Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Performance and emission characteristics of a gas-diesel engine Tao, Yinchu 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_tao_yinchu.pdf [ 8.38MB ]
JSON: 831-1.0080891.json
JSON-LD: 831-1.0080891-ld.json
RDF/XML (Pretty): 831-1.0080891-rdf.xml
RDF/JSON: 831-1.0080891-rdf.json
Turtle: 831-1.0080891-turtle.txt
N-Triples: 831-1.0080891-rdf-ntriples.txt
Original Record: 831-1.0080891-source.json
Full Text

Full Text

PERFORMANCE AND EMISSION CHARACTERISTICS OFA GAS-DIESEL ENGINEbyYINCHU TAOB. Eng., Jiangsu Institute of Technology (P. R. China), 1983A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF APPLIED SCIENCEinTHE FACULTY OF GRADUATE STUDIESMechanical Engineering DepartmentWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1993© Yinchu Tao, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of ke *-ekezit;cciL EnsincritiThe University of British ColumbiaVancouver, CanadaDate  AllaSt .) S / 113 DE-6 (2/88)iiABSTRACTThe performance and emission characteristics of high pressureinjection of natural gas with liquid pilot-diesel fuel (ie. gas-diesel operation) was investigated in a single-cylinder, two-strokecompression-ignition engine with a poppet-valve gas-dieselinjector.The investigated injector geometries and engine operatingparameters included: fuel injection angle, fuel jet interruptionratio, engine speed, load, beginning of injection timing, naturalgas injection pressure, pilot-diesel to total-fuel energy ratio andpilot-diesel cetane number. These parameters were found to havevery strong effects on thermal efficiency and exhaust emissions(ie. NOR, THC, CH4, CO, CO2 and BOSCH smoke index). The thermalefficiency and exhaust emissions were determined as a function ofload (ie. BMEP).The thermal efficiency of the optimum gas-diesel operation wasshown to exceed that of the conventional diesel operation at fullload, but was lower at low load.With this gas-diesel injector configuration, it was found thatthe pilot-diesel fuel was not mixed well enough to burn completely.A new gas-diesel injector designed to overcome this drawback is inprocess.A three-zone combustion and exhaust emission analysis modelwas established to deduce ignition delay from cylinder pressuredata with crank angle. At low loads, it was found that the ignitioniiidelay of natural gas was excessive with very late burning andconsequently low thermal efficiency. At high loads, the ignitiondelay of natural gas was considerably longer than that of dieselfuelling, but not so long as to affect the thermal efficiency.The three-zone model was also used to deduce maximum burned-gas temperature. It was found that equilibrium NO concentrationcalculated from this temperature and cylinder pressure can be usedto correlate the exhaust NO measured from engine exhaust pipe.ivTABLE OF CONTENTSABSTRACT^ iiTABLE OF CONTENTS^ ivLIST OF SYMBOLS ixLIST OF TABLES^ xviLIST OF FIGURES xviiACKNOWLEDGEMENTS^ xxiv1 INTRODUCTION^ 11.1^Introduction 11.2^The History of Federally Enacted Laws^ 11.3^The Diesel Engine Exhaust Emission Problems^31.4^Natural Gas as an Alternative Fuel^ 51.5^Objectives of This Research^ 61.6^Method^ 72 LITERATURE REVIEW^ 82.1^Introduction 82.2^Combustion Characteristics of Natural Gas^82.3^Gas-Diesel Engine Operation^ 112.4^Pollutant Formation in Diesel Engines^ 162.5^Exhaust Emission Studies in Compression-IgnitionEngines^ 213 EXPERIMENTAL APPARATUS^ 253.1^Introduction^ 253.2^Test Engine and Test Control System^ 253.2.1 Test Engine^ 263.2.2 Dynamometer, DCM and DDM^ 293.2.3 Engine Fuelling System 303.2.4 Engine Cooling System^ 353.2.5 Instrumentation^ 353.2.6 Data Acquisition System^ 383.3^The Exhaust Emission Analysis System A, (EEAS-A)^413.3.1 Arrangement of the EEAS-A^ 413.3.2 Principle of the Analyzers in the EEAS-A^433.3.3 Drawback of the EEAS-A^ 473.4^The Exhaust Emission Analysis System B, (EEAS-B)^473.4.1 Arrangement of the EEAS-B^ 483.4.2 Principle of the Analyzers in the EEAS-B^503.4.3 Relationship between the EEAS-A andthe EEAS-B Systems^ 533.5^Smoke Determination (BOSCH Smoke Meter)^544 EXPERIMENTAL PROCEDURES^ 564.1^Introduction^ 564.2^The Engine Performance and Emission Test Procedure^574.3^The engine Performance Calculation Procedure^624.4^The Engine Exhaust Emission Calculation Procedure^644.4.1 Conversion between Dry-Basis and Wet-BasisviEmissions^ 654.4.2 Determination of Brake Specific Emissions^675 EXPERIMENTAL RESULTS^ 715.1^Introduction 715.2^Diesel Baseline Test Results^ 715.3^Effect of Injector Geometrical Parameters^765.4^Effect of Engine Operating Parameters 895.5^Optimum Gas-Diesel Operation Condition^1106 NUMERICAL SIMULATION CALCULATION OF COMBUSTION--- ONE-ZONE MODEL^ 1186.1^Introduction 1186.2^Formulation of the Exhaust Emission Analysis (EEA)Model^ 1196.3^Mixture Compositions of the Unburned Gas^1236.3.1 Terminology^ 1246.3.2 Residual Gas and the Unburned GasComposition^ 1266.4^Determination of Initial Condition^ 1286.5^Thermodynamic Properties of the Unburned Gas^1296.6^Computation Results and Discussion^ 1306.6.1 Effect of Different Fuels on Burned-GasComposition^ 1306.6.2 Effect of the Residual Gas on Burned-GasComposition^ 135vii6.7^Summary^ 1367 NUMERICAL SIMULATION CALCULATION OF COMBUSTION--- THREE-ZONE MODEL^ 1387.1^Introduction 1387.2^Formulation of the Three-Zone Combustion Model^1397.3^Mass of Air Trapped in the cylinder andResidual Mass Fraction^ 1517.4^Unburned-Gas Composition 1557.5^Unburned-Fuel Mass Ratio in Exhaust^ 1567.6^Averaged Equilibrium NO Concentration 1577.7^Computation Results^ 1617.8^Summary^ 1698 CONCLUSIONS AND RECOMMENDATIONS^ 1718.1^Conclusions^ 1718.2^Recommendations 173REFERENCES^ 174APPENDICES^ 179Appendix A - NO Formation Rate^ 179Appendix B - Exhaust Emission Analysis System B OperatingProcedure^ 183Appendix C - Power Correction Factor Calculation Method^191Appendix D - Determination of Specific Humidity^192Appendix E - A More Exact Formula for Dry-Wet BasisviiiConversion Factor 194Appendix F - Repeatability of the Test Results 198Appendix G - Photograph of the Visualization Results ofNatural Gas Injection 204Appendix H - Program #1 208Appendix I - Program #2 210Appendix J - Procedures for Calculating ThermodynamicProperties 213Appendix K - Procedure for Computation of EquilibriumComposition 216Appendix L - Evaluation of the Residual Temperature 218Appendix M - Work Sheet of the Program XPRESSD 221Appendix N - Program XPRESSD 227LIST OF SYMBOLSA^areabs,^brake specific emissionsdiameter of the cylinder boreBMEP brake mean effective pressureconstantC2^constantc3^constantC4^constantCFK^correction factorCH1.8^dieselCH2^dieselCH,^methane (or natural gas)CHy^combined fuelCN62^cetane number 62 diesel fuelCO^carbon monoxideCO2^carbon dioxideC,^constant-pressure specific heatC,^constant-volume specific heatDF1^diesel fuel grade 1DF2^diesel fuel grade 2DP^degree of purity of the chargetotal internal energypower correction factormolal fractionixconversion factor of dry-to-wet basis emissionsFG^geometric functionFres^residual molal (or mass)fractionF,^emissivity functionF/A^fuel-air ratioF/AcNG CNG fuel-air ratioF/A,, diesel fuel-air ratiospecific enthalpyspecific humidityenthalpy of the systemHC^hydrocarbon1120^waterspecific heat ratiothermal conductivityKdwi^correction factorKcjw2^correction factorLHV^lower heating valuemass per cycleMa^mass of dry airMatrap^mass of air trapped in the cylinderMtrap^mass of the cylinder contents at inlet port closure111C^mass of water vapour in airth^mass flow ratethi^mass flow rate of emission component "i"112filei total fuel mass flow ratenumber of molesMW,^molecular weight of a component "i"polytropic constantn,^number of crank revolutions for each power stroke percylindern,^total moles of the combustion productsspeednumber of molesN,^nitrogenNMHC non-methane hydrocarbonNO^nitric oxideNO2^nitrogen dioxideNO^nitrogen oxides02^oxygen03^ozoneOH^hydroxyl grouppressurePa^partial pressure of dry airPb^brake powerbarometric pressurePM^particulate matterPr^Prandal numberP,^saturated pressure of water vapourPw^partial pressure of water vapour in airxixiiheat transfer per unit massheat transfer to the systemresidual-air molal ratiorm^pilot-diesel fuel to CNG fuel mass ratiogas constantR,^gas constant of dry airRe^Reynolds numberRw^gas constant of water vapourRFASTOIC equivalent stoichiometric fuel-air ratiotimeTo^standard temperaturetemperatureTb^engine torqueTd^dry bulb temperatureTHC^total hydrocarbonspecific energyinternal energy of the systemUFRAT unburned fuel mass ratio in exhaustspecific volumeV^volumeVd^cylinder displacement volumeVP is^piston velocitywork done on the pistonWc^engine shaft work per cyclex,^burned-gas mass fractionfuel mass-burned fractionX, molal fraction of a component "i" in exhaust emissionsatomic hydrogen to carbon ratioyCNG atomic hydrogen to carbon ratio of CNGYDSL^atomic hydrogen to carbon ratio of dieselY, signal-output voltage of the span gas "i"Z,^concentration of the span gas "i"Greek Letters: specific heat ratiorib^scavenged-blower efficiency11th^thermal efficiencyk^relative air-fuel ratioA^delivery ratiodynamic viscositydensityStefan-Boltzmann constant(1)^relative humiditySubscript: abox air boxair^trapped fresh airAVG^averagebg^burned gasbulk bulkcharge cylinder content at intake port closureCH,^dieselxivCH4^methaneCHy^combined fuelCNG^compressed natural gasCO^carbon monoxideCO,^carbon dioxideCONV convectiondry^dry-basis dataDSL^dieselexh^exhaust gasEQUIL equilibriumfbmax fuel burned at maximum temperaturefuel^fuelHC^hydrocarbonipc^intake port closureN2^nitrogenNMHC non-methane hydrocarbonNO^nitric oxideNO^nitrogen oxides0,^oxygenres^trapped residual gasRAD^radiationsurf^surfacetot^total mass of the cylinder contentTHC^total hydrocarbonub^unburned gasuf^unburned fuelWALL wallwet^wet-basis dataAbbreviations: ABDC after bottom dead centreATDC after top dead centreBDC^bottom dead centreBOI^beginning of injectionBTDC before top dead centreCA^crank angleCHEMI chemiluminescenceCI^compression ignitionCNG^compression natural gasEGR^exhaust gas recyclingEPA^environmental protection agencyFID^flame ionization detectorIPC^intake port closureNDIR non-dispersive infraredppm^parts per millionPW^pulse width of injectionRPM^revolution per minuteTDC^top dead centreXVxviLIST OF TABLESTable 1.1^EPA CAAA Emission Standards for UrbanBus Engines^ 3Table 3.1^General Specifications of 1-71 DieselEngine^ 27Table 3.2^Properties of Diesel Fuels^ 31Table 3.3^Composition of the B.C. Natural Gas^32Table 3.4^Properties of the B.C. Natural Gas 33Table 3.5^List of Steady State Data and HighSpeed Data^ 39Table 4.1^List of Variable Parameters and TestingRanges^ 59Table 5.1^List of Unburned CNG and Pilot-Diesel Ratio 112Table A.1^Typical Values of 121, R1/R2 and R1/(R2+R3)^181Table B.1^Typical Operating Ranges of EmissionAnalyzers^ 185Table B.2^Typical Zero and Span Gases^ 186xviiLIST OF FIGURESFigure 1.1^1994 and 1998 EPA Standard, NO and SootParticulates Emission Level of presentEngines^ 4Figure 2.1^Methods of Using Natural Gas in Gas-DieselEngine^ 12Figure 3.1^Schematic of Experimental Apparatus andInstrumentation^ 26Figure 3.2^Combustion Chamber Geometry and InjectorMounting of the 1-71 Diesel Engine^27Figure 3.3^Arrangement of Test Engine Cell 28Figure 3.4^Engine Control Console^ 29Figure 3.5^Schematic of Engine Fuelling System^30Figure 3.6^Schematic of Gas-Diesel Electronic UnitInjector^ 34Figure 3.7^Schematic of Data Flow in Data AcquisitionSystem^ 38Figure 3.8^Schematic of Exhaust Emission AnalysisSystem A^ 42Figure 3.9^Schematic Flow Diagram of NO/NO, Analyzer^43Figure 3.10^Schematic Flow Diagram of HC Analyzer^45Figure 3.11^Schematic of Non-Dispersive InfraredDetection System with Double-Beam Method^46Figure 3.12^Schematic Flow Diagram of the First Cabinetxviiiof Exhaust Emission Analysis System B^48Figure 3.13^Schematic Flow Diagram of the second Cabinetof Exhaust Emission Analysis System B^49Figure 3.14^Schematic of Non-Dispersive InfraredDetection System with Single-Beam Method^51Figure 3.15^Schematic of Oxygen Measurement Principle^52Figure 3.16^Schematic of BOSCH "Spot" Smokemeter^54Figure 4.1^Schematic of Injector with Castellated-EndSleeve^ 57Figure 4.2^Determination of the Best BOI PerformanceCurve^ 61Figure 5.1^Effect of Fuel Composition on Performance^72Figure 5.2^Effect of Fuel Composition on NitrogenOxides^ 73Figure 5.3^Effect of Fuel Composition on TotalHydrocarbon^ 74Figure 5.4^Effect of Fuel Composition on Non-MethaneHydrocarbon^ 75Figure 5.5^Effect of Fuel Composition on CarbonMonoxide^ 76Figure 5.6^Effect of Fuel Jet Interruption Ratioon Performance^ 77Figure 5.7^Effect of Fuel Jet Interruption Ratioon Total Hydrocarbon^ 79xixFigure 5.8^Effect of Fuel Jet Interruption Ratioon Nitrogen Oxides^ 80Figure 5.9^Effect of Fuel Jet Interruption Ratioon Carbon Monoxide^ 81Figure 5.10^Effect of Fuel Injection Angle onPerformance with 50% Shrouding^82Figure 5.11^Effect of Fuel Injection Angle onPerformance with 30% Shrouding^83Figure 5.12^Effect of Fuel Injection Angle onCarbon Monoxide^ 84Figure 5.13^Effect of Fuel Injection Angle onNitrogen Oxides^ 86Figure 5.14^Effect of Fuel Injection Angle onMethane^ 87Figure 5.15^Effect of Fuel Injection Angle onNon-Methane Hydrocarbon^ 88Figure 5.16^Effect of Pilot-Diesel Cetane Numberon Performance^ 90Figure 5.17^Effect of Pilot-Diesel Cetane Numberon Methane^ 91Figure 5.18^Effect of Pilot-Diesel Cetane Numberon Non-Methane Hydrocarbon^ 92Figure 5.19^Effect of Pilot-Diesel Cetane Numberon Nitrogen Oxides^ 93Figure 5.20^Effect of Diesel Ratio on ThermalEfficiency at Low Load (BMEP = 1 BAR)^94X XFigure 5.21^Effect of Diesel Ratio on TotalHydrocarbon at Low Load (BMEP = 1 BAR)^95Figure 5.22^Relationship between Mass Flow Ratio ofFuels and Load^ 96Figure 5.23^Effect of Diesel Ratio on Maximum ThermalEfficiency^ 97Figure 5.24^Effect of Diesel Ratio on Maximum LoadCapability^ 98Figure 5.25^Effect of Diesel Ratio on NitrogenOxides at High Load (BMEP = 3.8 BAR)^99Figure 5.26^Effect of CNG Injection Pressure onPerformance with 100 Injection Angle^100Figure 5.27^Effect of CNG Injection Pressure onPerformance with 20° Injection Angle^101Figure 5.28^Effect of CNG Injection Pressure onMethane^ 102Figure 5.29^Effect of CNG Injection Pressure onNon-Methane Hydrocarbon^ 103Figure 5.30^Effect of CNG Injection Pressure onNitrogen Oxides^ 104Figure 5.31^Effect of CNG Injection Pressure on BOI^105Figure 5.32^Effect of Engine Speed on Performance^106Figure 5.33^Effect of Engine Speed on Nitrogen Oxides^107Figure 5.34^Effect of Engine Speed on Methane^108Figure 5.35^Effect of Engine Speed on Non-MethaneHydrocarbon^ 109xxiFigure 5.36^Optimum Performance of Gas-DieselOperation^ 110Figure 5.37^Nitrogen Oxides of Optimum Gas-DieselOperation^ 112Figure 5.38^Methane of Optimum Gas-Diesel Operation^113Figure 5.39^Non-Methane Hydrocarbon of OptimumGas-Diesel Operation^ 114Figure 5.40^Carbon Monoxide of Optimum Gas-DieselOperation^ 115Figure 5.41^Carbon Dioxide of Optimum Gas-DieselOperation^ 116Figure 5.42^Bosch Smoke Index of Optimum Gas-DieselOperation^ 117Figure 6.1^The Modified Air-Standard Diesel Cycle^119Figure 6.2^Schematic of Unif low-ScavengedConfiguration^ 123Figure 6.3^Effect of Different Fuels on NOConcentration^ 131Figure 6.4^Effect of Different Fuels on AdiabaticFlame Temperature^ 132Figure 6.5^Effect of Different Fuels on EquilibriumCO Concentration in the Cylinder^133Figure 6.6^Effect of Different Fuels on EquilibriumCO2 Concentration in the Cylinder^134Figure 6.7^Effect of Residual Gas on NO Concentration^135xxiiFigure 7.1^Schematic of the Three-Zone Combustion Model 139Figure 7.2^Typical Scavenging Data Range of Two-StrokeDiesel.^ 154Figure 7.3^Cylinder Pressure Distribution (BMEP=4 bar) 161Figure 7.4^Temperature Distributions (BMEP=4 bar)^162Figure 7.5^Fuel Mass-Burned Fraction (BMEP=4 bar)^163Figure 7.6^Effect of BOI on Ignition Delay^164Figure 7.7^Effect of Fuel Jet Interruption Ratio onIgnition Delay^ 165Figure 7.8^Effect of EGR on Ignition Delay^166Figure 7.9^Correlation of Measured NO and Calculated NO 167Figure 7.10^Effect of EGR on Measured NO Concentration^168Figure 7.11^Effect of EGR on Calculated NO Concentration 169Figure F.1^Repeatability of Thermal Efficiency^198Figure F.2^Repeatability of Nitrogen Oxides Emissions^199Figure F.3^Repeatability of Total HydrocarbonEmissions^ 200Figure F.4^Repeatability of Unburned Methane Emission 201Figure F.5^Repeatability of Carbon Monoxide Emission^202Figure F.6^Repeatability of Carbon Dioxide Emission^203Figure G.1^Free Conical Sheet Jet with 100 InjectionAngle^ 204Figure G.2^Interrupted Conical Jet with 100 InjectionAngle^ 205Figure G.3^Free Conical Sheet Jet with 200 InjectionAngle^ 206Figure G.4^Interrupted Conical Jet with 20° InjectionAngle^ 207Figure L.1^Schematic of Residual Temperature EvaluationModel^ 218xx ivACKNOWLEDGEMENTSI express my sincere gratitude to Dr. P.G. Hill for hisinvaluable guidance and help, judicious advice and encouragementthroughout all phases of the project and the writing of thisthesis.A special thanks to Bruce Hodgins, research engineer andproject manager, for his advice and assistance in experimentalwork.A special thanks to Drs. R.W. & N.L Lewis, my Englishteachers, for their advice and help on English grammar.A Special thanks to Dehong Zhang for his helpful advice andsuggestions, to Brad Douville for his help on computer programming,to Patric Ouellette for providing reference photographs for thisthesis.A very special thanks to my wife and former colleague, QinZhou, who offered incessant moral support and encouragement,helpful advice and suggestions during my graduate work.Financial support for the first year of this work by the"Pao Yu-kong and Pao Zhao-long Scholarship for Chinese StudentsStudying Abroad" is gratefully acknowledged.11 . INTRODUCTION1.1 IntroductionOne of the key symbols that represents human civilization istransportation. Our present transportation systems are mainlypowered by internal combustion engines, which use petroleum-basedliquid fuels. From the so-called "oil-crisis" of the 1970's, whencrude oil supplies seemed to become uncertain and prices increasedsignificantly, people realized that for energy security it was timeto research and utilize alternative fuels. Since then manyalternative fuel research projects have emerged, as well as manynew theories and new techniques.Exhaust emission control for environmental protection is thelatest and the most important motivation which reinforces thenecessity of alternative-fuel research and utilization. Accordingto the "Journal of Air Pollution Control Association", automobileengine exhaust emissions contributed about 50% of pollutants fromall air pollution sources from 1960's to 1980's in the UnitedStates (USA). This clearly indicates the importance of automobileengine exhaust emission control. [1]1.2 The History of Federally Enacted LawsFrom the beginning of this century, with the rapid increase inthe number of automobiles, air quality declined and smog frequentlyoccurred in large city such as Los Angeles, New York, Chicago,2Mexico City and Tokyo. Based on this situation, some governmentsand Congress began to establish national laws to protect the globalenvironment. The following is a brief summary of the history of USAnational air pollution control laws which relate to automobileexhaust emissions. [2]The first air pollution control act was developed by the USACongress in 1955 (Public Law 84-159, July 14,1955). It was narrowin scope and potential and did not regulate exhaust emissions ofautomobile IC engines. An important amendment to the 1955 act waspassed by the USA Congress in 1960. In the face of worseningconditions in urban areas caused by mobile sources, Congressdirected the Surgeon General to conduct a thorough study of theeffects of motor vehicle exhaust emissions on human health.The first Clean Air Act was passed by the USA Congress in1963. Item 6 of this act, encouraged "efforts on the part ofautomotive companies and the fuel industries to prevent pollution".For the first time, this act provided for federal financial aid forresearch and technical assistance as well as a formal process forreviewing the status of the motor vehicle pollution problem.The Motor Vehicle Air Pollution Control Act was passed by theUSA Congress in 1965. This act formally recognized the technicaland economic feasibility of setting automotive emission standards,and the national standards should be set for automotive emissions.Finally it stipulated that controls would be tightened astechnological advances became available in conjunction withreasonable costs.3The Environmental Protection Agency (EPA) established in 1971is a US national administrative agency which is responsible forimplementing of the requirements of the Clean Air Act.1.3 The Diesel Engine Exhaust Emission ProblemThe most recent EPA Clean Air Act Amendments (CAAA) extend to1998 and include the emission standards for urban bus heavy-dutyengines shown in Table 1.1. In this table, NO refers to oxides ofnitrogen (NO or NO2 expressed as NO equivalent), HC designatesunburned non-methane hydrocarbon in gaseous form, CO denotes carbonmonoxide, and PM denotes particulate matter.Table 1.1: EPA CAAA Emission Standards for Urban Bus Engines.URBAN BUS HEAVY-DUTY ENGINE EMISSION STANDARDS(g/bhp-hr measured during EPA heavy-duty engine test)Year^NO ^CO^PM1990^6.0^1.3^15.5^0.601991^5.0^1.3^15.5^0.251993^5.0^1.3^15.5^0.101994^5.0^1.3^15.5^0.051998^4.0^1.3^15.5^0.05With electronic controls, present production diesel enginescan meet the restrictions of HC and CO on the 1993/94 EPA standardsNOx - PARTICULATE TRADE-OFFHEAVY DUTY TRANSIENT CYCLEProduct onEnginesODC Series 50 DiDDC 8V-92 Diesel-Trap^ CummInQ L-10 DIseal-TrapG DOC 8V-82TA M-100 CuminQ L-10 G1996 EPA 1998EPA020.190.18O 170.160.15O 140.130.12O 110.10.08O 080.070.08O 050.040.03O 020.014when they operate with conventional liquid diesel fuel. But to meetthe NO and soot particulates standards is difficult, as shown inFigure 1.1.2^4^8BRAKE SPECIFIC NOx (g/BHP-hr)Figure 1.1: 1996 and 1998 EPA Standards as well as NO and SootParticulates Emission Level of Present Engines.Figure 1.1 shows how far the 1996 and 1998 EPA standards forNO and soot particulates are below the NO and particulatesemission ranges of the most production engines. These are the majorproblems for diesel engines operating with conventional liquiddiesel fuel. As shown, the NO and soot particulates emissions of5DDC 6V-92 and Cummins L-10 diesel engines equipped withparticulates trap are close to the 1996 EPA standards; the NO andsoot particulates emissions of methanol fuelled DDC 6V-92TA M-100(compression-ignition engine) and CNG fuelled Cummins L-10 CNG(spark-ignition engine) meet the 1998 EPA standards.In order to solve NO and soot particulate emission problemswhile maintaining the high thermal efficiency of diesel engineoperation, natural gas has been considered as an alternative fuelfor engines. [3]1.4 Natural Gas as an Alternative FuelNatural gas consists mainly of methane. The conclusions ofprevious research indicate that natural gas is a promisingalternative fuel for internal combustion engines [4]. It producesless soot particulates and NO emissions than conventional dieselliquid fuels and is thus considered to be a potentially cleanburning fuel.Because natural gas is already a widely utilized fuel in NorthAmerica, existing natural gas extraction and processing techniquesmake it ready to use at low cost.Natural gas has a high autoignition temperature [5]. Since itwould not autoignite inside a diesel engine of the highest feasiblecompression ratio, a supplemental ignition device is needed toignite the natural gas. In a compression ignition engine (ie.diesel engine), a quantity of liquid diesel fuel (called pilot-diesel fuel) can be injected with natural gas to provide6temperature high enough for subsequent autoignition of the naturalgas. Thus a gas-diesel fuel injector and its electronic controlmodule are needed to be designed to achieve this concept. Pilot-diesel fuel is defined as a small quantity of liquid diesel fuelinjected with (or before) natural gas and burns before natural gasfor heating the cylinder environment, thereby igniting the naturalgas. The gas-diesel fuel injector is defined as an injector thatcan inject pilot-diesel and CNG fuels at a desired timing. A gas-diesel engine is defined as a compression-ignition engine operatingwith natural gas and pilot-diesel fuel.1.5 Objectives of This ResearchBased on environmental, economic and energy security concerns,the Alternative Fuels Laboratory of UBC is presently developing agas-diesel injection system for gas-diesel engines. The objectivesof my research are the following:1. To investigate the influences of injector geometry and engineoperating parameters on engine performance and exhaustemissions (ie. NO„ HC, CO etc.).2. To correlate engine experimental NO emission data withequilibrium calculations and results of numerical simulationof combustion.3.^To provide fundamental information to assist in the design ofgas-diesel fuel injectors.71 . 6 MethodThe research program is mainly experimental, and utilizes anexisting two-stroke diesel engine whose design is thecharacteristic of the engines most widely used for urban buses inNorth America. The engine is equipped with a dynamometer andinstrumentation for processing the engine performance and emissionmeasurements. A gas-diesel fuel injector with an electroniccontroller is supplied for fuelling the engine. The principalvariables of concern are:Injector geometrical parameters: -- Fuel injection angles.-- Fuel jet sheet configuration.Engine operating parameters: -- Engine load.-- Engine speed.-- Injection timing.-- Compression natural gas (CNG) injection pressure.-- Pilot-diesel/total fuel energy ratio.-- Pilot-diesel cetane number.The effects of these principal variables on the engineperformance and emissions are investigated and discussed in thesubsequent chapters.82 . LITERATURE REVIEW2.1 IntroductionA compression-ignition (CI) engine is defined as an engine inwhich the fuel is directly injected into the cylinder (orcombustion chamber), and autoignites in compressed air. A dieselengine may be defined as a CI engine operating with liquid fuel. Wemay define a natural gas engine as a CI engine operating withnatural gas fuel and a gas-diesel engine as a CI engine operatingwith natural gas and pilot-diesel fuels.In this chapter, the combustion characteristics of natural gasand the methodology of gas-diesel engine operation are discussedfirst. This is followed by a review of the theory of pollutantformation in diesel engines. The final section describes previousexperimental studies of exhaust emissions from compression-ignitionengines.2.2 Combustion Characteristics of Natural GasMethane (the main component of natural gas ) is the simplestand the most stable member of the hydrocarbon family of fuels. Itis a potentially clean burning fuel which can provide superiorexhaust emission characteristics under most operating conditions[6] [7] [8]. Exhaust emissions include soot, particulates and NO.Previous experimental studies [9] [10] [11] [12] of the shocktube ignition of methane-ethane-air mixtures, and Westbrook and9Pitz's modelling study [13] of the autoignition of premixed naturalgas at typical compression-ignition (CI) engine condition (10-30atm, 1000-1100 K) have shown that fuel composition andgas/environment temperature are the most important parametersaffecting the autoignition delay time; the initial environmentalpressure is less important. The autoignition delay time of the fuelcan be considered an index to indicate ignitability of the fuel. Anacceptable value for liquid diesel fuel used in a diesel engine isabout 2 milliseconds, which corresponds to 12 crank angle degreesat 1000 rpm.The low pressures (typically 1-4 atm) and high temperatures(1300-2100 K) of shock tube data may not be representative of CIengine environments, and Westbrook and Pitz's premixed natural gasautoignition may differ from the diffusion mixing in CI engines.However, the conclusions drawn from the above studies areconsistent with recent results of Fraser, Siebers and Edwards [5]who studied the autoignition delay time characteristics ofdirect-injected methane and natural gas under simulated CI enginecondition (5-55 atm, 600-1700 K). They found that a compressiontemperature of between 1200 to 1300 K is needed to operate a CIengine with natural gas (ie. achieve autoignition delay time lessthan 2 ms) in the absence of a supplementary ignition source andwithout a chemical ignition improver. This implies that in anaturally aspirated CI engine with a bottom-dead-centre (BDC)temperature of 325 K and pressure 1 atm, the compression ratiorequired to reach the 1250 K top-dead-centre (TDC) temperature10isentropically would be in excess of 29:1. In contrast, about 800K (compression ratio of 10:1) is needed to autoignite cetane in 2ms [14].The reason for the methane having such high autoignitiontemperature is that the first-broken C-H bond in methane needs moreenergy than the others and certainly more than the C-H bonds inlonger-chain hydrocarbons [15].In order to convert diesel engines into natural gas engines,several methods have been proposed to raise the end-of-compressiontemperature. The first method is to raise the compression ratioover 29:1; but this is impractical because of design difficulties(for example, the minimum clearance for valve lift, and themechanical loading associated with higher cylinder pressure). Thesecond method is to supercharge the intake air or to recirculatepart of the exhaust gas. The third method is to use a supplementalignition source.There are two supplemental ignition methods recommended byprevious work [4]: spark ignition and pilot-diesel-fuel-injectionignition. If the gas and air are mixed before entering the engine,the spark-ignition method (with throttling to control the fuel-airratio) is more suitable than the pilot-diesel-fuel method. This isbecause the air-fuel mixture is uniform inside the cylinder and canbe ignited easily with a spark plug (regardless of the location) ifthe spark timing can be controlled closely. The thermal efficiencyhowever, is lower because of the requirement to reduce compressionratio to avoid compression knock, and because of throttling losses.11Pilot-diesel-fuel-injection ignition is used together withdirect injection of high pressure natural gas into the cylinder, asdiscussed by Beck [16]. This will be discussed in the next section.2.3 Gas-Diesel Engine OperationThere are three main approaches for utilizing natural gas ingas-diesel engines. They are natural fumigation method, timed-portinjection method, and direct-injection method (ie. direct injectionof high pressure natural gas into the cylinder). The schematics ofthese three methods are shown in Figure 2.1. Natural fumigation isthe original method to utilize natural gas fuel, and directinjection is the latest method.Natural Fumigation Method: A schematic of the natural fumigation method is shown inFigure 2.1 (A). Natural gas as the main fuel is injected into theair-gas mixer upstream of the inlet manifold. In the mixer and themanifold, the incoming air and the natural gas are premixed. Thenthe premixture is introduced into the cylinder of the gas-dieselengine. A throttle is placed upstream of the mixer to control theair-fuel ratio of the mixture according to the load. A certainamount of pilot-diesel fuel is injected into the cylinder toinitiate combustion of the natural gas.Detailed experimental investigations of the characteristics ofgas-diesel operation with the natural fumigation method have beenmade by Simonson [17], Moore and Mitchell [18] in a single-NATURAL GASPREMIXEDFUEL/AIRGASMIXERAIR(A) NATURAL FUMIGATIONDIESEL FUELAIR DIESEL FUEL(13) TIMED PORT INJECTIONHIGH PRESSURE^DIESEL FUELNATURAL GASAIRI(C) DIRECT INJECTIONGAS INJECTIONFigure 2.1: Methods of Using Natural Gas in Gas-Diesel Engine.1213cylinder compression ignition engine with natural aspiration, andlater by Wong [19] in a twelve-cylinder two-stroke diesel engine(Detroit Diesel 12V-149T) as well as by Ding and Hill [20] in afour-cylinder pre-chamber turbocharged diesel engine (Caterpillar3304). One of the main findings of these studies is that gasengines using natural gas premixed with intake air (naturalfumigation method) have lower engine thermal efficiency than thatof straight diesel operation, especially at part load. This iscaused by throttling losses or poor combustion of the fuel.Although this method has poor part-load performance, it isstill used in stationary and mining engines for driving powergenerators, compressors and pumps. In these applications, however,the engines are usually operated at full-load and constant speedwith fully opened throttle.Timed-Port Iniection Method: A schematic of the timed-port injection method is shown inFigure 2.1 (B); the natural gas injector in this method is set atthe inlet manifold close to the intake valve. Low-pressure naturalgas is injected into the inlet manifold a short time before theintake valve closes. A throttling valve may be placed upstream ofthe inlet manifold to restrict the air entry for controlling air-fuel ratio. A certain amount of diesel fuel is injected directlyinto the cylinder to initiate combustion of the natural gas.In order to achieve the same thermal efficiency with dieseloperation as with gas-diesel operation, Beck and Johnson used the14timed-port injection method on a 0M-352 engine [16] and a BelarusD-144 diesel engine (for truck and bus applications) [21].Theoretically, partial stratification of the gas-air mixture mayimprove the part-load thermal efficiency. But experimental resultsindicate that the timed-port injection method does not fullyovercome the disadvantages of the natural fumigation method. One ofthe reasons might be that, within the time period from the gasinjection to the beginning of combustion, the mixing is such thatthe advantages of stratified charge are lost.Direct Injection Method: Figure 2.1 (C) shows the schematic of the direct injectionmethod. A gas-diesel fuel injector replaces the conventional dieselinjector. Pilot diesel is injected into the cylinder with (orbefore) high pressure natural gas (CNG). The combustion of pilotdiesel raises the local temperature high enough to ignite thenatural gas. The appropriate relative injection timing of pilotdiesel and natural gas, however, still needs to be determinedexperimentally. The gas-diesel fuel injector is the key element forthis method.Miyake and his coworkers [22] have successfully studied thefundamental behaviour of the high pressure gas spray jet and foundthat it is similar to the liquid fuel (ie. diesel) spray jet. Theyalso found that the higher the natural gas injection pressure, thehigher the indicated thermal efficiency. A combined gas-diesel fuelinjector was designed by Miyake and his coworkers [23]. They used15a pilot-diesel fuel supply system to control the injection timingand duration of natural gas. A medium speed large marine dieselengine (L35MC engine) was converted into a gas-diesel engine toperform the test. Their experimental results show that directinjection of natural gas with pilot diesel quantities as low as 5%can be achieved while high thermal efficiency and low emissions canbe obtained.Similar work has been done by Wakenell, O'Neal and Baker on atwo-stroke medium-speed locomotive diesel engine. They found thatthe pilot diesel fuel quantities can be reduced to 1.8% withoutknock and that a low emission level can be achieved while keepinggood thermal efficiency [24].A comparison study has been conducted by Lom and Ly [25] on aDDC 8V92TA engine. Three methods were evaluated: 1) post-pilotin-cylinder injection; 2) early in-cylinder injection; 3)timed-port injection. The post-pilot in-cylinder injection methodis the same as the direct injection method, but the natural gas iscontrolled to inject after the pilot diesel. The early in-cylinderinjection method is to inject natural gas into the cylinder justafter exhaust valve closure. In principle, this is similar to thenatural fumigation method. Experimental results indicated that thepost-pilot in-cylinder injection method (ie. direct inject method)is potentially the best method for fuelling the engine with naturalgas. It can rival the diesel operation in all aspects ofefficiency, peak pressure and cylinder pressure rate of increaseand also match closely the diesel operation on HC emission.16As concluded by Beck [16], the direct injection method hasthe following advantages:1. Uses basic diesel cycle with compression ignition of pilotfuel followed by high pressure gas injection.2. No detonation limit if gas injection is simultaneous withliquid fuel pilot injection.3. Unthrottled.4. Lean burn, requiring no mixture ratio control.5.^Diesel cycle efficiency.2.4 Pollutant Formation in Diesel EnginesOxides of nitrogen (nitric oxide, NO, and small amounts ofnitrogen dioxide, NO 2 , collectively known as NO R ), carbon monoxide(CO), unburned or partially burned hydrocarbons (HC) andparticulates are the main pollutants from the exhaust gas of theinternal combustion engine.In the diesel engine, the fuel is injected into the cylinderjust before combustion starts, so throughout most of the burningperiod the fuel distribution is nonuniform. The pollutant formationprocesses are strongly dependent on fuel distribution and how thatdistribution changes with time due to mixing as well as combustiontemperature. The mechanisms of the formation of the above pollutantspecies can be explained as following [26].Nitric Oxide (NO): Because NO and NO 2 are usually grouped together as NO17emission, NO is the predominant oxide of nitrogen produced insidean engine cylinder, so only NO formation will be discussed in thissubsection.In principle, the formation and destruction processes of NOare not part of the fuel combustion process, but they take place inan environment created by the combustion reactions. NO formsthroughout the high-temperature burned gases behind the flamethrough chemical reactions involving nitrogen and oxygen atoms andmolecules, which may not attain chemical equilibrium. The principalchemical reaction equations are the following:0+N2=NO+NN+02 =N0+0 (2.1)N+01-1=NO+HStarting from these three equations, the initial NO formationrate can be derived and expressed by the following formula(referring to Appendix A and [27] for the detailed derivation ofinitial NO formation rate):d[ ^_6x1016 exp(  -69 1 090 ) [02 6, "]^[AT2]dt^T1/2(2.2)where the unit of d[NO]/dt is mol/cm's and [ ]e denotes theequilibrium concentration. From the exponential term of thisformula, we see that the NO formation rate has very strongdependence on temperature.As described by Heywood [26], the NO formation process in thecylinder is as follows. When the burned gas temperature inside the18engine cylinder increases during the combustion reaction, the NOformation rate stays at a low level until the temperature reachesa threshold value. Above the threshold temperature, the NOformation rate increases sharply with temperature. Also, theformation rates are highest in the regions close to stoichiometric.Then, as the burned gas cools during the expansion stroke, the NOreaction suddenly freezes (ie. both forward and backward reactionsfreeze). It leaves the NO concentration at a level that is muchhigher than the level corresponding to equilibrium at exhaustcondition.As shown in Eq. (2.2), the concentrations of 02 and N, alsoaffect the NO formation rate (ie. reducing air concentration in thecylinder will also decrease the NO formation rate). This indicatesthat exhaust gas recycling (EGR) and residual gas kept in thecylinder are the methods to reduce NO formation.Soot and Particulates: Diesel particulates consist principally of combustiongenerated soot on which some organic compounds are absorbed. Mostparticulate material result from incomplete combustion of fuelhydrocarbons; some result from burning the lubricating oil.The results of fundamental studies of soot formation in simplepremixed and diffusion flames, stirred reactors, shock tubes andconstant-volume combustion bombs are available in a recent review[28]. As quoted by Heywood [26], soot particles form primarily fromcarbon in diesel fuel (C12H26). The formation process starts with a19fuel molecule containing 12 to 22 carbon atoms and an H/C ratio ofapproximately 2, and ends with particles typically a few hundrednanometres in diameter, composed of spherules 20 to 30 nm indiameter, each containing about 105 carbon atoms and having an H/Cratio of about 0.1. Although the results of these fundamentalstudies of soot formation may not fully apply to the diesel engineenvironment (which has high gas temperatures and pressures, complexfuel composition, and turbulent mixing), they imply that naturalgas fuel jets (CH 4 ) will likely form much less soot than dieselfuel sprays.Experimental studies of the soot formation in diesel engineshave been conducted by Whitehouse and his coworkers [29] on a largedirect-injection engine (30.5 cm bore and 38.1 cm stroke), byAoyagi and his coworkers [30] on a small direct-injection enginewith swirl, as well as by Duggal and his coworkers [31] on a IDIswirl chamber engine. They found that soot forms in the richunburned core of a fuel spray at high temperature and in low oxygenconcentration. In such regions, the fuel (including unvaporizedliquid diesel droplets and diesel vapour) is heated by burned gasbut with insufficient oxygen to oxidize it. Therefore, pyrolysis ofthe fuel takes place, and thereby soot is produced. As the sootcontacts oxygen in the flame zone, it is then oxidized.Approximately 90% of soot is oxidized before exhaust [26]. Swirlspeeds soot mixing with air and tends to reduce soot concentration[30].20Gaseous Hydrocarbons (HC): Gaseous hydrocarbons (HC) originate mainly from incompletecombustion of fuel. Fuel that vaporizes from the nozzle sac volumeduring the later stages of combustion is also a source of HC [26].After injection into a cylinder in the diesel engine, some ofthe fuel will have rapidly mixed with air to equivalence ratioslower than the lean limit of combustion (called locally over-leanmixture); some will be within the combustible range; and some willhave mixed more slowly and be too rich to burn (called locallyover-rich mixture). Slow reaction of the mixture, or even noautoignition and no flame propagation, caused by locally overleanor overrich mixtures is one source of the HC emissions. Theoverrich mixture may burn later by mixing with air oralready-burned gases within the time available before rapidexpansion and cooling occurs. The overlean mixture can be oxidizedonly by relatively slow thermal-oxidation reactions which areincomplete, and is believed to be an important part of HCemissions.At the end of the fuel-injection process, the injector sacvolume (the small volume left in the tip of the injector after theneedle seats) is filled with fuel. As the combustion and expansionprocesses proceed, this fuel is heated and vaporized, and entersthe cylinder at low velocity through the nozzle holes. This fuelvapour mixes relatively slowly with air and may escape the primarycombustion process, thereby becoming another source of HCemissions.21Carbon Monoxide (CO): Carbon monoxide (CO) also forms during the fuel combustionprocess, depending on the fuel/air equivalence ratio. Usually COforms in the locally fuel-rich mixture regions because ofinsufficient oxygen to oxidize all the carbon atoms in the fuel toCO2. High-temperature dissociation is another source of COemission. Diesel engines, however, always operate well on the leanfuel-air mixture side of stoichiometric, so diesel engine COemission is generally insufficient to be significant [26].2.5 Exhaust Emission Studies in Compression-ignition(CI) EnginesFor diesel engines, one of the main concerns is NO (or NOR)emission. It is well established in theory that NO is frozen atequilibrium levels corresponding to the local maximum flametemperature. This indicates that a reduction in maximum flametemperature will reduce the NO formation.Emission Studies in the Diesel Engines: An experimental investigation was made by Torpey, Whiteheadand Wright [32] on both direct-injection (DI) and swirl-chamber(IDI) versions of the engine (Ricardo E16 single-cylinder researchengine with 140-mm stroke and 121-mm bore). The purpose of theirwork was to study the effects of injection timing, exhaustrecirculation (hot and cold), water injection into the inletmanifold, compression ratios, and combustion chamber configurations22on exhaust emissions (NO„, HC and CO). Their major findings are asfollows:1. Retarding the injection timing from optimum performanceinjection timing significantly reduces the NO emission (witha decrease of 55% at 8° retard), while neither HC nor COemission show any significant changes. But, the thermalefficiency suffers.2. Recycling of 15% cold exhaust gases (to constitute 15% of theintake charge) reduces NO in half with 10% power penalty. 10%hot gas recycling reduces NO by 25% with 10% power loss.3. Inducing water into the intake system reduce NO by half withlittle effect on performance.4. Redesigning the combustion chamber and varying the compressionratio have very little effect on emissions.A similar study was made by Khan and Wang [33] on a DI dieselengine. Their results confirm that retarding injection timing cancause a significant reduction in NO emission. They also found thatan increase in the rate of injection or air swirl reduces theexhaust smoke emission.Herzog, Burgler and Winklhofer [34] suggested that controllingin-cylinder NO formation requires the control of both the mixingand the combustion process with respect to local oxygen-nitrogenconcentrations, local temperature, and their temporal development.Their latest experimental investigations were focused on thefollowing aspects:1.^In-cylinder charge conditions (charge temperature, pressure23and rotation).2. Fuel injection system parameters.3. Exhaust gas recirculation (cooled and uncooled EGR).4.^fuel formulation.Their conclusion is, in order to meet US 1998 (49 states)HD/MD Standard (ie. EPA 1998 Heavy-Duty/Medium-Duty Diesel EngineEmission Regulations); CAL 1995 MD-TLEV Standard (California 1995Medium-Duty, Transitional-Low-Emission Vehicle EmissionRegulations); and CAL 1998 MD-LEV/MD-ULEV Standard (California 1998Medium-Duty Low-Emission Vehicle/Medium-Duty Ultra-Low-EmissionVehicle Emission Regulations), optimized combustion system andreformulated fuels with higher cetane-numbers have to be used aswell as additional elements such as EGR (uncooled or cooled),injection rate control, and DEN0,-catalyst have to be added.Emission Studies in Gas-Diesel Engines: An experimental study of the exhaust emission characteristicsof gas-diesel operation with the natural fumigation method was madeby Ding and Hill [20] in a four-cylinder pre-chamber turbochargeddiesel engine (Caterpillar 3304). The variables of their studyincluded pilot diesel fuel proportion, intake air restriction,engine speed, and pilot diesel injection timing. Their conclusionsare as follows:1.^Generally speaking, gas-diesel operation (with naturalfumigation) produces higher HC and CO emission than straightdiesel operation, especially at low loads. This is caused by24poor combustion.2. At low loads, air restriction increases the mixture strength,leading to improvement of combustion and reduction of unburnedhydrocarbon emission.3. Advancing injection timing at low loads can reduce HC emissionwith little influence on CO and NO emissions. Retardinginjection timing at high loads can reduce NO emission to agreat extent with little effect on HC emission.Beck and Johnson [21] studied the exhaust emissioncharacteristics of gas-diesel operation with the timed-portinjection method. Their experimental results show that HC and NOemissions with gas-diesel operation are higher than those withstraight diesel operation.A study by Lom and Ly [25] on a DDC 8V92TA engine, asmentioned before in this chapter, evaluated three methods:post-pilot in-cylinder injection, early in-cylinder injection andtimed-port injection. Unfortunately, only HC emission was measuredin this study. The results indicate that the potentially bestmethod for fuelling an engine with natural gas is the post-pilotin-cylinder injection method. It can closely match the straightdiesel operation on HC emission.253 . EXPERIMENTAL APPARATUS3.1 IntroductionThis chapter describes the experimental apparatus used in thisresearch for this thesis. The test engine and its control and thedata acquisition system are described first. followed by thespecifications of the exhaust emission analysis systems.Two exhaust emission analysis systems (EEAS), EEAS-A and EEAS-B will be explained in this chapter. EEAS-A was established in 1985and was used until August, 1992. EEAS-B is an updated system. Adescription of both systems is necessary, because the experimentalresults discussed in subsequent chapters are acquired from thesetwo systems.3.2 Test Engine and Test Control SystemA schematic diagram of experimental apparatus andinstrumentation is shown on Figure 3.1. Intake air, compressednatural gas (CNG) and pilot diesel are measured and introduced intothe fully instrumented test engine. Compressed natural gas andpilot diesel fuels are supplied by the natural gas and dieselfuelling systems which will be explained in Subsection 3.2.3.Figure 3.1 also shows that the water-brake dynamometer with speedand load sensors is used to measure torque. An exhaust emissionanalysis system and a BOSCH smoke sampling pump are connected tothe engine exhaust pipe. The engine dynamometer control console is26Figure 3.1: Schematic of Experimental Apparatus andInstrumentation.used to control engine operation. A PC-based and monitored dataacquisition system is used to obtain and to record engineperformance, exhaust emission, and pressure data.3.2.1 Test EngineA Detroit Diesel Series 71 single-cylinder diesel engine was27Table 3.1: General Specifications of 1-71 Diesel Engine.Bore & Stroke:^4.25 in.(108 mm) & 5.0 in.(127 mm)Displacement: 70.93 Cu. in. (1.162 litres)Compression ratio:^16 to 1Maximum operating speed:^1600 RPMRated speed:^ 1200 RPMIdle speed: 500 RPMRated output:^ 10 kWMaximum bmep (break meaneffective pressure):^5 barScavenging type:^Unif low1, Injector-Fuel,3. PIn-Dowel,24, Rock-Control,28, Tip-Spray,36, Clamp-Injector.37. Stud-Injector' Clamp.38. Washer,39, Nut40. Tube-Injector.41. RIng-Seal.55, Shaft-Rocker,56. Arm-Injector Rocker.59, Shaft-Injector Control,60. Lever-Rock Control,67. Chamber-Combustion.68, Pipe-Fuel Inlet.69. Pipe-Fuel Outlet,70, Connector-Fuel Pipe.Figure 3.2: The Combustion Chamber Geometry and InjectorMounting of the 1-71 Diesel Engine.28used of this study. This engine is a 71 cubic inch displacement,two-stroke, direct-injection diesel engine with forced airaspiration and scavenging by a blower. The engine specificationcharacteristics are shown in Table 3.1.Figure 3.2 shows the geometry of the combustion chamber andthe injector mounting. A shallow bowl-shaped combustion chamber(67) was formed on the top of the piston. A mechanically-controlledfuel unit injector (1) was originally mounted on this engine, butit was replaced with an electronically-controlled unit injector tocontrol the injection timing depending on the temperature, load,and speed [35] [36]. Electronic control was based on the DetroitDiesel Electronic Control Module (ECM) shown in Figure 3.4.Figure 3.3: The Arrangement of the Test Engine Cell.(1) Test Engine, (2) Dynamometer, (3) Cooling Tower.293.2.2 Dynamometer, DCM and DDMFigure 3.3 shows the arrangement of the test engine cell. Thetest engine (1) was coupled to a 600 hp water-brake dynamometer (2)(GO-Power Model DA-316) which converts the rotating engine torqueto stationary torque that can be precisely measured. A coolingtower (3) has been installed to replace the radiator to control theengine cooling-water temperature.Figure 3.4: The Engine Control Console.(1) Dynamometer Control Module (DCM), (2) Data DisplayModule (DDM), (3) Electronic Control Module (ECM).As shown in Figure 3.4, a Dynamometer Control Module (DCM) (1)and a Data Display Module (DDM) (2) are both located on the enginecontrol console in the control room. The engine speed and load,which are controlled through the throttle and load control dial on30the DCM front panel, are detected by the tachometer and the loadcell, and then displayed on the DDM. The DDM also provides displaysof engine cooling water temperature, lubrication oil temperatureand pressure, and provides fault-mode automatic shutdown.3.2.3 Engine Fuelling SystemFigure 3.5: Schematic of Engine Fuelling System.As natural gas (the principal fuel) has a high autoignitiontemperature, a small amount of pilot-diesel is injected into thecylinder to initiate the combustion of natural gas. Two independentfuel supply systems, for diesel and natural gas, were establishedto achieve gas-diesel operation. Figure 3.5 shows the schematic of31the engine fuelling system.Diesel Fuelling System: Commercial grade 1 and grade 2 diesel as well as cetane number62 diesel were used in straight diesel operation. But only cetanenumber 62 diesel was used in gas-diesel operation because it hasbetter ignition quality (ie. shorter ignition delay time) and ismore effective than the others in igniting the natural gas. Theproperties of these diesel fuels are listed in Table 3.2.Table 3.2: Properties of Diesel Fuels.Grade 1^Grade 2^Cetane No.62(Shell)^(Chevron)^(Shell)Higher Heating Value (kJ/kg): 45,094^45,220^45,220Density (kg/m3):^ 860^836Cetane Number: .45^-45^62.2As shown in Figure 3.5, in the diesel fuelling system dieselflows by gravity from the fuel tank to the AVL730 Dynamic FuelMeter (AVL). After leaving the AVL, the diesel is pumped to theinlet of the gas-diesel injector through the filter and theemergency shut-off valve. Then a small amount of returned diesel issent back to the AVL directly from the outlet of the injector. Withthis closed-loop connection, the AVL can measure the net32consumption of diesel fuel.Natural Gas Fuelling System: The natural gas obtained from B.C. GAS through the city gaspipe line was the main fuel. Table 3.3 shows the typicalcomposition of B.C. natural gas, the properties of which are listedin Table 3.4.Table 3.3: Composition of the B.C. Natural Gas.Methane (CH4)^ 95.50^(Volume %)Ethane (C2H6) 3.00Propane (C3H8)^ 0.50Iso-Butane 0.05N-Butane^ 0.10Carbon Dioxide (CO2)^ 0.20Nitrogen (N2)^ 0.60Pentanes 0.04Hexanes^ 0.01Figure 3.5 (page 30) shows that the commercial oil-free 4-cylinder 4-stage Residential Refuelling Appliance (RRA) isconnected to the city gas pipe line. It compresses the natural gasfrom the pipe line pressure (5 psi) to about 3000 psi highpressure. The compressed natural gas (CNG) is stored in 3 gas33storage bottles.Table 3.4: Properties of the B.C. Natural Gas.Molecular Weight (kg/kmol):^16.689Density (kg/m3):^ 0.6903 (at 20°C, 101.325 Kpa)0.7023 (at 15°C, 101.325 Kpa)Lower heating value (kJ/kg):^49,098 (at 15°C, 101.325 kPa)The mass-flow meter (Micro Motion Model DH012), the solenoidshut-off valve, and the pressure sensor are installed on the pipeline between the gas storage bottles and the gas inlet of the gas-diesel injector, as shown in Figure 3.5. A regulator unit ismounted directly on the outlet of gas storage bottle to adjust theCNG back pressure of the injector.Both diesel and CNG shut-off valves are connected to DCM forfault-mode automatic shutdown.Gas-Diesel Fuel Injection System: This system consists mainly of a gas-diesel electronic unitinjector (EUI) (ie. gas-diesel fuel injector) and a modifiedElectronic Control Module (ECM).Figure 3.6 shows the schematic of the gas-diesel electronicunit injector. The gas-diesel fuel injector uses the hydraulicactuation of the original configuration of the Detroit Diesel EUI,34UNITINJECTORDIESELSUPPLY/RETURNINTENSIFIERSPOOLVALVE CHECK VALVERELIEFVALVEPUSHRODDDECSOLENOID_POPPET SEAT ANGLE/ U.S. Potent # 1067,467 November 1991ACCESSORY SHAFTDRIVEN ACTUATORDIESEL PILOTMETERING VALVERETURNSPRINGPOPPETNOZZLEBYPASSCHECKVALVECNG Storage(20-200 loar)LUNGERCHECKVALVEMIXINGRESERVOIRFigure 3.6: Schematic of Gas-Diesel Electronic Unit Injector.which is the electronic solenoid valve controlling injection timingand the cam-driven, and the plunger-bush pressurized device. A CNGpassage, a diesel hydraulic actuated poppet nozzle, a pressurecheck valve and a pilot-diesel metering valve were introduced toprovide gas-diesel injection.The modified ECM is used to control the beginning of injection(BOI) and the duration of injection (or pulse width, PW) of thegas-diesel injector through the solenoid valve. The BOI, the PW andthe pilot-diesel metering valve opening (to control the amount ofpilot diesel injected into the engine) can be set and adjusted fromthe front panel of the engine control console.353.2.4 Engine Cooling SystemTo keep the engine operating at the same temperature so thatreliable and repeatable performance and emission data can beobtained, a cooling tower was used (as shown in Figure 3.3, page28). An thermostatic valve was installed at the water inlet of thecooling tower to keep the engine cooling water temperature between80 and 95 °C and the lubrication oil temperature between 70 and 110°C.3.2.5 InstrumentationEngine speed, torque, inlet air flow rate, CNG mass-flow rateand diesel mass-flow rate are the raw data which can be measureddirectly by the instruments and used to calculate the other engineperformance data. The other raw data, such as intake temperatureand pressure, ambient temperature and pressure as well as relativehumidity, are measured for correcting engine performance data tostandard ambient conditions. Cylinder pressure is measured forcalculating the mass burned fraction and emission simulationcomputation. The following is the description of the instrumentsused to measure the raw data.Engine Speed: A speed sensor mounted on the end of the dynamometer was usedto measure the engine speed. This magnetic pickup type sensorconsisted of a 60-tooth gear and magnetic pickup (sensor). Thissensor assembly developed a frequency signal that was directly36proportional to RPM (ie. 1 RPM corresponds to 1 cycle per second).The frequency signals were displayed on the front panel of theengine control console (ie. on an digital instrument) as well asrun through a digital-to-analog converter (DAC) and sent to thedata acquisition system. This instrument was calibrated by a handdigital tachometer (Shimpo Model DT-205).Torque: A strain gauge load cell mounted on the trunnion support ofthe dynamometer was used to measure load (ie. rotating torque)applied on the engine. The low level analog load signals wereamplified and sent to the data acquisition system as well as to thedisplay of engine control console through a analog-to-digitalconverter (ADC). This instrument was calibrated by applying weightsto the torque arm.Flow Rate: A laminar flow element (Meriam Instrument Model 50MC2) mountedafter the air filter was used to measure air flow rate. Pressuredrop across the element was transmitted by a transducer and thensent to the data acquisition system.Amass-flow sensor (Micro Motion Model DH012) installed on theCNG intake line between the gas storage bottle and the inlet of thegas-diesel injector was used to measure the CNG mass flow rate.This instrument worked on the Coriolis acceleration principle. Thesignal from the mass-flow sensor was sent to the remote flow37transmitter (Micro Motion Model RFT9712) and converted to 4-20 mAcurrent signal which was then sent to the data acquisition system.The mass-flow sensor, in conjunction with a remote flowtransmitter, formed a complete mass flowmeter system.An AVL 730 Dynamic Fuel Consumption Measuring Equipment wasused to measure the diesel mass-flow rate. This instrument workedon a gravimetric measuring principle. The analog signal was sent tothe data acquisition system and displayed on the evaluation modulethrough an analog-to-digital converter (ADC).Ambient Conditions: Ambient temperature, pressure, and relative humidity were readfrom the digital gauges and recorded manually. Ambient temperatureand relative humidity were used to determine the humidity ratio.Humidity ratio and ambient pressure data were used by the dataacquisition system to calculate the correction factor for engineperformance data.Cylinder Pressure: A PCB piezo-electric pressure transducer was installed in asleeve in the cylinder head. The signal from the transducer wastransmitted to a Model 5004 Kistler Charge Amplifier and then tothe data acquisition system. A dead-weight tester was used to checkthe linear slope of the calibration and quality of the transducer.STEADY STATE DATAIBM-PCMULTIPLEX BOX383.2.6 Data Acquisition SystemSTEADY ^SPEED^COMMANDDATA FLOW ^DATA FLOW^____ ,..._ SIGNAL FLOWFigure 3.7: Schematic of Data Flow in Data Acquisition System.An IBM-PC-based data acquisition system was used to acquiresteady-state data and high-speed data from the test engine [37].Table 3.5 lists the names of the steady-state data and high-speeddata as well as their corresponding channel numbers. The dataacquisition system consisted of hardware and software. Figure 3.7shows the general diagram of the data flow in the data acquisitionsystem.The hardware of the data acquisition system included amultiplex box which functioned as a multiswitch, an IBM-PC and anISAAC which functioned as a computer to acquire high speed pressuredata. The multiplex box contained a multiplexer board, a screwterminal board (DT-709), a digital interface board and a vector39trigger board. The IBM-PC was equipped with a data acquisitionboard (PC Lab PCL-818) and a general purpose interface board (GPIB,IEEE-488).Table 3.5: List of Steady State Data and High Speed Data.Name^ Channel^TypeNatural gas mass flow:^No. 0 Steady state dataBeginning of injection:^No. 1^Steady state dataAmbient temperature: No. 2 Steady state dataPulse width:^ No. 3^Steady state dataTorque: No. 4 Steady state dataSpeed:^ No. 5^Steady state dataIntake pressure:^No. 6 Steady state dataDiesel mass flow: No. 7^Steady state dataAir flow (delta P):^No. 8 Steady state dataNatural gas pressure:^No. 9^Steady state dataCH 4 emission:^ No. 10^Steady state data0 2 emission: No. 11^Steady state dataCO2 emission:^ No. 12^Steady state dataCO emission: No. 13^Steady state dataTotal HC emission:^No. 14^Steady state dataNO emission:^ No. 15^Steady state dataCrank angle & BDC index:^ High speed dataCylinder pressure:^ High speed dataInjection hydraulic pressure:^ High speed data40As shown in Figure 3.7, the steady-state-data analog signals(total 16 channels) are directed by the multiplexer to the DT-709screw terminal board. From the board, the signals are sent to thePCL-818 data acquisition board in which they are converted from ananalog voltage to a 12 bit digital number. The digital numbers arethen saved on a specified floppy disk.The high-speed data flow is more complicated than the steady-state data flow as one can see from Figure 3.7. When commanded toacquire high-speed data, the GPIB initiates the ISAAC and gives thecommand to take a specified number of data points, and the PCL-818then sends a digital signal to the vector trigger board whichprepares a trigger for the ISAAC to begin taking high-speed data.Using the next BDC signal as a trigger and the crank angle signalas an external clock, the ISAAC then starts taking data. After thecompletion of data taking, the ISAAC sends a signal to the GPIBthat data acquisition is completed, and the data is transferred tothe designated drive in binary form.A menu-driven program, ENGDATA, controls the data acquisitionsystem. It allows the user to tailor the data acquisition system tohis specific needs. i.e. allows the building of configuration andcalibration files, directs data flow, does calculations with data,converts binary pressure data to ASCII, and allows data to be savedto disk.413.3 The Exhaust Emission Analysis System A, (EEAS-A)The exhaust emission analysis system A (EEAS-A) wasestablished in 1985 at the Alternative Fuels Laboratory of UBC[20]. The EEAS-A has a Model 951 Chemiluminescent N0/NOx Analyzer,a Model 400 Flame Ionization Detection (FID) Hydrocarbon Analyzer,and a Model 865 Non-Dispersive Infrared Analyzer (for CO). Afterthe 1990 modification [38], a Model 880 Non-Dispersive InfraredAnalyzer (for CO2), a Model 1054 02 Analyzer and a sample-gaschiller were added into this system. This system was used untilAugust, 1992. The Model 865 Non-Dispersive Infrared Analyzer (forCO), the Model 880 Non-Dispersive Infrared Analyzer (for CO2) andthe sample-gas chiller were re-equipped in EEAS-B.3.3.1 Arrangement of the EEAS-AFigure 3.8 shows the schematic diagram of the exhaust emissionanalysis system A (EEAS-A). The exhaust sample gas is taken by asampling probe which is mounted downstream on the exhaust pipeabout 2 meters away from engine exhaust manifold outlet flange.After entering the sampling line, the sample gas is pumped firstthrough a coarse filter to eliminate soot and particulates, thenthrough a chiller to remove water, and then through fine filtersand a relief valve before separating into two tubes. In one tube,the sample gas goes directly to HC, CO, CO2 and 02 analyzers. In theother tube, the sample gas is heated and sent to the NO analyzer.The flowmeter and valve after the analyzer are used to adjust thesample gas flow rate passing through the analyzer. Zero and spanCOARSE^CHILLER^FINE^RELIFE^FINEPUMPFILTER (CONDENSER) FILTER^VALVE^FILTER HEATERMAINBYPASS [IIOUTLETModel 105402 kta0er042SAMPLING PROBEFigure 3.8: Schematic of Exhaust Emission Analysis System A.gas cylinders are connected to the analyzer. Pressure regulatorsare used to set the right flow rate for zero and span gases flowinginto the analyzer. Zero and span gases are used to calibrate theanalyzer before tests. The best span gas concentration range isbetween 75% and 100% of full operating scale; the full operatingscale is determined by maximum engine experimental emission data.The output signals from this system are sent to the dataacquisition system and recorded on discs continuously by a IBM PC.433.3.2 Principle of the Analyzers in the EEAS-AThe following summarizes the principles of the major analyzerswhich were in the EEAS-A.Nitric Oxide Determination (Model 951): Figure 3.9: Schematic Flow Diagram of NO/NO, Analyzer.Figure 3.9 shows the schematic flow diagram of the NO/NO,analyzer and the chemiluminescent method. The chemiluminescentmethod for detection of nitric oxide (NO) is based on its reactionwith ozone (03) to produce nitrogen dioxide (NO2) and oxygen (02).Some of the nitrogen dioxide molecules thus produced are initiallyin an electronically excited state (NO:). These revert immediatelyto ground state with emission of photons. The reactions involvedare44NO+ 03 -'NO2 * 02Ar024--NO2+hvwhere: h = Planck's constant; v = frequency, HzAs shown in Figure 3.9, after flowing into the analyzer in thelower tube, the sample gas first passes to a NO,-to-NO converterwhere all NO2 in the sample gas will be convert to NO. Then, thesample gas passes to the reaction chamber. In the upper tube, theair flows into a ozone generator to produce 03 which is then sentto the reaction chamber of the analyzer. As NO and 03 mix in thereaction chamber, the chemiluminescent reaction produces lightemission that is directly proportional to the concentration of NO.This emission is measured by a photomultiplier tube and associatedelectronic circuitry. The NO concentration is then known since itis equal to the NO concentration.Hydrocarbon Determination (model 400): The Model 400A Hydrocarbon Analyzer utilizes the flameionization method of detection. The sensor is a burner in which aregulated flow of sample gas passes through a flame sustained byregulated flows of air and a fuel gas (hydrogen or a hydrogen-diluent mixture). Within the flame, the hydrocarbon components ofthe sample stream undergo a complex ionization that produceselectrons and positive ions. Polarized electrodes collect theseions, causing current to flow through electronic measuring45SAMPLE CAPILLARYSAMPLE INFILTER■,,_ CE00-MODEL 400 HYDROCARBON ANALYZERGAUGEREGULATER-7BURNER..e.....(11_C4L_H^1--,EXHAUSTOUTLETSAMPLE BYPASS BYPASSOUTLETFUELFigure 3.10: Schematic Flow Diagram of HC Analyzer.circuitry. Current flow is proportional to the rate at which carbonatoms enter the burner. Figure 3.10 shows the schematic flowdiagram of HC analyzer.Non-Dispersive Infrared Detection Method (Model 865 and 880): As shown in Figure 3.11, in the analyzer, infrared radiationis produced by two separate energy sources. Once produced, thisradiation is beamed separately through a chopper which interruptsit at a certain frequency. Depending on the application, theradiation may then pass through optical filters to reducebackground interference from other infrared-absorbing components.The infrared beams pass through two cells: one is a referencecell containing a non-absorbing background gas, and the other is asample cell containing a continuously flowing sample.oolore6:bloc)O(I cPCoo--ooo ooolooB9Isc92Ptec•• •CHOPPERSAMPLECELLDIAPHRAGMDISTENDEDDETECTOR• ••• ••• • ••• •••SIGNALCONDITIONINGCIRCUITRY• COMPONENT OF INTEREST0 OTHER MOLECULESINFRAREDSOURCEREFERENCECELLINFRAREDSOURCESAMPLE INSAMPLE OUTFigure 3.11: Schematic of Non-Dispersive Infrared DetectionSystem with Double-Beam Method.During operation, a portion of the infrared radiation isabsorbed by the component of interest in the sample such as CO andCO„ with the percentage of infrared radiation absorbed beingproportional to the component concentration. The detector is a "gasmicrophone" operating on the Luft principle. It converts the4647difference in energy between sample and reference cells to acapacitance change. This capacitance change, equivalent tocomponent concentration, is amplified and indicated on a meter.3.3.3 Drawback of the EEAS-AThe major drawback of this system was that precise results oftotal hydrocarbon (HC) and nitric oxides (NO) could not beobtained. The reason for that is the following. The sampling pipe,the filters, and the hydrocarbon analyzer were not heated, so watercondensed from the sample gas on the way to the analyzers. Asample-gas chiller was used to protect analyzers from watercorrosion. But the sample-gas chiller also condensed somehydrocarbons which are heavier than methane and some nitrogendioxides (NO2). This caused the readout of the hydrocarbon andnitric oxides analyzers lower than the true concentration. Toovercome this drawback, a new system, EEAS-B, has been establishedsince September, 1992.3.4 The Exhaust Emission Analysis System B, (EEAS-B)The major advantage of this new system is that it provides aprecise measurement for total hydrocarbon (THC) and nitric oxides(N0x). In contrast to the EEAS-A, the EEAS-B has a heated sample-gas sampling system which includes heated sample-gas tubes, heatedfilters, heated pump and heated valves. The new instrumentsinstalled on the EEAS-B are the heated total hydrocarbon (THC)analyzer (RATFISCH Model RS-55), the oxygen analyzer (SIEMENS48OXYMAT 5E), the CH4 and NO analyzer (SIEMENS ULTRAMAT 22P), andheated NO2 to NO converter. The instruments transferred from theEEAS-A to the EEAS-B are the carbon monoxide, the carbon dioxideanalyzers, and the sample-gas chiller.3.4.1 Arrangement of the EEAS-BFigure 3.12: Schematic Flow Diagram of the First Cabinet ofExhaust Emission Analysis System B (EEAS-B).Generally, the EEAS-B consists of two cabinets of instruments.Figures 3.12 and 3.13 show the schematic flow diagrams of the firstBACK MANCOSIEMENSNO/CH4ULTRA MATCABINET 2SC-1FM1#3COC^SC-3^SC-4 -1C SAMPLE^0F^M3#1#2NOx SAMPLEN2FM2BECKMANCO2FM4SIEMENS02OXYMAT49and the second cabinets of the EEAS-B. Cabinet 1 comprises all theheated instruments and the heated auxiliary as well as the sample-gas chiller; Cabinet 2 consists of all the "cool" instruments.Figure 5.13: Schematic Flow Diagram of the Second Cabinet ofExhaust Emission Analysis System B (EEAS-B).SC-2As shown in Figure 3.12, the exhaust sample gas is taken by asampling probe mounted downstream on the exhaust pipe about 2meters away from the engine exhaust manifold outlet flange (Sameconfiguration as in EEAS-A, but not shown in Figure 3.12). Afterentering the heated sampling line, the heated sample gas first50enters the first cabinet, passes through a heated coarse filter(DH) and a heated fine filter (BH) to remove soot and particulates,then separates into two tubes. In the downward tube, the heatedsample gas goes directly into the heated total hydrocarbon analyzer(RS55). In the upward tube, the heated sample gas passes through aheated pump before a second-time separating into two tubes. In onetube, the heated sample gas goes directly into a NO2 to NOconverter to convert all NO2 into NO, then passes through thechiller to condense out water, and finally goes to NO/CH4 analyzerwhich is located in Cabinet 2 shown in Figure 3.13 (The NO/CH4analyzer detects the NO and CH4 concentrations at same time). Inthe other tube, the sample gas flows out of heated closure and intothe chiller. After condensing, the dry cool sample gas flows out ofCabinet 1 into Cabinet 2.In Cabinet 2 as shown in Figure 3.13, the cool sample gasseparates into three tubes which lead to CO2, CO and 02 analyzers.The flowmeter is placed in front of each analyzer (except the totalhydrocarbon analyzer) to adjust the sample flow rate entering theanalyzer. Zero and span gases are used to calibrate the analyzerbefore a test.3.4.2 Principle of the Analyzers in the EEAS-B SystemTotal hydrocarbon (THC) analyzer, NO/CH, analyzer, oxygen (OAanalyzer, carbon monoxide (CO), and carbon dioxide (CO2) analyzersare five analyzers in the EEAS-B. The measurement principle of theCO and the CO2 analyzers was discussed in subsection 3.3.2. The new1. Detector cell2, Secondary detector'level3, Microflow sensor4. Primary detector level5. Window6, Analyzer cell7. Sample gas inlet8. Chopper9, IR source10, Reflector'11, Chopper motor51THC analyzer has the same detection principle as the old HCanalyzer, which operates on the flame ionization detection methoddescribed in subsection 3.3.2. The general principle of the NO/CH,analyzer and the 02 analyzer will be discussed in subsequentsubsections.Measurement Principle of NO/CH,  Analyzer: Figure 3.14: Schematic of Non-Dispersive Infrared DetectionSystem with Single-Beam Method.This instrument operates on the non-dispersive infraredabsorption principle using the single-beam method with an opto-pneumatic double-layer detector. Figure 3.14 shows the schematicdiagram of the detection system of the analyzer. The radiationspiral (9), heated to approx. 600°C, emits infrared radiation whichis modulated by a chopper (8). After passing through the analyzer52cell (6) which contains sample gas, the intensity of the radiationis measured selectively by a specific gas in a double-layerdetector cell (1). The gas in the detector cell is capable ofabsorbing radiant energy and converting it into temperatureincrement. A gas mass flow caused by the temperature incrementbetween two detector cells is thus produced, and measured by amicrof low sensor (3) which converts this flow signal into anelectrical output signal.Measurement Principle of 0? Analyzer: 1. REFERENCE GAS2, RESTRICTORS3, MICROFLOW SENSOR INMEASURING SYSTEM4, REFERENCE GAS CONDUITS5, SAMPLE GAS6. SAMPLE CHAMBER7, PARAMAGNETIC MEASURINGEFFECT8. ELECTROMAGNET WITHVARYING FLUX INTENSITY9. MICROFLOW SENSOR INCOMPENSATING SYSTEMFigure 3.15: Schematic of Oxygen Measurement Principle.Oxygen molecules in a non-uniform magnetic field are attractedto the strong part of the field because of their paramagneticproperty. If two gases having different oxygen contents are brought53together in a magnetic field, a pressure differential will begenerated between them. The pressure differential is proportionalto the oxygen content of the sample gas. This is the basicmeasurement principle of the 02 analyzer.As shown in Figure 3.15, in the analyzer, one of the gases isthe sample gas, and the other is a reference gas (N2 of maximumpurity, 02 or air). The reference gas is admitted to the samplechamber through two conduits (4). One stream of the reference gasmixes with the sample gas in the area of the magnetic field.Because the two conduits are interconnected, the pressure which isproportional to the oxygen content of the sample gas produces aflow which is measured by a microflow sensor (3) and is convertedinto an electrical signal.3.4.3 Relationship between the EEAS-A and the EEAS-B SystemsFrom the operation principles for each analyzer, as well asthe overall arrangements of both systems, we can draw someconclusions:1. CO and CO2 readout from both systems should be consistentbecause they are using the same analyzers and the readout isthe same on the dry basis.2. Total hydrocarbon (THC) readout from the EEAS-A system is onthe dry basis and has non-methane hydrocarbon condensationoccurring, whereas the readout from EEAS-B system is on thewet basis.3.^Although the NOx readout is on the dry basis from both54systems, the one from the EEAS-A system may be slightly lowerthan the one from the EEAS-B system, caused by some NO2condensing with water before it enters the NO2 to NO converterand detector in the EEAS-A system. Furthermore, the twoanalyzers used in the two systems use different detectionmethods, so that the original readout might also be slightlydifferent.3.5 Smoke Determination (BOSCH Smoke Meter)Figure 3.16: Schematic of BOSCH "Spot" Smokemeter.Independent of the above systems, the concentration of dieselexhaust soot particles is determined by photoelectric evaluation55using a portable Bosch "Spot" Smokemeter [39]. Figure 3.16 showsthe measurement principle of this instrument. A spring-operatedsampling pump draws a fixed volume of exhaust gas from the exhauststream through a controlled density paper filter disc (not shown inFigure 3.16). Soot particles from the sample are deposited on thefilter disc, causing it to darken in proportion to the sootparticle concentration. A separate 110 V AC or battery-poweredphotoelectric device measures the light reflected from the darkenedfilter disc. Readout is by a milliammeter calibrated in 0-10 units.564 . EXPERIMENTAL PROCEDURES4.1 IntroductionThis chapter discusses the three experimental procedures:engine performance and emission test procedure; engine performancecalculation procedure; and engine exhaust emission calculationprocedure.The experimental work was divided into two test categories:diesel baseline tests and gas-diesel operation tests. Gas-dieseltesting was carried out with different injector geometries andengine operating parameters. The detailed test procedure will bedescribed in a subsequent section.Nine indices were chosen to represent engine performance andemission characteristics: thermal efficiency, brake mean effectivepressure (BMEP), nitrogen oxides emission (NO,), total hydrocarbonemission (THC), non-methane hydrocarbon emission (NMHC), unburnedmethane emission (CH,), carbon monoxide emission (CO), carbondioxide emission (CO2), and Bosch smoke index.The engine performance indices, thermal efficiency, and brakemean effective pressure (BMEP), were not measured directly andrequire calculations discussed in the section on engine performancecalculation procedure (page 62).The engine exhaust emissions can be expressed as dry-basis,wet-basis or brake specific emissions. The first and secondemissions are stated in parts per million (ppm) by volume. The(a) (b)PILOT DIESEL—iINJECTORCNGSUPPLYCNG/DIESELMIXINGCHAMBERCASTELLATED-ENDSLEEVE-UEL INJECTIONANGLE57third is stated in a mass unit as kg/kW-hr. The conversion betweenthe dry-basis and the wet-basis emissions as well as thecalculation of brake specific emissions (from the wet-basisemissions) are explained in the section concerning engine exhaustemission calculation procedure (page 64).4.2 The Engine Performance and Emission Test ProcedureFigure 4.1: Schematic of Injector with Castellated-End Sleeve.Engine performance and emission tests were conducted with both58diesel and gas-diesel fuel injectors. Baseline testing with purediesel fuel was conducted for comparison with the results of gas-diesel fuelling. Most tests were done with the gas-diesel injectorsupplied with compressed natural gas fuel and pilot diesel (ie.ignition liquid). In the original configuration of the gas-dieselinjector (see Fig. 3.7 in chapter 3), a poppet nozzle was used sothat the fuel jet formed a conical sheet. The injection angle (ie.the poppet seat angle) is defined as the angle between the poppetseat and the cylinder head surface. Figure 4.1 shows the schematicof injector configuration with a castellated-end sleeve tointerrupt the conical sheet jet in part of the test. Fuel jetinterruption is a method to increase jet penetration, injectionspeed and tendency of jet stability [42].The injector geometries and engine operating parameters variedin the engine performance and emission tests are listed in Table4.1. The fuel jet interruption ratio is defined as the ratio of theshrouded area to the total column area; it is also referred to asshroud %.In the baseline tests, a production model electronic unitinjector (controlled by an electronic module with manufacturer-preprogrammed operating BOI and PW) was used to provide the dieseloperation. The engine was operated at constant speed with loadsetting at increments from 0.5 to 5 bar.With gas-diesel operation, a modified electronic module wasused to control the gas-diesel electronic unit injector. With thismodule, BOI, PW and diesel/total fuel energy ratio can be easily59adjusted to desired values within the testing range. The pilotdiesel used for most cases was cetane number 62 diesel. The enginespeed was kept constant at 1200 rpm for most cases by adjusting thePW (ie. the duration of the fuel injection). The injection angleswere adjusted by changing the poppet nozzle. The fuel jetinterruption ratio was adjusted by modifying the castellated end ofthe sleeve.Table 4.1 List of Variable Parameters and Testing Ranges:Injector Geometrical ParametersFuel injection angles:^ 100, 20° and 30°Fuel jet interruption ratio:Conical sheet:^ 0% shroudingConical sheetwith six interruptions:^30 ... 60 % shroudingEngine Operating ParametersSpeed:^ 1000 ... 1400 rpmLoad (BMEP):^ 0.5 ... 4.5 barBeginning of injection (BOI):^16 ... 40°BTDCPulse width of injection (PW): 5 ... 25 °CACNG injection pressure:^ 50 ... 90 barPilot-diesel/total fuel energy ratio:^15 ... 25 %Diesel-fuel cetane number:^-45 or 6260The following is the test procedure for obtaining the gas-diesel engine performance and emission data at a given speed.1. Modify the nozzle shroud to obtain a specific fuel jetinterruption ratio, e.g. 50% shrouding.2. Install a poppet nozzle with specific fuel injection angle,e.g. 100 to the cylinder head surface.3. Set a specific CNG injection pressure, e.g. 50 bar.4. Select a specific pilot-diesel/total fuel energy ratio,e.g. 15%, and keep it constant by adjusting the pilot-dieselmetering valve.5. Set a specific beginning of injection (BOI), e.g. 24°BTDC.6. Increase the load from 0.5 bar to 4.5 bar (or the maximumachievable) and keep speed and pilot-diesel/total fuel energyratio constant.7. Acquire engine performance and emission data for every loadpoint. The operating procedure of the exhaust emissionanalysis system B is presented in Appendix B.8. Repeat for the other BOI, ie. 28, 32, 36 and 40°.9. Repeat for the other pilot-diesel/total fuel energy ratios,ie. 20% and 25%.10. Repeat for the other CNG injection pressures, ie. 60, 70, 80and 90 bar.11. Repeat for the other fuel injection angles, ie. 20° to thecylinder-head surface.12. Repeat for the other fuel jet interruption ratios, ie. 50%,40%, 30% and 0% shrouding.BEST BOI PRERORMANCE CURVE36 DEG.32 DEG.28 DEG.24 DEG.20 DEG32302826242220181614121064261Most of the data were acquired at 1200 rpm, but some werecollected at 1400 rpm.1200 RPM, 60 BAR, 25% CN62, 20 DEG./30%0^1^ 2^3^4BMEP Cbar)---- 801 (DEG. BTDC)Figure 4.2: Determination of the Best BOI Performance Curve.(CNG injection pressure 60 bar, diesel-ratio 25%,CN62, 20° injection angle, 30% shrouding)Figure 4.2 shows typical results of the above test procedureand how the best BOI performance curve is determined as a envelopeof all performance curves for the whole BOI range. This isinteresting because:1.^The best BOI performance curve represents the maximum thermalefficiency that the engine can achieve at any given load.622.^The engine can be computer-controlled to operate along thebest BOI performance curve, as long as the operatingparameters are known.After determining the best BOI performance curve, theoperating conditions were selected for acquiring smoke data andcylinder pressure data. These measurements were made at low, mediumand high load, and best BOI.4.3 The Engine Performance Calculation ProcedureAs mentioned, thermal efficiency, brake mean effectivepressure (BMEP) and brake power are the indices that represent theengine performance, and they are not directly measured. Thevariables that are directly obtained by measurement are enginespeed, torque, cylinder displacement, CNG and pilot-diesel massflow rates. The following is the procedure to calculate the engineperformance indices with measured variables.Brake Power: Brake power Pb (kW) is the power delivered by the engine andabsorbed by the dynamometer (load); it is the product of enginespeed N (rev/min or rpm) and torque Tb (N•m) measured with adynamometer [26].N Tbx10-3^ (4.1)Brake power Pb (kW) can also be expressed as the product ofengine shaft work per cycle W, (kJ) and engine speed N (rpm).63Pb= 60R(4.2)where nR is the number of crank revolutions for each power strokeper cylinder (two for four-stroke cycle, one for two-stroke cycle).Brake Mean Effective Pressure: Brake mean effective pressure BMEP (bar) of the engine isdefined as the engine shaft work per cycle Wc (kJ) divided by thecylinder displacement volume Vd (CM3) [26].BmEP= --sxio4rid(4.3)BMEP can also be expressed in terms of torque Tb (N.m) byusing Eq. (4.2) and (4.1) for a two-stroke engine (nR=1).BMEP-21cTbx10 (4.4)VdBecause pressure, humidity and temperature of the ambient airinducted into a engine affect the air mass flow rate and the poweroutput, a correction factor is usually used to adjust brake powerand BMEP to the standard atmospheric conditions to provide a commonbasis for comparisons between engines. The procedure for computingthe correction factor is listed in Appendix C.Thermal Efficiency: Thermal efficiency is defined as the ratio of the engine brakepower to fuel enthalpy supplied per cycle per second [26].FI,64(4.5)n eh—lh fuel•LEV/3 600where Pb (kW) is the brake power, Arm,/ (kg/hr) is mass flow rateof the fuel, and LHV (kJ/kg) is the Lower Heating Value of thefuel. As there are two fuels inducted into the engine, we have2th fue1.1"=E Oh fuel) i'LliVi1-3.(4.6)where 1=1 for CNG fuel, i=2 for pilot diesel fuel.By substituting Eq. (4.1) and (4.6) into Eq. (4.5), thethermal efficiency of the engine becomes11th— 2 N2NT157l'x10-3 (4.7)g(thfuei) 1LEV1/36 00where N (rpm) is engine speed , Tb (N-m) is torque.4.4 The Engine Exhaust Emission Calculation ProcedureIn the engine tail-pipe exhaust, the main components arecarbon monoxide, carbon dioxide, nitrogen oxides, totalhydrocarbon, oxygen, nitrogen and water vapour. The water vapour inthe exhaust comes from hydrocarbon-fuel combustion and the humidityin intake air. The water vapour resulting from hydrocarbon-fuelcombustion is the main source.Depending on the experimental setup, the exhaust emission data65can be measured either on a wet or a dry basis. The wet-basisemission data are measured with exhaust samples which have the samewater vapour concentration as the tail-pipe exhaust gas (ie. nowater vapour condensed before measurement). The dry-basis emissiondata are measured with exhaust samples from which the water vapourhas been removed before the measurement. Thus the wet-basisemission data are the actual engine exhaust emission data.The EEAS-A system, for example, has a chiller located on theupstream sample line. Thus all emission data from analyzers aredry-basis emission data. With the EEAS-B system, the heated pipeand enclosure maintain the water vapour in the wet exhaust sampledrawn into the total hydrocarbon analyzer so that the unburned THCdata are wet-basis emission data. Another part of the exhaustsample flows through a chiller to remove water vapour beforeentering the other analyzers. Thus the CH„ the NO„ the CO„ the COand 02 data are measured on the dry basis.Usually, the emission measurement is based on volume. Thus theunit of measured emission data is ppm, which is the volumetricratio of 1 volume of measured component to 106 volume of exhaustsample.4.4.1 Conversion between Dry-Basis and Wet-Basis EmissionsThe dry-basis emission data can be converted to the wet-basisemission data by using a conversion factor F„,„ such thatppm (wet-basis) =Fdvxppm(dry-basis)^(4.8)The factor Fth, for gas-diesel operation is expressed as the66product of two factors Kdwi and Kdw2.Fdw=1Cdin •Kdw2^ (4.9)The correction factor Ktha includes the effect of the watervapour produced by hydrocarbon-fuel combustion. The correctionfactor Kdw2 contains the influence of the humidity of the inlet air.An empirical formula for the correction factor Kdwi for purediesel operation is given in the SAE recommendation [41] and isKchn=1-yam(F/A)mm^ (4.10)where yps,=1 . 8 [26] is the atomic hydrogen to carbon ratio of thediesel fuel; (F/A) DSL is the diesel fuel-air ratio (dry basis).The formula for the correction factor Ktha for gas-dieseloperation is obtained by modifying Eq. (4.10):Kohn =1 - [you (F/A) cw+YDsz, ( Fi A ) Dad =3. X1120^(4.11)where v.. CNG= 3.85 is the atomic hydrogen to carbon ratio of the B.C.CNG fuel; v.. DSL:= 1.8 is the atomic hydrogen to carbon ratio of thepilot diesel fuel; (F/A) CNG is the CNG fuel-air ratio (dry basis)and (F/A) DSL is the diesel fuel-air ratio (dry basis). Both (F/A) CNCand (F/A) DSL can be calculated from measured data of CNG mass flowrate, diesel mass flow rate and air mass flow rate. X is themolal fraction of the water vapour in the exhaust.The relation between dry-basis fuel-air ratio (F/A)dry and wet-basis fuel-air ratio (F/A)„,, is:(FIA) vet= (1-H/1000) (F/A) dry^(4.12)where H is the specific humidity (g of H20 per kg of dry air),67referring to Appendix D and [40] for calculation of the specifichumidity.Because the humidity of the inlet air has an effect on theamount of NO chemically formed in combustion, correction factor Kdw,should be applied to NO emission at each test point. According tothe SAE recommendation [41], for NO emission data the followingformula is applied: 1 (4.13)1+7B(H-10.714) +1.8•C(T-29.444)where B=0.044(F/A)-0.0038 and C=-0.116(F/A)+0.0053; F/A is fuel-airratio (dry basis); H is specific humidity (g of 1120 per kg of dryair) (see Appendix D); T is intake air temperature (°C).For the other emission data, 1<dw2 = 1.Amore exact formula for dry-to-wet basis conversion factor Fdwof the gas-diesel operation is presented in Appendix E. Thedifferences between the above method and the more exact method are:0.05% at BMEP - 0.5 bar, 0.7% at BMEP - 2 bar and 0.2% at BMEP - 4bar.4.4.2 Determination of Brake Specific EmissionsThe concentrations of gaseous emissions in the engine exhaustare usually measured in parts per million or percent by volume. Butanother more comparable emission indicator, brake specific emissionbse, is also used. Brake specific emission (bse) of a component isdefined as the mass flow rate of the component Ale per unit ofpower output Pb.bse= •Pb(4.14)68The brake specific emissions are calculated based on thefollowing assumptions:1. The engine exhaust consists of CO, CO„ NO„ H20, 02, N2 andunburned HC.2. A combined fuel (CH) is used to replace CNG (CH3.85) and diesel(CH18) fuels. The hydrogen-to-carbon atom ratio y of thecombined fuel is determined by Eq. (E.2) in Appendix E.3. The unburned HC has the same composition as the combined fuel.4. All measured emission data have already been converted to awet basis in ppm (or % by volume).Exhaust gas mass flow rate Aiuch is determined by summing theintake air mass flow rate k1 and the total fuel mass flow rateE /firmathexh_Ihair 9 E Ihfuel =itiair +1k:1W +IhDSL^ (4.15)Molal fraction of a component X1 can be evaluated asfollowing:io469(4.16)x1 = [ (ppm) i] where [(ppm)„], is concentration of a component "i" on the wetbasis. Xco, X002, XHc, Xo, and X02 can be determined by Eq. (4.16). X,20can be evaluated by Eq. (4.11) (or Eq. (E.9) in Appendix E). X,2 canbe calculated as follows:Xia=1-Xco-X032-Xac-Xmox-420-42^ (4.17)Mass flow rate of a component "i" can be determined by.X11 E ximwi m-mrh (4.18)where MW, is the molecular weight of a component "i". The molecularweight of CO, CO2, NO, NO2, H20 and 02 are known.According to assumptions #2 and #3 of this subsection, themolecular weight of unburned HC can be determined as follows:MEgmc=fDsz,MWDsz,+faveMWan^ (4.19)where DsL is the molal fraction of pilot diesel (CH1.8); MW„ =13.825 kg/kmol is the molecular weight of the pilot diesel; f CNGthe molal fraction of the compressed B.C. natural gas; MlacNG16.689 kg/kmol is the molecular weight of the compressed B.C.natural gas (see Table 3.4 in Chapter 3).Based on assumption #2 and #3, it is known that the unburnedTHC is the combination of 1 mole of B.C. CNG and (16.689/13.825)rmmoles of pilot diesel, so the molal fraction of the pilot dieselis:MW cm^ rMW InDSLfum- MP/1+__!212rWam I'(4.20)70where rm is the diesel to CNG mass ratio and can be determined byEq. (E.1) in Appendix E.The molal fraction of the B.C. compressed natural gas is:fayrd=1  fAgt.^ (4.21)Thereby, the brake specific emissions of the components can bedetermined by Eq. (4.14).715 . EXPERIMENTAL RESULTS5.1 IntroductionThe purpose of the chapter is to present and discuss theexperimental results. Primarily, diesel baseline and gas-dieseloperation tests have been completed and the results are presentedand summarized in the following sections. The repeatability of theresults is discussed in Appendix F.The diesel baseline test involves three diesel fuels. The gas-diesel operation test contains the injection geometrical parametersand the engine operating parameters. A comparison of optimum gas-diesel operation against the diesel baseline is presented as thesummation.5.2 Diesel Baseline Test resultsCommercial grade 1 (DF1), grade 2 (DF2), and cetane number 62diesel (CN62, an experimental high-cetane fuel) were tested to findthe effect of fuel composition on conventional diesel engineperformance and emissions. The cetane numbers of DF1 and DF2 areapproximately 45 and the other properties of these fuels are listedin Table 3.2 of Subsection 3.2.3. In the computation of thermalefficiency, the lower heating values of the three fuels were takento be identical. The differences in heating values is estimated tobe within 0.3%.72All tests were conducted at 1200 rpm with the samepreprogrammed BOI, although the BOI was not optimizedfor CN62.1200 RPM 323028262422201816141210a6420^1^2^3^4^5BMEP (Bar)CN62^+ DF 1^DF2Figure 5.1: Effect of Fuel Composition on Performance.Figure 5.1 shows the effect of fuel composition on thermalefficiency. As shown, the thermal efficiencies of the three fuelsare about the same at low load (BMEP 1 bar). As load increases,the thermal efficiency with CN62 fuel becomes highest, and thethermal efficiency with DF1 fuel is the lowest. The peak thermal121 . for CN62, DF1 and DF2 fuels are 29.0%, 28.4% and 28.6%respectively. The difference in peak thermal efficiency betweenCN62 and DF1 fuels is about 2%. The maximum achievable load withall three fuels is about 4.5 bar.1200 RPMa^1^ 2^3^4^ 5BMEP (Bar)0 CN62^+ DF1^o 3F2Figure 5.2: Effect of Fuel Composition on Nitrogen Oxides.Figure 5.2 through Figure 5.5 show the effect of fuelcomposition on exhaust emissions. Generally, CN62 fuel producesless nitrogen oxides (NO,), total hydrocarbon (THC), non-methanehydrocarbon (NMHC) and carbon monoxide (CO) emissions than DF1 fuel74over most of the load range. The emissions of DF2 fuel are betweenthose of CN62 and DF1 over most of the load range.Figure 5.3: Effect of Fuel Composition on Total Hydrocarbon.The difference between total hydrocarbon and non-methanehydrocarbon is the methane emission in the exhaust. The absolutevalue of methane emission is very low for diesel operation, beingin the range of 10 - 35 ppm or 6% - 17% of total hydrocarbonemission.0 I 2 3 4 $1200 RPM19018017016015014013081EP C Bar)O CN62^+ DF1^0 DF2Figure 5.4: Effect of Fuel Composition on Non-MethaneHydrocarbon.The performance and emission curves for 100% CN62 fuel werechosen to represent the engine baseline used in subsequentcomparison with gas-diesel operation in which CN62 is the pilot-fuel in most tests. The emission curves of engine baseline was onwet basis.75-----0 1 2 3 4 51200 RPM700SO 0500400300200103oBMEP CBar)0 CN62^+ DF 1^0 3F2Figure 5.5: Effect of Fuel Composition on Carbon Monoxide.5.3 Effect of Injector Geometrical ParametersThe effects of injector geometrical parameters on gas-dieselengine performance and emissions were investigated and the resultsare presented in this section. The geometrical parameters that werevaried were fuel jet interruption ratio and fuel injection angle.The emission results presented in this section were on dry-basis.,76----_-77Fuel Jet Interruption Ratio: Tests were conducted with different values of fuel jetinterruption ratio (ie. percentage shrouding or % SRD) at thefollowing test condition: 1200 rpm speed, 60 bar CNG injectionpressure, 25% CN62 pilot-diesel energy ratio, and 200 fuelinjection angle.1200 RPM, 60 BAR, 25% CETANE62, 20 DEG.323026249^22Li201816u_LU^1412106422^3^4^5BMEP (Bar)---- BASELINE^+ 0% SAD^* 6096 SRD^A 40% SRO^x 50% SADFigure 5.6: Effect of Fuel Jet Interruption Ratioon Performance.Figure 5.6 shows the effect of fuel jet interruption ratio (%SRD) on thermal efficiency. As exhibited, there is an optimum fuel78jet interruption ratio (- 40% SRD for this engine) which providesbest thermal efficiency at low and medium load, as well asrelatively high thermal efficiency at high load; the maximum loadcapability increases with the increment of the fuel jetinterruption ratio; the thermal efficiency of all cases is lowerthan that of the baseline at low load, while the peak thermalefficiency of the 40% SRD, 30% SRD and 0% SRD cases exceeds that ofthe baseline. The reason for this is that fuel jet interruptionincreases jet penetration, injection speed and jet stability [42](see Appendix G). The increased jet penetration and higherinjection speed of the 50% SRD case may disperse the fuel jet toomuch at low and medium load, thereby reducing thermal efficiency.At high load, the greater jet penetration and higher injectionspeed helps the fuel to mix with air, so that it improves thermalefficiency and maximum load capability. The shorter penetration ofthe 0% SRD case (ie. conical sheet without interruption) helpsthermal efficiency at low and medium load, but suffers fromreduction of the maximum load capability. The peak thermalefficiency of the 0% SRD case (around BMEP - 3.4 bar) also shiftsto lower load compared to that of the 50% SRD case (around BMEP -4 bar). This indicates that there should be an optimum jetinterruption ratio case between the 50% SRD and the 0% SRD cases.The effects of fuel jet interruption ratio on totalhydrocarbon (THC) and nitrogen oxides (NO,) are shown in Figures5.7 and 5.8, respectively. Higher THC and lower NO are observedfor the 50% SRD case at low and medium load because of overFigure 5.7: Effect of Fuel Jet Interruption Ratio on TotalHydrocarbon.dispersion; the THC emission of both the 40% SRD and the 30% SRDcases are very close, but the NO emission of the 40% SRD case islower than that the of the 30% SRD case; the NO emissions of alljet interruption cases are lower, however, than that of the dieselbaseline except at high load; the THC emission in all cases alsosurpasses that of the baseline, especially at low loads. All theseindicate that the low-load combustion quality needs improvement.Usually, high thermal efficiency is associated with highcombustion temperature, high NO formation rate and high NO791200 RPM, BO BAR, 25% CETANE82, 20 DEG80^2^3^4^5BMEP (Oar)---- BASELINE^+ 0% SAD^30% SRO^A 40% SAD^X 50% SADFigure 5.8: Effect of Fuel Jet Interruption Ratio on NitrogenOxides.emissions. But NO emission does not depend on combustiontemperature only, it also depends on combustion duration time andair concentration in the cylinder. If the air concentration isconstant and the fuel combustion rate could be increased (ie.reduce combustion duration) in some way, we could still obtainhigher thermal efficiency with lower NO emission. The 40% SRD casemight be an example of this condition - for which highest thermalefficiency in the gas-diesel operation is not associated withhighest NO,. RPM, 60 BAR, 25% CETANE62, 20 DEG.812^3^4^5BMEP (Bar)---- BASELINE^+ 0% SAD^o 30% SRO^A 40% SRO^X 50% SADFigure 5.9: Effect of Fuel Jet Interruption Ratio on CarbonMonoxide.It can also be seen in Figure 5.9 that the carbon monoxideemissions in all cases are much higher than that of the dieselbaseline (average about 10 times higher). This suggests that thecombustion in all cases took place in locally rich regions.Fuel Injection Angle: 1200 RPM, 60 BAR, 25% CETANE62, 5111% SRD82 3230282624222018161412106420^2^3^4^5BIMEP (Bar)BASELINE^+ 10 DEG. ANGLE^20 DEG. ANGLEFigure 5.10: Effect of Fuel Injection Angle on Performancewith 50% Shrouding.Tests were undertaken using two injection angles with four jetinterruption cases. Test condition was 1200 rpm speed, 60 bar CNGinjection pressure, and 25% CN62 pilot-diesel energy ratio. In thisdiscussion, two typical jet interruption cases (50% SRD and 30%SRD) were selected as representative of all situations.20 DEG.32302826249^222019016U-141210421200 RPM, 60 BAR, 25% CETANE62, 30% SRD830^-1^2^3^4^5BMEP (Bar)---- BASELINE^+ 10 DEG. ANGLE^20 DEG. ANGLEFigure 5.11: Effect of Fuel Injection Angle on Performancewith 30% Shrouding.Figures 5.10 and 5.11 show the effects of fuel injection angleon thermal efficiency with the 50% and the 30% SRD respectively. Inthe 50% SRD case, operation with the 100 angle provides betterthermal efficiency at low and medium load, but much lower peakthermal efficiency and lower maximum load capability than operationwith the 20° angle; and vice versa. In the 30% SRD case, the 100angle loses its advantage at low load, but retains itsBASELINE1200 RPM, 60 BAR, 25% CETANE62, 30% SRD84disadvantages at high load; the 200 angle becomes dominant over thewhole load range.32.82.8242.22a^an^1.6Li [IIW C^6me— th^X =^4zc^Li ^1.210.8080.40.200^2^3^4^5BMEP (Ber)---- BASELINE^+ 10 DEG. ANGLE^20 DEG. ANGLEFigure 5.12: Effect of Fuel Injection Angle on Carbon Monoxide.The reason for that can be explained with the flowvisualization results [42] (see Appendix G) and carbon monoxideemission results in Figure 5.12. The fuel jets with the 10° angletend to cling to the top wall (ie. top wall effect), so that thejet penetration has been reduced, but the 20° angle jets do not.85In the 50% SRD case, large jet interruption increases thepenetration of the 200 angle jets more than that of the 10° anglejets, so that the 10° angle jets have less dissipation and burnwith locally richer mixtures than the 20° angle jets do. Therefore,operation with the 10° angle provides better thermal efficiency atlow and medium load. However, longer penetration of the 20° anglejets still improves peak thermal efficiency and maximum loadcapability at high load.The jet penetration is believed to have been reduced when thejet interruption ratio decreased to 30% SRD. Thus the 10° and the20° angle jets have similar behaviour at low load. But the top walleffect at 10° angle still influences the peak thermal efficiencyand maximum load capability at high load. The differences ofthermal efficiency and maximum load capability between the 10° andthe 20° angle jets, however, have been reduced in the 30% SRD case,which indicates that reducing the top wall effect is associatedwith decreasing the jet interruption ratio.The effect of fuel injection angle on carbon monoxide is shownin Figure 5.12. It is observed that operation with the 10° angleproduces more carbon monoxide emission than with the 20° angle,which suggests that the top wall effect of the 10° angle may havecaused locally rich-mixture burning and top wall quenching.1200 RPM, 60 BAR, 25% CBTANG62, 301% SRD86 1.71.8151. . 80 . 50 . 40 . 30.20 . 10^2^3^4^5BMEP C Bar )- BASEL I NE^+ 10 DEG. ANGLE 0 20 DEG. ANGLEFigure 5.13: Effect of Fuel Injection Angle on Nitrogen Oxides.It is also seen in Figure 5.13 that operation with the 200angle produces almost the same amount of nitrogen oxides emissionsas with the 100 angle at low load, but more at medium and highload. These trends match the thermal efficiency trends in Figure5.11.Figure 5.14 shows the effect of fuel injection angle onunburned methane emission. Unburned methane emission comes mainlyfrom CNG fuel, and it is a measurement of the amount of CNG fuelsurviving combustion. It is observed that operation with the 10010 DEG.20 DEG.(BASELINE871200 RPM, 60 BAR, 25% CETANE62, 30% SRD4.543.53E-8acamy9oz .c2.52<1.510.50^1^2^3^4^5BMEP (Bar)---- BASELINE^+ 10 DEG. ANGLE^0 20 DEG. ANGLEFigure 5.14: Effect of Fuel Injection Angle on Methane.angle produces almost the same amount of unburned methane emissionas with the 200 angle at low load (BMEP - 0.8 bar), 50% moreunburned methane emission than with the 20° angle at medium load(BMEP - 2.5 bar) and 100% more unburned methane emission than withthe 20° angle at high load (BMEP - 3.5 bar). These trends alsomatch the thermal efficiency trends in Figure 5.11.The effect of the fuel injection angle on non-methanehydrocarbon emission is shown in Figure 5.15. Non-methanehydrocarbon emission primarily results from incomplete combustionof pilot-diesel fuel and lubrication oil (the latter probably being1200 RPM, 60 BAR, 25% CETANE62, 30% SRD88 3232. BASELINE0^2^3^4^5°MEP CBar)---- BASELINE^+ 10 DEG. ANGLE^20 DEG. ANGLEFigure 5.15: Effect of Fuel Injection Angle on Non-MethaneHydrocarbon.negligible). It is thus a measurement of the amount of pilot-dieselfuel surviving combustion. As shown, operation with the 100 angleproduces more non-methane hydrocarbon emission than with the 20°angle over the whole load range, which suggests that the pilot-diesel was not atomized well or did not mix sufficiently with air(ie. burned too rich), or that top wall quenching was significant,or that the gas-air mixture was too lean because of very longignition delay. This agrees with the carbon monoxide trends inFigure 5.12. The total hydrocarbon emission is the sum of unburned89methane and non-methane hydrocarbon emissions.5.4 Effect of Engine Operating ParametersIn this section, the effects of engine operating parameters ongas-diesel engine performance and emissions are presented. Theorder of presentation is pilot-diesel cetane number, pilot-diesel/total fuel energy ratio (diesel ratio), CNG injectionpressure and engine speed. The emission results presented here aredry-basis.Pilot-Diesel Cetane Number: Two different pilot-diesel fuels with different cetane numberswere tested. They were cetane number 62 diesel (CN62) andcommercial grade 2 diesel (DF2). The cetane number of DF2 isapproximate 45. Test condition was 1200 rpm speed, 60 bar CNGinjection pressure, 25% of pilot-diesel energy ratio, 200 fuelinjection angle, and 0% SRD.Usually, diesel engines perform better (ie. obtain higherthermal efficiency) when the fuel burns evenly. Combustion shouldbegin as soon as possible after injection instead of lagging untilpressure, temperature, and the accumulation of fuel build up to apoint conducive to detonation. For a given engine at a given speedand load, the ignition delay time depends on the ignition qualityof the fuel. Diesel cetane number represents the ignition qualityof the fuel. The higher the cetane number of a diesel fuel, thebetter its ignition quality.DF2 PILOTCN62 PILOT323028262422201816141210a421200 RPM, 60 BAR, 20 D, ANGLE, ax SAD0^2^3^4^5BMEP (Bar)---- BASEL I NE^+ 25% CN62 PILOT ^ 25% D F2 P I LOTFigure 5.16: Effect of Pilot-Diesel Cetane Numberon Performance.Figure 5.16 shows the effect of pilot-diesel cetane number onthermal efficiency. As shown, operation with the CN62 pilot-fuelprovides better thermal efficiency in the load range of BMEP from0.4 - 3.5 bar; the thermal efficiency with the DF2 pilot-fuelexceeds that with the CN62 pilot-fuel from BMEP equalling 3.5 bar;there are almost identical maximum load capabilities for operationwith both pilot-fuels.90___-__--^(i BASEL II NE4.543.532.521.5I0.501200 RPM, 60 BAR, 20 DEG. ANGLE, 0% SRD0^1^2^3^4^5BMEP ( Bar)---- BASELINE^+ 25% CN62 PILOT^0 25% DF2 PILOTFigure 5.17: Effect of Pilot-Diesel Cetane Numberon Methane.It is believed that the ignition delay time is relatively longwhen the engine operates at low and medium load. The good ignitionquality of the CN62 pilot-fuel reduces the ignition delay time andthe combustion delay time of fuel, thereby the fuel combustion isimproved and the thermal efficiency is high. This is consistentwith the lower methane and non-methane hydrocarbon emissions shownin Figures 5.17 and 5.18, as well as the higher nitrogen oxides91BASELINE92emissions shown in Figure 5.19 for loads less than 3.5 bar. Atloads greater than 3.5 bar, the lower thermal efficiency with theCN62 pilot-fuel may be caused by large amounts of fuel/air mixtureburned before TDC.1200 RPM, GO BAR, 20 DEC. ANGLE, 0% SRD3.^2.2orZ2Ci1.62x^141.21Ul0.012^0.6OA0.200^ 2^3^4^5BMEP (Bar)---- BASELINE^+ 25% CN62 PILOT^25% 0F2 PILOTFigure 5.18: Effect of Pilot-Diesel Cetane Numberon Non-Methane Hydrocarbon.21.91.8171.61.51.4131. RPM, 60 BAR, 20 DEG. ANGLE, 0% SRD 1^0^2^3^4^5BMEP (Bar)-BASELINE^+ 25% CN62 PILOT^25% DF2 PILOTFigure 5.19: Effect of Pilot-Diesel Cetane Numberon Nitrogen Oxides.Pilot-Diesel/Total Fuel Energy Ratio (Diesel Ratio): Tests were performed with three different diesel (DSL) ratios25%, 20% and 15% in almost all jet interruption cases. Testcondition was 1200 rpm speed, 60 bar CNG injection pressure, CN62pilot-diesel fuel, and 200 fuel injection angle.932019181716LI^1514(-3^13u_u_^1211109a51200 RPM, 20 DEG. ANGLE, BMEP = 1 BAR94 50^ 40^ 30^ 0FUEL JET INTERRUPTION RATIO (% SRD)---- BASELINE^4- 25% DSL RATIO^0 20% DSL RATIO^4 15% DSL RATIOFigure 5.20: Effect of Diesel Ratio on Thermal Efficiencyat Low Load (BMEP = 1 BAR).Figures 5.20 and 5.21 show the effects of diesel ratio onthermal efficiency and total hydrocarbon at low load (BMEP = 1 bar)respectively. The figures appear to indicate that poor low-loadcombustion quality was one of the serious problems with thisinjector configuration for gas-diesel operation in this engine. Asillustrated, operation with high diesel ratio (ie. 25% DSL ratio)provides high thermal efficiency and low total hydrocarbon at lowload through all jet interruption cases. The tests with the 15% DSLBASELINE1^95ratio with 0% SRD and 30% SRD were not done, because of excessivemisfiring, especially at low load.1200 RPM, 20 DEG. ANGLE, BMEP = 1 BAR50^ 40^ 30^ 0FUEL JET INTERRUPTION RATIO CAS SRD)- BASELINE^+ 25% DSL RATIO^o 20% DSL RATIO^A 15% DSL RATIOFigure 5.21: Effect of Diesel Ratio on Total Hydrocarbonat Low Load (BMEP = 1 BAR).As discussed in Section 2.2, CNG fuel will not self-ignite inthe typical diesel engine cylinder environment because of the highself-ignition temperature (- 1200 K [5]) of CNG fuel. A minimumamount of pilot-diesel injected into the cylinder is necessary toheat the CNG fuel to its auto-ignition temperature. The amount ofpilot-fuel depends on intake state, compression ratio, distribution4NE107543201200 RPM, 60 BAR, 20%CN62, 1003E/50%6RD96 2 62 42 221 81141210040.20^ 2^ 3^ 481MEP C6arp0 LIG DSL dm/dt^+ CNG &nickFigure 5.22: Relationship between Mass Flow Rate of the Fuelsand Load.of pilot-diesel and CNG fuels and load of a given engine. Generallyspeaking, the quantity of pilot-diesel that is necessary to heatthe CNG fuel to its auto-ignition temperature could be expected tobe approximately constant over the whole load range, thus thediesel ratio would become high at low load and reduced withincreased load (when more CNG fuel is injected). Figure 5.22 showsthe relationship between mass flow rate of the fuels (CNG andpilot-diesel) and load (BMEP). The test case was 1200 rpm speed, 60bar CNG injection pressure, 25% CN62 pilot-diesel energy ratio, 20097fuel injection angle and 50% jet interruption ratio. As shown,diesel mass flow rate remains almost constant at low and mediumload, and increases slightly at high load.1200 RPM, 20 D. ANGLE50^ 40^30FUEL JET INTERRUPTION RATIO (% SRD)BASELINE 1M 25% DSL RATIO 2E0 20% DSL RATIO r22:1 15% DSL RATIOFigure 5.23: Effect of Diesel Ratio on Maximum ThermalEfficiency.As shown in Figures 5.23 and 5.24 respectively, the effects ofdiesel ratio on peak thermal efficiency and maximum load capabilityare associated with the jet interruption ratio (ie. the penetrationand distribution of the fuel). For the large jet interruptionratio, low diesel ratio (15% DSL ratio) improves peak thermalefficiency and maximum load capability; High diesel ratio (25% DSL5. 1200 RPM, 20 DEG ANGLE9850^ 40^ 30^ 0FUEL JET INTERRUPTION RATIO (6 SRD)---- BASELINE^4- 25% DSL RATIO^o 20% DSL RATIO^b. 15% DSL RATIOFigure 5.24: Effect of Diesel Ratio on Maximum LoadCapability.ratio) dominates peak thermal efficiency and maximum loadcapability at small jet interruption ratio (from 30% SRD).one can also see in Figure 5.25 that higher diesel ratioproduces higher nitrogen oxides emissions at high load. FromFigures 5.20 and 5.21, we also see that high diesel ratio produceshigh nitrogen oxides emissions at low load.Figure 5.25: Effect of Diesel Ratio on Nitrogen Oxidesat High Load (BMEP = 3.8 BAR).CNG Iniection Pressure: Tests were conducted with different CNG injection pressuresand different injection angle. Test condition was 1200 rpm speed,60 bar CNG injection pressure, and 25% CN62 pilot-diesel energyratio.99BASELINE^ 70 BAR80 BAR50 BAR32302826242220181614121064201200 RPM, 25% CETANE62, 10 DEG,/50% SAD1000^-1^2^3^4^5BMEP (Bar)- BASELINE + 50 BAR o 60 BAR A 70 BARFigure 5.26: Effect of CNG Injection Pressure on Performancewith 10 DEG. Injection Angle.Figure 5.26 and 5.27 show the effects of CNG injectionpressures on thermal efficiency with 100 and 200 injection anglesrespectively. Operation with high CNG injection pressure improvespeak thermal efficiency and maximum load capability; The maximumachievable CNG injection pressure with 20° injection angle (80 bar)is higher than that with 10° injection angle (70 bar); Similar tothe effect of jet interruption ratio, there is an optimum CNGinjection pressure for both injection angles. As shown in Figure5.26, the optimum pressure for 10° injection angle is 60 bar, which1200 RPM, 25% CETANE62, 20 DEG./40% SRD101 3230282624222010161412104200^2^3^4^5BMEP (Bar)BASELINE^+ 60 BAR^BC BARFigure 5.27: Effect of CNG Injection Pressure on Performancewith 20 DEG. Injection Angle.provides good thermal efficiency at the load range of BMEP before2 bar and relatively good thermal efficiency at medium and highload. In Figure 5.27, 60 bar appears to be the optimum pressure forthe 200 injection angle. If there were a curve for 50 bar in Figure5.27, it could expected to be similar to the 100 injection anglecurve in Figure 5.26.The explanation of the effect of CNG injection pressure onperformance is also similar to that of jet interruption ratio. Thatis, higher CNG injection pressure increases the fuel jet75432102penetration and injection speed, so that it improves the high-loadthermal efficiency and maximum load capability but deteriorates thelow-load and medium-load thermal efficiency, and vice versa.1200 RPM, 25% CETANE62, 20 DEG./40% SRD0^2^3^4^5BMEP CBar)BASELINE^+ 60 BAR^BO BARFigure 5.28: Effect of CNG Injection Pressure on Methane.Considering the requirements of fuel jet penetration, it isreasonable to expect an optimum CNG injection pressure for a givenengine which matches the internal cylinder pressure trace (ordistribution) and provides appropriate thermal efficiency over thewhole load range. The internal cylinder pressure trace (withoutcombustion) for a given engine is fixed, depending only on intakeBASELINE103state (ie. natural aspiration or turbocharging) and compressionratio of the given engine.1200 RPM, 25% CETANE62, 20 DEG./40% SAD21.91.81? 13gi^^ 0.40 30.20.100^2^3^4^5BMEP (Bar)BASELINE^+ 60 BAR^0 BO BARFigure 5.29: Effect of CNG Injection Pressure on Non-MethaneHydrocarbon.The effects of CNG injection pressure on unburned methane andnon-methane hydrocarbon emissions (with 20° injection angle) areshown in Figures 5.28 and 5.29 respectively. It is observed that 80bar CNG injection pressure is too high for operating at low andmedium load. High injection pressure disperses the fuel jet andcauses higher methane and non-methane hydrocarbon emissions.1200 RPM, 25% CETANE62, 20 DEG /40% SRO104 1.41.3121.^ 2^3^4^5BMEP (Bar)- BASEL I NE + 60 BAR 0 80 BARFigure 5.30: Effect of CNG Injection Pressure on NitrogenOxides.Figure 5.30 shows that both pressure cases produce almost thesame nitrogen oxides emissions at low load; high CNG pressureproduces less nitrogen oxides at medium and high load.The effect of CNG injection pressure on best operating BOI isshown in Figure 5.31. It is interesting to see that, with higherCNG injection pressure, the engine is able to run with smaller BOI.1200 RPM, 25% CETANE62, 20 DEG./40% SRD 40383634323028262422201818141210805^15^25^35^45amEP (Bar)0 60 BAR^+ BO BARFigure 5.31: Effect of CNG Injection Pressure on BOI.Engine Speed: Tests were conducted with engine speeds of 1200 rpm and 1400rpm. Test condition was 60 bar CNG injection pressure, 25% CN62pilot-diesel energy ratio, 200 fuel injection angle, and 30% SRD.Figure 5.32 shows the effect of the engine speed on thermalefficiency. As can be seen, operation with 1400 rpm providesslightly better thermal efficiency than with 1200 rpm at low load;however, much better thermal efficiency is obtained by operatingwith 1200 rpm at medium and high load.1051200 RPM60 BAR, 2596 CETANE62, 20 DEG / 30% SAD106323028262422201816141210621^2^3^4^5BMEP (Bar)-BASELINE  + 1200 RPM^0 1400 RPMFigure 5.32: Effect of Engine Speed on Performance.Usually, thermal efficiency is affected by engine speedthrough load and heat transfer. At low load, operating at 1400 rpmmay have less heat loss (per unit time) than at 1200 rpm, therebyincreasing thermal efficiency slightly. At medium and high load,operating at 1400 rpm the combustion duration (counted by degreecrank angles) is longer than at 1200 rpm. More heat loss (per unitarea) and power loss, resulting from longer combustion duration,reduce the thermal efficiency at 1400 rpm.1200 RPM1.71.81 ,-'W^0(AC13 C .9W0 ID0.08g07z§^ BAR, 25% CETANE62, 20 DEG./30% SAD1072^3^4^5BMEP (Bar)-BASELINE + 1200 RPM^1400 RPMFigure 5.33: Effect of Engine Speed on Nitrogen Oxides.The effect of engine speed on nitrogen oxides emissions isshown in Figure 5.33. As can be seen from this figure, at 1400 rpmand low load slightly higher nitrogen oxides emissions areassociated with slightly higher thermal efficiency. The largedifference in nitrogen oxides emissions between 1200 rpm and 1400rpm at medium and high load is also consistent with the thermalefficiency difference.4.543.5321.50.51200 RPM60 BAR, 25% CETANE62, 20 DEG./30% SAD1081^2^3^4^5IBMEP (ear)---- BASELINE^+ 1200 RPM^1400 RPMFigure 5.34: Effect of Engine Speed on Methane.Figure 5.34 shows the effect of engine speed on unburnedmethane emissions. As shown, operation at 1400 rpm obviouslyreduces unburned methane emission at low load; with increasingload, the unburned methane emission at 1400 rpm rises at mediumload and eventually exceeds that of 1200 rpm at high load; thepoint at which unburned methane emission of 1400 rpm begins to risecorresponds to the point at where the thermal efficiency at thisspeed starts to suffer.2.62 RPM60 BAR, 25% CETANE62, 20 DEG./30% SRD1090^2^3^4^5BMEP (Bar)BASELINE^+ 1200 RPM^1400 RPMFigure 5.35: Effect of Engine Speed on Non-Methane Hydrocarbon.Figure 5.35 shows that non-methane hydrocarbon emission at1400 rpm and low load is much lower than that at 1200 rpm. The non-methane hydrocarbon emission of 1400 rpm ascends at the pointcorresponding to where the thermal efficiency of 1400 rpm starts tosuffer and finally catches up with that of 1200 rpm at high load.These indicate that less heat transfer at higher speed and low loadnot only improves CNG combustion but also help pilot-dieselburning.OPTIMLN DATA1105.5 Optimum Gas-Diesel Operation ConditionSummarizing all the effects of the injector geometricalparameters and the engine operating parameters on performance andemissions of the gas-diesel engine, an optimum operation conditioncan be selected based on the best engine performance (ie. thermalefficiency). The injector geometrical parameters and the engineoperating parameters of the optimum operation condition at 1200 rpmengine speed are as follows:1200 RPM, 60 BAR, 20 DEG./40% SAD323028262422201816141210420^2^3^4^5BMEP C6ar)---- BASELINE^+ OPTIMUM DATAFigure 5.36: Optimum Performance of Gas-Diesel Operation.1111. 40% of jet interruption ratio with 200 injection angle.2. 60 bar CNG injection pressure.3. 25% cetane 62 pilot-diesel energy ratio.The plots presented in this section are performance andemissions of the optimum gas-diesel operation condition againstthose of CN62 baseline. The dry-basis and wet-basis emissions ofthe same component are plotted together to give a clear idea.Generally, the dry-basis emission of a given component is higherthan the wet-basis emission of this component (The differencebetween them is due to the water vapour content in exhaust whichincreases with load).Figure 5.36 shows the optimum thermal efficiency of the gas-diesel operation. Compared with the baseline, the optimum thermalefficiency of the gas-diesel operation is about 15% lower at BMEP- 1 bar, identical at BMEP - 2.5 bar and higher 3% at BMEP - 3.5bar.The nitrogen oxides emissions of the optimum gas-dieseloperation are shown in Figure 5.37. As can be seen, the differencebetween dry-basis and wet-basis nitrogen oxides emissions of theoptimum gas-diesel operation is increased with load. This isbecause the amount of fuel injected into the cylinder is increasedwith load, so is the water vapour produced in the exhaust. Also,the wet-basis nitrogen oxides emissions of the optimum gas-dieseloperation are under the baseline, although the dry-basis data areslightly above baseline at high load.WET-BASIS DATABASELINEDRY-BASIS DATA1 21.10.9(-10.0aa(.0 0. 7wuo c— 00.65Ll0.3020.11200 RPM, 60 BAR, 20 DEG./40% SRD1120^2^3^4^5BMEP (Bar)— BASELINE^+ WET-BASIS DATA^* DRY-BASIS DATAFigure 5.37: Nitrogen Oxides of Optimum Gas-Diesel Operation.Table 5.1: List of Unburned CNG and Pilot-Diesel Ratio.BMEP (bar)^ 1.0^2.6^4.0Unburned CNG ratio:^ 0.2358^0.0555^0.0087Unburned pilot-diesel ratio:^0.2777^0.0610^0.0844WET-BASIS DATADRY-BASIS DATABASELINE1200 RPM, 60 BAR, 20 DEG/40% SAD11340^2^3^4^5BMEP (Bar)- BASELINE^+ WET-BASIS DATA^DRY-BASIS DATAFigure 5.38: Methane of Optimum Gas-Diesel Operation.The methane and non-methane hydrocarbon emissions of optimumgas-diesel operation are shown in Figure 5.38 and 5.39. Table 5.1lists the unburned fuel ratio at three specific loads. The unburnedfuel (CNG or pilot-diesel) ratio is defined as a mass ratio of theunburned fuel (CNG or pilot-diesel) in the exhaust to the injectedfuel (CNG or pilot-diesel). As Table 5.1 shows, A quite largepercent of pilot-diesel survives combustion. The ratio of theunburned pilot-diesel is higher than the ratio of the unburned CNGover the whole load range. The worst situation is at low load. ThisDRY-BASIS DATAWET-BASIS DATABASELINE21.91.8171.61.51.4131. RPM, 60 BAR, 20 DEG./410% SRD1140^ 2^3^4^5BMEP C8ar]- BASELINE^+ WET-BASIS DATA^0 DRY-BASIS DATAFigure 5.39: Non-Methane Hydrocarbon of Optimum Gas-DieselOperation.suggests that the pilot-diesel was not atomized well or did not mixsufficiently with air (ie. burned too rich), or that top wallquenching was significant, or that the gas-air mixture was too leanbecause of very long ignition delay. The low-load thermalefficiency could be improved if the combustion of pilot-diesel wasimproved. The lower diesel ratio operation might be achieved if thepilot-diesel was burned completely.3.50.5251.5DRY-BASIS DATAWET-BASIS DATABASELINE1200 RPM, 60 BAR, 20 DEG./40% SRD1150^1^2^3^4^5BMEP (Bar)- BASELINE^-I- WET-BASIS DATA^0 DRY-BASIS DATAFigure 5.40: Carbon Monoxide of Optimum Gas-Diesel Operation.The carbon monoxide and the carbon dioxide emissions of theoptimum gas-diesel operation are shown in Figures 5.40 and 5.41.High carbon monoxide and low carbon dioxide emissions indicate thatthe burned fuel/air mixture is still locally rich, although someunburned mixture is already too lean to burn. Wall quenching can beanother reason.-BASIS  DATAWET-BASIS DATABASEL I NE654321200 RPM, 60 BAR, 20 DEG . /40% SRD116^1^2^3^4B.EP (Bar)----^BASELINE^+ WET-BASIS DATA^* DRY-BASIS DATAFigure 5.41: Carbon Dioxide of Optimum Gas-Diesel Operation.Figure 5.42 shows that gas-diesel operation produces slightlyhigher smoke emission at low and medium load, but much higher smokeemission at high load than that baseline. It is believed that thehigher smoke emission resulted from incomplete burning of thelocally rich pilot-diesel fuel. The high smoke emission at BMEF4 bar indicates that the smoke limit of gas-diesel operation isabout 4 bar. The smoke limit is related to the maximum achievableload. The diesel baseline has slightly higher smoke limit becauseof the better penetration and atomization which allows better airentrainment.oprimum DATA1.51.41 31.21.16^0.90 0.8m0^ RPM, SO BAR, 20 DEG./40% SRD1172^ 4BMEP (Bar)---- BASELINE^+ OPTIMUM DATAFigure 5.42: Bosch Smoke Index of Optimum Gas-Diesel Operation.1186. NUMERICAL SIMULATION CALCULATIONOF COMBUSTION -- ONE ZONE MODEL6.1 IntroductionIn order to simulate the diesel engine working process andcalculate the exhaust emission compositions, a one-zone exhaustemission analysis (EEA) model was created. In this model, freshair, fuel (either natural gas or diesel) and residual gas fromprevious cycle underwent a Modified Air-Standard Diesel (MASD)Cycle. The constant-pressure heat transfer to the working fluid inthe Air-Standard Diesel Cycle was replaced by the constant-pressurecombustion process of air, fuel and residual gas mixture (ie.unburned gas mixture). The STANJAN program (ie. STANJAN ChemicalEquilibrium Solver version 3.60 written by Stanford University) wasused in place of constant-pressure combustion process to calculatethe equilibrium compositions of the combustion products [43].Primarily, two subjects were investigated theoretically withthe EEA model, the effect of different fuels (diesel fuel andnatural gas fuel) on exhaust emissions of the diesel engine and theeffect of exhaust gas recycling (EGR) on NO emission of the dieselengine.This chapter contains seven sections: the introduction; adetailed description about the formulation of EEA model; thecalculation of mixture compositions of the unburned gas; thedetermination of initial condition; the thermodynamic properties2 3119the computation results; the summary.6.2 Formulation of the Exhaust Emission Analysis Model24222018181412104200^0.2^0.4^0.6^0. 8^1.2^14^1.8VOLUME ( L tre)Figure 6.1: The Modified Air-Standard Diesel Cycle (MASD).Usually, there are four processes for an actual diesel cycle:exhaust and aspiration; compression; combustion; expansion. Inorder to simulate the actual diesel cycle, a Modified Air-StandardDiesel (MASD) Cycle is applied to the EEA model. As shown in Figure6.1, the MASD cycle is a closed cycle which consists of fiveprocesses: 1. Process 1-2 is an isentropic compression (ie.reversible adiabatic compression). 2. Process 2-3 is a constant-pressure and adiabatic combustion. 3. Process 3-4 is an isentropicexpansion. 4. Process 4-0 is a constant-volume exhaust. 5. Process0-1 is a constant-volume inlet. In some cases, in order to account120for the influence of heat transfer, the isentropic processes can bereplaced with the reversible polytropic processes.For the isentropic compression process (ie. process 1-2 inFig. 6.1), the relationships in term of the initial and finalstates can be expressed as follows: [40]= ( vl ) kV2(6.1)andT^P "c-1)/kT1 P12=( 2) (6.2)where P is the fresh-charge pressure inside the cylinder, T is thefresh-charge temperature inside the cylinder, v is the specificvolume of the fresh-charge inside the cylinder and k is the ratioof constant-pressure and constant-volume specific heats at zeropressure. The subscripts "1" and "2" denote the initial and finalstates of the isentropic compression process respectively.If we replace the specific heat ratio "k" with the polytropicconstant "n" in Eqs. (6.1) and (6.2), then the two equations whichare suitable for the reversible polytropic process can be writtenas follows:P2 =( VI )P1 v2(6.3)andT^p (0-1) / n2=( 2)TiThe total energy of the system is assumed to remain the samebetween the combustion in the constant pressure and adiabaticcombustion process (ie. process 2-3 in Fig. 6.1). By neglectingchanges in kinetic and potential energies, the first law ofthermodynamics applied to this constant-pressure process is [40]a Q= dU+ 6 W= dU+ PdV= dH= 0 ( 6.5)where Q is the heat transfer to the system, U is the internalenergy of the system, W is the work done to the piston, H isenthalpy of the system. By integrating Eq. (6.5) from state 2 tostate 3, we can get the important relationH2 =H3 (6.6)where subscript "2" and "3" present the initial and final states ofthe combustion process.An interactive program, STANJAN, has been used for theconstant pressure and adiabatic combustion process to perform thechemical equilibrium analysis by the method of element potentials[44].By using the same Eqs. (6.1), (6.2) or (6.3), (6.4) , andresetting the initial and final states "1" and "2" with "3" and "4"respectively, four corresponding equations can be established forthe isentropic or reversible polytropic expansion process (ie,process 3-4 in Fig. 6.1). These relationships can be used tocalculate the exhaust temperature which can be used to correlate122test engine exhaust temperature.Although the exhaust process (ie. process 4-0 in Fig. 6.1) inthe MASD cycle (close cycle) differs from that in the actual dieselcycle (open cycle), this process has less effect on pollutantformation, so the Air Standard Cycle representation still remains.The inlet process (ie. process 0-1 in Fig. 6.1) is speciallydesigned to simulate the inlet condition of the 1-71 test engine.This is another instance that differs from the Air Standard Cycle.Details of which are discussed in Section 6.4.To simplify the computation process, the following assumptionsare applied:1. The thermodynamic state inside the cylinder is considered tobe homogeneous and uniform.2. The fresh charge trapped in the cylinder after the inlet valveclosure (corresponding to the state "1" in Fig. 6.1) includesonly residual gas from the previous cycle and fresh air, andbehaves as an ideal gas.3. Fuel (either natural gas or diesel) is directly injected intocylinder at top-dead-centre (TDC) which correspond to state"2" in Fig. 6.1, and mixes with fresh charge immediately,uniformly and homogeneously (ie. the whole mixture has thesame air-fuel ratio, temperature and pressure immediately).4. Natural gas is considered to be methane (CH4), and thediesel fuel can be approximated by CH1.8. Vaporization of theinjected diesel fuel is assumed to take place very quickly, sothat the injected diesel fuel can be treated as a gaseousEXHAUST VALVEEXHAUSTROOTS BLOWER7A-IR-BOX PISTON123diesel fuel with allowance made for heat of vaporization. Allthe fuel vaporizes immediately after injection in thecylinder.5.^The combustion products are in thermodynamic equilibrium.6.3 Mixture Compositions of the Unburned GasFRESH AIRINLETFigure 6.2: Schematic of Unif low-Scavenged Configuration.As shown in Figure 6.2, the original test engine is a unif low-scavenged two-stroke cycle diesel engine. A separate Roots bloweris used to displace the burned gases from the previous cycle aswell as to supply fresh air to the engine cylinder. With thisconfiguration, however, some of the incoming fresh air escapes withthe burned gas and part of the burned gas still remains in thecylinder after exhaust port closure. That portion of burned gas124remaining from the previous cycle is defined as the residual gas ofthis cycle. To find the effect of the residual gas on theequilibrium compositions of the combustion products, thecompositions of unburned gas inside the cylinder have to bedetermined [26].6.3.1 TerminologyThe following terms are defined for use in the subsequentsections of this chapter.OFresh air:The portion of inducted air trapped in the cylinder.OFresh charge:The whole contents of a cylinder at the inlet port closure(IPC). It consists of fresh air and residual gas from theprevious cycle.OUnburned gas:The whole contents of a cylinder after fuel injection butbefore combustion. It consists of fresh air, residual gas andfuel.•Basic cycle:The last non-combustion cycle in engine starting state.OConsequent cycles:The combustion cycles after the basic cycle in the engineoperating state.OTransient operating state:The non-combustion cycles at the starting and first few125burning cycles in which the mixture compositions of theunburned gas differ from cycle to cycle.OSteady operating state:The operating state that the cyclic variations of theunburned-gas mixture composition are very small.0Moles of trapped fresh air, MairThe portion of inducted air moles per cycle trapped in thecylinder at IPC.•Residual moles, Mres :The moles of the combustion products remaining in the cylinderfrom the previous cycle.0Moles of trapped fresh charge, Mcharge :The cylinder content at IPC, which is the summation of themoles of air trapped and the residual moles.=Msir +MMcherge^rea(6.7)•Residual molal fraction, Fres :The ratio of the residual moles to the moles of fresh chargetrapped in the cylinder at IPC.F -^ - ^MT88 Yes Aff4'2c-barge (14air+14rea)(6.8)OResidual-air molal ratio, r:The ratio of the residual moles to the moles of air trapped.r= 2-013 ^FI&B Mair (1-Fres)126(6.9)6.3.2 Compositions of Residual Gas and Unburned GasFor the two-stroke test engine because of residual gas fromthe previous cycle, the working condition can be divided into twostages, the transient operating stage and the steady operatingstage. The transient operating stage includes the non-combustioncycles in the starting and first few transient burning cycles thatthe mixture compositions of the unburned gas are different fromcycle to cycle. After a few transient burning cycles, the enginewill run into a steady operating stage in which the cyclicvariations of the unburned-gas mixture composition are negligible.PROGRAM#2 is the computation program that is designed tosimulate the transient operating stage and calculate the mixturecompositions of the unburned gas at the steady operating stage. Itis designed to have the final results with the cycling variationless than 0.01%. This program is documented in Appendix I and theassumptions applied to it is as follows:1. The diesel cycle concept is used, so the cylinder contentsbeing compressed is only fresh air and residual gas fromprevious cycle.2. The residual gas is only the fresh air in the basic cycle,which contains 21% Oxygen and 79% Nitrogen.3.^The combustion products or the residual gas consist of onlyCO„ H20, 02 and N2 in the consequent cycles. The dissociationreactions can be neglected because the burning temperature is127low in the transient operating state.4. The volumetric percentage of the residual gas is constant forevery cycle (ie. the residual molal fraction is constant forevery cycle), and the mixture compositions of theunburned gas are different from cycle to cycle in thetransient operating stage.5. The relative air-fuel ratio is constant for every cycle. Onlyfresh air and fuel are considered to be involved incombustion, but residual gas is not.The combustion reaction equation used in the PROGRAM#2 for thegeneral fuel (CH) is CHy+21 (02+3 .76N2)+0 [ (X1c02) c02+ (y10) N20+ (.7002) 02+ (x.1„.2) N2](3. 41)1(1002) c02+ (-+X0) H20(21-1-X+PX102) 02+ (7 . 521+11XIN2) N24(6.10)in whichP=9.52r1 (6.11)where X, is the relative air-fuel ratio, r is the residual-air molalratio which was defined in Subsection 6.3.1. y is the atomichydrogen-to-carbon ratio of the fuel, for methane (CH,) y=4. X1 02,X'02 and X 2 are the molal fractions of cycle ni" for CO2, H20,02 and N2 respectively which are1281+PXI-1xi -aa ^CO2nT(6.12)Xi H2.0- 2^ mo (6.13)nTVI -0221 -1 --Y-1211,71-1(V4 (6.14)nr(6.15)n =X+9 • 521(1+r)r 4 where n, is the total moles of the combustion products.(6.16)6.4 Determination of Initial ConditionAs mentioned in Section 6.3 and shown in Fig. 6.2, ascavenged-blower on the 1-71 test engine is used to displace burnedgases from the previous cycle as well as to supply enough fresh airto the engine cylinder.To simulate the operation of the actual two-stroke engine moreclosely, the measured experimental data from 1-71 test engine isused to determine the initial state "1" of the MASD cycle. They arethe inlet temperature T0=30 °C (ie. 303 K), inlet pressure P0=103kPa (ie. 1.03 atm) and air-box pressure PI=140 kPa (ie. 1.4 atm).The inlet state "0" is very close to the ambient state. The air-boxnTand129state is the scavenged-blower exit which is the same as the intakestate "1" of the MASD cycle in Fig. 6.1. The air-box temperature,T„ can be calculated by the following equation [45].T1:=T0{1+— [ (—pAO k_1] }lb PO(6.17)where lb is the scavenged-blower efficiency, here assume w=0•75•k is the specific heat ratio of air, for isentropic process k=1.4.Thus, T, and P, are known.6.5 Thermodynamic Properties of the Unburned GasIn computing the combustion product compositions and theirthermodynamic properties in the constant pressure and adiabaticcombustion process (ie. process 2-3 in Fig. 6.1), it is importantto prepare the thermodynamic properties of the unburned gas atstate "2" as the input parameters for the STANJAN program. Becauseof the constant pressure and adiabatic combustion process, P2 = P3and H, = H3 are two important relations, thus P, and H, are the inputparameters required for running the combustion function of theSTANJAN program. The combustion function is one of the functions inSTANJAN, which deal with the combustion equilibrium analysis.STANJAN has the non-combustion function to compute thethermodynamic properties of the unburned gas. It can, for example,calculate internal energy U, enthalpy H, entropy S, specific volumev, temperature T and pressure P for unburned gas. This function canbe used to prepare the H, required for running the combustion130function.PROGRAM#1 is the program to compute the unburned-gas pressureP, and temperature T2 whichare the two input parameters for runningthe non-combustion function of the STANJAN. It involves Eq. (6.1),(6.2), (6.3), (6.4) and (6.17) (see Appendix H).The detailed procedures for calculating the thermodynamicproperties of the unburned gas are presented in Appendix J.6.6 Computation Results and DiscussionIn the diesel engine combustion process, the combustion iscarried on only when the air-fuel ratio and temperature areappropriate for burning. Much unburned gas still remained in theexhaust with over-lean (or over-rich) mixtures. Thus the exhaustgas concentrations of the diesel engine (ie. experimental results)are averaged concentrations of the burned and the unburned gases.The equilibrium calculation results of the combustion productcomposition can be considered as the burned-gas composition. Thecalculation procedure is presented in Appendix K.6.6.1 Effect of Different Fuels on Burned-Gas CompositionA computation was conducted to investigate the effect of twodifferent fuels on the combustion-product NO concentrations. Thefuels involved in this computation are the diesel and the naturalgas fuels. To simplify the computation, the residual-gas influenceis omitted (ie. set the residual molal fraction to zero). Theresults of equilibrium calculation are shown in Figures 6.3through 6.6.131B754 2 44+118^24^28^35RELATIVE AIR-FUEL  RATI0DIESEL  FUEL DEO NATURAL GAS FUEL ema EXPERIMENTAL DATAFigure 6.3: Effect of Different Fuels on NO Concentration.(MASD cycle, compression ratio = 16:1, specificheat ratio = 1.4, inlet temperature = 303 K andinlet pressure = 1.03 atm)Figure 6.3 shows the effect of two different fuels on NOconcentration of burned gas. The computation was conducted in therange of relative air-fuel ratio from 1.8 to 3.6, which correspondsto the break mean effective pressure (BMEP) 4.5 bar to 1 bar of thetest engine operating range. It can be seen from Figure 6.3 thatburning natural gas fuel reduces NO emission from 8.9% to 6%132compared to diesel fuel in the computation range of relative air-fuel ratio 1.8 to 3.6. This is because the adiabatic flametemperature of the natural gas fuel is lower, about 1.08% averagedover the computation range, than that of the diesel fuel, as shownin Figure 6.4. 2.3222.^2^22^24^28^28^3^32^34^36RELATIVE AIR- FUEL RATIO0 DIESEL FUEL + NATURAL GAS FUELFigure 6.4: Effect of Different Fuels on Adiabatic FlameTemperature.ligure 6.3 also shows that the equilibrium calculation resultsof NO concentration for both fuels are considerably higher than theexperimental result of NO concentration from the test engine. ThereFigure 6.5: Effect of Different Fuels on Equilibrium COConcentration in the Cylinder.are three possible reasons for this. First, the time period of NOformation reaction is very short in the test engine, thus thereaction may not have reached the equilibrium state. This meansthat the burned-gas NO concentration of the test engine is lowerthan the equilibrium reaction NO concentration. Secondly, themeasured NO concentration of the test engine is the averaged resultover the cylinder contents, and it is lower than the burned-gas NOconcentration. Thirdly, the existence of the residual gas in thetest engine cylinder will reduce the NO concentration, though this134possibility has not been considered in this calculation. The NOcalculation result, however, is the only one that can correlatewith the experimental result because of the freezing behaviour ofthe NO reactions. BO7060504030201018^24^28^3^3RELATIVE AIR-FUEL RATIODIESEL FUEL Du NATURAL GAS FUELFigure 6.6: Effect of Different Fuels on Equilibrium CO2Concentration in the Cylinder.Figures 6.5 and 6.6 show the effects of burning diesel andnatural gas fuels on CO and CO2 concentrations of burned gasrespectively. Burning natural gas fuel reduces average 25.94% COand 46.4% CO2 respectively, compared to burning diesel fuel.7a5432ow RESIDUALCOMPRESSION RATIO = 16:1SPECIFIC HEAT RATIO = 1.4INLET TEMPERATURE = 303 KINLET PRESSURE = 1.03 atmADIABATIC FLAME TEMPERATURE = 2000 K1900 K1900 K1700K1600 K10% RESIDUAL20% RESIDUAL30% RESIDUAL135Because CO will still be oxidized into CO, after the combustion andthe oxidization rate are unable to determine, the equilibriumcalculation results of CO and CO2 concentrations can not becorrelated with the experimental results.6.6.2 Effect of the Residual Gas on Burned-Gas Compositions1.4^1.8^2.2^2.6^3^3.4^3.8RELATIVE AIR- FUEL RATIOFigure 6.7: Effect of Residual Gas on NO Concentration.The effect of the cold residual gas on burned-gas compositionwas calculated. The results for natural gas fuel are presented asa map in Figure 6.7. The residual-gas influence is presented as the136residual molal fraction in the range from 0% to 30%. The adiabaticflame temperature is presented from 1600 K to 2000 K. Thecomputation range of relative air-fuel ratio is from 1.8 to 3.6.With this map, if one knows the residual molal fraction andrelative air-fuel ratio, the equilibrium NO concentration which isthe maximum critical NO concentration value from the test enginecan easily be found. The measured experimental results will belower than the critical value because the NO formation reaction inthe diesel engine has not reached the equilibrium state. Thefarther the NO formation reaction is from equilibrium, the less NOwill be formed. Thus one approach to reduce NO concentration mightbe to match the injection parameters with cylinder pressure andtemperature as well as possibly to shorten the combustion time.6 . 7 SummaryThe Exhaust Emission Analysis (EEA) model (one-zone) with aModified Air-Standard Diesel (MASD) cycle for diesel engine is wellestablished. Based on equilibrium calculation results, thefollowing conclusions can be drawn:1. Burning natural gas instead of diesel fuel in a compression-ignition engine reduces burned-gas NO concentration withinthe whole engine operation range.2. Recycling exhaust gas in diesel engines can reduce burned-gas NO concentration within the whole engine operation range.3.^A new method is established, which can be used to calculatethe burned-gas equilibrium NO concentration values for the137specific diesel engine.4. A possible approach to reduce NO concentration is to match theinjection parameters of injector with cylinder environment(ie. pressure and temperature) as well as possibly to shortencombustion time (ie. to shorten the NO formation time) sothat the actual NO formation reaction is far from equilibrium.5. Limitations of this model are: constant pressure and adiabaticcombustion; equilibrium dissociation at single-zone adiabatictemperature.1387. NUMERICAL SIMULATION CALCULATIONOF COMBUSTION -- THREE ZONE MODEL7.1 IntroductionIn order to reveal combustion quality and predict NO emissionof the gas-diesel engine more exactly, a three-zone combustion andexhaust emission analysis model XPRESSD was established. Withmeasured cylinder pressure distribution, it was used to compute theflame temperature and the mass-burned fraction of fuel withstoichiometric (or diffusion) combustion in the gas-diesel engine.With pressure and temperature distribution, as well asreactant C:H:O:N ratio of stoichiometric complete combustion,STANJAN [43] was used to calculate the equilibrium composition ofthe combustion products. The calculated equilibrium NOconcentration after dilution by unburned-gas was used to correlatethe measured test engine tail-pipe NO concentration.The principle of the correlation of the calculated equilibriumNO concentration to the measured engine tail-pipe NO concentrationwas based on the sudden-freezing theory of the NO formationreaction [26]. This theory indicates: the NO formation rateincreases exponentially with the burned-gas temperature inside thecylinder until reaching the peak temperature; as the burned gascools during the expansion stroke, the NO reaction suddenlyfreezes. According to this theory, the engine tail-pipe NOconcentration should close to the equilibrium NO concentration139corresponding to the peak burned-gas temperature.In the following text of this chapter, a detailed descriptionof this model will be given before the discussion of the calculatedresults, followed by a summary.7.2 Formulation of the Three-Zone Combustion ModelIn the gas-diesel engine, natural gas and diesel fuel areinjected together and proportionally, beginning at the crank angledenoted by BOI. The injection duration is called the pulse width(PW). Combustion is initiated by the pilot diesel when thetemperature of the cylinder contents is above the self-ignitiontemperature of the diesel. Natural gas is then ignited by locallyhigh temperature. Mainly, the fuels undergo a diffusion combustion(ie. nearly stoichiometric combustion).Figure 7.1: Schematic of the Three-Zone Combustion Model.140A three-zone combustion model designed for representing thecombustion in the gas-diesel engine is shown in Figure 7.1. "Threezone" indicates unburned-gas, unburned-fuel and burned-gas zones.They are denoted by subscripts "ug", "uf" and "bg" respectively.The following assumptions are applied to this model:1. The pressure inside the cylinder is uniform.2. The cylinder constituents (both unburned and burned) behave asideal gases.3. The thermodynamic state of each zone is considered to behomogeneous and uniform.4. The molecular formulae, CH, and CH2, are used to representnatural gas and diesel fuel respectively.5. The fuels (natural gas and diesel fuel) are injected into thecylinder at a uniform rate over the crank angle interval PWfrom BOI with the room temperature (298 K). Diesel fuelremains liquid in the unburned-fuel zone until it burns.Vaporization of the diesel fuel takes place quickly.6. A combined fuel (CH) is used to replace natural gas (CH4) anddiesel (CH2) fuels. The hydrogen-to-carbon atom ratio y of thecombined fuel is determined by:4+ 16.043 r .(2)y.„ 11, ^14.026 m C 16 043 1+ ^r14.026 m(7.1)in which rm is the diesel to natural gas mass ratio and can bedetermined by Eq. (E.1) in Appendix E.1417. Stoichiometric combustion reaction is taken to represent thediffusion combustion in the cylinder.8. Natural gas and diesel fuels are burned with the samemass-burned fraction at any instant during combustion(proportional burning).9. The content of the unburned-gas zone is the mixture of theintake air and the residual gas from the previous cycle whichis considered to consist of 02, N2, CO2 and 1120.10. The content of the unburned-fuel zone is the uniform mixtureof the natural gas and the diesel fuel. The natural gas andthe diesel have the same temperature as the temperature of theunburned-fuel zone.11. The content of the burned-gas zone is the combustion productswhich are in thermodynamic equilibrium.Properties of the Unburned-Gas Zone: The specific internal energy of the unburned gas uog (kJ/kg)is a function of the temperature of the unburned-gas zone Tog (K).uug= uug( Tug) =u° ug+ Cvug( Tug- To)where u'og (kJ/kg) is the specific internal energy of the unburnedgas at 298 K, To (298 K) is the standard temperature, and C,ig(kJ/kg.K) is the constant-volume specific heat of the unburned gas.The specific volume of the unburned gas vog (m3/kg) is afunction of the temperature of the unburned-gas zone Tug (K) and thecylinder pressure P (kPa).v -RugTugug^p142(7.2)where Rug (kii-m/kg•K) is the unburned-gas constant.Tug can be evaluated by applying the first law ofthermodynamics to the unburned gas.dq=dhug- vwdP= cpugdT ug - vugdP^ (7.3)where dq is the heat transfer with the cylinder wall which is,c A Qdq-  1mug(7.4)where let cl=c2(mug/mtot) be a constant, c2 is another constant. AQ„,(kW/CA) is the instantaneous heat transfer with the cylinder wall,mug (kg) is the instantaneous mass of the unburned gas, and mitot (kg)is the instantaneous total mass in the cylinder.Substituting Eq. (7.2), (7.4) and constant cl into Eq. (7.3),we havedT _  -1 dPi_  C2AQwALL   Y Tug^y P mtotCpugTug(7.5)where y=Cpug/Cvug is the specific heat ratio of the unburned gas, Cpugand Cvug (kJ/kg.K) are the constant-pressure and constant-volumespecific heats of the unburned gas.By applying Eq. (7.5) to two states which are one crank angledegree apart, we have,(Tug) 2 = ( Tug)1+ ( Tug) 1 ( Y - 1 ) 132- P1 ) C2AQWALL131^m totCpug(7.6)143where subscripts "1" and "2" denote starting and ending states. c,is a constant which can be adjusted in the program to satisfy themixture and unburned-gas temperatures at the state of exhaust-portopening. The range is found in 0.3-0.5 for gas-diesel operation.Properties of the Unburned-fuel Zone: The specific internal energy of the unburned CNG ucNG (kJ/kg)is a function of the temperature of the unburned-fuel zone Tof (K).CNG 1-1 CNC, ( T or) =ucl CNG vt:NG u f 0+C^'T -T )where u°CNG (-4821 kJ/kg) is the specific internal energy of theunburned CNG at 298 K, To (298 K) is the standard temperature, andCyCNGThe specific volume of the unburned CNG vCNG (m3/kg) is afunction of the temperature of the unburned-fuel zone Tof (K) andthe cylinder pressure P (kPa).ReNoTtitv CNG —^ (7.7)where 1104G (kN-m/kg-K) is the unburned-CNG constant.The specific internal energy of the unburned diesel ups,(kJ/kg) is a function of the temperature of the unburned-fuel zoneTot (K).UDSL=UDSL ( Tut) =U° DSL + CvDSL ( Tu — To )where u°DSL (-3216 kJ/kg) is the specific internal energy of theunburned diesel at 298 K, and CvDsL (kJ/kg-K) is the constant-volumespecific heat of the diesel fuel.(kJ/kg-K) is the constant-volume specific heat of the CNG fuel.(dT) (8mgCpeNG+81ndCposL) ( (7' Lir) 1-T0)ur flax—MC +M Cg pc:NG d -pDSL(7.9)144The specific volume of the unburned diesel v,„,, (m3/kg) can beneglected because vi,s, << v CNG'Assuming that the unburned fuel inside the cylinder undergoesa two-stage process.1. Adiabatic constant-pressure mixing with newly injected fuel.2. Isentropic compression (or expansion).Within the crank angle interval of fuel injection, theexisting unburned fuel in the cylinder has a adiabatic constant-pressure mixing with newly injected fuel. By applying the first lawof thermodynamics to this process with instantaneous unburned fuel,we havedil=ingepCNGdTu f 8MgCpCNG T - u f To) +127dCpDsLCITu f+8112depDSL Tuf- TO)^( 7 • 8)where mg and 8mg (kg) are the existing and newly injected masses ofCNG fuel; md and timd (kg) are the existing and newly injected massesof diesel fuel; C EA:14G and CpDSL (kJ/kg.K) are the constant-pressurespecific heats of the CNG and diesel fuel.By rearranging Eq. (7.8), we can determine the temperaturechange caused by mixing (dTuf).,•where subscript "1" denotes the state of the fuel injection.By applying the first law of thermodynamics to isentropiccompression (or expansion) process with instantaneous unburnedfuel, we haveIn gR CNG ( Tut) 1 (P2 -131)(dTur) comp- m C +RI Cg --pCNC, d jpDSL(7.11)145dH- VdP=mgCpc,NGdTuf+mdCr,DsLdTuf-mgvcA,GdP-mdvDsLdP=0^( 7.10 )Substituting Eq. (7.7) and v,s,=0 in to Eq. (7.10), we candetermine the temperature change caused by compression.^(dT )IngR CNGT u f ^dPu f camp m r,^m r,"`ge- peNG "V---pDsL Dorwhere subscripts "1" and "2" denote starting and ending states.Thus, the temperature of the unburned fuel at state "2" can beevaluated from the previous state "1".Tuf )^( Tut)1+ ( dTut) mix+ ( dTur) comp^ (7.12)Properties of the Burned-Gas Zone: Both the specific internal energy of the burned gas ubg (kJ/kg)and the specific volume of the burned gas vbg (m3/kg) are a functionof the temperature of the burned-gas zone Tbg (K) and the cylinderpressure P (kPa).u g =u (Tbg' P)b bg Tibg=1/bg(Tbg,The relation among instantaneous volumes of cylinder V,unburned-fuel zone Vuf, burned-gas zone Vbg and unburned-gas zone Vugis146V=Vut+Vbg+Vug=mgvg+mcivembgvbg+(mtot-mg-mci-mbg)vug^(7.13)where v (m3/kg) is the specific volume, and m (kg) is theinstantaneous mass. Subscripts "g", "d", "bg", "ug", "uf" and "tot"denote CNG, diesel, burned-gas, unburned-gas, unburned-fuel andtotal cylinder contents respectively.Dividing into, on both sides of the Eq. (7.13) and rearrangingit, we can get the mass conservation equation.111)6.7g g ug^d d ug  - m ^+ (1 -^ ) V ug14-m (v -v ) -m (v -v ) mH?totIltot 4w Intot(7.14)xi=mbg/into, is defined as the burned-gas mass fraction. Becausethe left side of Eq. (7.14) is the known quantity, let v,=EV-m9(vg-vug)-md(vd-vug)1/mtot. The mass conservation equation can now besimplified tovm =xvbg (T^+ ( 1 -x1 ) Vugg (7.15)The instantaneous relation of internal energy for the wholecylinder contents isE =U +U +U =m u +m u +m u +(m -m -m -m )u0 uf bg ug gg dd bg bg^tot g d bg ug (7.16)where Etot is the total internal energy of the cylinder contents, U(kJ) and u (kJ/kg) are the internal energy and the specificinternal energy respectively, and m (kg) is the instantaneous mass.Subscripts "g", "d", "bg", "ug", "uf" and "tot" denote CNG, diesel,burned-gas, unburned-gas, unburned-fuel and total cylinder contents147respectively.Dividing mt.„ on both sides of the Eq. (7.16) and rearrangingit, we can get the energy conservation equation.Etot -ilt (11gug)M d ( U d ug) _ M^bg u+ (1_ Mbg ) uub9112t0t^ M tottot(7.17)Because the left side of Eq. (7.17) is the known quantity, letum=i Et0C"'in g^ hatot^and^xi=mbghntot•^The^energyconservation equation can now be simplified toUm=XiUbg(Tbg, P) + ( 1-x1) Uug^ (7 .18 )By given P, vm, um, vug and uug, two equations, Eq. (7.15) and(7.18), can be used to solve Tbg and x, iteratively [47].x is defined as the mass-burned fraction of the fuel which canbe computed as follows:_^X (111^+M^+112^)1 trap CNG DSLX- (7.19)M fuel (mCNG ÷mDSL ) (1+ ^1 REASTOIC)where rnbf and Mfuel (kg/cycle) are masses of burned fuel and totalmass of the fuel per cycle, mtrap (kg/cycle) is the mass of thecylinder contents at IPC, which is the sum of the mass of airtrapped in the cylinder and the residual mass from the previouscycle. ma,G (kg/cycle) and mDs, (kg/cycle) are the total masses ofthe CNG and the diesel fuels per cycle. RFASTOIC is the equivalentstoichiometric fuel-air ratio.148Total Internal Energy of the System: If one considers the whole cylinder contents as a system, thefirst law of thermodynamics that applies to the system for a smalltime change At isdEt0t=80-8"in9hc-NGIn+8111dhpsunwhere Q is the heat transfer with the cylinder wall, W is the workdone to the piston, and E,, is the total internal energy of thesystem. &rig and &ad are injected masses of CNG and diesel fuel.Assume both fuels enter the cylinder at 298 K. Thus, h„„„ (-4667kJ/kg) is the enthalpy of CNG at 298 K, and hps,in (-3216 kJ/kg) isthe enthalpy of diesel at 298 K. Thus the total internal energy atstate "2" can be evaluated from previous state "1" over a giventime step At corresponding to 1 degree crank angle in this model.(Etot)^(Etot) 1+AQ1-2-41 W1-2 +13InghCNGin+8111dhDSLin^(7.20)Work Done to the Piston: The average pressure of the system from state "1" to "2" fora time step At can be defined as-p_  Pl +P22Thus the work done on the piston from system can be written asP +PAW1,2- 12 2 (V2-171) (7.21)where P is the pressure of the cylinder, and V is the volume of thecylinder.149Heat Transfer with Cylinder Wall: Usually, heat transfer in engine consists of convective andradiative heat transfers with cylinder wall (including piston topwall and cylinder head flame wall). In order to simplify thecomputation, in this model assume that only the unburned gascontacts the cylinder wall and has convective and radiative heattransfer with the cylinder wall. The total heat transfer to thecylinder wall QwL is the sum of the convective 0—CONV and theradiative Q heat transfer to the cylinder wall.()WALL= Qcolvv+ QRADThe heat transfer between state "1" and "2" for a given timestep At can be evaluated asA Q1 -.2^()WALL (360) )60where Qw„, is in kW and AQ1 is in kJ/CA, and N (rpm) is the enginespeed.The general formula of convective heat transfer 0—CONV SQcoNv=-hXAsuifX(Tbuik-^ (7.22)where h is the convection heat transfer coefficient (kW/m2.K), ASURF(m2) is the wall surface area which the unburned gas contacted, Tw11is the averaged cylinder wall temperature (assume Twall= 450 K in thismodel), and Tbu„ is the bulk temperature of the cylinder content[48] which is determined byM T +M TTbu 1 k- ug ug bg bg (1 - X1 ) Tug+ XI Tbgmug - Mbg150(7.23)where mug and Tug are the instantaneous mass and the temperature ofthe unburned-gas zone. mbg and Tbg are instantaneous mass andtemperature of the burned-gas zone. x, is the burned-gas massfraction.As recommended by Woschni [48], the Nusselt number (Nu) in theengine can be expressed asNu= hB =c3 (Re) C4k (7.24)The instantaneous heat transfer coefficient in the engine canbe expressed ash-c3k(Re)c4^ (7.25)where c3 is a coefficient and in the range of 0.35-0.8, c4 isanother coefficient and in the range of 0.75-0.87 for the gas-diesel operation, and B is the diameter of the cylinder bore (m).k=Cpug-R/Pr (kW/m•K) is the thermal conductivity of the unburnedgas, in which Cpug (kJ/kg-K) is the specific heat of unburned gas,(kg/m-s) is the dynamic viscosity of unburned gas, and Pr is thePrandtl number, Pr=0.7 in this model.Re=p-Vp„s-B/R is the Reynolds number, in which p (kg/m3) is thedensity of unburned gas, %/pis (m/s) is the piston velocity.151Radiative heat transfer Q can be computed with the followingequation.QRAD= - Fxlis „ x (Tug - T„ii)^ ( 7.26 )where F=FexFGxo=1.6x10-12 is the multiply of three factors, F, is theemissivity function, FG is the geometric function, and u is theStefan-Boltzmann constant with a valve of 5.669x10-8 W/m2•1<4.7.3 Mass of Air Trapped in the Cylinder and ResidualMass FractionIn the uniflow-scavenged, two-stroke test engine, the unburnedgas is the mixture of the fresh air trapped in the cylinder and theresidual gas from the previous cycle. To estimate the amount of airtrapped in the cylinder at the end of the scavenging process, thefollowing definitions are used.ODelivered air mass, mair:The mass of air delivered to the engine per cycle as measuredat the intake line.•klass of air trapped, Matra"):The portion of the delivered air mass per cycle trapped in thecylinder at the inlet port closure (IPC).OResidual mass, mres:The mass of the combustion products remaining from theprevious cycle.OTrapped mass, Intrap:The mass of the cylinder contents at IPC, which is the sum of152the mass of air trapped and the residual mass.^=M^+mtrap a trap resODelivery ratio, A:The ratio of the delivered air mass to the trapped mass.A-  mairtrap•Degree of purity of the charge, DP:The ratio of the mass of air trapped to the trapped mass.DP- ma trap —  Matra!? m trap matrap +OResidual mass fraction, f resThe ratio of the residual mass to the trapped mass.^f ^ raS^r res " m trap m EGS +matrapIf one assumes that the residual gas has the same molecularweight as air, the residual mass fraction is equal to the residualmol fraction.An iteration process is used for the computation of Matrap andf res • For given exhaust temperature Texh air box temperature T"",and estimated DP and mt,p, the initial temperature of the residual,T,. (K), can be determined fromTres-Te,thCpwth( Mai r 4-IncAIG +MDSL) TaboxCpair (Ma -DPxm^)trap Cpres DPxm trap+McNG Ina5L(7.27)where main MCNG and mm31, (kg/cycle) are the delivered air mass, total153injected mass of CNG per cycle and total injected mass of dieselper cycle respectively. Cpexh, Cpair and Cpres (kJ/kg•K) are theconstant-pressure specific heats of exhaust gas, air and residualrespectively. Detailed derivation of this equation is given inAppendix L.The temperature of the cylinder contents at IPC T (K) isTipc= T eboxDP+ Tres ( -DP)For given temperature, pressure and volume at IPC, the trappedmass in the cylinder at IPC and the delivery ratio can bedetermined with the formulae:P • V •ipc ipcIn^ —trap RugTi. pcA-  m". rm trapwhere Rug (kN.m/kg•K) is the gas constant of the unburned gas (ie.the cylinder contents at IPC has the same composition as theunburned gas).An empirical equation of the degree of purity as a function ofthe delivery ratio for the unif low scavenging [47] is used in thismodel, which isDP=0 .173611A3-0 .95982A2+1 .774305A-O.19642^(7.28)Figure 7.2 shows a typical uniflow-scavenging data range forthe two-stroke diesel engine [26] and the position of the abovePERFECT DISPLACEMENTUNIFLOW-SCAVENGING TOP LIMITUNIFLOW-SCAVENOING BOTTOM LIMITI I^ICOMPLETE MIXING1541.21 .10.90.8a0.^0.6^1.4^1.8^2.2DELIVERY RATIO---- EMPIRICAL EQUATIONFigure 7.2: Typical Scavenging Data Range of Two-Stroke Diesel.empirical equation curve.After entering the new DP and mtrap into the Eq. (7.27) anddoing the iteration until the new DP equals the previous DP, thenlaatrap (kg/cycle) and f- res are evaluated fromand=1)1):00/Inatraw^trap (7.29)48,7=1 -BP^ (7.30)1557.4 Unburned-Gas CompositionBecause of the residual gas, the composition of the unburnedgas varies during the first few cycles after firing. PROGRAM#2 andEqs. (6.10) through (6.16) in Subsection 6.3.2 are adopted intothis model to simulate the initial transient variation of thecomposition and calculate the composition of residual gas at steadystate. The following assumptions are made for simplifying thecalculation.1. Residual gas consists of only CO2, H20, 02 and N2. The molalfractions of these components (ie. X CO2 X1H20 X102 andr-,2) and the molecular weight of the residuals (ie. MWres)have already been calculated in the previous cycle. Residualmass fraction (ie. fres) is known.2. Fresh intake air consists of only 02 and N2. Mass of airtrapped in the cylinder (ie. Matrap ) is known.The number of moles of the residuals N,s is determined in theformula:_  fIGO  _Et.. 2Vre° 1— fr. mw.reffThe total number of moles of the unburned-gas beforecombustion taking place Ntotsg is determined in the formula:AT^1_25.2.M4V tOtug MWair'I"' ir0.79^+NrosIri N2—^ —Plyvair X112^112 "totug Ntotug(7.31)156The molal fractions of the unburned-gas components, X02, X,2,and X1120, are evaluated in the formulae:X 02 NO2 — ITIV totug0.23.( -82Ainyatr ) +NresX102'11rwair NtotugX — NCO2  _  Nreayir- CO2CO2 AT"totug^Ntotugv =  H20  _  resKi H20`k1120 ATavtotug^Ntotugwhere NO2, NN2, New and N1120 are the number of moles of 02, N2, CO2 andH20 in the unburned-gas. Mai,. (28.97 kg/kmol) is the molecularweight of the intake air.7.5 Unburned-Fuel Mass Ratio in ExhaustUnburned-fuel mass ratio in the exhaust, UFRAT, is defined asa mass ratio of unburned fuel in the exhaust to the total injectedfuels (including CNG and diesel fuel). It is used to normalize thefuel mass-burned fraction. It can be computed directly frommeasured engine exhaust emission data, air mass flow rate and fuelmass flow rate. Unburned fuel in the exhaust is measured as thetotal hydrocarbon emission HC. The following assumptions areapplied in the computation process.1.^The engine exhaust consists of CO, CO„ NO„ H20, 02, N, andunburned HC.1572. The unburned HC has the same composition as the combined fuel.3. All measured emission data have already been converted to wetbasis in ppm (or % by volume).Unburned-fuel mass ratio in the exhaust, UFRAT, can becalculated from the following equation.VFRAT- IIIme tasilfilion(7.32)where ginc , bum and zhca,G (kg/hr) are mass flow rate of unburnedHC, diesel and CNG respectively.The mass flow rate of unburned HC, then, can be calculated byusing Eqs. (4.15) through (4.21) in Subsection 4.4.2. In consistentto the molecular formulae of CNG and diesel, 16.043 kg/kmol and14.026 kg/kmol are used as the molecular weights of CNG and dieselfuels in Equ. (4.20). The mass flow rate of the diesel and the CNGare measured.7.6 Averaged Equilibrium NO ConcentrationTo simulate the diffusion combustion in the test engine, astoichiometric complete combustion pattern was applied in thismodel. The combustion equation for the fuel (CH) and the unburnedgas at the maximum burned-gas temperature Tbgn,ax and correspondingpressure PTbgmax i 5158Xu,"^X/ V,CH + (1+)7) (0 +—N^20+ "e-=-= CO2)4^2 X02 2 X02 ^X02y X v X y X=> [1+^CO2] CO +[-Y—+ (1+-.'–)] 112°H 0+ (1+ ") " N4 X02^2 2^4 X02^4 17— 202(7.33)where y is the hydrogen-to-carbon atom ratio of the combined fuelas determined by Equ. (7.1). Xw, X2, X„20 and )(co, are the molalfraction of oxygen, nitrogen, water and carbon dioxide in theunburned gas which are determined by Equ. (7.31) in Section 7.4.The C:H:O:N ratio at stoichiometric complete combustion can bedetermined as2^nn[1+— (1+—) X H2°]H _^n^4 X02 (7.34)n XCO23.4 (1+ —) v4 4102(14-22)(2+-tra+24Y1!"2)0_^4^'02^'02 (7.35)n1+(l+--) X.,24^4.1.02n2 (1+— )N_ ^4 X02 n Xco24 42(7.36)STANJAN is the program that is used to calculate theequilibrium burned-gas composition with dissociation combustion.Thus, the equilibrium NO concentration in the burned-gas zone[N062,,,L can be obtained by entering C:H:O:N ratio, Tbgmax and P Tbgmaxinto Stanjan. A correction factor, CFK, is used to average159equilibrium NO concentration over whole cylinder contents.[NO] AvG_Hom =CFKx [NO] Born^ (7.37)where [NO] AVG-EQUIL is the averaged equilibrium NO concentration overwhole cylinder contents.Determination of the CFK starts at maximum burned-gastemperature (ie. Tbg = Tbcmax).Mass of fuel per cycle Mf uel (kg/cycle) is determined in theformula:M fue1=MD9L4Mc:I3^ (7.38)where MDSL and me„,, (kg/cycle) are the mass of diesel and CNG percycle respectively, which can be calculated from the measured massflow rates of diesel and CNG.Mass of the fuel burned Mf bmax (kg/cycle) and number of moles offuel burned Nfbmax at maximum temperature Tbgmax are determined in theformulae:mfbmax=XTbgmaxXM fuelN fbmax mfbmax"ICwhere xTbgmax is the fuel mass-burned fraction at Tbx and MWgmathe molecular weight of the fuel (CH).Considering that the fuel (CH) undergoes a stoichiometriccomplete combustion with unburned gas at Tbgmax and Pmgmax Eq. (7.33)is used to determine the number of moles of burned gas Nbgmax atTbgmax PTbgmax •160v X^iv bmax=N ftwx [3. + + (1+ ) ^CO2 H20 N2 )2^4^X02Total mass and total number of moles of the engine exhaust,Mexh and Nexh areInexh=Mair+MfuelAT —^AiTA7""Wthwhere mair (kg/cycle) is delivered air mass determined from measuredair mass flow rate, mfuei (kg/cycle) is mass of the fuel determinedfrom Equ. (7.38), and MWexh is the molecular weight of the exhaust.Then the average equilibrium NO concentration, [NO1 AVG-EQUIL can becalculated by-bibpmxr[No] Av0_13001-L^LNOJ Homy Nwchwhere CFK is defined as:^CFK-  bgiaax^(7.39)exti120ORP1, 606AR, 25%CN62, 200EG./40%SRD140 160 190 200^220 240 260 2E107621617.7 Computation ResultsCRANK ANGLE (DEG ABDC)Figure 7.3: Cylinder Pressure Distribution (BMEP = 4 bar).Computations were conducted to investigate the fuel combustionpattern, the effect of various operation parameters on the ignitiondelay, the relationship of calculated NO to measured NO, and theeffect of EGR on NO concentration. The calculation procedure interms of work sheet is presented in Appendix M and the programXPRESSD is documented in Appendix N.The cylinder pressure distributions were the source data forthe calculation. Figure 7.3 shows a example of these data.' .. . ... .. .... . .. . ......^:: : :132.82.6242.,40.201200RPM, BOBAR, 25%CN62, 200EG./4096SRD140^180^180^200^220^240^250^280CRANK ANGLE (DEG. ABDC)0 UNBURNED-FUEL ZONE^+ UNBURNED-GAS ZONE^BURNED-GAS ZONE^A BULKFigure 7.4: Temperature Distributions (BMEP = 4 bar).The typical temperature distributions of the three-zone modelare shown in Figure 7.4. They were directly calculated from thecylinder pressure distribution shown in Figure 7.3. The bulktemperature was used to evaluate heat transfer to the cylinderwall.Combustion Pattern in the Gas-Diesel Engine: The three test conditions investigated were:1)^Straight diesel operation at 1200 rpm speed, 4 bar load, and12°BTDC BOI (ie. diesel baseline).1621632) Gas-diesel operation at 1200 rpm speed, 4 bar load, 60 bar gaspressure, 25% diesel ratio, 40% fuel jet interruption ratio(% SRD), 200 injection angle, and 32°BTDC BOI.3) Gas-diesel operation at 1200 rpm speed, 4 bar load, 60 bar gaspressure, 25% diesel ratio, 0% fuel jet interruption ratio(% SRD), 20° injection angle, and 32°BTDC BOI.1200 RPM1.10.92LLu_^0 . 70^08ft 0.50.4a0 33^0.20.1 140^180^180^200^220^240^280^280CRANK ANGLE (DEG. ABDC)^0 BASELINE (DIESEL) + 411MSR0 (GAS-DIESEL)^Q 0%SRD (GAS-DIESEL)Figure 7.5: Fuel Mass-Burned Fraction (BMEP = 4 bar).The combustion rates of three test cases are shown in Figure7.5. As shown, operation with straight diesel has shorter ignitiondelay than that with gas-diesel. Ignition delay is defined as acrank angle interval (or time period) from BOI to the point with 1%164of fuel burned. From two gas-diesel operation cases, it is seenthat pilot-diesel fuel burns first with a fast burning rate andnatural gas burns later with a slow burning rate. This may suggestthat considerable amount of pilot diesel is well mixed with airafter a long ignition delay, thereby suddenly burning with apremixture burning rate when the temperature is ready.Ignition Delay: 1200 RPM, 60 BAR,25%CN62,20DEG./096SRD525048464442403836343230282625^27^28^31^33^35^37^39BMEP [BAR)0 801=28 DEG.BTDC^+ 501=32 DEG.BTDC^0 501=36 DEG.BTDCFigure 7.6: Effects of BOI on Ignition Delay.The ignition delay was calculated according to abovedefinition in the model within 5% error range at 1200 rpm.165The effect of BOI on ignition delay is shown in Figure 7.6.Test condition was gas-diesel operation at 1200 rpm speed, 60 bargas pressure, 25% diesel ratio, 0% fuel jet interruption ratio (%SRD), and 200 injection angle. It is interesting to see that thereis almost no difference in ignition delay with BOI at 28°BTDC andat 32°BTDC; but a big difference results from increasing BOI to36°BTDC.1200 RPM 454035SRD CASE3025201540% SRD CASE10DIESEL BASELINE52^22^24^2^28^3^32^34^3^38^4BMEP (BAR)0 DIESEL BASELINE^+ 40% SRD CASE^0% 3RD CASEFigure 7.7: Effect of Fuel Jet Interruption Ratio on IgnitionDelay.Figure 7.7 shows the effect of fuel jet interruption on theignition delay. It can be seen that operating with gas-diesel has166much longer ignition delay than with straight diesel. The typicalignition delay for diesel operation is in the range of 7 - 9 °CA(or 0.97 - 1.25 ms) at 1200 rpm, and for gas-diesel operation is inthe range of 26 - 46 °CA (or 3.6 - 6.4 ms) at 1200 rpm. It is alsofound that ignition delay is shorter with 40% SRD than with 0% SRD(ie. no interruption), which indicates that fuel jet interruptionhelps fuel ignition.1200 RPM, 60 BAR025%CN62,20DEG./0%SRD 424140393937363534333225^27^29^31^33^35EIMER (BAR)0 NON-EGR CASE^+ EGR CASEFigure 7.8: Effect of EGR on Ignition Delay(with BOI = 28 °Bamc).The effect of EGR on ignition delay is shown in Figure 7.8. Inthis case, about 20% of exhaust gas was recycled into the cylinder167and the residual fraction in the cylinder increased about 35%. Itis found that EGR slightly reduces ignition delay and is moreeffective at low load than at high load.Relationship between Calculated NO and Measured NO: 1200 RPM 60 BAR,25%CN62,20DEG./40%5RD00029009007005005004003002001002^22^24^25^28^3^32^34^36^38^4EIMER CBAR)0 MEASURED NO^+ CALCULATED NOFigure 7.9: Correlation of Measured NO and Calculated NO.The equilibrium NO concentration of the combustion productswas calculated in the model. An example of the correlation of thecalculated NO concentration to the measured NO concentration (ortail-pipe NO concentration) is shown in Figure 7.9. The differencebetween calculated and measured NO was within 20% for the whole168gas-diesel operation cases and up to 40% difference for the dieselbaseline.Reduction of Tail-Pipe NO with EGR: 1200 RPM, SO BAR,2596CN62,20DEG./0%SRD300280260240220200180160140120100806040 25^27^29^31^33^356MEP CBAR)^0 NON-EGO [MEASURED)^+ EGR [MEASURED)Figure 7.10: Effect of EGR on Measured NO Concentration(with BOI = 28 ITIDC).The effects of EGR on both measured and calculated NOconcentration shown in Figures 7.10 and 7.11. In this case, about20% of exhaust gas was recycled into the cylinder and the residualfraction in the cylinder increased about 35%. It is found that EGRreduces NO emission significantly (about 50% reduction at low load169and 25% reduction at high load for measured data). The NO reductionwith measured data is larger than with calculated data.1200 RPM, 60 BAR,25%CN62,20DEG./096SRD300280260240220200190160140120100906040 25^27^29^31^33^asBMEP (BAR)^NON-EGR(CALCULATED)^+ EGR (CALCULATED)Figure 7.11: Effect of EGR on Calculated NO Concentration(with WI = 28 1Trpc).7 . 8 Summary1. The three-zone combustion and exhaust emission analysis modelXPRESSD was established on the mass and the energy conservationequations.2. The estimation of the amount of fresh air trapped in thecylinder and residual fraction followed the procedure recommended170by Heywood [26] has been employed.3. A method for simulating the initial transient variation of thecomposition and calculating the composition of the residual gas atsteady state was adopted into this model. The evaluation of theunburned-gas compositions was based on the result of this method.4. The stoichiometric complete combustion was applied in thismodel to simulate the diffusion combustion. The burned-gastemperature was calculated with a error range of ±25 K.5. The equilibrium NO concentration was evaluated with STANJAN[43] and correlated with the measured tail-pipe NO concentration.6. The unburned-fuel mass ratio in the exhaust determined by theunburned hydrocarbon in the exhaust was used to normalize the fuelmass-burned fraction curve.7. The heat transfer with the cylinder wall was applied in themodel. The heat transfer coefficients were used to adjust the fuelmass-burned fraction curve to a normalized value.8. The ignition delay was defined and evaluated within 5% errorat 1200 rpm. The ignition delay of the gas-diesel operation wasfound to be 3 - 5 times longer than the straight diesel operation.9. Fuel jet interruption was a method of reducing ignition delay.10. EGR was an efficient method to reduce NO emission.1718. CONCLUSIONS AND RECOMMENDATIONS8.1 ConclusionsThe influences of variable parameters on performance andemissions of a gas-diesel engine were investigated bothexperimentally and theoretically. The conclusions drawn are asfollows:1. Replacing diesel fuel by natural gas in a diesel engine canreduce NO and CO 2 emissions while meeting or exceeding dieselengine efficiency.2. With a poppet-valve gas injector and entrained diesel pilotignition fluid, a minimum diesel energy ratio of about 25% wasfound to be necessary to provide stable operation over thewhole load range.3. The optimum natural gas injection pressure was 60 bar, whichmeans an injection pressure ratio varying over the injectionperiod in the range 2.8 to 1.5; engine performance andexhaust emissions depended strongly on injection pressure.4. With a poppet-valve gas injector and entrained diesel pilotignition fluid, combustion was typically incomplete at lowload, as shown by excessive emissions of CO and unburnedhydrocarbon (THC).5.^Engine performance and exhaust emissions were sensitive toinjection angle and jet sheet geometry (ie. jet interruption);improvements made in the light of flow visualization studies172lead to small improvements in the engine efficiency andexhaust emissions.6. Diesel pilot fluid with high cetane number can improve engineefficiency and reduce unburned hydrocarbon (THC) emission atlow and medium load.7. Replacing diesel fuel with natural gas in a diesel engine cansignificantly reduce NO formation, as shown by bothmeasurements and equilibrium estimates of NO concentration.8. Ignition delay of the gas-diesel operation was found to bemuch longer than that of the straight diesel operation; properfuel jet sheet geometry (ie. jet interruption) can reduceignition delay.9. BOI retardation can significantly reduce NO with only slightreduction of thermal efficiency.10. Reduction of NO concentration by EGR (exhaust gas recycling)was been shown to be significant over the whole engineoperation range, but this also resulted in a slight increasein carbon monoxide (CO) and unburned hydrocarbon (THC)emissions.11. Correlation between NO concentration measured in the exhaustpipe and NO estimated from equilibrium concentration (at peaktemperature in the cylinder and averaged over the wholecylinder contents) was close at low and medium load.1738.2 RecommendationsFrom the above conclusions, to improve engine performance andemissions, especially at low load, the present injection patternand injector configuration ought to be modified. The followingchange are suggested:1. Separate the diesel-pilot-fluid passage in the injector fromthe CNG passage completely to control the quantity and BOI ofthe pilot diesel precisely.2. Inject the diesel pilot fluid first; inject the CNG into thehigh temperature zone close to the combustion zone of thediesel pilot fluid to ignite the in coming CNG/air mixture.3. Improve spray quality of the diesel pilot fluid and atomizediesel droplets for more rapid vaporization and mixing.4. In order to achieve above purpose, a multi-orifice injectorwith separate fuel passages is suggested. The size of CNGorifices, which should be larger than pilot-diesel orifices,can be determined experimentally. A injection angle of about200 is recommended. CNG injection pressure should be adjustedfor each engine.174REFERENCES[1] Journal of Air Pollution Control Association,  v24, January1974.[2] Wark, K. and Warner, C.F., Air Pollution -- Its origin and Control, First Edition, 1976.[3] Karim, G.A. and Wierzba, I., Comparative Studies of Methane and Propane as Fuels for Spark Ignition and Compression Ignition Engines, Society of Automotive Engineer, SAE Paper831196, 1983.[4] Karim, G.A., The Dual Fuel Engine of the Compression Ignition Type - Prospects, Problem and Solution - A Review, Society ofAutomotive Engineer, SAE Paper 831073, SAE Trans. v92, pp.569-577, 1983.[5] Fraser, R.A., Siebers, D.L. and Edwards, C.F., Autoignition of Methane and Natural Gas in a Simulated Diesel Environment,Society of Automotive Engineer, SAE Paper 910227, 1991.[6] Karim, G.A., On the Emission of Carbon Monoxide and Smoke fromCompression Ignition Engines; Including Natural Gas Fuelled Engines, Proceedings of 2nd Int. Clean Air Congress, AcademicPress, p. 617, 1970.[7] McJones, R.W. and Corbeil, R.J., Natural Gas Fuelled Vehicles; Exhaust Emissions and Operational Characteristics,Society of Automotive Engineer, SAE Paper 700078, 1970.[8] Fleming, R.D. and Allsup, J.R., Emission Characteristics of Natural Gas as an Automotive Fuel, Society of AutomotiveEngineer, SAE Paper 710833, 1970.[9] Nicholls, J.A., Sichel, M., Gabrijel, A., Oza, R.D., andVandermolen, R., Detonatability of Unconfined Natural Gas-Air Clouds, Seventeenth Symposium (International) onCombustion, The Combustion Institute, Pittsburgh, p. 1223,1978[10] Bull, D.C., Elsworth, J.E., and Hooper, G., Susceptibility of Methane-Ethane Mixtures to Gaseous Detonation in Air,Combustion and Flame, v34, p.327, 1979.[11] Vandermolen, R. and Nicholls, J.A., Blast Wave InitiationEnergy for the Detonation of Methane-Ethane Mixtures,Combustion Science and Technology, v21, p.75, 1979.175[12] Eubnk, C.S., Rabinowitz, M.J., Gardener, W.C., Jr., andZellner, R.E., Shock-Initiated Ignition of Natural Gas-Air Mixtures, Eighteenth Symposium (International) on Combustion,The Combustion Institute, Pittsburgh, p. 1767, 1980.[13] Westbrook, C.K. and Pitz, W.J., High Pressure Autoignition of Natural Gas/Air Mixtures and the Problem of Engine Knock, GasResearch Institute Technical Report No. GRI-87/0264, Sept.1987.[14] Siebers, D.L. and Edwards, C.F., Autoignition of Methanol and Ethanol Sprays under Diesel Engine Conditions,  Society ofAutomotive Engineer, SAE Trans. v96, pp. 5.140-5.152, 1987.[15] Glassman, I., Combustion, Second Edition, ACADEMIC PRESSINC., 1987.[16] Beck, N.J., Johnson, W.P., George, A.F., Petersen, P.W.,van der Lee, B., and Klopp, G., Electronic Fuel Injection for Dual Fuel Diesel Methane, Society of Automotive Engineer, SAEPaper 891652, 1989.[17] Simonson, J.R., Some Combustion Problems of the Duel-Fuel Engine, Engineering v178, p. 363, 1954.[18] Moor, N.P.W. and Mitchell, R.W.S., Combustion in Duel-Fuel Engine, Joint Conference on Combustion, ASME/Institute ofMechanical Engineering, p. 300, 1955.[19] Wong, J.K.S., Messenger, G.S., Moyes, B.W., and Chippiior, W.Conversion of a Two-Stroke Detroit Diesel Allison Model 12V-149T Diesel Engine to Burn Natural Gas with Pilot Injection of Diesel Fuel for Ignition, Society of AutomotiveEngineer, SAE Paper 841001, 1984.[20] Ding, X. and Hill, P.G., Emissions and Fuel Economy of a Prechamber Diesel Engine with Natural Gas Dual Fuelling,Society of Automotive Engineer, SAE Paper 860096, 1986, SAETrans. v95, pp. 612-625, 1986.[21] Johnson, W.P., Beck, N.J., Lovkov, O., van der Lee, A.,Koshkin, V.K. and Piatov, I.S., All Electronic Dual Fuel Injection System for the Belarus D-144 Diesel Engine, Societyof Automotive Engineer, SAE Paper 901502, 1990.[22] Miyake, M. and Ass., The Development of High Output, Highly Efficient Gas Burning Diesel Engines, CIMAC 15 Paper D112,Conference Proceeding, Paris-France, Jun. 1983.[23] Miyake, M., Endo, Y., Biwa, T., Mizuhara, S., Grone, 0. andPedersen, P.S., Resent Development of Gas Injection Diesel Engine, CIMAC 17 Paper.176[24] Wakenell, J.F., O'Neal, G.B. and Baker, Q.A., High Pressure Late Cycle Direct Injection of Natural Gas in a Rail MediumSpeed Diesel Engine, Society of Automotive Engineer, SAEPaper 872041, 1987.[25] Lom, E.J. and Ly, K.H., High Pressure Injection of Natural Gas in a Two Stroke Diesel Engine, Society of AutomotiveEngineer, SAE Paper 902230, 1990.[26] Heywood, J.B., Internal Combustion Engine Fundamentals,McGraw-hill Publishing Company, 1988.[27] Komiyama, K. and Heywood, J.B., Predicting NO Emissions andEffects of Exhaust Gas Recirculation in Spark-Ignition Engines, Society of Automotive Engineer, SAE Paper 730475,SAE Trans. v82, pp. 1458-1476, 1973.[28] Haynes, B.S. and Wagner, H.G., Soot Formation, Prog. EnergyCombust. Sci., v7, pp. 229-273, 1981.[29] Whitehouse, N.D., Clough, E. and Uhunmwangho, S.O., TheDevelopment of Some Gaseous Products during Diesel Engine Combustion, Society of Automotive Engineer, SAE Paper 800028,1980.[30] Aoyagi, Y., Kamimoto, T., Matsui, Y. and Matsuoka, S., A Gas Sampling Study on the Formation Processes of Soot and NO in a DI Diesel Engine, Society of Automotive Engineer, SAE Paper800254, SAE Trans. v89, 1980.[31] Duggal, V.K., Priede, T. and Khan, I. M., A Study of Pollutant Formation within the Combustion Space of a Diesel Engine, Society of Automotive Engineer, SAE Paper 780227, SAETrans. v87, 1978.[32] Torpey, P.M., Whitehead, N.J., Wright, M., Experiments in the Control of Diesel Emissions, Paper C124/71, "Air PollutionControl In Transport Engines", A Symposium arranged by theAutomobile Division and the Combustion Engines Group of theInstitution of Mechanical Engineers, 9th-llth November, 1971.[33] Khan, I.M., Wang, C.H.T., Factors Affecting Emissions of Smoke and Gaseous Pollutants from Direct Injection Diesel Engines, Paper C151/71, "Air Pollution Control In TransportEngines", A Symposium arranged by the Automobile Division andthe Combustion Engines Group of the Institution of MechanicalEngineers, 9th-llth November, 1971.[34] Herzog, P.L., Burgler, L. and Winklhofer, E., NOx Reduction Strategies for DI Diesel Engines, Society of AutomotiveEngineer, SAE Technical Paper Series 920470, 1992.177[35] Hames, R.J., Straub, R.D., and Amann, R.W., DDEC --Detroit Diesel Electronic Control, Society of Automotive Engineer,SAE Paper 850542, 1985.[36] Hames, R.J., Hart, D.L., Gillham, G.V., Weisman, S.M.,Peitsch, B.E., DDEC II -- Advanced Electronic Diesel Control, Society of Automotive Engineer, SAE Paper 861049,1986.[37] Yuen, D. and Hodgins, K.B., Data Acquisition System for Alternate Fuels Engine Testing, Unpublished Report, Mech.Eng. Dept. - The University of British Columbia, May 1991.[38] Rohling, N.R., Operation and Performance of Emissions Console, Unpublished Report, Mech. Eng. Dept.- The Universityof British Columbia, August, 1990.[39] Diesel Engine Smoke Measurement, Society of AutomotiveEngineer, SAE Recommended Practice, 1991 Handbook, v3,SAE J255a.[40] Van Wylen, G.J. and Sonntag, R.E., Fundamentals of Classical Thermodynamics, Third Edition, Jone Wiey & Sons, Inc.,New York, 1985.[41] Measurement of Carbon Dioxide, Carbon Monoxide, and Oxides of Nitrogen in Diesel Exhaust, Society of Automotive Engineer,SAE Recommended Practice, 1991 Handbook, v3, SAE J177 APR82.[42] Ouellette, P. and Hill, P.G., Visualization of Natural Gas Injection for a Compression Ignition Engine, Society ofAutomotive Engineer, SAE Paper 921555, 1992.[43] Reynold, Wm.C., Chemical Equilibrium Solver Ver.3.60 - An Application Software, Mech. Eng. Dept, Stanford University,1987.[44] Van Zeggeren and Storey, The Computation of Chemical Equilibria, Cambridge University, Press, 1970.[45] Hill, P.G. and Peterson, C.R., Mechanics and Thermodynamics of Propulsion, Second Edition, Addison-Wesley PublishingCompany, Inc., 1992.[46] Song, S. and Hill, P.G., Dual Fuelling of a Pre-Chamber Diesel Engine, with Natural Gas, J.Eng for Turbines and Power,Trans. ASME, Vol.107 pp914-921, October 1985.[47] Gunawan, H., Performance and Combustion Characteristics of a Diesel-Pilot Gas Injection Engine, Unpublished Master Thesis,Department of Mechanical Engineering, UBC. 1992.178[48] Woschni, G., A Universally Applicable Equation for the Instantaneous Heat Transfer Coefficient in the Internal Combustion Engine, Society of Automotive Engineer, SAE Paper670931, 1967.179APPENDICESAPPENDIX A^NO FORMATION RATEThe mechanism of NO formation from atmospheric nitrogen aregenerally understood [26] [27]. It is generally accepted that incombustion of near stoichiometric fuel-air mixtures the principalreactions governing the formation of NO from molecular nitrogen are02^N2 = NO + NN + 02 = NO + 0N + OH = NO + HThe forward (kj and reverse (kir) rate constants for thesereactions are as follows:ki = (7.6E+13)exp(-38,000/T)kir = 1.6E+13k2 = (6.4E+9)T•exp(-3150/T)k2r = (1.5E+9)T-exp(-19,500/T)k3 = 4.1E+13k3r = (2E+14)exp(-23,650/T)where the units are cm3/mol-s.The rate of formation of NO via reactions (A.1) to (A.3) is180d[NO] dt -k1[0] [N2] -kir[NO] [N] +k2[N] [02]-k2r[NO] [0] +k3[N] [OH] -k3r[NO] [H]where the brackets denote concentrations in units of moles/cm3.A similar relation to (A.4) can be written for d[N]/dt:d[N] dt -k1[0] [N2] -kir[NO] [N] -k2[N] [021+k2r [NO] [0] -k[N] [H] +k3 [NO] [H](A.5)In order to apply Eq. (A.4), two approximations are introduced:1) N atoms change concentration by a quasi-steady process.2) The combustion and NO formation processes are decoupled, and theconcentrations of 0, 021 OH, H and N2 can be calculated by theirequilibrium values at the local pressure and equilibriumtemperature.The first approximation means that d[N]/dt can be set to zero andEq. (A.5) can be used to solve [N]. Substitute [N] expression intoEq. (A.4). The NO formation rate then becomes1- [NO] 2 / K[02] [N2] cl[AV]  -2k1[0] [N2]dt^i+kir [NO] / (k2[02] +k3 [OH] )(A.6)where K = (kik,r)(k2/k2r)•The second approximation means that one can simply use the(A.4)equilibrium composition (ie. [Ole, [021e1 [H]e, [HO]e and [N2]e) to181determine the concentrations of 0, 0 2 f H, HO and N,. It isconvenient to use the following notations for the one-wayequilibrium rates^= k,[0],[N2], = kir[NO]e[N],^for Eq. (A.1)^R2 = k2[N]e[02]e = k2r[N0],[0],^for Eq. (A.2)^R3 = k3[N],[011], = k3r[NO]e[H]e^for Eq. (A.3)where [ ], denotes equilibrium concentration.Substituting [Ole, [02]el [OH]e, [H], and [N2]e for [0], [02], [OH],[H] and [NJ in Eq. (A.6). The NO formation rate then becomesd[NO] _^241- ( [NO] / [NO],)2} (A.7)dt^1+ ( [NO] / [NO] 9) R1/ (R2+R3)Typical values of R, R1/112 and R1/(R2+R3) are given in Table A.1.The difference between R1/122 and R1/(R2+R3) indicates the relativeimportance of adding reaction (A.3) to the mechanism.Table A.1 Typical Values of 121, 111/R2 and R1/(112-1-R3):Equivalence ratio^R1^R1/R2^R,/(R2+R3)0.8^ 5.8E-05^1.2^0.331.0 2.8E-05^2.5^0.261.2^ 7.6E-06^9.1^0.14The strong temperature dependence of the NO formation rate can bedetermined by considering the initial term of d[NO]idt, when182[NO]/[N0],<<1. Then from Eq. (A.7),d[ArCA -2121=2/(1[0]8[ATjedt (A.8)The equilibrium oxygen atom concentration is given byK [0 1/2[0] -  P 2e^(RT) 1/2(A.9)where Kp is the equilibrium constant for the reaction(1/2)02 = 0and is given byKp = (3.6E+13)exp(-31,090/T)^atm"2^(A.10)Combining Eqs. (A.8), (A.9) and (A.10) withkl = (7.6E+13)exp(-38,000/T), the initial NO formation rate maythen be written asd[ 0]_6x1016 exp( -69 ' 090 )[02]dt^T1/21912 [N2] e (A.11)where the unit of d[NO]/dt is mol/cm3-s. The strong dependence ofd[NO]/dt on temperature in the exponential term is evident. Hightemperature and high oxygen concentrations result in high NOformation rates.183APPENDIX BEXHAUST EMISSION ANALYSIS SYSTEM BOPERATING PROCEDUREAs mentioned in Section 3.4, the exhaust emission analysis systemB (EEAS-B) mainly includes the heated sampling system and sixemission analyzers. The CH„ NO, CO and CO 2 analyzers use the NDIRmethod; the THC analyzer uses the FID method. The EEAS-B operatingprocedure includes starting procedure, analyzer calibration andemission measurement.1. The Starting Procedure1) Turn on the power switches for the sample line heater (1B)and the heated enclosure (5A). Set the temperatures of thesample line and heated enclosure both to 190 ° C for dieselengine (130 ° C for gasoline engine). The main power switchbox is located in the right side of the first cabinet fromback. The temperature control switch box for the sampleline is near the main power switch box. The temperaturecontroller of the heated enclosure is located on the frontpanel of the first cabinet.2) Switch on the power to the sampling pump (5B), chiller(7A), NO 2 to NO converter (8A) and the six emissionanalyzers. Allow a minimum 2 hours warmup if the system hasbeen turned off.1843) Check the sample gas pressure inside system. A pressuremeter is located on the front panel of the second cabinet.A pressure less than 6 psig indicates a large restriction.Therefore, the filter cores of the sample gas must bechanged.4) Open all the zero and span gas cylinders, the air andfuel (hydrogen) cylinders used by the THC analyzer.5) Set the flow rate for each analyzer (except THC analyzer)from the flow rate meters on the front panel of the secondcabinet. A typical setting is 4.5 SCFH for CH4 and NOanalyzer, 1.5 SCFH for CO2 and CO analyzers, and 3 SCFH for02 analyzer. Set the same sample flow rate for zero, spanand exhaust sample gases by adjusting the regulators ofthe zero and the span gases to match exhaust sample gas.6) For the THC analyzer only. Turn on the analyzer power, pumpand heater oven. Set the inlet pressure consistent forzero, span and sample gas at 200 mbar by adjusting theregulators of the zero and span gas as well as theregulator on the analyzer. Set the air pressure to 0.8 bar.Do not try to ignite the detection burner until thetemperature of the heater oven reaches 90°C. Push "H2-"button to adjust fuel (ie. hydrogen) pressure to about 0.5bar. Try to ignite the detection burner by holding down andthe "1.12 - " button, and pushing the "ignition" button. Thefuel pressure can be adjusted slightly higher to produce aricher fuel mixture to provide better ignition. After the185burner has ignited, set the fuel pressure back to 0.4 bar.Let it warm up.2. The Analyzer CalibrationAfter the temperatures of the sample line and heated enclosurereach 190°C and the emission analyzers have been warmed, startcalibrating the analyzers. Table B.1 shows the typical operatingranges and corresponding signal-output voltage ranges of sixanalyzers.Table B.1: Typical Operating Ranges of Emission Analyzers.Name of the analyzers Operating range^Output range( ppm)^B, (volts)Methane (CR4): 0^... 5,000 0^... 5Nitric oxide (NO): 0^... 3,000^0^... 5Carbon dioxide (CO2): 0^... 20^% 0^... 5Carbon monoxide (CO): 0^... 10,000^0^... 5(0^... 2500 for diesel baseline)Oxygen^(02): 0^... 21^% 0^... 5Total hydrocarbon (THC): 0^... 10,000^0^... 10(0^... 1,000 for diesel baseline)Table B.2 shows the typical zero and span gases for the analyzersand the typical concentrations for the span gases.186Table B.2: Typical Zero and Span Gases.Name of analyzer^Zero gas Span gas and^Output voltageMethane^(CH,): N2concentrationZ,CH„ 3470 ppmof the spanY,3.47 voltsNitric oxide (NO): N2 NO, 2113 ppm 3.522 voltsCarbon dioxide (CO2): N2 CO2, 19.3% 4.825 voltsCarbon monoxide (CO): N2 CO, 510 ppm 0.255 voltsOxygen^(02 ) : N2 Air, 20.9% 02 4.976 voltsTotal hydrocarbon (THC): N2 CH“ 3940 ppm 3.94 voltsThe following equation is used to calculate the signal-outputvoltage of the span for a specific analyzer, Y„ which is shown inTable B.2.(B.1)where A, is the maximum scale of the operating range (ppm) for thespecific analyzer; B, is the maximum signal-output voltage (volts)for the specific analyzer; Z, is the concentration of the span gas(ppm or %) for the specific analyzer.1^CH, and NOx analyzer:(1) Turn on the pump.(2) Switch on the zero gas valve; wait for one minute before187pushing the zero calibration button. The analyzer has aninside microprocessor to do the zero calibration. If theanalyzer does not zero, repeat this step again. The signaloutput for the zero gas should be zero volts or veryclose to zero. Switch off the zero gas valve.(3) Switch on the span gas valve and wait until the signal-output voltage becomes steady. Adjust the potential meteruntil the output voltage matches the signal-output voltageof the span for CH„ YcH„ (or for NO, YNO, refer to TableB.2 and Eq. (B.1)). Switch off the span gas valve.CO2 analyzer:(1) Switch on the zero gas valve and wait for one minutebefore undertaking zero calibration. Push the "ZERO" keyand the "ENTER" key to start zero calibration. Use the "1"or the "1" key to adjust the potential meter to chose avalue which is zero or very close to zero, then press the"ENTER" again to accept this value. If the analyzer doesnot zero, repeat this step again. The signal output forthe zero gas should be zero volts or very close to zero.Switch off the zero gas valve.(2) Switch on the span gas valve and wait until the readoutbecomes steady. Push the "SPAN" and the "ENTER" keys tostart span calibration. Use the "T" or the "1" key toadjust the potential meter to chose a value which is lessbut very close to the CO2 span gas concentration, then188push "ENTER" again to accept this value. The analyzer hasa microprocessor inside to calculate the span. You canrepeat this step again and chose a smaller value if theanalyzer displays an "ERROR" message. Switch off the spangas valve.3)^CO analyzer:(1) This analyzer is the oldest one in this system. In orderto do the calibration easily and precisely, the spancalibration must be done first, following the zerocalibration. Repeat those two steps. Using this procedure,one will find that it is easier to achieve the desiredcalibration results.(2) Switch on the span gas valve and wait until the signal-output voltage becomes steady. Adjust the span knob to setthe output to the signal-output voltage of the span forCO, Yco (refer to Table B.2). Switch off the span gasvalve.(3) Switch on the zero gas valve and wait until the signal-output voltage becomes steady. Adjust the zero knob to setthe output voltage to zero. Switch off the zero gas valve.4^02 analyzer:(1) Push ".111" and " MEAS./CAL." into calibration mode frommeasurement mode.(2) Switch on the zero gas valve and wait until the signal-189output voltage becomes steady. Push "5" and "ENTER" andwait for the microprocessor of the analyzer to do the zerocalibration. If the analyzer does not zero, repeat thisstep. The signal output for the zero gas should be zerovolts or very close to zero. Switch off the zero gasvalve.(3) Switch on the span gas valve and wait until the signal-output voltage becomes steady. Push "8" and "ENTER" andwait for the analyzer to do the span calibration. If theanalyzer does not span well, repeat this step again. Thespan gas concentration should always be 20.9% oxygen.Switch off the span gas valve.(4) Push "MEAS./CAL." back to the measurement mode.Total HC analyzer:(1) Place the three-way valve in the zero gas position andwait until the signal-output voltage becomes steady.Adjust the zero knob to set the output voltage to zero.Switch the valve to sample position.(2) Switch the three-way valve to span gas position and waituntil the signal-output voltage becomes steady. Adjustfuel (ie. hydrogen) pressure to about 0.4 bar and look atthe values on the indicator. Set a fuel pressure whichcorresponds with the peak value on the indicator. Adjustthe span knob to set the output to the signal-outputvoltage of the span for THC, Yvic ( refer  to Table B .2 ) .190Switch the valve to sample position.3. The emission measurementAll the emission measurements for this study are based on steady-state operating conditions, which are different from EPA HighwayDriving Cycle or Transit Cycle. The reason for using steady-stateoperating condition is that proceeding EPA Transit Cycle needs verysophistical equipments that we do not have. The following is themeasurement procedure.1) Start the engine and warm it up under load until thetemperatures of the lubricate oil and the cooling waterboth reach 80°C.2) Calibrate every analyzer and make the system ready to work.3) Operate for at least 10 min in each test mode, record theemission data until the engine runs steady and the monitorshows steady emission data.4) Check and record the zero and span offset after each test.Turn off all the gas cylinders and some analyzers.191APPENDIX CPOWER CORRECTION FACTOR CALCULATION METHOD(in accordance with SAE J1349)1. The correction is made against the standard inlet airconditions, i.e.,Inlet Air Pressure (Absolute)^: 100kPaInlet Air Temperature^: 25 C(298K)Dry Inlet Air Pressure (Absolute) : 99kPa2. The correction factor fcorr applied to the observed brake powerdepends on the atmospheric factor fa and engine factor fm whichis calculated using the empirical relationship:Corr_ ( fa) fmThe atmospheric factor is calculated based on dry inlet airpressure Iiido and inlet air temperature t,fa= [99/Bdo] [( t+273)/298]°7The engine factor depends on the fuel flow F(g/s), the enginedisplacement D(dm3), the engine speed N(RPM), and the pressureratio r of inlet manifold pressure P. to inlet air pressure Bo.It has the following value:fm = (0.036 q/r) - 1.14^if 40 < (q/r) < 65fm = 0.3^ if (q/r) < 40fm = 1.2 if (q/r) > 65where q = (60,000 F)/(D N), for a two-stroke cycle engine.3.^^The correction factor used in the calculation is within therange of 0.90 to 1.10.192APPENDIX DDETERMINATION OF SPECIFIC HUMIDITYSpecific humidity H (g of H20 per kg of dry air) of an air-watervapour mixture is defined as the ratio of the mass of water vapormc (kg) to the mass of dry air ma (kg) [40].mH=1000--tma(D.1)The equations of state both for water vapor and for dry air can beexpressed in terms of partial pressure, volume, mass of substanceand temperature.PwV=mwRwT^ (D .2 )PaV=maRaT^ (D.3)where subscripts "w" and "a" denote water vapor and dry airrespectively. R, = 461.52 (J/kg-K) is the gas constant of watervapor. Ra = 287.0 (J/kg-K) is the gas constant of dry air.Eq. (D.2) divided by Eq. (D.3), thenm R___Lv-,(_02) (._.P ')ma Rw Pa(D.4)where Ra/Rw = 0.6219.Substitute Eq. (D.4) into Eq. (D.1):193H=621.9-^ (D.5)PaRelative humidity 4 is defined as a ratio of the partial pressureof water vapor Pw to the saturated pressure of water vapor at drybulb temperature P,.(D.6)Wet-air pressure (ie. ambient pressure or barometric pressure) Pbis defined as the sum of the partial pressure of dry air Pa and thepartial pressure of water vapor P.Pb= Pa + 2wSubstitute Eqs.(D.6) and (D.7) into Eq. (D.5):H621.9^Pb-4)Ps(D.7)(D.8)Barometric pressure Pb (inHg), relative humidity (1) (%) and dry bulbtemperature Td (°F) are measured directly from the gauges. P, (inhg)can be expressed in terms of Td (°F) in following equation, whichis the fitting curve equation of the Saturated Steam TemperatureTable [40] from 40 to 140 °F.Ps=-0 .461920+0 .029439 Td-0.00040Td2+0.000004Td3^(D.9)194APPENDIX EA MORE EXACT FORMULAFOR DRY—WET BASIS CONVERSION FACTORThe definitions that are used in the formula of the dry-wet basisconversion factor (E1w) are the following:Pilot-diesel fuel to CNG fuel mass ratio, r:The ratio of pilot-diesel mass flow rate Masi, to CNG mass flowrate itcla3 . Both of them are measured data.(E. 1)Atomic hydrogen-to-carbon ratio of the combined fuel, y: The ratio of hydrogen to carbon atoms of a combined fuel (CH)which is used to replace CNG (CH3.80 and diesel (CH1.8) fuels.3.85+ 16689 r (1.8)13.825 m-C^1+16.689r13.825 m(E.2)The more exact formula of the dry-wet basis conversion factor (Fdjis based on the following assumptions:1951. Considering the effect of the humidity of the intake air onthe reaction of fuel combustion, the reactants of thecombustion are combined fuel, air and water vapour in theintake air.2. Consider a real situation that the combined fuel is not 100%burned. The unburned fuel in the exhaust (which has the samecomposition as combined fuel) will affect the water vapour inthe exhaust. Thus the products of the combustion are carbondioxide, water vapour, oxygen, nitrogen and unburned fuel.A general equation of the combined-fuel combustion can be definedbased on above assumptions.Chry+n(02+3.761V2)+10120==>(1-x)CO2+[17(1-x)+m]ll202+[n -(1+17-) (1 -A7)]192+3.76rik+AtYly4where y is the atomic hydrogen-to-carbon ratio of the combinedfuel; n is the number of moles of oxygen in intake air; m is themoles of water vapour in 4.76n moles of dry air; x is the number ofmoles of unburned fuel in the exhaust. The following is thecalculation procedure for the variables n, m and x.1). Evaluation of n and m:From Eq. (E.3), the mass flow rate of combined fuel and intake aircan be determined.Mass of combined fuel per unit time, F=12.011+1.008y(E.3)196Mass of intake air per unit time, A=31.998n+105.330n+18.015m=137.328n+18.015m. Then,F/A- 12.011+1.008y 137.328n+18.015m (E. 4 )where F and A can be measured; y can be determined from Eq. (E.2).Referring to the general Eq. (E.3), the molal ratio of water vapourto dry air in the intake air can be expressed in terms of thespecific humidity H (g H20 / kg of dry air) which can be determinedby Eq. (D.8) in Appendix D.m 4.76n -1.608x10-3H (E.5)Solve for Eq. (E.4) and Eq. (E.5), n and m are determined,n-( A )(  12.011+1.008y B^137.328+0.1379H )andm=7 .654x10-3Hn(E.6)(E.7)2). Evaluation of x:Referring to Eq. (E.3), the total moles of the combustion productsis nT=x(1-y/4)+y/4+m+4.76n. The molal ratio of the unburned fuel tothe combustion products can be expressed in terms of a measuredvalue of total hydrocarbon z (ppm) in the wet-basis.197^-zx1 0 -6x(1-X)+X+m+4.76n4 4Thus, x can be determined by(17+m+4.76n)z4 x-106-(1-X)4(E.8)The volumetric fraction (or molal fraction) of the water vapour inthe exhaust can be expressed in terms of the variables n, m and x. y(1-x)  +n72 (E.9)x(1-X) +X +m+4 .76n4 4The more exact formula of the dry-wet basis conversion factoris,Fcbir=3- — X H20^ (E.10)2826242220181614121042198APPENDIX F REPEATABILITY OF THE TEST RESULTS1200 RPM, 80 BAR, 25%CN62, 200EG/30%SRD+4f-JAili1-*0^04^08^12^1^2^24^28BMEP (Bar)^0 DATA-DECO2,92^+ DATA-DEC04,92Figure F.1: Repeatability of Thermal Efficiency.(at BOI=24°BTDC)Repeatability of engine performance and exhaust emissions at agiven test condition are shown in Figure F.1 through Figure F.6.Tests were conducted in two different days. The exhaust emissiondata presented are on the wet-basis. The test condition is: 1200rpm speed, 60 bar CNG injection pressure, 25% cetane 62 pilot-diesel, 200 injection angle, 30% SRD and 24° BTDC of BOI.5004003300620010049E1El1200 RPM, 60 BAR, 25%CN62, 20DEGJ3096SRD199The following factors are consider to affect the repeatability ofdifferent-day data: gas-diesel fuel injector reliability,calibration off-set and detection error of analyzers (orinstrumentations) and manual error in the process of data taking.0^04^0 B^12^16^2^24^2BMEP (Bar)^0 0ATA-DECO2,92^+ DATA-DEC04,92Figure F.2: Repeatability of Nitrogen Oxides Emissions.(at BOI=24°BTDC)Figure F.1 shows the repeatability of thermal efficiency data. itappears that the thermal efficiency data of Dec.04 are slightlylower than that of Dec.02 over whole load range. The maximumdifference is about 13% at BMEP - 0.5 bar.0 04 08 12 1 2 24 285404F4qin1200 RPM, 60 BAR, 25%CN62, 20DEG/30%SRD200BMEP (Bar)DATA-DECO2,92^+ DATA-DEC04,92Figure F.3: Repeatability of Total Hydrocarbon Emissions.(at BOI=24°BTDC)The repeatability of nitrogen oxides emissions is shows in FigureF.2. There are quite big differences between two-day data at lowand high load. They are quite close to each other at medium load.The maximum difference is about 30% at BMEP - 0.5 bar. For 24° BTDCof BOI, the operating load range is BMEP from 0 to 2.5 bar, so thatBMEP - 2.5 bar is the high load. Figure F.3 shows the repeatabilityof total hydrocarbon emissions. The maximum difference is about 19%at BMEP - 0.5 bar.89 -foar0.52521.543.531200 RPM, 60 eAR, 25%CN62, 200EG/3096SRD2010^04^0^1 2^6^2^24^28BMEP (Bar)^0 OATA-DECO2,92^+ DATA-DEC04,92Figure F.4: Repeatability of Unburned Methane Emission.(at BOI=24°BTDC)It can be seen in Figure F.4 that the quite big differences betweentwo-day unburned methane data are at low and high load. Thedifferences are 21 % and 50% at BMEP - 0.5 bar and BMEP - 2.5 barrespectively.Figure F.5 shows the repeatability of carbon monoxide emission. Themaximum difference is at BMEP - 2.5 bar, which is about 17%. Carbondioxide emission has the best repeatability in whole data. As shown4 1200 RPM, BO EAR, 25%CN62, 200EG/3SRD2023.532 521.50.50 04 08 12 1 2 24 28BMEP (Bar)0 DATA-DECO2,92^+ DATA-DEC04,32Figure F.5: Repeatability of Carbon Monoxide Emission.(at BOI=24°BTDC)in Figure F.6, the maximum difference is 9% at BMEP - 0.5 bar.Summarizing from above, we know that the low (BMEP - 0.5 bar) andthe high load (BMEP - 2.5 bar) are two regions where create moreerrors. The cycling variations of the low-load data resulted byunsteady operation produce more error in the process of data takingat low load. The unsteady operation caused by load limitation isthe reason of creating error at high load.43.83.63 RPM, 60 BAR, 25%CN62, 20DEG/3SRD2030^04^08^12^16^2^2 4^2BMEP (Bar)^0 0ATA-DECO2,32^+ DATA-DEC04,92Figure F.6: Repeatability of Carbon Dioxide Emission.(at BOI=24°BTDC)In the determination of the best BOI performance curve, two placesmay include more errors, which are very low-load region (BMEP - 0.5bar) and very high-load region (BMEP 4.4 bar). The goodrepeatability can be obtained over the rest of the load range.204APPENDIX GPHOTOGRAPHS OF THE VISUALIZATION RESULTSOF NATURAL GAS INJECTIONThe four photographs presented in the following is from the flowvisualization results of natural gas injection [42] and courtesy ofPatric Ouelette. The configuration of the gas injector used inOuelette's flow visualization work is similar to the gas-dieselfuel injector used in engine test.Figure G.1: Free Conical Sheet Jet with 10° Injection Angle.(Pressure Ratio of 2; Lift of 0.056 mm)205Figure G.2: Interrupted Conical Jet with 100 Injection Angle.(Pressure Ratio of 2; Lift of 0.056 mm)Figure G.1 shows the free conical sheet jet with 10° injectionangle. The free conical sheet jet has the same condition as thegas-diesel fuel jet with 0% interruption ratio (or 0% SRD). Itappears that the free conical sheet jet with 10° injection anglehas very strong tendency of the top wall clinging, which is socalled top wall effect.Figure G.2 shows the interrupted conical jet with 10° injectionangle. The interrupted conical jet has the same situation as thegas-diesel fuel jet with interruption. Compared with Figure G.1,206Figure G.2 shows that the interruption of the conical sheet jetreduces the top wall effect and increases the jet penetration.Figure G.3: Free Conical Sheet Jet with 200 Injection Angle.(Pressure Ratio of 5; Lift of 0.056 mm)Figure G.3 and G.4 show the free conical sheet jet and theinterrupted conical jet with 20° injection angle respectively. Thefree conical sheet jet with 20° injection angle has the tendency ofcollapsing and the interrupted conical jet with 20° injection angledo not.The conclusion from above four photographs is that the interruptionof the conical sheet jet can increase the stability and penetrationFigure G.4: Interrupted Conical Jet with 200 Injection Angle.(Pressure Ratio of 5; Lift of 0.056 mm)of fuel jet.207208APPENDIX H^PROGRAM #11. Input Parameters*Engine inlet temperature, TO.*Engine inlet pressure, PO.*Engine air-box pressure, Pl.*Engine compression ratio, E.*Specific heat ratio (or polytropic constant), N.2. List of the PROGRAM #1THIS PROGRAM IS MADE TO FIND THE UNBURNED GAS TEMPERATUREAND PRESSURE (UNBURNED GAS INCLUDES FUEL, FRESH AIR ANDRESIDUAL GAS FROM PREVIOUS CYCLE).IMPLICIT REAL(N)OPEN(UNIT=3,FILE='B:\PROJOl\PROJ01.DAT ',STATUS='UNKNOWN')1^PRINT 22 FORMAT (1X,'PLEASE MAKE YOUR CHOICE:')PRINT 33^FORMAT(1X,'0==>QUIT THE PROGRAM'1^/1X,'1==>CONTINUE THE PROGRAM')READ *, JIF (J.EQ.0) THENGOTO 140ELSE IF (J.EQ.1) THENGOTO 4ELSEGOTO 1END IF4 PRINT 55 FORMAT (1X,'PLEASE INPUT ENGINE INLET TEMPERATURE (K)')READ *, TOPRINT 66 FORMAT (1X,'PLEASE INPUT ENGINE INLET PRESSURE (atm)')READ *, PO7 PRINT 88 FORMAT (1X,'PLEASE INPUT ENGINE AIR-BOX PRESSURE (atm)')READ *, P1IF (P1.LT.P0) THENGOTO 7END IF209PRINT 99^FORMAT (1X,'PLEASE INPUT ENGINE COMPRESSION RATIO')READ *, EPRINT 1010^FORMAT (1X,'PLEASE INPUT SPECIFIC HEAT RATIO',1^1X,'(OR POLYTROPIC CONSTANT)')READ *, NWRITE (3,15) TOPRINT 15, TO15^FORMAT (1X,'ENGINE INLET TEMPERATURE IS',1X,F7.2,1X,'(K)')WRITE (3,16) POPRINT 16, PO16^FORMAT (ix, 'ENGINE INLET PRESSURE IS',1X,F5.2,1X,'(atm)')WRITE (3,17) P1PRINT 17, P117^FORMAT(1X,'ENGINE AIR-BOX PRESSURE IS',1X,F5.2,1X,'(atm)')WRITE (3,18) EPRINT 18, E18^FORMAT (ix, 'ENGINE COMPRESSION RATIO IS',1X,F5.2)WRITE (3,19) NPRINT 19, N19^FORMAT(1X,'SPECIFIC HEAT RATIO',1^'(OR POLYTROPIC CONSTANT) IS',1X,F4.2)C^MAIN PROGRAM:C CALCULATE THE UNBURNED GAS TEMPERATURE AND PRESSUREA=(N-1)/NB=(P1/P0)**AT1=T0*(1+(B-1)/0.75)WRITE (3,100) TiPRINT 100, Ti100^FORMAT (1X,'ENGINE AIR-BOX TEMPERATURE IS',1 1X,F7.2,1X,'(K)')T2=T1*(E**(N-1))P2=P1*(E**N)WRITE (3,110) P2PRINT 110, P2110^FORMAT(1X,'UNBURNED GAS PRESSURE IS',1X,F5.2,1X,'(atm)')WRITE (3,120) T2PRINT 120, T2120^FORMAT (ix, 'UNBURNED GAS TEMPERATURE IS',1X,F7.2,1X,'(K)')WRITE (3,130)PRINT 130130^FORMAT (1X,'^,)GOTO 1140^CLOSE (UNIT=3)END210APPENDIX I^PROGRAM #21. Input Parameters*Relative air-fuel ratio, Lambda.*Residual molal fraction, FRES.*Atomic hydrogen-to-carbon ratio of fuel, RHC.2. List of the PROGRAM #2THIS PROGRAM IS MADE TO FIND THE FINAL UNBURNED GASCOMPOSITIONS (INCLUDING FUEL, FRESH AIR AND RESIDUAL GASFROM PREVIOUS CYCLE) INSIDE THE ENGINE CYLINDER AFTERCERTAIN CYCLES OF RUN. ASSUME BOTH THE RESIDUAL MOLALFRACTION AND THE RELATIVE AIR-FUEL RATIO REMAIN CONSTANTFOR EACH CYCLE.IMPLICIT REAL(M)OPEN (UNIT=3,FILE='B:PROJ03.DAT',STATUS="UNKNOWN')1^PRINT 22 FORMAT (1X,'PLEASE MAKE YOUR CHOISE:')PRINT 33^FORMAT(1X,'0==>QUIT THE PROGRAM'1^^/1X,'1==>CONTINUE THE PROGRAM')READ *, JIF (J.EQ.0) THENGOTO 340ELSE IF (J.EQ.1) THENGOTO 4ELSEGOTO 1END IF4^PRINT 55 FORMAT (1X,'PLEASE INPUT RELATIVE AIR-FUEL RATIO')READY = LAMBDA, IS RELATIVE AIR-FUEL RATIO10^PRINT 1515 FORMAT (lx, 'PLEASE INPUT RESIDUAL MOLAL FRACTION')READ *, FRESWRITE (3,25) YPRINT 25, Y25^FORMAT (lx, 'RELATIVE AIR-FUEL IS',1X,F4.2)WRITE (3,35) FRES211PRINT 35, FRES35^FORMAT (lx, 'RESIDUAL MOLAL FRACTION IS',1X,F4.2)WRITE (3,40) RHCPRINT 40, RHC40^FORMAT (1X,'H TO C ATOMIC RATIO OF THE FUEL',1X,F6.3)WRITE (3,45)PRINT 4545^FORMAT(1X,'CHn+2Y(02+3.76N2)1^+(9.52YR)(A*CO2+B*H2O+C*02+D*N2))WRITE (3,55)PRINT 5555^FORMAT(10X,'==>E*CO2+F*H20+G*02+H*N21)CALCULATION OF THE RESIDUAL-AIR MOLAL RATIO, R:100^R=FRES/(1-FRES)THE BASIC CYCLE (WITH RESIDUAL GAS OF 21% 02 AND 79% N2):THE REACTION EQUATION IS:CHn+2Y(02+3.76N2)+(9.52YR)(A*CO2+B*H2O+C*02+D*N2)=>E*CO2+F*H20+G*02+H*N2^(n=RHC)200^MRES=9.52*Y*RAA=0BB=0CC=0.21DD=0.79EE=l+MRES*AAFF=RHC/2+MRES*BBGG=2 *Y-1-RHC/4+MRES*CCHH=7.52*Y+MRES*DDMTOTAL=RHC/4+9.52*Y*(1+R)WRITE (3,210)PRINT 210210^FORMAT(4X,'A',8X,'B',8X,'C',8X,'D',8X,'E',8X,'F',1 8X,'G',8X,'H')WRITE (3,215)PRINT 215215^FORMAT(1X,'THIS IS BASIC CYCLE')WRITE (3,220) AA,BB,CC,DD,EE,FF,GG,HHPRINT 220, AA,BB,CC,DD,EE,FF,GG,HH220^FORMAT(8(1X,F8.5))MAIN PROGRAM:CALCULATE COMPOSITIONS OF COMBUSTION REACTANTS AND PRODUCTS300^I=1305 A=EE/MTOTALB=FF/MTOTALC=GG/MTOTALD=HH/MTOTALE=l+MRES*A212F=RHC/2+MRES*BG=2*Y-1-RHC/4+MRES*CH=7.52*Y+MRES*DWRITE (3,315) IPRINT 315, I315^FORMAT (lx, 'THIS IS CYCLE',1X,I4)WRITE (3,325) A,B,C,D,E,F,G,HPRINT 325, A,B,C,D,E,F,G,H325^FORMAT(8(1X,F8.5))A1=ABS(AA-A)/AIF (A1.GT.0.001) THENAA=ABB=BCC=CDD=DEE=EFF=FGG=GHH=HI=I+1GOTO 305END IFMCO2=EMH20=FMO2=GMN2=HWRITE (3,330) MCO2,MH20,MO2,MN2PRINT 330, MCO2,MH20,MO2,MN2330^FORMAT (1X,'MCO2=',F8.5,1X,'MH20=',F8.5,1X,'MO2=',F8.5,1 1X,'MN2=',F8.5)WRITE (3,335)PRINT 335335^FORMAT (1X,'^,)GOTO 1340^CLOSE (UNIT=3)END213APPENDIX JPROCEDURES FOR CALCULATING THERMODYNAMIC PROPERTIESThe following are the preparing procedures of the thermodynamicproperties of the unburned gas in three different cases:Case 1, the unburned gas only involves fresh air and natural gasfuel:1. Input the experimental data (inlet pressure Po, inlettemperature To and air-box pressure P1) into PROGRAM#1 tocompute the unburned-gas pressure P2 and temperature T2.2. Input the unburned-gas compositions, P2 and T2 into thenon-combustion function of the STANJAN to compute theunburned-gas thermodynamic properties which are U2f H„ S2 etc.Therefore, P2 and H2 are known.Case 2, the unburned gas contains fresh air, residual gas andnatural gas fuel:1. Use same procedure as the step 1 of case 1 to find T2 and P2.2. It is known that the unburned gas contains fuel, 02, N„ CO2and H20. Then, divide fuel (ex. CH4), 02 and N2 into Group A andCO„ H20 into Group B.3. Use same procedure twice as the step 2 of case 1 to calculatethe enthalpy of unburned gas, as H2A for Group A and as H28 forGroup B.4. Calculate the mass, mA and mL for Group A and Group B214respectively.5.^Use the following equation to calculate the total enthalpy ofthe unburned gas, H2AH _  H 2 m A +H 2.a- B2MA +MB(J.1)Therefore, P2 and H2 are known.Case 3, the unburned gas contains fresh air and diesel fuel:1. Use same procedure as the step 1 of case 1 to find T2 and P2.2. Put diesel fuel (CH18) in Group A and air in Group B.3. Calculate H2A, the enthalpy of gaseous diesel at state "2" inFig. 6.1 in chapter 6, with the following equationsH =H =H° +AH,2A Ta^298K^.2 ( J.2 )whereAH2.2= [2327.6 (T2)2+10418 (T2) -51714.3] /12^(J.3)andli°2981e-1852.4 (kJ/kg)where H°298A is the enthalpy of formation of the diesel (CH18)which is evaluated from HHV=-45220 kJ/kg for liquid diesel.vaporization of the diesel has already counted in. Then calculatethe mass of diesel, mA.4. Use same procedure as the step 2 of case 1 to calculate theenthalpy of Group B (ie. air), H2B. Then the mass of Group B,MB •5.^Determine the total enthalpy of the unburned gas, H2, byEq. (J.1). Therefore, P2 and H, are known.215216APPENDIX KPROCEDURE FOR COMPUTATION OF EQUILIBRIUM COMPOSITIONThe calculation procedure of the equilibrium compositions of thecombustion produces is as follows:1. Obtain the test engine specification data and experiment data.The following is the data needed for later calculations:*Engine inlet temperature, To.*Engine inlet pressure, Po.*Engine air-box pressure, P,.*Engine compression ratio, E.*The actual range of the relative air-fuel ratio whichcorrespond to actual load range of the engine.*Engine residual molal fraction, Fres.*Specific heat ratio, k (or polytropic constant, n ).*Engine exhaust temperature.*Atomic hydrogen-to-carbon ratio of the fuel.2. Calculate the unburned-gas temperature, T2 and pressure, P2with the PROGRAM#1 (as explained in Section 6.4, 6.5 andAppendix H).3. Determine the unburned-gas compositions with the PROGRAM#2(as mentioned in Subsection 6.3.2 and Appendix I).4. Evaluate the thermodynamic properties of the unburned gaswith the non-combustion function of the STANJAN. Only H, andP2 are needed for later calculation.5.^Compute the burned-gas equilibrium compositions of the217constant pressure and adiabatic combustion process with thecombustion function of the STANJAN by inputting the unburned-gas compositions, P2 and H2.6.^Determine the exhaust temperature of MASD cycle only when itis necessary.IM air MatraplTabox(rYlatrap+ M ifuel )/ (flair -I-M fuel)F - 7 iTexhOn air - Matra aboxm a 1r, Tabox(a) <b)M exh, TexhCYLINDERMIXING POINT218APPENDIX LEVALUATION OF THE RESIDUAL TEMPERATUREFigure L.1: Schematic of Residual Temperature Evaluation Model.Figure L.1 shows a schematic of the residual temperatureevaluation model. After inlet port opening, most combustionproducts are pushed out from the cylinder by intake air. Part ofthe intake air passes the cylinder with combustion products, whichis defined as passed air. We assumes that there is a mixing pointin the exhaust pipe and there is no mixing and heat exchange before219this point; combustion products and passed air mix adiabaticallywith constant pressure into exhaust gas at the mixing point. Thatis, before mixing, the combustion products remain at the sametemperature as the residual gas in the cylinder and the passed airkeeps the air-box temperature; after mixing, all the mixture hasexhaust temperature which can be measured.For the mixing point, we can write the mass conservation andenergy conservation equations as following:(Inatrap+M fuel) + (Blair—ma trap) = (Inair+Mfuel)^(L.1)and(Matraplinfuel) CpresTres+ (Mair—matrap) CpairTabaa^(L.2)= (Mair+Mfuel) CpexhTeichwhere mair, Matrap and Mfuel are the delivered air mass, mass of the airtrapped and total mass of the fuel per cycle respectively. Cpair,Cpres and Cpexh are the constant-pressure specific heat of the air,the residual gas and the exhaust gas respectively. Tabox, Tres and Texhare the temperature of the air box, the residual gas and exhaustgas respectively.From the energy conservation Eq. (L.2), we can determine theresidual temperature byT -(r11air +III^)fuel- Cpeuct2Texh— (Mair—ma ^CpairTabcutIOC^Onatrap+m fuel ) Cprat,(L.3)Substituting Matrap = DP X Mtrap and mfuel = m_CNG + m_DSL intoEq. (L.3), we can have_  (mair+Incivc+rapsd CpexhTexh- (filair-DP"Itrap) CpairTaboxTnm(DPXR1tr9p+In+Man) Cpres(L.4)220where DP is the degree of purity of the charge. mtcap is the trappedmass. MeNG and MDSL are the mass of the CNG and the diesel fuels.221APPENDIX M WORK SHEET OF THE PROGRAM XPRESSDXPRESSD.WK1/JULY 05, 93.Vipc/VTDC: 12.96558TEST CONDITION:1) 1200 RPM, 60 BAR, 20DEG. ANGLE (FOR GAS-DIESEL OPERATION).2) 1200 RPM DIESEL BASELINE.BLOCK 1: Intake and exhaust parameters.Speed^Ambient^Gage^Ambient^Air Box Air Box Exhaust## Pressure Pressure Temp. Pressure Temp.^Temp.(RPM) (kPa) (kPaG)^(K) (kPa)^(K) (K)1)^40% SRD, 25% CN62 (Humidity Ratio: 9.0 gH20/kgAIR):1 1212 102.32 24.4254^305.2485^126.7454^346.2125 5272 1215 102.32 23.90182 307.0763 126.2218 347.4647 5603 1193 102.32 25.65576 307.1113 127.9757 350.2453 5932) NON-EGR, 0% SRD,^25% #2^(Humidity Ratio:^11.37 gH20/kgAIR):1 1234 101.65 24.16911 314.4150 125.8191 354.1821 6862 1219 101.65 24.99256 314.9412 126.6425 355.6633 6883 1236 101.65 23.82741 315.4908 125.4774 356.2527 6654 1218 101.65 23.52503 316.3694 125.1750 356.2569 6185 1214 101.65 23.4193 316.7745 125.0693 355.6569 6346 1220 101.65 23.48546 316.8956 125.1354 356.2425 6337 1236 101.65 24.16105 317.0172 125.8110 357.7293 6788 1197 101.65 23.94655 317.2572 125.5965 357.7263 6249 1217 101.65 23.22568 317.3026 124.8756 356.2331 64110 1230 101.65 23.03895 317.3101 124.6889 356.6492 62211 1206 101.65 24.31713 317.3135 125.9671 357.4188 6573) EGR,^0% SRD,^25% #2^(Humidity Ratio:^13.26 gH20/kgAIR):1^1216^101.64^24.25391^317.3723^125.8939^371.1421 7202 1227 101.64 24.0868 317.3965 125.7268 374.2451 7103 1196 101.64 25.2216 317.4268 126.8616 377.4058 6944 1205 101.64 23.96791 317.4516 125.6079 380.2706 6935 1223 101.64 23.23388 317.4763 124.8738 380.4395 6856 1193 101.64 24.41961 317.4793 126.0596 381.2838 6737 1214 101.64 23.97048 317.4903 125.6104 383.302 7078 1222 101.64 24.31868 317.515 125.9586 385.6313 7069 1228 101.64 23.80001 317.5391 125.4400 385.885 6654) Baseline (Humidty Ratio: 4.52 gH20/kgAIR):1 1205 100.6^25 315 125.6 358.8775 5502 1222 100.6 23.3 316 124 357.3976 5803 1188 100.6 25 316 125.6 360.0168 600222BLOCK 2: Input and exput parameters.Air^Lig.fuel^CNG^Inject.^Inject.^Corr.^Corr.^ThermalMassflow^dm/dt^dm/dt^PW^BOI^Power^MEP Effic.(kg/hr)^(kg/hr)^(kg/hr) (°BTDC)^(kW)^(kPa)^(%)1) 40% SRD,^25% CN62^(Humidity Ratio:^9.0 gH20/kgAIR):113^0.372602^1.033945^8^24^4.590035 1.965267^24.4186116^0.485392^1.329538^10^32^7.185184 3.050826 29.53114113^0.63034^1.766822^13^32^9.479388 4.117704^29.484042) NON-EGR, 0% SRD, 25% #2 (Humidity Ratio:^11.37 gH20/kgAIR):109 0.648701^1.848526 11^28^5.798951 2.425195^17.38516107 0.667318^1.970625 11 28 7.210735 3.053528 20.46398109 0.765553^1.981506 12 28 8.330941 3.479745 22.72406106 0.552986^1.59209 10 32 5.946683 2.520408 20.72375105 0.571115^1.575998 10 32 7.25618 3.0863 25.28698106 0.660883^1.6943 11 32 8.427998 3.56694 26.80346107 1.010106^2.328081 5 32 9.494095 3.965298 21.33126102 0.495288^1.513703 12 36 5.703078 2.459298 21.19061108 0.54983^1.672513 12 36 7.11388 3.01803 23.91148109 0.589926^1.664481 12 36 8.37678 3.516791 27.77336105 0.79208^1.99166 14 36 9.324703 3.992163 25.080963) EGR,^0% SRD, 25% #2^(Humidity Ratio: 13.26 gH20/kgAIR):87.7 0.735073 1.862798 11 28 6.157891 2.614098^17.584588.8 0.782023 2.016833 12 28 7.215641 3.034691 19.1189583.4 0.749656 2.057158 12 28 8.208283 3.542856 22.0105687.9 0.569808 1.6342 11 32 6.00669 2.573048 20.1717389.6 0.69326 1.727946 12 32 7.341525 3.099348 22.4989584.7 0.73886 1.827525 12 36 8.129796 3.518398 23.5524188.1 0.560663 1.633711 12 36 5.974646 2.540025 20.1524192.4 0.587615 1.812251 12 36 7.15562 3.023678 22.0502390.9 0.60977 1.833738 13 36 8.623358 3.625296 26.089554) Baseline: (Humidty Ratio: 4.52 gH20/kgAIR):101.1^1.9231 0^9 12^4.63 1.98 19.1101.9 2.3399 0 12 12 8.22 3.48 27.894.9 2.9997 0 13 12 9.25 4.02 24.4223BLOCK 3: Exhaust emissions.CO^CO2^NOx^02^THC^H20(wet)^N2[wet]^[wet]^[wet]^[wet]^[wet]^[wet]^[wet](ppm)^(%vol)^(ppm)^(%vol)^(ppm)^(%vol)^(ppm wet)1) 40% SRD, 25% CN62 (Humidity Ratio: 0.0090):981.4663^2.775269^207.5841^15.19777^1345.601516.4374^3.465667^648.5362^13.82398^820.41022530.389^4.703483^825.9376^11.16361^850.37222) NON-EGR, 0% SRD, 25% #2 (Humidity Ratio:^4.674310^77.099185.746264^76.765548.033601^75.6786211.37 gH20/kgAIR):5136.23 4.828243 108.9531 9.840656 3267.516 8.630271^75.849555437.075 4.871955 126.206 9.636355 2646.596 8.776316 75.894385846.996 4.775346 241.8171 9.6462 2366.01 8.510363 76.222601928.955 4.012528 135.6853 12.28996 3171.17 6.916358 76.257562051.331 4.462478 248.7756 11.47011 1934.925 7.636873 76.007022939.885 4.750616 435.6068 10.80065 1204.173 8.134726 75.856046746.935 4.886815 376.0823 8.735898 3446.818 8.647666 76.672631356.98 3.79284 162.0435 12.72445 3666.546 6.532228 76.431921655.15 4.511855 194.582 11.38461 2575.331 7.77086 75.890161614.746 4.560831 319.5783 11.32321 1105.928 7.77232 76.039603609.03 5.209126 534.3126 9.900576 1625.96 8.93886 75.374503) EGR, 0% SRD, 25% #2 (Humidity Ratio: 13.26 gH20/kgAIR):5457.311 4.768828 68.85998 9.949586 3236.405 8.430988 75.974336142.453 4.630793 102.7665 9.778838 3097.733 8.300078 76.355996048.575 4.351735 166.7753 10.35601 2533.54 7.895495 76.521863261.598 4.218713 81.30516 11.69213 2871.175 7.40756 76.060184081.285 4.505443 106.5993 11.05431 1838.235 7.828541 76.009084841.933 4.543085 214.455 10.73595 1680.361 7.962108 76.085182884.738 4.42245 89.21936 11.39918 3221.893 7.720701 75.838073312.81 4.840675 103.829 10.4961 2027.468 8.529641 75.589173338.115 4.807608 332.9111 10.58241 1244.821 8.451136 75.667254) Baseline (Humidty Ratio: 4.52 gH20/kgAIR):64.2 3.2 483.15 15.5^158 2.9 78.3294679.75 4.97 735.23 13.7 178 4.5 76.73070143.5 6.13 1042.1 11.3 192 5.6 76.83224BLOCK 4: Heat transfer adjustment parameters (June 30, 93).C2^C3^C4^Ignition DelayA B (CA)^(sec)1) 40% SRD, 25% CN62 (Humidity Ratio: 0.0090):0.45 0.35/0.8 0.7/0.774^30.4^0.004180.45 0.35/0.8 0.7/0.786 30.2 0.0041420.45 0.35/0.8^0.7/0.79^26.04 0.0036372) NON-EGR,^0% SRD, 25% #2^(Humidity Ratio:^11.37 gH20/kgAIR):0.4 0.6/0.8 0.7/0.853 41.42 0.0055940.4 0.6/0.8 0.7/0.843 39.1 0.0053460.4 0.6/0.8 0.7/0.831 33.44 0.0045130.4 0.6/0.8 0.7/0.821 41.59 0.0056910.4 0.6/0.8 0.7/0.816 38 0.0052210.4 0.6/0.8 0.7/0.813 34.01 0.0046470.4 0.6/0.8 0.7/0.843 30.01 0.0040470.4 0.6/0.8 0.7/0.834 46.72 0.0065050.4 0.6/0.8 0.7/0.812 43.39 0.0059420.4 0.6/0.8 0.7/0.806 41.06 0.0055640.4 0.6/0.8 0.7/0.826 35.19 0.0048633) EGR, 0% SRD, 25% #2 (Humidity Ratio: 13.26 gH20/kgAIR):0.35 0.8/0.8 0.7/0.867 40.12 0.0054990.35 0.8/0.8 0.7/0.86 38.71 0.0052580.35 0.8/0.8 0.7/0.84 33.22 0.004630.35 0.8/0.8 0.7/0.844 43.27 0.0059850.35 0.8/0.8 0.7/0.83 39.8 0.0054240.35 0.8/0.8 0.7/0.832 35.57 0.0049690.35 0.8/0.8 0.7/0.804 48.37 0.0066410.35 0.8/0.8 0.7/0.825 46.7 0.0065030.35 0.8/0.8 0.7/0.813 39.56 0.0053694) Baseline (Humidty Ratio: 4.52 gH20/kgAIR):0.45^0.8^0.775 8.27^0.0011440.45 0.8 0.762 7.81 0.0010660.45 0.8 0.789 7.01 0.000985224225BLOCK 5:STORHCStoichiometric combustionSTOROC^STORNC^Tbmax(K)results^(June 30,^93).Pmax^EQUI.NO^CFK^AVG.NO(atm)^(ppm) (ppm)1) 40% SRD,^25% CN62^(Humidity Ratio:^0.0090):3.419^3.709^13.948^2380^35.5 2015 0.093 1873.389 3.695 13.892 2558 58.1 3416 0.174 5943.397 3.699 13.906 2590 64.1 3711 0.204 7572) NON-EGR, 0% SRD, 25% #2 (Humidity Ratio:^11.37 gH20/kgAIR):3.428 3.714 13.965 2160 24 873 0.116 1013.442 3.721 13.991 2209 31.1 1037 0.128 1333.387 3.694 13.888 2344 42.6 1715 0.15 2573.432 3.716 13.972 2169 29.4 877 0.12 1053.414 3.707 13.939 2300 38.6 1462 0.17 2493.383 3.692 13.881 2455 53.7 2467 0.189 4663.337 3.669 13.794 2449 53.3 2409 0.158 3803.456 3.728 14.017 2091 26 620 0.121 753.454 3.727 14.013 2249 35.2 1204 0.148 1783.423 3.712 13.956 2381 45.8 1943 0.19 3693.375 3.687 13.865 2479 59 2622 0.221 5273) EGR, 0% SRD, 25% #2^(Humidity Ratio:^13.26 gH20/kgAIR):3.378 3.689 13.871 2003 21.3 410 0.173 713.386 3.693 13.885 2072 28.9 555 0.194 1083.412 3.706 13.934 2230 42.8 1074 0.228 2453.431 3.715 13.969 2001 23.5 398 0.17 683.371 3.686 13.858 2156 33 816 0.2 1633.368 3.684 13.852 2289 47.4 1351 0.224 3033.437 3.718 13.981 2074 23.6 578 0.179 1033.456 3.728 14.017 2079 26 584 0.222 1303.449 3.725 14.005 2348 48.1 1670 0.246 4114) Baseline (Humidty Ratio: 4.52 gH20/kgAIR):2.017^3.009^11.313 2440^56.3 2124 0.148 3142.017 3.009 11.313 2458 50.6 2689 0.153 4112.017 3.009 11.313 2431 65 2340 0.265 620226APPENDIX N^PROGRAM XPRESSD1. Input Parameters*Engine speed, rpm.*Ambient temperature and engine exhaust temperature, K.*Ambient pressure and engine air-box pressure, kPa.*Delivered air mass, total mass of CNG and diesel, kg/cycle.*Crank angle of the first pressure record, CA.*Crank angle of the BOI.*Crank angle of the interval size, CA.*Crank angle interval of the PW, CA.*Number of lines of pressure records per cycle.*Number of CA intervals per cycle to 1c) analyzed after BOI.*Number of engine cycles to be analyzed.*Engine wet-basis emission data, ppm for CO, NO„ THC,%vol for CO2, 02, H2O.*Engine pressure data.2. List of the program XPRESSD.FORC XPRESSD.FORC This program takes engine pressure data at regular crank angleC increments DCA and determines mass-burned fraction.C The source program was XPRESSE.FOR, originated by Dr. P.G. Hill.C Modified for gas-diesel engine by H.Gunawan (May14,1992).C Modified for three-zone combustion and exhaust emission analysisC model of gas-diesel engine by Yinchu Tao and Dr. P.G. HillC (June 30, 1993).C THIS VERSION IS ONLY FOR STOICHIOMETRIC (OR DIFFUSION)C COMBUSTION (WITH SUBROUTINE QWALL CONNECTED).IMPLICIT REAL*8(A-H2O-Z)REAL*8 MAIR,MDSL,MGAS,MG,MDCREAL*8 MTRAP,MATRAP,MTOTREAL*8 MW02,MWN2,MWCH4,MWCH2,MWH20,MWCO2REAL*8 MWCO,MWNOX,MWTHC,MWEXHREAL*8 CA(360),P(360),XMB(360),T(360),X(360)REAL*8 PAVG(360),XAVG(360),TAVG(360)REAL*8 WRK(360),WAVG(360)REAL*8 QAVG(360),QWL(360)REAL*8 TU(360),TUAVG(360)REAL*8 TBRN(360),TBAVG(360)COMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/STATS/CABOI,RPM,N,NCYC,NCACOMMON/GEOM/BORE,STROKE,ROD,CLRHCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/PORT/PABOX,CAIPC,PIPC,TIPC,TEXH,TABOXCOMMON/AMBNT/PAMB,TAMBCOMMON/BURN/UBCOMMON/EMIS/YCO,YCO2,YNOX,Y02,YTHC,YH20,UFRATCOMMON/PREP/STORHC,STOROC,STORNC,RFASTOCOMMON/MW1/MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2COMMON/MW2/MWCO,MWNOX,MWTHC,MWEXHCOMMON/AVGNO/XTBMAXAVGCOMMON/MOLFRAC/X02,XN2,XCO2,XH20,WTMOLUC Specify the cylinder geometry of 1-71 engine:C BORE is cylinder bore(m), STROKE(m),C ROD^is conn rod length(m),C CR^is compression ratio,C CLRH is clearance height(m).C CAIPC is a crank angle (ABDC) after all ports are closed.C CAEXH is a crank angle (BBDC) that exhaust ports open.STROKE = 0.1270D0BORE = 0.10795D0ROD = 0.2540D0CR = 16.0D0CLRH = STROKE/(CR-1.0D0)CAIPC = 50.0D0CAEXH = 86.0D0Specify the molecular weights of the components:MWCO = 28.011DOMWCO2 = 44.011DOMWNOX = 30.006D0MWO2 = 31.999D0MWH20 = 18.015D0MWN2 = 28.013D0MWCH4 = 16.043D0MWCH2 = 14.026D0c*******************Initialize***********************C RDG^is the mass ratio of diesel to natural gas.C RHC^is the atomic ratio of H to C in combined fuel.227228C FRES is the residual mole fraction.C EQVR is the fuel-air equivalence ratio.C RU^is the gas constant for the unburned gas per kg andC CVU is the specific heat for unburned gas.C TIPC is the cylinder contents temp at ipc after mixingC^with residual gas. The residual gas mass fractionC is determined using a scavenging data typical ofC^two-stroke diesels.C ROC is the atomic ratio of 02 to C in unburned mixture.C RNC is the atomic ratio of N2 to C in unburned mixture.C UFRAT is the mass ratio of the unburned fuel to theC^total fuel per unit time.C CA^is crank angle.C PR1^subscript which refers to the first pressure record.C IPC subscript which refers to intake port closing.C BOI^subscript which refers to beginning of injection.OPEN(UNIT=2,FILE='BSLPD-3.DAT',STATUS='OLD')OPEN(UNIT=9,FILE='Y.OUT',STATUS='UNKNOWN')OPEN(UNIT=10,FILE='ESL-3.0UT',STATUS='UNKNOWN')OPEN(UNIT=16,FILE='CH4EQ1.DAT',STATUS='OLD')OPEN(UNIT=26,FILE='CH2EQ1.DAT',STATUS='OLD')C Read the input data:C PABOX is the pressure of air box (kPa).C TABOX is the temperature of air box (K).C TEXH is the exhaust temperature (K).C MAIR is the air mass flow (kg/cycle/cylinder).C MDSL is the diesel mass flow (kg/cycle/cylinder).C MGAS is the natural gas mass flow (kg/cycle/cylinder).C CAPRI is the crank angle of the first pressure record.C CABOI is the crank angle of the beginning of injection.C DCA is the crank angle interval size.C CAPW is the crank angle interval of the PW.C NPR is the number of lines of pressure records per cycle.C NCA is the number of CA intervals per cycle to be analyzedC^after BOI.C NCYC is the number of engine cycles to be analyzed.READ(2,*)RPM,PABOX,TABOX,TEXH,PAMB,TAMBREAD(2,*)MAIR,MDSL,MGASREAD(2,*)CA2R1,CABOI,DCA,CA2WREAD(2,*)NPR,NCA,NCYCREAD(2,*)YCO,YCO2,YNOX,Y02,YTHC,YH20C^^UNITS: ppm FOR CO,NOX,THC; % FOR CO2,02,H20UFRAT = 0.D0C Read pressure data:PRINT*,'reading pressure data'DO 1000 N = 1,NCYCDO 41 I = 1,NPR41^READ(2,*)CA(I),P(I)229C For PCB pressure transducer (PABOX not equal to PIPC) only:DDP = P(CAIPC) - PABOXDO 42 I = 1,NPR42 P(I) = P(I) - DDPC Estimate the mass of air trapped in cylinder MATRAPC and the equivalence ratio EQVR:PIPC = P(CAIPC)CALL INTAKE(MATRAP)MTRAP = MATRAP/(1.DO-FRES)MTOT = MTRAPC Note: MTOT is a instantaneous total mass which is changed duringC^the injection.IF (N .NE. 1) GO TO 100C Calculate the unburned fuel ratio UFRAT:CALL UNBFUELPRINT*,'UFRAT = ',UFRATPRINT*,'MWTHC = ,MWEXH = ,',MWTHC,MWEXHWRITE(10,50)NCYCWRITE(10,51)RPM,EQVR,CABOIWRITE(10,52)MAIR,TIPC,TABOXWRITE(10,53)FRES,MATRAP,MTOTWRITE(10,54)RDG,UFRAT,RFASTOWRITE(10,55)STORHC,STOROC,STORNC50^FORMAT(/,1X,I4,'cycles of reduced byXPGDSL.FOR')51^FORMAT(/,1X,'RPM',F7.1,' Equiv Ratio ',F6.3,'1^CAboi',F8.3)52^FORMAT(1X,'Mair kg ',D10.4,'^Tipc K',F6.1,'1^Tabox',F9.3)53^FORMAT(1X,'Fres',F8.3,' Matrap kg ',E10.4,' Mtot kg1^',E10.4)54^FORMAT(1X,'Mdsl/Mgas= ',F7.4,' UFRAT= ',F8.6,' RFASTO=1^',F8.6)55^FORMAT(1X,'Carbon = 1',2X,'STORHC = ',F7.3,' STOROC =1 ',F7.3, ' STORNC = ',F7.3)C Initialize subroutines:PRINT*, 'initializing subroutines'CALL BURNED(Q1,Q2,Q3,Q4,Q5,Q6,Q7,Q8,1)PBOIAV = 0.D0DO 90 I = 1,NCAPAVG(I) = 0.D0TAVG(I) = 0.D0TUAVG(I) = 0.D0TBAVG(I) = 0.D0QAVG(I) = 0.D0WAVG(I) = 0.D0XAVG(I) = 0.D0STBMAX = 0.D0230SPTBMAX = 0.D090^SXTBMAX = 0.D0100^CONTINUEC Calculate the temperature at BOI:C Assume fuel injection start in between BOI and BOI+1, thus thereC is no fuel in cylinder at BOI.KBOI = DINT((CABOI - CAPR1)/DCA) + 1VBOI = VCYL(CABOI)PBOI = P(KBOI)TUBOI = PBOI*VB0I/MTOT/RUCALL UNBURNED(TUBOI,UU,CVU,VISC,1)ETOT = MTOT*UUCA(1) = CABOI + DCADO 125 I = 1,NPR-KBOIIF(I .GT. 1)CA(I) = CA(I-1) + DCA125^P(I) = P(I+KBOI)PRINT*,'UU=',UUV1 = VBOIP1 = PBOITU1 = TUBOITi = TU1XMB1 = 0.D0XMB(I) = 0.D0TFU = 298.D0MG = 0.D0MD = 0.D0WRITE(10,103)WRITE(10,104)CABOI,PBOI,TUBOI103^FORMAT(7X,'CA',10X,'P kPa',9X,'Tu K',8X,'Tb K',8X,1 'Tfu K',11X,'X')104^FORMAT(1X,6(2X,E11.5))C Calculate conditions at end of each crank angle interval:C Assume: 1) Injection of fuel at constant rate over crank angleinterval CAPW.2) Partial injected fuel begine burning in BOI+1.XMAXP = 0.D0XMAX = 0.D0CAXMAX = CABOIWRK1 = 0.D0QWL1 = 0.D0NCA = NPR-CABOI-CAEXHDO 200 I = 1,NCAIF (I .GT. CAPW) GO TO 107MTOT = MTRAP+(DCA*I/CAPW)*(MDSL+MGAS)ETOT = ETOT +(DCA/CAPW)*(MDSL*(-3216)+MGAS*(-4667))107^CONTINUECALL UNBURNED(TU1,UU,CVU,VISC,2)G = 1.D0/(1.D0+CVU/RU)231GAMMA = 1.D0 + RU/CVUV2 = VCYL(CA(I))AA = 0.8BB = 0.789IF (XMB1 .LT. 0.1) GO TO 112AA=0.8BB=0.789112^ASURF = ACYL(CA(I))CALL QWALL(T1,V2,ASURF,AA,BB,DQWL)TU2 =TU1+TU1*G*(P(I)-P1)/P1+0.45D0*DQWL/(MTOT*(CVU+RU))CALL UNBURNED(TU2,UU,CVU,VISC,1)VU = RU*TU2/P(I)DWRK = (P1+P(I))/2.D0*(V2-V1)ETOT = ETOT - DWRK + DQWLWRITE(10,105)ASURF,DQWL,DWRK,ETOTC 105^FORMAT(1X,'ASURF=,DQWL=,DWRK=,ETOT=,',4(E9.3,1X))C Determine the temperature of unbrned fuel, TFU.IF (I .GT. CAPW) GO TO 109DMG = (DCA/CAPW)*MGASDMD = (DCA/CAPW)*MDSLMG = I*DMG - XMB1*MGASMD = I*DMD - XMB1*MDSLCALL UNBURNED(TFU,UU,CVG,VISC,3)CPG = CVG + 8.3143D0/MWCH4CALL UNBURNED(TFU,UU,CVD,VISC,4)CPD = CVDC TFU changes due to adiabatic constant-pressure mixing withinjected fuels.DTMIX = -(DMG*CPG+DMD*CPD)*(TFU-298.D0)/(MG*CPG+MD*CPD)109^^IF (I .LE. CAPW) GO TO 110MG = (1-XMB1)*MGASMD = (1-XMB1)*MDSLCALL UNBURNED(TFU,UU,CVG,VISC,3)CPG = CVG + 8.3143D0/MWCH4CALL UNBURNED(TFU,UU,CVD,VISC,4)CPD = CVDDTMIX = 0.D0110^CONTINUEC TFU changes due to isentropic compression of fuel.DTCOMP =1^MG*TFU*(8.3143D0/MWCH4)*(P(I)-P1)/P1/(MG*CPG+MD*CPD)TFU = TFU + DTMIX + DTCOMPUG = (-4821) + CVG*(TFU-298.D0)UD = (-3216) + CVD*(TFU-298.D0)VG = (8.3143D0/MWCH4)*TFU/P(I)VD = 0.D0VM = (V2 - MG*(VG-VU) - MD*(VD-VU))/MTOTUM = (ETOT - MG*(UG-UU) - MD*(UD-UU))/MTOTCALL BURNED(P(I),TB,VU,VM,UU,UM,VB,XMB(I),2)C CORRECTION OF X BECAUSE OF THE STOICHIOMETRIC BURNING.IF (XMB(I) .LT. 0.D0) XMB(I)=0.D0X(I) = XMB(I)C 19991111170XMB(I) = XMB(I)*MTOT/(MGAS+MDSL)/(IF (N .NE. 1) GO TO 1111WRITE(10,104)CA(I),P(I),TU2,TB,TFUWRITE(10,1999)V2,VM,VU,VB,UM,UU,UBFORMAT(1X,'V2,VM,VU,VB,UM,UU,UB=',CONTINUEIF(XMB(1) .LT. XMAXP) GO TO 170XMAXP = XMB(I)CAXMAX = CA(I)CONTINUETBRN(I) = TBTU(I) = TU2WRK(I) = WRK1 + DWRKQWL(I) = QWL1 + DQWL1+1/RFASTO),XMB(I)7(1X,D9.3))232C*************prepare for next stepV1 =V2P1 = P(I)TU1 = TU2QWL1 = QWL(I)WRK1 = WRK(I)T(I) = TB*X(I) + TU2*(1 - X(I))Ti = T(I)XMB1 = XMB(I)200^CONTINUEC*********do statistics for each cycleXMAX = 1.D0 - UFRATDO 280 I = 1,NCAIF (CA(I) .LT. CAXMAX) GO TO 280XMB(I) = XMAX280^CONTINUEDO 300 I = 1,NCAXAVG(I) = XAVG(I) + XMB(I)WAVG(I) = WAVG(I) + WRK(I)QAVG(I) = QAVG(I) + QWL(I)TAVG(I) = TAVG(I) + T(I)TBAVG(I) = TBAVG(I) + TBRN(I)TUAVG(I) = TUAVG(I) + TU(I)300^PAVG(I) = PAVG(I) + P(I)PBOIAV = PBOIAV + PBOICALL CYCSTATS(CA,P,XMB,2)C****Find the Max. burned temp. 'TBMAX' and the correspondingmass-burned fraction 'XTBMAX':TBMAX = 1600.D0DO 301 1=1,70IF (XMB(I) .LT. 0.05) GO TO 301IF (TBRN(I) .LT. TBMAX) GO TO 301TBMAX = TBRN(I)XTBMAX = XMB(I)PTBMAX = P(I)233301^CONTINUEPRINT*,'TBMAX =, PTBMAX =, XTBMAX =,',TBMAX,PTBMAX,XTBMAXSTBMAX = STBMAX + TBMAXSPTBMAX = SPTBMAX + PTBMAXSXTBMAX = SXTBMAX + XTBMAX1000^CONTINUEC****Calaulate the conversion factor for converting equilibriumNO (ppm) to tail-piper NO (ppm):TBMAXAVG = STBMAX/NCYCPTBMAXAVG = SPTBMAX/NCYCXTBMAXAVG = SXTBMAX/NCYCWRITE(10,302)TBMAXAVG,PTBMAXAVG,XTBMAXAVG302^FORMAT(1X,'TBMAXAVG =, PTBMAXAVG =, XTBMAXAVG1 =,',3(E11.5,1X))PRINT*,'TBMAXAVG =, PTBMAXAVG =, XTBMAXAVG =,',1^TBMAXAVG,PTBMAXAVG,XTBMAXAVGCALL CONVF(CFK)WRITE(10,303)CFK303^FORMAT(1X,'CFK = ',E11.5,/,1^'(NOTE: AVG NO (ppm) = CFK*(NO)bmax, (NO)bmax is from2^STANJAN)')C****do statistical analysis for all cyclesWRITE(10,102)102^FORMAT(1X,'ensemble-avgd pressures and mass-burned1 fractions')WRITE(10,101)101^FORMAT(6X,'CA',7X,'PAVG kPa',6X,'XAVG',7X,'TAVG K',1 6X,'TUAVG K',5X,'TBAVG K')PBOIAV = PBOIAV/FLOAT(NCYC)WRITE(10,1105)CABOI,PBOIAVDO 1500 I = 1,NCATIME = (CA(I) - CABOI)/6.D0/RPMXAVG(I) = XAVG(I)/FLOAT(NCYC)QAVG(I) = QAvG(I)/FLoAT(NCYC)WAVG(I) = WAVG(I)/FLOAT(NCYC)TAVG(I) = TAVG(I)/FLOAT(NCYC)TUAVG(I) = TuAVG(I)/FLOAT(NCyC)TBAVG(I) = TBAvG(I)/FLOAT(NCyc)pAvG(I) = pAvG(I)/FLoAT(NcYC)wRITE(10,1105)CA(I),PAVG(I),XAVG(I),TAVG(I),TuAvG(I),1^TBAVG(I)1105^FORMAT(1X,6(E11.5,1X))WRITE(6,104)CA(I),PAVG(I),XAVG(I),TIME1500^CONTINUEDO 1600 I=1,NCAIF (XAVG(I) .GT. 0.01) GO TO 1610PP1=CA(I)QQ1=XAVG(I)1600^CONTINUE1610^PP2=CA(I)234QQ2=XAVG(I)CAIGND =(0.01-QQ1)*(PP2-PP1)/(QQ2-QQ1) + PP1 - CABOITIGND=60*CAIGND/360/RPMPRINT*,'PP1,QQ1,PP2,QQ2,CAIGND,TIGND,',PP1,QQ1,PP2,1^QQ2,CAIGND,TIGNDWRITE(10,1620)CAIGND,TIGND1620^FORMAT(1X,'IGNITION DELAY',F8.2,1X,'(CA)',1X,F10.6,1 1X,'(SEC)')STOPENDC****************************************************************DOUBLE PRECISION FUNCTION VCYL(CA)IMPLICIT REAL*8 (A-H2O-Z)COMMON/GEOM/BORE,STROKE,ROD,CLRHC CA is crank angle degrees ABDC.PI = 3.14159D0APSTON = PI/4.D0*BORE**2CAR = CA*PI/180.D0Z = (1.D0 + 2.D0*ROD/STROKE + DCOS(CAR)1^- DSQRT((2.D0*ROD/STROKE)**2 - (DSIN(CAR))**2))2 * STROKE/2.D0 + CLRHVCYL = Z*APSTONRETURNENDc**********************************************************SUBROUTINE INTAKE(MATRAP)C**** ******************************************************IMPLICIT REAL*8(A-H2O-Z)REAL*8 NRESOUT,NO2,NN2,NH20,NCO2,NEXHREAL*8 MAIR,MDSL,MGASREAL*8 MTRAP,MATRAP,MTOTREAL*8 MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2REAL*8 D(3)COMMON/AMBNT/PAMB,TAMBCOMMON/PORT/PABOX,CAIPC,PIPC,TIPC,TEXH,TABOXCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/MW1/MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2COMMON/MOLFRAC/X02,XN2,XCO2,XH20,WTMOLUC****Calculate the blower-exit air temperature, TXBLO:C^EFFBLO = 0.47D0C GSTAR = 0.2857D0C^TABOX = TAMB*(1.D0+((PABOX/PAMB)**GSTAR-1.D0)/EFFBLO)C****Estimate the residual gas mass fraction,C^Fres=Mres/(Mres+Matrap)C^and the mass of air in trapped cylinder charge, MATRAP:RA = 0.287D0VTRAP = VCYL(CAIPC)C^Start iteration to find Degree of Purity:235MTRAP = 0.5*MAIRCPAIR = 1.0035CPEXH1 = CPAIRCPRES1 = CPAIR20^DEGP = 0.6030 D(1) = 0.D0D(2) = 0.D0D(3) = 0.D0I = 1M = 140^DEGP = DEGP + 0.05D0C Correct the residual temperature TRES.50^TRES = (TEXH*CPEXH1*(MAIR + MGAS + MDSL) - TABOX*CPAIR*1^MAIR - DEGP*MTRAP))/CPRES1/(DEGP*MTRAP + MGAS + MDSL)TIPC = TABOX*DEGP + TRES*(1.D0-DEGP)MTRAP = PIPC*VTRAP/RA/TIPCRDELIV = MAIR/MTRAPDEGPUR = 0.173611D0*RDELIV**3.D0-0.95982D0*RDELIV**2.D01^+1.774305*RDELIV - 0.19642D0Y = TIPC - (TABOX*DEGPUR+TRES*(1.D0-DEGPUR))IF (Y .GT. 0.D0) GO TO 52D(1) = DEGPGO TO 5352^D(2) = DEGP53 D(3) = D(1)*D(2)IF (D(3) .EQ. 0.D0) GO TO 40M=M+1DEGP = 0.5D0*(D(1)+D(2))IF (DABS(Y) .LE. 0.5D0) GO TO 55IF (M .LT. 50) GO TO 5055^CONTINUEFRES = 1.D0 - DEGPURTIPC = (1.D0-FRES)*TABOX + FRES*TRESMATRAP = (1.D0-FRES) * MTRAPC ****Calculate CPEXH AND CPRES, (kJ/kgK):CALL COMPOS(MATRAP,XX02,XXN2,XXCO2,XXH20)XTOTAL=RHC/4+9.52*(1+FRES/(1-FRES))/EQVRXCO2 = XXCO2/XTOTALXH20 = XXH20/XTOTALX02 = XX02/XTOTALXN2 = XXN2/XTOTALWTMOLU = X02*MWO2 + XN2*MWN2 + XCO2*MWCO2 + XH20*MWH20RU = 8.3143D0/WTMOLUCALL UNBURNED(TIPC,UU,CVRES,VISC,2)CPRES = CVRES + RUC No. of moles of escaped air.NO2 = ((MAIR-MATRAP)/28.97D0)*0.21NN2 = 3.76*NO2C^No. of moles of residual out.NRESOUT= (MTRAP+MGAS+MDSL)/WTMOLU - FRES*MTRAP/WTMOLUC^Total moles of exhaust.236NEXH = NO2 + NN2 + NRESOUTC Add escaped air and residual out together.NO2 = NO2 + NRESOUT*(XX02/XTOTAL)NN2 = NN2 + NRESOUT*(XXN2/XTOTAL)NCO2 = NRESOUT*(XXCO2/XTOTAL)NH20 = NRESOUT*(XXH20/XTOTAL)X02 = NO2/NEXHXN2 = NN2/NEXHXCO2 = NCO2/NEXHXH20 = NH20/NEXHWTMOLU = X02*MWO2 + XN2*MWN2 + XCO2*MWCO2 + XH20*MWH20RU = 8.3143D0/WTMOLUCALL UNBURNED(TIPC,UU,CVEXH,VISC,2)CPEXH = CVEXH + RUYY = DABS(CPRES-CPRES1)IF (YY .LT. 0.001) GO TO 57CPRES1 = CPRESCPEXH1 = CPEXHI = I+1IF (I .LT. 50) GO TO 5057^CONTINUECALL COMPOS(MATRAP,XX02,XXN2,XXCO2,XXH20)WRITE(10,58) X02,XN2,XCO2,XH20,WTMOLU,RU58^FORMAT(1X,'X 02, N2, CO2, 1120, WTMOLU, RU=',6(F7.4))WRITE(10,60)RDELIV,MTRAP,TEXH,TRES60^FORMAT(1X,'RDELIV,MTRAP,TEXH,TRES',4(1X,E10.4))RETURNENDC*** *******************************************************SUBROUTINE COMPOS(MATRAP,XX02,XXN2,XXCO2,XXH20)c**********************************************************IMPLICIT REAL*8(A-H2O-Z)REAL*8 NTOTU,NRES,NO2,NN2,NH20,NCO2REAL*8 MATRAP,MTOT,MAIR,MDSL,MGASREAL*8 MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2,MWRESCOMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/PREP/STORHC,STOROC,STORNC,RFASTOCOMMON/MW1/MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2COMMON/MOLFRAC/X02,XN2,XCO2,XH20,WTMOLUC ****CALCULATE FUEL-AIR EQUIVALENT RATIO, EQVR:C Stoichiometric complete combustion of a combined fuel (CHn):RDG is mass ratio of diesel-fuel(CH2) to gas(CH4)Replace CH4 + (16/14)*RDG CH2^by CHnwhereRHC = n = (4+16/14*RDG*2)/(1+16/14/*RDG), which iscombined-fuel hydrogen (H) to carbon (C) atomic ratio.CHn + (1+n/4)(02 + 3.76 N2) ===> (n/2)H20 + CO2 +(3.76)(1+n/4)N2237C^RFASTO is the stoichiometric fuel-air ratioC RFA is the fuel-air ratioJ= 1RDG = MDSL/MGASRHC = (28.D0+16.DO*RDG)/(7.D0+8.D0*RDG)RFASTO1 =1^(12.D0+RHC)/((1.D0+RHC/4.D0)*(32.D0+3.76D0*28.D0))RFA = (MDSL+MGAS)/MATRAP55^EQVR = RFA/RFASTO1C ****CALCULATE THE RESIDUAL COMPOSITIONS:C Assume:C^1) Both residual molal fraction and equivalent fuel-air ratioC remain constant for each cycle in cycling iterations.C^2) Residual molal fraction equals residual mass fraction.C^3) Basic cycle residual is pure air (with 21% 02 and 79% N2).C^4) The equation used is:C CHn + 2/EQVR*(02+3.76N2) + 9.52*RR/EQVR*(A*CO2+B*H2O+C*02+D*N2)C^----> E*CO2 + F*H20 + G*02 + H*N2^(RR=FRES/(1-FRES))RR=FRES/(1-FRES)BATA=9.52*RR/EQVRAA=0BB=0CC=0.21DD=0.79EE=l+BATA*AAFF=RHC/2+BATA*BBGG=2/EQVR-1-RHC/4+BATA*CCHH=7.52/EQVR+BATA*DDXTOTAL=RHC/4+9.52*(1+RR)/EQVRI=162^AAA=EE/XTOTALBBB=FF/XTOTALCCC=GG/XTOTALDDD=HH/XTOTALEEE=l+BATA*AAAFFF=RHC/2+BATA*BBBGGG=2/EQVR-1-RHC/4+BATA*CCCHHH=7.52/EQVR+BATA*DDDA1=DABS(AA-AAA)/AAAIF(Al .LT. 0.001) GO TO 64AA=AAABB=BBBCC=CCCDD=DDDEE=EEEFF=FFFGG=GGGHH=HHHI=I+1IF(I .LT. 100) GO TO 6223864^CONTINUEXXCO2=EEEXXH20=FFFXX02=GGGXXN2=HHHC****CALCULATE THE COMPOSITIONS OF THE CYLINDER CONTENTS BETWEEN• IPC AND BOI:• Assume:1) Residual gases are completely mixed with traped air.2) Residual gas fraction affect properties of unburned gas• RESIDUALS AT PREVIOUS COMBUSTION CYCLE:• H20: XXH20; CO2: XXCO2; 02: XX02; N2: XXN2.• SUM = XTOTAL=RHC/4+9.52*(1+RR)/EQVR.• No. of moles of intake air:NO2 = (MATRAP/28.97D0)*0.21D0NN2 = NO2*3.76D0• No. of moles of residuals:MWRES =(XXCO2*MWCO2+XXH20*MWH20+XX02*MW02+XXN2*MWN2)/XTOTALNRES = (FRES/(1-FRES))*(MATRAP/MWRES)• Total No. of moles of cylinder contents between IPC and BOI:NTOTU = NO2 + NN2 + NRES• PRINT*,'NO2=, NN2=, NRES=, ',NO2,NN2,NRES• ADD OXYGEN AND NITROGEN CONTENT IN RESIDUALS:NO2 = NO2 + NRES*XX02/XTOTALNN2 = NN2 + NRES*XXN2/XTOTALNCO2 = NRES*XXCO2/XTOTALNH20 = NRES*XXH20/XTOTALX02 = NO2/NTOTUXN2 = NN2/NTOTUXCO2 = NCO2/NTOTUXH20 = NH20/NTOTUWTMOLU = X02*MWO2 + XN2*MWN2 + XCO2*MWCO2 + XH20*MWH20RU = 8.3143D0/WTMOLUC ****CALCULATE THE STORHC, STOROC AND STORNC AT STOICHIOMETRICCOMPLETE COMBUSTION:C CHn + (1+n/4)(02 + (XN2/X02)N2 + (XH20/X02)H20 + (XCO2/X02)CO2)• __-> (1+(1+n/4)*XCO2/X02)CO2 + (n12+(1+n/4)*XH20/X02)H20+ ((1+n/4)*XN2/X02)N2^(n = RHC)STORHC is the hydrogen(H) to carbon (C) atomic ratioat stoichiometric combustion.STOROC is the oxygen (0) to carbon (C) atomic ratioat stoichiometric combustion.STORNC is the nitrogen (N) to carbon (C) atomic ratioat stoichiometric combustion.BOTTOM = 1+(1+RHC/4)*XCO2/X02STORHC = (RHC+2*(1+RHC/4)*XH20/X02)/BOTTOMSTOROC = (1+RHC/4)*(2+XH20/X02+2*XCO2/X02)/BOTTOMSTORNC = 2*(1+RHC/4)*XN2/X02/BOTTOM239RFASTO =1^(12+RHC)/(1+RHC/4)/(MW02+XN2*MWN2/X02+XH20*MWH20/X022 +XCO2*MWCO2/X02)IF (RFASTO1 .GT. RFASTO) GO TO 47A2 = DABS(RFAST01-RFASTO)IF (A2 .LE. 0.001) GO TO 49RFASTO1 = RFASTO1 + A2/2J = J+1IF (J .LT. 50) GO TO 55GO TO 4947^A2 = DABS(RFAST01-RFASTO)IF (A2 .LE. 0.001) GO TO 49RFASTO1 = RFASTO1 - A2/2J = J+1IF (J .LT. 50) GO TO 5549^PRINT*,'RFAST0=',RFASTORETURNENDC* *********************************************************SUBROUTINE UNBFUELC** ********************************************************IMPLICIT REAL*8(A-H2O-Z)REAL*8 MWCO,MWCO2,MWNOX,MW02,MWTHC,MWH20,MWN2REAL*8 MWCH2,MWCH4,MWEXHREAL*8 MAIR,MDSL,MGAS,MEXH,MTOT,MUFCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/EMIS/YCO,YCO2,YNOX,Y02,YTHC,YH20,UFRATCOMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/MW1/MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2COMMON/MW2/MWCO,MWNOX,MWTHC,MWEXHYCO = YCO/10000.D0YNOX = YNOX/10000.D0YTHC = YTHC/10000.D0YN2 = 100 - YCO - YCO2 - YNOX - Y02 - YTHC - YH20FF =MWCH4*RDG/MWCH2FDSL = FF/(1+FF)MWTHC = FDSL*MWCH2 + (1-FDSL)*MWCH4SUMYMW = YCO*MWCO + YCO2*MWCO2 + YNOX*MWNOX + YO2*MW021^+ YTHC*MWTHC + YH20*MWH20 + YN2*MWN2MEXH = MAIR + MDSL + MGASMUF = YTHC*MWTHC*MEXH/SUMYMWUFRAT = MUF/(MDSL+MGAS)MWEXH = SUMYMW/100RETURNENDc****************************************************************SUBROUTINE UNBURNED(TU,UU,CVU,VISC,L)C* ***************************************************************C THIS SUBROUTINE IS USED TO COMPUTE THE THERMODYNAMIC PROPERTIES240C OF UNBURNED AIR-RESIDUAL MIXTURE AT UNBURNED GAS ZONE, AND THEC THERMODYNAMIC PROPERTIES OF UNBRNED FUELS (CH4 AND CH2) ATC UNBURNED FUEL ZONE.IMPLICIT REAL*8(A-H2O-Z)REAL*8 MW02,MWN2,MWCH4,MWCH2,MWH20,MWCO2REAL*8 HF02,HFN2,HFH20,HFCO2REAL*8 DH02,DHN2,DHH20,DHCO2REAL*8 CP02,CPN2,CPCH4,CPCH2,CPH20,CPCO2COMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/MW1/MWCO2,MW02,MWH20,MWN2,MWCH4,MWCH2COMMON/MOLFRAC/X02,XN2,XCO2,XH20,WTMOLUHF02 = 0.0D0HFN2 = 0.0D0HFH20 = -241827.D0HFCO2 = -393522.D0TDIM = TU/100.0D0TDIMSQ = DSQRT(TDIM)TDIM2 = TDIM*TDIMTDIM3 = TDIM2*TDIMTDIM4 = TDIM3*TDIMTDIM14 = TDIM**0.25D0TDIM54 = TDIM**1.25D0TDIM74 = TDIM**1.75D0TDIM32 = TDIM**1.5D0TDIM52 = TDIM**2.5D0TDIM34 = TDIM**0.75D0C *****Calculate molar enthalpy differences between 298K and TU,DH02 = 3743.2D0*TDIM+0.80408D0*TDIM52+35714.0D0/1^TDIMSQ-23688.0DO/TDIM-23906.63D0DHN2 = 3906.0DO*TDIM+102558.0DO/TDIMSQ-107270.0D0/1^TDIM+41020.0DO/TDIM2-39673.0D0DHH20 = 14305.0DO*TDIM-14683.2D0*TDIM54+5516.73D0*1^TDIM32-184.95D0*TDIM2-11876.23D0DHCO2 = 6914.5D0*TDIM-40.265D0*TDIM74-40154.0D0*1^TDIMSQ+70704.0DO*TDIM14-43912.73D0IF(L .NE. 1) GO TO 10C ****Calculate the internal energy of unburned air and residuals,UU (kJ/kg):UU = (X02*(HF02+DH02)+XN2*(HFN2+DHN2)+XH20*(HFH2O+DHH20)+1^XCO2*(HFCO2+DHCO2))/WTMOLU - RU*TURETURN10^CONTINUEIF(L .NE. 2) GO TO 20C ****Calculate the specific heat of unburned air and residuals,CVU: CP (kJ/kmol-K) & CVU (kJ/kg-K)CP02=37.432D0+0.020D0*TDIM32-178.57D0/TDIM32+236.88D0/TDIM2CPN2=39.060D0-512.79D0/TDIM32+1072.7D0/TDIM2-820.40D0/TDIM3241CPH20 =1^^143.05D0-183.54D0*TDIM14+82.751D0*TDIMSQ-3.6989D0*TDIMCPCO2 = 69.145D0-.70463D0*TDIM34-200.77D0/TDIM5Q+1^176.76D0/TDIM34CVU = (X02*CP02+XN2*CPN2+XH20*CPH20+XCO2*CPCO2)/WTMOLU - RU20^CONTINUEIF(L .NE. 3) GO TO 30C ****Calculate specific heat of CH4, CPG (kJ/kmol-K) & CVU(kJ/kg-K):CPCH4 = -672.87D0+439.74D0*TDIM14-24.875D0*TDIM34+1^323.88DO/TDIMSQCVU = (CPCH4 - 8.3143D0)/MWCH4RETURN30^CONTINUEIF(L .NE. 4) GO TO 40• ****Calculate specific heat of CH2, CPD (kJ/kmol-K) & CVU(kJ/kg-K):CPCH2 = (104.18D0+46.55D0*TDIM)/12.D0CVU = CPCH2/MWCH2RETURN40^CONTINUEC ****Estimate the mean viscosity of unburned air-residualmixture:TM = TU**0.645D0VISC = (X02*MW02*5.09D0 + XN2*MWN2*4.57D0 +1^XH20*MWH2O*3.26D02 + XCO2*MWCO2*3.71D0)/WTMOLU*10.D0**(-7.D0)*TMRETURNENDC***************** ***********************************************SUBROUTINE BURNED(P,TB,VU,VM,UU,UM,VB,XMB,II)C* ***************************************************************IN THIS SUBROUTINE XMB MEANS Mburned/MTOT, NOT THE FUEL• MASS-BURNED FRACTION. MTOT VARIES DURING THE FUEL INJECTION• PERIOD.IMPLICIT REAL*8(A-H2O-Z)REAL*8 D(3)COMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/BURN/UBIF(II .NE. 1) GO TO 10CALL TABLE(Q1,Q2,Q3,Q4,Q5,1)XMB = 0RETURN10^CONTINUEC Find the linear relationship between UB and VB at the flame frontIF( UM .NE. UU)G0 TO 210242A = -1.D0/PB = VU - A*UUGO TO 220210^A = (VM - VU )/(UM - UU )B = VU - A*UUC^SOLVE FOR T,V,U JUST BEHIND FLAME AND MASS FRACTION X220^TB = 1000.D0230^D(1) = 0.D0D(2) = 0.D0D(3) = 0.D0M = 1240^TB = TB + 100.D0250^CALL TABLE(TB,P,SB,UB,VB,3)Y = VB - ( A*UB + B )IF( Y .GT. 0.D0) GO TO 252D(1) = TBGO TO 253252^D(2) = TB253^D(3) = D(1)*D(2)IF(TB .GT. 6000.D0 ) WRITE(10,261)IF(TB .GT. 6000.D0 ) go to 300IF( D(3) .EQ. 0.D0 ) GO TO 240M= M+1TB = 0.5D0*( D(1) + D(2) )IF(DABS(Y) .LE. 0.1D-7 ) GO TO 300IF( M .LT. 50) GO TO 250WRITE(10,260)260^FORMAT(1X,'NO CONVERGENCE ON FLAME TEMPERATURE IN 501^TRIES')261^FORMAT(1X,'FLAME TEMPERATURE EXCEEDS 6000 K')RETURN300^CONTINUEC CALCULATE MASS FRACTION XMB CORRESPONDING TO ASSUMED PRESSUREXMB = (VM - VU )/(VB - VU)321 RETURNENDc****************************************************************SUBROUTINE TABLE(TBX,PBX,SBX,UBX,VBX,L)C* ***************************************************************IMPLICIT REAL*8(A-H2O-Z)REAL*8 PTAB(10),TBTAB(10)REAL*8 SBTAB(10,10),UBTAB(10,10),AMTAB(10,10)REAL*8 SBTAB1(10,10),UBTAB1(10,10),AMTAB1(10,10)REAL*8 SBTAB2(10,10),UBTAB2(10,10),AMTAB2(10,10)REAL*8 SP(10),UP(10),AMP(10),X(10),Y(10)REAL*8 EQREAD(6)COMMON/PROPS/RDG,RHC,FRES,EQVR,RUIF(L .NE. 1) GO TO 20PSTORE = -1000.D0DO 12 I = 1,10243PTAB(I) = 0.D0SP(I) = 0.D0UP(I) = 0.D0AMP(I) = 0.D0TBTAB(I) = 0.D0DO 11 J = 1,10SBTAB(I,J) = 0.D0UBTAB(I,J) = 0.D0AMTAB(I,J) = 0.D011^CONTINUE12 CONTINUEC* ***************************************************************C^READ TABLE OF BURNED GAS PROPERTIESC Reads files for different equivalence ratios; unit 11(0.2),C unit 12(0.4), unit 13(0.6), unit14(0.8), unit15(0.9),16(1)C THIS VERSION ASSUME THAT FOR DIESEL ENGINES THE BURNED GASESC ARE THE RESULT OF STOICHIOMETRIC COMBUSTION.DO 120 J = 1,10DO 110 I = 1,10DO 103 K = 6,6READ(K+10,*)TBTAB(I),PTAB(J),AMTAB1(I,J),VB,UBTAB1(I,J),1^ H,SBTAB1(I,J)READ(K+20,*)TBTAB(I),PTAB(J),AMTAB2(I,J),VB,UBTAB2(I,J),1 H,SBTAB2(I,J)103^CONTINUEEQREAD(1) = 0.2D0EQREAD(2) = 0.4D0EQREAD(3) = 0.6D0EQREAD(4) = 0.8D0EQREAD(5) = 0.9D0EQREAD(6) = 1.0D0C Set a burned-gas properties table for a given H to C ratio RHCC by linearly interpolate the CH4 and CH2 tables at the sameC equivalence ratio EQVR.AMTAB(I,J)=AMTAB2(I,J)+(RHC-2)/2*(AMTAB1(I,J)-AMTAB2(I,J))UBTAB(I,J)=UBTAB2(I,J)+(RHC-2)/2*(UBTAB1(I,J)-UBTAB2(I,J))SBTAB(I,J)=SBTAB2(I,J)+(RHC-2)/2*(SBTAB1(I,J)-SBTAB2(I,J))PTAB(J) = PTAB(J)*101.325D0UBTAB(I,J) = UBTAB(I,J)*0.001D0109^FORMAT(1X,'TAB TB,P,AM,UB,SB=',5(1X,D10.4))110^CONTINUE120^CONTINUEPRINT*,'A table of proportional-model burned gas1^properties'PRINT*,'^for a given EQVR is formed'RETURNC*** *************************************************************C^GIVEN P AND SB OR TB20^IF(PBX .EQ. PSTORE) GO TO 160PSTORE = PBXNP = 10XSET = PBXDO 131 I = 1,10DO 130 J = 1,10X(J) = PTAB(J)130^Y(J) = UBTAB(I,J)CALL CUBICS(NP,X,Y,XSET,UP(I))131^CONTINUEDO 141 I = 1,10DO 140 J = 1,10140^Y(J) = AMTAB(I,J)CALL CUBICS(NP,X,Y,XSET,AMP(I))141^CONTINUEDO 151 I = 1,10DO 150 J = 1,10150^Y(J) = SBTAB(I,J)CALL CUBICS(NP,X,Y,XSET,SP(I))151^CONTINUE160^CONTINUEIF( L .NE. 2 ) GO TO 300C**************************************XSET = SBXDO 230 I =1,10X(I) = SP(I)230^Y(I) = TBTAB(I)CALL CUBICS(NP,X,Y,XSET,TBX)DO 240 I = 1,10240^Y(I) = UP(I)CALL CUBICS(NP,X,Y,XSET,UBX)DO 250 I = 1,10250^Y(I) = AMP(I)CALL CUBICS(NP,X,Y,XSET,AMX)VBX = 8.3143D0/AMX*TBX/PBXRETURNC* *************************************300^XSET = TBXDO 330 I = 1,10X(I) = TBTAB(I)330^Y(I) = SP(I)CALL CUBICS(NP,X,Y,XSET,SBX)DO 340 I = 1,10340^Y(I) = UP(I)CALL CUBICS(NP,X,Y,XSET,UBX)DO 350 I = 1,10350^Y(I) = AMP(I)CALL CUBICS(NP,X,Y,XSET,AMX)VBX = 8.3143D0/AMX*TBX/PBXRETURNENDC* ********************************************************SUBROUTINE CUBICS(NP,X,Y,XSET,YCALC)C*********************************************************NP IS NUMBER OF X,Y DATA PAIRS (I RUNS FROM 1 TO N)244245IMPLICIT REAL*8(A-H2O-Z)REAL*8 X(10),Y(10),D(10),E(10),F(10),G(10)M = NP -1MM = NP -2C CALCULATION OF SECOND DERIVATIVES G(I)G(1) = 0.D0G(NP) = 0.D0DO 100 I = 2,MD(I) = X(I) - X(I-1)E(I) = 2.D0*( X(I+1) - X(I-1) )F(I) = X(I+1) - X(I)100^G(I) = 6.D0/F(I)*(Y(I+1)-Y(I))+6.D0/D(I)*(Y(I-1) - Y(I))DO 1040 I = 2,MMFA = D(I+1)/E(I)E(I+1) = E(I+1) - FA*F(I)G(I+1) = G(I+1) - FA*G(I)1040^CONTINUEDO 1070 I = 2,14G(NP+1-I)=(G(NP+1-I)-F(NP+1-I)*G(NP+2-I))/E(NP+1-I)1070^CONTINUEC CALCULATION OF INTERPOLATED VALUE YCALC AT X=XSETD(NP) = X(NP) - X(NP-1)I = 1200^I = I + 1IF( XSET .GE. X(I) .AND. I .LT. NP) GO TO 200DELM = XSET - X(I-1)DELP = X(I) - XSETYCALC = G(I-1)/6.D0/D(I)*DELP**3 + G(I)/6.D0/D(I)*DELM**31^+(Y(I-1)/D(I) -G(I-1)*D(I)/6.D0)*DELP2 +(Y(I)/D(I)^-G(I)*D(I)/6.D0 )*DELMRETURNENDC** ***********************************************************SUBROUTINE CONVF(CFK)C****** *******************************************************IMPLICIT REAL*8 (A-H2O-Z)REAL*8 MAIR,MDSL,MGAS,MFUEL,MTOT,MEXHREAL*8 MFBMAX,NFBMAX,NBMAX,NEXHREAL*8 MWFUEL,MWTHC,MWEXH,MWCO,MWNOXCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/PROPS/RDG,RHC,FRES,EQVR,RUCOMMON/MW2/MWCO,MWNOX,MWTHC,MWEXHCOMMON/MOLFRAC/X02,XN2,XCO2,XH20,WTMOLUCOMMON/AVGNO/XTBMAXAVGC****Calculate the No. of moles of burned gas that burned at max.C temperature TBMAX:C MFBMAX is the mass of fuel (CHn) burned at max. temperature.C NFBMAX is No. of moles of fuel (CHn) burned at max. temperature.MFUEL = MDSL + MGAS246MFBMAX = XTBMAXAVG*MFUELMWFUEL = MWTHCNFBMAX = MFBMAX/MWFUELC Combined fuel CHn complete combustion equation at max.C temperature is:(NFBMAX)CHn + (NFBMAX)(1+n/4)[02 + (XN2/X02)N2+ (XH20/X02)H20 + (XCO2/X02)CO2] ---->(NFBMAX)[1+(l+n/4)(XCO2/X02)]CO2+ (NFBMAX)[n/2+(1+n/4)(XH20/X02)1H20 +(NFBMAX)(1+n/4)(XN2/X02)N2C where RHC = n is the atomic ratio of hydrogen (H) to carbon (C)C in the fuel (CHn).C NBMAX is No. of moles of burned gas at max. temperature TBMAX.NBMAX = NFBMAX*(1 + RHC/2 + (1+RHC/4)*(XCO2+XH20+XN2)/X02)C****Calculate the total No. of molea of engine exhaust and theconversion factor for vonverting the equilibrium NO (ppm) totail-piper NO (ppm):C MEXH is the total mass of the engine exhaust per cycle.C NEXH is the total No. of moles of the engine exhaust per cycle.MEXH = MAIR + MDSL + MGASNEXH = mExii/mwExiiCFK = NBMAX/NEXHPRINT*,'CFK =',CFKRETURNENDc*************************************************************SUBROUTINE PIKSR2(N,ARR,BRR)c*************************************************************C Sorts an array ARR of length N into ascending numerical orderC by straight insertion. N is input; ARR is replaced on outputC making the corresponding rearrangement of array BRRREAL*8 ARR(N),BRR(N)DO 12 J = 2,NA = ARR(J)B = BRR(J)DO 11 I = J-1,1,-1IF(ARR(I) .LE. A) GO TO 10ARR(I+1)=ARR(I)BRR(I+1)=BRR(I)1 1^CONTINUEI = 01 0 ARR(I+1) = ABRR(I+1) = B12^CONTINUERETURNENDC*************************************************************DOUBLE PRECISION FUNCTION ACYL(CA)247C*** **********************************************************C Calculates the cylinder surface area for a given degree CAIMPLICIT REAL*8 (A-H2O-Z)COMMON/GEOM/BORE,STROKE,ROD,CLRHPI = 3.14159D0APSTON = PI/4.DO*BORE**2CAR = CA*PI1180.D0Z = ( 1.D0 + 2.DO*ROD/STROKE + DCOS(CAR)1^-DSQRT((2.DO*ROD/STROKE)**2+(DSIN(CAR))**2))*STROKE/2.D02^+ CLRHACYL = Z*PI*BORE + 2.DO*APSTONRETURNENDC** *******************************************************SUBROUTINE QWALL(TM1,V2,ASURF,AA,BB,DQWL)c*********************************************************C Calculates the heat transfer from the gas to the cylinder wallC using Annand's and Woschni's correlation.IMPLICIT REAL*8 (A-H2O-Z)REAL*8 MTOT,MAIR,MDSL,MGAS,NUCOMMON/GEOM/BORE,STROKE,ROD,CLRHCOMMON/MASS/MTOT,MAIR,MDSL,MGASCOMMON/STATS/CABOI,RPM,N,NCYC,NCACOMMON/PROPS/RDG,RHC,FRES,EQVR,RUPISVEL = RPM * STROKE / 30.0DENS = MTOT / V2CALL UNBURNED(TM1,UU,CVU,VISC,2)RENUM = DENS * PISVEL * BORE / VISCNU = AA*RENUM**(BB)CPG = RU + CVUTRMLCO = CPG * VISC / 0.7D0H = AA * TRMLCO / BORE * RENUM**(BB)C The wall temperature is assumed to be constantTW = 450.0D0QCONV = -ASURF * H * (TM1-TW)QRAD = -(1.6E-12)*ASURF* ( TM1**4 - TW**4 )WRITE(9,26)RENUM,TRMLCO,H,NU26 FORMAT(1X,'RENUM=,TRMLC0=,H=,NU=,',4(E11.4,1X))PRINT*,'RENUM=, TRMLC0=, ',RENUM,TRMLCODQWL = (QCONV+QRAD)DQWL = DQWL * (60./RPM/360.)RETURNENDC** ***************************************


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items