. There is a slight drop of the CO2 from methanol, it drops from
83
13%vol to 12.76%vol from 2bar to 11bar. When converted to energy-specific
emissions in (gCO2/MJ), methanol leads to 7% lower specific CO2 emissions than
gasoline.
The amount of CO varies only slightly since the lambda of all these test points was kept
at 1. The CO emission by gasoline decreases from 0.71%vol to 0.62%vol firstly, then it
increases to 0.89%vol from 4.6bar to 11bar. For ethanol, its CO drops from 0.65%vol to
0.62%vol from 2bar to 4bar, then it increases to 0.74%vol at 6.9bar then continue drops
to 0.56%vol from 6.9bar to 11bar. The CO released by methanol has the same trends
as ethanol. It drops from 0.71%vol to 0.62%vol then increase to 0.78%vol from 2bar to
6.9bar. After that, it decreases to 0.63%vol from 6.9bar to 11bar. The amount of CO in
the exhaust gas is mainly affected by Lambda (shown in Figure 4-3).
The NOx released by gasoline, ethanol and methanol increases with the increase of
load as the combustion temperature increases. For the NOx released by gasoline, it
increases from 882ppm to 1768ppm with the increase in load. The NOx from ethanol
increases from 463ppm to 2103ppm and that of methanol increase from 262ppm to
1780ppm.
THC decreased slightly with the load as further oxidation can take place in the hotter
exhaust gas and the THC released by methanol and ethanol is lower than gasoline.
84
100 135
130
pBoostCCP:1 [kPa abs]

90
125
pManifold [kPaA]
80
120
70 115
110 Higher boost press pre throttle
60
105
50
100
40 95
94 44
43
Plenum Air Temp [°C]

42
92 41
40
39
38
90 37
CoolantTemperature degC
36
35
88 34
33
95
86
OilTemperature [degC]
90
84
82 85
80 80
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
NMEP [Bar] NMEP [Bar]
Figure 4-5 Engine conditioning pressure and temperature
Most boundary conditions of this test are shown in Figure 4-5. The manifold pressure
of different fuels has small difference with each other, it increases from 46kPa to
97kPa as the load is increased. The coolant and oil temperature are kept around
90°C. The Intake air temperature are kept around 40°C.
85
90000000 90000000
80000000 80000000
70000000 70000000
60000000 60000000
total PN
total PN
50000000 50000000
40000000 40000000
30000000 30000000
20000000 20000000
10000000 10000000
0 0
80000000 90000000
70000000 80000000
60000000 70000000
PN 10-1000nm
PN 10-1000nm
60000000
50000000
50000000
40000000
40000000
30000000
30000000
20000000 20000000
10000000 10000000
0 0
16000000 90000000
14000000 80000000
12000000 70000000
PN 23-1000nm
PN 23-1000nm
60000000
10000000
50000000
8000000
40000000
6000000
30000000
4000000 20000000
2000000 10000000
0 0
0 2 4 6 8 10 12 0.0 2.0 4.0 6.0 8.0 10.0 12.0
Figure 4-6 Particle numbers of methanol and Ethanol
Due to the equipment breakdown, the particle number’s data of gasoline was not
collected. Figure 4-6 shows the total particle numbers, PN 23-1000nm (Euro 5
regulation) and PN 10-1000nm (Euro 6 regulation) for ethanol and methanol
operations. The total particle number of ethanol drops from 8.6e7 to 5e7 with the
increase of load while the total PN of methanol decreases slower than ethanol,
dropping from 5.2e7 to 5e7. The number of particles from 10-1000nm of ethanol
drops from 7e7 to 2.4e7 and methanol drops from 3.8e7 to 2.7e7.
The particle number is mainly affected by injection timing and rail pressure which are
shown in Figure 4-3. From 2bar to 11bar IMEP, the start of injection is retarded from
320 Deg BTDCf to 310 Deg BTDCf. Early injection could lead to fuel impingement on
combustion chamber surfaces and hence pool fires—a source of high levels of PM
emissions [109]. The rail pressure was increased from 55 bar to 90 bar and increased
fuel injection pressure enhances the air-fuel mixing; at higher injection pressures,
fuel droplets become smaller which leads to better evaporation. The boiling point of
methanol (64.7°C) is lower than that of ethanol (78.4°C) so that methanol could lead
to less liquid impingement on the piston, and that’s why the particle numbers of
methanol is lower than ethanol at the same injection timing and rail pressure. [109]
86
Ethanol’s PN drops quickly at 6.9bar since the spark timing is slightly advanced at
this point.
Figure 4-7 2000rpm load sweep particle size spectral size density
Figure 4-7 shows the particle size spectral density distribution. It can be seen that
the size spectral density curve of each test point is peaked around 10-20nm, which
means that most particles are in nucleation mode and there are limited soot
emissions in accumulation mode. The particle released by ethanol and methanol’s
size is around 5nm to 60 nm.
87
3000rpm load sweep for gasoline, ethanol and methanol were done from 2bar to
28bar IMEP in order to explore the difference between these three fuels at a medium
speed and higher load condition. The maximum load of 28bar was imposed by the
severe knocking combustion of gasoline. For ethanol and methanol, longer
combustion duration caused misfire after 28bar which also couldn’t be controlled.
baseline_3000rpm load swipe.nc (NIMEP,BSFC_Corr42LHV)

Brunel Single Cylinder - 3000rpm Load Sweep ethanol_3000rpm processed data_3000rpm.nc (NIMEP,BSFC_Corr42LHV)
Alternative Fuels Comparison methanol_3000rpm processed data_3000rpm.nc (NIMEP,BSFC_Corr42LHV)}
650 35
600 50MBFav0 [°CA ATDCF]
30
550
BSFC_Corr42LHV
500 25
450
20
400
350 15
300
10
250
200 5
25 24
23
20
Spark Timing [°BTDC]
10-90 Burn Duration
22
15 21
20
10
19
5 18
17
0
16
-5 15
40 0.8
0.7
35
Eng_EffTherma [%]
0.6
KPmaxav0 [bar]
30
0.5
25 0.4
0.3
20
0.2
15
0.1
10 0.0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Figure 4-8 Combustion characteristic of 3000rpm load sweep
For 3000 rpm load sweep, most test points were opearated with the MBT timing set
by means of CA50 at 8 Deg ATDCf. For the test points where knocking intensity was
greater than 1, the spark timing was retarded. The highest BSFC of gasoline is
577.7g/kWh at 2bar then it decreases to 234g/kWh at 10bar IMEP. after that,
gasoline’s BSFC increases to 277 g/kWh at 28bar IMEP. The BSFC trends of
ethanol and methanol are similar to the gasoline one. The BSFC of ethanol
88
decreases from 517 g/kWh at 2bar IMEP to 227 g/kWh at 18bar IMEP then increase
to 247 g/kWh at 28bar IMEP. For methanol, its BSFC drops from 636 g/kWh at 2bar
IMEP to 226 g/kWh at 18bar IMEP and 20bar IMEP then increase to 243 g/kWh at
28bar IMEP. The spark timing of gasoline is retarded slowly from 24.9Deg BTDC to
11.1 Deg BTDC at 16bar IMEP. But after 16bar, the spark timing of gasoline was
retarded from 11.1 Deg BTDC to -2.5 Deg BTDC quickly because of knocking
intensity was too high which is also shown in Figure 4-8. For methanol and ethanol,
their spark timings are also retarded with load from 2bar to 16bar IMEP. After 16bar
IMEP, the spark timing is delayed to keep the Pmax below 120 bar limit and the
Rmax (pressure rise rate) at 6. The knocking intensities for ethanol and methanol are
always under limit even at higher load because their anti-knock characteristic is far
better than gasoline. The brake thermal efficiency (BTE) for these fuels could also be
found in Figure 4-8. The BTE of gasoline increase from 14.83% to 36.5% at 10bar
IMEP then decreases to 30.9% from 10bar IMEP to 28bar because combustion
duration increases. The BTE of ethanol increases from 16.5% to 37.7% at 18bar
IMEP then decrease to 34.6% at 28bar IMEP because of the engine’s limit. The BTE
of methanol has the same trends as ethanol, it increases from 13.46% to 37.9% at
18bar IMEP and 20bar IMEP then decrease to 35.3%. The maximum BTE of ethanol
and methanol increase by 3.2% and 3.8% compared with gasoline. The CA50 is also
shown in Figure 4-8. Before 16bar IMEP, the CA50 for gasoline, ethanol and
methanol are kept at 8 Deg ATDC. After 16bar, the CA50 of gasoline is retarded to
32 Deg ATDC at 28bar IMEP due to knocking, and the CA50 of ethanol and
methanol are retarded to 17.3 Deg ATDC and 17.8 Deg ATDC because of engine
limit. The combustion duration of gasoline, ethanol and methanol are shown in
Figure 4-8. The combustion duration of gasoline decreases from 22.2 crank angle
degree (CAD) to 16.8 CAD at 16bar IMEP then increase to 23.1 CAD at 28bar IMEP.
At 6bar IMEP and 8bar IMEP, their CA50 are slightly retarded which cause their
combustion duration longer. The combustion duration of ethanol decreases from 21
CAD to 16.4 CAD then increases to 19.3 CAD while that of methanol decreases from
21.4 CAD to 15.7 CAD then increases to 18.1 CAD.
Overall, the combustion duration of methanol is shorter than ethanol and gasoline.
Below 16bar IMEP, gasoline has the longest combustion duration. After 16bar IMEP,
gasoline’s combustion duration increases fast as a result of knocking. The
89
combustion duration of ethanol and methanol also increase because of the retarded
spark timing to keep the peak in-cylinder pressure and maximum pressure rise rate
limits of this engine. The knocking intensity (KI) is shown in the last diagram of
Figure 4-8. The knocking intensity of gasoline increase from 0.08 to 0.76 with the
load increase from 2 to 16bar IMEP, then KI keeps stable at around 0.55. Ethanol’s
KI increase from 0.08 to 0.79 at 22, 24, and 26bar then drops to 0.71. In addition,
methanol’s KI increases from 0.07 to 0.76 at 24bar IMEP then decreases to 0.74.

methanol_3000rpm processed data_3000rpm.nc (NIMEP,BSFC_Corr42LHV)}
Alternative Fuels Comparison
24 40
22 APmax [°CA ATDCF] 35
20 30
18 25
16 20
14 15
12 10
13.5 120
13.0
100
12.5
12.0
80
Pmax [bar]
11.5
11.0
60
10.5
10.0 40
9.5
9.0 20
8.5
8.0 0
11.0 6
10.5
5
10.0
Rmax [bar/dCA]
4
9.5
9.0 3
8.5
2
8.0
1
7.5
7.0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
The different combustion phasings are shown in Figure 4-9. From 2bar to 16bar
IMEP, the duration between the spark timing and CA10 of gasoline decreases from
22.5 CAD to 12 CAD as the flame speed increases with higher temperature at higher
load. After that, the CA10 increases from 12CAD to 16.1 CAD because the spark
timing is retarded to avoid knocking. Ethanol and methanol’s CA10 decrease from
19.5CAD to 12.5 CAD then increase to 14.7 CAD. The CA10-50 of gasoline
decreases from 11.2CAD to 8.7CAD at 14bar and 16bar IMEP, then increase to
90
13.2CAD between 16bar and 28bar IMEP. The CA10-50 of ethanol drops from
10.5CAD to 8.7CAD from 2bar to 16bar IMEP, then it increases to 10.3CAD with the
load increase to 28bar IMEP. Methanol’s CA10-50 has similar trends with ethanol, it
decreases from 10.5CAD to 8.5CAD then increases to 9.8CAD. CA50-90 is also
shown in Figure 4-9. Gasoline’s CA50-90 drops from 11CAD to 9.6CAD at 4bar
IMEP then increase to 10.3CAD from 4-8bar IMEP. After that, it decreases from
10.3CAD to 8CAD from 8bar to 16bar IMEP. In addition, it keeps around 9.7CAD
during the load from 18bar to 28bar IMEP. Between 2 to 6bar IMEP, the CA50-90 of
ethanol decreases from 10.5CAD to 8.7CAD and then slightly increases to 9.0CAD.
After that, it drops to 7.9CAD at 18bar IMEP then increases to 9.0CAD at 28bar
IMEP. Methanol’s 50%-90% burn duration decreases from 10.9CAD to 12.4CAD at
14bar then increases to 8.3CAD at 28bar. Before 16bar, the CA10, CA10-50, and
CA50-90 of ethanol and methanol are faster than gasoline since alcohol fuel’s burn
speed is faster than gasoline. After 16bar, as a result of knocking, gasoline needs to
retard the spark timing more than alcohol fuel and got longer combustion duration.
The maximum in-cylinder pressure correspond to crank angle (APmax) is shown in
figure 4-9. Before 16bar, APmax of gasoline, ethanol and methanol are similar to
each other around 15CAD ATDCf. After 16bar, the APmax of gasoline increases to
29.3 CAD ATDCf. The APmax of ethanol and methanol increases from 15 CAD
ATDCf to 23.9 CAD ATDCf. Between 18bar and 26bar, the APmax of methanol is
more retarded than that of ethanol since the spark timing of ethanol is more
advanced than methanol. The peak cylinder pressure (Pmax) and peak pressure rise
rate (Rmax) are also shown in Figure 4-9. The Pmax of gasoline increases from 14.4bar
to 80.7bar which happens at 16bar, then it keeps around 75bar because the spark
timing is retarded very close even after the engine’s top dead centre. The Pmax of
ethanol and methanol increases from 14.7bar to 113.6bar from the load 2bar to
26bar then it drops to 110.7bar. The limit of Pmax is 120bar and the maximum Rmax is
6bar/dCA so that even the knock intensity of ethanol and methanol is under the limit,
their spark timing also needed to be retarded after 16bar. That’s the reason why the
combustion duration of ethanol and methanol are increasing after 16bar. Since the
spark timing of gasoline is retarded more than ethanol and methanol, no attention
was paid to Pmax and Rmax of gasoline.
91
1.02 320

300
1.00
280
0.98 260
Lambda [#]
0.96 240
220
0.94 200
0.92 180
160
0.90
140
0.88 120
1000
180
980°C Temp Limit
160
950 Conventional Multi Cyl Limit
140
pRail[bar]
900 120
100
850
80
ExhaustTemp [degC]
780°C Temp Limit 60

800 Revised Single Cylinder Limit
40
750 14
12

700 10
8
650
6
4
600
2
550 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
The lambda values of all test points are shown in Figure 4-10. The lambda of
gasoline was kept at 1 before 26bar IMEP. It is decreased to 0.97 at 26bar IMEP and
0.89 at 28bar IMEP to reduce the exhaust gas temperature. For ethanol and
methanol, their lambda is kept at 1 even their combustion duration increases after
16bar IMEP, their exhaust temperature (shown in figure 4-10) is still under the limit of
780°C so that no over-fuelling is needed. The exhaust gas temperature of gasoline
increases from 599°C to 684°C from 2 bar to 8bar IMEP, then it drops to around
670°C from 10bar to 16bar IMEP. The reason why exhaust gas temperature at 8bar
is higher is because the spark timing at 8bar IMEP is slightly retarded which
increases the combustion duration. After that, it increases fast because its spark
timing is retarded which causes the combustion duration longer. It almost reaches
the limit 780°C at 24bar IMEP. At 26bar and 28bar IMEP, over-fuelling is introduced
because of charge cooling effect can reduce the in-cylinder temperature. During
testing, over-fuelling is achieved by decreasing lambda slowly as well as increasing
the load step by step, at the same time, the exhaust gas temperature is controlled to
be lower than 780°C. The exhaust gas temperature of ethanol increases from 592°C
92
to 680°C with the load increase from 2 bar to 28bar while that of methanol increases
from 582°C to 657°C. Overall, the exhaust gas temperature is lower than ethanol
and gasoline. That’s because methanol has the lowest lower heating value (LHV) so
that more fuel is injected into the cylinder, charge cooling effect can decrease the in-
cylinder temperature. Ethanol’s LHV is between methanol and gasoline so its
exhaust gas temperature is between methanol and gasoline.
The injection timings (i.e. the start of injection) are kept the same for different fuel as
boundary condition which is shown in Figure 4-10. From 2bar to 8bar IMEP, the
injection starts at 320 CAD BTDC. After that, injection timing is retarded from 316
CAD BTDC to 303 CAD BTDC and is kept at 303 CAD BTDC from 16bar to 24bar
IMEP. It’s advanced more to 297 CAD BTDC at 28bar. The injection pressure of
these fuels is increased from 55bar to 169bar. The fuel flow rates are also shown in
Figure 4-10. With the load increases from 2bar to 28bar IMEP, the fuel flow rate of
gasoline increases from 0.65kg/h to 7.71kg/h, that of ethanol increase from 0.91kg/h
to 10.3kg/h, and methanol’s fuel flow rate increases from 1.25kg/h to 13.4kg/h. Since
the lower heating value of methanol is the lowest, so the injection duration of
methanol is the longest. As a result of this, the fuel consumption and fuel flow rate of
methanol and ethanol are higher than gasoline at the same load. After 24bar IMEP,
the gasoline’s injection end timing is retarded because of increased fuelling so that
the fuel flow rate increases after 24bar IMEP.
93
baseline_3000rpm load swipe.nc (NIMEP,CO2)
Brunel Single Cylinder - 3000rpm Load Sweep ethanol_3000rpm processed data_3000rpm.nc (NIMEP,CO2)
methanol_3000rpm processed data_3000rpm.nc (NIMEP,CO2)}
Alternative Fuels Comparison
14.5 1100
1000
14.0 900
800
CO2 [%Vol]
THC [ppm]
CO2 [%]
13.5 700
600
13.0 500
400
12.5 300
200
12.0 100
4.0 1.8
3.5 1.6
3.0 1.4
CO (High) [%]
CO [%Vol]
O2 [%]
2.5 1.2
2.0 1.0
1.5 0.8
1.0 0.6
0.5 0.4
2400 0 5 10 15 20 25 30
2200 NIMEP [bar]
2000
1800
NOx [ppm]
1600
1400
1200
1000
800
600
400
200
0 5 10 15 20 25 30
NMEPav0 [bar]
Most of the emissions data are shown in Figure 4-11. The CO2 produced by gasoline
decreases from 14.12%vol to 13.16%vol, then it increases to 13.92%vol and drops
to 12.26%vol quickly because of over-fuelling. The CO2 of ethanol reduced from
13.72%vol to 12.9%vol with the load increase and the CO2 of methanol has the
same trends as ethanol, it decreases from 13.5%vol to 12.04%vol. The CO2
produced by ethanol and methanol is lower than gasoline since alcohol fuel leads to
lower specific CO2 because of CO2’s energy-specific basis (g CO2/MJ). The CO
produced by gasoline increases from 0.79%vol to 3.73%vol.
The CO produced by ethanol increases from 0.59%vol to 1.05%vol and that of

methanol increases from 0.57%vol to 1.29%vol. The amount of CO in the exhaust
gas has the opposite trend with CO2 and it’s mainly affected by Lambda (shown in
Figure 4-10). The amount of CO keeps stable for methanol and ethanol since all the
combustion are stoichiometric. In comparison to gasoline, methanol contains only
about 44% carbon, which is directly transformed into CO during combustion, thereby
reducing the formation and emission of CO. For gasoline, the content of CO is also
94
at a low level before 24bar IMEP. After 24bar, the CO increase because over-fuelling
is introduced to decrease the in-cylinder temperature.
The NOx released by gasoline increases from 762ppm to 2208ppm at 10bar IMEP.
After that, it decreases to around 1800ppm between 12bar and 24bar IMEP, and it
decreases quickly to 680ppm from 24bar to 28bar IMEP. The NOx produced by
ethanol increases from 566ppm to 2184 ppm with load increase, and methanol’s NOx
also increases from 244ppm to 2181ppm. The production of NOx is affected by the
peak in-cylinder pressure (shown in figure 4-9) because higher peak cylinder
pressure results in higher peak burned gas temperature then causes higher NOx.
The NOx of methanol and ethanol increase with the increase of peak cylinder
pressure before 16 bar IMEP. After that, the spark timing of them were retarded
because the maximum in-cylinder pressure 120bar IMEP has achieved and the peak
in-cylinder pressure was kept at 120bar so that the NOx release by methanol and
ethanol keeps stable after 16bar IMEP. For gasoline, the NOx increases from 2bar to
16bar IMEP with the increase of peak in-cylinder pressure. Afterwards, to keep the
knocking intensity under the limit, the spark timing was retarded, and the peak in-
cylinder pressure is around 80bar so that the NOx keeps stable between 16bar and
24bar IMEP. When operating at 26bar IMEP, the lambda decreases, more fuel was
injected and less oxygen is available therefore NOx decreases fast. Methanol
releases the lowest NOx since it has the highest evaporation enthalpy with the same
fuel quantity. Furthermore, methanol has the lowest LHV so that more methanol was
injected result in the lowest in-cylinder temperature (charge cooling effect) which
cause methanol produces the lowest NOx. THC released by methanol and ethanol
are lower than gasoline. The THC released by gasoline drops from 1031ppm to
172ppm at 26bar IMEP, then increases to 515ppm at 28bar IMEP. The THC
produced by ethanol reduced from 960ppm to 431ppm at 22bar IMEP, then
increases to 810ppm at 28bar IMEP. Methanol’s THC also drops from 336ppm to
166ppm at 16bar IMEP, then it increases to 503ppm at 28bar IMEP. Before 24bar
IMEP, alcohol fuels’ THC drops because their injection timing was retarded, and the
injection pressure were increased which cause less fuel impingement on the piston
and cylinder wall. After 24bar, ethanol and methanol’s THC increase since their
injection end timing are so retarded with the load increase which result in wall
wetting and their THC increases. Methanol has the lowest THC since it has the
95
lowest boiling point so that methanol evaporates very fast in the combustion
chamber which result in the air-methanol mixture quality is the best of the three fuels.
As a result of this, less fuel impingement on the cylinder wall and piston which leads
to less THC. The O2 released by these fuels have the same trend which increase
firstly then decrease and increase again after that. O2 released by gasoline
increases from 1.04%vol to 1.61%vol at 16bar IMEP, then keeps around 1.25%vol
from 18 to 24bar, and it decreases from 1.24%vol to 0.63%vol from 24bar to 28bar
because of over-fuelling. O2 produced by ethanol increases from 0.87%vol to
1.57%vol at 16bar, then it decreases to 1.2%vol at 22bar, and increases again to
1.56%vol at 28bar. The O2 released by methanol increases keeps around 0.7%vol
from 2bar to 16bar IMEP, then it increases to 1.23%vol at 18bar IMEP. After that, it
drops to 0.99%vol at 20bar then it increases to 1.44%vol at 28bar IMEP.
it should be pointed out that the flame ionization detector (FID) is less sensitive to
the alcohol and their partially oxidized products, the measured value of THC from
alcohol fuels is less than the true values [110].
96
220 240
200 220

180
pManifold [kPaA]
200
160
140 180
120 160
100
140
80
60 120
40 100
96 48
46

44
94 42
40
38
92
36
34
90 32
30
95
88 94
93
92
86 91
90
89
84 88
87
86
82 85
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Figure 4-12 Engine conditioning temperature and pressure
Most boundary conditions of this test are shown in Figure 4-12. The intake manifold
pressure of three different fuels are increased with the load increase. The intake
manifold pressure of gasoline increases from 45kPa to 215kPa, while that of ethanol
increases from 45kPa to 190.7kPa. The intake manifold pressure of methanol
increases from 45kPa to 175.8kPa. Above 16bar IMEP, the intake manifold pressure
of gasoline is higher than alcohol fuels because of its higher stoichiometric air-fuel
ratio. The coolant and oil temperature are kept around 90°C. The Intake air
temperature is kept around 40°C by the intake heater. The boost pressure is also
shown in Figure 4-12. The Pboost of gasoline is kept at 101kPa by means of natural
aspirate from 2 to 6bar IMEP. Because of the test bed’s issue since the boost
system is not stable with wide open throttle and low boost pressure. From 8 to 16bar
IMEP, the Pboost is kept at 130kPa, the load is controlled by the throttle. After 16bar
IMEP, the throttle is 100% open, the load increases with the increase of P boost, the
Pboost achieves at 220kPa for the load to achieve 28bar IMEP. The boost pressure of
ethanol increases from 100kPa to 196kPa, while that of methanol increases from
100kPa to 180kPa.
97
220000000
200000000
180000000
160000000
140000000
total PM
120000000
100000000
80000000
60000000
40000000
20000000
0
180000000
160000000
140000000
PM 10-1000nm
120000000
100000000
80000000
60000000
40000000
20000000
0
70000000
60000000
PM 23-1000nm
50000000
40000000
30000000
20000000
10000000
0
0 5 10 15 20 25 30
NMEPav0 [bar]
Figure 4-13 Particle numbers of gasoline, ethanol and methanol
Figure 4-13 shows the total particle numbers of all three fuels at 3000rpm, 23nm-
1000nm particle numbers and the 10nm-1000nm particle numbers. Due to
equipment’s issue, only some gasoline points’ PN data are collected. The total PN of
gasoline increases from 3.47e7 to 2.02e8 at 14bar then drops to 2.13e7 at 22bar
IMEP and increases again to 6.98e7 at 28bar IMEP. The total PN of ethanol reduces
from 7.3e7 to 6.7e6 at 16bar then increases to 5.46e7 at 28bar. The total particle
number of methanol increases from 3.77e7 to 6.48e7 at 12bar IMEP then reduces to
2.14e7 at 28bar IMEP slowly. The PN 10nm-1000nm of gasoline increases from
2.54e7 to 1.69e8 at 14bar then drops to 2.06e7 and increases to 6.97e7 at 28bar
IMEP. That of ethanol decreases from 5.94e7 to 4.47e6 at 16bar IMEP then
increases to 5.35e7 at 28bar IMEP. Ethanol’s 10-1000 PN keeps around 2.5e7 from
2bar to 12bar IMEP, then it drops to 2e7 from 14bar to 28bar. The 23-1000nm PN of
gasoline increases from 6.34e6 to 1.89e7 at 8bar then drops to 1.14e7 at 14 and
18bar IMEP. After that, the 23-1000nm PN of gasoline increases from 1.47e7 to
6.7e7 from 22bar to 28bar IMEP. That of ethanol reduces from 1.44e7 to 8.43e5 at
98
14bar then increases to 4.19e7 at 28bar IMEP. The 23-1000nm PN of methanol
increases slowly from 3.41e6 to 1.41e7 with the load increasing from 2bar to 28bar
IMEP. There are two reasons to explain why ethanol’s total PN drops from 8 bar to
18bar IMEP. The first one is that injection timing (Figure 4-10) is retarded from 320
CAD BTDC to 303 CAD BTDC which could reduce fuel impingement on the cylinder
wall and piston. The second one is that the rail pressure is increased from 120bar to
168bar which can improve the liquid atomization process [109]. The lower total PN of
ethanol than methanol is likely caused by the longer injection duration of methanol
and hence there is less time for atomisation and mixing to take place after the end of
the injection. After 16bar IMEP, gasoline and ethanol’s PN increase as more liquid
fuel is injected.
The explanation for the lower PN of methanol than gasoline is that methanol has a
34.8% gravimetric oxygen content. Since methanol formula, “CH4O” contains only
one carbon atom, and atomic mass of oxygen accounted for 50% of the relative
molecular mass of methanol. Since methanol is a typical fuel which contains a small
amount of carbon and a large amount of oxygen. Based on this reason, there are
few PM emissions when the engine is fuelled with pure methanol [111].
Figure 4-14 2-10bar IMEP Size spectral density
99
The size spectral density of gasoline, ethanol and methanol from the load 2-10bar
IMEP are shown in Figure 4-14. It’s clear that at the lowest load of 2 bar IMEP the
particle size distributions are mano-peak shapes with most particles between 5nm to
20nm by means of nucleation mode. Only gasoline’s 8bar IMEP gets dual-peaks in
which the left first peak is by the particle size belong to nucleation mode caused by
the oxidation of THC and the right peak is the particle size distribution from the
accumulation mode. What’s more, the particle number in accumulation mode is less
than nucleation mode.
Figure 4-15 10-20bar IMEP size spectral density
IMEP are shown in Figure 4-15. All the Gasoline’s curves are dual-peaks curves. At
14bar and 18bar IMEP operations, most particles released by gasoline belong to the
nucleation mode but for 22bar IMEP, the numbers of particulate in nucleation mode
and accumulation mode are almost equal to each other which means some small
particles stick to each other and form particles with the bigger size. Except at 20bar
IMEP, all other curves of methanol are mano-peak and most particle size is in
nucleation mode. Methanol’s 22bar’s size distribution mode is also a dual-peaks
curve but only limited particle size is in accumulation mode.
100
Figure 4-16 20-28bar IMEP size spectral density
IMEP are shown in Figure 4-16. At high load, the particle released by gasoline
mostly belongs to accumulation mode as a result of small particle sticking with each
other. All ethanol’s curves are also mano-peak and more than half particle’s size is
between 20-100nm. For methanol, the curves of its high load points are dual-peaks
curves and most particle size is under 20nm by means of nucleation mode.
4000rpm load sweep for gasoline, ethanol and methanol were done from 3.1bar to
26.9bar IMEP to explore the difference between these three fuels at high speed. The
maximum load is limited to 28bar IMEP by the knocking combustion of gasoline at
high load, and unstable combustion of ethanol and methanol.
101
Threshold: (y = 10)
Threshold: (y = 8)
Threshold: (y = 8):Polygon
baseline_4000rpm load swipe.nc (NIMEP,CA50)
Brunel Single Cylinder - 4000rpm Load Sweep ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,CA50)
methanol_4000rpm load swipe_Sheet1.nc (NIMEP,CA50)
4200 30
4150
50MBFav0 [°CA ATDCF]

25
4100
Speed [rpm]
4050 20
4000
3950 15
3900
10
3850
3800 5
20 23
18 22
10-90 Burn Duration

iOut[Deg CA BTDC]
16
21
14
12 20
10 19
8
18
6
4 17
2 16
1100 1.0
1000 0.9
900 0.8
BSFC [g/kWh]
800
0.7
700
0.6
600
0.5
500
400 0.4
300 0.3
200 0.2
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Figure 4-17 Combustion characteristic of 4000rpm load sweep
For 4000 rpm load sweep, all test points are running at the MBT spark timing if
possible. The spark timing and CA50 are shown in Figure 4-17. The spark timing of
gasoline is retarded from 18.3 Deg BTDC to 2.2 Deg BTDC with the load increases
from 3.1bar to 26.9bar IMEP. The spark timings are similar for ethanol and methanol,
they are retarded from 18.3Deg BTDC to 10.7 Deg BTDC. The CA50 of gasoline is
retarded from 8.1 Deg ATDC to 27.8 Deg ATDC, while ethanol and methanol’s CA50
are retarded from 8.2 Deg ATDC to 14.9 Deg ATDC. At 19.7bar and 23.6bar IMEP,
ethanol’s CA50 is slightly retarded than methanol since the spark timing of ethanol is
retarded more as well as the laminar flame speed of ethanol is slower than
methanol. Below 15.1bar IMEP all CA50 were kept at 8 Deg but after 15.1bar IMEP,
the spark timing of gasoline is retarded because of knocking intensity (shown in
Figure 4-17) is higher than 1 which is the maximum limit. For methanol and ethanol,
their spark timings are retarded because the Pmax (peak in-cylinder pressure) limit
of this engine is 120 bar and the Rmax (peak pressure rise rate) is 6bar/dCA. The
knocking intensity of gasoline increases from 0.2bar to 0.69bar at 15.4bar IMEP then
102
is kept around 0.65bar. Ethanol and methanol’s knocking intensity increases from
0.22bar to 0.89bar and 0.20bar to 0.96bar separately. The knocking intensities for
ethanol and methanol are always under the limit of 1 even at highest load because
their anti-knock characteristic is better than gasoline. At 11.2bar IMEP, methanol’s
spark timing is more retarded than gasoline and ethanol which cause methanol’s
knocking intensity at 11.2bar IMEP lower than the other fuels. The brake specific fuel
consumption (BSFC) of the three different fuels could also be found in Figure 4-17.
BSFC of gasoline decreases from 468g/kWh to 244g/kWh at 15.4bar, then increases
to 333g/kWh at 26.9bar IMEP. With the load increases, ethanol’s BSFC decreases
from 727g/kWh to 359g/kWh at 15.4bar IMEP, and increases to 369g/kWh.
Methanol’s BSFC is similar to ethanol which decreases from 1068g/kWh to
489g/kWh at 15.4bar IMEP and increases to 523g/kWh at 26.9bar IMEP. The BSFC
of different fuels have the same trend while the reason why the BSFC of methanol is
higher than ethanol and gasoline is because the LHV (lower heating value) of
methanol is the lowest of the three different fuels. The BSFC of gasoline increases
after 19.7bar IMEP is because over-fuelling is introduced to decrease the exhaust
gas temperature. For gasoline at 19.7bar, the spark timing could be advanced more
since there is still some margine between its knocking intensity and the limit. The
burn duration of gasoline increases from 17.5CAD to 22.6CAD with the load
increase. The burn duration of alcohol fuel keeps around 17.3CAD from 3.1bar to
15.4bar IMEP, then ethanol increases to 19.7CAD at 26.9bar IMEP and methanol
increases from 17.3CAD to 18.6CAD. The burn duration of methanol is shorter than
ethanol with similar spark timing after 15.4bar IMEP since methanol’s laminar flame
speed is higher than ethanol. From 3.1bar to 15.4bar IMEP, the knocking intensity of
these three fuels keeps increasing from 0.2bar to 0.72bar. After that, gasoline’s
knocking intensity keeps around 0.7bar because of retarding spark timing. Even the
spark timing of ethanol and methanol are also retarded, their knocking intensity
increases to 0.89bar and 0.96bar.
103
ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,APmax)
Brunel Single Cylinder - 4000rpm Load Sweep methanol_4000rpm load swipe_Sheet1.nc (NIMEP,APmax)
18.0 35
17.5
APmax [°CA ATDCF]

17.0 30
16.5
25
16.0
15.5
20
15.0
14.5 15
14.0
13.5 10
13.0 120
12.5 110
12.0 100
11.5 90
Pmax [bar]
80
11.0
70
10.5
60
10.0 50
9.5 40
9.0 30
8.5 20
10.0 6.0
5.5
9.5 5.0
Rmax [bar/dCA]
4.5
9.0 4.0
3.5
8.5 3.0
2.5
8.0 2.0
1.5
7.5 1.0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
The combustion characteristics of the three fuels are shown in Figure 4-18. The 10%
mass fraction burn of gasoline decreases from 17.5 CAD to 13.7 CAD at 15.4bar
IMEP, then increases to 17 CAD at 26.9bar IMEP. That of ethanol decreases from
17.6 CAD to 14.6 CAD then increases to 15.9CAD, and methanol’s CA10 decreases
from 17.2CAD to 14.8CAD at 19.7bar IMEP then increases to 15.5CAD. Before
15.4bar IMEP, the CA10-50 keeps around 9CAD for these three different fuels. After
that, CA10-50 of gasoline increases to 13CAD, and that of ethanol increases to
10.4CAD, methanol to 10CAD. The CA50-90 of gasoline keeps around 8.6CAD
before 15.4bar IMEP, after that it increases to 9.7CAD and then keeps around that.
The CA50-90 of ethanol increases from 7.7 CAD to 9.3 CAD, and that of methanol
increases from 8.0 Cad to 8.5 CAD. At 11.2bar IMEP of methanol, the CA50-90 is
slightly longer than other points since its spark timing is slightly retarded. Before
15.4bar IMEP, the CA10, CA10-50, and CA50-90 of them are similar since the spark
timing is almost the same. After 15.1bar IMEP, as a result of knocking, gasoline
needs to retard the spark timing more than alcohol fuel and got a longer combustion
duration. As mentioned before, gasoline at 19.7bar IMEP is not at MBT which
104
causes the combustion phasing to be retarded. Because methanol’s laminar flow
rate is faster than ethanol and gasoline so that all the mass fraction burn of methanol
is shorter than ethanol and gasoline. The APmax(maximum in-cylinder pressure
correspond to crank angle)can be found in Figure 4-18. The APmax of gasoline is
retarded from 14.8CAD ATDC to 34.9CAD ATDC. Because ethanol and methanol’s
spark timing aligns with each other, their APmax also keeps the same which is
retarded from 15.1CAD ATDC to 21.5CAD ATDC. The peak cylinder pressure (Pmax)
and peak pressure rise rate (Rmax) are also shown in Figure 4-18. The Pmax of
gasoline increases from 21.6bar to 78.8bar IMEP at the load of 15.4bar IMEP, then
the Pmax keeps around that even with the load increase. The Pmax of ethanol and
methanol align with each other which increase from 21.3bar to 112.6bar IMEP. The
Rmax of gasoline increases from 1.2bar/dCA to 4.5bar/dCA at the load 15.4bar and
Rmax is kept around 3bar/dCA with the load increasing up to 26.9bar. For ethanol, its
Rmax increases from 1.2bar/dCA to 5.1bar/dCA while methanol’s Rmax increases from
1.1bar/dCA to 5.5bar/dCA. The limit of Pmax is 120bar and the maximum Rmax is
6bar/dCA so that even the knock intensity of ethanol and methanol is under the limit,
their spark timing also needed to be retarded after 15.4bar IMEP. That’s the reason
why the combustion duration of ethanol and methanol are also increasing slowly
after 15.4bar IMEP. After 15.4bar IMEP, the Rmax of methanol is higher than ethanol
since methanol’s flame speed is faster than ethanol which causes the burn duration
shorter than ethanol. Because the spark timing of gasoline is retarded more than
ethanol and methanol, there is no need to pay attention to Pmax and Rmax.
105
baseline_4000rpm load swipe.nc (NIMEP,lMeasured_CCP_1)
ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,lMeasured_CCP_1)
Brunel Single Cylinder - 4000rpm Load Sweep methanol_4000rpm load swipe_Sheet1.nc (NIMEP,lMeasured_CCP_1)
1.05 330

325
1.00
320
0.95 315
Lambda [#]
310
0.90
305
0.85 300
295
0.80
290
0.75 285
1000
180
980°C Temp Limit 170
950 Conventional Multi Cyl Limit 160
150
pRail[bar]
140
900 130
120
110
ExhaustTemp [degC]
850
100
90
780°C Temp Limit
800 80
Revised Single Cylinder Limit
18
16
750

14
12
700 10
8
6
650
4
2
600 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
The lambda and exhaust gas temperature of all test points are shown in Figure 4-19.
The exhaust gas temperature of ethanol increases from 654°C to 759°C, and that of
methanol increases from 641°C to 726°C with the load increasing from 3.1bar to
26.9bar IMEP. For ethanol and methanol, their lambda is kept at 1 since even if their
combustion duration increase after 15.4bar IMEP, their exhaust temperature is still
under the limit 780°C so over-fuelling is not needed. For gasoline, the spark timing
was retarded more than alcohol fuel after 15.4bar IMEP because of knocking. After
that, the exhaust temperature increases as a result of combustion duration increase
and it reaches the limit at 23.6ba IMEP r. After 23.6bar IMEP, over-fuelling is
introduced because of charge cooling effect can reduce the in-cylinder temperature.
During testing, over-fuelling is achieved by decreasing lambda slowly as well as
increasing the load step by step, at the same time, the exhaust gas temperature is
also focused lower than 780°C. The lambda is 0.84 at 23.6bar IMEP and 0.78 at
27.9bar IMEP. The exhaust gas temperature of methanol is always lower than
ethanol and gasoline because the lower heating value of methanol is the lowest of
these fuels, more methanol is injected into cylinder and the charge cooling effect of
106
methanol is stronger than ethanol and gasoline which causes the lowest exhaust gas
temperature. The injection timing and injection pressure are kept the same for
different fuels as boundary conditions which are shown in Figure 4-19. The injection
start is advanced from 287CAD BTDC to 328Deg BTDC, and rail pressure increases
from 86bar to 169bar with load increases. The last diagram of Figure 4-19 shows the
fuel flow rate of gasoline, ethanol and methanol. The fuel flow rate of gasoline
increases from 1.18kg/h to 11.52kg/h. That of ethanol increases from 1.74kg/h to
12.71kg/h and methanol increases from 2.53kg/h to 17.12kg/h. Because the lower
heating value of methanol is lower than ethanol and lower than gasoline, so the
injection duration of methanol is the longest one of these fuels. As a result of this, the
fuel consumption and fuel flow rate of methanol and ethanol are higher than gasoline
at the same load. After 19.7bar IMEP, gasoline’s fuel flow rate increases faster
because of over-fuelling.
baseline_4000rpm load swipe.nc (NIMEP,HC)

ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,HC)
Brunel Single Cylinder - 4000rpm Load Sweep methanol_4000rpm load swipe_Sheet1.nc (NIMEP,HC)
14.5 1600
14.0 1400
13.5 1200
13.0
CO2 [%Vol]
THC [ppm]
1000
12.5
800
12.0
600
11.5
11.0 400
10.5 200
10.0 0
8 1.2
7 1.1
1.0
6
0.9
CO (High) [%]
O2 [%Vol]
5 0.8
4 0.7
3 0.6
0.5
2
0.4
1 0.3
0 0.2
3000
2500
2000
NOx [ppm]
1500
1000
500
0
0 5 10 15 20 25 30
NIMEP [bar]
107
Most of the emissions data are shown in Figure 4-20. The CO2 released by gasoline
keeps around 14%vol from 3.1bar to 19.7bar IMEP, then it starts to reduce from
23.6bar to 26.9bar IMEP which contains CO2 10.2%vol. The CO2 produced by
ethanol keeps around 13.5%vol during the load sweep. Methanol’s CO2 reduce
slowly from 13.46%vol to 12.28%vol. The CO2 gasoline decreases quickly from
19.7bar IMEP as a result of over-fuelling. The CO2 produced by ethanol and
methanol is lower than gasoline since alcohol fuel leads to lower specific CO2
because of CO2’s energy-specific basis (g CO2/MJ). The CO released by gasoline
keeps around 0.85%vol from 3.1bar to 19.7bar IMEP, then it increases quickly to
7.63%vol at 26.9bar IMEP. The CO produced by ethanol and methanol increases
slowly during this 4000rpm load sweep. Ethanol’s CO increases from 0.70%vol to
1.03%vol while that of methanol increases from 0.49%vol to 1.41%vol. The amount
of CO in the exhaust gas has the opposite trend with CO2 and it’s mainly affected by
Lambda (shown in Figure 4-19). The amount of CO keeps stable for methanol and
ethanol since all the combustion are stoichiometric. For gasoline, the content of CO
is also similar to alcohol fuel before 19.7bar IMEP. After 19.7bar IMEP, the CO
increase because over-fuelling is introduced. Additional, NOx is also shown in Figure
4-20. The NOx of gasoline increases from 1522ppm to 2811ppm at 11.2bar IMEP,
then it decreases to 276ppm at 26.9bar IMEP. The NOx produced by ethanol
increased from 926ppm to 1901ppm at 23.6bar IMEP, and it keeps around 1700ppm
at the rest of load. Methanol’s NOx increases from 280ppm to 1530ppm at 23.6bar
then it decreases to 1382ppm at 26.9bar. The content of NOx is affected by the peak
in-cylinder pressure (shown in figure 4-18) because higher peak cylinder pressure
results in higher peak burned gas temperature then causes higher NOx. The NOx of
methanol and ethanol increases with the increase of peak cylinder pressure before
15.4bar IMEP. After that, the spark timing of them were retarded because the
maximum in-cylinder pressure 120bar has achieved and the peak in-cylinder
pressure was kept at 120bar so that the NOx release by methanol and ethanol keeps
stable after 15.4bar IMEP. For gasoline, the NOx increases from 3.1bar to 15.4bar
IMEP with the increase of peak in-cylinder pressure. Afterwards, to keep the
knocking intensity under limit, the spark timing was retarded and the peak in-cylinder
pressure keeps around 80bar so that the NOx keeps stable between 15.1bar and
19.7bar IMEP. When achieving 23.6 bar, the lambda decreases, more fuel was
injected and evaporation cause the in-cylinder temperature decrease then NOx
108
decreases fast as a result of this. Overall, methanol’s NOx is the lowest of the three
fuels since its exhaust gas temperature is lower than ethanol and gasoline as a
result of more effective charge cooling effect.
THC is shown after NOx in Figure 4-20. The THC released by gasoline decreases
from 967ppm to 363ppm at 19.7bar IMEP because the in-cylinder thermal dynamic
condition is better with the load increases so that the combustion is more sufficient
with the unburn hydrocarbon decreases. After that, THC increases to 1572ppm at
26.9 bar IMEP because of over-fuelling, more fuel is injected into cylinder but not
burned. THC has the same trend with the reference from MAHLE powertrain and the
THC released by methanol and ethanol are lower than gasoline because of flame
ionization detector (FID) which has been explained in the previous section. The O 2
released by gasoline keeps around 1%vol from 3.1bar to 19.7bar, then it drops to
0.21%vol because of over-fuelling. Ethanol’s O2 decreases from 0.6%vol to
0.51%vol from 3.1bar to 11.2bar IMEP, then it increases to 1.02%vol at 19.7bar
IMEP and decreases to 0.85%vol at 26.9bar IMEP. The O2 produced by methanol
increases from 0.63%vol to 1.02%vol from 3.1bar to 19.7bar IMEP, and it reduces to
0.75%vol at 26.9bar IMEP.
109
baseline_4000rpm load swipe.nc (NIMEP,pManifold_CCP_1)
ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,pManifold_CCP_1)
Brunel Single Cylinder - 4000rpm Load Sweep methanol_4000rpm load swipe_Sheet1.nc (NIMEP,pManifold_CCP_1)
240 260
220 240

200
220
pManifold [kPaA]
180
160 200
140 180
120 160
100
140
80
60 120
40 100
94 50
48

92 46
44
90 42
40
38
88
36
95
86
84
90
82
80 85
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Figure 4-21 Engine conditioning temperature and pressure
Most boundary conditions of this test are shown in Figure 4-21. The manifold
pressure is shown in the first diagram of Figure 4-21. The Pmanifold of gasoline,
ethanol and methanol increase from 45kPa to 235.8kPa, 45kPa to 214.4kPa and
45kPa to 195.8kPa. The reason why gasoline’s intake pressure is higher since its
stoichiometric air-fuel ratio is higher than ethanol and methanol, to get the same
power, more air is needed for gasoline. The coolant and oil temperature are kept
around 90°C for all test points of this load sweep. The Intake air temperature are
kept around 40°C. Pboost set point is also shown in figure 4-21. Before 15.4bar, the
Pboost for these three fuels are the same, they are 100kPa at 3.1bar IMEP then
increases to 130kPa from 7.1bar to 15.4bar IMEP. After 15.4bar IMEP, the P boost of
gasoline, ethanol and methanol are increased to 240.9kPa, 219.1kPa and 200.2kPa
separately. The manifold pressure trends are similar with boost pressure and the
reason why Pboost of gasoline is higher than ethanol and methanol is the same with
intake manifold pressure. The intake air temperature is kept at 40°C all the time by
an intake heater which located upstream of the plenum.
110
baseline_4000rpm load swipe.nc (NIMEP,Eng_EffTherma)
Brunel Single Cylinder - 4000rpm Load Sweep ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,Eng_EffTherma)
methanol_4000rpm load swipe_Sheet1.nc (NIMEP,Eng_EffTherma)
aInjectionEndCCP:1 [deg BTDCf]

40 300
250
Brake_EffTherma [%]
35
200
30
150
25
100
20 50
15 0
550 550
500 500
450
BSFC_Corr42LHV
450
ISFCn [g/kWh]
400
350 400
300 350
250
300
200
150 250
100 200
20 35
18
Rmax_max [bar/dCA]
APmax [dCA ATDCF]

16 30
14
12 25
10
8 20
6
4 15
2
0 10
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Figure 4-22 Combustion characteristics of 4000rpm load sweep
The brake thermal efficiency (BTE) and fuel consumption are shown in Figure 4-22.
The BTE of gasoline increases from 18.6% to 35.7% at 15.4bar IMEP, then it
decreases to 26.2%. The BTE of ethanol increases from 18.1% to 36.6% then it
drops to 35.7% at 26.9bar. Methanol’s BTE increases from 16.9% to 36.9% and it
decreases to 34.6% with load increases to 26.9bar IMEP. Before 15.4bar IMEP, the
BTE of the three fuels are similar. After that, the BTE of them decrease because the
spark timing was retarded by the effect of knocking for gasoline and engine limit for
alcohol fuels. The spark timing was more retarded for gasoline and its BTE decrease
faster as a result of this. In order to compare the BSFC of the three different fuels,
their original BSFC is timed by their LHV (lower heating value) then divided by 42.
The BSFC corrected by 42 is shown after BTE. Gasoline’s BSFC drops from
460.4g/kWh to 247.1g/kWh at 19.7bar, then it increases to 327.2g/kWh because of
over-fuelling. That of ethanol and methanol are aligned with each other which drops
from 473.5k/kWh to 234g/kWh and keeps around that for the rest of load. The
111
injection end timing is also shown in Figure 4-22. The injection end of gasoline,
ethanol and methanol is retarded from 265CAD BTDC to 190CAD BTDC, 256CAD
BTDC to 180CAD BTDC and 245 CAD BTDC to 137CAD BTDC separately.
Because they have the same injection start timing, so that the injection duration of
methanol is longer than ethanol and gasoline because the LHV (lower heating value)
is lower.
baseline_4000rpm load swipe.nc (NIMEP,totalPM)

ethanol_4000rpm load swipe_Sheet1.nc (NIMEP,totalPM)
Brunel Single Cylinder - 4000rpm Load Sweep methanol_4000rpm load swipe_Sheet1.nc (NIMEP,totalPM)
120000000
100000000
80000000
total PM
60000000
40000000
20000000
0
120000000
100000000
PM 10-1000nm
80000000
60000000
40000000
20000000
0
120000000
100000000
PM 23-1000nm
80000000
60000000
40000000
20000000
0
0 5 10 15 20 25 30
NMEPav0 [bar]
Figure 4-23 Particulate number of gasolines, ethanol and methanol
The particle numbers of gasoline, ethanol and methanol are shown in Figure 4-23
which includes total particle numbers, 10-1000 particle numbers as well as 23-
1000nm particle numbers. Gasoline’s PN data is only collected from 19.7bar to
26.9bar IMEP. The total particle number of gasoline is shown in the first diagram of
figure4-23. It increases from 2.99e7 to 1.2e8 with the load increasing from 19.7bar to
26.9bar IMEP. The total PN released by ethanol reduces slowly from 3.81e7 to
1.72e7 at 19.7bar IMEP, it then increases to 6.25e7 at 26.9bar IMEP. The total
number of particles produced by methanol keeps around 1.7e7 during the load
112
sweep. The second diagram of Figure 4-23 shows the number of particle which size
between 10 to 1000nm. Gasoline PN 10-1000nm increases from 2.92e7 to 1.18e8.
That of ethanol drops from 3.81e7 to 1.09e7 at 7.1bar, then it keeps around 1.5e7
from 11.2bar to 19.7bar IMEP. After that, it increases fast to 6.25e7 at 26.9bar IMEP.
The PN 10-1000nm of methanol decreases slowly from 1.02e7 to 3.75e6 at 15.4bar
then it increases gradually to 1.56e7 at 26.9bar. The last diagram of Figure 4-23
shows the number of particle size between 23-1000nm. Gasoline’s PN 23-1000nm
also increases from 2.33e7 to 1.14e8 quickly. The particle size from 23-1000nm
released by ethanol decreases from 2.88e7 to 1.79e6 at 7.1bar IMEP, then it
increases slowly to 3.87e7 at 26.9bar IMEP. Methanol’s PN 23-1000nm reduces
from 5.91e5 to 2.95e5 at 11.2bar. After that, it increases gradually from 4.82e5 to
1.56e7 at 26.9bar IMEP with the load increases. It’s easy to find that there’s a tiny
difference between gasoline’s PN with these three ranges which means that the
most particles released by gasoline, their size is larger than 23nm. This could be
verified by the spectral size density analysis afterwards. The reason why gasoline,
ethanol and ethanol’s PN increase from 19.7 to 26.9bar is because the injection
timing is advanced as well as the rail pressure is high which causes fuel
impingement on the piston and cylinder wall. What’s more, the in-cylinder
temperature increases which caused by the burn duration also has a negative effect
on the PN number. Gasoline’s total PN is higher than ethanol and methanol from
19.7bar to 26.9bar IMEP because of over-fuelling, the rich mixture leads to the sharp
increase of PN. Both ethanol and methanol’s total PN drops with the load increases
from 3.1bar to 7.1bar IMEP because the spark timing is retarded, more sufficient
time for air-fuel mixing. What’s more, the injection is advanced from 287CA BTDC to
294CA BTDC which also gives more time for air-fuel mixing.
113
Figure 4-24 Size spectral density of 4000rpm load sweep
The particle size distribution of 4000rpm load sweep is shown in Figure 4-24. The
particle size curves of gasoline are all mano-peak curves and their size is between
50 to 100nm so most particle released by gasoline is accumulation mode. As a result
of over-fuelling, the THC produced by gasoline increases at 23.6bar and 26.9bar
IMEP. The THC is oxidation into soot particle in nucleation mode and then stick
together to form bigger size particle. That’s the reason why the particle numbers and
size increase sharply after over-fuelling. Most particle released by ethanol, its size is
between 10 to 100nm so that it contains both nucleation mode and accumulation
mode. All methanol’s particle size is around 10nm so that no particle of accumulation
mode is released by methanol.
4.3 Summary
For the 2000rpm load sweep, ethanol and methanol exhibited similar combustion
characteristics to E10 gasoline when the load is limited to 11bar IMEP. Even ethanol
and methanol’s burn speed are faster than gasoline, but it doesn’t reflect clearly on
114
combustion phasing and burn duration. There’s a reduction of CO2 released by
ethanol and methanol on the energy-specific basis. What’s more, the THCs released
by ethanol and methanol are also lower than gasoline partially due to the
ineffectiveness of FID in detecting the partially oxidised HCs.
For 3000rpm load sweep, the load ranges from 2bar to 28bar IMEP. Below 16bar
IMEP, the indicated thermal efficiency of gasoline, ethanol and methanol are similar
since all fuels’ CA50 were controlled at 8 CAD ATDC at the MBT timing. Above
16bar IMEP, all these fuels’ spark timings were retarded. For gasoline, its spark
timing was retarded since knocking happened at higher load than 16bar IMEP. For
alcohol fuels, the spark timings were limited by either or both peak in-cylinder
pressure and pressure rise rate. But gasoline’s spark timing needs to be retarded
more than alcohol fuels to prevent knocking. As a result of this, ethanol and
methanol can achieve higher thermal efficiency than gasoline at higher load
operations. Methanol achieved the highest brake thermal efficiency 37.9% at 18bar
IMEP and ethanol’s highest BTE was 37.7% also at 18bar. Gasoline’s best BTE was
36.5% at 10bar. For gasoline, over-fuelling was introduced at 26bar and 28bar IMEP
to keep the exhaust gas temperature below 780°C. For ethanol and methanol,
because their evaporation enthalpy is higher than gasoline which lead to lower in-
cylinder temperature so that their exhaust gas temperature is under limit. As a result
of this, over-fuelling is not required for ethanol and methanol. A reduction of NOx
emission was observed from ethanol and methanol engine operations. HC released
by alcohol fuels are also lower than gasoline because of two reasons. The first one is
FID which has mentioned before. Another is that, alcohol fuel evaporates faster than
gasoline so that they got longer time to mix with air which can makes air-fuel mixture
quality better than that of gasoline so that the amount of unburned hydrocarbon is
less than gasoline. Across the 3000rpm load sweep, ethanol and methanol release
less particulate emission compared with gasoline. There’s up to 90% reduction of PN
number by ethanol and methanol. What’s more, the particle produced by ethanol and
methanol whose size is also less than that of gasoline. The disadvantage of ethanol
and methanol is that their lower heating values are lower than gasoline which cause
longer injection duration and higher fuel consumption.
4000rpm load sweep’s conclusion is similar to the 3000rpm load sweep.
115
5. Chapter Five: Effects of Injection Timing, Rail Pressure on Ethanol and
Methanol Engine’s Performance and Emissions
5.1 Introduction
This chapter focuses on the effects of injection timings and rail pressure on the fuel
efficiency, combustion characteristics, emissions as well as particle numbers from
the single cylinder spark ignition engine operated with ethanol or methanol. The
baseline results are obtained for E10 Euro 6 gasoline. Two groups of fuel matrix
tests which are 2000rpm 4.6bar IMEP and 3000rpm 16bar IMEP were selected to
investigate the difference between alcohol fuels and gasoline. These two groups of
tests cover medium-speed medium load and high-speed high load. By adjusting the
injection timing and rail pressure, the best fuelling strategies for each fuel can be
found. Also, injection timing and rail pressure have a big effect on particle number
and particle size which will also be explained in this chapter. At 2000rpm 4bar IMEP,
there is no knocking combustion since the load is low. At 3000rpm 16bar IMEP, as
mentioned in chapter 4, gasoline’s knocking intensity is high and the spark timing
needs to be retarded but the 50% mass fraction of burn (CA50) of methanol and
ethanol can be controlled at 8 Deg ATDCf.
5.2 Experimental Setup and Test Conditions
All the testing work were carried out on the same single-cylinder direct injection
engine. Two groups of fuel matrix tests which are 2000rpm 4.6bar IMEP and
3000rpm 16bar IMEP and the details of these fuel matrix test points are shown in
Table 5.1.
116
Table 5-1 Fuel matrix test point
Speed & IMOP EMOP SOI Rail Pressure Spark

Lambda
Load (ATDC) (BTDC) (CAD BTDCf) (bar) timing
2000rpm
4.6bar 100 100 DBL 1
IMEP
[275,300,325,350] [50,100,150,200]
3000rpm
16bar 82 100 DBL 1
IMEP
For each group of testing, the IMOP and EMOP are the same and the injection
timing is adjusted from 275 Deg BTDCf to 350 Deg BTDCf with a gap of 25 Deg, and
the rail pressure is adjusted from 50 bar to 200 bar with a gap of 50bar so that there
are 16 test points in each group of testing.
117
5.3 Results and Discussion
5.3.1 2000rpm 4bar IMEP Fuel Matrix Analysis
Gasoline Ethanol
BSFC_cor BSFC_cor BSFC_cor Methanol
220 220 220
296 297 302 303 304 287 288 300 294 304 285 293 294 298 304
200 200 200
302 302 302
180 300 180 300 180 300
298 298 298
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 297 298 302 303 160 288 288 293 297 160 282 291 296 295
296 296 296
140 294 140 294 140 294
292 292 292
120 290 120 290 120 290
300 298 297 302 290 292 294 299 285 289 297 299
288 288 288
100 299 100 294 100 297
286 286 286
80 284 80 284 80 284
296 282 286 282 282
60 300 299 301 60 291 288 298 60 288 285 294 296
280 280 280
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
Injection Start [°BTDCf] Injection Start [°BTDCf] Injection Start [°BTDCf]
BSFC BSFC BSFC
220 220 220
302 302 307 308 441.3 442.0 460.5 452.0 601 619 621 628
200 630 200 630 200 630
600 600 600
180 570 180 570 180 570
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 303 304 307 309 540 160 442.8 442.0 449.8 455.9 540 160 595 615 624 623 540
510 510 510
140 140 480 140
480 480
120 450 120 450 120 450
306 304 303 308 445.6 448.8 451.0 459.5 420 602 611 626 631
420 420
100 305 100 451.1 390 100 627
390 390
80 80 360 80
360 360
330
302 330 439.6 330
60 305 304 307 60 447.2 441.9 458.0 300 60 608 601 620 624
300 300
40 40 40
270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-1 BSFC of different inject timing and rail pressure
The sequence of the fuel are gasoline, ethanol and methanol from left to right, and
each column shows the same fuel in all the diagrams in this chapter.
The 3 diagrams on the top of Figure 5-1 shows the brake fuel consumption corrected
by 42 of gasoline, ethanol and methanol. The X-axis shows the injection start timing
and Y-axis shows the rail pressure (injection pressure). It’s much clear to find the
difference between these three different fuels by correcting their BSFC by a specific
value since they have different lower heating values. The actual BSFC times the
lower heating value of each fuel than divided by 42 is the value of BSFC_Cor. The
lowest BSFC_cor point of gasoline is at the injection timing 300CA BTDC and rail
pressure 50bar, and its BSFC_cor is 296.4g/kWh. Keep rail pressure the same, and
the BSFC_cor decreases with the injection timing is advanced from 275CA BTDC to
118
300CA BTDC since there is more sufficient time for air-fuel mixing. After that, the
BSFC_cor increases with the injection timing advanced to 350CA BTDC since the
injection timing is too early and the piston is very close to the injector then causing
wall-wetting on the piston so that the BSFC_cor increases. At the late injection points
of injection timing at 275CA BTDC, the BSFC_cor decreases with the rail pressure
since higher injection pressure can decrease the size of fuel droplets hence the
evaporation of fuel is better, then it can give a positive effect on the air-fuel mixing.
What’s more, higher rail pressure also advances the end ofinjection which can give
more time for air-fuel mixing before the spark ignition. The maximum BSFC_cor for
gasoline is 303.1g/kWh which occurs at injection start 350CA BTDC, rail pressure
150bar. At the points which injection timing is 350CA BTDC, their BSFC gets worse
as the rail pressure increases since early injection and high rail pressure will cause
fuel impingement on the top of the piston hence pool fire. Ethanol’s average
BSFC_cor is lower than gasoline since its boiling point is lower than gasoline. The
minimum BSFC_cor point of ethanol of 286.1g/kWh happens at the injection timing
325CA BTDC, rail pressure of 50bar. This is because the lower heating value of
ethanol is lower than gasoline, to get the same power, more fuel is injected into
cylinder so that the injection end timing of ethanol is more retarded than gasoline. As
a result of this, ethanol’s best BSFC_cor point’s injection time is more advanced than
gasoline to get more sufficient time for air-fuel mixing. The high BSFC_cor regions
happen at early injection& high rail pressure region and the maximum BSFC_cor
occurs at injection timing 325CA BTDC, rail pressure 200bar which value is
299.7g/kWh. Methanol’s BSFC_cor is shown in the last column of Figure 5-1. It’s
easy to find that methanol has the lowest average BSFC_cor of these three fuels
since it has the highest evaporation enthalpy thus the fuel-power transfer rate is the
highest. Methanol’s lowest BSFC_cor is at injection start 275CA BTDC, rail pressure
150bar which value is 281.9g/kWh. The maximum BSFC_cor happens at 350CA
BTDC, rail pressure 100bar and its value is 298.9g/kWh. High BSFC_cor region also
occurs at injection early points by the reason of fuel impingement on the top of the
piston. At the same injection timing and rail pressure, methanol has lower BSFC_cor
than ethanol and gasoline because it got the highest evaporation enthalpy so its air-
fuel mixture is the best. Compare with gasoline’s best BSFC_cor, ethanol’s best
BSFC_cor is reduced by 3.5% and that of methanol is reduced by 4.8%.
119
The second column of Figure 5-1 shows the brake specific fuel consumption (BSFC)
of gasoline, ethanol and methanol. The highest and lowest points of BSFC are the
same as BSFC_cor. The lowest BSFC of gasoline is 301.8g/kWh and the highest is
308.6g/kWh. The best BSFC of ethanol is 439.6g/kWh while the worst point’s value
is 460.5g/kWh. Methanol’s minimum BSFC is 594.9g/kWh and its maximum BSFC is
630.9g/kWh. Methanol gets the highest average BSFC because its lower heating
value is lower than ethanol and gasoline so that more methanol is injected into the
cylinder to get the same power as ethanol and gasoline.
Figure 5-2 Brake Thermal Efficiency of gasoline, ethanol and methanol
The brake thermal efficiency (BTE) of these fuels are shown in Figure 5-2. The
maximum BTE of gasoline is 28.92% which happens with the start of injection at
300CA BTDC, rail pressure of 50bar while the minimum BTE of gasoline is 28.28%.
The best BTE of ethanol is 29.96% with an injection timing at 325CA BTDC and rail
pressure at 50bar. At injection start timing 275CA BTDC and rail pressure 150bar,
methanol gets its highest BTE of 30.41%. Compared with gasoline’s best BTE,
ethanol’s best BTE improved by 3.6% and that of methanol improved by 5.2%. The
difference between the best and worst BTE of gasoline, ethanol and methanol are
0.64%, 1.36% and 1.74% which means that injection timing and rail pressure have
more effect on methanol than ethanol and gasoline.
120
Spark Timing [°BTDC] Gasoline Spark Timing [°BTDC] Ethanol Spark Timing [°BTDC] Methanol
220.0 220.0 220.0
15.75 17.25 16.25 16.75 18 16.25
16 16.75
17 15.25
15 15.25
15 18 15.75 15.75 14.25 14.25 18
200.0 200.0 17.75 200.0
17.75 17.75
180.0 17.5 180.0 17.5 180.0 17.5
17.25 17.25 17.25
Rail Pressure [bar]
Rail Pressure [bar]

160.0 16.25 160.0 16.25 160.0 15.75
pRailCCP:1 [bar]
17.25 16.25 16.75 16 16.75

17 15.25
15 15.25
15 17 15.75 14.75 14.75
17 17
16.75 16.75 16.75
140.0 140.0 140.0
16.5 16.5 16.5
120.0 16.25 120.0 16.25 120.0 16.25
15.75 17.25 16.75 16.75 16.25
16 16.75
17 16.75
17 15.25
15 16 15.75 15.25 15.25 14.25
16 16
100.0 16.75 100.0 15.25
15 100.0 15.75
15.75 15.75 15.75
80.0 15.5 80.0 15.5 80.0 15.5
15.25 15.25 15.25
60.0 15.75 17.75 16.75 16.75 60.0 16.25
16 15.25
15 16.75
17 15.25
15 60.0 15.75 15.25 16.25 14.25
15 15 15
40.0 40.0 40.0
270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
aInjectionStartCCP:1 [deg BTDCf] Injection Start [°BTDCf] Injection Start [°BTDCf]
Figure 5-3 Spark Timing of gasoline, ethanol and methanol
The spark timing of gasoline is shown in the first diagram of Figure 5-3. The spark
timing of gasoline is between 15.75CA BTDC and 17.75CA BTDC. The most
advanced spark timing occurred when the injection starts 300CA BTDC and rail
pressure at 50bar while the most retarded one happened at the injection timing
275CA BTDC, rail pressure 50bar. The reason why late injection and low rail
pressure point got the most retarded spark timing is because the last injection
causes there is insufficient time for air-fuel mixing. What’s more, low injection
pressure increases the droplet’s size hence give negative effect on air-fuel mixing. At
the other rail pressure, with the injection timing advanced from 275CA BTDC to
350CA BTDC, the spark timing was advanced from 16CA BTDC to more than 17CA
BTDC since there is more time for air-fuel mixing.
Ethanol’s spark timing is slightly retarded than gasoline when the CA50 is controlled
at 8CA ATDC because ethanol’s laminar flame speed is faster than gasoline. Also, at
the same rail pressure, ethanol’s injection duration is longer than gasoline since
ethanol’s lower heating value is lower than gasoline, and more fuel was injected into
cylinder. The injection end timing of ethanol is also more retarded than gasoline, this
is another reason why ethanol’s spark timing is more retarded. The most retarded
spark timing of ethanol is 15.25CA BTDC which happened at the test points whose
injection timing is 350CA BTDC and 325CA BTDC, rail pressure at 100& 150bar.
The most advanced spark timing is 16.75CA BTDC which occurred at some points
whose injection timing are 300CA BTDC and 325CA BTDC. But only the point with
injection timing 300CA BTDC and rail pressure 100bar gets the CA50 (shown in
Figure 5-4) at 8CAD ATDC, the other points’ spark timings are slightly retarded.
121
Methanol got the most retarded average spark timing since it has the fastest laminar
flame speed of the three fuels. What’s more, methanol got the longest injection
duration of these three fuels so that more time is required for air-fuel mixing, and the
fuel quantity is larger also causes the mixing time longer. The most advanced spark
timing is 16.25CA BTDC which happened at injection timing 325CA BTDC and rail
pressure 50bar and the most retarded spark timing of 14.25CA BTDC occurred at
the early injection region.
In theory, keeping injection timing constant, increase rail pressure will cause the
injection end timing advanced thus the spark timing is advanced. Also, the fuel’s
droplet becomes smaller which can enhance the quality of air-fuel mixture as a result
of increasing rail pressure. But by looking through the spark timing of gasoline,
ethanol and methanol, increasing rail pressure has a limit effect on spark timing.
Even at earler and later injections, the spark timing remains the same as the rail
pressure is changed.
Gasoline Ethanol
Combustion Phasing 50%MFB Combustion Phasing 50%MFB Combustion Phasing 50%MFB Methanol
220 220 220
9 8 8 8 9 8 6 8 7 9 9 8 8 8 9
200 8.8 200 8.8 200 8.8
8.6 8.6 8.6
180 8.4 180 8.4 180 8.4
8.2 8.2 8.2
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]
160 8 8 8 8 8 160 8 7 8 8 8 160 8 8 8 8 8

7.8 7.8 7.8
140 7.6
140 7.6
140 7.6
7.4 7.4 7.4
120 120 120
9 8 7 8 7.2 8 8 6 8 7.2 8 8 8 9 7.2
100 8 7 100 9 7 100 8 7
6.8 6.8 6.8
80 6.6 80 6.6 80 6.6
6.4 6.4 6.4
60 9 8 8 8 6.2 60 8 8 7 8 6.2 60 9 8 8 9 6.2
6 6 6
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
Burn Duration 10-90% Burn Duration 10-90% Burn Duration 10-90%
220 220 220
19 19 19 19 17.7 17.6 18.2 18.0 18 19 18 18
22 22 22
200 200 200
21.5 21.5 21.5
180 180 180
21 21 21
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]
160 18 20 19 19 20.5 160 17.7 19.2 18.5 18.3 20.5 160 18 20 18 18 20.5
20 20 20
140 140 140
19.5 19.5 19.5
120 19
120 19
120
19 21 19 19 17.6 20.0 18.0 18.7 18 19 19 19 19
100 20 18.5 100 19.4 18.5 100 19 18.5
18 18 18
80 80 80
17.5 17.5 17.5
60 19 21 20 19
17
60 18.3 17.9 18.6 18.7
17
60 18 18 20 19
17
40 40 40
270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-4 Mass fraction burn (MFB) and burn duration of gasoline, ethanol and
methanol
122
On the first column of Figure 5-4, it shows the combustion phasing of all test points.
All the test points at 2000rpm fuel matrix testing were run at by means of CA50 at 8
crank angle degree ATDCf. Gasoline and methanol’s CA50 were controlled better
than ethanol, most of their test points’ CA50 are around 8CAD ATDC. For ethanol, at
injection timing 275CA BTDC rail pressure 200bar and injection start 300CA BTDC
rail pressure 100bar, their spark timings still has some space to be retarded to get
better fuel economies.
The burn duration of these three different fuels are shown in the second column of
Figure 5-4. Gasoline has the burn duration from 18.2CAD to 21.3CAD. The shortest
burn duration happened at the injection timing 275CA BTDC and rail pressure at
150bar while the longest burn duration happened at injection timing 300CA BTDC
and rail pressure at 50bar. Ethanol’s burn duration is between 17.6CAD to 20.0CAD.
At the injection timing of 300CA BTDC, the minimum burn duration is achieved with a
rail pressure at 200bar and the maximum point’s rail pressure is 100bar. The
minimum burn duration of methanol is 17.7CAD which happened at 275CA BTDC
rail pressure 100bar while the maximum is 20.38CAD which occurred with the
injection timing 300CA BTDC and rail pressure 150bar. Alcohol fuels got shorter
average burn duration than gasoline since their laminar flame speed are faster than
gasoline.
123
Gasoline Ethanol
THC THC THC Methanol
220 220 220
866 845 921 1153 386 569 495 521 491 307 383 377
200 1200 200 1200 200 1200
1100 1100 1100
180 180 180
1000 1000 1000
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 876 845 916 1073 900 160 404 348 455 650 900 160 524 298 354 378 900
140 800 140 800 140 800

700 700 700
120 600 120 600 120 600
916 853 903 1052 467 473 430 573 556 296 310 470
100 951 500 100 615 500 100 561 500
400 400 400
80 80 80
300 300 300
60 909 850 846 932 200 60 435 758 498 687 200 60 581 290 345 470 200
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
NOx NOx NOx
220 220 220
1285 1228 1283 1422 897.0 943.0 725.0 919.0 520 593 653 602
200 1500 200 1500 200 1500
1400 1400 1400
180 1300 180 1300 180 1300
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 1248 1299 1333 1370 1200 160 856.0 735.0 821.0 967.0 1200 160 551 466 619 458 1200
1100 1100 1100
140 140 1000 140 1000
1000
120 900 120 900 120 900
1110 1115 1286 1360 724.0 653.0 918.0 864.0 800 561 471 533 588 800
800
100 1217 100 709.4 700 100 597 700
700
80 80 600 80 600
600
500 500
500
60 1163 1168 1191 1332 60 745.0 784.0 856.0 767.0 400 60 614 447 567 554 400
400
40 40 40
270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-5 THC and NOx emission of fuel matrix
The THC released by gasoline, ethanol and methanol are shown in the first column
of Figure 5-5. The THC produced by gasoline has the minimum value of 845ppm
with the injection timing 300CA BTDC and rail pressure 150bar, the maximum THC
1153ppm happened at injection start 350CA BTDC and rail pressure 200bar. With
earlier injection, the THC increased from 932ppm to 1153ppm with the rail pressure
increased from 50bar to 200bar since the top of piston is close to the injector, and
increased rail pressure will lead fuel propagate further thus more fuel will
impingement on the piston. with the later injection, the THC decreased from 909ppm
to 866ppm with the rail pressure increasing from 50bar to 200bar. This is because
fuel droplet’s size decreases with the rail pressure. At the same rail pressure, both
early and late injections got higher THC than the other two injection timings since the
injection timing between early and late have the advantage not only avoiding fuel
impingement on the piston but also having enough timing for air-fuel mixing.
Ethanol got its lowest THC of 348ppm at injection timing 300CA BTDC, rail pressure
150bar. The maximum THC of 758ppm during the ethanol operation occurs with the
injection timing 300CA BTDC and rail pressure 50bar. At the same injection timing,
ethanol’s THC decreased with the increases of rail pressure since higher rail
124
pressure can improve the quality of the air-fuel mixture. What’s more, it is easier for
ethanol to evaporate in the combustion chamber so that less fuel impingement on
the top of the piston at high rail pressure conditions. The THC produced by methanol
reaches the minimum of 290ppm with an injection timing 300CA BTDC and rail
pressure 50bar and the maximum 581ppm happens at injection start 275CA BTDC
and rail pressure 50bar. The average THC released by methanol is lower than
ethanol and gasoline since its lowest boiling point. What’s more, considering that the
flame ionization detector (FID) is relatively insensitive to alcohol fuels, the effect on
HC emissions is less clear. HC emissions trends for alcohol fuel blends are well
documented to be misinterpreted when measured using conventional FIDs when
oxygenates are present in the exhaust stream. Compare with gasoline, the lowest
THC produced by ethanol decreases by 58.8% and methanol reduced by 65.7%.
The NOx released by these three fuels are shown in the second column of Figure 5-
5. Gasoline has its lowest NOx at injection timing 275CA BTDC rail pressure 100bar
which value is 1110ppm. The highest THC produced by gasoline located at injection
start at 350CA BTDC and rail pressure 200bar and its value is 1422ppm. The NOx
released by ethanol has the highest value 967ppm with the start of injection at
350CA BTDC and the lowest NOx 653ppm at injection timing 300CA BTDC and rail
pressure 100bar. Methanol’s NOx reaches its highest NOx emission of 653ppm when
the injection timing is 325CA BTDC and the rail pressure 200bar whilst the lowest
NOx happened at injection start 300CA BTDC and rail pressure 50bar.
At the same fuel quantity, methanol’s evaporation enthalpy is higher than ethanol
and that of gasoline so more heat is absorbed by the alcohol fuel’s evaporation than
gasoline thus gasoline’s exhaust gas temperature is higher than ethanol and
methanol. What’s more, because methanol’s lower heating value is lower than
ethanol and gasoline, more methanol was injected into the cylinder so that charge
cooling effect is more obvious which will lower the exhaust gas temperature thus
lowering the NOx in the exhaust gas. Compare with gasoline, the lowest NOx
produced by ethanol decreases by 41.2% and methanol reduced by 58.7%.
125
EU6 Gasoline
EU6 Gasoline_2000rpm 4.6bar_Fuel Matrix
Total PN PN 10-1000nm PN 23-1000nm
220 220 220
23993082
1e+08 1e+08
200 200 1e+08 200
9e+07
9e+07 9e+07
180 180 180 8e+07
8e+07 8e+07
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 160 160 5124340 7e+07
7e+07
7e+07 6e+07
140 140 6e+07 140
6e+07 5e+07
120 120 5e+07 120
5e+07 4e+07
4e+07
100 100 100 3e+07
4e+07 3e+07
2e+07
80 3e+07 80 2e+07 80
1e+07
60 98687576 2e+07 6017430758 98629880 1e+07 60 98442000
0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
Figure 5-6 Particle numbers released by gasoline of different regulations
Ethanol
Ethanol_2000rpm 4bar_Fuel Matrix
Total PM Total PN PM 10-1000nmPN 10-1000nm PN 23-1000nm
PM 23-1000nm
220 220 220
57391760 44827896
1e+08
200 1e+08 200 1e+08 200
9e+07
9e+07
180 9e+07 180 180 8e+07
8e+07
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 8e+07 160 160 17849136 7e+07
7e+07
7e+07 6e+07
140 140 6e+07 140
6e+07 5e+07
120 120 5e+07 120
83172600 66381000 4e+07
5e+07 4e+07
100 100 100 3e+07
4e+07 3e+07
2e+07
80 80 2e+07 80
3e+07 1e+07
60 60 1e+07 60 8396831
2e+07 0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
Figure 5-7 Particle numbers released by ethanol of different regulations
Methanol
Methanol_2000rpm 4.6bar fuel matrix
total PM Total PN PN 10-1000nm
PM 10-1000nm PN
PM 23-1000nm 23-1000nm
220 220 220
15601868.0
1e+08 1e+08 1e+08
200 200 200
9e+07 9e+07
180 9e+07 180 180
8e+07 8e+07
8e+07
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]
16041316196 16035583824 7e+07 160 7e+07

7e+07 6e+07
140 140 6e+07 140
6e+07 5e+07
5e+07
120 120 120 4e+07
60129792 5e+07 47369220 4e+07 9320546.0
100 100 100 3e+07
4e+07 3e+07
2e+07
80 3e+07 80 2e+07 80
1e+07
60 2e+07 60 1e+07 60 0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350
Figure 5-8 Particle numbers released by methanol of different regulation
126
The total PN produced by gasoline is shown in the first diagram of Figure 5-6. The
lowest total PN achieved by gasoline is 2.4e7 which happened at injection start
325CA BTDC and rail pressure 200bar. The maximum total PN of gasoline occurred
at injection timing 350CA BTDC and rail pressure 50bar and its value is 9.87e7. This
is because early injection causes fuel impingement on the top of the piston. What’s
more, low rail pressure has a negative effect on air-fuel mixing thus more fuel
impingement so that the total PN is high at this engine operation. In addition, low rail
pressure will cause the injection duration longer which has negative effect on particle
numbers. The second diagram shows the PN 10-1000nm which is the regulation of
ERUO 5. PN 10-1000nm of gasoline got the lowest value 1.74e7 at injection timing
275CA BTDC and rail pressure 50bar while the highest value is 9.86e7 which
located at injection start 350CA BTDC and rail pressure 50bar. Particle numbers of
23-1000nm is the regulation of EURO6. PN 23-1000 of gasoline has the lowest value
5.12e6 at injection start 275CA BTDC and rail pressure 150bar, and the highest PN
23-1000nm occurred at the same point with total PN and PN 10-1000 which value is
9.84e7.
Total PN produced by ethanol is shown in the first diagram of Figure 5-7. The lowest
total PN achieved by ethanol is 5.74e7 which happened at injection start 350CA
BTDC and rail pressure 200bar. The maximum total PN of gasoline occurred at
injection timing 275CA BTDC and rail pressure 100bar and its value is 8.32e7. The
second diagram shows the PN 10-1000nm. The maximum and minimum values of
PN 10-1000nm have the same location with total PN whose values are 6.64e7 and
4.48e7. PN 23-1000 of ethanol has the lowest value 8.40e6 at injection start 325CA
BTDC and rail pressure 50bar, and the highest PN 23-1000nm occurred at injection
timing 350CA BTDC and rail pressure 150bar which value is 1.78e7. For ethanol, the
best efficiency point (shown in figure 5-2) matches with the lowest PN size 23-
1000nm.
Total PN produced by methanol is shown in the first diagram of figure 5-8. The
lowest total PN achieved by methanol is 4.13e7 which happened at injection start
275CA BTDC rail pressure 150bar. The maximum total PN of methanol occurred at
injection timing 325CA BTDC rail pressure 100bar and its value is 6.01e7. PN 10-
1000nm of methanol got the lowest value 3.56e7 at injection timing 275CA BTDC rail
127
pressure 150bar while the highest value is 4.74e7 which is located at injection start
325CA BTDC rail pressure 100bar. PN 23-1000 of methanol has the lowest value
9.32e6 at injection start 325CA BTDC rail pressure 100bar, and the highest PN 23-
1000nm occurred at injection timing 300CA BTDC and rail pressure 200bar which
value is 1.56e7.
Figure 5-9 Size spectral density of gasoline ethanol and methanol injection start at
275CA BTDC
All the points’ size spectral density which injection start at 275CA BTDC are shown
in Figure 5-9. Apart from gasoline rail pressure at 50bar& 100bar and methanol
injection pressure at 150bar, other curves have a single peak between 10 to 40nm,
which mean that those particles are in nucleation mode. At gasoline rail pressure at
50bar& 100bar and methanol injection pressure at 150bar, the particle size
distributions are characterised by dual-peak curves, most of particles are formed by
nucleation mode with sizes between 5nm and 30nm, but there are also some
particles fomed by the accumulation mode in the range of 100nm.
128
300CA BTDC
The size spectral density of the points with an injection timing of 300CAD BTDC are
shown in Figure 5-10. Most of the particle’s size is between 10 to 100nm by means of
nucleation mode. Only the test point fueled by methanol which rail pressure is 200bar
has some particles in accumulation mode. It is clear that ethanol release more smaller
particles than methanol and gasoline at injection timing 300CAD BTDC.
129
325CA BTDC
The size spectral density of the points which injection timing is 325CAD BTDC are
shown in Figure 5-11. All the particles produced by ethanol and methanol whose size
is less than 100nm are belong to nucleation mode. Also, with the increase of rail
pressure, ethanol and methanol’s PN decrease since higher injection pressure result
in small fuel droplets which has positive effect on air-fuel mixing. Most particles
released by gasoline are in accumulation mode.
130
350CA BTDC
All the points’ size spectral density with injection start at 350CA BTDC are shown in
Figure 5-12. Ethanol and methanol’s PN curves are mano-peak curves, and most of
particle produced by them are smaller than 100nm so they are nucleation mode
particles. Most particles released by gasoline their size is larger than 100nm by
means of accumulation mode. What’s more, at this injection timing, gasoline
produces more particles than ethanol and methanol since this injection timing is too
early and gasoline cannot evaporate as fast as alcohol fuel. As a result of this, some
fuel impingement on the top of piston which causes big amount of particle emissions.
5.3.2 3000rpm 16bar IMEP Fuel Matrix Analysis
Compared with fuel matrix testing at 2000rpm 4.6bar IMEP, the 3000rpm 16bar
IMEP testing is more critical since the speed is higher which causes a shorter time
for air-fuel mixing and the load is higher will cause the fuel quantity injected into
cylinder more than 2000rpm so that charge cooling effect play a more important
during testing, there are more difference between gasoline, ethanol and methanol on
fuel efficiency and emissions. On the other hand, more fuel means that the injection
131
duration is longer than 2000rpm 4.6bar IMEP so the injection end timing of ethanol
and methanol are more retarded hence there is less time for air-fuel mixing. What’s
more, gasoline starts to knocking from 16bar IMEP according to the data from
3000rpm load sweep in the last chapter.
BSFC_LHV Corrected Gasoline BSFC_LHV Corrected Ethanol Methanol

BSFC_Corrected
220 220 220
IMOP EVC IVO IMOP EVC IVO 231
IMOP 228 235EVC 251 IVO 285
285 285 280
200 280 200 280 200
275
275 275 270
180 270 180 270 180 224 265
265 265 260
236 260 260 255
160 160 160
pRailCCP:1 [bar]
pRailCCP:1 [bar]
pRailCCP:1 [bar]
255 232 225 232 257
255 250
250 250 246
140 246 140 246 140 245
245 245 244
120 244 120 244 120 243
243 229 243 242
242 235 226 227 247
242 241
100 241 100 241 100 227
240 240
240 239
80 239 80 239 80 238
238 238 237
237 237 236
60 236 60 236
60 269 241 233 235
40 40 40
260 280 300 320 340 360 380 400 260 280 300 320 340 360 380 260 280 300 320 340 360 380 400
Combustion Phasing 50%MFB Combustion Phasing 50%MFB Combustion Phasing 50%MFB

220.0 220.0 220.0
9 9 12 15 9 8 8 8 8 9 9 6
200.0 6 200.0 6 200.0 6
8 8 8
180.0 8 180.0 8 180.0 9
10 10 10
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160.0 9 9 12 13 160.0 9 8 9 8 160.0 8 8 8 8
12 12 12
140.0 14 140.0 14 140.0 14
120.0 16 120.0 16 120.0 16
8 9 12 12 9 9 8 9 8 8 8 7
100.0 9 18 100.0 8 18 100.0 8 18
20 20 20
80.0 80.0 80.0
22 22 22
60.0 9 9 12 13 60.0 9 9 8 8 60.0 9 8 8 8
24 24 24
40.0 40.0 40.0
260.0 280.0 300.0 320.0 340.0 360.0 380.0 260.0 280.0 300.0 320.0 340.0 360.0 380.0 260.0 280.0 300.0 320.0 340.0 360.0 380.0
Figure 5-13 BSFC_cor and CA50 of gasoline, ethanol and methanol
The BSFC corrected by 42 of different fuels are shown in the first column of Figure
5-13. Gasoline’s BSFC_cor is shown in the first diagram. The highest BSFC_cor of
gasoline is 281g/kWh which happened at injection timing 350CA BTDC rail pressure
200bar, and the best BSFC point occurred at injection timing 300CA BTDC rail
pressure 150bar which value is 236g/kWh. The best BSFC point occurred at a
medium injection high rail pressure since the injection started not too early and too
late so that less fuel would impingement on the cylinder wall and top of the piston. In
addition, higher rail pressure will promote gasoline evaporation hence reducing
impingement so that the BSFC_cor decreases. At late injection points which injection
timing is 275 and 300CA BTDC, the BSFC_cor decreases with the increase of rail
pressure since higher rail pressure can increase the penetration of fuel droplets in
132
cylinder. Also, high rail pressure has a positive effect on air-fuel mixing by promoting
evaporating. Last but not the least, the injection duration is very long at low rail
pressure so there is no sufficient time for air-fuel mixing. At the injection timing 325
and 350CA BTDC, the BSFC_cor increases with the increase of rail pressure since
the piston is close to the top dead centre, increase rail pressure causes fuel
impingement on the top of the piston hence pool fire. Another reason for the high
BSFC_cor at early injection is because of knocking. Because early injection timing
will result in air-fuel mixture staying in a high temperature and pressure environment
for a long time which can lead to auto-ignition and knocking combustion. The 50%
mass fraction burnt is shown in the second column of Figure 5-13. The CA50
becomes more retarded with the earlier injection. At the same rail pressure, with the
advance of injection timing, BSFC_cor decrease firstly then increase. This is
because earlier injectioin is more likely to cause fuel impingement on the piston top
while late injection causes fuel impingement on cylinder wall.
The BSFC_cor of ethanol is shown in the second diagram of Figure 5-13. Compared
with gasoline, ethanol has better anti-knock characteristics so that all the points were
run at the MBT timing. The best BSFC_cor 229g/kWh happened at injection timing
350CA BTDC and rail pressure 100bar, and the worst BSFC_cor 261g/kWh occurred
at injection timing 275CA BTDC and rail pressure 50bar. Compared with gasoline,
ethanol’s lower heating value is lower so that ethanol’s injection duration is longer
than gasoline at the same rail pressure. At the point injection start 275CA BTDC and
rail pressure 50bar, ethanol got the shortest air-fuel mixing time since the injection is
too late and injection duration is too long so that the BSFC_cor at this point is higher.
In addition, the penetration of fuel is not further at low rail pressure which causes
wall wetting hence pool fire. At early injection points (injection start 350CA BTDC),
there is more time for air-fuel mixing. What’s more, even the piston is very close to
TDC at early injection points, but it’s easier for ethanol to evaporate than gasoline so
that less fuel impingement on the top of piston. At the same rail pressure, the
BSFC_cor decreases with the advance of injection timing since there’s more time for
air-fuel mixing. Compared with gasoline, the lowest BSFC_cor of ethanol is reduced
by 18.5%.
133
Methanol’s BSFC_cor is shown in the third diagram of Figure 5-13. The best
BSFC_cor 224g/kWh occurs at the reference point when the injection timing is
303CA BTDC and rail pressure at 166bar. This point’s injection strategies were
taken from 16bar IMEP 3000rpm load sweep. Beyond this reference point, the best
BSFC_cor 225g/kWh happens at injection timing 300CA BTDC and rail pressure
100bar, similar to the reference point.The worst BSFC_cor is 269g/kWh which
happened at injection timing 275BTDC and rail pressure 50bar. Because methanol
has the lowest lower heating value of these three fuels, it has the longest injection
duration and most retarded injection end timing. There is not only no enough time for
air-fuel mixing but also fuel droplets penetration not further and results in fuel
impingement on the cylinder wall hence pool fire. With retarded injection timings at
275 and 300CA BTDC, the BSFC_cor decreases with the increase of rail pressure
since high rail pressure decreases the size of fuel droplets and makes the air-fuel
mixture better. Moreover, because the injection timing is retarded, high rail pressure
makes fuel penetrate further into cylinder and less fuel will impingement on the top of
piston since the injection timing is late, the piston is far away from the top dead
centre. At injection timing 350CA BTDC, the BSFC_cor increases with the rail
pressure. That’s because methanol has the longest injection duration compared with
gasoline and ethanol at the same rail pressure, more fuel was injected into cylinder
and some of them impinge on the piston. Compared with gasoline, the lowest
BSFC_cor of methanol has reduced by 20.3%.
Figure 5-14 Indicate specific fuel consumption of gasoline, ethanol and methanol
134
The indicate specific fuel consumption (ISFC) of these three fuels are shown in
Figure 5-14. The analysis of ISFC is similar to BSFC_cor. Because methanol’s lower
heating value is the highest while that of gasoline is the lowest so that methanol has
the highest average ISFC since more fuel is needed to get the same power as
ethanol and gasoline.
Gasoline Ethanol Methanol
Figure 5-15 Burn duration of gasoline, ethanol and methanol
The burn duration of gasoline is shown in the first diagram of Figure 5-15. Gasoline
has its shortest burn duration 15.52CAD with the start of fuel injection at 300CA
BTDC and rail pressure at 200bar, and the longest burn duration 18.35CAD at
injection start at 350CA BTDC and rail pressure at 200bar. The shortest and longest
burn duration occurred at the same positions with the best and worst ISFC of
gasoline. At injection timing 300CA BTDC and rail pressure of 200bar, the injection
starts neither too early nor too late so that there is minimum fuel impingement on the
piston and cylinder wall. Also, high rail pressure decreases the size of fuel droplets.
At injection timing 300CA BTDC rail pressure 200bar, the air-fuel mixture quality is
the best among all gasoline’s fuel strategies so that all the heat can be released in
the shortest time for the best fuel efficiency.
The longest burn duration point occurs with the most retarded combustion phasing.
At late injection points (injection timing 275 and 300CA BTDC), the burn duration
decreases with the increase of rail pressure since the air-fuel mixture quality
increase. After that, all points whose injection timing before 325CA BTDC (including
135
325CA BTDC) have their burn duration longer than 17CAD since injection timing is
advanced and pre-ignition caused knocking at these control strategies.
Ethanol got its shortest burn duration 14.84CAD at injection timing 350CA BTDC and
rail pressure 200bar, and its longest burn duration 18.53CAD at injection timing
275CA BTDC and rail pressure 50bar. At early injection and low rail pressure
condition, because of ethanol’s lower LHV, the injection duration is longer than
gasoline so that there’s no sufficient time for air-fuel mixing properly, and that’s why
ethanol got its worst in-cylinder combustion result. At early injection high rail
pressure points, ethanol got its shortest combustion duration since there is sufficient
time for air-fuel mixing properly and high rail pressure reduce the size of fuel droplets
which can also improve air-fuel mixture quality. The laminar flow speed of ethanol is
faster than gasoline so that the average burn duration of fuel matrix testing is shorter
than gasoline.
Methanol got its shortest and longest combustion duration at the same fuel injection
timing and pressure as ethanol and the reasons why those points got the shortest
and longest burn duration are also similar to methanol. But methanol has the fastest
laminar flow rate of these three fuels so that methanol got the shortest average
combustion duration. The shortest combustion duration point could burn even faster
since its CA50 is at 6CAD ATDC, if the ignition timing is slightly retarded, the heat
could release in a shorter time.
136
Exhaust Gas Temp Exhaust Gas Temp Exhaust Gas Temp
220 220 220
668 670 685 654 655 652 649 655 646
700 700
200 200 700 200
690 690
690
180 660 680
180 647 180 680
680
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 664 667 684 655 670 160 649 655 652 656 160 670
670
660 660
140 140 660 140
650 650
650
120 640 120 120 640
661 668 677 658 642 652 660 659 640
100 675 630 100 650 100 630
630
620 620
80 80 620 80
610 610
610
60 669 667 680 672
600
60 634 646 645 659 60 600
600
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
CO2 CO2 CO2
220 220 220
12.98 13.16 13.08 12.54 13.6 12.86 13.00 13.00 13.06 13.6 12.76 12.96 12.24 11.40 13.6
200 13.4 200 13.4 200 13.4
13.2 13.2 13.2
180 13.04 180 12.92 180 12.88
13 13 13
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 12.98 13.28 13.06 12.56 12.8 160 12.78 12.98 12.98 12.98 12.8 160 12.60 12.92 12.30 11.22 12.8
12.6 12.6 12.6
140 12.4 140 12.4 140 12.4
12.2 12.2 12.2
120 12
120 12
120 12
12.92 13.26 13.14 12.56 12.64 12.94 13.00 12.96 12.42 12.76 12.48 11.80
100 13.14 11.8 100 13.06 11.8 100 12.86 11.8
11.6 11.6 11.6
80 11.4 80 11.4 80 11.4
11.2 11.2 11.2
60 13.00 13.12 13.16 12.98 60 12.28 12.66 12.80 13.12 60 11.60 12.26 12.20 12.28
11 11 11
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-16 Exhaust gas temperature and CO2
The exhaust gas temperature is shown in the first column of Figure 5-16. Exhaust
gas temperature can reflect the in-cylinder combustion temperature. The highest
exhaust gas temperature 685°C of gasoline happened at injection timing 325CA
BTDC and rail pressure 200bar which value is. The lowest exhaust gas temperature
654°C occurred at injection timing 350CA BTDC and rail pressure 200bar. The
highest point got the longest combustion duration since its spark timing was retarded
to avoid knocking. The lowest point even got a longer duration but its injection timing
is between the intake valve open (IVO) and exhaust valve close (EVC), some fuel
droplet was absorbed into the exhaust manifold so that lean combustion happened
at that moment, and the in-cylinder temperature is lower than stoichiometric
combustion. That’s why this point got the lowest exhaust gas temperature even
though its combustion duration is longer than that of the highest point.
Ethanol’s got its highest exhaust gas temperature 660°C at injection timing 325CA
BTDC and rail pressure 100bar, while the lowest exhaust temperature 634°C is at
injection timing 275CA BTDC and rail pressure 50bar. Ethanol’s average exhaust
gas temperature is lower than gasoline since the enthalpy of evaporation is higher
than gasoline and more heat is absorbed during the evaporation process compare
137
with the same quantity of gasoline. What’s more, more ethanol is injected into
cylinder and more heat is absorbed then causing lower in-cylinder temperature and
thus lower exhaust gas temperature.
Methanol got its highest exhaust gas temperature 646°C at injection timing 300CA
BTDC and rail pressure200bar, and its lowest exhaust gas temperature 583°C
happened at injection timing 275CA BTDC rail pressure 50bar. Methanol got the
lowest average exhaust gas temperature since it has the highest enthalpy of
evaporation.
The CO2 released by gasoline is shown in the second column of Figure 5-16. The
highest CO2 of gasoline happened at injection timing 300CA BTDC rail pressure
150bar which value is 13.26% vol. The lowest CO2 produced by gasoline 12.54% vol
located at injection timing 350CA BTDC rail pressure 200bar. The lowest CO2
happened at early injection high rail pressure point is because this injection timing is
between the intake valve open and the exhaust valve close, some fuel is absorbed in
the exhaust pipe and an lean combustion is formed as a result of this. What’s more,
increases lambda will cause a decrease in CO2 content in the exhaust gas. That’s
why the CO2 released by early injection points are lower than at other points. At
injection timing 350CA BTDC, CO2 decreases with the increase of rail pressure. This
is because the injection duration is shorter with the increase of rail pressure then
more fuel is absorbed into the exhaust pipe at higher injection pressure, there’s still
some time for fuel injection after EVC at low injection pressure points so that the
lambda at high rail pressure is greater than low rail pressure so that CO2 decreases.
The CO2 released by ethanol has the greatest value of 13.12%vol at injection timing
350CA BTDC and rail pressure 50bar, and the minimum CO2 happened at injection
timing 275CA BTDC and rail pressure 50bar which value is 12.28%vol. The average
CO2 released by ethanol is lower than gasoline since ethanol leads to lower specific
CO2 because of CO2’s energy-specific basis (g CO2/MJ). Compared with gasoline,
the lowest CO2 released by ethanol has dropped by 2.07%.
The CO2 released by methanol has the greatest value of 12.96%vol at injection
timing 300CA BTDC and rail pressure 2000bar, and the minimum CO2 occurred at
injection timing 350CA BTDC rail pressure 150bar which value is 11.22%vol. The
138
average CO2 released by methanol is lower than gasoline and ethanol since
methanol leads to 7% lower specific CO2 emissions compared to gasoline.
Compared with gasoline, the lowest CO2 released by methanol has dropped by
10.5%.
THC Gasoline THC Ethanol THC Methanol

220 220 220
561 528 1283 3422 461 397 481 580 445 255 266 851
200 3600 200 3600 200
3200 3600
180 180 3200 180
709 531 175 3200
2800 2800
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 781 641 1356 3262 160 566 375 471 411 160 564 308 254 730 2800
2400 2400 2400
140 2000 140 2000 140 2000
1600 1600
120 1600 120 120
1200 1200
609 656 811 2891 1200 917 544 457 419 267 307 241 466
100 754 100 637 100 190 800
800
800 400
80 80 400 80 0
400
0
60 709 787 698 2293 0 60 1073 588 575 376 60 258 339 263 226
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
NOx NOx NOx

220 220 220
1806.00 1727.00 1761.00 2526.00 1687.00 1663.67 1716.00 1687.00 1346.00 713.00 1147.00 1332.00
2600 2600 2600
200 200 200
2400 2400 2400
180 1852.00 180 1805.00 180 1227.00
2200 2200 2200
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 1844.00 2008.00 2096.00 2406.00 2000 160 1937.00 1520.00 1840.00 1726.00 2000 160 1388.00 1276.00 1180.00 1185.00 2000
1800 1800 1800
140 140 140
1600 1600 1600
120 1400
120 1400
120 1400
1740.00 1993.00 1969.00 2433.00 2371.00 2029.00 1802.00 1911.00 1450.00 1482.00 1310.00 1291.00
100 1987.00 1200 100 2120.00 1200 100 1373.00 1200
1000 1000 1000
80 80 80
800 800 800
60 1806.00 1998.00 1954.00 2204.00
600
60 1710.00 2038.00 2042.00 1624.00
600
60 1008.00 1362.00 1380.00 830.00
600
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-17 THC and NOx released by gasoline, ethanol and methanol
The THC released by gasoline is shown in the first diagram of Figure 5-17. Gasoline
got its lowest THC at injection timing 300CA BTDC and rail pressure 200bar which
value is 528ppm. The highest THC produced by gasoline is 3422ppm. The reason
why THC is high at injection early region is because at early injection timing the
distance between injector and piston is very short which causes a big amount of
unburned fuel impingement on the top of the piston. At injection start at 325CA
BTDC and 350CA BTDC, with the increase of rail pressure THC also increase since
fuel droplets penetrate further at high rail pressure, but the piston is close to the
injector at this stage so that THC increases. At injection starts at 275CA BTDC and
300CA BTDC, THC decreases with the increase of rail pressure since the piston is
far from the injector at this region, increase rail pressure causes the size of fuel
139
droplets smaller hence better to evaporate so that fewer fuel droplets impingement
on the cylinder wall and THC drops as a result of this.
The THC produced by ethanol is shown in the second diagram of Figure 5-17. The
maximum THC 1073ppm released by ethanol happened at injection start 275CA
BTDC and rail pressure 50bar, and the lowest THC is 375ppm which occurred at
injection timing 300CA BTDC rail and pressure 150bar. Because ethanol’s lower
heating value is lower than gasoline, ethanol got more retarded injection end timing
at late injection low rail pressure point. As a result of this, a big amount of unburned
ethanol impingement on the cylinder wall which causes high THC. At injection timing
275, 300 and 325CA BTDC, THC decreases with the increase of rail pressure since
the fuel droplets’ size is smaller at high rail pressure, and it’s easier for smaller
droplets to evaporate. At injection timing 350CA BTDC, THC increases with the
increase of rail pressure since the piston is close to the injector at early injection
points, high rail pressure causes ethanol impingement on the top of the piston which
results in high THC. From the 3D legend, it is clear that the average THC produced
by ethanol is lower than that of gasoline since ethanol is easier to evaporate so that
less fuel impingement on the cylinder wall and piston. Compare with gasoline, the
lowest THC released by ethanol dropped by 28.9%.
THC released by methanol is shown in the last diagram in column 1 of Figure 5-17.
The highest THC happened at injection timing 350CA BTDC rail pressure 200bar
which value is 851ppm. The lowest THC released by methanol is 175ppm occurred
at injection start 303CA BTDC rail pressure 166bar which is the reference point
(same injection strategy with 3000rpm 16bar during load sweep). There are two
reasons why the highest THC happened at early injection high rail pressure points.
The first is because fuel has further penetration at high rail pressure which causes a
big amount of methanol impingement on the top of the piston. The second reason is
that methanol’s lowest LHV result in the longest injection duration which will also
increase the fuel impingement on the piston when the piston is close to the injector.
The first reason has more influence on the THC so that at injection starts 350CA
BTDC, THC increases with the increase of rail pressure. Methanol got the lowest
average THC during 3000rpm fuel matrix testing since it’s the easiest to evaporate
so that less fuel impingement on the piston and cylinder wall compared with ethanol
140
and gasoline. The lowest THC released by methanol has reduced by 66.9%
compared with gasoline.
For gasoline, ethanol and methanol, the lowest THC all occurred at injection timing
300CA BTDC and rail pressure 150bar and 200bar since injection is not too early to
have fuel impingement on the piston, and not too late to produce fuel impingement
on the cylinder wall. What’s more, high rail pressure has a good effect on promoting
evaporate which can also decrease THC.
The NOx released by gasoline, ethanol and methanol is shown in the second column
of Figure 5-17. Gasoline has its highest NOx of 2526ppm which happened at
injection timing 350CA BTDC and rail pressure 200bar, and the lowest NOx which
occurred at injection timing 300CA BTDC and rail pressure 200bar. With the injection
timing advanced from 300CA BTDC to 325CA BTDC, NOx increases because the
knocking happened from injection start 325CA BTDC hence their spark timings are
retarded so that burn duration increases, and in-cylinder temperature increases
which causes NOx in the exhaust gas increase. The NOx at injection timing 350Ca
BTDC is higher than other injection timing is because this injection starts during the
overlap (when both intake valve and exhaust valve open), some fuel is absorbed into
the exhaust pipe then causes lean combustion so that the NOx is high at this region.
At injection start 350CA BTDC, the NOx increases with the increase of rail pressure
since more fuel is injected before the exhaust valve closed and lambda increases
with the rail pressure hence NOx increases.
The NOx released by ethanol has the maximum value 2371ppm at injection timing
275CA BTDC rail pressure 100bar, and its lowest NOx 1520ppm occurred at
injection timing 300CA BTDC rail pressure 150bar. From figure’s colour, it’s clear
that the average NOx value is lower than that of gasoline since the in-cylinder
temperature is lower. Ethanol has higher evaporation enthalpy, more heat is
absorbed during the evaporation process compared with the same amount of
gasoline. Also, because the lower heating value of ethanol is lower than gasoline,
more fuel is injected into the cylinder to get the same power as gasoline so that the
charge cooling effect is more effective than gasoline.
141
Methanol got its highest NOx 1482ppm at injection timing 300CA BTDC rail pressure
100bar and the best NOx point at injection timing 300CA BTDC rail pressure 200bar.
Methanol releases the lowest NOx of these three fuels since its best evaporation
enthalpy and lowest lower heating value. Compared with gasoline, the lowest NOx
released by ethanol is reduced by 11.9% and 58.7% for that of methanol.
O2 Gasoline O2 Ethanol O2 Methanol

220 220 220
1.70 1.56 1.54 2.05 1.35 1.36 1.35 1.32 1.30 1.23 1.60 2.54
200 3 200 3 200
2.8 2.8 3
180 1.51 180 1.39 180 1.14 2.8
2.6 2.6
2.6
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 1.68 1.58 1.60 2.07 2.4 160 1.42 1.36 1.43 1.40 2.4 160 1.35 1.17 1.44 2.71
2.4
2.2 2.2 2.2
140 140 140
2 2 2
120 120 1.8 120 1.8
1.8
1.62 1.50 1.57 2.04 1.50 1.51 1.43 1.42 1.6 1.47 1.31 1.41 2.22 1.6
100 1.44 1.6 100 1.35 100 1.13 1.4
1.4
1.4 1.2
80 80 1.2 80 1
1.2
1
60 1.56 1.54 1.51 1.60 1 60 1.71 1.65 1.53 1.26 60 2.36 1.60 1.59 1.50
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
CO CO CO
220 220 220
1.54 1.44 1.29 1.18 1.31 1.25 1.22 1.20 1.14 1.03 1.21 1.60
2 2 2
200 200 200
1.9 1.9 1.9
180 1.49 180 1.36 180 0.95
1.8 1.8 1.8
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]
160 1.60 1.40 1.36 1.17 1.7 160 1.35 1.24 1.23 1.23 1.7 160 1.20 0.97 1.20 1.81 1.7
1.6 1.6 1.6
140 140 140
1.5 1.5 1.5
120 1.4
120 1.4
120 1.4
1.67 1.39 1.36 1.24 1.45 1.23 1.26 1.21 1.36 1.16 1.16 1.50
100 1.42 1.3 100 1.29 1.3 100 0.98 1.3
1.2 1.2 1.2
80 80 80
1.1 1.1 1.1
60 1.62 1.41 1.38 1.11 60 1.62 1.47 1.35 1.13 60 1.48 1.48 1.45 1.35 1
1 1
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-18 O2 and CO released by gasoline, ethanol and methanol
The O2 data is shown in the first column of Figure 5-18. Gasoline got its highest O2
2.07 Vol% at injection timing 350BTDC and rail pressure 150bar, and the lowest O 2
happened at injection timing 320CA BTDC and rail pressure 90bar which value is
1.44 Vol%. The point which got the lowest O2 is another reference point which
injection timing and rail pressure is the same as the idle mode. The O2 data at
injection start 350CA BTDC can also reflect that the in-cylinder combustion states
are oxygen-rich since the O2 released at this region is higher than that of other
injection timings.
Ethanol has its highest O2 1.71 vol% at injection timing 275CA BTDC and rail
pressure 50bar since late injection timing and low rail pressure which causes long
142
injection duration result in fuel impingement so that this point is oxygen-rich. The
lowest O2 occurred at injection timing 350CA BTDC rail pressure 50bar which value
is 1.26vol%.
The O2 released by methanol has its highest value of 2.71vol% at injection timing
350CA BTDC and rail pressure 150bar. The lowest O2 occurred at the same position
as gasoline which value is 1.13vol%. There are two oxygen-rich regions during
methanol’s testing. The first one is at injection timing 275CA BTDC rail pressure
50bar since low injection pressure results in longer injection duration which causes
big amount of fuel impingement on the cylinder wall, then there’s excess oxygen.
Another high oxygen region is at injection timing 350CA BTDC. There are two
reasons why this region has excess oxygen. The first one is that this injection timing
is between the intake valve open and exhaust valve close which causes fuel droplets
absorbed into the exhaust pipe. The second reason is that at high rail pressure, fuel
droplets have further penetration but the piston is very close to the injector at early
injection so that fuel impingement on the top of piston so that there is excess
oxygen.
The second column of Figure 5-18 shows the CO released by gasoline, ethanol and
methanol. Gasoline has its highest CO at injection timing 275CA BTDC and rail
pressure 100bar, while the lowest CO occurred at injection timing 350CA BTDC rail
pressure 50bar which value is 1.11vol%. Ethanol has its best CO 1.13vol% produced
at injection timing 350CA BTDC and rail pressure 50bar, and the worst CO 1.62vol%
happened at injection timing 275CA BTDC rail pressure 50bar. The CO released by
methanol has its lowest value 0.95vol% at injection timing 303CA BTDC and rail
pressure 166bar. The highest CO released by methanol happened at injection timing
350CA BTDC and rail pressure 150bar which value is 1.81vol%.
143
throttle position% Gasoline throttle position% Ethanol throttle position% Methanol
220 220 220
37.83 37.85 37.80 66.36 27.88 26.94 26.91 31.78 17.74 17.31 19.02 31.16
200 66 200 66 200
61 61 66
180 37.85 180 23.30 180 17.13 61
56 56
56
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 37.82 37.83 37.77 54.32 51 160 27.91 27.94 26.91 31.75 51 160 17.74 16.96 18.22 30.38
51
46 46 46
140 140 140
41 41 41
120 120 36 120 36
36 16.76
37.84 37.82 37.79 50.25 31.65 27.91 26.91 30.93 31 17.73 18.21 24.07 31
100 37.85 31 100 23.19 100 17.20 26
26
26 21
80 80 21 80 16
21
16
60 37.80 37.82 37.80 44.61 16 60 37.74 27.88 26.90 28.58 60 27.67 18.91 18.37 19.59
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Spark Timing [Deg BTDC] Spark Timing [Deg BTDC] Spark Timing [Deg BTDC]
220 220 220
12.22 11.69 9.67 9.17 13.22 13.23 11.66 12.23 13.19 12.19 13.19 14.21
200 18 200 18 200 18
180 17 180 17 180 17

12.19 13.22 12.18
16 16 16
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 12.17 11.71 9.66 10.19 160 13.19 13.23 11.22 11.22 160 13.69 12.68 12.70 14.20
15 15 15
140 14 140 14 140 14
120 13 120 13 120 13

13.69 11.70 9.71 10.21 15.23 12.72 12.73 10.73 15.69 13.70 12.69 13.69
12 12 12
100 12.14 100 13.22 100 12.69
11 11 11
80 10 80 10 80 10
60 14.67 12.71 10.23 9.69 9 60 17.20 15.71 14.18 12.67 9 60 18.19 17.19 14.70 14.19 9
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-19 Throttle position and spark timing
The throttle position of gasoline, ethanol and methanol are shown in the first column
of Figure 5-19. From the figure, it’s easy to find that more throttle is opened for
gasoline than ethanol and methanol since gasoline’s stoichiometric air-fuel ratio
(14.7:1) is higher than that of ethanol (9:1) and methanol (6.5:1) so that more air is
needed to get stoichiometric combustion for gasoline. What’s more, because ethanol
and methanol have higher evaporation enthalpy as well as lower LHV so that the
better charge cooling effect will result in lower intake air temperature so that
methanol has the highest intake air density, and that of ethanol is also higher than
gasoline. That’s why gasoline’s throttle position is more open than ethanol and
methanol. For gasoline, the widest throttle open point is at injection timing 350CA
BTDC and rail pressure 200bar. At injection timing 350CA BTDC, their throttle was
open wider than other injection timings since engine has knocking at that injection
timing so that their spark timing were retarded. Because these points were not
running at MBT, more air and fuel are needed to get the same power as the other
test points. Ethanol has its widest open throttle at injection timing 275CA BTDC and
rail pressure 50bar since early injection cannot provide enough time for fuel
evaporation. What’s more, low injection pressure results in bigger fuel droplet size
which also gives negative effect on evaporation. Combining the above reasons, the
144
charge cooling effect of this point is not good enough which decreases the intake
air’s density, and the throttle opens more as a result of this. Either advancing the
injection timing or increasing rail pressure will reduce the throttle position which is
also shown in Figure 5-19.
LNV_1[%] Gasoline LNV_1[%] Ethanol LNV_1[%] Methanol

220 220 220
95.31 94.12 96.59 92.36 93.45 95.15 97.43 96.40 96.24 95.11 93.91 96.52
200 200 200 100
100 100
90 90 90
180 95.31 180 95.79 180 94.65
80 80 80
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 95.17 96.95 97.74 89.71 160 93.96 95.90 96.35 96.87 70 160 94.43 96.04 94.91 94.39
70 70
60 60 60
140 140 140
50 50
50
120 40 120 40 120
95.67 97.48 97.41 96.33 94.50 95.57 95.25 96.91 96.38 94.95 96.96 95.08 40
30 30
100 92.60 100 94.83 100 97.47 30
20
20
80 80 10 80 20
10
0 10
60 96.10 96.80 97.55 97.67 0 60 94.34 96.14 94.22 97.01 60 94.05 96.76 96.82 97.01
0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
combustion Stability COV IMEP combustion Stability COV IMEP combustion Stability COV IMEP
220 220 220
1.27 1.82 1.05 1.73 1.73 1.22 0.77 0.91 1.36 1.79 1.56 1.16
200 4 200 4 200 4
180 1.85 3.5 180 1.22 3.5 180 1.28 3.5
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 1.43 0.96 0.71 1.32 3 160 1.65 1.08 0.97 0.92 3 160 1.39 1.09 1.21 1.74 3
2.5 2.5 2.5
140 140 140
2 2 2
120 120 120
1.50 0.80 0.72 1.11 1.5 1.41 1.45 1.00 1.21 1.5 1.40 1.08 0.89 1.53 1.5
100 3.31 100 1.21 100 0.89
1 1 1
80 0.5 80 0.5 80 0.5
60 1.43 0.87 0.82 0.79 0 60 1.67 1.08 1.47 1.14 0 60 2.19 1.14 1.14 1.15 0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-20 LNV and combustion stability of gasoline, ethanol and methanol
The LNV is a parameter which shows the trend to misfire and there is more chance
to have misfire when LNV is lower than 90% so that the lowest valve of LNV is more
focused. Gasoline got its lowest LNV at injection timing 350CA BTDC rail pressure
150bar which value is 89.71%, and it’s more potential for this point to have misfire.
Only this point has its LNV lower than 90% because injection timing is early, air-fuel
mixture with high temperature and high pressure stay in the cylinder for a long time
hence cause knocking, the spark timing of this point is needed to be slightly
retarded. The lowest LNV for ethanol and methanol are 93.45% and 93.91%, both of
them is greater than 90% so there’s a small chance for ethanol and methanol to
have misfire.
145
The combustion stability is shown in the second column of Figure 5-20. Only one point
fuelled by gasoline with injection timing 320CA BTDC and rail pressure 90bar has its
Coefficient of Variation (COV) IMEP greater than 3 as a result of knocking, the spark
timing of this point should be slightly retarded to reduce COV IMEP less than 3.
Pmax Gasoline Pmax Ethanol Pmax Methanol

220 220 220
83.99 84.38 76.11 72.22 82.06 82.70 83.50 83.01 83.95 83.27 81.70 88.40
200 200 200 90
90 90
180 84.69 87.5
180 81.96 87.5 180 83.54
87.5
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 83.16 83.38 77.48 73.85 85 160 82.42 82.48 81.15 82.14 85 160 84.96 85.46 83.62 82.86 85
82.5 82.5 82.5

140 140 140
80 80
80
120 77.5
120 77.5 120
83.58 81.41 75.59 73.97 82.34 79.29 81.51 82.12 85.61 84.71 84.72 85.49 77.5
100 82.44 75 100 81.72 75 100 83.13
72.5 75
72.5
80 80 80
70 70 72.5
60 80.80 81.86 75.49 73.80 60 80.52 80.85 82.87 84.49 60 80.73 84.56 83.67 83.49 70
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Rmax Rmax Rmax

220 220 220
4.72 5.06 3.79 3.34 4.79 4.84 5.07 5.25 5.11 4.97 4.67 5.59
200 6 200 6 200 6
5.5 5.5 5.5
180 4.88 5 180 4.76 5 180 5.06 5
4.5 4.5 4.5
Rail Pressure [bar]
Rail Pressure [bar]
Rail Pressure [bar]

160 4.56 5.20 3.99 3.54 160 4.58 4.91 4.60 5.12 160 5.15 5.32 4.86 4.52
4 4 4
140 3.5 140 3.5 140 3.5
3 3 3
120 2.5 120 2.5 120 2.5
4.35 4.79 3.83 3.48 2 4.23 4.54 4.68 4.93 2 4.85 5.18 4.79 4.72 2
100 4.48
1.5
100 4.67
1.5
100 4.72
1.5
80 1 80 1 80 1
0.5 0.5 0.5
60 4.21 4.49 3.91 3.39 0 60 4.08 4.20 4.52 4.77 0 60 4.35 4.62 4.64 4.42 0
40 40 40
270 280 290 300 310 320 330 340 350 270 280 290 300 310 320 330 340 350 360 270 280 290 300 310 320 330 340 350 360
Figure 5-21 Peak in-cylinder pressure and maximum pressure rise rate
The peak in-cylinder pressure and the maximum pressure rise rate of gasoline,
ethanol and methanol opearations are shown in Figure 5-21. Gasoline has its lowest
Pmax at injection timing 350CAD BTDC rail pressure 200bar since its combustion
phasing is retarded as a result of knocking. Because of ethanol and methanol’s anti-
knock characteristic, their combustion phasing are more advanced than gasoline so
their average Pmax are higher than that of gasoline. The Rmax of ethanol and methanol
are also higher than gasoline and the description is the same with Pmax.
146
Figure 5-22 Total particle number of gasolines, ethanol and methanol
The total particle number of gasoline and alcohol fuel are shown in Figure 5-22.
Gasoline got its lowest PN at injection timing 275CAD BTDC and rail pressure
150bar which value is 1.36e7. The highest PN released by gasoline is at injection
timing 350CAD BTDC rail pressure 150bar, and the maximum PN is 2.7e8. For
gasoline, with the advance of injection timing, the total PN increase because of early
injection causes fuel impingement on the cylinder wall. What’s more, with the
increase of rail pressure, particles produced by gasoline increase since fuel
propagates further into the cylinder which causes fuel impingement on the top of the
piston hence the particle emission increase. Ethanol has its lowest PN at injection
timing 300CAD BTDC rail pressure 200bar which value is 9.4e5, and its highest PN
at injection timing 275CAD BTDC and rail pressure 50bar. There are two reasons
why the late injection and low injection pressure have the highest PN. The first one is
that ethanol’s injection timing is longer than gasoline so its injection end timing is
more retarded than gasoline, resulting in less timing for air-fuel mixing. The second
reason is that lower rail pressure causes larger fuel droplets which has negative
effect on air-fuel mixing.
5.4 Summary
In this chapter, a series of fuel injection parameter characteristic matrix testing was
carried out for all three fuels in order to evaluate the effect of injection strategies on
the combustion characteristic, efficiency and emissions. The experiment was
147
separated into two groups, the first one was running at 2000rpm 4.6bar IMEP, and
the second one was running at 3000rpm 16bar IMEP. All the tests were running at
stoichiometric and the 50% mass fraction burn is controlled at MBT except knocking.
The injection parameter was found has little impact on the combustion and emission
at the 2000rpm 4.6bar IMEP operation, and combustion characteristic and
combustion efficiency are similar between gasoline, ethanol and methanol. Ethanol
and methanol’s combustion efficiency is slightly higher because of their slightly faster
flame speed. THC released by alcohol fuel is lower than gasoline since alcohol fuel’s
better evaporation character and FID. What’s more, NOx produced by ethanol and
methanol is also lower since more heat is absorbed during evaporation. The
particulate emission of alcohol fuels is also lower. For ethanol, the best efficiency
point matches the lowest PN size 23-1000nm, and methanol's best efficiency
condition also has the lowest total PN and PN 10-1000nm.
At the 3000rpm 16bar IMEP fuel operation, there are more differences between
alcohol fuel and gasoline. Because ethanol and methanol’s higher RON values, they
don’t have knocking combustion. But alcohol fuel’s LHV is lower than gasoline, so
their injection durations are longer than gasoline hence there’s less time for air-fuel
mixing. Therefore, ethanol and methanol have their lowest combustion efficiency at
injection timing 275CAD BTDC and rail pressure 50bar because air fuel does not mix
properly. But ethanol and methanol’s peak efficiency area are still larger than
gasoline. Alcohol fuel still produces fewer emissions than gasoline during the
3000rpm 16bar IMEP fuel matrix testing.
148
6. Chapter Six: Analysis of the Spark Retard Capability of Alcohol Fuels for
Fast Catalyst Light-off
6.1 Introduction
Nowadays, most modern ICE vehicle calibrations utilise a catalyst light off strategy to
be able to conform to modern legislative drive cycle regulations. The final study on
the alcohol fuels was a lower temperature spark retard sweep under steady state
conditions. This test is aimed to find the most retarded spark timing and the
maximum exhaust gas temperature during the spark sweep for the catalyst light off.
6.2 Experimental Set-up
The engine was operated at 1500rpm 2bar IMEP with the coolant temperature, oil
temperature as well as intake temperature kept below 40°C, as shown in Table 6.1.
As the spark timing is retarded, combustion becomes more unstable. The test was
continued until the coefficient of variation (COV) IMEP achieved 15% then stopped.
Table 6-1 Fuel matrix test point
Speed& IMOP EMOP SOI Rail Spark Lambda Coolant Oil

load (CAD Pressure(bar) timing temperature temperature
BTDCf)
1500rpm
120 110 320 60 DBL 1 40°C 40°C
2bar
149
6.3 Result and discussion
The IMEP standard deviation which can reflect the combustion stability is shown in
Figure 6-1a. Gasoline’s IMEP_SD increases from 0.02bar to 0.30bar with the spark
timing retarded. That of ethanol increases from 0.03bar to 0.24bar and methanol’s
IMEP_SD increases from 0.02bar to 0.19bar as the spark timing is retarded.
Ethanol’s stability is slightly worse than gasoline when the combustion phasing
CA50, is retarded from 8CAD ATDC to 15CAD ATDC. After that, ethanol performed
better during the rest of the test. Methanol has the best combustion stability at all test
points.
Figure 6-1 Combustion characteristic of cold spark sweep
The spark timing is shown in Figure 6-1b. The spark timings of gasoline and ethanol
are retarded from 23.6CA BTDC to 5.4CA ATDC with the CA50 retarded from 8CAD
ATDC to 45CAD ATDC. The spark timing of ethanol can be more retarded than
150
gasoline since the laminar flame speed of ethanol is faster than gasoline. Thus to
achieve the same CA50, ethanol can be ignited later than gasoline. Methanol’s spark
timing is retarded from 18.5CAD BTDC to 7.6CAD ATDC during the spark timing
sweep. The spark timing of methanol can be more retarded than ethanol and
gasoline since methanol has the fastest laminar flame speed.
The indicated specific fuel consumption is shown in Figure 6-1c. The ISFC of
gasoline, ethanol and methanol increase from 350.3g/kWh to 537.5g/kWh,
490.8g/kWh to 803.6g/kWh and 651.5g/kWh to 1010.1kWh with the same trend
since the combustion efficiency becomes lower with the retarded CA50. Coefficient
of variation (COV) IMEP (shown in Figure 6-1d) is another parameter which can
reflect combustion stability. For this test, COV IMEP was limited under 15% to
confirm engine stability. At CA50 8CAD ATDC, all these fuels have very good
combustion stability with cold coolant and oil. At the end of the test, the COV IMEP
of gasoline, ethanol and methanol are 14.3%, 12.1% and 9.7%. It seems there is still
some margin for ethanol and methanol’s spark timing to be retarded more, but
misfire and partial burn occurred when their CA50 were retarded to 50CAD ATDC.
The burn duration is shown in Figure 6-1e. The burn duration of gasoline increases
from 19.8CAD to 30.9CAD. Ethanol’s burn duration is shorter than gasoline which
increases from 20.3CAD to 29.8 CAD since it has a faster laminar flame speed. At
CA50 11CAD ATDC, ethanol has a longer burn duration than gasoline since its
CA50 is more retarded. Methanol has the fastest laminar flame speed of these fuels
so that its burn duration is always shorter than ethanol and gasoline which increases
from 17.5CAD to 28.1CAD at CA50 40CAD ATDC. The knocking intensity (KI) of
gasoline, ethanol and methanol are shown in Figure 6-1f. All KI values of these fuels
are under 1 which means there’s no knocking happened during testing.
151
Figure 6-2 NV, maximum in-cylinder pressure and maximum pressure rise rate
The lowest normalized value (LNV) is a parameter which can reflect the trend to
misfire which is shown in Figure 6-2a. Gasoline and methanol have their LNV higher
than 90% before CA50 11CAD ATDC, after this point their LNV decreases to 84.8%,
86.3% which means there is more chance to misfire. Methanol has its LNV higher
than 90% before and includes CA50 15CA ATDC whose value is 91.0%. Gasoline
and methanol have their lowest LNV at CA50 40CAD ATDC whose values are
40.1% and 71.5%, while ethanol has its lowest LNV at CA50 45CAD ATDC whose
value is 59.4%. Methanol has the lowest chance to misfire with the spark timing
retarded, and it’s more potential for gasoline has misfire after CA50 20CAD ATDC.
The maximum in-cylinder pressure and maximum pressure rise rate are shown in
Figure 6-2b and Figure 6-2c. After CA50 15CA ATDC, these fuels’ Pmax curves
almost align with each other, from CA50 8 to 15CAD ATDC, methanol has higher
Pmax since it has the most retarded spark timing which is most close to the top dead
152
centre. The Rmax of ethanol decreases from 0.70bar/dCA to 0.35bar/dCA, and
there’s a small difference between these fuels.
baseline_cold start swipe.nc (CA50,CA10)

ethanol_cold start 2_Sheet1.nc (CA50,CA10)
Brunel Single Cylinder_Ethanol_1500rpm 2bar_Cold spark sweep methanol_cold start _Sheet1.nc (CA50,CA10)}
13.8 2200
13.6 2000
1800
13.4
1600
CO2 [%Vol]
THC [ppm]
13.2
HC [ppm]
CO2 [%]
1400
13.0 1200
12.8 1000
800
12.6
600
12.4 400
12.2 200
0.80 2.2
0.75 2.0
1.8
CO (High) [%]
0.70
CO [%Vol]
O2 [%Vol]
O2 [%] 1.6
0.65
1.4
0.60
1.2
0.55 1.0
0.50 0.8
500 14.5
450
400 14.0
It would be useful to calculate AFR..
AFR (Spindt) [#]
350 13.5
NOx [ppm]
300
250 13.0 No PN # Data
200
150 12.5
100 12.0
50
0 11.5
5 10 15 20 25 30 35 40 45 5 10 15 20 25 30 35 40 45
CA50 [dCA ATDCF] CA50 [dCA ATDCF]
Figure 6-3 Emissions released by gasoline, ethanol and methanol
The gaseous emissions are shown in Figure 6-3. The CO2 released by gasoline
increased slowly from 13.46%vol to 13.72%vol with the spark timing retarded. The
CO2 released by ethanol keeps around 13.3%vol, and that of methanol keeps
around 12.6%vol. The CO released by gasoline reduces slowly from 0.69%vol to
0.59%vol. Ethanol’s CO decreases from 0.69%vol to 0.6%vol with the CA50 retarded
from 8CAD ATDC to 20CAD ATDC. After that, it increases to 0.78%vol at CA50
45CAD ATDC. The CO produced by methanol keeps around 0.5%vol from CA50
8CAD ATDC to 30CAD ATDC, then increases to 0.73%vol at CA50 CAD ATDC. The
reason why CO of ethanol and methanol increase is because of incomplete
combustion as a result of poor atomisation of alcohol fuels under low temperature.
The NOx released by gasoline dropped from 460ppm to 158ppm with the CA50
retarded from 8CAD ATDC to 45CAD ATDC. The NOx released by ethanol and
153
methanol are aligned with each other which is reduced from 245ppm to 50ppm. With
the retarded of spark timing, all fuels’ NOx decreases since in-cylinder thermal
dynamic environment not good when the spark timing is very retarded. With the
same amount of gasoline, ethanol and methanol, ethanol and methanol’s
evaporation enthalpy are higher than gasoline, and the charge cooling effect of
ethanol and methanol are better than gasoline which decreases the peak in-cylinder
temperature. What’s more, alcohol fuels’ lower heating value are lower than gasoline
so that their charge cooling effect is more effective than gasoline then their in-
cylinder temperature are lower than gasoline. As a result of this, ethanol and
methanol produce less NOx but more CO emissions than gasoline. The THC
released by gasoline reduced from 1624ppm to 882ppm with the retarded of spark
timing. Ethanol’s THC keeps around 1280ppm from CA50 8CAD ATDC to 30CAD
ATDC. After that, at late spark timing points, its THC increases to 2102ppm because
of incomplete combustion and poor air-fuel mixing. The THC released by methanol
drops from 1576ppm to 390ppm at CA50 30CAD ATDC and it keeps stable after
that. The THC of methanol at a low level at late combustion phasing is because
methanol’s O/C ratio is the highest of these three fuels. In general, the oxidation and
evaporation of HC is promoted by higher in-cylinder temperature as a result of
retarded spark timing.
154
Figure 6-4 Particle numbers data of gasoline, ethanol and methanol
The three different size ranges of particle numbers are shown in Figure 6-4. Because
of equipment’s issue, only part of gasoline’s data was collected. PN 10-1000nm is
the regulation of EURO 5, and PN 23-1000nm is the regulation of EURO 6. The total
PN is shown in Figure 6-4 a), the total PN of gasoline reduces from 8.9e7 to 4.9e7
with the CA50 retarded from 8CAD ATDC to 30CAD ATDC. The particle released by
ethanol which numbers are around 2e7 during the cold start spark timing sweep.
Methanol’s total PN decreases from 3.9e7 to 2.0e7 at CA50 25CAD ATDC. After
that, it increases and stays around 2.8e7 from CA50 30CAD ATDC to 40CAD ATDC,
then it drops to 1.1e7 at CA50 45CAD ATDC. PN 10-1000nm is shown in figure6-4
b). Gasoline’s PN 10-1000nm reduces from 8.8e7 to 4.8e7 during testing. The PN
10-1000nm of ethanol and methanol are around 2e7. PN 23-1000nm of gasoline
drops from 8.8e7 to 3.9e7, that of ethanol and methanol keeps around 5e6 during
155
the whole testing. There is little difference between the PN of each size group of
gasoline which mean that most particle released by gasoline whose size is larger
than 23nm, and there are some small particles released by ethanol and methanol.
Figure 6-5 Size spectral density of gasoline, ethanol and methanol
Figure 6-5 shows the particle size distribution of these three fuels. The total PN
released by gasoline is greater than methanol and ethanol. Most particles released
by gasoline whose size is greater than 100nm by means of accumulation mode. With
the spark timing retarded, the particulate in nucleation mode increase but there’s a
decrease in accumulation mode. The particulate emission produced by methanol is
more than that of ethanol.
156
6.4 Summary
These tests were undertaken to assess whether alcohol fuels are able to perform as
good as gasoline under catalyst light-off conditions. The result shows that both
ethanol and methanol could operate with more retarded spark timings to increase the
exhaust gas temperature for catalyst-light-off. What’s more, the particle emissions
released by alcohol fuels are lower and smaller.
157
7. Chapter Seven: Conclusion
7.1 Introduction
In this experimental study, the performance, efficiency and emissions of a direct

injection spark ignition engine operating with ethanol and methanol have been
measured and analysed, and then compared with EU6 E10 gasoline (RON 95). The
testing was carried out on a single cylinder highly downsized spark ignition engine
with no hardware change for different fuels.
Three series of testing were performed, including load sweeps at three engine
speeds, fuel injection strategy matrix and cold spark sweep for methanol, ethanol
and baseline gasoline fuel. Combustion characteristics and emissions, particle
numbers and particle size spectral density results, are analysed and discussed.
The Load sweep testing (2000rpm & 3000rpm & 4000rpm) aims to have general
understanding of the effect on engine performance, efficiency and emissions by
ethanol and methanol at different speeds and loads. In particular, the potential
benefits of a spark ignition engine operating with methanol or ethanol at high load
conditions were assessed because of their higher-Octane number, charging cooling
effect and faster laminar flame speeds.
In the second series of testing, the injection timing and injection pressure sweep is
used to find the optimal injection timing and pressure which has the best fuel
efficiency and the point which get the lowest emission. (2000rpm 4.6bar IMEP &
3000rpm 16bar IMEP)
Finally, a cold start spark timing sweep testing was carried out to identify the
capability of alcohol fuels for fast catalyst light off through retarded spark timings.
158
7.2 Conclusions
7.2.1 Engine Performance, Combustion and Emissions of Alcohol Fuels
For the 2000rpm experiment, ethanol and methanol performed similarly to the E10
gasoline in terms of efficiency, combustion characteristics and emissions, because
the engine was operated at part load conditions limited to 11bar IMEP. Although
ethanol and methanol’s burn speed are faster than gasoline, the combustion
durations are similar for all three fuels when the spark timings were set to MBT.
There’s a reduction of CO2 released by ethanol and methanol. What’s more, the
THC released by ethanol and methanol is also lower than gasoline, which is due to
the lower FID response to partially oxidised alcohols).
For the 3000rpm load sweep, the load range was extended and varied from 2bar to
28bar IMEP. Below 16bar IMEP, the indicated thermal efficiency of gasoline, ethanol
and methanol are similar since all fuels’ CA50 were controlled at 8 CAD ATDC when
the MBT spark timing was applied. At loads higher than 16bar IMEP, spark timings
were retarded due to different reasons for gasoline and alcohol fuels. For gasoline,
its spark timing was retarded to avoid knocking combustion. For alcohol fuels, the
most advanced spark timings were limited by the maximum peak in-cylinder
pressure 120bar and the maximum rate of the pressure rise of 6 bar/CAD at 16bar
IMEP. Gasoline’s spark timing had to be retarded more than alcohol fuels to prevent
knocking at all higher load operations. As a result of this, ethanol and methanol led
to higher thermal efficiency than gasoline at high loads. Methanol achieved the
highest brake thermal efficiency 37.9% at 18bar IMEP, and ethanol’s highest BTE
was 37.7% at 18bar IMEP as well. Gasoline's best BTE was 36.5% happened at
10bar IMEP.
For gasoline, over-fuelling was introduced at 26bar IMEP and 28bar IMEP in order to
keep the exhaust gas temperature below the limit of 780 Deg C. Because of their
higher enthalpy of evaporation enthalpy, ethanol and methanol combustion took
place at lower temperature so that their exhaust gas temperature was always below
the limit. As a result of this, over-fuelling was not required for ethanol and methanol.
159
The lower combustion temperature also led to lower NOx emissions released by
ethanol and methanol.
At 3000rpm, HC released by alcohol fuels are also lower than gasoline because of
two reasons. The first one is the lower response of FID to the partially oxidised
alcohol fuels and the other alcohol fuels evaporates faster than gasoline when the
engine is hot so that they can make the air-fuel mixture quality better. In particular,
ethanol and methanol engine operations produced significantly less particles than
gasoline. There’s up to 90% reduction of PN number by ethanol and methanol.
What’s more, the particles produced by ethanol and methanol are of smaller sizes
than that of gasoline, but smaller size particle has more effect on human’s health.
The disadvantage of ethanol and methanol is that their lower heating value are lower
than gasoline which causes longer injection duration and higher fuel consumption.
The results obtained during the 4000rpm engine experiments are similar to those of
the 3000rpm load sweep.
7.2.2 Effect of fuel injection timing and pressure
The fuel injection strategy studies were carried at two operating conditions: low load
operation at 2000rpm 4bar IMEP and high load operation at 3000rpm 16bar IMEP.
The results demonstrated combustion characteristic and engine efficiency of all three
fuels were little affected by the injection timing at 2000rpm 4bar IMEP.
Ethanol and methanol’s combustion efficiency is slightly higher than gasoline. THC
released by alcohol fuel is lower than gasoline since alcohol fuel’s better evaporation
character and lower FID response to the partially oxided alcohol fuels. NOx produced
by ethanol and methanol is also lower due to their lower combustion temperature.
The particulate emission of alcohol fuels is much lower than gasoline. For ethanol,
the best efficiency point matches the lowest PN size 23-1000nm, and methanol's
best efficiency condition also has the lowest total PN and PN 10-1000nm.
160
At 3000rpm 16bar IMEP, gasoline engine operation was limited by the knocking
combustion with earlier injection timings but engine operations with both ethanol and
methanol were free from knocking combustion. But alcohol fuel’s LHVs are lower
than gasoline so their injection durations are longer. As a result of this, ethanol and
methanol have their lowest combustion efficiency with later injection timing 275CAD
BTDC and at rail pressure 50bar because of poorer atomisation at lower injection
pressure. But, ethanol and methanol’s peak efficiency area are still larger than
gasoline. Alcohol fuels produce less emissions than gasoline for the same reason as
the operation at 2000rpm 4.6bar IMEP.
7.2.3 Cold Start Spark Timing Sweep
For the cold start spark timing sweep, the result shows that both ethanol and
methanol could operate with more retarded spark timings than gasoline to enable by
faster catalyst-light-off. What’s more, the particulate emissions released by alcohol
fuels are much less and of smaller sizes.
7.3 Recommendations for Future Work
In order to facilitate future research works to be carried out on the single cylinder
engine, it would be desirable to update the intake air pressure control system to
enable more stable supply of compressed air to the engine. An automatic coolant&
oil temperature control system is preferred for high-load engine testing in place of the
manual control system in use to minimise the risk during high-speed & load testing.
After the completion of the current study, a number of areas has been identified
which would need to be investigated further, including
1) How to improve the cold start operation of ethanol and methanol spark ignition
engine (e.g. by injector heater& intake air heater)
161
2) Effects of alcohol on non-regulated emissions from alcohol, such as
aldehydes.
3) In this study, the maximum thermal efficiencies of ethanol and methanol
engine operations are limited by the engine’s mechanical design (Pmax
120bar, Rmax 6bar/CA). It would be useful to repeat the studies in an engine
with higher Pmax and Rmax limit to achieve higher engine thermal efficiency.
4) Higher compression ratio and higher dilution tolerance could be introduced to
increase the engine's thermal efficiency further.
162
Reference
[1] Smil V. The two prime movers of globalization: history and impact of diesel engines and gas
turbines. Journal of global history 2007;2(3):373-394.
[2] Cairns A, Stansfield P, Fraser N, Blaxill H, Gold M, Rogerson J, Goodfellow C. A Study of Gasoline-
Alcohol Blended Fuels in an Advanced Turbocharged DISI Engine. SAE International Journal of Fuels
and Lubricants 2009;2(1):41-57.
[3] Larsson T, Mahendar SK, Christiansen-Erlandsson A, Olofsson U. The effect of pure oxygenated
biofuels on efficiency and emissions in a gasoline optimised disi engine. Energies (Basel)
2021;14(13):3908.
[4] Martins FP, Lacava PT, De Andrade CR, Garzuzi S. Alternative Fuels: A Review about Anhydrous
and Hydrous Ethanol Properties. SAE technical paper series 2016;127082(October).
[5] Michael S, Timothy B, Ingemar D, Bengt J. CI Methanol and Ethanol combustion using ignition
improver. SAE Technical Papers 2019(December).
[6] Sarathy SM, Oßwald P, Hansen N, Kohse-Höinghaus K. Alcohol combustion chemistry. Progress in
energy and combustion science 2014;44:40-102.
[7] Dan Gearino. California Just Banned Gas-Powered Cars. Here’s Everything You Need to Know.
2022;2022(September 1,).
[8] Hu T, Wei Y, Liu S, Zhou L. Improvement of Spark-Ignition (SI) Engine Combustion and Emission
during Cold Start, Fueled with Methanol/Gasoline Blends. Energy Fuels 2007;21(1):171-175.
[9] Miller P, Solomon M. A brief history of technology-forcing motor vehicle regulations. EM

(Pittsburgh, Pa.) 2009:8.
[10] Delphi Technologies P. Delphi-worldwide-emissions-standards-passenger-cars-light-duty-2017-

2018. 2017.
[11] DieselNet. EU Emission Standards for Passenger Cars. 2017.
[12] New York Auto Show Receives Global Recognition from International Organization of Motor
Vehicle Manufacturers. PR newswire 2008.
[13] SMMT. New Car CO2 Report 2017 - The 16th Edition. 2017.
[14] Z. Yang, A. Bandivadekar. 2017 Global update: Light-duty vehicle greenhouse gas and fuel
economy standards. 2017.
[15] WardsAuto. 2016 WardsAuto Survey | DuPont Automotive | DuPont USA. 2016.
[16] Handbook of Air Pollution from Internal Combustion Engines: Pollutant Formation and Control.
Mechanical Engineering 1998;120(7):90.
163
[17] Harrington JA, Shishu RC. A single-cylinder engine study of the effects of fuel type, fuel
stoichiometry, and hydrogen-to-carbon ratio on CO, NO, and HC exhaust emissions. SAE technical
paper series 1973.
[18] Shimotani K, Oikawa K, Horada O, Kagawa Y. Characteristics of gasoline in-cylinder direct

injection engine. JSAE Rev 1996;17(3):267-272.
[19] Bowman CT. Kinetics of pollutant formation and destruction in combustion. Progress in energy
and combustion science 1975;1(1):33-45.
[20] Johnson GL, Caretto LS, Starkman ES. The kinetics of CO oxidation in reciprocating engines.
1970.
[21] Zhao F, Lai M, Harrington DL. The spray characteristics of automotive port fuel injection-a
critical reviews. SAE technical paper series 1995.
[22] Daniel WA. Flame quenching at the walls of an internal combustion engine. Symposium,
International, on Combustion 1957;6(1):886-894.
[23] Wentworth JT. The piston crevice volume effect on exhaust hydrocarbon emission. Combustion
Sci Technol 1971;4(1):97-100.
[24] Stone R. Introduction to internal combustion engines. 2012.
[25] Lavole GA, Heywood JB, Keck JC. Experimental and theoretical study of nitric oxide formation in
internal combustion engines. Combustion Sci Technol 1970;1(4):313-326.
[26] Zeldovich YA, Frank-Kamenetskii D, Sadovnikov P. Oxidation of nitrogen in combustion. 1947.
[27] Komiyama K, Heywood JB. Predicting NOx emissions and effects of exhaust gas recirculation in
spark-ignition engines. SAE technical paper series 1973.
[28] Kummer JT. Catalysts for automobile emission control. Progress in energy and combustion
science 1980;6(2):177-199.
[29] Heywood JB. Internal combustion engine fundamentals. 1988.
[30] Taylor CF. Internal Combustion Engine in Theory and Practice: Combustion, Fuels, Materials,
Design. 1985;2.
[31] Zhao H. Advanced direct injection combustion engine technologies and development: Volume 1,
Gasoline and gas engines. 2010.
[32] Reif K, Dietsche K, Robert Bosch GmbH, STAR Deutschland GmbH. Automotive handbook. 2014.
[33] Weberbauer F, Rauscher M, Kulzer A, Knopf M, Bargende M. Generally applicate split of losses
for new combustion concepts. MTZ worldwide 2005;66(2):17-19.
[34] International Comparison of Light-Duty Vehicle Fuel Economy 2005-2015. 2017;2017.
164
[35] Van Basshuysen R, Spicher U. Gasoline engine with direct injection: processes, systems,
development, potential : with 399 figures. 2009.
[36] Inoue T, Matsushita S, Nakanishi K, Okano H. Toyota lean combustion system - The third
generation system. SAE technical paper series 1993.
[37] Kühn M, Abthoff J, Kemmler R, Kaiser T. Influence of the inlet port and combustion chamber
configuration on the lean-burn behaviour of a spark-ignition gasoline engine. SAE technical paper
series 1996.
[38] Borges LH, Hollnagel C, Muraro W. Development of a mercedes-benz natural gas engine M 366
LAG, with a lean burn combustion system. SAE technical paper series 1996.
[39] Zhao F, Lai M-, Harrington DL. Automotive spark-ignited direct-injection gasoline engines.
Progress in energy and combustion science 1999;25(5):437-562.
[40] Zhao H. Advanced direct injection combustion engine technologies and development: Gasoline
and gas engines. 2009:9.
[41] Krebs R, Pott E, Kreuzer TP, Göbel U, Glück K. Exhaust gas aftertreatment of volkswagen FSI fuel
stratified injection engines. SAE technical paper series 2002.
[42] Rohr F, Peter SD, Lox E, Kögel M, Müller W, Sassi A, Rigaudeau C, Juste L, Belot G, Gélin P,
Primet M. The impact of sulfur poisoning on NOx-storage catalysts in gasoline applications. SAE
technical paper series 2005.
[43] Merdes N, Enderle C, Vent G, Weller R. The New Turbocharged Four-Cylinder Gasoline Engine by
Mercedes-Benz. MTZ worldwide 2011;72(12):16-23.
[44] Zhao H. HCCI and CAI engines for the automotive industry. 2007.
[45] Chen T, Xie H, Li L, Zhang L, Wang X, Zhao H. Methods to achieve HCCI/CAI combustion at idle
operation in a 4VVAS gasoline engine. Appl Energy 2014;116:41-51.
[46] Yamasaki Y, Umahashi S, Uesugi Y, Ma Q, Kaneko S, Hikita T, Mizuno S, Tsumura Y, Hashiguchi T.

Development of Dynamic Models for an HCCI Engine with Exhaust Gas Rebreathing System. 2015.
[47] Konigstein A, Grebe UD, Wu K, Larsson P. Differentiated analysis of downsizing concepts. MTZ.
Motortechnische Zeitschrift 2008;69(6).
[48] M. Kratzsch, M. Günther. Knocking in Gasoline Engines. DCM Druck Center Meckenheim GmbH
2013.
[49] Low Speed Pre-ignition (LSPI), Oxidation and Wear. Lubrizol360 2017.
[50] Sroka ZJ. Work Cycle of Internal Combustion Engine Due to Rightsizing. 2021:Ch. 1.
[51] W. Hannibal. Variable Valve Timing Systems on Modern Spark Ignition Engines. 2009.
[52] Potul S, Nachnolkar R, Bhave S. Analysis Of Change In Intake Manifold Length And Development
Of Variable Intake System.
165
[53] Moriya Y, Watanabe A, Uda H, Kawamura H, Yoshioka M, Adachi M. A newly developed
intelligent variable valve timing system-continuously controlled cam phasing as applied to a new 3
liter inline 6 engine. SAE technical paper series 1996.
[54] Leone TG, Christenson EJ, Stein RA. Comparison of variable camshaft timing strategies at part
load. SAE technical paper series 1996.
[55] Tuttle JH. Controlling Engine Load by Means of Late Intake-Valve Closing. SAE transactions
1980;89:2429-2441.
[56] Hentschel W, Block B, Hovestadt T, Meyer H, Ohmstede G, Richter V, Stiebels B, Winkler A.

Optical diagnostics and CFD-simulations to support the combustion process development of the
Volkswagen FSI® direct-injection gasoline engine. SAE technical paper series 2001.
[57] Szengel R, Middendorf H, Voeltz S, Laumann A, Tilchner L, Theobald J, Etzrodt T, Krebs R. The TSI
engine with 90 kW: Extension of the economical petrol engine series by Volkswagen. MTZ worldwide
2007;68(7-8):8-11.
[58] Roberts M. Benefits and Challenges of Variable Compression Ratio (VCR). 2003.
[59] Hoppe F, Thewes M, Baumgarten H, Dohmen J. Water injection for gasoline engines: Potentials,
challenges, and solutions. International journal of engine research 2016;17(1):86-96.
[60] hakariya M, Toda T, Sakai M. The New Toyota Inline 4-Cylinder 2.5L Gasoline Engine. 2017.
[61] J. Atkinson. Gas Engine. 1886(USA Patent 336505).
[62] J. Atkinson. Gas Engine. 1887(USA Patent 367496).
[63] Ting DS, Checkel MD. The effects of turbulence of spark-ignited, ultra lean, premixed methane-
air flame growth in a combustion chamber. SAE technical paper series 1995.
[64] Kato T, Saeki K, Nishide H, Yamada T. Development of CNG fueled engine with lean burn for
small size commercial van. JSAE Rev 2001;22(3):365-368.
[65] Surisetty VR, Dalai AK, Kozinski J. Alcohols as alternative fuels: An overview. Applied Catalysis A:
General 2011;404(1-2):1-11.
[66] Siwale L, Kristóf L, Bereczky A, Mbarawa M, Kolesnikov A. Performance, combustion and

emission characteristics of n-butanol additive in methanol–gasoline blend fired in a naturally-
aspirated spark ignition engine. Fuel Process Technol 2014;118:318-326.
[67] Ramadhas AS. Alternative fuels for transportation. 2011.
[68] Reed TB, Lerner RM. Methanol: A Versatile Fuel for Immediate Use. Science
1973;182(4119):1299-1304.
[69] He B, Jian-Xin Wang, Hao J, Yan X, Xiao J. A study on emission characteristics of an EFI engine
with ethanol blended gasoline fuels. Atmos Environ 2003;37(7):949-957.
166
[70] Hu T, Wei Y, Liu S, Zhou L. Improvement of Spark-Ignition (SI) Engine Combustion and Emission
during Cold Start, Fueled with Methanol/Gasoline Blends. Energy Fuels 2007;21(1):171-175.
[71] Kim S, Dale BE. Environmental aspects of ethanol derived from no-tilled corn grain:
nonrenewable energy consumption and greenhouse gas emissions. Biomass Bioenergy
2005;28(5):475-489.
[72] Kim S, Dale BE. Environmental aspects of ethanol derived from no-tilled corn grain:
nonrenewable energy consumption and greenhouse gas emissions. Biomass Bioenergy
2005;28(5):475-489.
[73] Solomon BD, Barnes JR, Halvorsen KE. Grain and cellulosic ethanol: History, economics, and
energy policy. Biomass Bioenergy 2007;31(6):416-425.
[74] Awad OI, Mamat R, Ali OM, Sidik NAC, Yusaf T, Kadirgama K, Kettner M. Alcohol and ether as
alternative fuels in spark ignition engine: A review. Renewable & sustainable energy reviews
2018;82:2586-2605.
[75] Thomas V, Kwong A. Ethanol as a lead replacement: phasing out leaded gasoline in Africa.
Energy Policy 2001;29(13):1133-1143.
[76] Jeuland N, Montagne X, Gautrot X. Potentiality of Ethanol As a Fuel for Dedicated Engine. Oil &
gas science and technology 2004;59(6):559-570.
[77] Yüksel F, Yüksel B. The use of ethanol–gasoline blend as a fuel in an SI engine. Renewable
Energy 2004;29(7):1181-1191.
[78] Mills JE. THE INTERNAL HEAT OF VAPORIZATION. J Am Chem Soc 1909;31(10):1099-1130.
[79] Koç M, Sekmen Y, Topgül T, Yücesu HS. The effects of ethanol–unleaded gasoline blends on
engine performance and exhaust emissions in a spark-ignition engine. Renewable Energy
2009;34(10):2101-2106.
[80] Milpied J, Jeuland N, Plassat G, Guichaous S, Dioc N, Marchal A, Schmelzle P. Impact of Fuel
Properties on the Performances and Knock Behaviour of a Downsized Turbocharged DI SI Engine –
Focus on Octane Numbers and Latent Heat of Vaporization. SAE International Journal of Fuels and
Lubricants 2009;2(1):118-126.
[81] Munack A. Book Reviews: Biodiesel - The Comprehensive Handbook. Edited by Martin
Mittelbach, Claudia Remschmidt. CLEAN - Soil, Air, Water 2007;35(1):14.
[82] Daniel R, Tian G, Xu H, Wyszynski ML, Wu X, Huang Z. Effect of spark timing and load on a DISI
engine fuelled with 2,5-dimethylfuran. Fuel 2011;90(2):449-458.
[83] Demirbas A. Combustion characteristics of different biomass fuels. Progress in Energy and
Combustion Science 2004;30(2):219-230.
[84] Alternative Fuels Data Center. Biobutanol. ;2022(October 17,).
[85] Rask KN. Clean air and renewable fuels: the market for fuel ethanol in the US from 1984 to
1993. Energy Econ 1998;20(3):325-345.
167
[86] El-Emam SH, Desoky AA. A study on the combustion of alternative fuels in spark-ignition
engines. Int J Hydrogen Energy 1985;10(7):497-504.
[87] Pourkhesalian AM, Shamekhi AH, Salimi F. Alternative fuel and gasoline in an SI engine: A
comparative study of performance and emissions characteristics. Fuel 2010;89(5):1056-1063.
[88] Nakata K, Utsumi S, Ota A, Kawatake K, Kawai T, Tsunooka T. The effect of ethanol fuel on a
spark ignition engine. SAE technical paper series 2006.
[89] Mahendar SK, Venkataraman V, Erlandsson AC. The Impact of Miller Valve Timing on
Combustion and Charging Performance of an Ethanol- and Methanol-Fueled Heavy-Duty Spark
Ignition Engine. SAE International journal of engines 2021;14(5):733-748.
[90] Çelik MB, Özdalyan B, Alkan F. The use of pure methanol as fuel at high compression ratio in a
single cylinder gasoline engine. Fuel (Guildford) 2011;90(4):1591-1598.
[91] Vancoillie J, Demuynck J, Sileghem L, Van De Ginste M, Verhelst S, Brabant L, Van Hoorebeke L.
The potential of methanol as a fuel for flex-fuel and dedicated spark-ignition engines. Appl Energy
2013;102:140-149.
[92] Pannone GM, Johnson RT. Methanol as a fuel for a lean turbocharged spark ignition engine. SAE
transactions 1989:243-253.
[93] VANCOILLIE J, DEMUYNCK J, SILEGHEM L, VAN DE GINSTE M, VERHELST S, BRABANT L, VAN

HOOREBEKE L. The potential of methanol as a fuel for flex-fuel and dedicated spark-ignition engines :
Advances in sustainable biofuel production and use. Applied energy 2013;102:140-149.
[94] Dhaliwal B, Yi N, Checkel D. Emissions Effects of Alternative Fuels in Light-Duty and Heavy-Duty
Vehicles. 2000.
[95] Kabasin D, Hoyer K, Kazour J, Lamers R, Hurter T. Heated Injectors for Ethanol Cold Starts. SAE
International Journal of Fuels and Lubricants 2009;2(1):172-179.
[96] Golzari R, Li Y, Zhao H. Impact of Port Fuel Injection and In-Cylinder Fuel Injection Strategies on
Gasoline Engine Emissions and Fuel Economy. SAE technical paper series 2016;2016-.
[97] Hancock D, Fraser N, Jeremy M, Sykes R, Blaxill H. A new 3 cylinder 1.2l advanced downsizing
technology demonstrator engine. SAE technical paper series 2008.
[98] Bassett M, Hall J, Hibberd B, Borman S, Reader S, Gray K, Richards B. Heavily Downsized
Gasoline Demonstrator. SAE International Journal of Engines 2016;9(2):729-738.
[99] Hall J, Bassett M, Hibberd B, Streng S. Heavily Downsized Demonstrator Engine Optimised for
CNG Operation. SAE International Journal of Engines 2016;9(4):2250-2261.
[100] Bassett M, Hall J, Cains T, Underwood M, Wall R, Richards BG. Dynamic Downsizing Gasoline
Demonstrator. SAE International Journal of Engines 2017;10(3):884-891.
[101] Bassett M, Vogler C, Hall J, Taylor J, Cooper A, Reader S, Gray K, Wall R. Analysis of the
Hardware Requirements for a Heavily Downsized Gasoline Engine Capable of Whole Map Lambda 1
Operation. SAE technical paper series 2018;2018-.
168
[102] Lumsden G, OudeNijeweme D, Fraser N, Blaxill H. Development of a Turbocharged Direct
Injection Downsizing Demonstrator Engine. SAE International Journal of Engines 2009;2(1):1420-
1432.
[103] Zhang Y. Experimental investigation of CAI combustion in a two-stroke poppet valve DI engine.
2015.
[104] Zhao H, Ladommatos N. Engine combustion instrumentation and diagnostics. 2001.
[105] Coates BP. Investigation of engine design parameters on the efficiency and performance of the
high specific power downsized SI engine. 2012.
[106] Zhao H, Ladommatos N. Engine combustion instrumentation and diagnostics. 2001.
[107] Randolph AL. Methods of processing cylinder-pressure transducer signals to maximize data
accuracy. SAE technical paper series 1990.
[108] De Melo, Tadeu Cavalcante Cordeiro, De Brito, Marcos Fernando Mendes, MacHado GB, Paiva
CEF. Procedure for uncertainty of measurement determination of spark ignition engine emission
tests. SAE technical paper series 2012.
[109] Raza M, Chen L, Leach F, Ding S. A Review of particulate number (PN) emissions from gasoline
direct injection (gdi) engines and their control techniques. Energies (Basel) 2018;11(6):1417.
[110] Yanju W, Shenghua L, Hongsong L, Rui Y, Jie L, Ying W. Effects of Methanol/Gasoline Blends on
a Spark Ignition Engine Performance and Emissions. Energy Fuels 2008;22(2):1254-1259.
[111] Qin J, Li X, Pei Y. Effects of Combustion Parameters and Lubricating Oil on Particulate Matter
Emissions from a Turbo-Charged GDI Engine Fueled with Methanol/Gasoline Blends. 2014.
169
170

Addsis Alem Thesis.

Uploaded by

Copyright:

Available Formats

Addsis Alem Thesis.

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Addsis Alem Thesis.

Uploaded by

Copyright:

Available Formats

Abstract

In order to reduce emissions of carbon dioxide (CO2) to minimize global warming

Finally, my most sincere gratitude goes to my family - my mother, my father and my

ABDC After Bottom Dead Centre

As a result of the popularity of internal combustion engines in the automotive sector,

In terms of decarbonizing fleet CO2 for internal combustion engines, renewable

On 25 August 2022, California announced that the sale of new gasoline-powered

1.3 Project Objectives

The specific objectives of the study include quantification of:

1.4 Outline of Thesis

Figure 2-1 Toxic Emissions Standards for Passenger Cars [10]

2.2 Main Component of Exhaust Gas

2.2.1 Carbon Dioxide (CO2)

• Improved combustion and engine control methods

• Improved transmission efficiency and powertrain integration

• Intelligent thermal management

• Brake/ kinetic energy recovery

• Exhaust/heat energy recovery

• Improved aerodynamics and rolling resistance

• Light-weighting design and materials

2.2.2 Carbon Monoxide (CO)

The kinetic chemical formation process of CO is shown below, in which R is a

2.2.3 Unburned Hydrocarbon (THC)

2.2.5 Particle Matter (PM)

PM comprises carbon particles (soot) and condensates of liquid (sulphate, unburned

2.3 Internal Combustion Engine Efficiency

As a means of reducing the fuel consumption and CO2 emissions of automobiles, it

Objectives Limitation of gasoline engine operation

1) Abnormal combustion: knocking, low speed

Increase compression ratio 2) The design and reduction of friction result of

1) The operation of a spark ignition (SI) engine is

a) The air-fuel ratio (AFR) has a big effect

b) In order to decrease the CO, HC and

2) Because the combustion type is stoichiometric

At part load conditions, the throttle is used to reduce

1) Combustion time losses are caused by the

2) Knocking at high load condition pushes the

i. Throttling at a lower load leads to the intake pressure decrease

i. For the gasoline engine, pumping loss decrease with the

ii. However, knocking combustion happens when the engine run at

• The low-speed pre-ignition (LSPI) may occur at a low-

i. Knocking tendency reduces at high-speed high load.

ii. Enrich AFR is adopted at high load high-speed condition to

The mechanical losses increase with the increase of engine speed

Recently, a series of new technologies have been researched and adopted on

2.4.1 Gasoline Direct Injection

2.4.3 CAI and HCCI

Controlled Auto-ignition (CAI) which is also called Homogeneous Charge

2.4.4 Gasoline DI Engines and Boosting (Downsizing)

It is essential necessary to decrease the compression ratio of boosted downsizing

2.4.5 Variable Valve Actuation and Air Intake

Variable intake manifolds can be realised by step adjusting or continuous adjusting

2.4.6 Variable Compression Ratio

2.4.7 Water Injection

2.4.9 Miller Cycle

• For a supercharged SI engine, thermal load and knocking/ pre-ignition

2.5 Spark Ignition Engines with Alternative Fuels

2.5.1 Natural Gas Engine

Natural gas, whose main composition is methane, is an alternative fuel to gasoline

2.5.2 Alcohol as Alternative Fuel in Spark Ignition Engine