Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Performance and combustion characteristics of a diesel-pilot gas injection engine Gunawan, Hardi 1992

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

Item Metadata


831-ubc_1992_fall_gunawan_hardi.pdf [ 5.3MB ]
JSON: 831-1.0080923.json
JSON-LD: 831-1.0080923-ld.json
RDF/XML (Pretty): 831-1.0080923-rdf.xml
RDF/JSON: 831-1.0080923-rdf.json
Turtle: 831-1.0080923-turtle.txt
N-Triples: 831-1.0080923-rdf-ntriples.txt
Original Record: 831-1.0080923-source.json
Full Text

Full Text

PERFORMANCE AND COMBUSTION CHARACTERISTICS OF A DIESEL-PILOT GAS INJECTION ENGINE  by HARDI GUNAWAN B.Eng., Trisakti University (Indonesia), 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Mechanical Engineering Department  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA June 1992 © Hardi Gunawan 1992  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  Mer,hanic4a Fhpe&Ifil.  The University of British Columbia Vancouver, Canada Date jaint  DE-6 (2/88)  aci I tqli?„  11  ABSTRACT  High pressure injection of natural gas with diesel liquid pilot fuel has been investigated in a single-cylinder two-stroke diesel engine of low compression ratio. The thermal efficiency and emissions of NO N , CO and HC were determined at 1200 RPM as a function of load. Variations of the pilot-diesel energy to total heat energy ratio (in the range 15 25%) strongly affected efficiency and emissions rate. Gas injection pressure was also shown to be an important variable. The thermal efficiency of the gas-diesel engine was shown to exceed that of the conventional diesel at full load, and is less at low load in the present configuration. The combustion rate analysis has been used to determine the pressure rise (ignition time) delay and combustion duration as well as the characteristic burning pattern. At low load late burning high cyclic variations, and incomplete combustion are associated with peak compression temperature lower than 900 K.  111  TABLE OF CONTENTS  ABSTRACT^ LIST OF SYMBOLS^  ii xiii  LIST OF TABLES^  xv  LIST OF FIGURES^  xvi  ACKNOWLEDGEMENT^  xix  1 INTRODUCTION^  1  1.1^Background^  1  1.2^Use of Natural Gas in Diesel Engine ^  4  1.3^Objective of This Research^  7  1.4^Methodology^  7  2 LITERATURE REVIEW^  9  2.1^The Diesel Engine^  9  2.2^Natural Fumigation Method^  12  2.3^Timed Port Injection Method^  15  2.4^Direct Injection Method^  16  2.5^Analysis of Combustion Rate^  17  2.6^Summary^  23  iv 3 EXPERIMENTAL APPARATUS ^  25  3.1^Introduction^  25  3.2^Engine and Test Bed^  28  3.3^Fuel System^  31  3.3.1 Fuel Supply System^  32  3.3.2 Fuel Injection System^  33  3.4^Instrumentation^  35  3.4.1 Engine Speed^  36  3.4.2 Torque (Load Cell)^  36  3.4.3 Gas Fuel Mass-flow Rate^  36  3.4.4 Pilot Diesel Fuel Consumption^  37  3.4.5 Emission Instrumentation^  37  3.4.6 Cylinder Pressure^  39  3.5^Data Acquisition System^  40  4 EXPERIMENTAL PROCEDURE^  43  4.1^Testing Procedure^  43  4.2^Engine Performance Calculation Procedure^  46  4.2.1 Thermal Efficiency^  46  4.2.2 Brake Mean Effective Pressure (BMEP)^  48  4.2.3 Brake Specific Emissions^  48  V  5 MEASUREMENTS OF ENGINE PERFORMANCE^  53  5.1 Conventional Diesel Performance^  53  5.2 Natural Gas-Diesel Performance ^  57  5.3 Effects of Gas Injection Pressure^  58  5.4 Effects of Diesel Ratio ^  63  5.5 Effects of Pilot-Diesel Cetane Number^  66  5.6 Summary^  68  6 COMBUSTION RATE ANALYSIS^  70  6.1^General^  70  6.2^Formulation of the Combustion Model^  72  6.3^Mixture Composition^  76  6.3.1 Scavenged Air^  78  6.3.2 Residual Gas^  81  6.3.3 The Unburned Gas Composition ^  82  6.4^Thermodynamics Properties of the Unburned Gas^  83  6.5^Thermodynamics Properties of the Burned Gas^  86  6.6^Heat Transfer to the Cylinder Wall ^  88  6.7^Calculation Procedure^  90  6.8^Pressure Measurements^  92  6.9^Indicated Work^  95  ^  vi 6.10 Cyclic Variation  96  6.11 Mass-burned Fractions  98  6.12 Summary  101  7 CONCLUSIONS AND RECOMENDATIONS  103  7.1^Conclusions  103  7.2^Recommendations  105  8 REFERENCES  107  9 APPENDICES  111  9.1^Appendix A - Diesel Engine Emission Standards  111  9.2^Appendix B - Calibration Curves  113  9.3^Appendix C - Properties of Test Fuels  118  9.4^Appendix D - Composition of Natural Gas  119  9.5^Appendix E - Exhaust Sampling Probe in the Engine Exhaust Pipe  120  9.6^Appendix F - Schematic of the Fuels Injector  121  9.7^Appendix G - Power Correction Factor Calculation Method  122  9.8^Appendix H - Determination of the Cylinder Volume  123  9.9^Appendix I - Determination of Fuel-air Equivalence Ratio  126  9.10 Appendix J - Iteration Procedure  129  9.11 Appendix K - Mass-burned Fraction Program Listing  131  vii 9.12 Appendix L - Verification of the Computation Procedure for the Constant Volume Case^147  viii  LIST OF SYMBOLS  a^constant bsco brake specific of carbon monoxide b so^brake specific emissions bsfic brake specific of unburned hydrocarbon bsNox brake specific of oxides of nitrogen B^Cylinder bore BMEP Brake mean effective pressure Bdo^  Dry inlet air pressure  B o^Inlet air pressure  c^clearance height CH 4 Methane CO Carbon monoxide CO 2 Carbon dioxide Cp^Molar specific heat at constant pressure C,^Molar specific heat at constant volume  D^Engine displacement DP^Degree of purity of the charge Etot^Total energy of the system  f^fraction  ix fa^Atmospheric  factor  FA^Fuel-air ratio FA.), Stoichiometric fuel-air ratio fe.,^Correction factor to observed brake horse power fai^Engine factor h e^Convective heat transfer coefficient Enthalpy of formation at standard state condition H^Enthalpy 11 2^Hydrogen H2 O Water HC Unburned hydrocarbon k^Thermal conductivity LHV Lower heating value of fuels m^mass mad  ^Delivered air mass to the engine per cycle  male^Mass of inducted air M at,^Mass  of air trapped in the cylinder per cycle  Mdsl^Mass  of diesel fuel injected into the combustion chamber per cycle  me^Mass of emissions mep Mean effective pressure m exh Mass of exhaust gas  me^Mass of fuel m g. Mass of natural gas injected into the combustion chamber per cycle Rot^Mass of cylinder charge during combustion process  mfr^Trapped mass M^Molecular weight n^Polytropic exponent Hydrogen-to-carbon atomic ratio Mole number N^Engine speed N2^Nitrogen  N„^Mole number of trapped charge NO Nitric oxide NO, Nitrogen dioxide NO Oxides of nitrogen Ni,^Nusselt number 0 2^Oxygen p^Distance between the piston pin to the piston top surface P^Pressure Power P,^Initial pressure Pb^Pressure  of cylinder at any crank position during combustion  Per^Corrected pressure  xi Pc3, 1^Cylinder pressure  P.^Maximum-measured pressure PM^Particulate Matter Pee., Maximum pressure Pmeas Measured cylinder pressure  Pinot^Polytropic compression or expansion pressure obtained without combustion Po^Inlet manifold pressure Ps^Pressure of cylinder at beginning of combustion Q„^Heat transfer from the cylinder contents to the wall r^Crank radius Inlet manifold to inlet air pressure ratio Diesel-to-gas mass ratio re^Compression ratio R^Universal gas constant Ra^Air gas constant  R e^Reynolds number RPM Revolution per minute  Sp^Mean piston velocity t^time temperature (in degree Celcius) T^Temperature Tb^Engine torque output  xi i Tm^Mass-average gas temperature T„,^Cylinder wall temperature u^Specific internal energy of the system U 0^Internal energy of the cylinder contents at a reference point v^Specific volume V^Volume V l^Initial volume Ve^Cylinder clearance volume Vd^Displacement  volume  Vipc^Cylinder volume when piston at inlet port closure position Vmax Maximum volume v ,a Pa Specific volume at ambient temperature and pressure ,  v o^Specific volume at temperature t and pressure P W^Work done on the piston We^Work produced per cycle x^Mass-burned Fraction xmax Maximum mass-burned fraction assuming no heat transfer Y^Mole fraction  Greek Letters  Specific heat ratio Scavenged-blower thermal efficiency libth^  11th  ^  Brake thermal efficiency  Thermal efficiency Crank angle  A  ^  Delivery ratio Viscosity of the mixture  p  ^  Density of the mixture  ^ Pt,P  Density of air at temperature t and pressure P  Fuel-air equivalence ratio  Subscript  abox Air box amb Ambient  b^Burned gas cf^Correction factor  i^Species i res^Residual gas tr^trapped  u^Unburned gas  xiv Abbreviations ABDC After Bottom Dead Centre ATDC After Top Dead Centre BDC Bottom Dead Centre BOI Beginning of fuels injection BTDC Before Top Dead Centre CA Crank Angle CHEMI Chemiluminiscense EPA Environmental Protection Agency EVC Exhaust Valve Close EVO Exhaust Valve Open FID Flame inoization detector IPC Inlet Port Closure LFE Laminar Flow Element LNG Liquified Natural Gas NDIR Non-dispersive Infrared PW Pulse width TDC Top Dead Centre  XV  LIST OF TABLES  Table 3.1  Table 4.1  Table 5.1  ^  ^  ^  Engine Specifiations  Testing Ranges  ^  ^  28  44  Effect of Gas Injection Pressure on Unburned Gas Ratio^62  Table 6.1^List of Temperatures and Pressures used in Burned Gas Properties Table^87  Table A.1  ^  Table A.2  Table C.1  ^  Heavy-duty Truck Engine Emission Standard ^112  ^  Table D.1  Urban Bus Heavy-duty Engine Emission Standard ^111  ^  Properties of Test Fuels^  118  Composition of Natural Gas^  119  xvi  LIST OF FIGURES  Figure 1.1^Status of Conventional Diesel Emissions towards US EPA Standards 2 Figure 1.2^Methods of Using Natural Gas in Diesel Engines^4  Figure 2.1^Graphical Method of Estimated Mass-burned Fraction ^19 Figure 2.2^Determination of Combustion Pressure from Measured Pressure 20  Figure 3.1^Schematic of Experimental Apparatus and Instrumentation ^26 Figure 3.2^Test Engine and Dynamometer in a Cell^ 29 Figure 3.3^Engine Control Panel and Data Acquisition System^30 Figure 3.4^Schematic of Fuel Supply System^  32  Figure 3.5^Schematic of Fuel Injection System ^  34  Figure 3.6^Exhaust Emission Sampling System^  38  Figure 3.7^Pressure Transducer Location in the Cylinder Head^39 Figure 3.8^Data Flow in the Data Acquisition System ^41  xvii Figure 5.1^Diesel Baseline Performance Curve^  54  Figure 5.2^Diesel Oxides of Nitrogen Emissions^  55  Figure 5.3^Diesel Unburned Hydrocarbon Emission ^ 56 Figure 5.4^Determination of Best BOI Performance^ 57 Figure 5.5^Effect of Gas Injection Pressure on Thermal Efficiency^59 Figure 5.6^Effect of Gas Injection Pressure on Oxides of Nitrogen ^60 Figure 5.7^Effect of Gas Injection Pressure on Unburned Hydrocarbon^61 Figure 5.8^Effect of Diesel Ratio on Thermal Efficiency^63 Figure 5.9^Effect of Diesel Ratio on Oxides of Nitrogen ^64 Figure 5.10^Effect of Diesel Ratio on Unburned Hydrocarbon^65 Figure 5.11^Effect of Pilot-Diesel Cetane Number on Thermal Efficiency^66 Figure 5.12^Effect of Pilot-Diesel Cetane Number on NO„^67 Figure 5.13^Effect of Pilot-Diesel Cetane Number on unburned HC^68  Figure 6.1^Typical Cylinder Pressure Data^  71  Figure 6.2^Schematic of the Combustion Model^  72  Figure 6.3^Schematic of the Uniflow-scavenged Configuration^73 Figure 6.4^Scavenging Data Typical of Large Two-stroke Diesels ^81 Figure 6.5^Typical Mass-burned Fraction Results ^  88  Figure 6.6^Typical Pressure Crank-Angle Diagram at different Loads^93 Figure 6.7^Typical Log P - Log V Diagram^  94  Figure 6.8^Comparison of Indicated Work to Brake Work^96  xviii Figure 6.9^Standard Deviation and Relative S.D. of the Brake Work^97 Figure 6.10^Superposition of 10 Successive Cycles ,^98 Figure 6.11^Cylinder Pressure, Temperature and Mass-burned Fraction Distribution^99 Figure 6.12^Combustion Pattern at different Load^ 100 Figure 6.13^Combustion Pattern at different Gas Injection Pressure ^101  Figure B.1^Speed Calibration Curve ^  113  Figure B.2^Torque Calibration Curve ^  114  Figure B.3^Diesel Fuel Mass-flow Calibration Curve  ^  115  Figure B.4^Cylinder Pressure Transducer Calibration Curve ^116 Figure B.5^Laminar Flow Element Calibration Curve ^117  Figure E.1  Figure F.1  ^  ^  Figure H.1  ^  Figure H.2  ^  Exhaust Sampling Probe in the Engine Exhaust Pipe ^120  Schematic of the Two-fuel Unit Injector ^121  Piston Top Surface Profile  Geometry of the Cylinder  ^  ^  123 124  xix  ACKNOWLEDGEMENTS  I wish to express my sincere gratitude to Dr. P.G. Hill for his invaluable guidance, help and encouragements throughout all phases of the project and the writing of this thesis. Thanks are due to Bruce Hodgins, research enginer and project manager, for his advice and assistance in experimental work. Thanks also are due to Dale Nagata for the help in the data acquisition system, and all the technical staff of the Mechanical Engineering Department for the help in construction of the experimental apparatus. Special thanks to my fellow graduate students for their helpful advice, suggestions, to Yinchu Tao for his help performing the experiments, and to Patric Ouellette who helped taking the pictures for this thesis. A very special thanks to my wife, Lenny, and all the children (Novi, Yani, Andrew, and Lita), who were a constant source of moral support and motivation during my graduate work. Financial support for this work by the Eastern Indonesia University Development Project - Simon Fraser University is gratefully acknowledged.  1  1. INTRODUCTION.  A 'diesel-pilot gas injection engine' is defined as a diesel engine running on  high-pressure gaseous fuel which is directly injected into the combustion chamber together with a small portion of diesel fuel to initiate the combustion. This method of using natural gas in diesel engines has been used before but not with the same injector for both fuels. Note that a 'diesel-pilot gas injection engine' is not a 'dual-fuel' engine since it can not run on either diesel or gas alone. It may operate with a wide range of gas/diesel fuel ratios. The general purpose of this investigation was to study the performance, and the combustion and emission characteristics of a single-cylinder diesel-pilot gas injection engine. The intent was to provide a method for investigating the effect of injector design on engine performance and emissions in an effort to meet the new emissions regulations for heavy duty bus engines.  1.1. Background.  Since its invention about 100 years ago, the diesel engine has become the most efficient prime mover that has found wide application. It is used for stationary purposes, such as in pumping stations, as well as in land and sea transportation. The use of diesel engines in urban buses and trucks is facing stricter environmental  2 regulations than in the last decade. A new set of emission standards for urban bus heavy-duty engines, and heavy-duty truck engines has been established by the Environmental Protection Agency (EPA) of the United States (Appendix A). In 1993, for urban bus engine these standards require a reduction in allowable oxides of nitrogen (NO,) and particulate matter (PM) to a level that is less than 5.0 g/bhp-hr and 0.1 g/bhp-hr, respectively. For trucks, the same emission concentration levels will be applied in 1994. An indication of what this means is shown in Fig. 1.1 which illustrates the typical status of a conventional diesel engine emissions in terms of its PM and NO. concentration towards the legislated standards.  0 . 6 m.11■1■11M, 0.5 .4  .2 o. .c  -  0.3ca tcc^0 . 2 'et a. 0  Production Engines  NO: - PARTICULATE TRADE-OFF HEAVY DUTY TRANSIENT CYCLE  -  0. 1 -  •••■ •••■•..  lofts  •••■• •■■■• ammo •■••• May...11s. a  •••••  1994 Legislated Standard OEM =NM Ogia 411■1  0 .0 0  1  Development Engines  ^ ^ ^ ^ 2 3^4^5 6 7 NOx (g/bhph)  Figure 1.1 : Status of Conventional Diesel Emissions relative to U.S EPA Emissions Standard.  3  Particulate matter consists of soot on which some organic compound have become absorbed. Most of it results from incomplete combustion of fuel hydrocarbons. The primary constituent of NO is nitric oxide (NO) which forms throughout the high-temperature burned gases behind the flame through chemical reactions involving nitrogen and oxygen atoms and molecules, which do not attain chemical equilibrium. The other NO constituent is nitrogen dioxide (NO 2 ) which is rapidly converted from NO in the flame zone. As can be seen in Fig. 1.1, the new emissions standards are quite challenging to the diesel engine manufacturer. With present technology, it seems that it will be difficult to meet the EPA 1993/4 particulates and NOx emission standards with conventional diesel engines. Alternative solutions should be developed if diesel engines are to power future urban buses and trucks. One possible solution is to use natural gas as an alternative fuel for the diesel engine. Natural gas fuelling can solve the diesel particulate emission problem; it can also offer the flexibility to apply NOx reduction strategies which will not work with conventionally-fuelled diesels. Besides its potential for reducing particulate emissions, natural gas is preferred because of its ready availability, and relatively low cost. However, natural gas has the disadvantage of low volumetric heat content even when compressed as a gas to 20 MPa. The primary drawback to the use of compressed natural gas in transportation is that massive and bulky tanks are required to store enough compressed natural gas for reasonable vehicle range. This fuel storage problem puts a high value on increasing efficiency with natural gas.  4 1.2. Use of Natural Gas in Diesel Engines.  There are three principal methods of using natural gas in diesel engines as shown schematically in Fig. 1.2. In the following a brief explanation of each method will be presented together with a short description of previous experiences. A more detailed descriptions of the combustion characteristics and previous work on these topics will be given in Chapter 2. The first method is known as natural fumigation. As shown in Fig. 1.2a, the natural gas is injected into the inlet manifold. It mixes naturally with the inducted air, forms a fully pre-mixed fuel-air mixture which then enters the engine. Thus in the compression stroke this mixture will be compressed instead of air alone. This situation does not favor the two-stroke engine due to a considerable fuel loss which would occur in the scavenging process. The advantage of the natural fumigation method is associated with its simplicity. This method is relatively easy to adapt to an existing diesel engine. However, previous experience on using this method in four-stroke diesel engines showed that it generally has serious drawbacks during part load operation. Unless a high proportion of diesel pilot is used emissions are unsatisfactory and thermal efficiency is low due to incomplete combustion. In addition knock may be encountered as a consequence of premixing and high compression ratio. For these reasons this method is applied mainly to four-stroke stationary diesel engines which operate at relatively constant load and speed.  5 NATURAL GAS PREMIXED FUEL/AIR  AIR GAS MIXER  DIESEL FUEL  <A) NATURAL FUMIGATION  GAS INJECTION^I DIESEL FUEL AIR  (B) TIMED PORT INJECTION  HIGH—PRESSURE NATURAL GAS  DIESEL FUEL  AIR  (C) DIRECT INJECTION  Figure 1-2 : Methods of Using Natural Gas in Diesel Engines.  6  The second approach is known as timed-port injection. This method is a step ahead of the natural fumigation for this method would allow the used of natural gas in two-stroke diesels as well as four-stroke. Figure 1.2b shows schematically how this method works. In this second approach, medium-pressure (perhaps 20 atm) natural gas is directly injected into the inlet manifold close to the inlet valve. With precise timing and little mixing in the cylinder, a stratification of the gas in the cylinder could overcome some of the disadvantages encountered at part-load operation in the natural fumigation method. However, work reported to date on timed-port injection shows that knock and incomplete combustion at low loads are still a concern. The last method of using natural gas in diesel engines is direct injection of high-pressure gas into the combustion chamber as shown in Fig. 1.2c. High-pressure gas is injected directly into the combustion chamber near the end of the compression stroke. Thus a full stratification of the fuel-air mixture can be obtained with good flammability over the entire load range and with only small quantities of pilot diesel fuel needed. Successful operation has been demonstrated in applying this method in a medium speed diesel railway engine. In this case separate diesel and natural gas injectors were used with as low as 2% diesel pilot injection quantity. Previous work on marine diesel engines has shown that natural gas directly injected in the cylinder coupled with diesel pilot ignition can provide high efficiency and low emissions. It appears that direct injection of high-pressure natural gas with diesel pilot ignition is one of the most promising ways of meeting the 1993/4 EPA emissions standard. The  7 disadvantage of this option is the need for pressurization of the injected gas. A research group at UBC is addressing this problem by development of an intensifier/injector concept.  1.3. Objectives of This Research.  This research includes both experimental and analytical work on a diesel-pilot gas injection engine, and has the following objectives : (a)  To measure the engine performance over a wide range of load. The dimensionless parameter used to measure the engine performance in this work is thermal efficiency. It is defined as the ratio of the work produced per cycle to the amount of fuel energy supplied per cycle that can be released in the combustion process.  (b)  To determine the rate of combustion by analysis of cylinder pressure development.  (c)  To investigate the dependence of emissions on operating conditions.  (d)  To provide a method for investigation of the effect of injector design on engine performance and emissions.  1.4. Methodology.  A two-stroke Detroit Diesel of series-71 engine was used. This type of engine was chosen because of it is the engine type most widely used for transit buses in North America. The test engine has been instrumented and converted to an electronically controlled test engine, and equipped with a computerized data acquisition system. Performance and emissions data acquired are directly processed by the data acquisition software used, and  8 can be recorded on a floppy disk. Pressure development data can also be taken, viewed, and recorded for further analysis. Cylinder pressure data taken at specified engine operating conditions is then analyzed to revealed its combustion rate using a fuel mass-burned fraction program that is modified for this specific engine from an existing program.  9  2. LITERATURE REVIEW.  This chapter reviews some terminology and the available literature on natural gas fueling of diesel engines and on methods of estimating the mass-burned fraction. Sec. 2.1 reviews briefly the history of the natural gas fuelling of diesel engines and clarifies some of the terminology. Sec. 2.2, Sec. 2.3, and Sec. 2.4, briefly discuss the performance and the combustion characteristics of the diesel engines as affected by the three different methods of natural gas fuelling. Previous work on each method will be reviewed in the same section. The fifth section reviews methods of estimating the mass-burned fraction. Then, a summary will be presented before moving to the next chapter which will describe the experimental apparatus.  2.1. The Diesel Engine. The diesel engine is the internal combustion engine which was invented by Dr. Rudolph Diesel in 1892. His invention appeared only a few years after the first prototype of the Otto spark-ignition engine ran in 1876. In the spark-ignition engine, air is throttled so that a nearly stoichiometric fuel-air mixture is available to feed the engine while the combustion is initiated by an electric spark. In the diesel engine, only unthrottled air is inducted and compressed. The fuel is injected into the hot air near the end of the  10 compression stroke. Combustion is initiated by self-ignition of the injected fuel. Since only air is inducted to the cylinder, the compression ratio in the diesel engine can be much higher than that for the spark-ignition engine. For that reason, and because it operates without throttling losses, the thermal efficiency of the diesel engine is considerably higher than that of the spark-ignition engine. Conventionally, diesel engines are fuelled by liquid fuels derived from petroleum and known as diesel oils or diesel fuels. The diesel combustion process is considered to start after the beginning of fuel injection. As diesel fuel is injected, it atomizes and penetrates the hot air in the combustion chamber. The fuel vaporizes and mixes with the air. After a delay period, which is termed the ignition delay, spontaneous ignition of portions of already mixed fuel and air occurs because the air temperature are above the fuel ignition point. After ignition occurs, the flame propagates through the region in which fuel vapour and air are sufficiently mixed. Compression of the unburned portion of the charge (due to burning and piston movement) shortens the delay before ignition for the fuel and air which has mixed to within combustible limits and then burned rapidly. It also reduces the evaporation time of the remaining liquid fuel. Injection continues until the desired amount of fuel has entered the cylinder. Atomization, vaporization, fuel-air mixing, and combustion continue until all the fuel passes through each process. Mixing of the air remaining in the cylinder with burning and already burned gases continues throughout the combustion and expansion processes. Gaseous fuel has been used with supplementary ignition means in diesel engines for more than 60 years. The necessity of other ignition means for a natural-gas-fuelled diesel  11 engine was experienced by the C.& G. Cooper Company in 1927, as reported by Boyer and Crooks [11 1 . This is because natural gas by itself will not self-ignite in the combustion chamber of diesel engine at normal compression ratios. As a test, a small portion of diesel was also injected to self-ignite and subsequently ignite the natural gas which was injected at high pressure (1000 psi or about 7 MPa) at the end of the compression stroke. The small portion of diesel injected here is termed as pilot diesel. It was a successful attempt, and this event can be considered as the birth of the 'gas-diesel' engines. For some years this method was not popular because it was considered that the high pressure gas injection equipment needed were costly, and rather difficult to maintain. A diesel engine is defined as a 'dual-fuel' diesel engine if it is designed to operate on either diesel oil or natural gas, or both at the same time. Its mode of operation is defined as dual-fuel if the two fuels are used at the same time, and straight diesel if only diesel oil is used. Using this definition, a gas-diesel engine is then similar to a dual-fuel engine. It is important to note here that a dual-fuel diesel engine requires a different means of introducing the gaseous fuel into the combustion chamber in order to provide fuel flexibility in using either gaseous or liquid fuel. In contrast, in this experimental work, both the gaseous fuel and the diesel pilot are injected simultaneously through the same injector. This kind of diesel engine can not operate solely with gaseous fuel so it can not strictly be classified as a dual-fuel diesel engine. Thus, it is necessary to define another term. A 'diesel-pilot gas injection engine' is defined as a diesel engine which is fuelled with  Number in the bracket is the number of the reference as listed in references.  12 gaseous fuel and uses a small amount of diesel fuel to initiate combustion. A single injector is used to inject gaseous fuel and diesel-pilot into the combustion chamber. Combustion occurs in the conventional diesel engine combustion within small zones where the fuel-air ratio is suitable for combustion. The use of natural gas as the main fuel in the dual-fuel diesel engine consequently affects the combustion process. The combustion characteristics in dual-fuel operation differ from those of diesel operation, and depend on which method of using natural gas is applied. As briefly discussed in Sec. 1.2, we can distinguish three different methods of using natural gas in diesel engines. The three following sections discuss the combustion characteristics as well as review previous work on each method.  2.2. Natural Fumigation Method.  As shown in Fig. 1.2a, in the natural fumigation method the engine ingests a fully pre-mixed air/fuel mixture. Obviously the combustion occurs in a nearly homogeneous fuel-air mixture. This mixture is first compressed by the piston movement which increases the mixture temperature and pressure but ideally not high enough to cause auto-ignition. Near the end of this compression, a small amount of diesel fuel is injected into the hot homogeneous gas-air mixture. The injected pilot diesel subsequently goes through the ignition delay before it disintegrates into diesel vapour and ignites to initiate flame fronts which propagate through the gas-air mixture. Thus, the propagation of flame fronts is largely responsible for subsequent combustion of the remaining gas-air mixture. In part-load  13 operation, there will be a failure in flame propagation due to mixture weakness. In such a case misfiring will occur or some of the fuel will survive the combustion process. Natural fumigation as a concept of using natural gas in diesel engines has been investigated by a number of researchers over the decades. Much work on this concept has been done. A review of problems associated with the application of natural fumigation in dual-fuel diesel engines was made by G.A. Karim in 1983 [2]. He recognized that the performance of a dual-fuel diesel engine of this type at light load was inferior to that of the conventional diesel due to poor combustion of lean mixtures. He concluded that using larger pilot quantity, preheating the charge, partly restricting the air flow, recirculating exhaust gas, or using a combination of these measures, could partially improve the light load performance of this dual-fuel engine. The other problem encountered with natural fumigation engines discussed was knock, which is observed when very high outputs or very high intake temperatures and pressures are involved in the engine operation. The knock phenomenon encountered here is of an autoignition nature. The onset of knock can be delayed by employing a lower compression ratio and slightly later fuel injection. However, the engine will lose some of its thermal efficiency if the compression ratio is lowered. The two problems discussed in the G.A. Karim [2] review paper are typical problems encountered by a natural fumigation type dual-fuel diesel engine due to the compression and the combustion of a fully pre-mixed fuel/air mixture. Work using the natural fumigation method on a Caterpillar 3304 engine [3] [4], in 1985 and 1986 respectively, showed that the method has serious drawbacks at part-load.  14 The test engine was a four-stroke turbo-charged diesel engine which has a 93 kW rated output at 2000 rpm. The engine is mechanically controlled. It was found that the natural fumigation method was inherently unsuitable for part-load operation because of low flammability 2 , low efficiency, and excessive emissions compared to straight diesel operation due to poor combustion of the lean gas/air mixture. Throttling of the inlet air with existing turbocharged engines to improve flammability introduces the danger of compressor surge. In any case, throttling means pumping loss which will lower the thermal efficiency of the diesel engine. In other work on a naturally fumigated dual-fuel engine, a microprocessor was used to control the liquid pilot/ gas ratio as a function of load and speed [5] [6]. The test engines were a normally aspirated Caterpillar 3208 and a turbocharged John Deere 6466T diesel engine. Both are four-stroke engines. The amount of diesel and natural gas was varied, depending on engine speed and load in order to optimize the dual-fuel engine performance The results show that a significantly lower efficiency was encountered, while no emissions data were reported.  2  Flammability limits are defined [20] as the upper and lower limits of volume percentage composition of mixtures of fuel and air, within which flame propagation takes place when the mixture is ignited.  15 2.3. Timed Port Injection Method.  As shown in Fig. l .2b, with timed port injection relatively low-pressure natural gas is injected into the inlet manifold close to the inlet valve. It might be expected that some of the problems associated with compressing a fully pre-mixed fuel/air mixture which are encountered in the natural fumigation method can be eliminated or reduced. Precise timing might allow for stratification of the gas in the cylinder. Gas stratification, together with little in-cylinder mixing, would improve the mixture flammability under part-load conditions. If the injected gas mixes rapidly with the air, the engine will compress a nearly premixed fuel-air mixture and its combustion characteristics would not differ much from those of natural fumigation. In this case, it would be expected its combustion is also strongly governed by the propagation of flame fronts. Following work on this concept with the Mercedes OM-352 naturally aspirated diesel engine, it has been reported [7] that above 50 percent speed and load, the engine can be operated on gas fuel with an unthrottled air supply. This result suggests that timed-port injection does not fully overcome the disadvantages of natural fumigation. Thus, as discussed before, compression ratio reduction, throttling, or increased diesel proportion or a combination of theses measures may be necessary to provide adequate part load flammability of the gas mixture and prevent knock. The first two of these methods reduce thermal efficiency, with consequent penalty in fuel economy, and satisfactory emissions levels have not yet been reported.  16 2.4. Direct Injection Method.  As illustrated in Fig. 1.2c, the high-pressure gas is injected into the combustion chamber near the end of compression. This method of using natural gas in diesel engines enables a full stratification of the fuel-air mixture with good flammability over the entire load range and with only small quantities of pilot diesel fuel needed. In this method, air alone is compressed by piston movement. The combustion, in contrast to the two previous methods, is governed by the mixing process which is typical of straight diesel operation. The combustion process in the direct injection method of using natural gas is largely due to auto-ignition of diesel fuel; whereas that of the first two method operation depends heavily on both the auto-ignition characteristics of pilot diesel and the propagation of flame fronts. Of three previous investigations of this method which will be briefly reviewed in the following one was done on a locomotive diesel engine, and two on marine diesel engines, using separate injectors for gas and diesel oil. Early work in 1983 on large bore diesel engines [8] [9] has shown that high-pressure natural gas directly injected in the cylinder coupled with diesel pilot ignition can provide high efficiency and low emissions. Miyake et al [8] confirmed that performance with gas fuel almost equal to that of the oil burning diesel engine was obtained by adopting a combustion system like that of the usual diesel engine. The test engine used in their work was a single-cylinder marine diesel engine which had a bore of 320 min and a stroke of 450 mm. This engine had a rated power of 520 kW at 500 rpm rated speed. The natural gas used in the experiment contains 99% methane. Although the  17 authors obtained their results with a four-stroke diesel engine, they argued that similar results should also be obtainable with two-stroke diesel engines. Einang, et al [9] showed that the high pressure gas injection system is a viable concept and suited to dual-fuel operation with gas as the main source of energy. Their experiments were done on a single-cylinder, two-stroke marine diesel engine which has 300 mm bore, and 450 mm stroke. Rated output of the test engine was 375 kW at 375 rpm. The gas used is methane which is boil-off gas from LNG and injected into the cylinder from a gas injector. An electro-hydraulic system, which is electronically controlled, actuated the gas valve. Wakenell et al in 1987 reported [10] that a successful operation had been achieved using a separate diesel and natural gas injectors in a locomotive research engine. This is an Electro-Motive Division (EMD) 567B two-stroke, two-cylinder medium speed diesel engine, which has a bore of 8.5 inches and a stroke of 10.0 inches, with a compression ratio of 16:1. A liquified natural gas (LNG) fuel was vaporized before being injected directly at 6000 psi (or about 40MPa) to the combustion chamber near top dead centre on the compression stroke. It was found that rated speed and power were obtained with as low as 2% or 3% diesel pilot injection quantity without reducing compression ratio; although with slightly lower thermal efficiency.  23. Analysis of Combustion Rate. Combustion rate or fuel mass burning rate analysis is used to estimate from pressure  18 data the fuel mass-burned fraction during the combustion. The pressure data are the cylinder pressures (recorded at each crank angle degree) over the compression and expansion strokes of the engine operating cycle. Several means for estimating mass-burned fraction from pressure data have been proposed by researchers. Most of the methods are based on the result that for constant-volume combustion the mass-burned fraction x can be shown to be  P -  (2.1) P1  where P is the measured pressure, P 1 is the initial pressure, and P^is the maximum pressure achieved in the combustion. In the engine the volume changes but the pressure P., which would have been developed if the volume had stayed constant (without combustion) can be estimated from  Pte,,.  =pV  ^  (2.2)  in which n is the polytropic exponent. Thus to a first approximation,  ^P  V"  x  -  -  P. Vmez  P,V,"^  )"  -  P1 V1 "  (2.3)  where^is the maximum-measured pressure. The last equation was proposed by McCuiston et. al. in 1977 [11]. This method is very similar to that of Marvin [12] who in 1927 proposed a graphical method equivalent to Eq. (2.3) except that subscript 1 denotes the TDC volume in the above equations. Illustrating Marvin's procedure, Fig. 2.1 shows states s and e at the beginning and the end  19 of the combustion process, respectively, on a logarithmic PV diagram. An intermediate pressure Pb at state b is corrected to the equivalent pressure P c corresponding to TDC volume. Pressure correction is made to this volume by drawing a line parallel to the compression polytrope, i.e., points a, c, and d at TDC volume are the correction pressures for s, b, and e respectively.  ln(P  )  exhausts opens  intake closes TDC r=> In(V)  Figure 2.1 : Graphical Method of Estimated Mass-burned Fraction.[28]  Thus, according to Marvin, the mass burned fraction at crank position b could be approximated as Pc Pa  (2.4)  20  Pcyl  autasured pressure  ri  Pea  Peon.  Pai-1  ei-i^ei^soi => Crank Angle  Figure 2.2 : Determination of Combustion Pressure from Measured Pressure [16].  Another method which is based on constant-volume combustion and requires only cylinder pressure data was proposed by Rassweiler and Withrow in 1938 [13]. Their approach utilizes pressure corrected to constant-volume for each small crank angle step. Fig. 2.2 , shows how for a step from 0 ;_, to 0 ; the measured pressure P ei differs from the pressure P co which would have been obtained with compression only. The equivalent ,  1  combustion pressure rise AP, for this step is P oi - P a  ..,  . Rassweiler and Withrow showed  that for a number of such steps the mass-burned fraction can be calculated from  21  xe  E  e=e,  Pei Pcord  e„  Ve (2.5) V  E Pei 0=0, V1 Pcon)  where V 1 is the initial volume, and again subscripts s and e denote beginning and end of combustion. Shayler et al [14] in 1990 investigated the Rassweiler and Withrow method and addressed the problem of uncertainties in the value of polytropic exponent. They proposed a method of determining the correct polytropic exponent from the smoothened pressure data. Amman in 1985 [15] has reviewed these three methods and concluded that they are similar in approach, although the procedures are different. Since these three different methods need only measured pressures and corresponding cylinder volumes, they offer a relatively handy analysis. However, these approaches are based on approximations which could be avoided by using the energy equation, and avoiding the polytropic assumption (to correct for piston motion). Krieger and Borman in 1966 [16] presented in detail a computational method which attempted accurately to represent the thermodynamic properties of the cylinder contents. It computes the mass of fuel burned during each crankangle increment by solving the equations of energy and mass continuity together with the equations of state, internal energy. The disadvantage of this method, in addition to the approximation that the burned and unburned gases are handled as a mixture, is that it requires a correlation for gas-side  22 heat-transfer coefficient and an estimate of metal temperature for the exposed cylinder surface in order to account the effect of heat transfer. The amount of heat transfer from the cylinder contents to the wall Q w is given by  Q w = A h, ( T - T. ) (2.6) where h e is the convective heat transfer coefficient, T is the temperature of gases in contact with the cylinder wall, T. is the wall temperature, and A is the cylinder wall area. One problem in estimating the amount of heat transfer from the cylinder contents to the wall is how to determine the temperature of the gases in contact with the cylinder wall. The methods available assume that all gases in contact with the wall have a mean temperature T. which is the mass-average or bulk mean gas temperature obtained from an equation of state with given pressure [17]. In the following, two of the widely used correlations are briefly reviewed. Annand [18], after reviewing the existing formulae for internal combustion engine instantaneous heat transfer rate in 1963, concluded that the experimental convective heat transfer data can be represented by a Nusselt-Reynolds Number relation in the following form.  Nu = a (Re)"^  (2.7)  where a is a constant having a value in the range between 0.35 and 0.8. In the above correlation the characteristic length is taken to be the cylinder bore B, and the Nusselt Number and Reynolds Number are expressed as follows.  Nu=kB/k Re = p S p B /  23 where he is the convective heat transfer coefficient, k is the thermal conductivity of the mixture, p is density of the mixture, S p is the mean piston velocity, and IA is the viscosity of the mixture. In 1967, Woschni [19] proposed an equation for determining the heat transfer coefficient for internal combustion engines. His correlation is identical with the dimensionless equation N. = 0.035 (Rd"  (2.8)  These two methods of determining the convective heat-transfer coefficient are of similar form , differing primarily in the constant and the exponent of the Reynolds Number. Both models use bulk mean gas temperature as an approximation of the gas temperature in contact with the wall. The uncertainty associated with this assumption constitutes considerable physical uncertainty (perhaps as much as 20%) in estimating the instantaneous convection heat transfer in the internal combustion engine. After reviewing the available heat transfer model available in 1987, Borman and Nishiwaki [16] concluded that "a satisfactory predictive heat transfer model for engines has not yet evolved".  2.6. Summary. It has been shown experimentally that the part-load performance of dual-fuel with natural fumigation diesel engines is inferior to that of the conventional diesel due to poor combustion of lean mixtures. Using microprocessor control of diesel pilot fuel it is possible to avoid knock and maintain part-load fuel economy, but at the expense of large diesel fuel consumption at low load. The second method, timed port injection, was expected would  24 overcome the disadvantages encountered in the natural fumigation method. However, work to date on a coach diesel engine does not indicate that timed port injection fully overcome the disadvantages of the first method. It is unlikely that adequate part-load efficiency will be obtained with natural fumigation or timed-port injection method, or that these two methods will offer the emissions reduction of direct injection of natural gas into the cylinder. Work on large marine diesel engines and railway engine suggests that direct injection method can provide high efficiency and low emissions. Work on the direct injection method indicates that the traditional diesel mode of combustion (stratified fuel charge) could be employed with natural gas, so that the efficiency advantage of the diesel engine can be preserved. In previous works different injectors were used for natural gas and diesel fuel. However, no literature found on employing a two-fuel unit-injector which simultaneously injects natural gas and diesel fuel in medium-size bus and truck diesel engines. As a tool, analysis of combustion rate from the calculated mass-burned fraction gives a pattern of how combustion occured. There are several methods for estimating massburned fraction which could be classified into two categories, i.e., methods using only pressure crank-angle data, and methods employing the mass and energy equations and cylinder pressure information.  25  3. EXPERIMENTAL APPARATUS  3.1. Introduction.  The purpose of this chapter is to describe the test engine, the apparatus, and the data acquisition system. The test engine was fuelled with compressed natural gas and diesel fuel used as an ignition source (diesel pilot). It was coupled with a water-brake dynamometer, and fully instrumented. Principal measurements are torque, engine speed, fuel mass-flow, cylinder pressure, and emissions. The first three operating variables were needed to calculate brake thermal efficiency which is used to describe engine performance. Cylinder pressure data were taken in order to determined combustion rate. Ambient conditions were measured to determine the correction [201 applied to the calculated power. The composition of NO R , CO, HC, CO2 , and 0 2 of the exhaust gas were measured to determine dependence of emissions on operating condition.  26  Figure 3-1 : Schematic of Experimental Apparatus and Instrumentation.  The arrangement of the apparatus and instruments is shown schematically in Figure 3.1. The test engine used was a single-cylinder, two-stroke, diesel engine. A water-brake dynamometer was connected to the engine to regulate load. The fuel system consists of two sub-systems, i.e., the compressed natural gas sub-system and the diesel pilot sub-system.  27 Each has a mass-flow measuring instrument. A laminar flow element was installed in the intake line to measure intake air flow. Ambient pressure and temperature sensors were positioned close to intake air filter. Exhaust gas was analyzed by flowing it through a console. This console has a gas sampling system which is capable of measuring five exhaust gas compositions simultaneously. A data acquisition system which consists of a multiplex box, a data acquisition card, an ISAAC% and an IBM2 compatible standard personal computer (IBM-PC) was provided. The engine test bed together with almost all the instrumentation is located in the engine test cell. Engine operation and data acquisition were controlled from a control room next to the test cell. Data were classified into two categories: performance data and pressure data. The first one is related to performance and emissions; these data are time-averaged steady state data taken at particular operating conditions. The pressure data are cylinder pressure measurements taken at relatively high frequency, e.g., every crank angle degree. Both are taken at steady engine operating condition, and both are handled by the data acquisition system. An Ethernet system was added to the IBM-PC to allow a transfer of data from the data acquisition system to a main frame computer. The data flow will be elaborated later. All instruments installed are electronic except the ones used to read the barometric pressure, relative humidity and dry bulb temperature. They were selected to fit the data acquisition system adopted. Calibrations of some principal instruments are documented in  The trade mark of a high speed data acquisition computer. 2  IBM = International Business Machines. A company name.  28 the Appendix B. The following section gives details on the test engine used and its test bed. Fuel system, which is considered one of the most important engine operation system, is discussed afterwards. The testing and calculation procedure will be covered in the next chapter.  3.2. Engine and Test Bed. As stated in Sec. 1.4, a Detroit Diesel Series 71 single-cylinder diesel engine was used in this experimental work. The engine is a two-stroke naturally aspirated, direct injection diesel engine which has specification [36] as shown in the following table.  Bore x Stroke^ 4.25 in.(108 mm) x 5.0 in.(127 mm) Displacement^ 70.93 cu in. (1.162 liters) Rated Output^ 15 HP (11.2 kW) @ 1200 RPM Rated BMEP^ 70 psi (4.8 bar) Compression Ratio^16 : 1 Fuel Injection^ Direct, Unit Injector Scavenging Type^Uniflow No. of Intake Ports^18 No. Exhaust Valve^2 (two)  Table 3.1 : Engine Specifications.  29 The engine had been used to power a 10 kW alternator at 1200 RPM with mechanical governor control of a unit-injector. This fuel injection system has been converted to electronic control to gain accurate control and variability of the injection timing and the amount of fuel injected to the combustion chamber. Figure 3.2 shows the test bed installation in a UBC engine test cell. The test engine (1) was mounted on a movable test bed. It was coupled with a Go Power Model DA316D water-brake dynamometer (2). A cooling tower (3) has been installed to replace the radiator so that the cooling fan could be removed and so that closer control of engine temperature could be achieved.  Figure 3-2 : The Test Engine and Dynamometer in a Cell. (1) Test Engine, (2) Water-brake Dynamometer, (3) Cooling Tower, (4) Dynamometer-controlled Valve  30  Figure 3.3 : The Engine Control Panel and Data Acquistion System. (1) Dynamometer Control Module (DCM),(2) Data Display Module (DDM).  The engine control panel and data acquisition system are in the control room. Figure 3.3 shows a photograph of these equipment. A Dynamometer Control Module (DCM) and a Data Display Module (DDM) from Environmental Dynamics Inc. are used. These two modules are labelled as (1) and (2) respectively, and are parts of the engine control panel also shown in Figure 3.3. The load and the speed of the engine are controlled through the variable resistor on the DCM front panel. The DDM displays engine speed, dynamometer load, fuel rate, and one out of eighteen temperature options. The DCM also has a safety  31 shutdown feature. If one of the engine operating indicators, e.g., engine cooling water temperature, oil temperature, and oil pressure, is higher than a preset level, the DCM will shut the engine down by closing the fuel supply valves. Resistance of the brake is controlled by adjusting the water flow rate passing through the dynamometer. An electro-pneumatically operated water inlet valve, (4) on Figure 3.2, linked to a load setting dial in the control panel is used. The capacity of the dynamometer is 130 lb.ft (176.2 N.m) at 1500 rpm or 57 lb.ft (77.3 N.m) at 1000 rpm. Engine torque is measured by load cell attached to the dynamometer case. A magnetic pickup type speed sensor develops a frequency signal (1 Hz = 1 RPM) which is then converted to an analog voltage proportional to speed. Both load (torque) and speed signals are sent to data acquisition system and are displayed on the engine control panel.  3.3. Fuel System. As shown in Figure 3.1, the engine was provided with gaseous as well as liquid fuel for its operation as a dual-fuel diesel engine. Unlike conventional dual-fuel diesel engines, compressed natural gas is injected directly into the combustion chamber. Natural gas does not ignite easily. Therefore, a small portion of diesel fuel is injected from the same injector to self-ignite and then ignite the natural gas. The following two sub-headings will elaborate on the fuel supply and fuel injection systems.  32 3.3.1. Fuel Supply System.  Figure 3.4 shows the fuel supply system which consists of two sub systems, i.e., natural gas, and diesel fuel supply sub systems. The properties of both fuels and a typical chemical composition of the natural gas used are found in Appendix C, and Appendix D respectively. BLD. GAS LINE PRESSURE REGULATOR COMPRESSOR  DIESEL FUEL .ELECTRONIC. I CONTROL1 L COMPRESSED NATURAL GAS  MASS-FLOW METER I SHUT-OFF  SHUT-OFF VALVE  VALVE  FILTER  PUMP  DUAL FUEL UNIT INJECTOR  V  Figure 3.4: Schematic of Fuel Supply System.  The diesel fuels used are a commercial Grade-2 Diesel Fuel Oil and a high-cetane number diesel fuel. The fuel is pumped from the fuel tank through a filter to a gravimetric mass-flow measuring instrument (AVL) before it is injected to the combustion chamber.  33 The return line from the injector is then directed to the AVL, so that the net consumption of diesel fuel can be determined. Compressed natural gas is drawn from commercial pressure bottles. Natural gas from the main supply at a pressure of 14 kPA is compressed to 19.0 MPa by a commercially available Residential Refueling Appliance (RRA). One bottle is used at a time by manually setting the appropriate valves. The compressor in the appliance is a four cylinder, 4-stage, air-cooled, oil-free unit. It is driven by a 1.5 HP electric motor and is equipped with a control system consisting of pressure and temperature sensors, relief valves and electrically operated valves. The compressor stops automatically when the pressure reaches the set level. Storage bottles are connected in parallel in a system which allows one cylinder to be used while the other cylinders are being filled. The supply gas pressure was manually regulated through a pressure regulator. A solenoid shut-off valve was installed in each of fuel supply line to stop the fuel flow to the unit injector in case of emergency. They are controlled by the engine's electronic control system which is an integral part of the fuel injection system.  3.3.2. Fuel Injection System. In the mechanically controlled injector of conventional diesel engines, injection timing and injection rate are mechanically controlled by ports and helices machined in the bushing and plunger assembly. An electronic unit-injection engine governing system known as Detroit Diesel Electronic Control (DDEC) was developed by Detroit Diesel [21] to  34 enhance flexibility and precision. This electronic system was used to control the engine in the baseline test series.  DIESEL FUEL  COMPRESSED NATURAL GAS  ^  I  ^ ■DEG. OF INJECTION ^ CRANK *PULSE WIDTH POSITION *DSL THROTTLE POS. ^SIGNAL  I INPUT SIGNAL COND, UNIT I I MICRO-CONTROLLER BOARD I  I OUTPUT DRIVE CIRCUIT I^ I ELECTR. DISTRIBUTOR LNIT  I ^- I HYDR. PRESS ^ SELEMED  I  STEPPER MOTOR  INJECTOR  Figure 3.5 : Schematic of Fuel Injection System.  A microprocessor controlled unit-injector for gas-fueling diesel engines has been developed at UBC Combustion and Alternative Fuel Laboratory which is part of an "Intensifier-Injector Technology" [22]. Figure 3.5 shows a schematic of this gas injector in the fuel injection system. The injector injected compressed natural gas and diesel pilot into the combustion chamber near top dead center. It utilizes a cam-driven plunger to provide high pressure diesel fuel to actuate the poppet valve which permits flow of the gas and entrained diesel fuel to enter combustion chamber near top dead center. The diesel pilot  35 fuel is gas-blast atomized by the natural gas flow past the diesel ports. A modification of the DDEC is adapted in developing an electronic control system for this two fuel unitinjector. The beginning of injection (BOI), and the duration of injection ( or pulse width,  PW ) settings are fed to the electronic control system. They were manually set through separate variable resistors mounted on the engine control panel. These input signals, together with crank position signal, were used to actuate a solenoid which will close a valve and consequently develops a hydraulic pressure (Fig. 3.5). The developed hydraulic pressure will open the poppet valve. High-pressure gas and gas-blast atomized diesel fuel enters the combustion chamber with high velocity in a form of conical sheet. A needle valve, which is coupled with a stepper motor, is used to throttle the flow to meter the amount of diesel fuel injected. Needle position was electronically controlled by a variable resistor. The amount of pilot diesel can be manually controlled through a variable resistor mounted on the engine control panel in this way.  3.4. Instrumentation. Four principal instruments used to measure engine performance, i.e., engine speed, torque, gas mass-flow, and diesel mass-flow, are described together with emissions instrumentation. Cylinder pressure instrumentation used is also discussed. Calibration procedure of most these principal measurements is presented. Other instruments were installed to support these measurements, or to monitor the engine working condition.  36 3.4.1.Engine Speed.  A speed sensor attached to the shaft of the dynamometer is used to measure engine speed. This magnetic speed sensor provides a signal frequency proportional to engine speed. The frequency signal is then converted into an analog signal which is displayed on the engine control panel and sent to a data acquisition system. A hand digital tachometer (Shimpo, model DT-205) was used in the calibration. This hand tachometer sends out a continuous light beam and counts the pulses reflected from a piece of reflective tape attached to the engine shaft. Calibration showed that the standard error of the measurement is 0.1% (Appendix B.1).  3.4.2. Torque (Load Cell).  A strain-gauge load cell mounted on the dynamometer casing is used to measure the load applied to the engine. The strain-gauge signals are processed and displayed on the engine control display. This low level signal is also amplified in a circuit and then sent to the data acquisition system. The load cell was calibrated by loading and unloading the load cell with standard weights placed on an arm bolted to dynamometer casing. A standard deviation of 0.1% was found in the calibration (Appendix B.2).  3.4.3. Gas Fuel Mass-flow Rate.  Compressed natural gas mass-flow is measured with a Micro Motion Model D12 mass-flow meter. This instrument works on the Coriolis acceleration principle. The gas  37 flows through a U-shaped tube which vibrates at a frequency which is directly proportional to the product of the fluid density and velocity. The frequency of the signal produced was converted to a 4-20 mA current signal and sent to the data acquisition system by a remote flow transmitter. The manufacturer's calibration for this meter was used, which showed that the averaged measurement error is ±0.4% .  3.4.4. Pilot Diesel Fuel Consumption.  An AVL Dynamic Fuel Consumption meter series 730 is used to measure diesel mass-flow rate. This determines flow rate by sensing the weight of fluid filling a vessel. Both the supply line and the return line of pilot diesel are connected to this vessel. The diesel fuel consumption is determined through the effective change in weight of the vessel.  The system permits measurement of the total consumption over a selected measuring period as well as the instantaneous fuel consumption. A five second measuring period is used for the measurements. The analog signal is sent to the data acquisition system, and its digital value is displayed on the evaluation module placed on the engine control panel. Calibration is done by weighing the fuel leaving the vessel in a set time period. It was found that the measurement standard error is 0.2% (Appendix B.3).  3.4.5. Emissions Intrumentation.  Emissions instrumentation used was assembled in 1985 at UBC [23]. It can measure the concentration of unburnt hydrocarbons, oxides of nitrogen, and carbon monoxide; it is  38 equipped with a gas sample handling system. Modification in 1990 [24] allows measurement of carbon dioxide and oxygen. However, the oxygen analyzer did not work properly. The four emission analyzers and gas cylinders needed for analyzer calibration and operation are mounted in a movable cabinet. Figure 3.6 illustrate the schematic diagram of the emissions console. Signals are sent to the data acquisition system.  To  Exiausr <zi STACK  G  FROM ENGINE  SAW LIN6 PROSE  OUTLET  Figure 3-6 : Exhaust Emission Sampling System.  ^  39 An exhaust sampling probe was inserted in the exhaust line at about 1.2 m from the engine manifold. A drawing of the sampling probe located in the exhaust line is documented in Appendix E. Calibration of each analyzer was made before taking data with span gas and zero gas. After taking the data, a calibration check were made to ensure that the analyzer works properly during measurements.  3.4.6. Cylinder Pressure.  An air-cooled PCB piezoelectric pressure transducer with maximum frequency of 80 kHz was used. The frequency used when taking one pressure datum at each crank angle degree at engine speed 1200 rpm is 7.2 kHz.  ^1  PRESSURE SIGNAL TO CHARGE AMPLIFIER  EXHAUST VALVE  ADAPTOR SLEEVE CYLINDER HEAD /  _L /^-1-- -'--- /X A^ l^I^ /^ 1 FUEL/ INJECTOR \-%.. I ...--'i^1 —...._ ■-123' ^ 1^ ^N. 1 A /^I^\ X^ \^‘^1^/ ‘^  / /1  /  INTAKE VALVE PRESSURE TRANSDUCER (PCB Mod. 112A05) SECTION A-A  Figure 3-7: Pressure Transducer Location in the Cylinder Head.  40 Figure 3-7 shows how the pressure transducer is mounted in the cylinder head through an adaptor sleeve. The sensing surface of the transducer is installed 3 0 mm above the fire deck. The adaptor passage has the same diameter as that of the sensing surface (5.6 mm) This mounting technique is applied to minimized the effect of thermal shock [25] [26] [27]. The signal was amplified in a Model 5004 Kistler Charge Amplifier and transmitted to the data acquisition system. The pressure transducer was statically calibrated using a dead-weight tester for a pressure range of 0 to 1900 psi. Standard error is 0.6% (Appendix B.4).  3.5. Data Acquisition System.  The two engine test cells, with one engine in each cell, are equipped with an IBMPC based data acquisition system [28]. This system is controlled by a program which allows the user to choose the engine, to configure the system to his needs and to calibrate input channels. The software does calculations, allows data to be saved to a disk, and converts cylinder pressure data from binary to ASCII 3 . Hardware includes a multiplex box, a PC which has a PCL-818/Data Acquisition Board, an IEEE-488/General Purpose Interface Board, and an ISAAC which is a computer to acquire pressure data. The multiplex box consists of a multiplexer board, DT-709/screw terminal board, "Vector" trigger board, and a digital interface board. It does all the  ASCII = American Standard Code for Information Interchange, which is a set of 256 codes that many computers use represent letters, digits, special characters, and other symbols. 3  41 switching in the data acquisition system. Figure 3-8 shows a schematic diagram of data flow.  IBM-PC  MULTIPLEX BOX  Tc_ Li  STEADY STATE DATA  MICRO NOTION FREQ/VOLT --al 0 DI21H-RFT9712 CONV ERT BEGINSIG OF INJECTION (Manual setting) ^ lb I AMBIENT AIR TEMP.^I Electr.Sensorl^ i• 2 PULSE WIDTH^(Manual setting) ^ • 3 TIME^LOAD CELL ^4 DCN SPED^ Sensor ^ .5 INTAKE PRESSURE^IDDEC Press.Sensor • 6 ^ AVL Fuel DIESEL MSS FLOW 7 Calc. Fuel Bat ^ ^ AIR FLOW (DELTA P) LanFlow lien. Delta Press. 8 GAS MASS FLOW  DISK DRIT; E PCL - 818^ (Data Acq. Board)  !  L  LI  I  GAS PRESSURE^[Heise 620 ^ #TURNS OF STEPPER (Manual setting) ^ EMISSIONS (NOKHC,OE1CIE02) I DOSSIERS CENSILE  HIGH SPEED DATA  ^  CR. ANGLE & BBC INDEX ^  ISIFTPOT SP-360I^ I PCB Piezo-Elect. I-4Charue Aro. ^ HYDRAULIC PRESSURE I PCB Piezo-Elect. It Par. SuDilv  9 10 1115 AUX.  i  I  1^"Cr  IEEE-488 /  I  ^I  L Ma  z  _4.  P6  _-n ^►  CYLINDER PRESSURE  ISAAC^I  Figure 3-8 : Data Flow in the Data Acquisition System.  Up to 16 performance data, e.g., engine speed, torque, etc., can be directed by the multiplexer board to the DT-709/screw terminal board (Fig. 3.8). The incoming signal was conditioned by the DT-709 with a gain of one. This board converts the 16 differential signals to 16 single-ended signals with a common ground. Analog voltage signals are then converted to a 12-bit digital number by the PCL-818 data acquisition board. Each of the 16 channels data is sampled 100 times, averaged, and standard deviation calculated.  42 Averaged data are used to calculate engine performance characteristics, e.g., brake power, brake mean effective pressure, thermal efficiency. Any eight of the steady state data and or calculated values can be displayed simultaneously on the PC monitor screen. Data can be saved to a specified floppy disk on command. Because it is acquired at a relatively higher frequency, the cylinder pressure data flows in a different way compare to that of the performance data. Crank angle signals, together with bottom dead centre (BDC) signals, are needed to clock data taking. An auxiliary multiplexer board directs these signals to a trigger board. When commanded to acquire pressure data, the PCL-818 sends a digital signal to the trigger board which prepares a trigger for the ISAAC to take cylinder pressure data. Using the next BDC signal after the power stroke as a trigger and crank angle as an external clock, the ISAAC will then start taking data. Up to 100 cycles of cylinder pressure data can be taken. Upon completion of this process, the ISAAC sends signal through the IEEE-488 General Purpose Interface Board, and data is transferred in binary form, one cycle at a time to the specified file. Converting binary pressure data to ASCII format can be done by selecting the relevant option on the main menu. An Ethernet system is employed to connect the PC to a main frame which allows transferring and processing pressure data for further analysis.  43  4. EXPERIMENTAL PROCEDURE  4.1. The Testing Procedure. Tests were divided into two groups. The first group was done with the diesel injector to obtain the diesel performance data. The second group used a prototype two-fuel unit injector. The injector has an approximately 10° poppet seat angle. The schematic of the injector is filed in Appendix F. Note that the gaseous and diesel fuels injection have a conical-sheet formed jet when leaving the injector. The conical sheet jet was interrupted by the castellated-shrouded the poppet since flow visualization [29] had shown that this configuration could enhance gas penetration in the cylinder which subsequently could affect combustion rate. The 50% shrouding' was devised so that there were six jets instead of a continuous conical-sheet jet. Natural gas injection pressure, diesel ratio', diesel-fuel cetane number, and certain engine operating parameters were varied, while engine speed was maintained at 1200 RPM to determine their effects on thermal efficiency, emissions and combustion characteristics. The variable parameters ranges are given in Table 4.1.  I 2  Approximately 50% of the initial jet-flow area blocked by the shroud. Diesel ratio is defined as the energy percentage of diesel pilot to total fuel energy supplied to the engine.  44  Injection Gas Pressure^50 ... 70 bar Diesel Ratio^  15 ... 25  Diesel-fuel Cetane Number 3^—45, or 62.2 Load (BMEP)^  0.5 ... 4.5 bar  Beginning of Injection (B01)^24 ... 40°BTDC  Table 4.1 : Testing Ranges.  Brake mean effective pressure (BMEP), which is defined in Section 4.2.2, is used to represent the load applied to the engine. Engine speed is kept constant by manually adjusting the fuel injection duration. Injection gas pressure is adjusted with a natural gas supply pressure regulator as shown in Fig. 3.4. Diesel ratio is adjusted by metering the amount of diesel fuel injected. The metering valve position is controlled by a stepped motor as described in Sec. 3.3. The beginning of injection (B01) indicates the crank position where the injector solenoid is energized which is followed by pressurization of the diesel fuel and simultaneous injection of both liquid and gaseous fuels. Prior to the tests of the gas injector, a series of tests were completed with the diesel injector. This first group of tests was done to acquire the diesel performance data against  3  Diesel-fuel properties are listed in Appendix C.  45 which to compare the data on performance, emissions for natural gas fuelling. The procedure below was followed in obtaining the gas-diesel engine performance data for the chosen gas injector at the specified engine speed.  1.  Set the gas pressure at a value, i.e., 50, 60, or 70 bar.  2.  Set the beginning of injection (BOI) at a value, i.e., 24, 28, 32, 36, or 40°BTDC.  3.  Set the diesel ratio by adjusting the pilot-diesel metering valve.  4.  Take data for the chosen combination of gas injection pressure, diesel ratio, and BOI at different loads in the range 0.5 bar to 4.5 bar (or the maximum achievable load) at constant speed.  5.^Repeat for all the other BOI.  The main purpose of engine performance evaluation is to determine the dependence of thermal efficiency on BMEP; this is termed the performance curve. For a particular gas injection pressure and a specific diesel ratio, the performance curve for every BOI setting was plotted. At a given load, the BOI which gave the best thermal efficiency was determined by cross-plotting BMEP vs BOI at that load. Best performance for a given gas injection pressure and diesel ratio is obtained by plotting 2 variables vs load (BMEP): maximum thermal efficiency, and best BOI. This curve termed "best BOI performance curve" and used to compare engine performance for different gas injection pressure, diesel  46 ratio, and cetane number. Pressure data were acquired at intervals of one degree crank angle. They were taken at low, medium, and maximum load in each group of tests. Pressure data were analyzed to show combustion rate by estimating the mass-burned fraction of the fuel at each crank angle interval. Combustion rate analysis is discussed in Chapter 6.  4.2. The Engine Performance Calculation Procedure.  4.2.1. Thermal Efficiency. Thermal efficiency ri th is defined as the ratio of the work produced per cycle W, to the amount of fuel enthalpy supplied per cycle ( h f = m f LHV ) We th  m • LHV  (4.1a)  where m f is the mass of fuel inducted per cycle, and LHV is the lower heating value of the fuel. By taking the time derivatives of the numerator and the denominator of the Eq. (4.1a), thermal efficiency can be expressed as  th - •  m  in which P is the power.  f  P • LHV  (4.1b)  47 Power was obtained from speed and torque measured with the water-brake dynamometer. Brake power Pb (kW) was calculated from engine torque Tb (N.m) and the engine speed N (rev/min or RPM), using : P = 2 It  60  T x10 -3 b  Then by substituting the above equation into Eq. (4.1b) (and considering there are j types of fuel burned) the brake thermal efficiency of the engine at a defined engine operating condition can be calculated as follows: 2 It  11  0 Tb  x10-3  E mf; LHVi / 3600  (4.2)  where N is the engine speed (RPM), T is the torque (N.m), riz.0 is the mass flow of the  th fuel burned (kg/hr) which are j=1 for diesel and j=2 for natural gas, and LHVi is the Lower Heating Value or the enthalpy of combustion of the i th fuel. The engine speed, the torque, and the mass flow rate of the diesel-pilot liquid and the compressed natural gas are the measured quantities which are acquired during the experiments. The enthalpy of combustion (LHV) of the diesel-pilot is 45220 kJ/kg, the natural gas is 49098 kJ/kg (Appendix C). Since it is not practical to run the experiments at the typical standard inlet air condition, a correction factor to the observed power was applied conforming to the SAE Standard J1349 JUN85 [20] to provide a common basis of comparison. The procedure to  48 compute the correction factor is filed in Appendix G. This correction factor is also applied to the calculated thermal efficiency or the engine performance and the specific emissions.  4.2.2. Brake Mean Effective Pressure (BMEP). The mean effective pressure MEP is defined as the work per cycle W e per unit cylinder displacement volume Vd. This parameter is a performance measure independent of engine size and which expresses engine load. MEP =  ^ V  (4.3)  Substituting the work per cycle with the engine speed N (rpm), and brake torque Tb (N.m), the bmep (bar) of the engine at a defined operating conditions can be calculated using: BMEP -  where the cylinder displacement volume  Vd  2  1G Tb  X10^  Vd  (4.4)  is in cm 3 , and 1 bar is 100 kPa.  4.2.3. Brake Specific Emissions. The measured concentrations of gaseous emissions in the dry exhaust gases are oxides of nitrogen ( NO, and NO, grouped together as NO„ ), carbon monoxide (CO), unburned hydrocarbons (HC), and carbon dioxide (CO 2). They are measured in parts per million except for CO, which is measured in percent by volume. The brake specific emissions is used to indicate the level of emissions since it is a  49 more comparable indicator. Brake specific emissions b se are defined as the mass flow rate of pollutant M E per unit brake power output P b .  bse  ^ m e^  (4.5)  Pb  The following assumptions are made in calculating the brake specific emissions of NO N , and the unburned HC.  1.  The dry exhaust gases consist of CO 2 , CO, 0 2 , N2 , NO N , and unburned HC.  2.  The unburned HC has the same composition as that of the compressed natural gas fed into the engine.  3.  The NO is considered to be NO.  4.  The intake air consists of 79.0% N 2 and 21% 0 2 by volume, and is considered as an ideal gas.  The first assumption was made because the concentration of the other constituents of the exhaust gas, e.g., 0, OH, H, and N, are very small compared to the concentrations of 0 2 or N 2 . Since the concentration of the emissions gas are measured volumetrically, conversion of these pollutant volumetric flows into mass flows are needed. Dry exhaust  50 mass flow rate must be calculated first. Dry exhaust gas mass flow rate th exh is calculated from the intake air mass flow  rate in , which is assumed to be an ideal gas and water free, and the total fuel mass  flow rate E thfue by conserving mass flow.  m ph = m ay + E  fuel  (4.6)  Compressed natural gas mass flow rate and diesel-pilot mass flow rate are individually measured. Mass flow rate of intake air is calculated from its standardized measured (volume) flow rate 1)41' since its density at the adopted instrument standard condition  Pt,P  is known. Their relationship is =n t,P r  Standardized intake air flow rate is computed from actual flow rate  (4.7)  1. %Pa  by  applying the correction factors for pressure Pe f and temperature T a . =  'ta,Pa  Pcf Tcf^  (4.8)  A Laminar Flow Element (LFE) is used to determined the actual flow rate of the incoming air (Fig. 3.1). Output of LFE is a pressure difference which is translated into flow rate  51 using the manufacturer calibration curve (Appendix B.5). The base standard condition of the curve is 70°F(21.1°C) and 29.92"Hg(101.03 kPa). The correction factor for pressure is  _  Pa  cf^101.03  ,and for temperature [30] is 294.1 T,„ - [ ^ [ 1 - 0.00231 (^- 21.1 ) ] ta +273  where P. , and t o are ambient pressure and temperature respectively. The first factor in the temperature correction adjusts the volume changes due to temperature. The second factor in the temperature correction account for the dependency of viscosity on temperature which is a linear fit applied for range of 20°C to 40°C. Intake air mass flow is calculated using Eg. 4.9 obtained by substituting standardized intake air flow rate in Eq. (4.5) with actual flow rate as given in Eq. (4.8).  =p  tp  i,rapa  [(1 -0.002317(ta -21.1)] [  2.1^Pa ta 9473 +2 [ 101.03]  (4.9)  The next step is converting the mole fraction of a pollutant Y i , e.g. NO R , CO, or unburned HC in the exhaust gas, to its mass flow rate using the equation,  (4.10)  52 where M 1 is the molecular weight of the P h pollutant species. Then by inserting the mass flow rate of the pollutant calculated from Eq.(4.10) into the Eq.(4.5), the brake specific emissions of the pollutants are : ihNO.  Pb bsC0  CO  Each has the unit g/kW.h  (4.5b)  Pb (4.5c)  h  v'snc  (4.5a)  Pb  53  5. MEASUREMENTS OF ENGINE PERFORMANCE  5.1 Conventional Diesel Performance. Conventional diesel performance (termed diesel baseline) was taken with the electronically controlled diesel fuel injector installed. Gas-diesel test results were compared to this baseline. As described in Sec. 4.2, the engine performance is expressed by its brake thermal efficiency at different loads and plotted as a thermal efficiency' versus BMEP (defined in Sec. 4.2); this is termed a performance curve. The corresponding engine exhaust emissions, e.g., oxides of nitrogen and unburned hydrocarbons, are presented as brake specific emissions (defined in Sec. 4.2.3). Since the diesel-fuel injector was electronically controlled the amount of diesel fuel injected into the combustion chamber and the injection timing, i.e., the beginning of injection timing (B01) and the duration of injection (PW), are automatically updated from time to time to keep the engine running smoothly at its best performance for a given load and speed. The performance curve corresponding to best operating conditions for this engine is termed the best BOI performance curve or performance curve for short.  The term thermal efficiency used for the rest of this thesis implies the brake thermal efficiency.  54  28 A 26  0.9  24  0.8 0 17 0.7  22 20 18  0.8 0  16  0.5  14  0.4 5  12  0.3  10  0.2  8  0.1  6  ct  cc -J w  0 BMEP ( bar ) r>  Figure 5.1 : Diesel Baseline Performance Curve  The diesel engine performance curve is presented in Fig. 5.1, while Fig. 5.2 and Figure 5.3 show the corresponding emissions. As shown in Fig. 5.1, the maximum achievable load at 1200 RPM with the diesel injector was about 4.5 bar which is 6% lower than the engine specification. This test engine (manufactured in 1939) has a relatively low compression ratio (16:1) and low load capability compared with currently typical naturallyaspirated two-stroke diesel engines which have BMEP values in the range 7 to 9 bar [31]. It was found that the engine maximum thermal efficiency is 25.7%.  55 Fig. 5.1 also shows the fuel-air equivalence ratio of the fuel-air mixture in the combustion process. The fuel-air equivalence ratio 4) is the ratio of the actual fuel-air mass ratio to the stoichiometric mass ratio which is the theoretical proportions of fuel and air. It is shown in Fig. 5.1 that the equivalence ratio of the diesel ranges from 0.23 to 0.69, which means the diesel engine operates with excess air of 335% to 45% from low to high load.  A  60  1100  55  1000  50  900  -C^45 co  0  800  40 35  700  30  600  25  500  20  400  15 10  300 EPA 93 /  200  5 0  a  4  BMEP ( bar ) L^> Figure 5.2 : Diesel Oxides of Nitrogen Emission.  100  A  Eca. CL X  0  •  56  Concentration of oxides of nitrogen in the exhaust gas increases with load in the range of 440 to 995 ppm as shown in Fig. 5.2. This is higher than the standard. At low load (BMEP — 1 bar) it is about three times higher, while at high load (BMEP — 4 bar) it exceeds the standard by approximately 80%.  5.5  120  5  115  ff^4.5  -C^4  105  3.5  100  C.)^3  F  2.5  85  2 ▪  1.5  75  0.5  70  0  4  1  I  80  EPA '93/ '94  1  0  A  110  5  65  BMEP ( bar ) ^-,›  Figure 5.3 : Diesel Unburned Hydrocarbons Emission.  As shown in Figure 5.3, the unburned hydrocarbons concentration in the diesel emissions vary with load in the range from 75 to 115 ppm. From medium (BMEP 1.5 Bar) to high load the concentration meets the standard, although at low load it falls considerably above the standard.  57 5.2 Natural Gas Diesel Performance. The best BOI performance curve is used to represent the diesel baseline performance. The gas-diesel performance at a given gas injection pressure and diesel ratio is also presented in best BOI curve. However, the injection timing in the gas-diesel engine is manually adjusted to run the engine smoothly for a given load and speed. The testing procedure decribed in Sec. 4.1 was followed.  4^4 BMEP ( bar )  Figure 5.4 : Determination of Best BOI Performance. ( Gas injection pressure 60 bar, diesel-ratio 25% DF2, Shrouding 50% )  In each test, a set of performance data includes the whole range of BOI and load listed in Table 4.1. The individual BOI performance curve was plotted. By cross-plotting  58 the maximum thermal efficiency and BOI versus BMEP, the best BOI performance curve is obtained. Fig. 5.4 shows the typical individual BOI performance curves and the BOI performance curve that represents the engine performance at a given gas injection pressure and diesel ratio (defined in Sec. 4.1).  5.3 Effect of Gas Injection Pressure. Tests to determine the effect on thermal efficiency of gas injection pressure were performed for the whole range of load while keeping the engine speed and the diesel ratio constant. Fig. 5.5 shows the effect of injection pressure on thermal efficiency for the whole range of load with diesel ratio maintained at 25%. The tests for three different gas pressures exhibit improvements of the maximum load capability when the gas pressure is increased. This is not surprising since higher gas pressure, to a certain degree, increases the gaseous fuel mass flow with the same injection duration. With either 50 or 60 bar gas injection pressure, thermal efficiency appears to be about the same up to medium load (BMEP — 2.5 bar); although with 70 bar injection presssure its thermal efficiency is 2-4% lower than with 50 or 60 bar. However, these results clearly indicate that high thermal efficiency (exceeding the baseline by about 2%) is obtained at high load with high injection pressure.  59  30 ^ 28A 28  2422 2018-  us-  14 1210 8-  RPM^: 1200 Diesel Ratio : 25% Shrouding : 50%  42 0  4 BMEP (bar) I.^>  Figure 5.5 : Effect of Injection Pressure on Thermal Efficiency.  Gas injection pressure affects the NO R concentration in the exhaust gas (Fig. 5.2). At low load, higher gas injection pressure reduces the NO R . In all cases a substantial reduction in NO R as compared to the baseline is observed up to medium load although at high load the NO R concentration is slightly higher.  60  50 A  45  RPM^: 1200 Diesel Ratio : 25% Shrouding : 50%  -  4035  a  -  30 -  0x 25 - DIESEL •••• Z^BASELINE 20  Pgas : 70 bar 80 bar^\  -  15a  10 -  50 bar  5-  4  00 ^  5  BMEP ( bar ) =>  Figure 5.6 : Effect of Gas Injection Pressure on Oxides of Nitrogen.  The effect of gas injection pressure on unburned hydrocarbon (HC) is shown in Figure 5.7. HC emissions are strongly dependant on gas pressure and the best gas pressure depends on load. At high load the HC for any gas injection pressure exceeds the emission standard by at least 100%. However, the emission standard regulates only non-methane unburned hydrocarbons.  ^  61  2824 22 20 1818141 = 12-  a  RPM^: 1200 Diesel Ratio : 25% Shrouding ; 50%  Fa 108-  2^fs.  -  0^4  2^^ 0^  DIESEL BASELINE EPA S3P94  A^4^6  BMEP ( bar ) ,===>.  Figure 5.7 : Effect of Gas Injection Pressure on Unburned Hydrocarbon.  It can be seen from Fig. 5.7 and 5.5 that the concentration of the unburned hydrocarbon in the exhaust gas is related to the thermal efficiency. A a higher HC in the exhaust gas suggests a lower thermal efficiency, while a lower HC is associated with a higher thermal efficiency. The amount of the unburned fuel gas in the exhaust gas is a good indication of how the combustion progressed. Assuming all measured unburned hydrocarbon in the exhaust gas is methane (CH 4), gaseous fuel which survives the combustion process can be  62 estimated. Table 5.1 lists the unburned gas ratio' for three injection pressure and for high and low load cases. The high load value at each injection pressure corresponds to minimum HC emissions and highest thermal efficiency.  Pio , bar  50  60  70  BMEP, bar  0.4  2.8  0.5  2.6  0.5  4.0  Unburned Gas Ratio  0.544  0.019  0.447  0.023  0.374  0.031  Table 5.1 : Effect of Gas Injection Pressure on Unburned Gas Ratio.  As shown in Table 5.1, the gas injection pressure significantly affects the unburned gas ratio. The lowest unburned gas ratio is 2.4% which is experienced with 60 bar gas injection pressure at its optimum load 3 (BMEP = 2.6 bar). At low load operation, higher gas injection pressure greatly reduces the unburned gas ratio. At low loads 37-54% of the injected gaseous fuel survives combustion. This is associated with very late burning rather than misfiring, especially with higher gas pressure. Injection gas pressure appears to be one of the important factors controlling the combustion process. This suggests that the fluid mechanics of gas distribution in the combustion chamber are of great importance.  2  3  The unburned gas ratio is defined as the ratio of the unburned gas in the exhaust to the injected gaseous fuel. Optimum load is defined as the load that has maximum thermal efficiency.  63  5.4 Effect of Diesel Ratio.  31A 292725 23 21191715131197531 0^0:5^1:5^2^2:5^6^3:5^4^46^6  BMEP ( bar) ==>  Figure 5.8 : Effect of Diesel Ratio on Thermal Efficiency.  Tests were carried out with different diesel ratios to study the effect of pilot quantity on thermal efficiency. Fig. 5.8 shows the effect of diesel ratio on thermal efficiency. With 25% diesel ratio the thermal efficiency is about the same as with 20%. Lowering the diesel ratio to 15% reduced the thermal efficiency at low load operation by about 1.5%, but the engine operation is less smooth than with the 20% and 25% diesel ratio operation. However, at high load the 15% diesel ratio operation is associated with  64 thermal efficiency of about 3% higher compare with the other diesel ratios. This suggests that the gas-diesel engine needs a minimum quantity of pilot fuel to operate smoothly and efficiently.  BMEP ( bar) ^ Figure 5.9 : Effect of Diesel Ratio on Oxides of Nitrogen.  The specific NO in the gas-diesel engine emission is significantly affected by the diesel ratio (Fig. 5.9). Lower diesel ratio produces lower specific NO„ in most of load ranges except at very high load (BMEP > 3.7 bar). Fig. 5.9 suggests that increasing the  65 amount of pilot-diesel tends to increase the NO„ concentration in the exhaust gas. However, regardless of the diesel ratio, the NO„ in the gas-diesel engine is lower than that of the diesel engine.  24 A  Diesel Ratio ; 25%  22  Zo 18 16  15%  14 12 10  8  a 4  RPM^: 1200 Pgas^80 bar Shrouding 50% ,DIESEL  20% NE  EPA '93/'94  2 0  4 BMEP ( bar ) ^  Figure 5.10: Effect of Diesel Ratio on Unburned Hydrocarbon.  Fig. 5.10 shows the specific unburned hydrocarbon in the exhaust gas of the gasdiesel engine operated with 15, 20, and 25% diesel ratio. The minimum in HC emissions is associated with 20% diesel ratio. Operating the engine up to BMEP — 3.5 bar with 15% diesel and gas injection pressure of 60 bar gives large amounts of unburned hydrocarbon  66 compared with that of the 20 and 25%. This indicates particularly poor combustion with the 15% diesel ratio at low load.  5.5 Effect of Pilot-Diesel Cetane Number.  31 292725 2321 19 17 15 13 11 9 75-  36  4 BMEP ( bar) ,===>  Figure 5.11: Effect of Pilot-Diesel Cetane Number on Thermal Efficiency.  Tests were carried out using a higher cetane number diesel fuel to examine the effect of ignition delay of the pilot-diesel. Fig. 5.11 shows the effect of pilot-diesel cetane number on thermal efficiency. A significant improvement in the part-load thermal  67 efficiency results from using higher cetane number pilot-fuel (CN62) although at higher load the thermal efficiency is substantially lower. Decreasing the quantity of the higher cetane number pilot fuel by lowering the diesel ratio seems to overcome the drawback of using higher cetane number pilot fuel at higher load operation. As shown in Fig. 5.12, the NO in the exhaust gas is affected by changing the pilot-diesel cetane number. The use of higher cetane number pilot-diesel causes the load range to shift. The concentration of NO shifted in same pattern as the loads thus the NO. concentration at the same load is higher for a higher cetane number in most cases.  55 50 45 40  DIESEL BASELINE  : RPM Pgas : Diesel Ratio : Shrouding :  1200 60 bar 20% 50%  35 30 25 20  PILOT-DIESEL : DF2 CN62  15 10 5  EPA '93/ '94  0 ^^ 1 0 BMEP ( bar ) ^  Figure 5.12:  5  Effect of Pilot-Diesel Cetane Number on Oxides of Nitrogen.  68  Fig. 5.13 shows the effect of using two different pilot-fuel cetane numbers on the concentration of the unburned hydrocarbon in exhaust gas (HC). The use of higher-cetanenumber diesel-pilot significantly reduces the amount of HC, and provides better combustion. This is one of the reason for using a higher cetane number pilot-diesel (CN62).  20 PILOT-DIESEL : DF2 N\ilk  RPM^: 1200 Pgas^: 80 bar Diesel Ratio : 20% Shrouding : 50%  10 8 8 4 2  0  DIESEL BASELINE EPA 13/14  4  A  BMEP ( bar ) ^  Figure 5.13 : Effect of Pilot-Diesel Cetane Number on Unburned Hydrocarbons.  69  5.6 Summary. The effect of engine performance and emissions of the gas injection pressure, diesel ratio, and pilot-fuel cetane number in a gas-diesel engine have been examined. The results were compared against the diesel baseline. The gas-diesel has better thermal efficiency at full load while at low load the thermal efficiency is nearly as good as that of the conventional diesel. Relative to the 1994 EPA bus standard, emissions are a serious problem both with conventional diesel and gas-diesel pilot opeation. At full load NO„ is excessive, while at part load unburned HC (probably mostly unburned fuel) considerably exceeds the standards. High unburned HC associated with low thermal efficiency. Gaseous fuel injection pressure affected the load capability. Increasing gas pressures tends to extend load capability, improve thermal efficiency at full load, and reduce NO up to high load operation. However, higher gas pressure at low load operation produces relatively higher unburned HC. Results of the tests done with different diesel ratios show that relatively smooth operation at low load was achieved with high diesel ratio, although at high load relatively higher thermal efficiency was obtained with low diesel ratio. Tests carried out using a higher-cetane-number pilot-diesel demonstrated that by reducing the pilot-diesel ignition delay time, the thermal efficiency is better and the engine operation is smoother. At the same time the concentration of unburned HC is reduced. These results indicate problems with combustion which are further investigated in Chapter 6 with the aid of combustion rate analysis.  70  6. COMBUSTION RATE ANALYSIS  The analysis of combustion rate in internal combustion engines is a technique used to obtain the fuel mass-burned fraction or the combustion rate from a measured time history of the cylinder pressure. This chapter is divided into twelve sections. The first section provides a brief explanation of the analysis. The second section describes the combustion model used. The mixture composition of the unburned gases is discussed in the third section. Its thermodynamics properties are evaluated in section four. Section five explains how the table for burned-gases properties is formed. The method to account the effect of heat transfer to the cylinder wall is described in section six. The computation procedure is explained in the seventh section. Measurements of cylinder pressure and results of analysis are presented and discussed in the last three sections before the summary.  6.1. General. Engine pressure data is the principal data in the combustion rate analysis. It is the combustion part of the pressure history that the analysis needed. The combustion process is considered to start at the beginning of injection (BOI) near the end of compression stroke, although the compression auto-ignition occurs a few crank angle degrees after BOI.  71 This delay is termed ignition delay. A typical cylinder pressure distribution is shown in Figure 6.1. The fuel mass-burned fraction at a certain crank position is directly related to the pressure difference AP between the measured pressure P.. (with combustion) and the polytropic compression or expansion pressure P rnot obtained without combustion.  PeY  BOl  ^  TDC 522*  °CA  Figure 6.1: Typical Cylinder Pressure Data  In the thermodynamic analysis a two-zone model has been chosen to simplify the complex combustion process in the engine. Governing equations of this analyis are the mass conservation equation and energy conservation equation.  72  6.2. Formulation of the Combustion Model. Combustion in the gas-diesel engine is initiated by spontaneous ignition of a portion of the injected diesel fuel when the air temperature and pressure in the combustion chamber are above the ignition point of the diesel fuel. This happens after a delay period of a few crank angle degrees from the beginning of fuel injection and near the end of compression stroke; it could happen at several places in the cylinder. The result is that the local temperature is increased enough to autoignite and bum the gaseous fuel.  FUELS INJECTOR  Time-step C.A  Figure 6.2: Schematic of the Combustion Model  73 A two-zone combustion model has been adopted to simplify the combustion process to estimate the combustion process. Figure 6.2 shows a schematic of the thermodynamic system in the engine cylinder while combustion is in progress. Consider the presence of two zones in the control volume, i.e., those of the burned and the unburned gases. The following assumptions are applied : 1.  The pressure inside the cylinder is uniform.  2.  The cylinder constituents (both unburned and burned) behave as ideal gases.  3.  The thermodynamic state of each zone is considered to be homogeneous and uniform.  4.  The natural gas is considered to be methane (CH 4), and the diesel fuel can be represented by CH 2 . Vaporization of the injected diesel fuel is assumed to take place very quickly, with allowance made for heat of vaporization. All fuel is assumed to be present in the combustion chamber in vapour one degree after the beginning of injection.  5.  The combustion products are in thermodynamic equilibrium.  6.  The mass-burned fraction of diesel fuel and natural gas are the same at any instant during the combustion process. This is called the proportional burning assumption.  Temperatures of these two zones are Tb and T. respectively. Specific energy content in the burned zone is u b which is a function of both temperature Tb and pressure P since combustion products are partially dissociated. The unburned zone has a specific energy  74 content of u u which is a function of temperature. For polytropic compression of the unburned gases, their temperature T. will be a function of cylinder pressure P. At any crank angle position, the system specific volume v and its specific energy u can be evaluated as follows  V(0) v -^ m tot ,and^  u =  (6.1a)  U0 - W + Q., tot^  (6.2a)  mu.^mtot  where V is the cylinder volume, 0 is the crank angle, m,„, is the total mass of the cylinder contents, E t., is the total energy of the system which is a summation of is the internal energy of the cylinder contents at some reference point U. , the work done on the piston W , and the heat transfer to the wall Q., (a negative quantity). Define x as the fuel mass-burned fraction at any instant during the combustion process , v the specific volume, and u the specific energy of the system (and let subscripts u and b denote unburned and burned gas properties). The properties of the unburned and burned gas at a given time can be expressed by the following two governing equations : mass conservation,^v = x. vb(P,Tb) + (1-x)• energy conservation, ^u = x- u b(P,Tb)  vu(P)^  + ( 1-x)• u.(P)  ^  (6.3) (6.4)  Given P, v, and u , the system of the above two equations can be solved iteratively for Tb and x.  75  The calculation is done stepwise in a small time interval corresponding to a 1 degree crank angle. The volume of the cylinder at any crank angle position V(0) is determined from the geometry of the cylinder (Appendix H). The total mass of the cylinder contents m is the summation of the mass of incoming air trapped in the cylinder, the exhaust gas remaining from the previous cycle (residual gas), and the fuel masses. The procedure used to calculate the residual gas fraction, and the incoming air trapped in the cylinder is explained later. The specific volume of the system v is the mean specific volume in the time interval which is taken to be the specific volume at the end of the step. It is calculated by using the Eq. (6.1a). The energy of the system is the total energy up to the end of the chosen time interval which is calculated by using Eq. (6.4a), if the total internal energy of the system E tm is known. The latter is determined from the first law of thermodynamics which is applied to the system for a small time change At as follows  dE = 8Q  -  8W  where Q is the heat transfer to the system and W is the work done to the piston. Then, integrating the Eq. (6.5) from state 1 to state 2 for a given time step At which is corresponding to 1 degree crank angle, will gives, DES = .1Q2 - .11312^  (6.5a)  76 By defining the average pressure of the system for the given time step as  -  P1 + P2 2  the work done to the piston can be written as P1 +P +P2 . 1 W2 - ^2 (V2 Vi) -  (6.6)  The method of accounting for the effect of heat transfer is explained in Sect. 6.6. For the moment we note only, that if SQ = 0 then substituting Eq. (6.6) into Eq. (6.5a) gives the energy of the system at the end of the step as P1 +P E2 =^(^2 ) V2 - )  2  (6.5b)  Using this last relationship in Eq. (6.2a), we can compute the specific energy of the system. U =  E2^  m wr  (6.2b)  6.3. Mixture Composition.  The combustion process is preceded by compression of the cylinder charge. The composition of the cylinder charge depends on the type of gas exchange process in between two consecutive cycles.  ^ ^  77  EXHAUST VALVE OPEN ; WAIDC CLOSE ; dTPABDC  J FRESH AIR =-1) AIR FROM^ SLOWER^1110X  1,/"..*°.".--- r=:;• EXHAUST  INTAKE. PORTS (IE)  rosmoNs 73•ARDC IIDC  1  Figure 6.3 : Schematic of the Uniflow-scavenged Configuration.  The test engine is a two-cycle with a uniflow-scavenged configuration as illustrated schematically in Fig. 6.3. A root-type blower develops a slightly higher pressure in the air box surrounding the cylinder. Through inlet ports the fresh air enters the cylinder and displaces the previous-cycle cylinder contents. A part of the incoming fresh air mixes with the residual gas and is expelled with it. However, some of the burned gas remains in the cylinder after this process. To determine the cylinder charge composition, it is important to estimate the amount of this remaining gas or residual gas as well as the portion of the fresh air trapped in the cylinder.  78 6.3.1. Scavenged Air. To estimate the amount of air trapped in the cylinder at the end of the scavenging process the following definitions are used. ■ The delivered air mass, m ad , is the mass of air delivered to the engine per cycle as measured at the air-intake line. ■ The mass of air trapped, m ai,. , is the portion of the delivered air mass per cycle trapped in the cylinder when the piston is at the inlet port closure (IPC) position. ■ The residual mass, m res , is the mass of the combustion products remaining from the previous cycle. ■ The trapped mass, m fr , is the mass of the cylinder contents at IPC which is the summation of the mass of the air trapped and the residual mass. = matr + inns ■ The delivery ratio, A , is defined as the ratio of the delivered air mass per cycle to the trapped mass. A -  Mad  M a.  ■ The degree of purity of the charge, DP , is defined as the ratio of the mass of air in trapped cylinder charge to the mass of trapped cylinder charge. DP = tr^M res + M atr  79 It indicates the degree of mixing of the delivered air with burned gases in the scavenging process, and is shown later to be a function of delivery ratio. DP = f(A) ■ The residual fraction, f res , is defined as the mole ratio of the residual gas n r. to that of the trapped charge in the cylinder at IPC n iz. .  Tres  ^Nes tr  If it is assumed that the residual gas has the same molecular weight as that of the fresh air, we can write fres  co  res _  co res  Itt^+ co t,. tr^res^  or^fres = 1 - DP  There are two limiting ideal models in the scavenging process [31] [32], i.e., perfect displacement, and complete mixing. In perfect displacement, the incoming fresh gases totally displace the burned gases without any mixing. Complete mixing occurs if entering fresh mixture mixes instantaneously and uniformly with the cylinder contents. Following the method described by Heywood [31], imagine a boundary surface in between the fresh charge and the burned gases; then we can write DP = A^for A s 1^  (6.7a)  DP = 1^for A > 1^  (6.7b)  80 For the complete mixing, during a small time interval dt in the scavenging process the delivered air entering the cylinder is dm ad while the instantaneous air mass in cylinder is m at, and the instantaneous total mass in cylinder is m om . Assuming the incoming mass flow rate is equal to the exit mass flow rate while keeping in mind that it is a fully mixed process, we can write  dmaa.  °") = dm ad dmad ( m mtr  If mtr is constant during the process, then "t ar  )  nta^  = 1 - ( marr  m a.  m u.  or^- In(1-  ) tr^  mgr M  +C  Taking the intital condition, mad /ma. = m a,,/ma = 0 , C = 0. Then  ?fl  air  M a.  = 1 - e  and by using the definition of Degree of Purity, DP , and Delivery Ratio, A we get^DP = 1 - C A  ^  (6.8)  In the actual scavenging process, the scavenging configurations of the the engine affects the degree of purity. Typical data for large two-stroke diesels is shown in Fig. 6.4. It shows that the uniflow scavenging is the most effective scavenging configuration.  81 1.0  0.8  1 1 1 ,1i  111 11111111 1 1 ili^  6>42\, 2 4r •  ^  E. 0.6  ^/  \  \  '44  s to,•"'% t! 0.4  Unillow scavenging Loop scavenging 0.2  0  0.4  Cross scavenging  0.6^0.8  1.0^1.2  1.4^1.6  Delivery ratio A  Figure 6.4 : Scavenging Data Typical of Large Two-stroke Diesels [31].  In the determination of the amount of air trapped in the cylinder, a fit to the average value of the degree of purity as a function of delivery ratio for the uniflow scavenging area as shown in Fig. 6.4 is used. This third-degree polynomial curve fit is implemented in the calculation of the fuel-air equivalence ratio filed in Appendix I.  6.3.2. Residual Gas.  The composition of the residual gas is the same as that of the burned gas. However, in the determination of the residual gas composition, complete combustion (with the simplifying approximation that the products consist of CO 2 , H 2 O , 0 2 , and N2 ) is  82 assumed. If r is the diesel-to-gas mass ratio and 4) is the fuel-air equivalence ratio as calculated in Appendix 1, the complete combustion equation for the methane and diesel fuel (CH 2 ) is 2 + 3 16 1 6^2 14 r CH4 + r CH2 +^( 02 + 3.76 N2 ) 14 4) -  16 16 (1+- r) CO2 + (2+- r) H2O 14^14 3 16 (2+- — r)  (1-4))  3 16 02 + (2+- — r) 3.76 N2 2 14^•^2 14  (6.9)  To determine the amount of residual gas, the residual fraction ^as defined in subsection 6.3.1 is used.  63.3. The Unburned Gas Composition. The fuels are assumed to be injected within the first time step after the beginning of injection. Then the unburned gas in the beginning of the combustion process consist of the trapped air, residual gas, and the fuels. Taking 1 mole methane in a complete combustion as shown in Eq. (6.9), the number of moles of the i species Ft; of the unburned gases are as follows : nCH4  n CH2  =1  14 =—r 16  83  3 16 r 2+2 14^fres 1-, [ 1 + 1 f  02  -  3 16 r 2+2 14^ f ( 1^r_ 1 -f  nN2  co2  n H20  = (1 +  (1+  =  ) (336)  16^fres r)  14^1-4. 16^fres r) 14^1 —fres  and the sum of the moles of the unburned gases is 6  En i^n  + n CH 2 + n + n N + nCO2^ + n H0 2 c H4^2^  where n i is the number of moles of the unburned gases species. The relative population of the unburned gases species is determined using the following relationship.  YE  ni -  6  n E a=i  i  6.4. Thermodynamics Properties of the Unburned Gas. In the combustion process, the properties of the unburned gas are determined from knowledge of its composition as derived in Sect. 6.3. Its composition is constant in the  84 combustion process because of the assumption of proportional burning in the combustion model. The molecular weight M. and the gas constant R. of the unburned gas are calculated as follows  (6.10)  Mu  =^  My  (6.11)  where n, is the number of moles, M 1 is the molecular weight of the constituent gases, and R is the universal gas constant. The unburned gas temperature T. at the end of a step is calculated from its pressure increment (or decrement) in a polytropic compression (or expansion) process. Note that the injection of the fuels is assumed to occur in the first crank angle degree after BOI, so that the combustion starts one degree after BOI. The unburned gas temperature at BOI is computed from the total mass of cylinder charge using the ideal gas relation since its pressure and volume at BOI are known. The molar specific heat at constant volume C„ of the unburned gas is calculated from its molar specific heat at constant pressure C p„ which is a function of its temperature only [331. C. = Cpu 6  but, Cpu  _  En  E  [ CdT ) l E Mw  85 6  Then,  E^Cp0(T ) C.  i+ 1  (6.12)  R.  Mu  The isentropic exponent of the unburned gas is determined from its constant-volume specific heat value C,,„ and gas constant R„ . R  (6.13)  1 +C."  The specific volume of the unburned gas is calculated from the ideal gas equation,  Vu  R. T.  (6.14)  P  where P is the cylinder pressure.  The unburned gas enthalpy H. is the summation of its constituent enthalpies H 1 each of which is a function of temperature as a result of the ideal gas assumption.  Hu  E 6  ^[  f  /7f  1= 1  where Y i is the i th unburned gases species, h  EpoM err  ],  298.15  f  is its enthalpy of formation at standard  state condition (298.15K and 0.1 MPa), Cpo is the constant-pressure specific heat of the  86 unburned gas species. The heat of vaporization of diesel fuel is accounted for at this point. Using the definition of enthalpy,^u = h -Pv the internal energy of the unburned gas is  -  P v,,^  (6.15)  .  6.5. Thermodynamic Properties of the Burned Gas Mixture. The burned gas internal energy u b and specific volume v b are found from a properties table by specifying the pressure and temperature. This table of properties is generated using STANJAN [34] which is a chemical equilibrium solver using the JANAF thermodynamic property tables. Since the engine consumes two different fuels, i.e., diesel fuel and natural gas, we need two different sets of tables as data. To cover adequately all possible equivalence ratios of the engine working condition, a set of six tables were generated with equivalence ratio of 0.20, 0.40, 0.60, 0.80, 0.90, and 1.00. In generating the tables, it was assumed that the diesel fuel can be represented by CH 2 , and the natural gas by CH 4 . Each table lists the properties for ten different temperatures and ten different pressures as listed in Table 6.1. These two sets of tables provide the burned-gas property data.  87 Temperature  Pressure  K  atm.  1500  1  1700  2  1900  3  2100  5  2300  10  2500  20  2700  40  2900  60  3100  80  3300  100  Table 6.1 : List of Temperatures and Pressures used in Burned Gas Properties Table  As stated in Sect. 6.2, the combustion of diesel fuel and natural gas are assumed to be proportional so that at any time the mass-burned fraction of two fuels are the same. If r is the mass ratio of diesel to natural gas, then by replacing ( CH 2 + 16/14 r CH 2 ) with CH. as described in Appendix I, the burned gas properties can be determined from a table for a given n ( 2 < n < 4 ) at the given fuel-air equivalence ratio. The procedure to form this table form the two sets of tables ( CH 4 and CH 2 ) is as follows. 1.^For each fuel, interpolate the tables to obtain a table for the given fuel-air equivalence ratio.  88 2.^Interpolate the above two tables to form a table for a given n (carbon to hydrogen atom ratio). Then, the set-up table serves specifically a particular engine operating condition, i.e. at certain diesel to gas mass ratio at a given fuel-air equivalence ratio.  6.6. Heat Transfer to the Cylinder Wall.  The amount of heat transfer to the cylinder wall directly affects the calculation of the system total energy as can be seen in Eq. (6.2b) and subsequently the calculation of the mass-burned fraction. Methods of direct estimation of the instantaneous spatially averaged heat flux to the cylinder wall have uncertainties as shown in Sect. 2.5.  X  X  0  8 min  ^  0 max  e Figure 6.5 : Typical Mass-burned Fraction Results.  89  To account for the effects of heat transfer, an indirect method has been used. Exhaust composition measuments provide the concentration of total unburned hydrocarbons. If it is assumed that all of these hydrocarbons represent unburned (methane) fuel, then the maximum mass-burned fraction x. can be determined. This will substantially differ from the value obtained in the adiabatic calculation because the effect of heat transfer is substantial. Fig. 6.5 illustrates the difference between the calculated mass-burned fraction and the estimated actual values. The broken line in Fig. 6.5 (for the adiabatic calculation) indicates negative values of x before the combustion pressure rise begins; this is a direct result of heat transfer and is corrected for by the following procedure. Let X, be the adiabatic calculation value and x be the estimated true mass-burned fraction. The approximate correction procedure is as follows. (i)  For^eboi^< em i o  x  =  e  ^)(-x) °min °kg  (6.16a)  -  (ii)  For Oimn < 0 < 0. x = xa  +^+ (  e-  mb,  ) (x - xl  + xmin)  0 - Omin^(6.16b)  (iii)^For^0^0,nu (6.16c)  90 This procedure for estimating the distribution of mass-burned fraction is quite approximate. The chief reason for this is that the cylinder heat transfer rate is predictable only with a wide range of uncertainty. However it should be borne in mind that the important features of the calculation are : (i)  The delay period before there is appreciable burning.  (ii)  The characteristic burning duration.  (iii) The apparent presence (or absence) of distinctive burning periods for the pilot-diesel fuel and natural gas. The curves in Fig. 6.5 do not indicate such a distinction.  6.7. Calculation Procedure.  In calculating the combustion rate, determination of the initial conditions is important. The mass of air trapped in the cylinder has to be calculated first. This parameter affects directly the fuel-air equivalence ratio whose calculation procedure is filed in Appendix I. The combustion rate calculation starts when the fuels are injected into the combustion chamber. The pressure data is used to determined the work done to the piston and subsequently the specific energy of the system is determined as explained in Sect. 6.2. A table of properties for the burned gases is formed to use in iteratively solving for the mass-burned fraction. The combustion rate calculation procedure is as follows: 1.^Determine the trapped air mass, m atr , and the fuel-air equivalence ratio, (1).  91 2.  Determine the gas constant of the unburned gas,^(Eq. 6.11).  3.  Set the burned-gas table for the given fuel-air equivalence ratio and hydrogen to carbon ratio.  4.  Obtain the total mass in the combustion chamber from the trapped mass, rn tr , and the fuel masses.  5.  Set the initial condition of the combustion process, i.e., the unburned gases temperature and internal energy (T„ and u u ). The unburned gas temperature is determined using the total mass, pressure, and the volume information.  6.  Calculate the isentropic compression (or expansion) coefficient of the unburned gases, yu , and the unburned gas temperature at the end of the calculation step. As mentioned in Sect. 6.2, the calculation step is taken to be 1 crank angle degree.  7.  Evaluate the unburned gas properties at the end of the time step. Specific volume, vu , is calculated from the ideal gas relationship. Its internal energy, u u , is determined by knowing its temperature (Eq. 6.15).  8.  Compute the work done during the time step using the cylinder pressure information (Eq. 6.6).  9.  Evaluate the specific energy of the system at the end of the step u m from Eq. 6.2b, and its specific volume v m from Eq. (6.1a).  10.  Given P, T, , vu , um , and vm , iteratively solved the two governing equations for burned gases temperature, Tb , and fuel mass-burned fraction, x, by looking at the burned table. The iteration procedure is documented in Appendix J.  11.  Prepare for next calculating step by resetting the initial condition.  92 12.^Repeat step 6 to 11 for the next calculating step.  To implement the above computation procedure, an existing mass-burned fraction computer program for a four-stroke spark ignition engine was modified. The listing of the is documented in Appendix K together with the burned-gas tables. Appendix L contains the record of a verification of the computation procedure for the constant volume case.  6.8. Pressure Measurements. The engine cylinder-pressure data were acquired using a PCB 112A05 air-cooled piezoelectric pressure transducer. As described in Sec. 3.4.6 and shown in Fig. 3.7, it was mounted through an adaptor sleeve which has a single passage. This mounting technique avoided the problem of thermal shock encountered when the transducer was flush mounted and thus directly exposed to the combustion gas. As explained in Sec. 3.5, the cylinderpressure transducer analog signals were digitized and recorded. To interpret the signals, they must be referenced to a known pressure at some point in the engine cycle. The air-box intake port pressure was chosen to be the reference pressure at BDC. Fig. 6.6 shows a typical pressure crank-angle diagram at low and high load. Each curve represents the ensemble-averaged pressure for 100 consecutive cycles at optimum B01. Note that the peak pressure magnitude is higher for the higher load than for the lower one. The location (in terms of degree CA) seems to be moving towards the TDC as the load increases; this suggests a late burning at low-load operation. Operating the engine at  93 higher load resulted in higher cylinder pressure since more fuel was burned in the combustion process.  1i0 160 1'0 180 i90 260 210 220 260 240  e (*ABDO ) ==> Figure 6.6 : Typical Pressure Crank-Angle Diagrams at different Loads.  The cylinder-pressure data can also be plotted against the cylinder volume using logarithmic scales to show the polytropic of the compression and expansion strokes explicitly. Fig. 6.7 shows a typical log PV diagram for one cycle each of high and low load. As can be seen from Fig. 6.7, the polytropic exponent n comp of the compression stroke is similar for both high and low loads. The value of n c,„„ p deduced from the compression shown in Fig. 6.7 is 1.23 (at both loads). For high load expansion stroke (after combustion), the value of the polytropic exponent nem, is 1.29. At low load it was not  94 determined because of the late burning situation.  4i^AS^4A^4.4^•^as  Ln (Vcyl)  ^-7!^44  1=:.  (a)  A  V  a  5  4A^4i^4.4^4.4^4^-Ti  Ln (Vcyl)  ^  42  ^  4A  4=>  (b) Figure 6.7 : Typical Log P - Log V Diagram.  95  6.9 Indicated Work.  Work transfer from the gas to the piston, which is calculated from the pressure data, is defined as indicated work Wind. It is equal to the area enclosed in the linear P-V diagram which is obtained by integrating around the curve. Wind =  c  P dV^  (6.17)  Brake work Wb is defined as the measured work per cycle which is calculated from the brake mean effective pressure BMEP. Wb = 100 BMEP Vd  ^  (6.18)  where brake work is in kJ, BMEP in bar and displaced volume V d in m 3 . Comparing the indicated work with the brake work is one way to check the validity of the pressure data acquired. Since the indicated work has to overcome the friction in engine parts and to drive engine accessories, the indicated work is greater than the brake work. Fig. 6.8 shows the comparison of the indicated work to the brake work of the test engine operated with 60 bar gas injection pressure and 20% diesel ratio at engine speed 1200 rpm. As seen in Fig. 6.8, the slope of the line fitted to the indicated work is very nearly parallel to that of the brake work which is exactly a straight line. This demonstrates that the friction loss is essentially independent of load.  96  0.6 0.5 OA 0.3 0.2 0.1  RPM^: 1200 P9as^: 60 bar Dose' Ratio : 20% Shrouding : 50%  0 2^3  ^  4  BMEP ( bar ) c==>  Figure 6.8 : Comparison of Indicated Work to Brake Work.  6.10 Cyclic Variation.  Variations of the pressure data, and indicated work, from cycle to cycle are termed cyclic variations. Assessing cyclic variations is one way to evaluate the quality of the combustion process. Fig. 6.9 shows the standard deviation and the relative standard deviation of the indicated work at different loads. This information reflects the cycle-by-cycle combustion variation which is important to note in interpreting the mass-burned fractions resulting from the corresponding pressure data.  97 0.8 0.5 0.4 0.3 02 0.1 0 1  ^ ^ ^  2  3  3.8  BMEP ( bar) =>  Figure 6.9 : Standard Deviation and Relative S.D. of the Brake Work.  The cyclic variation of the high load case (BMEP = 3.8 bar) of the gas-diesel engine operating with 60 bar gas injection pressure and 25% diesel ratio is shown in Fig. 6.10 which displays the pressure-volume data for ten successive cycles. At this load, the relative standard deviation of indicated work for 100 cycles (as shown in Fig. 6.9) is 5.6%. At low load (BMEP = 1 bar), the relative standard deviation is 36.7% which indicates poor combustion.  98  fa  •  fi‘  a  pa,  •  4  41  ^  41  ^  41  ^ ^ 4  4S  ^  as  ^  41  Ln ( Ay! ) r==',>  Figure 6.10 : Superposition of 10 Successive Cycles.  6.11 Mass-burned Fractions.  We can defined the ignition time delay period as the crank angle difference between the first appearance of burning and the actual BOI which is about 3 ° greater than BOI. Fig.6.11 shows an example of measured cylinder pressure, estimated cylinder pressure, and normalized mass-burned fraction for the operating condition at 1200 RPM, BMEP = 3.8 bar, 20% diesel(CN62) ratio, 60 bar gas injection pressure, 10° poppet angle,  50% shrouding. Here the electronic BOI is 32°BTDC, actual BOI is approximately  99 29°BTDC, and the apparent ignition delay is about 31 CA degree (4.3 ms). The unburned gas temperatures during ignition delay period is estimated to be less than 1000 K. This ignition delay time is approximately an order of magnitude less than those of methane alone presented by Fraser, Siebers and Edwards [36].  TDC Unburned-gm^; Temp. (K/1000)  1.2 1.0  -^ -^ -  0.8  MEP:4 MR 1 OM  Mass - bum*  OS  -1--FesitimParm-)  OA 0.2 0 140  100^lap^200  220  delay-- _combustion period ---  Crank Angle ('ABDC ) r==> Figure 6.11 : Cylinder Pressure, Temperature and Mass-burned Fraction Distribution.  Fig. 6.12 shows mass-burned fraction curves at different loads while the engine worked with the same injection pressure and diesel ratio at best BOI. We see from Fig. 6.12 that the ignition delay period 30-35° and is not strongly dependent on load (except for  100 the 1 bar case). The combustion period' is roughly twice as long for the 2 bar case as for the 3.8 bar case. The mass-burned fraction curves do not show any sign of two distinct stages of burning (for the pilot liquid diesel and the natural gas).  1.0 A  0.8 0.8  LL.  0.4  E -9  2  RPAI^- 1200  0.2  • 00 ber Mt Rob - 20% Shicidna^50%  0 140 180 180 200 220 240 280 280 300 Crank Angle ( °ABDC ) ===>  Figure 6.12 : Combustion Pattern at different Loads.  Fig. 6.13 shows the normalized mass-burned fraction curve for 3 different gas injection pressures with the engine operated at about 4 bar. For the highest pressure the ignition (i.e., pressure rise) delay period is about 4° less than with 50 bar and the  1  The combustion period is defined as the crank angle difference between the x = 0.05 and the x = 0.95 points.  101 combustion period not much affected by gas pressure. The shape of the burning curved is much the same in all 3 cases; no distinct burning period for the pilot fuel can be observed.  1.1 1 0.9 0.8 0.7 LL  f  0.60.5 -0.4-  0.3 2 0.20.10140  Crank Angle ( °ABDC ) ^ Fig 6.13 : Combustion Pattern for different Gas Injection Pressures.  6.12 Summary.  The combustion rate analysis utilized to estimate the mass-burned fraction from the cylinder pressure data has been elaborated. The analysis used a two-zone combustion model which solved the mass and the energy conservation equations. Estimation of the amounts of fresh air trapped in the cylinder, and the remaining  102 combustion products from the previous cycle in the scavenging process followed the procedure recommended by Heywood [31] has been employed. The burned gas properties table used in the computation is generated using STANJAN [34]. An indirect method to account the effect of the heat transfer by correcting and normalized the mass-burned fraction curve resulted from an adiabatic calculation has been adopted. The unburned hydrocarbons in the exhaust gas is considered in normalizing the curve. The cylinder pressure data have been measured and recorded for selected engine operating conditions. The indicated work calculated from the pressure data has been compared with the brake work to check the validity of the pressure data acquired. A high cyclic variation in the cylinder pressure data indicates poor combustion at low load as does the high concentration of unburned hydrocarbons in the exhaust gas. The calculated unburned gas temperatures at TDC ranged from 900 K at low load to 1000K at high load; these are much lower than reported autoignition values for methane (approaches 1300 K). The calculated mass-burned fractions suggest that the amount of heat transfer is large. Combustion rate analysis shows that the burning rate increases as the load increases. Longer ignition delay time and longer burning time are associated with low load operation. The burning rate seems to be increased by increasing the gas injection pressure.  103  7. CONCLUSIONS AND RECOMMENDATIONS.  7.1 Conclusions. Performance, emissions and cylinder pressure data of a diesel-pilot gas injection engine have been investigated. A combustion rate analysis was employed with the pressure data to study the burning rate pattern. The following may be concluded :  1.  With high pressure injection of natural gas and about 20% diesel-pilot energy ratio the thermal efficiency exceeds that of the conventional diesel engine efficiency at high load. It is less than that of the conventional diesel engine efficiency at low load in the present configuration.  2.  With the same pilot-diesel ratio, lower gas injection pressure provides better thermal efficiency at low load than that higher injection pressure. The opposite is true at high load.  3.^Low-load thermal efficiency of the gas-diesel engine depends strongly on the pilot diesel energy ratio; this indicates that the pilot diesel ratio plays an important role in helping the natural gas to ignite.  104 4.  Using a higher cetane number pilot-diesel decreases the ignition delay time and improves thermal efficiency at low load.  5.  Lower pilot-diesel ratio produces less oxides of nitrogen (NO) emissions which shows the use of natural gas lowered the NO concentration in the exhaust gas.  6.  Although methane concentration in the exhaust gas was not measured directly, high unburned hydrocarbons emissions suggest that a considerable amount of injected gaseous fuel survived the combustion at low load.  7.  High cyclic variation in the cylinder pressure data is one evidence of poor combustion at low load. Another is the high rate of unburned hydrocarbons emission.  8.^The calculated mass-burned fractions indicated a large amount of heat transfer in this low-compression-ratio low-speed test engine. They also serve to indicate, especially at low load, long pressure rise (ignition) delay times and long burning times. No distinct burning curve for the pilot-diesel was observed.  105 9.  Combustion rate analysis shows that the burning rate depends on load; the higher load the higher burning rate than that of the lower load.  10.  Calculated temperatures of the unburned gas at TDC ranged from 900 K at low load to 1000 K at high load. These low temperature (associated with low compression ratio and heat transfer) are probably the main reason for the low quality of combustion, particularly at high load.  7.2 Recommendations.  1.  Continue experimental observation using a better unburned hydrocarbons analyzer which is capable of measuring the methane concentration in the exhaust gas. Determination of the unburned methane helps to provide a better understanding of the combustion process.  2.  Consider the possibility to supply the engine with a higher air temperature to study the effect of the unburned gas temperature on the ignition delay time and the combustion quality.  3.^If analytical work on the combustion rate analysis of the cylinder pressure data is to continue, an engine heat transfer correlation model should be  106 considered in the computation. This is to give a more accurate estimation of the mass-burned fraction.  107  8 REFERENCES  1^Boyer, R.L. and Crooks, W.R., "The Modem Gas Engines," ASME Paper No51-0GP-4, 1951. 2^Karim, G.A., "The Dual Fuel Engine of the Compression Ignition Type - Prospects, Problems and Solutions - A Review," SAE Paper No.831073, SAE Trans. vol.92 pp 569-577, 1983. 3^Song, S. and Hill, P.G., "Dual Fueling of a Pre-Chamber Diesel Engine, with Natural Gas," J.Eng for Turbines and Power, Trans. ASME, Vol.107 pp 914-921, October 1985. 4^Ding, X. and Hill, P.G., "Emissions and Fuel Economy of a Pre-Chamber Diesel Engine with Natural Gas Dual Fueling," SAE Paper 860069, SAE Trans Vol. 95 pp.612-625, 1986. 5^Gettle, L.E., Perry, G.C., Boisvert, J. and O'Sullivan., P.J., "Dual Fuel Engine Control Systems for Transportation Applications," J.of Eng. for Gas Turbines and Power, Vol 109, Oct. 1987, pp. 435-438. 6^Gettle, L.E., Perry, G.C., Boisvert, J. and O'Sullivan., P.J., "Microprocessor Dual Fuel Diesel Engine Control System, SAE Paper No.861577, Oct. 1986. 7^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", SAE Technical Paper 891652, Aug. 1989. 8^Miyake, M., Biwa, T., Endoh, Y., Shimotsu, M., Murakami and S., Komoda, T., "The Development of High Output, Highly Efficient Gas Burning Diesel Engines," CIMAC Paper D112, Conference Proceeding, Paris-France, Jun. 1983.  108 9^Einang, P.M., Koren, S., Kvamsdal, R., Hansen, T. and ^Sarsten, A., "High-Pressure, Digitally Controlled Injection of Gaseous Fuel in a Diesel Engine, with Special Reference to Boil-Off from LNG Tankers," Proceeding CIMAC Conference, Paris-France, Jun. 1983. 10^Wakenell, J.F., O'Neal, G.B. and Baker, Q.A., "High Pressure Late Cycle Direct Injection of Natural Gas in a Rail Medium Speed Diesel Engine," SAE Technical Paper 872041, Nov. 1987. 11^Mc.Cuiston, F.D.,Jr., Lavoie, G.A. and Kaufmann, "Validation of a Turbulent Flame Propagation Model for a Spark Ignition Engine," SAE Trans Vol 86, pp.200-221, 1977. 12^Marvin, C.F.,Jr.,"Combustion Time in the Engine Cylinder and its Effects on Engine Performance," NACA Tech Report 276, 1927. 13^Rassweiler, G.M. and Withrow,L., "Motion Pictures of Engine Flames Correlated with Pressure Card," SAE Journal (Trans.), Vol 42 pp 185-204, May 1938. 14^Shayler, P.J., Wiseman, M.W. and Ma, T., "Improving the Determination of Mass Fraction Burnt," SAE Paper 900351, 1990. 15^Amann, C.A., "Cylinder-Pressure Measurement and Its Use in Engine Research", SAE Paper 852067, 1985. 16^Krieger, R.B. and Borman, G.L., "The Computation of Apparent Heat Release for Internal Combustion Engines," ASME Paper 66-WAIDGP-4, 1966. 17^Borman, G. and Nishiwaki, K., "Internal Combustion Engine Heat Transfer," Prog Energy Combust Sci, Vol.13 pp.1_46, 1987. 18^Annand, W.J.D., "Heat Transfer in the Cylinders of Reciprocating Internal Combustion Engines," Proc Instn Mech Engrs, Vol.177 No.36, 1963.  109 19^Woschni, G., "A Universally Applicable Equation for the Instantaneous Heat Transfer Coefficient in the Internal Combustion Engine," SAE Paper 670931, 1967. 20 " 1990 SAE Handbook 1990 Volume 3," Society of Automotive Engineers, Inc., 1990. 21^Hames, R.J., Straub, R.D. and Amann, R.W., " DDEC - Detroit Diesel Electronic Control, " SAE Paper No. 850542, 1985. 22^Hill, P.G., Pierik, R.J. and Hodgins, K.B., " Intensifier-Injector Technology," US Patent No. 5 067 467, 1991. 23^Ding, X., " Operation Manual for Exhaust Analysis System," Report No. AFL-85-04 , Mech Eng Dept - The University of British Columbia, Dec.1985. 24^Rohling, N.R., " Operation and Performance of Emissions Console," Unpublished Report, Mech Eng Dept - The University of British Columbia, Aug.1990. 25^Randolph, A.L., "Cylinder-Pressure-Transducer Mounting Techniques to Maximize Data Accuracy," SAE Paper 900171, 1990. 26^Lancaster, D.R., Krieger, R.B. and Lienesch, J.H., "Measurement and Analysis of Engine Pressure Data," SAE Paper 750026, SAE Trans., Vol. 84, pp 155-172, 1975. 27^Kach, R.A. and Adamczky, A.A., "Effects of Thermal Loading on Pressure Measurement in a Combustion Bomb," Rev. Sci. Instrum. 56, American Institute of Physics, 1985. 28^Yuen, D., Hodgins, K.B., " Data Aquisition System for Alternate Fuels Engine Testing," Unpublished Report, Mech Eng Dept - The University of British Columbia, May 1991.  110 29^Ouellette, P., "High Pressure Injection of Natural Gas for Diesel Engine Fueling," M A.Sc Thesis, Mech Eng Dept, The University of British Columbia, 1992. 30  ^,"Installation and Operation Instructions - Meriam LFE Laminar Flow Element," Meriam Instrument, 1981.  31^Heywood, J.B., "Internal Combustion Engine Fundamental," McGraw-Hill Inc., New York, 1988. 32^Benson, R.S. and Whitehouse, N.D., "Internal Combustion Engines," Pergamon Press, Oxford, 1979. 33^Van Wylen, G.J. and Sonntag, R.E., "Fundamental of Classical Thermodynamics, Third Edition," John Wiley & Sons, New York, 1987. 34^Reynold, Wm.C., "Chemical Equilibrium Solver Ver.3.60 - An Application Software ," Mech Eng Dept, Stanford University, 1987. 35  ^, "Single Cylinder Two-cycle Diesel Engine Power Plant - Model 171, Description and Operating Manual," Diesel Engine Div - General Motor, 1939.  36^Fraser, R.A., Siebers, D.L. and Edwards, C.F., "Autoignition of Methane and Natural Gas in a Simulated Diesel Environment," SAE Paper 910227, 1991.  111  9. APPENDICES  9.1 Appendix A  DIESEL ENGINE EMISSION STANDARDS'  Nox  HC2  CO  PM  1990  6.0  1.3  15.5  0.60  1991  5.0  1.3  15.5  0.25  1993  5.0  1.3  15.5  0.10  1994  5.0  1.3  15.5  0.053  1998  4.0  1.3  15.5  0.05  Model Year  Table A.1 : Urban Bus Heavy-duty Engine Emission Standards  g/ measured during EPA heavy duty engine test 2  Non Methane Unburned Hydrocarbon Gas.  3  Proposed level. However, may be relaxed to 0.007 g/ if technology is not available to meet the proposed level.  112  Model Year  NOx  HC2  CO  PM  1990  6.0  1.3  15.5  0.60  1991  5.0  1.3  15.5  0.25  1994  5.0  1.3  15.5  0.10  1998  5.0  1.3  15.5  0.10  Table A.2 : Heavy-duty Truck Engine Emission Standards  113  9.2 Appendix B CALIBRATION CURVES  1. Engine Speed.  ENGINE SPEED CALIBRATION CURVE (MAGNETIC TYPE SENSOR)  1^ ci w w  1400  1200  i 1 2  ^1000  11  NO  800  400  400^800^800^1000^1200  ^  1400  imi44 ENGINE SPEED, RPM (AS OF DIGITAL HAND TACHOMETER)  Figure B.1 : Speed Calibration Curve  114  2. Load Sensor.  TORQUE SENSOR CALIBRATION (STRAIN-GAGE LOAD CELL) E 140 i ur 120  D 0 CC MO 0 to  w 80  CC  D  Q < w 2  I  eo 40 20 0  0^20^40 60^80 100 120 140 + IDADIPE^• UNLIICON  100 APPLIED TORQUE, N.m Figure B.2 : Torque Calibration Curve  115 3. Diesel Fuel Mass-flow.  DIESEL FLOW CALIBRATION AVL FUEL BALANCE  6  AVL output cm Ilbratton, 4.994 C ko/nr)/Volt 5  L m^4 ■ o Y  LL  3  2^  2  I  0  0  ^  2  ^  4  APPLIED FLOW, ko/hr + Calibration Data ---- Error-free Reading  Figure B3 : Diesel Mass-flow Calibration Curve  ▪^  116  4. Cylinder Pressure Sensor.  CALIBRATION OF PRESSURE TRANSDUCER PC8 112A sin 10118, Sens 1 17 mVipst  2 100 2.000 1.900 1.600  Scale : 200 psi/Volt  1.700 1.600 1 500 ▪  1.400  ui^1.300 0^1.200 m 13c^1.100 w° Ic a. 1.000  D  o  o W ix  • •  LI  0.900 ^0.600  0.700 0.600 0.500 0.400 0.300 ^ 200 0.100 0.000  I^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1^1  0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 1.600 1.6 00 2.000 (Thousands) APPLIED PRESSURE, PSI ^  MEASURED PRESSURE - IDEAL READING  Figure B.4 : Cylinder Pressure Sensor Calibration Curve  117 5. Intake Air Flow.  Figure B.5 : Laminar Flow Element Calibration Curve [30].  118  9.3 Appendix C  CNG'  DF22  CN623  Lower Heating Value, kJ/kg :  49,051  45,220  45,220  Density, kg/m'  0.7114  860  836  Cetane Numbers  n/a  ---45  62.2  Table C.1 : Properties of Test Fuels  Natural Gas (BC Gas). Averaged data in the period of April 1989 up to December 1991. Diesel Fuel No.2 (Chevron). High Cetane Number Diesel Fuel. At STP (25°C,101.3kPa). The cetane number of a diesel fuel is defined (SAE J313 JUN89 [20]) as the percentage by volume of normal cetane in a blend with heptamethylnonane required to match the ignition quality of the test fuel.  119 9.4 Appendix D  Constituent  Percent by Volume  Methane^,CH4  95.5  Ethane^,C2H6  3.0  Propane^,C3H 8  0.5  Butanes plus'  0.2  Nitrogen  0.6  Carbon Dioxide,CO 2  0.2  Total^100.0  Table D.1 : Composition of Natural Gast  'This includes butane (C 4H io) and all heavier hydrocarbons. 2Nominal  gas analysis for 1989, BC Gas.  120  9.5 Appendix E  TO EMISSION CONSOLE  05' TUBE STAINLESS STEEL  S VAG ELOK FITTINGS STAINLESS STEEL  0.5' SUIAG.-NPT SS UNION 3/4-1/2' REDUCER  " (BRASS) -  2-3/4' REDUCER (GAL V.IR ON) ^ 2' TEE JOINT (GAL V. IRON)  <=I TO EXHAUST STACK  ^ FROM ^ ENGINE  Figure E.1 : Exhaust Sampling Probe in the Engine Exhaust Pipe  ^^ ^  121 9.6 Appendix F.  DIESEL SUPPLY UNE  ACCESSORY SHAFT DRIVEN ACTUATOR  DDEC SOLENOID fri ir *1^ I• te  • I^  I•  illirf ^  SPOOL VALVE  • An r ,Ordrir Sir^4 7,,  UNIT INJECTOR  4.02\1.... ^. rezo i 21/ A ■ '  ^Vi metering valve  INTENSIFIER  1  ‘^1  &\ I1^ ..—^  CHECK VALVE  RELIEF VALVE PUSH ROD BYPASS  Ili  CHECK ::^\ % ■ s,.1k RETURN^CHECK VALVE^ : -4 VALVE ^SPRING ,, \ Ns" P \ s.,4k^OPPET ^NOZZLE NIXING^\N * RESERVOIR  Pol;Pot^  CNG Storage (20-200 bar)  U.S. Patent • M067A67 Hoverter 1991  Figure F.1 : Schematic of the Two-fuel Unit Injector'.  Implementation of the "Intensifier-Injector Technology" [12]. Courtesy of K.B. Hodgins.  122 9.7 Appendix G POWER CORRECTION FACTOR CALCULATION METHOD  (in accordance with SAE J1349)  1.  The correction is made against the standard inlet air conditions, i.e., Inlet Air Pressure (Absolute)^: 100 kPa Inlet Air Temperature^: 25°C (298K) Dry Inlet Air Pressure (Absolute) ^: 99 kPa  2.  The correction factor fc,,,.,. applied to the observed brake power depends on the atmospheric factor t and engine factor f„, which is calculated using the empirical relationship :^fc. = ( fa )  fm  The atmospheric factor is calculated based on dry inlet air pressure  Bd o  and  inlet air temperature t,  t = [ 99 / Bdo ] [(t+273) / 298 ] °.7 The engine factor depends on the fuel flow F(g/s), the engine displacement D(dm3 ), the engine speed N(RPM), and the pressure ratio r of inlet manifold pressure Po to inlet air pressure B o . It has the following value : to = (0.036 q/r) - 1.14^if 40 < (q/r) < 65 fm = 0.3^if (q/r) < 40 fn, = 1.2^if (q/r) > 65  where q = (60,000 F) / (D N) since the engine is a two-stroke cycle engine. 3.^The correction factor used in the calculation is within the range of 0.90 to 1.10.  123 9.8 Appendix H  DETERMINATION OF THE CYLINDER VOLUME  The volume of the cylinder is a function of the crank angle. The reference position is taken when the piston at its farthest postion from the cylinder head, i.e. when piston at its bottom dead centre or BDC. The nearest the piston could approach the cylinder head is the top dead centre or TDC. The distance in between BDC and TDC is the stroke of the engine, s.  Figure A6.1 : Piston Top Surface Profile.  The top of the piston has a shallow bowl shape as shown schematically in Figure a6.1. Its volume was measured by pouring oil into the bowl (60 cm3 ). Adding the volume above the piston (17 cm 3 ) to the piston-bowl volume gives us the clearance volume of the cylinder V, (77 cm 3 ). Subsequently the compression ratio rc is found to be 16.02 which is determined as the ratio of the maximum cylinder volume ( displacement volume plus  124 clearance volume) to the minimum cylinder volume (clearance volume).  s/2 cos  Figure A6.2 : Geometry of the cylinder.  The geometry of the cylinder, piston, and connecting rod is shown in Fig.A6.2 where B(0.10795 m) is cylinder bore, s(0.127m) is stroke, 1(0.254m) is connecting rod length, r is crank radius ( r = s/2 ), e is crank angle. The clearance height c is defined 2as the height in the cylinder for the clearance volume, while p is the distance between  125 t2he piston pin axis to the piston top surface. The cylinder volume at any crank position V(0) is V(8) =  4  B2 z  ^  (H.1)  where z is the distance between the piston top and the cylinder head. Referring to Fig.A6.1, z is determined in the following way. s z = c + —s + (p+0 + — cose -^t2 - sine? ®+ p 2 2 2  or^z = c +  2  +1+  2  cose - l2 -(.sine?  ]  ^(H.2)  Since the geometrical properties of the engine is known, the cylinder volume at any crank position V(0) is determined from Eq. (H.1) and Eq. (H.2). The accuracy of the total cylinder volume calculation is limited by the accuracy of both the clearance volume measurement and the determination of crank angle. Determination of crank angle depends on the accuracy of determining the reference point that is the BDC. It is assigned by using a dial indicator to locate the lowest point of piston position. In addition, to minimized the effects of changes in mechanical clearances determine the BDC, procedure as suggested by Lancaster et al [26] was followed.  126 9.9 Appendix I  DETERMINATION OF FUEL-AIR EQUIVALENCE RATIO  1. Definitions  ■^The diesel to gas ratio r is defined as the mass ratio of the diesel fuel mom, to the natural gas nigas injected into the combustion chamber at the beginning of injection (BOI). r =  dist  ■^The combined-fuel hydrogen to carbon atomic ratio, n , is defined as the ratio of the total hydrogen to the total carbon atoms of the two fuels in a proportional burning assumption. We can replace^( CH 4 + 16/14 r CH 2 )^by^CH„ 16 r(2) 4+ — H ^14^ 28 + 16 r n =^-^C^1 + 16 ^7 + 8 r —T  where  (L1)  14  Using the above definitions, the complete combustion equation for the two fuels can be written as, CH.  1+^ 11 4 I 1.1^n  +^  16..2  + 3.76 N2 )^CO,  n^It pr^4^4 ) 02 + 1+  •^AA2....  • I,^1  2^  41^2  127 From which the stoichiometric fuel-air ratio is, (12 + n)  FA sa ( 1 +  )  4  (L2)  (32 + 3.7648)  2. Calculation procedure  Given the experimental data as follows, Pam b Tamb Pabox  9  mad 9 Inds' mg.  where Pamb is the ambient pressure, T aro is the ambient temperature,  Paw,  is the air-box  temperature, and assuming the scavenged-blower efficiency is li mo , the air gas constant is R.  and its specific heat ratio is y , and the residual gas temperature is T, , , the fuel-air  equivalence ratio calculation procedure is : 1.  Compute the the air temperature in the airbox,  Ta b or  ,  p  T^= T [1+  2.  Assume Tit..  3.^Calculate the trapped mass. P V abox ipc  Ra To,  4.^Determine the delivery ratio. mad —  y -1  °x)Y --^6 ---#— ^P aid)  ^lio  i.e. at scavenged-blower exit. -1}]  128 5.  Obtain the delivery ratio, determine the degree of purity DP using the fit curve of the typical uniflow-scavenged data as shown in Figure 6.4. DP = 1.1527 A3  6.  -  0.4094 A2 + 0.0251 A  Recalculate the T ipc , Tip: = Tab°, DP +^(1 DP) -  7.  If^I Tip: - T1 ^5_ 0.5 , proceed to the next step. If not, revise Tipc and repeat step 2 up to 7.  8.  Compute the mass of air trapped. inau. = DP m u.  9.  Calculate the residual fraction f .es = 1 ,  10.  -  DP  Obtain n from the diesel to gas ratio using Eq. (I.1), and the calculate the stoichiometric fuel-air ratio FAswich using Eq. (I.2).  11.  Compute the fuel-air ratio. FA  12.  m, + m -  ^  gas  Determine the fuel air equivalence ratio. -  •  -  FA FA  ,  129 9.10 Appendix J ITERATION PROCEDURE  The two governing equations, i.e. Eq. (6.3) and Eq. (6.4), can be rewritten as follows Vr = X' Vb + (1—X)* V ag  ^  um = .x• ub + (1—x)• u.^  (J.1) (J.2)  Rearranging, Vat — Va = X' ( Vb — Vs ) — Um = X' ( Ub — Us )  From which, X ^V  Vm — Va — b  Let^A =  UN  =  — Vu^Ub — UN  um^a — u^ub — UN =  =  (J.3)  A• (ub — u.)  A ub — A u. + v.  If^B = — A u1 then^Vb  —  — V^ — V^ ^ in^a = Vb vu  hence^Vb — Vii or^Vb  UM  —  + vm^  A ub + B^  (J.4) (J.5)  130 In the beginning of the iteration, the following parameters are given, T2  9  P2  9  Vm^Vu  9  Um  9  Vu  where vm is the system specific volume, u m is its specific internal energy , and v. is the unburned gas specific volume, u u its specific energy. Then the iteration procedure is as follows : v-v — uY  1.  Calculate^A =  2.  Calculate^B = — A u. +  3.  Assume T b  4.  Look up the value of the burned gas specific volume v b and specific energy u b the burned gas table for the assumed value of  Tb and  the measured  pressure P2 . 5.  Check whether the equation Eq. (J.5), which is derived from the two conservation equations, is satisfied. vb = A ub + B  6.  If the above equation is satisfied then proceed to the next step. If not, revise Tb and  repeat step 4 and 5.  7.^Calculate the mass-burned fraction using Eq. (J.1) x =  8.^Iteration completed.  Vas  —  Vat  vb — v.  131  9.11 Appendix K LISTING OF THE MASS-BURNED FRACTION PROGRAM C This is program xpgdsl.for which takes engine pressure data C at regular crank angle increments DCA and determines massC burned fraction. C modified for gas-diesel engine by H.Gunawan (June19,1992) C^from xpresse.for. IMPLICIT REAL*8(A-H 2 O-Z) REAL*8 MAIR,MDSL,MGAS REAL*8 MREF,MTRAP,MATRAP,MTOT REAL*8 CA(360),P(360),XMB(360),T(360) REAL*8 PAVG(360),XAVG(360),TAVG(360) REAL*8 WRK(360),WAVG(360) REAL*8 QWL(360),QAVG(360) REAL*8 TU(360),TUAVG(360) REAL*8 TBRN(360),TBAVG(360) CHARACTER*50 PDAT COMMON/PROPS/RDG,RHC,FRES,EQVR,RU COMMON/STATS/N,NCYC,NCA,CABOI,RPM COMMON/GEOM/BORE,STROKE,ROD,CLRH COMMON/MASS/MTOT,MA1R,MDSL,MGAS COMMON/PORT/PABOX,CAIPC,PIPC,T1PC,TRES COMMON/AMBNT/PAMB,TAMB COMMON/BURN/UB C  C Specify the cylinder geometry C BORE is cylinder bore(m), STROKE(m), ROD is corm rod length(m), C CLRH is clearance height(m). STROKE = 0.1270D0 BORE = 0.10795D0 ROD = 0.2540130 CLRH = STROKE/15.0D0 C  c*******************Initi alize *********************** C^EQVR is the fuel-air equivalence ratio; subscripts C^ipc and boi refer to intake port closing and beginning C^of injection, respectively. C^CA is crank angle and subscript PR1 C^refers to the first pressure record. RU is the gas constant C^for the unburned gas per kg and CVU is the specific heat. C^NPR is the number of lines of pressure records per cycle. C^NCYC is the number of engine cycles to be analyzed; C^NCA is the no. of CA intervals after spark to be analyzed C^for each cycle. C^TIPC is the cylinder contents temp at ipc after mixing C^with residual gas. The residual gas mass fraction C^is determined using a scavenging data typical of C^two-stroke diesels. OPEN(UNIT=2,FILE='PDATA.DAT',STATUS='OLD') OPEN(UNIT=10,FILE= XPGDSL. OUT' ,STATUS='NEW' ) OPEN(UNTT=11,FILE='CH4EQ20.DAT',STATUS=' OLD') OPEN(UNIT=12,FTLE='CH4EQ40.DAT' ,STATUS='OLD') OPEN(UNTT=13 ,FILE='CH4EQ60. DAT' ,STATUS=' OLD')  132 OPEN(UNIT=14,FILE='CH4EQ80.DAT',STATUS='OLD') OPEN(UNIT=15,FTLE='CH4EQ90.DAT',STATUS='OLD') OPEN(UNTT=16,FILE='CH4EQ10.DAT',STATUS='OLD') OPEN(UNTT=21,FILE='CH2EQ20.DAT',STATUS='OLD') OPEN(UNIT=22,FILE='CH2EQ40.DAT',STATUS='OLD') OPEN(UNIT=23,FILE='CH2EQ60.DAT',STATUS='OLD') OPEN(UNIT=24,FILE='CH2EQ80.DAT',STATUS='OLD') OPEN(UNIT=25,FILE='CH2EQ90.DAT',STATUS='OLD') OPEN(UNTT=26,FILE='CH2EQ10.DAT',STATUS='OLD') READ(2,*)RPM,PAMB,TAMB,PABOX,TRES READ(2,*)MAIR,MDSL,MGAS,UHCRAT READ(2,*)CAPR1,CAIPC,CABOI,DCA READ(2,*)NPR,NCA,NCYC WRITE(6,40) READ(6,41)PDAT WRITE(1 0,42)PDAT 40 FORMAT(1X,'PDATA.DAT = ?') 41 FORMAT(A50) 42 FORMAT(1X/IX,'Pressure data : ',A50) C  C Estimate the mass of air in trapped cylinder charge MATRAP C ,and the equivalence ratio EQVR CALL INTAKE(MATRAP) MTRAP = MATRAP/(1.D0-FRES) MTOT = MTRAP+MDSL+MGAS C  WRII'E(10,50)NCYC WRITE(10,51)RPM,EQVR,CABOI WRITE(10,52)MAIR,TIPC,CAIPC WRITE(10,53)FRES,MATRAP,MTOT WRITE(10,54)RDG,RHC,UHCRAT 50 FORMAT(/,1X,14,'cycles of reduced by XPGDSL.FOR') 51^FORMAT(/,1X,'RPM',F7.1,' Equiv Ratio ',F6.3,' CAboi',F8.3) 52^FORMAT(1X,'Mair kg ',D10.4,' Tipc K',F6.1,' CAipc',F9.3) 53^FORMAT(1X,'Fres',F8.3,' Matrap kg ',E10.4,' Mtot kg ',E10.4) 54 FORMAT(1X,'Mdsl/Mgas ',F7.4,' RHC, Ruhc = ',2(1X,F5.3)) WRITE(10,103) print*,Init.ializing subroutines' CALL UNBURNED(Q1,Q2,Q3,Q4,1) CALL BURNED(Q1,Q2,Q3,Q4,Q5,Q6,Q7,Q8,1) CALL CYCSTATS(Q1,Q2,Q3,1) PBOIAV = 0.D0 DO 90 I = 1,NCA PAVG(I) = 0.D0 TAVG(I) = 0.D0 TUAVG(I) = 0.D0 TBAVG(I) = 0.D0 QAVG(I) = 0.D0 WAVG(I) = 0.D0 90 XAVG(I) = 0.D0 print*,'reading pressure data' DO 1000 N = 1,NCYC DO 100 I = 1,NPR 100 READ(2,*)CA(I),P(I) ICIPC = DINT((CAIPC - CAPR1)/DCA) + 1 1001 = DINT((CABOI - CAPR1)/DCA) + 1 VIPC = VCYL(CAIPC) VBOI = VCYL(CABOI) PBOI = P(KBO1) CA(1) = CABOI + DCA  133 DO 125 I = 1,NPR-KBOI .GT. 1)CA(I) = CA(I-1) + DCA 125 P(I) = P(I+KBOI) PBOI1 = P(1) CABOI1 = CABOI+1.D0 VBOI1 = VCYL(CABOI1) TUBOI1 = PBOIl*VB0I1/RU/MTOT PRINT*,'TUBOI1 = ',TUBOI1 CALL UNBURNED(TUBOILUU,CVU,VISC,2) ETOT = UU*MTOT V1 = VBOI1 P1 = PBOI1 TU1 = TUBOI1 Ti = TU1 XMB1 = 0.D0 C WRITE(10,104)CABOILPBOILTUBOI1 103 FORMAT(8X,'CA',10X,'P kPa',8X,'Tu K',10X,'Tb K',11X,'X') 104 FORMAT(1X,5(2X,E12.6)) C****calculate conditions at end of each crank angle interval****** XMAXP = 0.D0 XMAX = O.DO XMIN = 0.D0 CAXMIN = CABOI CAXMAX = CABOI DO 200 I = 2,NCA WR1TE(6,*)'calculating step ',I,' of cycle ',N CALL UNBURNED(TU1,UU,CVU,VISC,3) G = 1 .D0/(1.DO+CVU/RU) GAMMA = LDO + RU/CVU TU2 = TU1*(P(I)/P1)**G CALL UNBURNED(TU2,UU,CVU,VISC,2) VU = RU*TU2/P(I) V2 = VCYL(CA(I)) ASURF = ACYL(CA(I)) CALL QWALL(T1,V2,XMBI,ASURF,DQWL,1) DWRK = (Pl+P(I))/2.D0*(V2-V1) ETOT = ETOT - DWRK + DQWL VM = V2/MTOT UM = ETOT/MTOT CALL BURNED(P(I),TB,VU,VM,UU,UM,VB,XMB(I),2) IF (N .NE. 1) GO TO 1111 WRITE(10,104)CA(I),P(I),TU2,TB,XMB(I) C^WRITE(10,1104)CA(I),P(I),XMB(I),T1,GAMMA,TB C WRITE(10,1999)V2,VM,VU,VB,UM,UU,UB 1104 FORMAT(1X,6(E11.5,1X)) 1999 FORMAT(1X,'V2,VM,VU,VB,UM,UU,UB=',7(1X,D9.3)) 1111 CONTINUE IF(XMB(I) .GT. XMAXP) GO TO 170 IF(XMB(I) .GT. )(MIN) GO TO 180 XMIN = XMB(I) CAXMIN = CA(I) GO TO 180 170 XMAXP = XMB(I) CAXMAX = CA(I) 180 CONTINUE TBRN(I) = TB TU(I) = TU2 T(I) = TB*XMB(I) + (1-XMB(I))*TU2 WRK(I) = WRK1 + DWRK QWL(I) = QWL1 + DQWL C*************prepare for next step  CINH dOIS  (r£11AIXVVD)SIVISDAD TIVD HfINILNOD 0051 HIA111:(1)DAVX(1)DAVX(1)VD(170I'01)a1111A1 005I3 ((XI `5'I I H)VXIAVIM111103 5011 (I)DAV (1)DAVIIIUDAVIADDAVMDAVX(1)VD(50 I I '0 Oanxtik bAatsavold/Wonvd = (1)DAVd (DADMIVOINWDAVELL = (1)DAVEII (DADN).LVOINWDAVIlL = (1)0AVf1I (DADN),LVOINWDAVI = WDAVI (DADN)IV0INWDAVA1 = (T)DAVM (DADIN)IVO'13/WDAVO = (1)DAV6 (DADN).LVOINWDAVX = (I)DAVX 1A1d11/0(19/(10EIVD - (I)VD) = HVEL VDNI = I 0051 00 AVIOHXIOHVD(170 I '0 (DADN).LVOIVAVIOEld = AVIOHd Cs m1111:Xe.X`XII`API d.`X0r.VD:X8/LVW2103 IOI (IOVOI)HIIHM Csuope.g paumq-ssum ptre samssaid p8As-aicumasuo: x iAv mod zoi (ZO sapko Ae 10j spiCretm popstims HIINILNOD 0001 (z`HIAIXVVD)SIVISDAD TIVD I0Hd + AVIOHd = AVIOHd (I)d + (1)DAVd = (1)DAVd 00£ WfIL + (1)DAVIlL = (1)DAVIII (1)NIIELL + WDAVELL = WDAVHI WI + (1)DM/I = (I)DAVI (1)'1M) + (I)DAVO = (I)DAVO (1)3RINt + (I)DAVM = WDAVAA + (I)DAVX = (I)DAVX XVIAIX/WELIAIX = WHIAIX VDNI = I 00£ ou HIINILNOD 08Z XVIAIX = WHIAIX 08Z OI OD MOD + WHIAIX = WmArx xvinuu*avorrauKisaixvp - (Dva) + raysix- = mop OZZ 08Z 01 OD + (Dam = Wasix  (NalAix-) * IIAIONRCIA 'MEND (I)vD) = ozz oI OD (CHAIXVD^(Dva) OtiZ oI oD (XVIAIXVD^(I)VD)  VDNI = I 08Z OCI NLIAIXVDNIIAIX.= NHAIXVD'NIIAIX.'*.LNIHd XVIAIXVD`dXVIADC.= XVIAIXVD'dXVIAIX.'*INIad NHAIX + dXVIAIX - XVIAIX = XVIAIACI NIFAIXVD - XVIAIXVD = ZIAIONHCI II0EIVD - NIIAIXVD = - OCT = XVIAIX 21343 11"9 1°3 S°TISFWI S ° P*********D HfINILNOD 00Z (I)MAIX = tam = TI (!)}BIM = 13111M (1)1Mt) = MAO =  (Da =  ZA = IA  PEI  135 c**************************************************************** DOUBLE PRECISION FUNCTION VCYL(CA) IMPLICIT REAL*8 (A-H O-Z) COMMON/GEOM/BORE,STROKE,ROD,CLRH C CA is crank angle degrees ABDC. PI = 3.14159D0 APSTON = PI/4.DO*BORE**2 CAR = CA*P1/180.D0 Z = (1.D0 + 2.D0*ROD/STROICE + DCOS(CAR) 1^- DSQRT((2.D0*ROD/STROKE)**2 - (DSIN(CAR))**2)) 2^* STROKF-2.D0 + CLRH VCYL = Z*APSTON RETURN END C********************************************************** 2  SUBROUTINE INTAKE(MATRAP) C********************************************************** IMPLICIT REAL*8(A-H O-Z) REAL*8 MAIR,MDSL,MGAS REAL*8 MREF,MTRAP,MATRAP,MTOT REAL*8 D(3) COMMON/AMBNT/PAMB,TAMB COMMON/PORT/PABOX,CAIPC,PIPC,TIPC,TRES COMMON/MASS/MTOT,MAIR,MDSL,MGAS COMMON/PROPS/RDG,RHC,FRES,EQVR,RU C Calculate the blower-exit air temperature, TXBLO EFFBLO = 0.751)0 GSTAR = 0.2857D0 TABOX = TAMB*(1.D0+((PABOX/PAMB)**GSTAR-1.D0)/EFFBLO) C Estimate the residual gas mass fraction, Fres=Mres/(Mres+Matrap) C and the mass of air in trapped cylinder charge, MATRAP RA = 0.2871)0 PIPC = PABOX PIPC = 200.D0 VTRAP = VCYL(CAIPC) C Start iteration to find Degree of Purity 20 DEGP = 0.60 30 D(1) = 0.D0 D(2) = 0.D0 D(3) = 0.D0 M=1 40 DEGP = DEGP + 0.05D0 50 TIPC = TABOX*DEGP + TRES*(1.D0-DEGP) MTRAP = PABOX*VTRAP/RA/TIPC RDELIV = MAIR/MTRAP DEGPUR = 0.173611D0*RDELIV**3.D0-0.95982D0*RDELIV**2.D0 1^+1.774305*RDELIV - 0.19642D0 Y = TIPC - (TABOX*DEGPUR+I RFS*(1.D0-DEGPUR)) IF (Y .GT. 0.D0) GO TO 52 D(1) = DEGP GO TO 53 52 D(2) = DEGP 53 D(3) = D(1)*D(2) IF (D(3) .EQ. 0.1)0) GO TO 40 M=M+1 DEGP = 0.5D0*(D(1)+D(2)) PRINT*,'Deg.of Purity = ',DEGP IF (DABS(Y) .LE. 0.5130) GO TO 100 IF (M .LT. 50) GO TO 50 100 CONTINUE FRES = 1.D0 - DEGPUR 2  .  136 TIPC = (1.DO-FRES)*TABOX + FRES*TRES MATRAP = (1.DO-FRES) * MTRAP WRITE(10,60)TIPC,RDELIV,MTRAP 60 FORMAT(1X,'Tipc,Rdeliv,Mtrap=',3(4X,E10.4)) C RDG is mass ratio of diesel-fuel(CH2) to gas(CH4) C Replace CH4 + (16/14)*RDG CH2 by CHn C where n = RHC = (4+16/14*RDG*2)/(1+16/14/*RDG) C  C CHn + (l+n/4)/EQVR ( 02 + 3.76 N2 ) C ^> CO2 + n2 H2O + (l+n/4)(1-EQVR)/EQVR 02 + (l+n/4)/EQVR*3.76 N2 C  C RFASTO is the stoichiometric fuel-air ratio C RFA is the fuel-air ratio RDG = MDSLAVIGAS RHC = (28.D0+16.D0*RDG)/(7.D0+8.D0*RDG) RFASTO = (12.D0+RHC)/((l.D0+RHC/4.D0)*(32.D0+3.76D0*28.D0)) RFA = (MDSL+MGAS)/MATRAP EQVR = RFA/RFASTO PRINT*,'FRES,EQVR,RHC =',FRES,EQVR,RHC RETURN END C  c**************************************************************** SUBROUTINE UNBURNED(TU,UU,CVU,VISC,L) c**************************************************************** IMPLICIT REAL*8(A-H 2 O-Z) REAL*8 MWO2,MWN2,MWCH4,MWCH2,MWH2O,MWCO2 REAL*8 HFO2,HFN2,HFCH4,HFCH2,HFH2O,HFCO2 REAL*8 DHO2,DHN2,DHCH4,DHCH2,DHH2O,DHCO2 REAL*8 CP02,CPN2,CPCH4,CPCH2,CPH20,CPCO2 COMMON/PROPS/RDG,RHC,FRES,EQVR,RU IF(L .NE. 1) GO TO 10 MWO2 = 31.999D0 MWN2 = 28.013D0 MWCH4 = 16.043D0 MWCH2 = 14.026D0 MWH2O = 18.015D0 MWCO2 = 44.011DO C  c******** methane _diesehair mixtures ***************************** C Residual gas fraction affect properties of unburned gas only C CH4 + (16/14)RDG CH2 + (2+(32)(16/14)RDG)/EQVR (02+3.76N2) C ^> (1+(16/14)RDG) CO2 + (2+(16/14)RDG) H2O C^+ (2+(32)(16/14)RDG) (1-EQVR)/EQVR 02 C^+ (2+(32)(16/14)RDG) 3.76 N2 C  RDG1 = (16.D0/14.D0)*RDG YAIR = (2.D0+(3.D02.D0)*RDG1)/EQVR FSTAR = FRES/(1.D0-FRES) BOTTOM = LDO + RDG1 + YAlR*4.76D0 + FSTAR* 1^(3.D0+2.D0*RDG1+YAIR*(1.D0-EQVR)/EQVR+YA1R*3.76D0) XCH4 = 1/BOTTOM XCH2 = RDG1/BOTTOM X02 = YAIR*(1.D0+FSTAR*(1.D0-EQVR)/EQVR)/BOTTOM XN2 = YAIR*3.76D0*(1.DO+FSTAR)/BOTTOM XCO2 = (1.DO+RDG1)*FSTAR/BOTTOM XH2O = (2.DO+RDG1)*FSTAR/BOTTOM TOP = MWCH4 + RDG1*MWCH2 1 + X02*BOTTOM*MWO2 + XN2*BOTTOM*MWN2 2 + XCO2*BOTTOM*MWCO2 + XH20*BOTTOM*MWH20 WTMOL = TOP/BOTTOM  137 RU = 8.3143D0/WTMOL HFO2 = 0.0D0 HFN2 = 0.0D0 HFCH4 = -74873.0D0 HFCH2 = -32059.0D0 HFH2O = -241827.D0 HFCO2 = -393522.D0 PR1NT*,'WTMOL,RU =',WTMOL,RU C WRITE(9,45) WITVIOL,RU C WRITE(9,46) XCH4,XCH2,X02,XN2,XCO2,XH20 45 FORMAT(1X,'WTMOL, RU = ',2(1X,F8.3)) 46 FORMAT(1X,'X CH4,CH2,02, N2, CO2, H20 =',6(1X,F8.4)) RETURN 10 CONTINUE TDIM = TU/100.0D0 TDIMSQ = DSQRT(TDIM) TDIM2 = TDIM*TDIM TDIM3 = TDIM2*TDIM TDIM4 = TDIM3*TDIIVI TDIM14 = TDIM**0.25D0 TDIIvL54 = TDIM**1.25D0 TDIM74 = TDIIVI**1.75D0 TDIM32 = TDIM**1.5D0 TDIM52 = TDIM**2.5D0 TDIM34 = TDIM**0.75D0 C  IF(L .NE. 2) GO TO 20 C *****Calculate molar enthalpy differences between 298K and TU, C^then calculate the internal energy UU DI102 = 3743.2D0*TDIM+0.80408D0*TDIM52+35714.0D0/ 1^TDIMSQ-23688.0DO/IDIM-23906.63D0 DHN2 = 3906.0DO*TDIM+102558.0DO/TDIMSQ-107270.0D0/ 1^TDIM+41020.0DO/TDIIV12-39673.0D0 DHCH4 = -67287.0D0*TDIM+35179.2D0*TDIM54-1421.43D0* 1^TDIM74+64776.0DO*TDIMSQ-39436.89D0 DHCH2 = 10418D0*TDIM+2327.6DO*TDIM2-51714.3D0 DHCH2 = DHCH2/12.D0 DHH2O = 14305.0D0*TDIM-14683.2DO*TDEV154+5516.73D0* 1^TDIM32-184.95DO*TDIM2-11876.23D0 DHCO2 = 6914.5D0*TDIM-40.265D0*TDIM74-40154.0D0* 1^TDIIVISQ+70704.0D0*TDIM14-43912.73D0 UU = (X02*(HF02+DH02)+XN2*(HFN2+DHN2)+ 1 XCH4*(HFCH4+DHCH4)+XCH2*(HFCH2+DHCH2)+XH20*(HFH2O+DHH20)+ 2 XCO2*(HFCO2+DHCO2))/WTMOL - RU*TU RETURN 20 CONTINUE IF(L .NE. 3) GO TO 30 C ******Calculate specific heat CVU ********** CP02=37.432D0+0.020D0*TDIM32-178.57D0/TDIM32+236.88DO/TDIM2 CPN2=39.060D0-512.79D0/TDIM32+1072.7D0/TDIM2-820.40D0/1DIM3 CPCH4 = -672.87D0+439.74DO*TDIM14-24.875D0*TDIM34+ 1^323.88D0/TDIMSQ CPCH2 = 104.18D0+46.55D0*TDIM CPCH2 = CPCH2/12.D0 CPH2O = 143.05D0-183.54DO*TDIM14+82.751D0*TDIMSQ-3.6989D0*TDIM CPCO2 = 69.145D0-.70463D0*TDIM34-200.77DO/TDIMSQ+ 1^176.76D0/TDIM34 CVU = (X02*CP02+XN2*CPN2+XCH4*CPCH4+XCH2*CPCH2+XH20*CPH20+ 1^XCO2*CPCO2)/WTMOL - RU C WRITE(9,55)CPCH4,CPCH2,CP02,CPN2,CPCO2,CPH20 C WRITE(9,56)CVU  138 55 FORMAT(1X,'CP CH4,CH2, 02, N2, CO2, H2O =',6(1X,F8.4)) 56 FORMAT(1X,'CVU = ',F8.4) RETURN 30 CONTINUE C ********Estimate the mean viscosity of gas mixtures TM = TU**0.645D0 VISC = (X02 *MW02*5.09D0 + XN2*MWN2*4.57D0 + 1^XCH4*MWCH4*3.35D0 + XCH2*MWCH2*1.33D0 + 2^XH2O*MWH2O*3.26D0 + XCO2*MWCO2*3.71D0) 3^/WTMOL*10.D0**( -7.D0)*TM RETURN END c**************************************************************** SUBROUTINE BURNED(P,TB,VU,VM,UU,UM,VB,XIVIB,II) c***************************************************************************** IIVIPLICTT REAL*8(A-H O-Z) REAL*8 D(3) COMMON/PROPS/RDG,RHC,FRES,EQVR,RU COMMON/BURN/UB 2  C IF(H .NE. 1) GO TO 10 CALL TABLE(Q1,Q2,Q3,Q4,Q5,1) TBM1 = 1000.D0 XMl = 0.D0 XMB = 0 RETURN 10 CONTINUE C Find the linear relationship between UB and VB at the flame front IF( UM .NE. UU)GO TO 210 A = -1.DO/P B = VU - A*UU GO TO 220 210 A=(VM-VU )/(UM-UU ) B = VU - A*UU C SOLVE FOR T,V,U JUST BEHIND FLAME AND MASS FRACTION X C220 TB = TBM1 - 400.D0 220 TB = 1000.D0 230 D(1) = O.DO D(2) = 0.D0 D(3) = 0.D0 M=1 240 TB = TB + 100.D0 250 CALL TABLE(TB,P,SB,UB,VB,3) Y = VB - ( A*UB + B) IF( Y .GT. 0.D0) GO TO 252 D(1) = TB GO TO 253 252 D(2) = TB 253 D(3) = D(1)*D(2) IF(TB .GT. 6000.D0) WRITE(10,261) IF(TB .GT. 6000.D0) go to 300 IF( D(3) .EQ. 0.D0) GO TO 240 M= M+1 TB = 0.5D0*( D(1) + D(2) ) IF(DABS(Y) .LE. 0.1D-7 ) GO TO 300 IF( M .LT. 50) GO TO 250 WRITE(10,260) 260 FORMAT(1X,'NO CONVERGENCE ON FLAME TEMPERATURE IN 50 TRIES') 261 FORMAT(1X,'FLAME TEMPERATURE EXCEEDS 6000 K') RETURN 300 CONTINUE  139 C CALCULATE MASS FRACTION XMB CORRESPONDING TO ASSUMED PRESSURE XMB = (VM - VU )/(VB - VU) TBM1 = TB XM1 = XMB 321 RETURN END c**************************************************************** C  SUBROUTINE TABLE(TBX,PBX,SBX,UBX,VBX,L) C c**************************************************************** C^L =1 INITIALIZATION C L =2 GIVEN P AND SB C L =3 GIVEN P AND TB C IMPLICIT REAL*8(A-H O-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 AMREAD(6),UBREAD(6),SBREAD(6),EQREAD(6) REAL*8 AMREAD1(6),UBREAD1(6),SBREAD1(6),EQREAD1(6) REAL*8 AMREAD2(6),UBREAD2(6),SBREAD2(6),EQREAD2(6) COMMON/PROPS/RDG,RHC,FRES,EQVR,RU C IF(L .NE. 1) GO TO 20 PSTORE = -1000.D0 DO 12 I = 1,10 PTAB(I) = 0.D0 SP(I) = O.DO UP(I) = 0.D0 AMP(I) = 0.D0 TBTAB(I) = 0.D0 DO 11 J = 1,10 SBTAB(I,J) = 0.D0 UBTAB(I,J) = 0.D0 AMTAB(I,J) = 0.D0 11 CONTINUE 12 CONTINUE c**************************************************************************** C READ TABLE OF BURNED GAS PROPERTIES C Reads files for different equivalence ratios; unit 11(0.2), C unit 12(0.4), unit 13(0.6), unitl4(0.8), unitl5(0.9),16(1) DO 120 J = 1,10 DO 110 I = 1,10 DO 103 K = 1,6 READ(K+10,108)TBTAB(I),PTAB(J),AMREAD1(K),VB,UBREAD1(K), 1^H,SBREAD1(K) READ(K+20,108)TBTAB(I),PTAB(J),AMREAD2(K),VB,UBREAD2(K), 1^H,SBREAD2(K) 103 CONTINUE EQREAD(1) = 0.2D0 EQREAD(2) = 0.4D0 EQREAD(3) = 0.6D0 EQREAD(4) = 0.8D0 EQREAD(5) = 0.9D0 EQREAD(6) = LODO C Set a burned-gas properties table for a given H to C ratio RHC C by linearly interpolate the CH4 and CH2 tables at the same C equivalence ratio EQVR. 2  .  140 CALL CUBICS(6,EQREAD,AMREAD1,EQVR,AMTABla,n) CALL CUBICS(6,EQREAD,AMREAD2,EQVR,AMTAB2(I,J)) AMTAB(I,J)=A/vITAB2(1,1)+(RHC-2)/2*(AMTAB1(1,J)-AMTAB2(1J)) CALL CUBICS(6,EQREAD,UBREAD1 ,EQVR,UBTAB1 CALL CUBICS(6,EQREAD,UBREAD2,EQVR,UBTAB2(I,J)) UBTAB(1,J)=UBTAB2(I,J)±(RHC-2)/2*(UBTAB1(1,J)-UBTAB2(1,J)) CALL CUBICS(6,EQREAD,SBREADLEQVR,SBTAB1(1,J)) CALL CUBICS(6,EQREAD,SBREAD2,EQVR,SBTAB2(I,J)) SBTAB(1,J)=SBTAB2(1,1)+(RHC-2)2*(SBTAB1(1,J)-SBTAB2(1,J)) C READ(1,108)TBTAB(I),PTAB(J),AMTAB(I,J),VB,UBTAB(I,J),H, C 1^SBTAB(I,J) 108 FORMAT(1X,2(1X,F6.1),1X,F8.4,4(1X,D11.3)) C PTAB(J) = PTAB(J)* 101.325D0 UBTAB(I,J) = UBTAB(I,J)*0.001D0 C WRITE(10,109)TBTAB(1),PTAB(J),AMTAB(1,1),UBTAB(1,J),SBTAB(I,j) 109 FORMAT(1X,'TAB TB,P,AM,UB,SB=',5(1X,D10.4)) 110 CONTINUE 120 CONTINUE PRINT*,'A table of proportional-model burned gas properties' PRINT'*,'^for a given EQVR is formed' RETURN c******************************************************************************  am)  C GIVEN P AND SB OR TB 20 IF(PBX .EQ. PSTORE) GO TO 160 PSTORE = PBX NP = 10 XSET = PBX DO 131 I = 1,10 DO 130 J = 1,10 X(J) = PTAB(J) 130 Y(J) = UBTAB(I,J) CALL CUBICS(NP,X,Y,XSET,UP(I)) 131 CONTINUE DO 141 I = 1,10 DO 140 7 = 1,10 140 Y(J) = AMTAB(I,J) CALL CUBICS(NP,X,Y,XSET,AMP(I)) 141 CONTINUE DO 151 I = 1,10 DO 150 J = 1,10 150 Y(J) = SBTAB(I,J) CALL CUBICS(NP,X,Y,XSET,SP(I)) 151 CONTINUE 160 CONTINUE IF( L .NE. 2) GO TO 300 C************************************** .  XSET = SBX DO 230 I =1,10 X(I) = SP(I) 230 Y(I) = TBTAB(I) CALL CUBICS(NP,X,Y,XSET,TBX) DO 240 1 = 1,10 240 Y(I) = UP(I) CALL CUBICS(NP,X,Y,XSET,UBX) DO 250 1 = 1,10 250 Y(I) = AMP(I) CALL CUBICS(NP,X,Y,XSET,AMX) VBX = 8.3143D0/AMX*TBX/PBX RETURN C**************************************  141 300 XSET = TBX DO 330I = 1,10 X(I) = TBTAB(I) 330 Y(I) = SP(I) CALL CUBICS(NP,X,Y,XSET,SBX) DO 340 I = 1,10 340 Y(I) = UP(I) CALL CUBICS(NP,X,Y,XSET,UBX) DO 350 I = 1,10 350 y(I)= AMP(I) CALL CUBICS(NP,X,Y,XSET,AMX) VBX = 8.3143D0/AMX*TBX/PBX RETURN END C************************************************* SUBROUTINE CUBICS(NP,X,Y,XSET,YCALC) C NP IS NUMBER OF X,Y DATA PAIRS (I RUNS FROM 1 TO N) IMPLICIT REAL*8(A-H O-Z) REAL*8 X(10),Y(10),D(10),E(10),F(10),G(10) M = NP -1 MM = NP -2 C CALCULATION OF SECOND DERIVATIVES G(I) G(1) = 0.D0 G(NP) = 0.D0 DO 100 I = 2,M D(I) = X(1) - X(I-1) E(I) = 2.D0*( X(I+1) - X(I-1) ) 2  RD= x(1-F1) - X(I) 100 Gm= 6.D0/FaNya-m-y(I))+6.Domor(Y(I-i) - Y(1))  DO 1040 I = 2,MM FA = D(I+i)/E(1) E(I+1) = E(I+1) - FA*F(I) G(I+1) = G(I+1) - FA*G(1) 1040 CONTINUE DO 1070 I = 2,M G(NP+1 -1)=(G(NP+1 -1)-F(NP+1-1)*G(NP+2-I))/E(NP+1 -I) 1070 CONTINUE C CALCULATION OF INTERPOLATED VALUE YCALC AT X=XSET D(NP) = X(NP) - X(NP-1) I=1 200 1=1+ 1 IF( XSET .GE. X(I) .AND. I .LT. NP) GO TO 200 DELM = XSET - X(I-1) DELP = X(I) - XSET YCALC = G(I-1)/6.D0/D(I)*DELP**3 + G(1)/6.D0/D(1)*DELM**3 1^+(Y(I-1)/D(I) -G(1-1)*D(I)/6.D0)*DELP 2^+(Y(I)/D(1) -G(1)*D(1)/6.D0 )*DELM RETURN END c**************************************************************** SUBROUTINE CYCSTATS(CA,P,X,L) IMPLICIT REAL*8(A-H O-Z) REAL*8 CA(360),P(360),X(360),CAPMAX(200) REAL*8 T10(200),T20(200),T30(200),T40(200),T50(200),PMAX(200) COMMON/STATS/N,NCYC,NCA,CABOI,RPM C^L = 1 : Initialize C^L = 2 : Compute times/secs from spark to 10%, 20%, 30%,40%,50% C^of mass-fraction burned for each cycle C^L = 3 : Compute averages and std deviations for all cycles IF( L .NE. 1) GO TO 10 DO 5 I = 1,NCYC 2  142 T10(1) = 0.D0 T20(I) = 0.D0 T30(1) = 0.D0 T40(1) = 0.D0 T50(1) = 0.D0 5 PMAX(1)= 0.D0 RETURN 10 IF(L .NE. 2) GO TO 200 CA10 = 0.130 CA20 = 0.D0 CA30 = O.DO CA40 = 0.D0 CA50 = 0.D0 DO 50 I = 2,NCA CAI = CA(I) CAIM1 = CA(I-1) XI = X(1) XIM1 = X(I-1) DENOM = XI - XEM1 IF(DENOM .EQ. 0.D0) GO TO 50 DCADX = (CM - CAIM1)/DENOM IF(XI .LT. 0.1D0) GO TO 50 IF(CA10 .NE. 0.D0) GO TO 20 CA10 = CM - DCADX*(XI - 0.1D0) - CABOI GO TO 50 20 IF(XI .LT. 0.21)0) GO TO 50 IF(CA20 .NE. 0.D0) GO TO 30 CA20 = CM - DCADX*(XI - 0.2D0) - CABOI GO TO 50 30 IF(XI .LT. 0.3D0) GO TO 50 1F(CA30 .NE. O.DO) GO TO 40 CA30 = CM - DCADX*(XI - 0.3130) - CABOI GO TO 50 40 IF(XI .LT. 0.4D0) GO TO 50 IF(CA40 .NE. 0.130) GO TO 45 CA40 = CM - DCADX*(XI - 0.4D0) - CABOI GO TO 50 45 IF(XI .LT. 0.5130) GO TO 50 IF(CA50 .NE. 0.130) GO TO 50 CA50 = CM - DCADX*(XI - 0.5D0) - CABOI 50 CONTINUE T10(N) = CA10/6.D0/RPM T20(N) = CA20/6.DO/RPM T30(N) = CA30/6.DO/RPM T40(N) = CA40/6.DO/RPM T50(N) = CA50/6.D0/RPM C^Determine Pmax for each cycle PMAX(N) = 0.130 DO 175 I = 2,NCA IF(P(1) .GT. PMAX(N) ) CAPMAX(N)=CA(I) 175 IF(P(I) .GT. PMAX(N) ) PMAX(N) = P(I) C^write time elapsed(sec) for 10%,20%,30%,40%,50% mass burned WRITE(10,51)T10(N),120(N),T30(N),T40(N),T50(N),PMAX(N),CAPMAX(N) 51 FORMAT(1X,'T10,T20,T30,T40,T50,Pmax,CApmax=',7(1X,D12.4)) RETURN 200 CONTINUE C^Determine averages for all cycles: 10%,20% etc burned TlOME = 0.D0 T2OME = 0.130 T3OME = 0.130 T4OME = 0.130  143 T5OME = 0.130 DO 300 N = 2,NCYC TlOME = TlOME + T10(N) T2OME = T2OME + T20(N) T30ME = T30ME + T30(N) T4OME = T4OME + T40(N) T5OME = T5OME + T50(N) 300 CONTINUE TOTALN = FLOAT(NCYC-1) TlOME = T1OME/TOTALN T2OME = T20ME/TOTALN T30ME = 'T30ME/TOTALN T4OME = T40ME/TOTALN T5OME = T5OME/TOTALN C^Determine standard deviation of times to 10%,20% etc burned SIG10 = 0.D0 SIG20 = 0.D0 SIG30 = 0.D0 SIG40 = 0.D0 SIG50 = 0.D0 DO 400 N = 2,NCYC SIG10 = SIG10 + (T10(N) - TlOME)**2 SIG20 = SIG20 + (T20(N) - T20ME)**2 SIG30 = SIG30 + (T30(N) - T30ME)**2 SIG40 = SIG40 + (T40(N) - T4OME)**2 SIG50 = SIG50 + (T50(N) - T5OME)**2 400 CONTINUE SIGIO = DSQRT(SIG10/TOTALN) SIG20 = DSQRT(SIG20/TOTALN) SIG30 = DSQRT(SIG30/TOTALN) SIG40 = DSQRT(SIG40/TOTALN) SIG50 = DSQRT(SIG50/TOTALN) C^Determine normalized covariances for 10%,20% etc burned CR1020 = 0.D0 CR1030 = 0.130 CR1040 = 0.130 CR1050 = 0.D0 CR2030 = 0.D0 CR2040 = 0.130 CR2050 = 0.130 C^Determine Std Dev of 10% to 50% burning times SIG5M1 = 0.130 DO 500 N = 2,NCYC CR1020 = CR1020 + (T10(N)-T10ME)*(T20(N)-T20ME) CR1030 = CR1030 + (T10(N)-T10ME)*(T30(N)-T30ME) CR1040 = CR1040 + (T10(N)-T10ME)*(T40(N)-T40ME) CR1050 = CR1050 + (T10(N)-T10ME)*(T50(N)-T50ME) CR2030 = CR2030 + (T20(N)-T20ME)*(T30(N)-T30ME) CR2040 = CR2040 + (T20(N)-T20ME)*(T40(N)-T40ME) CR2050 = CR2050 + (T20(N)-T20ME)*(T50(N)-T50ME) SIG5M1 = SIG5M1 + (T50(N) - T10(N))**2 500 CONTINUE CR1020 = CR1020/SIG10/SIG20/TOTALN CR1030 = CR1030/SIGIO/SIG30/TOTALN CR1040 = CR1040/SIG10/SIG40/TOTALN CR1050 = CR1050/SIGIO/SIG50/TOTALN CR2030 = CR2030/SIG20/SIG30/TOTALN CR2040 = CR2040/SIG20/SIG40/TOTALN CR2050 = CR2050/SIG20/SIG50/TOTALN SIG5M1 = DSQRT(SIG5MI/TOTALN)  144 WRITE(10,71) WRITE(10,711) NCYC VVRITE(10,72) WRITE(10,73)T1OME,T20ME,T3OME,T40ME,T5OME WRITE(10,74)SIGIO,SIG20,SIG30,SIG40,SIG50 WRITE(10,76) WRITE(10,77)CR1020 WRITE(10,78)CR1030 WRITE(10,79)CR1040 WRITE(10,80)CR1050 WRITE(10,81)CR2030 WRITE(10,82)CR2040 WRITE(10,83)CR2050 WRITE(10,831)SIG5M1 CA10 = T1 OME*6.DO*RPM CA20 = T2OME*6.D0*RPM CA50 = T5OME*6.D0*RPM WRITE(10,84)CA10,CA20,CA50 71 FORMAT(1X,'Statistics of Time from Spark to Given X') ') 711 FORMAT(1X,I4,' Cycles from Data Set^ 72 FORMAT(1X,'X = ',18X,'10%',9X;20%',9X,'30%',9X;40%',9X,'50%') 73 FORMAT(1X,'Mean Time (sec)',3X,5(2X,D10.4)) 74 FORMAT(1X,'Std Dev (sec)',5X,5(2x,D10.4)) 76 FORMAT(1X,'Cross Correlation Factors') 77 FORMAT(1X,'10% - 20% Times',F10.3) 78 FORMAT(1X,'10% - 30% Times',F10.3) 79 FORMAT(1X,'10% - 40% Times',F10.3) 80 FORMAT(1X,'10% - 50% Times',F10.3) 81 FORMAT(1X,'20% - 30% Times',F10.3) 82 FORMAT(1X,'20% - 40% Times',F10.3) 83 FORMAT(1X,'20% - 50% Times',F10.3) 831 FORMAT(1X,'Std Dev of 10% to 50% burning time =',F10.3) 84 FORMAT(1X,'Crank Angles to X3.1,0.2,0.5=',3(1x,f10.2)) C^Determine mean value of Pmax PMAXME = 0.D0 DO 600 N = 2,NCYC 600 PMAXME = PMAXME + PMAX(N)/TOTALN C.....Determine standard deviation of Pmax SIGP = 0.D0 DO 700 N = 2,NCYC 700 SIGP = SIGP + (PMAX(N) - PMAXME)**2 SIGP = DSQRT(SIGP/TOTALN) SPREL = SIGP/PMAXME WRITE(10,85)PMAXME,SIGP,SPREL 85 FORMAT(///,1X,'Mean Pmax= ',D12.6,3x,'Std Dev Pmax= ',D12.6, 1 3x,'Relative Std Dev Pmax= ',D12.6) C^Sort in order of 10% times and write times for all cycles NSORT = NCYC CALL PIKSR2(NSORT,T10,Pmax) DO 1100 I = 2,NCYC WRITE(10,56)I,T10(l),Pmax(I) 56 FORMAT(1X,'Ncycle =',I4,3X,'10% time/s = ',D12.6, 1 ' Pmax/lcPa = ',D12.6) 1100 CONTINUE C^Sort in order of Pmax and print for all cycles  145 NSORT = NCYC CALL PIKSR2(NSORT,Pmax,T10) DO 1200 I = 2,NCYC WRITE(10,87) I,Pmax(I),T10(I) 87 FORMAT(1X,'Ncycle = ',14,2X,'Pmax = ',D12.6, 1 ' 10% time/s = ',D12.6) 1200 CONTINUE C^Determine normalized covariance for 10% time and Pmax CRT1Pm = O.DO DO 1300 I = 2,NCYC CRT1Pm = CRT1Pm + (T10(l)-T10ME)*(Pmax(I)-Pmaxme) 1300 CONTINUE CRT1Pm = CRT1Pm/SIG10/SigP/TOTALN WRITE(15,88)CRT1Pm 88 FORMAT(1X,'Normalized Covariance for 10% burning time 1 and maximum pressure = ',D10.4) RETURN END C************************************************************* SUBROUTINE PIKSR2(N,ARR,BRR) C************************************************************* C Sorts an array ARR of length N into ascending numerical order C by straight insertion. N is input; ARR is replaced on output C making the corresponding rearrangement of array BRR C  REAL*8 ARR(N),BRR(N) DO 12 J = 2,N A = ARR(J) B = BRR(J) DO 11 I = J-1,1,-1 IF(ARR(I) .LE. A) GO TO 10 ARR(I+1)=ARR(I) BRRa+1 BRR(I) 11^CONTINUE I=0 10^ARR(1+1) = A BRR(I+1) = B 12^CONTINUE RETURN END C  C******************************************************** DOUBLE PRECISION FUNCTION ACYL(CA) C Calculates the cylinder surface area for a given degree CA IMPLICIT REAL*8 (A-H O-Z) COM.MON/GEOM/BORE,STROKE,ROD,CLRH 2  C  PI = 3.14159D0 APSTON = PI/4.D0*BORE**2 CAR = CA*PI/180.D0 Z = ( LDO + 2.D0*ROD/STROKE + DCOS(CAR) 1 -DSQRTa2.D0*ROD/STROKE)**2+(DSIN(CAR))**2))*STROKE/2.D0 2 + CLRH ACYL = Z*PI*BORE + 2.D0*APSTON RETURN END C********************************************************* SUBROUTINE QWALL(TM1,V2,XMI3,ASLTRF,DQWL,L) C********************************************************* C Calculates the heat transfer from the gas to the cylinder wall  146 C using Annand's and Woschni's correlation. C IMPLICIT REAL*8 (A-H O-Z) REAL*8 MTOT COMMON/GEOM/BORE,STROKE,ROD,CLRH COMMON/MASS/MTOT,MAIR,MDSL COMMON/STATSNNCYC,NCA,CABOLRPM C IF (L .NE. 1) GO TO 20 DQWL = 0 RETURN 20 CONTINUE AAA = 0.38D0 BBB = 0.75 CCC = 1.6E-12 PISVEL = RPM * STROKE / 30.0 DENS = MTOT / V2 CALL UNBURNED(TM1,UU,CVU,VISC,4) RENUM = DENS * PISVEL * BORE / VISC CALL UNBURNED(TM1,UU,CVU,VISC,3) TRMLCO = CPG * VISC / 0.7D0 C^The wall temperature is assumed to be constant TW = 450.0D0 QCONV = ASURF * AAA * TRMLCO / BORE * RENUM**(BBB) * (TM1-TW) QRAD = ASURF * CCC * ( TM1**4 - TW**4 ) PRINT*,'QCONV,QRAD=',QCONV,QRAD DQWL = (QCONV+QRAD) DQWL = DQWL * (60./RPM/360.) RETURN END C C***************************************** 2  147 Appendix 6.5 VERIFICATION OF THE COMPUTATION PROCEDURE FOR THE CONSTANT VOLUME CASE  1.  Run the program for an arbitrary pressure data. Obtain the unburned gas properties at the beginning of the first calculating step which become the initial condition of the combustion process. The composition of the unburned gas is found to be: Constituent  Mole fraction  CH4  0.0436  CH2  0.0133  02  0.1845  N2  0.7345  CO2  0.0087  H2 O  0.0154 (1.0000)  Its specific volume, v. = 0.182067 m3/kg ; specific energy,^u„ = - 275.282 kJ/kg. molecular weight, M. = 28.028 kg/kmol.  2.  The above unburned gas mole fractions give the atom relative populations of C , H , 0 , N : 0.0656, 0.2318, 0.4018, 1.469, respectively.  148 3.  Run STANJAN using the above initial condition, i.e. atom relative populations, specific volume v. , and specific energy u. , to determine the pressure P.,„, and temperature T^of the mixture at the end of combustion. It was found that :^= 3328.3 kPa, T  ,  ,  = 2055.68 K.  While the burned gas molecular weight, Mb = 28.205 kg/kmol; which is 0.6% higher than that of the unburned gas.  4.  In the mass-burned fraction program, consider the first step calculation as a constant volume combustion (V2 = V2) and specify the pressure at the end of the step P2 = Pte . Ideally, the output of the program will show that the burned gas temperature is  equal to that of STANJAN, Tb = T  ,  ,  . Moreover, the program result should  demonsrates that all the fuel is burned at the end of the process, x = 1.0.  5.^Output of the program are : (i)  Burned Temperature, Tb = 2062.87 K. This result is within 0.35% compare to the one we get from STANJAN.  (ii)  Mass-burned fraction, x = 0.995260, which is within 0.47% to the ideal case.  


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