UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Swirling combustion of premixed gaseous reactants in a short cylindrical chamber Pierik, Ronald Jay 1987

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

Item Metadata

Download

Media
831-UBC_1987_A7 P57.pdf [ 10.87MB ]
Metadata
JSON: 831-1.0097192.json
JSON-LD: 831-1.0097192-ld.json
RDF/XML (Pretty): 831-1.0097192-rdf.xml
RDF/JSON: 831-1.0097192-rdf.json
Turtle: 831-1.0097192-turtle.txt
N-Triples: 831-1.0097192-rdf-ntriples.txt
Original Record: 831-1.0097192-source.json
Full Text
831-1.0097192-fulltext.txt
Citation
831-1.0097192.ris

Full Text

SWIRLING COMBUSTION OF PREMIXED GASEOUS REACTANTS IN A SHORT CYLINDRICAL CHAMBER BY RONALD JAY PIERIK B.Sc, The University of Michigan, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Mechanical Engineering We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1987 (c) Ronald Jay Pi e r i k , 1987 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. Department of Mechanical Engineering The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 D a t e August- 1987 DE-6(3/81) i i ABSTRACT The e f f e c t s of s w i r l and spark l o c a t i o n on combustion duration were studied i n a constant volume c y l i n d r i c a l chamber of length-to-diameter r a t i o of 0.5. A chemically balanced methane-air mixture was swirled up to 628 radians per second by tangential i n j e c t i o n . The chamber was closed by a valve before i g n i t i o n by a spark gap of v a r i a b l e l o c a t i o n and electrode geometry. The burning duration, indicated by repeated measurements of combustion pressure r i s e , was found to be a strong function of s w i r l i n t e n s i t y and spark loca t i o n . Increased swirl resulted i n decreased burning duration; mid-radius i g n i t i o n l o c a t i o n combined with high s w i r l resulted i n the shortest combustion durations. Spark gap was found to have an important e f f e c t on the standard deviation of the burning duration, e s p e c i a l l y with high s w i r l . Various "flame holders" were i n s t a l l e d to achieve shorter burning durations and lower c y c l i c v a r i a t i o n . Results indicated that the best i g n i t i o n source geometry was an unshielded, low-drag probe. This gave the l e a s t burning durations and the l e a s t c y c l i c v a r i a t i o n at the higher s w i r l values. i i i TABLE OF CONTENTS /Abstract i i Table of Contents i i i L i s t of Tables v i i i L i s t of Figures i x Acknowledgments x i i i Dedication xiv Chapter One; Introduction 1 1.1 Fast Burning i n Engines 1 1.1.1 Advantages 1 1.1.2 Disadvantages 2 1.1.3 Mass Burning Rate 4 1.2 Swirling Combustion i n Engines 6 1.2.1 Methods of Swirl Generation 6 1.2.2 Turbulence Intensity 8 1.2.3 Buoyancy E f f e c t s 8 1.2.4 Combined E f f e c t s of Swirl and Squish 9 1.3 Research Objectives 10 Chapter Two: Literature Review 11 2.1 Introduction 11 2.2 Swirling Combustion In Constant Volume Chambers 11 2.2.1 Dyer (1979) 11 i v 2.2.2 Inoue, Nakanishi, Noguchi, and Iguchi (1980) 12 2.2.3 Zawadzki and Ja r o s i n s k i (1983) 13 2.2.4 Hanson and Thomas (1984) 14 2.3 Swirling Combustion In Single Cylinder Research Engines 15 2.3.1 Wakuri, Kido, Ono, Nakashima, and Murase (1981) 15 2.3.2 Witze and V i l c h i s (1981) 16 2.3.3. Witze (1982) 17 2.3.4 Groff and Sinnamon (1982) 18 2.4 Swirling Combustion In Mult i c y l i n d e r Engines 19 2.4.1 Mayo (1975) 19 2.4.2 Nagayama, Araki, and Iioka (1977) 22 2.4.3 Nagao and Tanaka (1983) 23 Comparison of Observations 25 2.5 Swirling V e l o c i t y Distributions 2 5 2.6 E f f e c t of Swirl on Burning Duration 26 2.7 Spark Location 27 2.8 Flame Adhesion 2 9 2.9 C y c l i c V a r i a t i o n 30 2.10 Lean Limit 31 2.11 Volumetric E f f i c i e n c y 31 2.12 Emissions 32 2.13 Heat Transfer 3 3 2.14 Summary 3 3 V Chapter Three; Apparatus 36 3.1 Introduction 3 6 3.2 Combustion Chamber 36 3.3 Valves 3 6 3.4 Flow C i r c u i t 38 3.5 Instrumentation 40 3.6 Ig n i t i o n and Control 42 3.7 Flame Holders 43 3.8 Schlieren Photographic Arrangement 43 3.9 C a l i b r a t i o n 44 3.9.1 Combustion Chamber Swirl Rate 44 3.9.2 Pressure Transducer 46 3.9.3 Ignition Delay Timer 46 3.9.4 Large Low-Range Pressure Gauge 47 3.9.5 Barometer 47 3.9.6 Gas Chromatograph 47 3.10 Testing 48 3.10.1 System Pressure and Vacuum 48 3.10.2 Fan Shaft Seal Leakage 49 3.10.3 Valve Closure Time 49 Chapter Four: Procedures 52 4.1 Mixture Preparation 52 4.2 Chamber F i r i n g Test Sequence 53 4.2.1 Start Up 53 4.2.2 Repeat F i r i n g 55 4.3 Photography 56 4.4 Data Processing 57 4.5 Measurement Uncertainties 58 4.5.1 Introduction 58 4.5.2 Swirl Intensity 60 4.5.3 Gas Chromatograph and Mixtures 60 4.5.4 Pre-Ignition Pressure 61 4.5.5 Data Processing 61 Chapter Five: Results 63 5.1 Introduction 63 5.2 Repeatability Of Measurements 66 5.3 V a r i a t i o n Of Estimated Standard Deviation With Sample Size 66 5.4 E f f e c t Of Swirl On Burning Duration And Heat Transfer 70 5.5 E f f e c t Of Spark Location On Burning Duration 71 5.6 E f f e c t Of Swirl And Spark lo c a t i o n On C y c l i c V a r i a t i o n 74 5.7 E f f e c t Of Spark Gap On Burning Duration 76 5.8 E f f e c t Of Spark Gap On C y c l i c V a r i a t i o n 76 5.9 E f f e c t Of Flame Holders 77 Chapter Six: Conclusions 80 6.1 E f f e c t of Swirl on Burning Duration and V a r i a t i o n 80 6.2 E f f e c t of Spark Location on Burning Duration 80 6.3 E f f e c t of Spark Gap 81 v i i 6.4 E f f e c t of Flame Holders 82 6.5 Recommended Future Work 82 Figures 85 Tables 145 Bibliography 156 Appendix A ; Gas Chromatograph Operation 159 Appendix B: Mixture Preparation Calculations 164 Appendix C: Pressure Transducer C a l i b r a t i o n 165 Appendix D: Eguipment Specifications 174 Appendix E: Datanic L i s t i n g 187 Appendix F: Riahi L i s t i n g 191 v i i i LIST OF TABLES Chapter Four 4.1 Equipment Switch Settings 4.2 Test of "RIAHI" 145 146 Chapter Five 5.1 Summary of Burning Duration For Mid-Axis Spark Locations 147 5.2 Summary of Standard Deviation For Mid-Axis Spark Locations 148 5.3 Comparison Of Burning Duration For Mid-Plane And Endwall Spark Locations 149 5.4 E f f e c t Of Spark Gap On Burning Duration 150 5.5 E f f e c t Of Spark Gap On C y c l i c V a r i a t i o n 151 5.6 E f f e c t of "Flame Holders" On Burning Duration 152 5.7 E f f e c t of "Flame Holders" On C y c l i c V a r i a t i o n 153 Appendix A A . l Oven Temperatures and Retention Times 154 Appendix C C.1 Range Capacitor Values 155 ix LIST OF FIGURES Chapter One 1.1 Various Heat-Addition Processes 85 1.2 E f f i c i e n c y Comparison of Ideal Engine Cycles 86 1.3 Mass Entrainment Parameters 87 1.4 Swirl Generation Methods 88 1.5 E f f e c t of Swirl Axis Perpendicular To Cylinder Axis 89 1.6 May " F i r e b a l l " Head 90 1.7 Secondary Flows 91 Chapter Three 3.1 Apparatus Assembly Drawing 92 3.2 Combustion Chamber Cross-Section 93 3.3 Swirl Valve 94 3.4 No-Swirl Valve 95 3.5 C i r c u l a t i o n Fan 96 3.6 Fan Seal 97 3.7 Instrumentation / Ignition / Camera Arrangement 98 3.8 Spark Gap Arrangement 99 3.9 Schematic of Delay Timer 100 3.10 Delay Timer Switch Arrangement 101 3.11 Flame Holder Designs 102 3.12 Flame Holder Arrangements 103 3.13 Schlieren Photographic Arrangement 104 3.14 Paddle Wheels 105 3.15 Chamber Swirl Rate Versus C i r c u l a t i n g Fan Speed 106 X 3.16 Swirl Decay 107 3.17 I g n i t i o n Time Delay C a l i b r a t i o n Curve 108 3.18 Chromatograph C a l i b r a t i o n Curve: Methane (CH4) 109 3.19 Seal Pressure Retention 110 3.21 Apparatus Vibration 111 3.22 Apparatus Vibration 112 3.23 Apparatus Vibration 113 Chapter Four 4.1 High-Speed Camera 114 Chapter Five 5.1 Experiment Repeatability 115 5.2 Standard Deviation Versus Sample Size: Central Ign. 116 5.3 Standard Deviation Versus Sample Size: Outside Ign. 117 5.4 E f f e c t of Swirl on Burning Duration: Outside Location 118 5.5 E f f e c t of Swirl on Burning Duration: Central Location 119 5.6 E f f e c t of Swirl on Burning Duration: Mid-radius Location 120 5.7 E f f e c t of Swirl on t 1 Q 121 5.8 E f f e c t of Swirl on t 1 0 _ Q 0 121 5.9 E f f e c t of Spark Location on Burning Duration: 0 RPM 122 5.10 E f f e c t of Spark Location on Burning Duration: 2000 RPM 123 5.11 E f f e c t of Spark Location on Burning Duration: 4000 RPM 124 5.12 E f f e c t of Spark Location on Burning Duration: x i 6000 RPM 125 5.13 Pressure Curves for Central Ig n i t i o n : 2000 RPM 126 5.14 Pressure Curves for Central Ig n i t i o n : 4000 RPM 127 5.15 Pressure Curves for Central Ignition: 6000 RPM 128 5.16 Pressure Curves f o r Outside Ig n i t i o n : 6000 RPM 129 5.17 Pressure Curves for Mid-Radius Ignition: 6000 RPM (HSM1) 130 5.18 Pressure Curves for Mid-Radius I g n i t i o n : 6000 RPM (HSM2) 131 5.19 E f f e c t of Spark Location on Standard Deviation: 0 RPM 132 5.20 E f f e c t of Spark Location on Standard Deviation: 2000 RPM 133 5.21 E f f e c t of Spark Location on Standard Deviation: 4000 RPM 134 5.22 E f f e c t of Spark Location on Standard Deviation: 6000 RPM 135 5.23 E f f e c t of Spark Gap on C y c l i c V a r i a t i o n 136 5.24 E f f e c t of Spark Gap on Burning Duration 136 5.25 Schlieren Photographic Sequence: no flame holder 137 5.26 Schlieren Photographic Sequence: no flame holder 138 5.27 Schlieren Photographic Sequence: flame holder number four 139 5.28 Schlieren Photographic Sequence: flame holder number two 140 Appendix A A . l Gas Chromatograph 141 X l l A.2 Chromatograph Calibration Curve: Ethane (C 2H 6) 142 A.3 Chromatograph Calibration Curve: Propane (C 3H 8) 143 Appendix C C.l Pressure Transducer Calibration Curve 144 x i i i ACKNOWLE DGEMENTS I wish to express my sincerest appreciation to my advisor, Professor P h i l i p G. H i l l , whose dedication, support, and coaching throughout t h i s e n t i r e project helped me stay on track and see i t through enjoyably. I also l i k e to thank Professors Martha E. Salcudean, and Edward G. Hauptmann, for t h e i r contributions on the defense committee. Thanks also to Mr. Tony Besic whose s k i l l at operating t o o l s made the seemingly impossible possible. To Ms. Paula Parkinson of the Environmental Engineering Laboratory who che e r f u l l y aided me i n c a l i b r a t i n g and using the Gas Chromatograph, I say thanks. I also appreciated the help of Mr. Mike Savage with the schlieren photography. In addition, I would l i k e to thank the faculty, s t a f f , and students of the department of Mechanical Engineering for t h e i r i n t e r e s t , support, and tolerance of t h i s noisy project. F i n a l l y , t h i s project was made possible by funding from the Natural Science and Engineering Research Council. This i s dedicated to Carla, and Adriana and Derk who are my loving and supportive wife and parents. 1 1. INTRODUCTION 1.1 Fast Burning In Engines 1.1.1 Advantages The benefits of fas t burning i n spark i g n i t i o n engines (which are increased e f f i c i e n c y and output and decreased engine octane requirement) have been known for decades; some of the f i r s t fast-burning designs were incorporated i n engines of the 1920's (Hempson 1976). The main motivation f o r fast burning i s high e f f i c i e n c y . Figs. 1.1 and 1.2, which show the e f f e c t s of compression r a t i o on three i d e a l heat addition processes, indicate that the cycle with constant volume heat addition ( i . e . , i n f i n i t e l y f a s t combustion), has the highest e f f i c i e n c y of the three id e a l heat addition processes. Fast-burning engines tend to require lower octane fuel because of the lower time avai l a b l e f o r end gas heating. Detonation of end gases increases with temperature and time so that a decrease i n combustion time with fa s t e r burning r e s u l t s i n a decrease i n detonation. Because spark advance i s reduced and i g n i t i o n takes place at higher temperature i n fast-burning engines, flammability l i m i t s can be widened. I t i s possible to operate at considerably leaner mixtures than with slow-burning engines. In general, fast-burning engines have higher peak pressures, lower exhaust temperatures, and greater e f f i c i e n c y than engines with slow burning. Fast burning requires less spark advance f o r best torque, but tends to produce more 2 nitrous oxides because of higher combustion temperatures. However, fast-burning engines can t o l e r a t e more r e c i r c u l a t e d exhaust gas; t h i s means that nitrous oxide l e v e l s can be reduced to t o l e r a b l e l e v e l s without reduced e f f i c i e n c y (Nagao and Tanaka, 1983, p. 153). 1.1.2 Disadvantages Usually f a s t burning implies rapid mixture motion. I f such motion i s caused by constricted ports or valve areas the penalty may be reduced volumetric e f f i c i e n c y and engine output. However, i f mixture motion i s created by a l t e r a t i o n of piston and head geometry, or by imparting s w i r l i n the chamber, the adverse e f f e c t s of c o n s t r i c t i o n on volumetric e f f i c i e n c y can be avoided. Enhanced mixture motion increases heat tr a n s f e r to the walls of the combustion chamber. The r a p i d l y moving mixture i s heated by the cylinder walls before combustion; t h i s decreases mixture density, volumetric e f f i c i e n c y , and engine power. Heat transfer from the burned gas during and a f t e r combustion decreases e f f i c i e n c y since t h i s reduces pressure at a given crank angle. Another disadvantage of fas t burning i s the increased pressure r i s e rate which increases engine noise and roughness. Mechanical f a i l u r e may also r e s u l t , e s p e c i a l l y at high compression r a t i o s . In the past noise, roughness, and mechanical f a i l u r e tended to discourage development of f a s t -burning designs. Techniques to reduce the problems of engine noise, 3 combustion roughness, and mechanical f a i l u r e include retarding the spark timing to reduce the rate of pressure r i s e , and strengthening the engine members so that there i s less v i b r a t i o n . Both techniques r e s u l t i n a net loss of performance since retarding the spark timing reduces e f f i c i e n c y , and strengthening e n t a i l s more weight. Fast burning may have l i t t l e e f f e c t on hydrocarbon emissions. The increased mixture motion t y p i c a l of f a s t -burning chambers may help to scrub the cool cy l i n d e r walls of unburned hydrocarbons and thus make burning more complete. However, with the greater heat loss t y p i c a l of fast-burning chambers, the exhaust temperature i s reduced so that unburned hydrocarbons leaving the combustion chamber are le s s l i k e l y to burn i n the engine exhaust system. Mattavi (1980, p. 2789) attr i b u t e s the unpredictable nature of unburned hydrocarbons to the more e f f i c i e n t combustion t y p i c a l of fast-burning chamber design. On one hand, increased e f f i c i e n c y r e s u l t s i n lower exhaust temperatures which reduces the oxidation rate of unburned hydrocarbons. This r e s u l t s i n increased unburned hydrocarbon concentration. On the other hand, increased e f f i c i e n c y reduces the required mass flow rate through the engine which may more than compensate for the increased concentration due to lower exhaust temperature. This may r e s u l t i n o v e r a l l decreased unburned hydrocarbon emissions on a mass per unit time or distance traveled. To permit use of lower octane unleaded fuel s , designers 4 reduced compression r a t i o s with the r e s u l t of longer burning duration and lower s p e c i f i c output. In an attempt to reduce oxides of nitrogen, designers i n the 1970*s implemented exhaust gas r e c i r c u l a t i o n which increased burning duration. Current a c t i v i t y i n fast-burning chamber design can be seen as an e f f o r t to ree s t a b l i s h the rates of combustion that once existed, while maintaining low emissions (Mattavi, 1980, p. 2783) . 1.1.3 Mass Burning Rate Before surveying current methods to promote f a s t burning, the mass entrainment rate equation w i l l be considered. Although mass burning rate strongly depends on mass entrainment rate, these are not the same (except under conditions l i k e a planar and constant-thickness-flame burning) . Mass entrainment rate i s defined as the rate of mass entering an advancing flame front while mass burning rate i s defined as the chemical reaction rate i n the flame. The mass entrainment rate equation does i l l u s t r a t e which parameters designers can modify to achieve greater mass burning rates. Applying the p r i n c i p l e of mass conservation to a propagating flame front (as shown i n Fi g . 1.3) leads to an expression for mass entrainment rate (Mattavi, 1980, p. 2793): m = 0 * A f * S b where; m = mixture mass entrainment rate, (kg/s) O = density of the unburned mixture, (kg/irr) 5 A f = flame front area, (m2) S b = burning v e l o c i t y r e l a t i v e to unburned gas, (m/s) The f i r s t term i n t h i s expression, ^ , i s determined by the a i r - f u e l r a t i o , exhaust gas r e c i r c u l a t i o n rate, intake manifold pressure, residual mass f r a c t i o n , and compression r a t i o . Since c- does not r e l a t e to chamber design except through compression r a t i o , the chamber designer can not a l t e r t h i s term extensively. The designer does have i n d i r e c t control over flame front area, A f. By changing spark location, c y l i n d e r head geometry, and mixture motion, the area of the propagating flame can be increased so that mass entrainment rate i s increased. The burning v e l o c i t y , S b, i s a strong function of mixture motion; the i n t e n s i t y of turbulence i s p a r t i c u l a r l y important. Damkohler (1947) postulated that turbulence wrinkles the propagating flame front thus providing increased surface area and increased mass entrainment rate. Higher port v e l o c i t y causes stronger turbulent mixture motion. But turbulence generated at the i n l e t port decays s u b s t a n t i a l l y before i g n i t i o n so that i t i s only p a r t i a l l y e f f e c t i v e during combustion (Nagayama, et a l . , 1977). One method of i n t e n s i f y i n g mixture turbulence j u s t before i g n i t i o n i s to provide a small clearance (one millimeter may be t y p i c a l ) between the cylinder head and the outer portion of the piston. This arrangement i s c a l l e d "squish". As the piston approaches top dead center on the compression stroke, the compressed mixture i n the small clearance zone i s forced J 6 r a d i a l l y inward. Such rapid inward motion increases the turbulence l e v e l around the i g n i t i o n source which encourages rapid flame propagation. Squish also helps to reduce engine knock since end gases are cooled more e a s i l y with the squish geometry than without. However, cooling of end gases may increase the l e v e l of unburned hydrocarbons. Spark lo c a t i o n a f f e c t s burning rate by decreasing flame t r a v e l distance. In the absence of s w i r l , burning duration i s shortest with central i g n i t i o n . However, with s w i r l , burning duration may be shortest for some off-center i g n i t i o n l o cation l i k e mid-radius where the l e v e l of turbulence (and thus flame burning veloc i t y ) may be higher. Chamber geometry a f f e c t s heat transfer since an increase i n the surface-to-volume r a t i o of a chamber w i l l increase the heat tr a n s f e r to and from the reactants. Ideally, the chamber should have the le a s t surface-to-volume r a t i o . Therefore, an open chamber (which i s a chamber that i s nearly spherical i n cross-sectional shape) with a central spark source i s more e f f i c i e n t than a wedge chamber (which has a cross-sectional shape l i k e a wedge) with a spark source located at one end of the chamber. 1.2 Swirling Combustion In Engines 1.2.1 Methods Of Swirl Generation F i g . 1.4 i l l u s t r a t e s some methods how s w i r l can be generated within a cylinder. Large s w i r l i n g rates (up to four revolutions per engine stroke) can be created i n the cylinder with a shrouded intake valve so that tangential motion i s 7 imparted to the incoming mixture. This method i s not used i n commercial engines because of the penalty of reduced volumetric e f f i c i e n c y due to reduced valve area and flow r e s t r i c t i o n . Shrouded valves are often incorporated i n research engines where reduced volumetric e f f i c i e n c y may be unimportant. The i n l e t port to the engine cylinder can be arranged to create s w i r l i n the chamber. A tangential i n l e t passage to the chamber d i r e c t s the inflowing mixture to flow with a sw i r l i n g pattern. Swirl can also be induced within the chamber by placing flow d i r e c t i n g vanes within the i n l e t passage d i r e c t l y upstream of the i n l e t valve. The e f f e c t of the vanes would be s i m i l a r to that of tangential port entry and shrouded valves i n that a nonradial i n j e c t i o n into the chamber creates a sw i r l i n g flow f i e l d . Swirl can also be created with s p e c i a l valve geometry and i n l e t port arrangements which, for example, place the swi r l axis of rot a t i o n perpendicular to that of the cylinder as shown i n F i g . 1.5. As the piston approaches top dead center, the s w i r l mean radius i s reduced and i n t e n s i f i e d because of the p r i n c i p l e of conservation of angular momentum. Chamber shape can have a swi r l promoting e f f e c t . A current example of t h i s i s the May " F i r e b a l l " c y l i n d e r head which has app l i c a t i o n i n some Jaguar automobile engines. Fig. 1.6 shows a cross section of t h i s cylinder head. Swirl i s introduced within the cylinder by tangential port entry. The 8 mixture i s compressed into a small volume with a short radius below the exhaust valve as the piston r i s e s to top dead center on the compression stroke. Swirl i s i n t e n s i f i e d because of the p r i n c i p l e of conservation of angular momentum. The squish e f f e c t i s also present and spark lo c a t i o n i n the May chamber has been optimized to promote fast burning. The engine operates successfully with a i r - f u e l r a t i o s of up to 20 to 1 and compression r a t i o s of up to 16 to 1. This engine demonstrates improved f u e l economy owing to the combination of fas t burning, high compression r a t i o , and lean f u e l - a i r r a t i o . 1.2.2 Turbulence Intensity With large s w i r l i n g motion, turbulence i s generated throughout the compression stroke by shear forces within the mixture. This r e s u l t s i n greater turbulence i n t e n s i t y at the time of i g n i t i o n which d i r e c t l y a f f e c t s the burning v e l o c i t y . 1.2.3 Buoyancy E f f e c t s In addition to the e f f e c t s on turbulence generation, s w i r l induces a ce n t r i f u g a l body force. For a t y p i c a l swirl number (which i s the r a t i o of swi r l rpm to engine rpm) of four and a mean radius of twenty f i v e millimeters, the ce n t r i f u g a l acceleration would be between 500 and 10,000 g for engine speeds of 1000 to 5000 rpm. Thus, an off-center flame kernel (which has one-sixth the density of unburned mixture) tends to be drawn inward to the swi r l axis which may increase the e f f e c t i v e flame area. I t has been suggested (Witze, 1982, p. 171) that s w i r l may also increase burning rate by stretching the flame kernel which remains attached to the spark source or 9 a part of the chamber. The r e s u l t i s increased flame area during the early stage of combustion. Along with the swi r l induced c e n t r i f u g a l force f i e l d , there i s a swi r l induced secondary f l u i d flow. In Fig . 1.7 t h i s c i r c u l a t i o n i s driven i n the boundary layer on the f l a t end walls by the strong r a d i a l pressure gradient, dP/dr, established by the main sw i r l (cal l e d primary f l u i d flow i n Fig . 1.7). In the boundary layer, the force r e s u l t i n g from the r a d i a l pressure gradient i s not balanced by the i n e r t i a l force, <?Ve2/r, due to slower (than mainstream) tangential v e l o c i t i e s . Thus, with s w i r l i n g flow i n a short c y l i n d r i c a l volume, mixture moves r a d i a l l y outward i n the mid-plane area between the two end walls and r a d i a l l y inward i n a t h i n zones along the end walls. The secondary flows vary i n i n t e n s i t y with the diameter-to-length r a t i o and s w i r l . 1.2.4 Combined E f f e c t s Of Swirl And Squish Swirl and squish seem to complement each other. Some researchers (Nagayama, et a l . , 1977, p. 998) found that squish and s w i r l i n combination increased burning rate and reduced c y c l i c v a r i a t i o n (which i s the v a r i a t i o n i n burning rate, burning duration, and peak pressure within a serie s of cycles) more than eith e r s w i r l or squish alone. They observed that s w i r l tends to reduce early burning duration and squish tends to reduce main burning duration. When incorporated together, s w i r l and squish are e f f e c t i v e methods of increasing burning rate i n in t e r n a l combustion engines. 1.3 Research Obi ectives The purpose of t h i s work was to examine the conditions under which s w i r l improves combustion of premixed f u e l i n a short c y l i n d r i c a l chamber. Since the burning duration of a sw i r l i n g mixture i s a strong function of spark location (as well as spark plug geometry and orientation) , spark location was also included i n t h i s study. The s p e c i f i c objectives were to determine the ef f e c t s of s w i r l and spark location on combustion rate and duration, and on c y c l i c v a r i a t i o n . The main questions of in t e r e s t are: A) Given a c e r t a i n spark location, does s w i r l always r e s u l t i n decreased burning duration? B) Given a c e r t a i n spark l o c a t i o n and s w i r l i n t e n s i t y ( i . e . , the average rpm of the mixture i n the chamber), how do spark gap and shape a f f e c t burning duration and c y c l i c v ariation? C) How do s w i r l , spark gap, spark-source shape, and spark l o c a t i o n decrease burning duration and c y c l i c variation? What i s the physical mechanism whereby the burning duration i s changed? D) F i n a l l y , i s i t possible to have too much swi r l or a spark gap that i s too wide? Is i t possible that burning duration and c y c l i c v a r i a t i o n may increase with excessive s w i r l and other parameters? This question b a s i c a l l y asks: what i s the ide a l s w i r l l e v e l , spark location, and spark gap? 11 2. LITERATURE REVIEW 2.1 Introduction The scope of t h i s review i s r e s t r i c t e d to sw i r l i n g combustion of premixed fuels i n various types of spark i g n i t i o n chambers. Though much work on s w i r l i n g combustion has been done on, for example, the s t r a t i f i e d charge engine or the d i e s e l engine, with the t y p i c a l deep bowl-in-piston geometry, t h i s i s not included i n t h i s review because i t i s beyond the scope of t h i s research project. 2.2 Swirling Combustion In Constant Volume Chambers 2.2.1 Dyer (1979) Dyer measured mean v e l o c i t y d i s t r i b u t i o n s (before and a f t e r combustion), turbulence, pressure, and flame p o s i t i o n i n a constant volume combustion chamber. The chamber walls were heated, and propane (or methane) f u e l - a i r mixture was injected through a shrouded poppet valve. Dyer showed that s i g n i f i c a n t s w i r l existed long a f t e r the intake valve was closed. Aft e r waiting one hundred milliseconds f o r the s w i r l r a d i a l v e l o c i t y d i s t r i b u t i o n to s t a b i l i z e , Dyer observed that the s w i r l decayed by approximately 10% within ten milliseconds (which i s the t y p i c a l length of time between i n l e t valve closure and i g n i t i o n of an engine turning 3000 rpm). When a spark plug was i n s t a l l e d , o v e r a l l s w i r l i n t e n s i t y was reduced by about 2 0% thus i n d i c a t i n g that there i s considerable drag induced by the plug. 12 The t a n g e n t i a l f l o w d i s t r i b u t i o n i n D y e r ' s c h a m b e r h a d a v e l o c i t y p e a k a p p r o x i m a t e l y a t m i d - r a d i u s . D y e r m e a s u r e d t h e a x i a l d i s t r i b u t i o n o f v e l o c i t y a t v a r i o u s r a d i i . The r e s u l t s show t h a t t h e t a n g e n t i a l v e l o c i t y v a r i e s l i t t l e a c r o s s t h e c h a m b e r p a r a l l e l t o t h e c h a m b e r a x i s i n d i c a t i n g t h a t s w i r l i n a n a r r o w c y l i n d e r i s a p p r o x i m a t e l y t w o - d i m e n s i o n a l . The r e a s o n f o r t h i s i s t h a t p r e s s u r e g r a d i e n t s i n t h e a x i a l d i r e c t i o n a r e n e c e s s a r i l y v e r y s m a l l , b e c a u s e a x i a l v e l o c i t i e s a r e much s m a l l e r t h a n t a n g e n t i a l . S i n c e t h e p r e s s u r e t e n d s t o b e a f u n c t i o n o f r a d i u s o n l y a n d s i n c e t h e r a d i a l p r e s s u r e g r a d i e n t , d P / d r , i s b a l a n c e d b y t h e c e n t r i f u g a l t e r m c^>V e 2/r, t h e t a n g e n t i a l v e l o c i t y a t a g i v e n r a d i u s t e n d s t o b e v e r y n e a r l y c o n s t a n t e x c e p t i n t h e t h i n r e g i o n s n e a r t h e w a l l . I n t h i s r e g i o n , s e c o n d a r y f l o w i s d o m i n a n t . D y e r d i d n o t m e a s u r e t h e t a n g e n t i a l v e l o c i t y a t t h e e n d w a l l a n d h i s r e s u l t s do n o t i n d i c a t e t h e s e c o n d a r y f l o w s t h a t m u s t h a v e e x i s t e d . D y e r o b s e r v e d t h e e f f e c t o f s w i r l i n t e n s i t y o n b u r n i n g d u r a t i o n b y i g n i t i n g t h e m i x t u r e ( a t t h e o u t e r w a l l ) a t v a r i o u s d e l a y t i m e s a f t e r i n j e c t i o n . W i t h t h e i n t r o d u c t i o n o f s w i r l ( w i t h a mean v e l o c i t y o f a p p r o x i m a t e l y t h i r t e e n m e t e r s p e r s e c o n d ) , b u r n i n g d u r a t i o n d e c r e a s e d b y more t h a n a f a c t o r o f t w o . W i t h t h e f l a m e p o s i t i o n p r o b e s , D y e r o b s e r v e d t h a t t h e f l a m e b u r n e d a c r o s s t h e c h a m b e r a t a c o n s t a n t v e l o c i t y w h i l e t h e mass b u r n i n g r a t e i n c r e a s e d w i t h f l a m e a r e a . 2.2.2 I n o u e , N a k a n i s h i , N o g u c h i , a n d I g u c h i (1980) I n o u e , e t a l . , s t u d i e d s w i r l i n g c o m b u s t i o n i n a d i s c s h a p e d , c o n s t a n t v o l u m e c h a m b e r , a s w e l l a s i n e n g i n e s . The 13 chamber was 120 mm i n in t e r n a l diameter and twenty seven millimeters between end walls. Swirl was generated by i n j e c t i o n of f u e l - a i r mixture into the previously evacuated chamber, eithe r i n the r a d i a l d i r e c t i o n (which resulted i n no swirl) or at t h i r t y degrees to the r a d i a l d i r e c t i o n . The mean sw i r l i n g v e l o c i t y d i s t r i b u t i o n was approximately solid-body type r o t a t i o n with the peak v e l o c i t y close to the outer wall. Turbulence i n t e n s i t i e s were observed to be considerably higher and to l a s t longer with the sw i r l i n g case than with the no-sw i r l case. Flame photography of the combustion zone i n the short c y l i n d r i c a l chamber displayed a tendency of the burning gases to be driven toward the center of the chamber with off-center i g n i t i o n . The photography also indicated a tendency of the flame kernel to remain attached to the spark source with the so-called flame holding e f f e c t . Inoue, et a l . , observed that burning duration depended on i g n i t i o n timing a f t e r the i n i t i a l j e t i n j e c t i o n ( i . e . , depended on swi r l i n t e n s i t y ) . They observed that, with s w i r l , burning duration with central i g n i t i o n was longer than with mid-radius l o c a t i o n and that burning duration was considerably shorter with s w i r l than without s w i r l . 2.2.3 Zawadzki and Jar o s i n s k i (1983) The c y l i n d r i c a l constant volume chamber of Zawadzki and Ja r o s i n s k i was seventy millimeters i n diameter and f i f t y millimeters i n length. I t had variable tangential v e l o c i t i e s between t h i r t y and ninety meters per second and a maximum root 14 mean square turbulent v e l o c i t y range of two to ten meters per second. Homogeneous mixtures of methane and a i r were burned at constant pressure. They concluded that i n t e n s i f y i n g the s w i r l increases the combustion duration because s w i r l caused the flame to become more laminar-like, and hence, propagate at a lower speed, e s p e c i a l l y with lean mixtures. 2.2.4 Hanson and Thomas (1984) Hanson and Thomas examined the e f f e c t of well s p e c i f i e d s w i r l i n g motions on flame development i n two constant volume c y l i n d r i c a l chambers. The chambers were ei t h e r 420 or 3 0 mm i n length, and both were 200 mm i n diameter. Both chambers were rotated about the c y l i n d r i c a l axis at speeds up to 750 rpm. Since the chambers were rotated, viscous drag was eliminated and thus there was no continuous turbulence generation. The r a d i a l d i s t r i b u t i o n was assumed to be of solid-body type rotation and s w i r l i n t e n s i t y was considered the same as bomb rotation speed. A spark probe was located mid-plane at d i f f e r e n t r a d i i . Data was c o l l e c t e d from pressure records and photography. Hanson and Thomas describe an e f f e c t (relevant to the large chamber) which they c a l l " p e n c i l l i n g " . They said that "flames developing on the axis elongate along the axis and tend toward the c y l i n d r i c a l i n shape with greater surface area than i f they remained spherical, and thus explosions are accelerated. The e f f e c t i s greater at higher r o t a t i o n a l speeds and with more slowly burning mixtures [ i . e lean mixtures] ." (1984, p. 278) In the case of the disk shaped 15 chamber, the a x i a l elongation of the c e n t r a l l y i g n i t e d flames resulted i n early quenching by the vessel end walls. In t h i s case, higher r o t a t i o n a l speeds resulted i n slower burning rates. But f o r off-center i g n i t i o n , higher r o t a t i o n a l speeds s t i l l accelerated the combustion by avoiding quenching at the cylin d e r walls. Hanson and Thomas observed that shear flow (obtained by having a counter rot a t i n g concentric c y l i n d e r within the longer chamber) resulted i n faster burning than quiescent mixtures. Lean mixtures had greater increases i n burning rate with increased shear than chemically balanced mixtures. In considering the relevance of t h i s r o t a t i n g chamber study to combustion i n engines, i t i s important to note that i n the r o t a t i n g cylinder, secondary flows (which can be strong i n short cylinders with stationary walls) would be absent. Secondary flow, generated as a r e s u l t of end wall shear stress, would oppose the " p e n c i l l i n g " motion along the central axis. 2.3 Swirling Combustion In Single Cylinder Research Engines 2.3.1 Wakuri, Kido, Ono, Nakashima, and Murase (1981) Wakuri and colleagues studied the e f f e c t of s w i r l , spark loc a t i o n , and engine speed on burning v e l o c i t y . A uniflow scavenging-type engine running at 3 00 and 600 rpm with v a r i a b l e d i r e c t i o n vanes on the scavenging ports was used throughout the experiment. Swirl i n t e n s i t y was defined as the sine of the angle between the tangential i n l e t vane and the cy l i n d e r radius. The spark locations were r/R = 0 (center), 16 0.5 (mid-radius), or 0.75 (outer). Pressure data, photography, and hot wire measurements were used to observe the e f f e c t s of the variables. The conclusions of Wakuri, et a l . , are that central spark l o c a t i o n y i e l d s the shortest burning duration, that burning duration decreases with increased engine rpm, and that i n general, s w i r l i n t e n s i t y has l i t t l e e f f e c t on the burning duration. They observed, with photography, a strong tendency of flame holding at the plug with peripheral i g n i t i o n and also of inflow of hot gases toward the center owing to buoyancy forces. They observed that the mixture turbulent k i n e t i c energy appeared to decline much less r a p i d l y with s w i r l than with no s w i r l . 2.3.2 Witze and V i l c h i s (1981) Witze and V i l c h i s investigated the e f f e c t s of s w i r l on burning rate and c y c l i c v a r i a t i o n s i n a L-head research engine which had a cylinder-wall-mounted shrouded intake valve, an outer wall spark location, and a compression r a t i o of 5.4 : 1. They defined t h e i r s w i r l v a r i able as the cosine of the angle of the shrouded valve opening to the horizontal plane. For lean mixtures, the 90% burning duration was reduced by more than a factor of two when the shrouded valve was rotated from no-tangential i n j e c t i o n to t o t a l - t a n g e n t i a l i n j e c t i o n . There was l i t t l e reduction of 90% burning duration for the r i c h mixture case. Witze and V i l c h i s observed reduced c y c l i c v a r i a t i o n for shroud angles of seventy f i v e and zero degrees ( i . e . , low and 17 high swirl) and that the percent reduction was greater for the lean than the r i c h mixture case. They suggested that r e l a t i v e l y large c y c l i c v a r i a t i o n s at medium s w i r l l e v e l s may have been due to the random detachment of the flame from the i g n i t i o n source. In these experiments, r i c h mixture burning rate and c y c l i c v a r i a t i o n improved l i t t l e with increased s w i r l . However, lean mixture burning rate and c y c l i c v a r i a t i o n c o nsistently improved with sw i r l as long as there was no flame detachment from the i g n i t i o n source. 2.3.3 Witze (1982) While continuing research on the e f f e c t s of s w i r l i n the same engine, Witze focused on the in t e r a c t i o n of spark plug type, spark location, and s w i r l . Since the v e l o c i t y d i s t r i b u t i o n was nonlinear, Witze defined s w i r l number as the r a t i o of the equivalent solid-body angular v e l o c i t y i n the cyli n d e r to the engine speed. The central spark location was most e f f e c t i v e f or the reduction of burning duration except at the highest sw i r l where outside-wall i g n i t i o n was better than c e n t r a l . This was p a r t i c u l a r l y evident for the lean mixture at the maximum swi r l l e v e l . The ide a l s w i r l l e v e l seemed to be at S = 3.2 where the burning duration was minimum for both the lean and chemically balanced correct mixtures. At t h i s s w i r l l e v e l the ide a l spark lo c a t i o n was ce n t r a l . When there was no s w i r l , the burning duration was a function only of flame t r a v e l distance so that central spark location yielded shortest 18 durations. The c y c l i c v a r i a t i o n s at S = 3.2 were also less than at S = 0. The t e s t i n g of d i f f e r e n t spark plug types ( i . e . , protruding or surface gap) indicated that flame holder e f f e c t s were important at higher sw i r l l e v e l s . In a l l cases, the protruding spark plug gave shorter burning durations than the surface gap plug, e s p e c i a l l y at high sw i r l numbers. Witze took laser shadowgraphs of the s w i r l i n g combustion with peripheral i g n i t i o n i n h i s engine and showed that when ign i t e d away from the cylinder axis, the flame was held by the spark plug such that burning was dominated by the swi r l i n g convection of the unburned mixture into the s t a b i l i z e d reaction zone. He observed that the fast e s t burning was associated with highest sw i r l and that the flame remained attached to the spark plug (which was located at the outer radius). Under c e r t a i n conditions random detachment of the flame was observed; t h i s was associated with large c y c l i c v a r i a t i o n s . 2.3.4 Groff and Sinnamon (1982) Groff and Sinnamon studied the e f f e c t of spark location i n a s w i r l i n g f i e l d on combustion duration, thermal e f f i c i e n c y , c y c l i c v a r i a b i l i t y , and knock s e n s i t i v i t y . They used two engines; engine A was of Bowditch design where a single valve i s used for the i n l e t and exhaust strokes and the mixture was ign i t e d every eight strokes. This arrangement allowed o p t i c a l access to h a l f of the combustion chamber but resulted i n a rather high percentage of exhaust residuals 19 (between 11 and 20% at 1000 to 1900 rpm) . A shrouded valve was incorporated to increase the s w i r l number (defined as the r a t i o of mixture s w i r l to the engine angular veloc i t y ) from four to seven. Five r a d i a l spark locations were tested. Engine B was a conventional engine of b a s i c a l l y the same dimensions and chamber shape as engine A. I t was operated at 1900 rpm, with two spark locations, and at a s w i r l number of four. Both engines used propane at various equivalence r a t i o s . Engine B only ran on chemically balanced mixtures but on gasoline i n addition to propane for knock s e n s i t i v i t y t e s t s . Groff and Sinnamon conclude from t h e i r experiments that central i g n i t i o n produces shorter burning durations than outer wall i g n i t i o n . The exception was for the high sw i r l case with very lean mixtures (down to 0.65 of the chemically balanced r a t i o ) . But the high-swirl case they found reduced thermodynamic e f f i c i e n c y . They also observed that, for lean mixtures and peripheral i g n i t i o n location, avoiding flame detachment from the i g n i t i o n source was d i f f i c u l t . With a s w i r l number of four, c y c l i c v a r i a t i o n and exhaust emissions were i n s e n s i t i v e to spark location. They observed flame holding and buoyancy e f f e c t s from photographic records. They conclude that neither of these e f f e c t s reduced the burning duration. 2.4 Swirling Combustion In Mult i c v l i n d e r Engines 2.4.1 Mayo (1975) Mayo used a production Ford 351C engine as the base 20 engine and modified the f i r s t and eighth cylinders to provide a l l combinations of the following: 1) Squish (which i s the reduced clearance between the piston and the outer part of the cylind e r head r e s u l t i n g i n r a d i a l inflow and increased mixture turbulence at top dead center). He tested the normal production combustion chamber, an open chamber (zero squish), and a chamber with a 3 0% squish area. 2) Spark plug p o s i t i o n . Mayo tested the e f f e c t of the spark plug location, moving i t by means of an extended length spark plug from the production chamber lo c a t i o n to a po s i t i o n 17 mm clo s e r to the bore centerline. 3) Swirl. The e f f e c t s of the r e l a t i v e l y low s w i r l case of the production engine and 5,000 rpm s w i r l created by a shrouded valve were compared. 4) Charge v e l o c i t y . Mayo investigated the e f f e c t of increasing the charging v e l o c i t y using a high v e l o c i t y port. Mayo measured the burning duration (from pressure data) of 17 d i f f e r e n t configurations. For comparison, burning duration was measured on a fast-burning 4 60 engine. Mayo found that although s i g n i f i c a n t reduction i n burning duration could be achieved through various configurations of the 351C engine, the burning duration was always longer than i n the 460 fast-burning engine. The 460 engine was a highly-turbulent, fast-burning engine with approximately 75% piston squish, 8400 rpm sw i r l , 21 central spark plug location, and a bowl-in-piston combustion chamber. The compression r a t i o was comparable to that of the 351C engine. Mayo found that the burning duration decreased by 24% when s w i r l was increased from 0 to 5000 rpm and by 20% when the i g n i t i o n point was moved from the normal l o c a t i o n i n the production engine to the central l o c a t i o n . I t should be noted that when these and other features were combined, the r e s u l t i n g burning duration reduction was not the sum of the i n d i v i d u a l contributions. C y c l i c v a r i a t i o n (defined as the r a t i o of the standard deviation of the peak pressure to the mean peak pressure), was generally reduced with reduced burning duration. The greatest reduction i n c y c l i c v a r i a t i o n , when using a s w i r l i n g flow, was with off-center i g n i t i o n . A comparable reduction occurred with squish, s w i r l , and high port v e l o c i t y . Central i g n i t i o n and high port v e l o c i t y seemed to have l i t t l e e f f e c t i n percent reduction of c y c l i c v a r i a t i o n when used i n an open chamber, high-swirl engine. But the l e a s t amount of c y c l i c v a r i a t i o n occurred when squish, s w i r l , and central i g n i t i o n were used together. Mayo found that with a decrease i n the 10-9 0% burning duration (the so-called main burning duration) from t h i r t y to f i f t e e n crank angle degrees, nitrogen oxide production increased by approximately 30% and the unburned hydrocarbon emissions by approximately 250%. 22 2.4.2 Nagayama, Araki, and Iioka (1977) The objective of Nagayama, and co-workers was to study the e f f e c t of sw i r l and squish on burning v e l o c i t y , c y c l i c v a r i a t i o n , lean l i m i t , emissions, f u e l economy, and d r i v e a b i l i t y . They modified the i n l e t port angle of a four cylinder, 1.4 l i t e r engine to increase s w i r l , and the head geometry to provide squish. They measured s w i r l i n t e n s i t y with a paddle wheel inside a pistonless cyli n d e r with an intake flow rate equivalent to average intake flow rate during operation. The burning duration was longest for the standard cy l i n d e r head and shortest for the combination s w i r l and squish head with various mixture strengths. The e f f e c t of sw i r l i n t e n s i t y was dominant i n the early burning duration. Nagayama, et a l . , found the e f f e c t of squish i n t e n s i t y was greater than that of s w i r l i n t e n s i t y during the main burning duration. With unusually lean mixtures ( a i r - f u e l r a t i o s greater than 17.5), the main burning duration with the swi r l head ra p i d l y decreased, i n d i c a t i n g that for lean mixtures, i t was better to have s w i r l than squish. The early and main burning durations increased with increased a i r - f u e l r a t i o . Nagayama, et a l . , defined c y c l i c v a r i a t i o n as the r a t i o of the standard deviation to the mean indicated mean e f f e c t i v e pressure. The r a t i o was largest for the standard head and for lean mixtures. Swirl seemed more e f f e c t i v e at reduction of c y c l i c v a r i a t i o n than squish, but not as e f f e c t i v e as a s w i r l -squish combination. 23 Although the swirl-squish head had the shortest burning durations, t h i s head also had the highest nitrogen oxide emissions at a l l a i r - f u e l r a t i o s . Comparing the d i f f e r e n t c y l i n d e r head types, the s w i r l head produced s l i g h t l y less nitrogen oxide than the squish head for chemically balanced mixtures and the reverse was true for lean mixtures. Unburned hydrocarbon production seemed to depend on the m i s f i r e l i m i t of each head. The standard head had the lowest lean m i s f i r e l i m i t and the swirl-squish head had the highest r e s u l t i n g i n unburned hydrocarbon production highest for the standard head and lowest for the swirl-squish head. I t appeared that swirl provided a higher m i s f i r e l i m i t than squish and thus produced les s unburned hydrocarbon at a given a i r - f u e l r a t i o . 2.4.3 Nagao and Tanaka (1983) The objective of Nagao and Tanaka was to define an optimum s w i r l i n t e n s i t y considering combustion duration, c y c l i c v a r i a t i o n , nitrogen oxide emission (with variable exhaust gas r e c i r c u l a t i o n ) , volumetric e f f i c i e n c y , combustion noise, combustion s t a b i l i t y , and thermal e f f i c i e n c y . Their apparatus was a four cylinder, 1.5 l i t e r engine with three d i f f e r e n t heads; a hemispherical head, a bathtub head, and a high compression head. For s w i r l generation they used shrouded valves and what the authors c a l l a dual induction system, which consisted of two i n l e t passages to each i n l e t port of the t e s t chambers. The f i r s t passage was tangential to the c y l i n d e r while the other was r a d i a l . By varying the r a t i o of flows between the 24 two passages, s w i r l i n t e n s i t y could be varied during engine operation. Swirl was measured on a t e s t r i g with a paddle wheel i n steady state flow. Swirl was also measured with a hotwire anemometer i n the engine. Flow v i s u a l i z a t i o n was achieved with an o i l f i l m on top of the piston which generated streak l i n e s . Nagao and Tanaka found that as s w i r l number increased, burning duration decreased, and turbulence i n t e n s i t y at top dead center of the compression stroke increased. They also found that the optimal s w i r l l e v e l depended on the objective, e.g., whether i t be minimum unburned hydrocarbon and nitrogen oxide emission, extended lean l i m i t , extended exhaust gas r e c i r c u l a t i o n l i m i t , reduced brake s p e c i f i c f u e l consumption, optimal amount of cooling, or reduced i g n i t i o n advancement. Nagao and Tanaka present a concept of a variable s w i r l engine where at any given operating condition the engine optimizes s w i r l to achieve the desired r e s u l t . In summary, Nagao and Tanaka conclude that " i n t e n s i f i e d s w i r l brings with i t improvement i n combustion s t a b i l i t y as represented by lean l i m i t and exhaust gas r e c i r c u l a t i o n l i m i t , and knock resistance, as well as better brake s p e c i f i c fuel consumption. But such b e n e f i c i a l e f f e c t s do not increase i n proportion to increasing s w i r l i n t e n s i t y because of a steep r i s e i n cooling losses; there i s an optimum s w i r l i n t e n s i t y at which maximum e f f e c t s are yielded."(1983, p. 156) 25 Comparison of Observations 2.5 Swirling V e l o c i t y D i s t r i b u t i o n s Dyer (1979) observed b a s i c a l l y the same type of v e l o c i t y d i s t r i b u t i o n as Witze and V i l c h i s (1981) which was that swirl was of nearly solid-body type configuration. The v e l o c i t y peak was at about r/R = 0.5 i n Dyer's case, and at r/R = 0.8 i n case of Witze and V i l c h i s . Inoue, et a l . (1980) , also observed solid-body type r o t a t i o n with the v e l o c i t y peak near the wall i n t h e i r constant volume chamber study. Witze and V i l c h i s did not mention the p o s s i b i l i t y of having the r o t a t i o n a l axis of s w i r l perpendicular to axis of the c y l i n d e r as shown i n F i g . 1.4. Their shrouded valve i n the. no-swirl configuration may have generated s w i r l as just described. Such sw i r l may be i n t e n s i f i e d with the p r i n c i p l e of conservation of angular momentum since the mean rotation radius i s reduced by the r i s i n g piston. Thus, t h e i r r e s u l t s for the no-swirl case may not be i n d i c a t i v e of actual engine experience since there may be s i g n i f i c a n t bulk mixture motion e f f e c t s on burning duration. There was some in d i c a t i o n of t h i s trend i n t h e i r paper since they remarked that the turbulence i n t e n s i t y for the no-swirl case was greater than for the s w i r l i n g cases. Nagayama, et a l . (1977), observed approximately s o l i d body r o t a t i o n and noticed that the rate of r o t a t i o n at top dead center on the compression stroke had dropped to one-sixth of the i n i t i a l r o t a t i o n a l v e l o c i t y during the intake stroke. 26 However, Dyer (1979) has shown that s w i r l i n h i s constant volume chamber decayed by about 10% i n ten milliseconds which was le s s than that observed by Nagayama, et a l . F i n a l l y , researchers frequently observed that the sw i r l axis of rot a t i o n seldom coincided with the cylind e r axis. 2.6 E f f e c t Of Swirl On Burning Duration In general, most c i t e d authors conclude that swirl reduces combustion duration. Nagao, et a l . (1983), observed that s w i r l decreased burning duration because of turbulence i n t e n s i t y . Mayo observed a 24% reduction i n burning duration when s w i r l was increased from the production l e v e l ( i . e . , low swirl) to 5000 rpm. Hanson and Thomas (1984) had the same observations; the main burning duration was decreased. Hanson and Thomas observed that "as the bomb was rotated f a s t e r there was l i t t l e e f f e c t over the early stages of combustion, but divergences then developed and combustion duration was shortened i n the faster rotating mixtures."(1984) Their theory was that the flame kernel during the early burning period was too small for ce n t r i p e t a l e f f e c t s . But as the kernel grew, i t was driven to the axis of s w i r l r o t a t i o n where i t " p e n c i l l e d " , creating a large flame area. Subsequently, i t burned r a p i d l y outward. Some authors conclude that s w i r l i s dominant mostly i n the early combustion period. With increased s w i r l i n most chamber types tested by Mayo (1977), the early burning duration was reduced. Nagayama, et a l . (1979), observed that while squish was e f f e c t i v e f or the main burning duration, s w i r l was dominant during early burning period. They also observed that when there was a lean mixture, the e f f e c t s of squish were lessened and that s w i r l was more e f f e c t i v e to decreasing burning duration. Nagayama, et a l . (1977), said that s w i r l with r i c h or chemically balanced mixtures was not very e f f e c t i v e for decreasing main burning duration, but only the early burning stage. Wakuri, et a l . (1981), observed that s w i r l had l i t t l e , i f any, e f f e c t on burning duration. Witze and V i l c h i s (1981) said that s w i r l reduced the main burning duration as well as the early burning duration for lean mixtures though s w i r l had l i t t l e e f f e c t f o r chemically balanced or r i c h mixtures. Wakuri, et a l . (1981), agreed but emphasized that spark l o c a t i o n was also important. Nagayama, et a l , (1977) said s w i r l and very lean mixtures reduced the main burning duration. Zawadzki and J a r o s i n s k i (1983) contradict the findings of the other authors when they conclude that s w i r l has l i t t l e e f f e c t on burning duration. B a s i c a l l y , they say, that the turbulent burning v e l o c i t y approaches or only s l i g h t l y exceeds the value of the laminar burning v e l o c i t y . Wakuri, et a l , (1981) also conclude that s w i r l has l i t t l e e f f e c t on burning duration. 2.7 Spark Location For t h e i r rotating disk chamber studies, Hanson and Thomas (1984) observed that, with central i g n i t i o n and increased r o t a t i o n rate, peak pressure declined. But the pressure rose fa s t e r and to a higher value when the spark plug 28 was located off-center and the chamber rotated f a s t e s t . The reasons appeared to be that with increased chamber r o t a t i o n a l speed and off-center i g n i t i o n , the low-density kernel was centered more vigorously by c e n t r i f u g a l force and suffered l e s s quenching e f f e c t s by the cool walls. For c e n t r a l l y i g n i t e d flames, the e f f e c t of the c e n t r i f u g a l force was to stre t c h the flame along the axis of ro t a t i o n and quench the flame by the end walls. Groff and Sinnamon (1982) d i f f e r e d with the conclusion of Inoue, et a l . (1980), who said that, at s w i r l numbers of three, the outer wall i g n i t i o n l o c ation was associated with shorter burning duration than the central l o c a t i o n . Groff and Sinnamon concluded that: f o r a s w i r l number of four no s i g n i f i c a n t differences with i g n i t i o n l o c a t i o n were measured i n thermal e f f i c i e n c y , heat transfer losses, exhaust temperature, c y c l i c v a r i a b i l i t y , or exhaust emissions. But with peripheral i g n i t i o n , the engine had greater knock s e n s i t i v i t y , lower knock l i m i t e d peak pressure and had a m i s f i r e problem. For a sw i r l number of seven, an increase i n heat transfer losses and a corresponding decrease i n thermal e f f i c i e n c y were measured for peripheral i g n i t i o n . Thus with intake generated s w i r l , c e n t r al i g n i t i o n was found to be desirable i n the two engines tested. (1982, p. 3838) Mayo (1975) observed a 20% reduction i n burning duration when the spark location was central with s w i r l i n g flow as compared to the production off-center l o c a t i o n with l i t t l e s w i r l . Quoting from Smith (1979) and from h i s own observations, Witze (1982) concludes that central i g n i t i o n i s desirable f o r the no-swirl and a l l s w i r l l e v e l s but 8.3. For sw i r l number of 8.3, outside-wall-ignition l o c a t i o n with a 29 properly designed spark plug give the fastest combustion rate. For such swirl intensities, the flame attaches to the spark plug or even to the cold wall. At a lower swirl number of 3.2, flame holding s t i l l takes place, but, not as well for lean mixtures. At swirl numbers higher than 3.2, the flame burning rate appears to be reduced. The reason, Witze submits, i s that at these swirl numbers, there are aerodynamic i n s t a b i l i t i e s which hamper the flame holder effects of the spark plug. He also found that the use of two spark plugs (located at opposite half-radius points) gave l i t t l e decrease in the burning duration when compared to one optimally located plug. He concludes that central ignition results in shortest burning duration except for high-swirl numbers where outside wall spark location i s fastest. Wakuri, et a l . (1981), say that the fastest burning spark location was the central position for their test conditions. 2.8 Flame Adhesion Wakuri, et a l . , (1981) observed a relationship between swirl, spark location, and flame adhesion. With increased swirl, the flame area was increased as long as the flame remained attached to the plug. This decreased the length of the average flame propagation path. Witze and Vilchis (1981) suggested that due to convection of the flame when the spark plug acted as a flame holder, short burning durations were observed. With lean mixtures, they observed the same trends but not the same burning rates as chemically balanced or rich mixtures. With random detachment, though, cyclic variation 30 increased. Witze (1982) found that flame holder e f f e c t s were important at high s w i r l i n t e n s i t i e s . In a l l cases, the protruding spark plug gave shorter burning durations than the surface gap plug, e s p e c i a l l y at high s w i r l l e v e l s . Dyer (1979) and Inoue, et a l . (1980), display photos i l l u s t r a t i n g flame holding with buoyancy e f f e c t s but they do not comment on the e f f e c t s of flame holding. Groff and Sinnamon (1982, p. 3854) report that "with peripheral i g n i t i o n , flame holding at the spark electrodes was consistently observed at a l l engine operating conditions", though they did not f i n d that off-center i g n i t i o n provided a net gain i n combustion rate. Flame adhesion was d i f f i c u l t and i r r e g u l a r with peripheral i g n i t i o n l o c ation and lean mixtures. 2.9 C y c l i c V a r i a t i o n Witze and V i l c h i s (1981) observed that c y c l i c v a r i a t i o n was strongly dependent on the detachment of the early flame from the spark plug. They observed low c y c l i c v a r i a t i o n with low- and high-swirl i n t e n s i t i e s . They conclude that for the medium-swirl range, there must be some aerodynamic i n s t a b i l i t i e s that caused the flame detachment. Groff and Sinnamon (1982) said that c y c l i c v a r i a t i o n was unaffected by spark l o c a t i o n for a s w i r l number of four. Mayo (1975) observed that a reduction i n burning duration generally resulted i n a reduction i n c y c l i c v a r i a t i o n . He had the least c y c l i c v a r i a t i o n when a l l the fast-burning techniques were combined into one chamber. Witze (1982) contradicts Mayo when he says that standard deviation increases with decreased 31 burning duration. Nagayama, et a l . (1977), observed that s w i r l was more e f f e c t i v e i n reducing c y c l i c v a r i a t i o n than squish. They observed that for a given l e v e l of c y c l i c v a r i a t i o n and a i r -f u e l r a t i o with the standard cylinder head, the s w i r l head tolerated increased a i r - f u e l r a t i o more than the squish head. The reason c y c l i c v a r i a t i o n i s reduced say Nagao and Tanaka (1983) i s that the turbulent scale decays more slowly with increased s w i r l , thus implying that the scale of turbulence i s the important factor i n c y c l i c v a r i a t i o n . C y c l i c v a r i a t i o n during medium s w i r l was greatest i f the flame detached from the spark plug. 2.10 Lean Limit The researchers that investigated emissions agreed that s w i r l helped extend the lean l i m i t . Nagayama, et a l . (1977), observed that s w i r l was very e f f e c t i v e at extending the lean l i m i t by 1.0 a i r - f u e l , i . e . , from 14-1 to 15-1. Nagao and Tanaka (1983) conclude that since s w i r l reduces the 10-90% burning duration, the lean l i m i t i s improved. Swirl, i n t h e i r research engine extended the maximum a i r - f u e l r a t i o from 19-1 to 22-1. They stated that the exhaust gas r e c i r c u l a t i o n l i m i t was also extended, but too much swi r l reduced both the lean and exhaust gas r e c i r c u l a t i o n l i m i t s . 2.11 Volumetric E f f i c i e n c y Nagayama, et a l . (1977), observed that with increased s w i r l , volumetric e f f i c i e n c y decreased. Nagao and Tanaka (1983) presented two reasons why t h i s may be the case: 32 1) i n general, current s w i r l generation techniques reduce flow area and create flow r e s t r i c t i o n s and, 2) s w i r l creates turbulence due to the viscous shear action at the cylinder walls. This increases heat transfer from the c y l i n d e r walls and reduces charge density. During and a f t e r combustion, the hot turbulent gases give up more heat energy to the walls so that both mechanical and volumetric e f f i c i e n c y s u f f e r . Since most current swirl inducing methods strongly reduce volumetric e f f i c i e n c y , more research i s needed on finding other methods of generating and maintaining s w i r l which r e s u l t i n less reduction on volumetric e f f i c i e n c y . 2.12 Emissions In t h e i r research, Nagayama, et a l . (1977), observed that the nitrogen oxide emissions of t h e i r t e s t engine increased with s w i r l . According to Mayo (1975), quoting from Blumberg and Kummer (1971), there e x i s t s t h e o r e t i c a l l y , an optimum burning duration for minimum nitrogen oxide emissions at a given engine operating condition. Thus, any f a s t e r or slower burning w i l l r e s u l t i n increased emissions. The mechanism of the optimal burning rate for l e a s t nitrogen oxide emissions i s that, as burning duration i s reduced, peak cycle temperature increases and duration of exposure decreases. Shorter time exposure at a given temperature reduces nitrogen oxide emissions, but higher temperature increases nitrogen oxide emissions and v i c e versa. Nagao and Tanaka (1983) show emission of nitrogen oxides 33 and even possibly of unburned hydrocarbons may increase somewhat with higher s w i r l i n t h e i r multi-cylinder engine. They show that though the nitrogen oxide emissions have gone up, there i s substantial improvement i n the exhaust gas r e c i r c u l a t i o n l i m i t from perhaps 15% to 30% with increase i n s w i r l i n t e n s i t y . Mayo (1975) observed that, when the main burning duration decreased from t h i r t y to f i f t e e n crank angle degrees, oxides of nitrogen increased by about 30% and unburned hydrocarbons increased by 250%. Groff and Sinnamon (1982) found l i t t l e e f f e c t of spark l o c a t i o n on emissions. The exception was f o r emissions of carbon monoxide which were nearly doubled for peripheral i g n i t i o n than for the central l o c a t i o n . 2.13 Heat Transfer Dyer (1979) observed that having an extended t i p spark plug at the outer wall increased the heat tr a n s f e r losses because of the increased turbulence generated by the spark plug and s w i r l . Thus probe shape, s i z e , and l o c a t i o n i s important on heat transfer. Wakuri, et a l . , (1981) appeared to contradict t h i s observation i n s t a t i n g that spark location had nothing to do with heat transfer losses. They did observe, though, that heat transfer losses increased with increased s w i r l . 2.14 Summary In summary, s w i r l i n an engine has been generally observed to be of solid-body type. Researchers do not agree about the amount of decay of s w i r l at the time of i g n i t i o n . One engine study concludes that s w i r l decays to one-sixth the o r i g i n a l value at the time of i g n i t i o n while a constant volume chamber study concludes that s w i r l decays only 10% i n ten milliseconds (which i s the time for a compression stroke of an engine turning at 3000 rpm). A l l authors but one conclude that s w i r l decreases burning duration. There are differences of opinion on whether sw i r l has i t s greatest e f f e c t on early, or on main, burning duration. Although outside spark location under c e r t a i n swirl conditions may lead to improved combustion, most authors that studied spark location conclude that central l o c a t i o n i s best. Those researchers that photographed the combustion process observed that the early flame remained attached to the spark source. Some correlated c y c l i c v a r i a t i o n with the occasional detachment of the early flame from the probe, but most do not observe s i g n i f i c a n t combustion improvement with flame attachment. Swirl seems to extend the lean flammability l i m i t ; i t s e f f e c t on emissions i s not well known. Swirl seems to decrease volumetric e f f i c i e n c y by increasing heat transfer. Of course volumetric e f f i c i e n c y also suffers with reduced area i n l e t ports and valves which are common with current swirl generating methods. This l i t e r a t u r e review indicates that, though much work has been done on the e f f e c t s of s w i r l on various combustion 35 c h a r a c t e r i s t i c s , there i s need for further research. Opinions vary on what i s the e f f e c t of s w i r l on spark location, emissions, early and main burning duration, c y c l i c v a r i a t i o n , and on other factors. Some of the differences i n conclusions are due to differences of constant volume and engine combustion. Other differences may be due to use of nonproduction engines with unusual design c h a r a c t e r i s t i c s l i k e the Bowditch design of Groff and Sinnamon (1982), the uniflow scavenging design of Wakuri, et a l . (1981), and the side-mounted shrouded valve design of Witze (1981, 1982). 36 3. APPARATUS 3.1 Introduction Swirl i s generated by tangential i n j e c t i o n of a high-speed j e t of f u e l - a i r mixture into a c y l i n d r i c a l chamber through one side of a T-shaped valve as shown i n F i g . 3.1. In steady state, a c i r c u l a t i n g fan supplies mixture to one side of the T-shaped valve and withdraws mixture from the chamber through the other side. The valve i s closed suddenly just before combustion by a s t i f f valve spring released by a solenoid. Combustion i s i n i t i a t e d by a v a r i a b l e location spark probe. The flow c i r c u i t i s sealed from the atmosphere while i n operation. During steady state operation before f i r i n g , the f u e l - a i r mixture i s constantly c i r c u l a t e d through the chamber and fan to ensure homogeneity of the mixture and to allow time for sw i r l measurement. The apparatus s p e c i f i c a t i o n s are l i s t e d i n Appendix D. 3.2 Combustion Chamber A steady precombustion sw i r l chamber design was chosen. An a l t e r n a t i v e arrangement of generating s w i r l within a short c y l i n d e r i s by tangential i n j e c t i o n into a previously evacuated chamber as used by Dyer (1979) . One problem with t h i s a l t e r n a t i v e i s that the s w i r l i n g f i e l d i s not steady and easy to measure. Another problem i s that the mixture temperature r i s e due to the work of compression of the mixture as i t enters the evacuated chamber. This can complicate the s w i r l measurement process; p a r t i c u l a r l y with hot wire 37 anemometry. The combustion chamber, as shown in Fig. 3.2, i s one hundred millimeters in diameter and f i f t y millimeters in length. Machined from a block of stainless steel alloy, the chamber was designed to withstand an internal pressure of ten MPa. The maximum expected working pressure was approximately one MPa with a precombustion pressure of one hundred kPa. End plates for the chamber are made of either clear polycarbonate plastic or optical quality quartz glass. These provide an optical path through the chamber for photographic work. Polycarbonate was chosen for ease of machining a spark probe mounting. Furthermore, polycarbonate i s extremely tough (with typical applications including bullet proof windows and glazing in high crime areas) so that in the event of excessive pressure in the combustion chamber, the windows would deform and tear instead of turning into dangerous, high-speed shrapnel. In addition to these end plates, polycarbonate side plates are installed for observation of flame propagation parallel with the cylindrical axis. 3.3 Valves The valve mechanisms shown in Figs. 3.3 and 3.4 consist of valve block, T-shaped valve, end plates (not shown), seals, and bushings. The valve blocks are interchangeable to f a c i l i t a t e various chamber configurations and machining. The end plates serve as valve guides and prevent inflowing mixture from bypassing the T-shaped valve. The valves are made of 4140 steel for maximum strength. 38 The valve l i f t i s adjustable to any height under seven millimeters; for these experiments i t was set at f i v e millimeters. The valve i s attached to the p u l l r o d by a machined thread. The pullrod, i n turn, passes outside the main chamber body through a bushing and 0-ring seal to a valve spring under compression. When released by a solenoid pin, the p u l l r o d and the T-shaped valve are pulled shut by the valve spring, c l o s i n g the combustion chamber from the piping system. A f t e r a set delay time, the s w i r l i n g mixture i s i g n i t e d and observed. A no-swirl valve i s shown i n F i g . 3.4. This valve was made so that experiments (which have not yet been performed) could be made on combustion i n the same chamber with approximately the same rate of turbulence generation at the i n l e t but with no s w i r l . Again, the valve i s T-shaped but the flow passages to and from the chamber are curved so that the flow i s r a d i a l instead of tangential. The distance between the edge of the T-shaped valve and the curved edge of the valve block i s f i v e millimeters so that approximately the same j e t v e l o c i t y , turbulence i n t e n s i t y , and turbulence scale can be generated as with the s w i r l valve. 3.4 Flow C i r c u i t F i g . 3.5 shows the explosion proof fan which c i r c u l a t e s the f u e l - a i r mixture through the combustion chamber. The rotor and motor were obtained from a i n d u s t r i a l vacuum cleaner whose s p e c i f i c a t i o n s are given i n Appendix D. The vacuum cleaner could have provided the desired flow c h a r a c t e r i s t i c s 39 but would not have been able to withstand substantial i n t e r n a l pressure from system charging or from possible explosions. The housing material i s aluminum (which has l e s s tendency to spark when rubbed than steel) of a minimum thickness of t h i r t y millimeters to withstand an i n t e r n a l pressure of 500 kPa. The 150 mm impeller can be rotated by the direct-current type, 1200 watt motor to speeds up to 23,000 rpm. This single stage compressor discharges through a d i f f u s e r made from another vacuum cleaner impeller which has nearly i d e a l blade angles when used as a d i f f u s e r . The fan shaft material i s s t a i n l e s s s t e e l and i s supported by three high-speed, shielded bearings. The motor i s located away from the c i r c u l a t i n g fan to avoid i g n i t i o n , by the sparking commutator of the motor, of any mixture leaking from the fan shaft s e a l . The motor cooling fan helps to prevent any leaking mixture from entering the motor. The fan speed i s c o n t r o l l e d by a variable transformer whose s p e c i f i c a t i o n s are given i n Appendix D. F i g . 3.6 shows the labyrinth shaft seal used to i s o l a t e the system from the ambient atmosphere. The seal consists of ten stationary rings concentric with the fan shaft. The r a d i a l clearance between the shaft and the i n t e r n a l diameter of the rings i s less than 0.013 mm; the distance between rings i s approximately 3.8 mm. These multiple low clearance rings repeatedly t h r o t t l e the high pressure mixture to minimize the amount l o s t . System pressure during fan operation i s always set above ambient so that seal leakage i s outward to prevent d i l u t i o n of the mixture. Leakage i s removed from the seal and motor by a vacuum pump. Measurements of the leakage rate through the shaft seal are presented i n section 3.10. Although the apparatus was designed for high explosive pressures, measures were taken to protect the operator and equipment against harm and damage. A blow-off valve i s i n s t a l l e d to vent the mixture i n the event of excessive pressure. The operator i s also protected by a polycarbonate s h i e l d surrounding the apparatus. I f shrapnel i s ejected by the system, i t would be constrained and redirected away from the operator. 3.5 Instrumentation Provisions are made for chamber evacuation, for pressure measurement, f o r l o c a l v e l o c i t y d i s t r i b u t i o n measurement (with hot wire or l a s e r doppler anemometers) , and f o r temperature measurement. Local v e l o c i t y d i s t r i b u t i o n measurement and temperature measurement were not performed f o r t h i s work. Fi g . 3.7 shows, i n schematic form, the instrumentation and how i t i s connected to the apparatus. Not a l l of the instrumentation mentioned next i s i l l u s t r a t e d i n the figure because some instruments are not d i r e c t l y connected to the apparatus. The instruments include: a K i s t l e r p i e z o e l e c t r i c pressure transducer and charge amplifier, a Nicolet d i g i t a l storage oscilloscope, a Compaq personal computer, a large diameter Bourdon tube pressure gauge with a range of -780 mm to +780 mm of mercury, a large range Bourdon tube pressure gauge, and a high-speed motion camera. A F o r t i n type mercury column barometer, a stroboscope, paddle wheels, and a Simpo 4 1 hand-held tachometer are also included although not illustrated in Fig. 3.7. Specifications for this equipment are l i s t e d in Appendix D. The piezoelectric pressure transducer i s mounted in the side of the combustion chamber at the midpoint of the chamber length. Since a piezo crystal i s not suitable for static pressure measurements, pre-ignition pressure was set by the large diameter Bourdon tube pressure gauge. The barometer was used to convert gauge pressures to absolute pressures. The pressure record from the transducer i s fed through the charge amplifier and oscilloscope to the computer for storage and processing. A Hycam high-speed motion camera in conjunction with a 1.5 watt Argon laser was used for schlieren photography of the flame propagation within the chamber. The film used was Kodak Eastman Ektachrome 7250 High-Speed Film which had an Tungsten ASA rating of 400. This film was obtained in double perforated, one hundred foot lengths and stored in a freezer before being exposed. For these experiments, quartz glass windows replaced the polycarbonate windows and special spark probes were installed. The paddle wheels and stroboscope were used to relate the steady state swirl intensity to fan speed. The stroboscope was arranged to illuminate the rotating paddle wheel in the combustion chamber so that the paddle wheel speed could be monitored. The tachometer i s used to monitor the motor rpm for swirl repeatability. 42 3.6 Ig n i t i o n and Control The i g n i t i o n voltage i s supplied by a Heathkit automotive capacitive discharge i g n i t i o n system whose s p e c i f i c a t i o n s are given i n Appendix D. F i g . 3.7 shows that the i g n i t i o n system i s triggered by a time delay c i r c u i t which, i n turn, i s triggered by a microswitch tripped by the cl o s i n g valve. The i g n i t i o n voltage i s discharged across a spark gap, shown i n Fi g . 3.8, which i s two millimeters unless otherwise stated. The spark electrodes consist of two low-aerodynamic-drag probes. These are made from 1.6 mm diameter, cold-drawn s t e e l wire. One electrode i s mounted on the combustion chamber wall (which i s e l e c t r i c a l ground) and the other i s mounted on the polycarbonate end plate. This combination of probes i s adjustable to any radius. The delay timer i s based on a standard 555 e l e c t r o n i c timer and the time constant i s cont r o l l e d by a r e s i s t o r -capacitor combination. The time delay range i s one to one hundred milliseconds and i s repeatable within one microsecond. The output from the timer t r i g g e r s both the oscilloscope and the i g n i t i o n system simultaneously so that pressure traces have i g n i t i o n at time t = 0. F i g . 3.9 i s a schematic diagram of the e l e c t r o n i c c i r c u i t of the delay timer. F i g . 3.10 i s a diagram of the switch arrangement on the timer box cover. F i g . 3.16 i s the c a l i b r a t i o n curve of the actual recorded delay time versus time indicated by the adjustment potentiometer. The s p e c i f i c a t i o n s of the timer are given i n Appendix D. The solenoid that p u l l s the valve release pin i s activated with a standard 120 v o l t pushbutton switch. When high-speed films are taken, the camera t r i g g e r switch activates the solenoid instead of the pushbutton switch. Fig. 3.7 d e t a i l s how the camera was wired into the c i r c u i t . 3.7 Flame Holders The discussion i n Chapter One suggests that, i f the flame remains attached to the spark source i n a s w i r l i n g mixture, shorter and more r e l i a b l e burning duration may r e s u l t due to increased flame area. Such flame holders would also increase turbulence i n t e n s i t y and reduce large v e l o c i t y at the spark gap to aid i n early flame development. But the increased drag of these flame holders may increase the s w i r l decay rate and p a r t i a l l y negate the e f f e c t of flame holding. A number of d i f f e r e n t flame holders were tested to observe i f the theory was correct. F i g . 3.11 shows the d i f f e r e n t flame holders tested. F i g . 3.12 shows how these flame holders were i n s t a l l e d i n the combustion chamber. Flame holders number two and three were i n s t a l l e d as b l u f f bodies upstream of the spark source as shown i n upper diagram i n F i g . 3.12. The remaining two flame holders were i n s t a l l e d as i n the lower drawing i n F i g . 3.12. 3.8 Schlieren Photographic Arrangement F i g . 3.13 displays the schlieren photographic arrangement. The lase r beam passed through a diverging lens to spread the l i g h t to the one hundred millimeter chamber diameter. Another lens was arranged to take t h i s diverging 44 beam and make i t parallel once again. This parallel light beam then passes through the chamber and, after passing through a third and fourth lens, was focused. A pin hole of approximately 0.9 mm diameter was placed at the focal point of this converging beam. The light that passed through the pin hole i s focused on the film of the camera. 3.9 Calibration 3.9.1 Combustion Chamber Swirl Rate In steady state operation, the relation between swirl rate and fan speed was obtained with four paddle wheels of different shapes shown in Fig. 3.14. The paddle wheels were free to rotate inside the chamber with the axis of rotation coinciding with the radial axis of symmetry of the chamber. The paddle wheels were made of two thin aluminum blades supported by two small brass bushings on a 3.2 mm steel shaft. This shaft was held normal to the polycarbonate end wall with a rubber compression f i t t i n g . The rotational velocity of the paddle wheels was measured with a stroboscope. With the swirl valve in the open position, the paddle wheels could not be the same diameter as the combustion chamber due to the five millimeter valve l i f t . The diameters chosen were ninety and seventy millimeters. The paddle wheels were either forty-nine or thirty-five millimeters wide. Thus, four paddle wheel combinations were possible and were tested. The reason for the different paddle wheel dimensions was to learn about the "shape" of the swirl. Shape refers to whether the swirl was of solid-body type rotation, free 45 vortex, or some combination of these two. The results from these experiments are displayed in Fig. 3.15. The top two lines (for the cases of large diameter but different width paddle wheels) were nearly identical showing that, at this paddle wheel diameter, there was l i t t l e effect of paddle wheel width. The fan speed chosen for combustion experiments was based on the swirl calibration curve of the large diameter, large width paddle wheel since i t represented the closest approximation to the integral of the angular momentum. The small diameter paddle wheels had slower swirl rates than the large diameter paddle wheels, a finding which was counter intuitive. With solid body rotation, the small diameter paddle wheels should have rotated faster than the large diameter, since they were not inhibited by the boundary layer formed at the outer circular wall. The fact that the small diameter paddle wheels were slower indicated that, when the valve was open and the fan was operating, solid-body type rotation did not exist. This may have been due to the high-speed jet of mixture at the outer wall. A measure of the rate of swirl decay was obtained by observing a high-speed schlieren film of the rotation of the burned gases after combustion. The film was exposed at 5000 frames per second (so each frame represented 0.2 ms duration) and the pre-ignition swirl rate was 6000 rpm. The number of frames for the mixture to rotate ninety degrees around the chamber was recorded at one hundred frame intervals and the corresponding swirl rpm was calculated and plotted in Fig. 3.16. This figure shows that, during the f i r s t twenty milliseconds after ignition (which i s the typical maximum length of total burning duration for these experiments), the swirl decayed about 1000 rpm. 3.9.2 Pressure Transducer The piezoelectric transducer i s inherently suitable for the dynamic pressure measurements because i t is not limited by the inertia of mechanical component displacement. A weakness of the piezoelectric transducer i s the tendency of the output signal to d r i f t and decay over time periods of several seconds. This is due to the pressure induced electric charge leaking through cable insulation and charge amplifier input impedance (which, at approximately one hundred teraohms, i s s t i l l not high enough to prevent leakage) . This decay rate i s not a problem for the dynamic combustion pressures but requires specific procedures for setting the static precombustion pressure and for transducer calibration. The pressure transducer was calibrated using a dead weight type pressure tester. Appendix C describes the calibration procedure. 3.9.3 Ignition Delay Timer The delay-timer-indicated time was calibrated against actual delay time recorded by the oscilloscope as shown in Fig. 3.17. calibration was less crucial for the timer since the delay was easily measured with the oscilloscope during experiments. For the swirl tests, the timer was set for the 47 shortest av a i l a b l e time, which was approximately one millisecond. 3.9.4 Large Low Range Pressure Gauge The large low range pressure gauge was c a l i b r a t e d by simultaneously applying vacuum to a column of mercury and the pressure gauge. The readings were compared and recorded. This t e s t showed the gauge to be accurate to within 0.5 mm of mercury with applied vacuum. The gauge was assumed accurate i n the p o s i t i v e pressure range. I f the gauge was not accurate i n the p o s i t i v e range, the r e s u l t i n g error would be small since, the highest p r e - i g n i t i o n was about f i f t y millimeters of mercury gauge pressure, which was j u s t outside the c a l i b r a t e d range. 3.9.5 Barometer Because of the technical d i f f i c u l t i e s of doing so, the barometer was not rec a l i b r a t e d . According to the manufacturer, the barometer was accurate to 0.3 mm of mercury and because pressure measurements of t h i s p r e c i s i o n were never needed, i t was assumed that the barometer error had l i t t l e , i f any, e f f e c t on the accuracy of the experimental r e s u l t s . 3.9.6 Gas Chromatograph Accurate mixture preparation was e s s e n t i a l i n t h i s research since flame speed i s strongly dependent on mixture strength. For her experimental thesis on the laminar burning speed of various fuels and mixtures, Hung (1986) prepared a set of mixture c a l i b r a t i o n plots for the Hewlett Packard 5750B gas chromatograph. The same gas chromatograph was used for 48 mixture strength measurements in this work. In addition, a new set of calibration plots were made for this work and compared with those of Hung. Strong correlation was found between the two sets. Fig. 3.18 i s a calibration plot of methane and a i r which shows the results of Hung and this work. Appendix A i s a manual explaining how the gas chromatograph works, how i t was operated, and how i t was calibrated. 3.10 Testing 3.10.1 System Pressure and Vacuum The complete system was regularly tested for pressure and vacuum leaks. It was necessary to test for both types of leaks since, in some cases, a leak could occur in only one direction. For example, an O-ring that seated properly and seals when a vacuum was applied may have been l i f t e d off the seat when high pressure was applied. The pressure test was conducted by pressurizing the closed system with air to approximately 689 kPa. Though this was not a recommended method of testing for leaks, i t was performed for the following reasons. F i r s t l y , the piping system (which was considered the weakest system component) had already passed a water pressure strength test to approximately 2.62 MPa. Secondly, there were many close tolerance bearings and components which would be damaged by corrosion from water. Thirdly, using o i l instead of water would have been d i f f i c u l t to clean out. Finally, i t was easier to detect a gas leak than a liquid leak. With the system pressurized, the pressure supply was shut 49 off from the system and the rate of pressure decline monitored. A mixture of liquid soap and water was used to test the various joints, connections, and welds for leakage and porosity. The necessary repairs were performed where leaks were indicated and the system retested. The vacuum test was similar to the pressure test except that the system was evacuated and the rate of pressure increase was observed. Repairs consisted of replacing or install i n g seals like gaskets, 0-rings, or furrels; applying commercial weld porosity sealer; and carefully tightening loose connections and joints. The system was never perfectly sealed but eventually the leakage was almost undetectable. 3.10.2 Fan Shaft Seal Leakage The circulating fan shaft seal effectiveness was tested in a similar way as the system pressure retention described in section 3.10.2. After the system had passed the pressure and vacuum tests outlined in the last section, the apparatus was repressurized with a i r and the leakage due to the shaft seal was monitored. Fig. 3.19 i s a plot of seal pressure versus time. During experiments, the system pressure was only slightly above atmospheric so there was l i t t l e leakage past the seal. 3.10.3 Valve Closure Time Valve closure time was measured with an oscilloscope. The valve open position was detected with an optical diode and the closed position with the timing box trigger microswitch. A number of tests were repeated and the result was a 50 consistent valve closure time of 1.5 ms. The valve was designed to close i n about three milliseconds and thus, the design s p e c i f i c a t i o n s , i n t h i s case, were exceeded. A close look at the pressure h i s t o r i e s i n Chapter Five reveals a superimposed r i p p l e . This r i p p l e was caused by the apparatus v i b r a t i n g from the impact of valve closure. Figs. 3.20, 3.21, and 3.22 show the v i b r a t i o n at various scales of magnification. The chamber i n these cases had no combustible mixture inside and, therefore no combustion was taking place. There i s no evidence of v i b r a t i o n on the quiescent combustion p l o t s i n Chapter Five since, i n these cases, there was no valve closure. In summary, an apparatus was designed to create steady precombustion s w i r l up to 6000 RPM within a short c y l i n d r i c a l chamber. The s w i r l generating mechanism i s a T-shaped valve through which mixture enters and e x i t s the chamber tan g e n t i a l l y . The mixture i s c i r c u l a t e d with an explosion proof fan driven by a 12 00 watt motor to speeds up to 2 3,000 rpm. The main source of burning duration data i s derived from a p i e z o e l e c t r i c pressure transducer and the sig n a l i s stored on a computer. Homogeneous mixture i g n i t i o n i s from a va r i a b l e l o c a t i o n spark probe made from 1.6 mm s t e e l wire. The apparatus i s adaptable to provide schlieren photography. With minor modifications, the apparatus can provide va r i a b l e length-to-diameter r a t i o by having end plates move i n or out of the chamber and variable precombustion pressure by c o n t r o l l i n g charging pressure. In addition, the apparatus can 5 1 be al t e r e d to vary precombustion temperature by heating the mixture, and turbulence l e v e l s with no-swirl by c o n t r o l l i n g the i n j e c t i o n v e l o c i t y , valve " l i f t " and i g n i t i o n delay. By c a l i b r a t i o n and t e s t i n g , the apparatus was found to meet the design c r i t e r i a for "steadiness", r e p e a t a b i l i t y , and control. 52 4. PROCEDURES 4.1 Mixture Preparation Commercially available extra dry a i r (which had less than ten parts per m i l l i o n moisture) and instrument grade methane (which was greater than 99.5 percent pure) were used to make chemically balanced mixtures. The gases were mixed and then analyzed by a Hewlett Packard gas chromatograph following the procedure given i n Appendix A. The mixtures were stored i n s t e e l b o t t l e s under pressure and were metered with a pressure regulator. The maximum storage pressure i n the bot t l e s was li m i t e d to about 2.1 MPa so that i f , however u n l i k e l y , there was combustion within the the bot t l e , the burned pressure would not exceed the 17.0 MPa maximum working pressure. The p a r t i a l pressures of methane and a i r were calculated from the ide a l gas law for a chemically balanced mixture. Appendix B i s a sample of the mixture c a l c u l a t i o n s . A f t e r evacuating another b o t t l e , the fu e l was slowly added u n t i l the desired p a r t i a l pressure and room temperature were achieved. Then a i r was slowly added to the previously evacuated b o t t l e u n t i l the f i n a l desired 2.1 MPa pressure was achieved. This new mixture was allowed to homogenize f o r at lea s t twelve hours before a sample was taken for analysis i n the gas chromatograph. Mixture sample tests indicated that twelve hours was not long enough to achieve complete homogeneity and high accuracy. But t h i s time was long enough to get a "rough" idea of the stoichiometry of the mixture for further 53 i t e r a t i o n s . As the mixture approached the desired r a t i o and accuracy, the time for d i f f u s i o n was increased to ensure complete homogeneity. 4.2 Chamber F i r i n g Test Seguence The following procedures were used for combustion chamber f i r i n g . Later, a procedure i s l i s t e d f or operating the combustion chamber i n conjunction with the high-speed camera. 4.2.1 Start Up The sequence for the f i r s t f i r i n g which i s presented here was d i f f e r e n t than for repeated f i r i n g . A l l e l e c t r i c a l equipment was allowed to warm up for at l e a s t two hours. Special attention was given to the pressure transducer charge am p l i f i e r and oscilloscope which, u n t i l warm, d r i f t e d excessively. Next, the combustion chamber, supply l i n e s (as shown i n F i g . 3.7), pipes, fan, and shaft seal were evacuated and flushed with mixture. The fan shaft seal 0-ring, as shown i n Figs. 3.5 and 3.6, was seated for e f f e c t i v e evacuation. The f i n a l evacuation of the system was within a few millimeters of mercury of absolute zero pressure, which was v e r i f i e d with the barometer and low range pressure gauge. I f a low pressure could not be achieved, there was a s i g n i f i c a n t a i r leak somewhere within the system and t h i s was located and repaired before continuing. As a safety precaution, the spark lead was disconnected from the probe and attached to e l e c t r i c a l ground on the apparatus. Next, the sw i r l valve was raised with the reset lever and the solenoid r o l l e r pin brought into p o s i t i o n so that the 54 valve remained open. With the oscilloscope turned ON and ready, the valve-triggered-microswitch was manually tripped to determine i f the t r i g g e r voltage on the oscilloscope was set such that the l i g h t i n the t r i g g e r mode switch flashed. Next, the capacitive discharge i g n i t i o n system was checked by placing the spark lead approximately ten millimeters away from the e l e c t r i c a l ground and observing a spark as the microswitch was tripped. The oscilloscope was zeroed by adjusting the coarse and f i n e zero adjustments while a l l input switches were set to ground. With the oscilloscope set on AUTO t r i g g e r , the charge ampl i f i e r was also adjusted to zero while i n the RESET mode. Table 4.1 l i s t s the instrumentation switch positions. At t h i s time, the chamber and a l l the associated plumbing were evacuated, the i g n i t i o n tested, and the charge amplifier and oscilloscope were ready for data. The evacuated system was charged with mixture by se t t i n g the pressure regulator on the mixture b o t t l e to approximately 110 kPa, and then opening the supply valve to the system. The system was set at 800 mm of mercury absolute pressure. The i g n i t i o n lead f o r the spark probe was reattached and the charge amplifier was switched from the RESET mode to the OPERATE mode. The oscilloscope was set for the storage mode so that when the combustion chamber was f i r e d , the recorded pressure trace could not be erased by a possible second t r i g g e r pulse. Storage mode was obtained by simultaneously pressing the LIVE and STORE push button switches so that both the red and green l i g h t s i n the switches remained ON. After 55 a l l valves to the system were closed, the chamber was ready for f i r i n g . When the safety shield was in place, the fan motor was turned ON and set for the desired swirl speed by adjusting the variable transformer motor control. After approximately fifteen to twenty seconds, when the flow was assumed f u l l y developed within the apparatus, the solenoid push-button switch was pressed to f i r e the combustion chamber. 4.2.2 Repeat Firing After the f i r s t f i r i n g , the entire system was not evacuated, only the chamber. This saved mixture and time. The following procedure was used for repeated f i r i n g of the chamber. Immediately after the combustion chamber was fired, the fan motor was turned OFF and the combustion chamber evacuated. As the evacuation process continued, the data recorded in the oscilloscope was transferred to the computer. The combustion chamber was evacuated to about -73 5 mm of mercury gauge pressure. Absolute zero pressure could not be achieved in the combustion chamber because of the imperfect seal between the swirl valve and block which allowed some unburned mixture from the piping system to leak into the combustion chamber as i t was evacuated. When the lowest pressure was reached, fresh mixture from the mixture bottle was introduced into the combustion chamber. When a pressure higher than atmospheric was reached within the combustion chamber, the T-shaped valve was reset (with the spark probe e l e c t r i c a l l y grounded to prevent accidental ignition). Enough 5 6 mixture was allowed to flow into the system u n t i l the desired precombustion pressure was reached. With the oscilloscope reset and the safety s h i e l d replaced, the combustion chamber was again ready for f i r i n g . 4.3 Photography The procedure f o r taking high-speed schlieren photographic films of the combustion process i s s i m i l a r to the chamber f i r i n g t e s t sequence given i n sections 4.2 and 4.3. There are a few additional steps dealing with l a s e r and camera operation. The camera was set on a t r i p o d and adjusted to the same height as the center of the combustion chamber. Fi g . 4.1 displays camera d e t a i l s . The distance between the camera and pin hole (as shown i n F i g . 3.11) was adjusted so that the image of the chamber occupied about three-quarters to seven-eighths of the camera. The chamber s i z e was observed by turning the l a s e r ON and observing the image projected through the camera view port on a piece of paper. The chamber image was focused d i r e c t l y on the f i l m as no lens was used. The energizing of the valve solenoid was co n t r o l l e d by a f i l m footage sensor switch (which i s c a l l e d the event timer) i n the camera instead of the pushbutton solenoid switch. This switch was set for about f i f t y to s i x t y f i v e (assuming a one hundred foot r o l l of f i l m was used) and the leads from the switch were wired, with a turn-off switch, i n p a r a l l e l with the solenoid switch as shown i n F i g . 3.7. Next, the f i l m was loaded with the room darkened and the 57 operator and camera covered with a dark hood to prevent stray l i g h t from exposing the f i l m prematurely. The camera speed was set at 5000 frames per second. To help measure the speed of the camera, an o s c i l l a t o r was connected to the camera which gave a small f l a s h of l i g h t on the edge of the f i l m every millisecond. This device was turned ON before the camera was operated. The main power cable to the camera was plugged i n fo r at l e a s t one hour before f i r i n g the chamber so that the camera e l e c t r o n i c speed control could warm up and s t a b i l i z e . The camera t r i g g e r cut-off switch was turned ON, the l a s e r was set at approximately 210 m i l l i w a t t s and turned ON, and the fan motor turned ON when the camera a c t i v a t i o n switch was pressed. Immediately a f t e r f i r i n g and filming, the camera a c t i v a t i o n switch, the fan power switch, the l a s e r power switch, and the camera t r i g g e r cutoff switch were turned OFF and the combustion chamber was evacuated. The pressure trace was stored, and the f i l m was unloaded i n a light-subdued room. 4.4 Data Processing The data a c q u i s i t i o n program "Datanic" l i s t e d i n Appendix E takes data from the Nicolet oscilloscope and stores i t on a computer disk, i n eit h e r Binary or ASCII format. The program prompts the operator to press the "p" computer key when the pressure trace i s displayed on the oscilloscope screen and the cable from the oscilloscope to the computer i s connected. A f t e r a short time, the computer signals the operator that the data transmission i s complete and the computer redisplays the pressure h i s t o r y to v e r i f y successful data transmission. 58 A f t e r pushing the carriage return key, the operator has the choice of storin g the data i n eithe r ASCII of Binary format. Once the choice i s made and the data f i l e name s p e c i f i e d , the data i s stored on the diskette. The computer returns a prompt when data storage i s complete. The data processing program c a l l e d "RIAHI", which i s l i s t e d i n Appendix F, was written by Ardeshir Riahi who i s a graduate student at the University of B r i t i s h Columbia i n the Department of Mechanical Engineering. This menu driven FORTRAN program reads up to twenty eight ASCII or Binary data f i l e s ; each f i l e may contain up to 4000 points. A f t e r loading the program, the main menu appears and asks i f the operator wants to read i n data f i l e s , ensemble average these f i l e s , c a l c u l a t e the 10, 50, 90, and 10-90% peak pressures with the associated standard deviations, or f i n d the ensemble average with respect to time. Each of these menu items has the necessary prompts for the operator to input the proper information. The program was tested by comparing the output with c a l c u l a t i o n s performed manually from data obtained from the oscilloscope. Section 4.6 has d e t a i l s of these t e s t s . 4.5 Measurement Uncertainties 4.5.1 Introduction A measure of the o v e r a l l random error i s given by the r a t i o of the standard deviation of the burning duration to the burning duration at a cer t a i n no-swirl condition. This c a l c u l a t i o n was performed on the three no-swirl spark 59 locations and the outside location had the highest r a t i o for the early burning duration. The highest error f o r the main burning duration was for the mid-radius l o c a t i o n . The early burning duration random error was defined and t y p i c a l l y evaluated as: SD t 1 Q 157 * 10" 6 — = — = 0.0113 or 1.13% error t 1 Q 13.9 * 10" 3 The main burning duration random error was defined and t y p i c a l l y evaluated as: SD t 1 0 _ g o 520 * 10" 6 — = — = 0.0268 or 2.68% error t 1 0 _ 9 0 19.4 * 10 3 The errors calculated above include the error associated with oscilloscope and computer d i g i t i z a t i o n . The pressure traces presented i n Chapter Five have a pressure resolution of 0.195 kPa ( i . e . , 800 kPa/4096 points) and a time re s o l u t i o n of between 0.005 and 0.02 ms ( i . e . , 20/4000 = 0.005 and 80/4000 = 0.02) depending on time scale. Variations i n the delay timer are also accounted for i n the above random error estimates. The timer v a r i a t i o n was checked with the oscilloscope and the delay time was repeatable within one microsecond. Systematic or absolute error may e x i s t i n the approximation of mass f r a c t i o n burned as a function of f r a c t i o n a l pressure r i s e as given i n equation 5.1. This equation i s exact for adiabatic combustion with no d i s s o c i a t i o n . The combustion within t h i s chamber exhibits heat t r a n s f e r and d i s s o c i a t i o n e f f e c t s so that the calculated mass f r a c t i o n i s possibly affected by up to 2% error. 60 4.5.2 Swirl Intensity During experiments, the s w i r l rate was set by c o n t r o l l i n g the motor voltage with a variable transformer. A paddle wheel could not be used i n the chamber during combustion since i t hindered combustion. The c o r r e l a t i o n between the paddle wheel r o t a t i o n a l v e l o c i t y and the fan v e l o c i t y was found to be very high for each paddle wheel tested. The lowest c o r r e l a t i o n c o e f f i c i e n t was approximately 0.9996. But each paddle wheel shape, as shown i n F i g . 3.14, had d i f f e r e n t c o r r e l a t i o n curves as shown i n F i g . 3.15. The paddle wheel used f o r these experiments was the largest since i t represented the closest approximation to the t o t a l angular momentum of the s w i r l . Since d e t a i l e d measurements of the s w i r l i n g flow f i e l d were not taken, the error i n absolute s w i r l i n t e n s i t y i s estimated to possibly be as high as 10%. The error associated with the tachometer i s two rpm i n the range of 6000 to 30,000 rpm. The e f f e c t of absolute s w i r l i n t e n s i t y on burning duration may be approximately 5%. However, the random error e f f e c t s of s w i r l i n t e n s i t y are believed n e g l i g i b l e (due to high c o r r e l a t i o n c o e f f i c i e n t ) . 4.5.3 Gas Chromatograph and Mixtures The Hewlett Packard research gas chromatograph was used to evaluate the stoichiometry of the prepared mixtures. The machine was c a l i b r a t e d by evaluating d i f f e r e n t known mixtures. This c a l i b r a t i o n may r e s u l t i n a mixture strength uncertainty of 2% and the r e s u l t on burning duration i s considered n e g l i g i b l e . 61 The gas chromatograph operation yielded variable readings for mixture samples taken from the same sample v i a l . The indicated ratio varied in some cases up to 1%. Tests that gave the richest results were generally believed to be the most reliable since, conceivably, a i r from the ambient atmosphere, and not fuel, could leak into the test sample. The effect of impurities (including water vapor) in the fuel and a i r was shown by Hung (1986) to be less than 1% on mixture strength. The mixture was analyzed regularly and found to remain stable. 4.5.4 Pre-Ignition Pressure The pre-ignition pressure may have varied by about fifteen millimeters of mercury absolute pressure. A conservative estimate on the effect of peak pressure i s that a variation in pre-ignition pressure has a ten fold effect on peak pressure. Thus the variation due to i n i t i a l pressure in this case i s about 150 mm of mercury absolute pressure (i.e. 10 * 15 = 150 mm) which corresponds to a peak pressure variation of 20 kPa (since 7.5 mm of Hg i s about one kPa) . This represents an error of less than 3% on the peak pressure (i.e., 20 kPa uncertain pressure/700 kPa peak pressure = 0.0286) and less random and systematic error effect on the burning duration and rate. The effect of variable i n i t i a l pressure on mass fraction burned i s considered negligible. 4.5.5 Data Processing Additional error may be found in the computer program used to analyze data. A close look at the pressure traces in 62 Chapter Five reveals a superimposed ripple which tests have shown to be caused by apparatus vibration from valve closure. This ripple may affect the data obtained by the program. When the program reads a certain pressure (e.g., the 10% peak pressure), more than one time may represent a certain pressure as a result of the superimposed ripple. A test was performed by comparing the computer output with that from the Nicolet oscilloscope. The results, which are displayed in Table 4.2, indicate that the computer was within about 0.9% for burning rate calculations, and within about 9% on standard deviation calculations. In summary, the total error i s believed to be less than 5% for burning duration and within 15% for standard deviation. Many of the errors mentioned above can be controlled through proper procedure and understanding of equipment. 63 5. RESULTS 5.1 Introduction The following paragraphs outline the strategy used to collect data. The f i r s t set of experiments with an unshielded spark gap demonstrated how swirl and spark location affected burning rate and cyclic variation. Four levels of swirl (0, 2000, 4000, and 6000 rpm) with three spark locations (central, "mid-radius" [actually r/R = 0.6], and outside) gave twelve possible experimental combinations. In addition, tests were made of the effect of spark shielding on burning duration and cyclic variation. Various low-aerodynamic-drag bodies were placed upstream of the mid-radius located spark source in the swirling flow. This data provided information about the effect of spark source geometry on combustion duration and variation. Information was sought about the very early period of combustion when the flame zone was small, and about the subsequent development of the burning zone. Of particular interest was the question of whether the flame detaches from the spark source in a highly swirling flow f i e l d . A series of photographic experiments were conducted using schlieren techniques with various swirl intensities, spark locations, and spark shields. The experimental conditions were as follows: A) A l l experiments were conducted with samples from a single batch of chemically balanced methane-air mixture. B) Unless otherwise indicated, the spark gap for a l l experiments was two millimeters. C) The valve l i f t i n a l l experiments was f i v e millimeters. D) The p r e - i g n i t i o n pressure was 800 mm of mercury absolute pressure. E) The i g n i t i o n delay timer was set f o r one ms following valve closure. As shown by Lewis and von Elbe (1961, p. 374) , the mass f r a c t i o n burned, X, i n an adiabatic constant volume chamber i s nearly equal to the f r a c t i o n a l pressure r i s e even with the ef f e c t s of d i s s o c i a t i o n , i . e . ; P(t) - P 0 X(t) = (5.1) P - P rmax o where; X(t) i s the mass f r a c t i o n burned, at a given time, P(t) i s the chamber pressure at a given time, P Q i s the p r e - i g n i t i o n pressure, I?max i s the maximum pressure following combustion. D i f f e r e n t i a t i n g equation 5.1 r e s u l t s i n a mass burning rate equation as a function of the rate of change i n pressure: dX 1 dP = (5.2) d t pmax " p o d t where; dX/dt i s the mass burning rate, 65 P Q i s the p r e - i g n i t i o n pressure, P m a x i s the maximum pressure following combustion, dP/dt i s the rate of change of pressure. The maximum pressure due to combustion w i l l be affected somewhat by heat transfer from the chamber which depends on the extent to which the burned gases are i n contact with the walls during the combustion period. However, the assumption i s made that even with heat transfer, equation 5.1 i s v a l i d i f the experimental (rather than the adiabatic) value of P m a x i s used. This assumption i s s a t i s f a c t o r y because heat transfer, during the short period ( t y p i c a l l y less than twenty milliseconds) of combustion, has a small e f f e c t on the pressure time curve. The term "burning duration" re f e r s to the time from i g n i t i o n to a given f r a c t i o n a l pressure r i s e . Two measures of burning duration were defined. The early burning duration ref e r s to the period from i g n i t i o n to 10% peak pressure and the main burning duration re f e r s to the period from 10-90% peak pressure, i . e . , 10-90% mass f r a c t i o n burned. Ensemble averages for repeated tests were taken with respect to pressure to determine the mean burning curve of pressure versus time. The term "standard deviation" r e f e r s to va r i a t i o n s from t e s t to t e s t i n the time from i g n i t i o n to a s p e c i f i c mass f r a c t i o n burned. This d e f i n i t i o n was chosen, rather than the standard deviation i n the pressure at a given time because of the importance of burning duration i n engines. 66 5.2 Repeatability of Measurements Fig . 5.1 demonstrates experimental r e p e a t a b i l i t y . The i n i t i a l conditions before combustion were central i g n i t i o n and no s w i r l and the pressure curves were taken from the same mixture batch used for the other experiments. Eight pressure h i s t o r i e s were plotted. Experimental r e p e a t a b i l i t y was demonstrated by the extent to which these pressure curves agree i n the absence of turbulence or other mixture motion. 5.3 V a r i a t i o n of Estimated Standard Deviation with Sample Size With s w i r l and turbulent i n i t i a l conditions, repeated t e s t s might be expected to show more v a r i a t i o n than with quiescent i n i t i a l conditions. Witze (1982), and Groff and Sinnamon (1982) have shown that turbulent combustion i n engines i s associated with v a r i a t i o n s from cycle to cycle i n combustion duration; the same might be expected i n the constant volume chamber. With t h i s i n mind, te s t s were undertaken to determine the minimum acceptable sample s i z e for estimating the standard deviation i n combustion duration. Since i t was desirable that a l l the experiments should be obtained from one mixture batch and since the t e s t s were time consuming, i t was desirable that the sample s i z e not be too large. The apparatus was operated t h i r t y times with two sets of d i f f e r e n t i n i t i a l conditions. The standard deviations of the early and main burning duration were calculated a f t e r each run of the apparatus and plotted i n Figs. 5.2 and 5.3. The i n i t i a l conditions of F i g . 5.2 were: central i g n i t i o n , three 67 millimeter spark gap, and 6000 rpm s w i r l . The i n i t i a l conditions of F i g . 5.3 were outside i g n i t i o n , two millimeter spark gap, and 6000 rpm s w i r l . The d e f i n i t i o n of standard deviation used i n c a l c u l a t i n g the r e s u l t s shown i n Figs. 5.2 and 5.3 i s : <Tt - A / V n where; CT^. = standard deviation with respect to time t ^ = time from i g n i t i o n for sample number i , and n = sample s i z e . These te s t s indicated that a sample s i z e of f i f t e e n could provide an estimate of the standard deviation though (as Fig 5.3 suggests), the standard deviation i n the i n i t i a l burning period (0-10%) i s uncertain by probably + 15%. This uncertainty, due to the smallness i n sample s i z e , has to be kept i n mind when reviewing c e r t a i n of the following r e s u l t s . 5.4 E f f e c t Of Swirl On Burning Duration And Heat Transfer Fig. 5.4 i l l u s t r a t e s the large e f f e c t of s w i r l on burning duration f o r the outside spark location shown i n F i g . 3.8. Each curve represents the ensemble average of f i f t e e n pressure traces with the same i n i t i a l conditions. The number at the pressure peak of each curve i s the time i n milliseconds from i g n i t i o n to the pressure peak. The curves d i f f e r only i n pre-i g n i t i o n s w i r l i n t e n s i t y . The slowest burning case occurred with quiescent i n i t i a l 68 conditions. At 2000 rpm, the burning duration was su b s t a n t i a l l y reduced. Increasing s w i r l to 4000 rpm and 6000 rpm further decreased the burning duration but only by small amounts. In F i g . 5.4, peak pressure i s seen to increase with increased s w i r l . This would be the e f f e c t of les s heat tra n s f e r out of the chamber with shorter combustion time. Since peak pressure i s almost proportional to peak temperature, the r i s e i n peak pressure indicates the reduction i n heat transferred out. With increased s w i r l , and thus decreased burning duration, there would be less time for heat to be transferred to the chamber walls. But with increased s w i r l , there would be an increase i n the heat t r a n s f e r f i l m c o e f f i c i e n t ; i t seems that the increase i n f i l m c o e f f i c i e n t from s w i r l had les s e f f e c t on the heat t r a n s f e r during combustion than the substantial decrease i n burning duration. F i g . 5.5 demonstrates the same trends as F i g . 5.4 but for central spark locat i o n . There i s continued reduction i n burning duration from the no-swirl to the 6000 rpm case. During combustion with central i g n i t i o n , the hot gases do not touch the walls u n t i l combustion i s nearly complete. As a r e s u l t the heat transfer i s small and remains nearly the same for a l l s w i r l i n t e n s i t i e s . This trend i s shown i n F i g . 5.5 where the peak pressures are almost i d e n t i c a l at a l l sw i r l l e v e l s (except f o r the quiescent case). F i g . 5.6 (for the mid-radius spark location) also shows that burning duration decreases continuously with increased 69 s w i r l . The peak pressure f i r s t r i s e s with increased s w i r l , then drops. In t h i s case, the increase i n heat tr a n s f e r rate due to the further increase i n s w i r l may have greater e f f e c t than the reduction i n burning duration. Fig . 5.7 shows burning durations derived from the same data plotted i n Figs. 5.4 to 5.6. I t indicates the e f f e c t of s w i r l on t 1 0 for the three spark locations. With central i g n i t i o n , there i s l i t t l e decrease i n the early burning duration and t h i s i s i l l u s t r a t e d by the near horizontal curve. With mid-radius spark location, the decrease i s greater, e s p e c i a l l y with the increase i n s w i r l from 0 to 4000 rpm. At 6000 rpm and mid-radius location, there are two points because t h i s one case was repeated for reasons that w i l l be explained l a t e r . The greatest e f f e c t of s w i r l on the early burning duration i s on the outside spark location during the r i s e i n s w i r l from 0 to 2000 rpm; t h i s i s i l l u s t r a t e d by the difference i n burning duration for these two i n i t i a l conditions. Fig. 5.8 i l l u s t r a t e s the e f f e c t of s w i r l on t 1 0 _ g o . The greatest e f f e c t of sw i r l i s i n the 0 to 2000 rpm sw i r l i n t e n s i t y range with outside spark location, but a l l three spark locations exhibit the large e f f e c t s i n t h i s swirl i n t e n s i t y range. Further increase i n s w i r l r e s u l t s i n l i t t l e decrease i n main burning duration for a l l spark locations. Thus, once i g n i t i o n has taken place and the flame kernel i s well developed, the main burning duration i s not strongly dependent on eithe r spark location or on s w i r l i n t e n s i t y 70 (above a c e r t a i n minimum l e v e l ) . Figs. 5.7 and 5.8 were plotted on the same page with the same scales so that the early and main burning durations could be compared. Such a comparison reveals that, for swirl i n t e n s i t i e s greater than 2000 rpm, the early burning duration (representing a pressure r i s e of only 10%) i s always longer than the main burning duration (representing a pressure r i s e of 80%). In the case of central i g n i t i o n , the r a t i o of t 1 0 to t 1 0 _ 9 0 i s about two. Table 5.1 i s a summary of the burning durations for a l l mid-plane spark locations. The tables referred to here use abbreviations f o r the experimental i n i t i a l conditions. Three uppercase l e t t e r s are arranged to indicate s w i r l i n t e n s i t y with the f i r s t two l e t t e r s , and spark l o c a t i o n with the t h i r d l e t t e r . For example, NSC stands for No-Swirl i n t e n s i t y and Central spark loca t i o n . The other abbreviations are LS_ for Low-Swirl, MS_ for Medium-Swirl, and HS_ f o r High-Swirl. The other spark locations are M for Mid-radius, and O for Outside. In Table 5.1, the underlined numbers are the time i n milliseconds of the shortest burning durations at a given s w i r l l e v e l f o r a given type of burning duration. The table was designed to compare d i f f e r e n t spark locations with the same s w i r l l e v e l and t h i s i s discussed i n the next section. 5.5 E f f e c t Of Spark Location On Burning Duration Fi g . 5.9 shows the e f f e c t of spark l o c a t i o n on burning duration with no s w i r l . Each curve represents the ensemble average of f i f t e e n pressure traces' with the same i n i t i a l conditions. The numbers at the pressure peaks are the time i n milliseconds from i g n i t i o n to peak pressure. As would be expected, the outside i g n i t i o n l o cation has the longest burning duration because i t has the greatest flame t r a v e l distance. Likewise, the central i g n i t i o n case has the least burning duration because of the l e a s t flame t r a v e l distance. The slopes of the curves i n F i g . 5.9 show that the burning rate i s d i f f e r e n t i n the three cases and that the early burning duration i s approximately the same i n the three cases. The differences, which occur i n the main burning duration, are not due to v a r i a t i o n s i n the laminar burning v e l o c i t y , but due to spark location, i . e . , due to flame t r a v e l distance. Because of heat transfer e f f e c t s during combustion the peak pressures r i s e as the spark location i s moved toward the central l o c a t i o n . With outside i g n i t i o n , combustion flames are constantly i n contact with the cool outer wall. Therefore, a greater amount of heat transfer takes place than i n the central i g n i t i o n case. In t h i s case, heat tr a n s f e r i s large c l o s e r at the end of combustion as the burned gas comes i n f u l l contact with the walls. Fi g . 5.10 shows that with the f i r s t introduction of s w i r l ( i . e . , 2000 rpm), the burning duration i s shortest f o r the outside spark loca t i o n . The main burning durations are approximately the same for the three spark locations and the differences between the curves i s at t r i b u t e d to v a r i a t i o n s i n the early burning duration. Peak pressure also drops s l i g h t l y 72 when the spark location i s moved to the outside due to the increased heat tr a n s f e r of the outside spark l o c a t i o n . The same trends observed i n the preceding figure are displayed i n F i g . 5.11 where the e f f e c t of spark location at medium s w i r l i s plotted. Outside i g n i t i o n i s the fastest burning location, but mid-radius location i s only s l i g h t l y slower than outside. Again peak pressure drops s l i g h t l y for the outside spark l o c a t i o n due to heat transfer. F i g . 5.12 displays the e f f e c t of spark l o c a t i o n with 6000 rpm s w i r l . This figure shows that the shortest burning duration occurs for the mid-radius spark l o c a t i o n . F i g . 5.11 shows the same trend taking place since the burning rate with mid-radius spark location, i s only s l i g h t l y slower than with outside spark loca t i o n . This trend may seem unusual. The outside location implies i g n i t i o n i n a high-speed mixture zone and i n the presence of high wall-shear-generated turbulence. Such a zone seems i d e a l f o r f a s t burning. Since however, burning duration i s shortest for the mid-radius spark location, there must be some other mechanism taking place here. Here are two possible explanations. F i r s t , there must be an off-center spark l o c a t i o n where the flame t r a v e l distance i s l e a s t f o r the revolving flow f i e l d . Such a p o s i t i o n would y i e l d equal turbulent burning durations to the center and to the outside wall. This factor may outweigh the high v e l o c i t y and the turbulence of outside spark loca t i o n . Grupta, et a l . , (1984, p. 420) conclude that the fastest-burning spark l o c a t i o n with s w i r l i s at r/R = 0.6 ( i . e . , located at 60% of the radius) which was the "mid-radius" l o c a t i o n i n these t e s t s . Second, buoyancy a f f e c t s the development of the low density flame kernel. As mentioned i n Chapter One, the s w i r l i n g flow f i e l d generates high c e n t r i f u g a l forces. With a density one-sixth that of the unburned mixture, the c e n t r a l l y i g n i t e d flame kernel i s hindered by the opposing c e n t r i f u g a l force. This e f f e c t would be most prominent during the early burning duration as i s shown i n F i g . 5.12. Examination of the slopes of the main burning sections of the curves of F i g . 5.12 reveals that the main burning rates are approximately the same. The curves are separated mainly by differences i n early burning duration. This indicates that the e f f e c t of changing spark location i s i n the early burning duration. In one t e s t , the spark source was also moved a x i a l l y from the mid-plane location to near the f l a t end wall. The purpose of t h i s experiment was to study the e f f e c t on burning duration and c y c l i c v a r i a t i o n of end wall spark source because t h i s i s c l o s e r to actual engine conditions. The t e s t l o c a t i o n was mid-radius and two millimeters from the end wall. A l l four s w i r l values were tested. The r e s u l t s , which are given i n Table 5.3, show an increase i n burning duration for most cases when compared to the same i n i t i a l conditions for the mid-radius, mid-plane spark l o c a t i o n . The only cases where the near-end-wall 74 l o c a t i o n gave shorter burning duration was f o r the no-swirl, low-swirl, and medium-swirl main burning duration; these changes are probably within the experimental uncertainty. In general, the increase i n burning duration was small and probably only a function of the s l i g h t increase i n flame t r a v e l distance. 5.6 E f f e c t of Swirl and Spark Location on C y c l i c V a r i a t i o n Fi g . 5.1 (for quiescent combustion) represents the least " c y c l i c v a r i a t i o n " case and may be considered as a reference case to which other sets of i n i t i a l conditions may be compared. Figs. 5.1 and 5.13 through 5.16 are a ser i e s of figures showing the tendency for c y c l i c v a r i a t i o n to increase with s w i r l ; t h i s i s indicated by the spread of pressure h i s t o r y curves f o r d i f f e r e n t s w i r l i n t e n s i t i e s with central i g n i t i o n . Figs. 5.16 through 5.18 display the e f f e c t of spark l o c a t i o n on apparent c y c l i c v a r i a t i o n . Each figure has the same pressure scale, but the time scales are d i f f e r so that the r e l a t i v e spread must be viewed accordingly. Table 5.2 shows the estimated standard deviation i n the early and main burning durations deduced from the data. Fi g . 5.17 (for high-swirl, mid-radius l o c a t i o n <HSM1>) displays the highest c y c l i c v a r i a t i o n f o r a l l i n i t i a l conditions. This high c y c l i c v a r i a t i o n was deemed uncharacteristic, e s p e c i a l l y when compared to the other two spark locations at the same swi r l l e v e l . Another set of tests with the same i n i t i a l conditions was performed to determine i f the high c y c l i c v a r i a t i o n was 75 repeatable or due to apparatus or operator error. The repeat experiment r e s u l t s (HSM2) are displayed i n F i g . 5.18. In t h i s case, the c y c l i c v a r i a t i o n i s greatly reduced as i l l u s t r a t e d by the reduction i n spread of the various pressure h i s t o r i e s . Although the reasons for the differences between the two sets of curves are not clear, there may have been low battery voltage at the time of the f i r s t set. The battery supplies power to the i g n i t i o n system and low power could have affected the early burning flame kernel s i z e . An e r r a t i c e arly burning flame kernel s i z e could r e s u l t i n experimentally large v a r i a t i o n i n the early burning duration and, therefore, increase the c y c l i c v a r i a t i o n . Table 5.2 summarizes the standard deviation f o r a l l burning times for a l l experimental i n i t i a l conditions ( i . e . , SD t 1 Q , SD t 5 Q , SD t 9 Q , SD t 1 Q _ 9 0 ) . Figs. 5.19 through 5.22 were derived from Table 5.2. F i g . 5.19, which represents no-s w i r l i n i t i a l conditions, shows that standard deviations of the early and main burning durations may increase as the spark l o c a t i o n i s moved away from the center. This may be more an e f f e c t of burning duration than spark l o c a t i o n because burning duration s u b s t a n t i a l l y increases as the spark source i s moved outward. Figs. 5.2 0 and 5.22 indicate a s l i g h t increase i n standard deviation while F i g . 5.21 indicates a s l i g h t decrease i n standard deviation as the spark source i s moved away from the central l o c a t i o n . The present r e s u l t s do not indicate that there i s a 76 strong e f f e c t of spark location or s w i r l i n t e n s i t y on standard deviation of burning duration. 5.7 E f f e c t Of Spark Gap On Burning Duration To obtain more knowledge of c y c l i c v a r i a t i o n , a supplemental experiment was conducted. With High Swirl, the spark gap was varied between 0.5 and 3.5 mm at the mid-plane, Mid-radius spark l o c a t i o n (abbreviated as HSM). Previous t e s t s indicated that t h i s set of i n i t i a l conditions with a two millimeter gap yielded high c y c l i c v a r i a t i o n (HSM1). The e f f e c t of spark gap s i z e on burning duration i s demonstrated i n Table 5.4 and F i g . 5.23, where burning duration i s plotted against spark gap si z e (with high s w i r l ) . The upper set of points correspond to the early burning duration; the lower to the main burning duration. There i s no s i g n i f i c a n t e f f e c t of spark gap s i z e on the mean value of the main-burning duration and only a s l i g h t decrease i n the early burning duration with increased spark gap. The two points at the two millimeter gap represent the HSM1 and HSM2 cases. 5.8 E f f e c t Of Spark Gap On C y c l i c V a r i a t i o n The r e s u l t s of t h i s experiment, as shown i n F i g . 5.24 and Table 5.5, indicates that c y c l i c v a r i a t i o n i s an inverse function of the spark gap s i z e . The two curves displayed i n F i g . 5.24 were derived from the repeated data for the High Swirl, Mid-radius, two millimeter spark gap case, the upper being the o r i g i n a l r e s u l t s with the high c y c l i c v a r i a t i o n ( i . e . , HSM1) and the lower being the repeated r e s u l t s ( i . e . , HSM2). 77 For a small gap of 0.5 mm, the c y c l i c v a r i a t i o n was quite su b s t a n t i a l . But when the gap was increased to two millimeters ( r e f e r r i n g to the repeat data (HSM2) and not the o r i g i n a l data), c y c l i c v a r i a t i o n dropped down by a factor greater than four. For the o r i g i n a l data (HSM1), the c y c l i c v a r i a t i o n only s l i g h t l y dropped from that for the 0.5 mm gap. A further increase i n gap si z e did not decrease the c y c l i c v a r i a t i o n when compared to the repeated case (HSM2). 5.8 E f f e c t Of Flame Holders An experiment was conducted to observe the e f f e c t s of d i f f e r e n t spark gap geometry on combustion duration and c y c l i c v a r i a t i o n . Various flame holders, as shown i n F i g . 3.11, were i n s t a l l e d on the mid-radius spark location probes. Schlieren films of the combustion process, as shown i n Figs. 5.25 and 5.26, indicated that the flame occasionally detached from the spark probe. I t was thought that i f the flame remained attached to the spark source, shorter combustion durations with le s s c y c l i c v a r i a t i o n would r e s u l t . F i g . 5.25 i s a sequence of sch l i e r e n pictures approximately one millisecond between frames of the bare probes. The number below each figure i s the approximate time i n milliseconds from i g n i t i o n . In these photographs, i g n i t i o n takes place between the r a d i a l probes that slant about twenty degrees from the v e r t i c a l and s w i r l i s i n the clockwise d i r e c t i o n . The v e r t i c a l wire i n these pictures l i e s along the quartz window and does not i n t e r f e r e with the f l u i d flow. F i g . 5.26 i s s i m i l a r to F i g . 5.25 except that the flame kernel 78 detaches from the spark source. This ser i e s of schlieren photograph has a d i f f e r e n t time i n t e r v a l and the entire chamber diameter i s not shown. Also, i g n i t i o n takes place between the v e r t i c a l probes and the probe twenty degrees from the v e r t i c a l l i e s along the quartz window. The r e s u l t s of the flame holder experiments are displayed i n Tables 5.6 and 5.7. The base condition, as mentioned i n Tables 5.6 and 5.7, was an unshielded probe i n the same po s i t i o n . In general, the main e f f e c t of the use of these flame holders was increased burning duration. There were a few exceptions to t h i s , one of which was the second flame holder at low s w i r l . In t h i s case, the burning duration was s l i g h t l y shorter than the base case. But at higher sw i r l rates, t h i s flame holder, l i k e the others, d i d not reduce burning duration. This same flame holder at low s w i r l also exhibited a tendency toward reduced c y c l i c v a r i a t i o n but once again, t h i s was only true for low swi r l and not for higher s w i r l rates. Figs. 5.27 and 5.28 show photographs of some the flame holders. F i g . 5.27 i s the screen flame number four (as shown i n F i g . 3.11) and Fi g . 5.28 i s the 6.35 mm diameter, number two flame holder. A l l the photographs (Figs. 5.25 through 5.28) were taken at a swi r l i n t e n s i t y of 4000 rpm. A l l the figures with show a tendency of the flame to elongate near the wall and t h i s may be due to the high j e t v e l o c i t y of the mixture i n t h i s zone. Tables 5.6 and 5.7 also show that two of the flame 79 holders had high incidence of m i s f i r e , p a r t i c u l a r l y at higher s w i r l rates. In general, the fastest-burning rates were achieved with the p l a i n , open gap with no flame holders. 80 6. CONCLUSIONS Swirl and spark location experiments performed with s w i r l i n g combustion i n a constant volume chamber lead to the following conclusions: 6.1 E f f e c t Of Swirl On Burning Duration And V a r i a t i o n Increasing s w i r l generally decreases combustion duration and increases burning rate. However, the amount of burning duration reduction decreases as swi r l i n t e n s i t y increases. Over a large range of swi r l i n t e n s i t i e s (quiescent conditions not included), the main burning duration ( t 1 0 - 9 o ) i s considerably shorter than the early burning duration ( t 1 0 ) . When s w i r l i s present, the main burning duration decreases s l i g h t l y as s w i r l i n t e n s i t y increases. The early burning duration i s a weak function of s w i r l i n t e n s i t y . As swi r l increases, early burning duration decreases. Present r e s u l t s with f i f t e e n samples do not indicate a strong e f f e c t of swi r l on the standard deviation of burning duration ( c y c l i c v a r i a t i o n ) . One case, where the standard deviation i n burning duration was very high, was not repeatable. This may point to a need f o r clo s e r control of i g n i t i o n parameters during such experiments. 6.2 E f f e c t of Spark Location On Burning Duration Spark l o c a t i o n has a strong e f f e c t on burning duration, but the e f f e c t i s dependent on s w i r l i n t e n s i t y . With no s w i r l , c e n t r al i g n i t i o n y i e l d s the shortest burning duration 8 1 (as would be expected since burning duration i n t h i s case i s only a function of flame t r a v e l distance). For low (2000 rpm) and medium s w i r l (4000 rpm), the outside l o c a t i o n r e s u l t s i n the shortest burning durations. At high s w i r l (6000 rpm), the mid-radius spark l o c a t i o n i s associated with the shortest burning duration. Relocation of the spark source from centre to outside can reduce the early burning duration by a factor of two for a l l s w i r l i n g i n i t i a l conditions. Main burning duration i s not affected by spark location i n the presence of s w i r l . An i g n i t i o n source located at the mid-radius end wall l o c a t i o n generally r e s u l t s i n increased early burning duration and equivalent main burning duration when compared to the mid-radius, mid-plane locat i o n . The reason may be the increased flame t r a v e l distance of the end wall loca t i o n . The trend toward reduced burning duration with increased s w i r l i n t e n s i t y , for central i g n i t i o n , may imply that central i g n i t i o n and s w i r l much greater than 6000 rpm would r e s u l t i n shortest burning duration when compared to other spark locations. With no s w i r l , the standard deviation of the early and main burning durations appear to increase as the spark source i s moved outward from the center. This conclusion i s tentative since the only eight samples were taken. 6.3 The E f f e c t of Spark Gap Spark gap v a r i a t i o n has no e f f e c t on the mean value of main burning duration. With increased gap, there i s a s l i g h t 82 reduction in early burning duration. Spark gap appears to have a strong effect on the standard deviation of burning variation. Increasing the spark gap from 0.5 to 3.5 mm reduced the standard deviation by a factor of four. 6.4 The Effect Of Flame Holders As verified by schlieren photography, the flame holders do hold the flames in the swirling flow f i e l d and unshielded probes occasionally shed the flames. However, flame attachment does not reduce the burning durations. Cyclic variation appears to increase with the application of flame holders. 6.5 Recommended Future Work i) The no-swirl valve should be tested to provide information on the separate effects of turbulence in the absence of swirl. The valve was designed to produce similar valve-induced turbulence as the swirl valve but without swirl. One problem i s that the swirling mixture also produces turbulence through shearing action against the walls. This turbulence cannot be generated by the no-swirl valve without bulk f l u i d motion. Thus the no-swirl valve i s only a step towards separating the effect of turbulence and swirl on burning duration. i i ) Further work is needed on the effects of swirl on cyclic variation. The apparatus used for this series of experiments could be used with a larger sample size i f testing were automated. 83 i i i ) F u r t h e r s c h l i e r e n p h o t o g r a p h y c o u l d r e v e a l more a b o u t f l a m e h o l d e r e f f e c t s a n d t h e c o n d i t i o n s u n d e r w h i c h f l a m e h o l d e r s i m p r o v e c o m b u s t i o n . The e f f e c t s o f s e c o n d a r y f l o w s a n d b u o y a n c y c o u l d a l s o b e f u r t h e r e x p l o r e d w i t h t h i s t e c h n i q u e . i v ) C a r e f u l v e l o c i t y m e a s u r e m e n t s w i t h b o t h s t e a d y a n d t r a n s i e n t f l o w a r e n e e d e d . The e f f e c t s o f t h e s p a r k p r o b e s a n d f l a m e h o l d e r s on d r a g a n d t u r b u l e n c e g e n e r a t i o n i s n o t w e l l known. F u r t h e r e x p l o r a t i o n i s a l s o n e e d e d on how p r o b e d r a g a f f e c t s t h e t r a n s i e n t d e c a y a n d t h e v e l o c i t y d i s t r i b u t i o n o f t h e s w i r l i n g f l o w f i e l d . V e l o c i t y d i s t r i b u t i o n c a n be c h a n g e d a n d o p t i m i z e d f o r b u r n i n g d u r a t i o n a n d v a r i a t i o n t o f i n d o u t w h a t i s t h e p r e f e r r e d s h a p e o f s w i r l . v ) T u r b u l e n c e m e a s u r e m e n t s a r e n e e d e d t o o b s e r v e t h e e f f e c t o f s w i r l o n t u r b u l e n c e g e n e r a t i o n , i n t e n s i t y , s c a l e , a n d d e c a y . v i ) V e r y h i g h s w i r l r a t e s ( o f a b o u t 10,000 t o 20,000 rpm) c o m b i n e d w i t h t h e e f f e c t s o f s p a r k l o c a t i o n , f l a m e h o l d e r s , a n d s e c o n d a r y f l o w s s h o u l d b e t e s t e d . One m i g h t f i n d some s u r p r i s i n g r e s u l t s s u c h a s c e n t r a l i g n i t i o n b e i n g t h e s p a r k l o c a t i o n r e s u l t i n g i n t h e s h o r t e s t b u r n i n g d u r a t i o n s . v i i ) F u r t h e r c o n s t a n t v o l u m e s w i r l c h a m b e r r e s e a r c h i s n e e d e d on t h e e f f e c t o f l e n g t h - t o - d i a m e t e r r a t i o ( i n p a r t i c u l a r , t h e e f f e c t o f s e c o n d a r y f l o w s o n c o m b u s t i o n d u r a t i o n , e t c . ) , o f f u e l - a i r r a t i o , o f p r e - i g n i t i o n t e m p e r a t u r e a n d p r e s s u r e , a n d o f s p a r k e n e r g y a n d o r i e n t a t i o n . I n f o r m a t i o n a b o u t t h e s e t o p i c s w o u l d be more e a s i l y o b t a i n e d 84 from constant volume chamber work and the r e s u l t i n g knowledge base could then be applied to further engine studies. v i i i ) I t would also be i n t e r e s t i n g to extend t h i s work to an engine study. One important question that remains unanswered i s what i s the id e a l s w i r l number (the r a t i o between s w i r l rpm and engine rpm). Other areas where more knowledge i s needed i n an engine study i s the e f f e c t of s w i r l on volumetric e f f i c i e n c y , f u e l - a i r r a t i o , emissions, compression r a t i o , and octane requirement. 85 1 V O L U M E F I G . 1 . 1 V A R I O U S H E A T - A D D I T I O N P R O C E S S E S (MATTAVI 1980) I L I I I I 4 8 1 2 1 6 2 0 C O M P R E S S I O N R A T I O . 1 . 2 E F F I C I E N C Y C O M P A R I S O N O F I D E A L E N G I N E C Y C L E S (MATTAVI 1980) DENSITY OF UNBURNED ZONE,p-FLAME AREA, Af FLAME BALL (BURNING OUTWARDS) FIG. 1.3 MASS ENTRAINMENT PARAMETERS 00 VANES TANGENTIAL PORT ENTRY F I G . 1.6 MAY " F I R E B A L L " H E A D (GRUPTA, e t . a l . , 1984) CHAMBER WALL FIG. 1.7 SECONDARY FLOWS LOWER HALF OF CHAMBER NOT SHOWN FOR CLARITY CHAMBER AXIS FORCE VECTOR PRIMARY FLUID FLOW SECONDARY FLIUD FLOW i rzzU c F I G . 3 . 1 A P P A R A T U S A S S E M B L Y D R A W I N G VO 93 F I G . 3.2 C O M B U S T I O N C H A M B E R C R O S S - S E C T I O N F I G . 3.3 S W I R L V A L V E F I G . 3 . 4 N O - S W I R L V A L V E 5 C I R C U L A T I O N F A N F I G . 3 . 6 F A N S E A L VO FIG. 3 . 7 INSTRUMENTATION / IGNITION / CAMERA 12 VOLT BATTERY 00 99 C R O S S - S E C T I O N OF CHAMBER F I G . 3 . 8 S P A R K G A P A R R A N G E M E N T P W R S W 1 2 V 1 2 V T O C D I S o T = l . l R C w h e r e , T i n ms R i n kohms C i n m i c r o f a r a d s 4005 T R I G 1 2 4 I O O U T 4 1 2 6 ^ — I 2 N 4 9 2 3 0 V F I G . 3 . 9 S C H E M A T I C O F D E L A Y T I M E R 101 IGNITE SW SET DELAY RESET SW O MANUAL SW O LED O DELAY OUT ( B N C ) TRIGGER TRIGGER SW 0 D © (CONN.) 12 VOLTS 0 VOLTS IN IN 1 2 VOLTS TO C D . PWR SW ON LED © © © ® @F I G . 3.10 DELAY TIMER SWITCH ARRANGEMENT 1 0 2 FLAME HOLDER NUMBER FOUR ALL DIMENSIONS ARE IN MILLIMETERS F I G . 3.11 FLAME HOLDER DESIGNS 103 ALL DIMENSIONS ARE IN MILLIMETERS F I G . 3.12 FLAME HOLDER ARRANGEMENTS LENS 30 DIA. ALL DIMENSIONS ARE IN CENTIMETERS NOT TO SCALE FIG. 3.13 SCHLIEREN PHOTOGRAPHIC ARRANGEMENT o 4^  105 9 0 ALL DIMENSIONS ARE IN MILLIMETERS. 3.14 PADDLE WHEELS 0 4 0 0 0 8 0 0 0 1 2 0 0 0 1 6 0 0 0 2 0 0 0 0 2 4 0 0 0 MOTOR/FAN RPM F I G . 3 . 1 5 C H A M B E R S W I R L R A T E V E R S U S C I R C U L A T I O N F A N S P E E D o FIG. 3.16 SWIRL DECAY CURVE INDICATED DELAY T I M E , t , inS F I G . 3.17 I G N I T I O N D E L A Y T I M E R C A L I B R A T I O N C U R V E o 00 F I G . 3 . 1 8 CHROMATOGRAPH CALIBRATION CURVE: METHANE (CH 4 ) o VO 1000 500 0 400 800 1200 1600 2000 TIME, t , S FIG. 3.19 SEAL PRESSURE RETENTION its cu x S to W a cu Z < > H a w TIME, t , n»S FIG. 3.20 APPARATUS VIBRATION 112 o o o o o o o in o io in o tn ' d ' 3 H f l S S 3 H d INTIVAIflOa f i — VIEW PORT r - FRAME ADJUST KNOB EVENT TIMER CONNECTION C A LASER LIGHT ENTERS HERE OSCILLATOR CONNECTION MOTOR FILM DRAG CONTROL SPEED SELECTION CAMERA ACTIVATION SWITCH BASE FIG. 4 . 1 HIGH-SPEED CAMERA 800-700-TIME, t, mS FIG. 5 .1 EXPERIMENTAL REPEATABILITY 0 . 5 0 0 . 4 0 H NUMBER OF TESTS FIG. 5.2 STANDARD DEVIATION VERSUS SAMPLE SIZE: CENTRAL IGN. 1 1 1 1 0 5 10 15 20 25 30 NUMBER OF TESTS FIG. 5 . 3 STANDARD DEVIATION VERSUS SAMPLE SIZE! OUTSIDE IGN. i—1 i— 1 800 T IME, t , mS F I G . 5.4 EFFECT OF SWIRL ON BURNING DURATION: OUTSIDE LOCATION 8 0 0 TIME, t , mS F I G . 5 . 5 EFFECT OF SWIRL ON BURNING DURATION: CENTRAL LOCATION 800 TIME, t , mS F I G . 5.6 EFFECT OF SWIRL ON BURNING DURATION: MID-RADIUS LOCATION g 30 z o cj w C O 2 O Q O z M z D CQ 25 20 - 15 10 FIG. 5.7 0 • A 2 0 0 0 O CENTRAL • MID-RADIUS A OUTSIDE 0 4 0 0 0 SWIRL INTENSITY, N , RPM E F F E C T OF SWIRL ON t 10 4 600 30 to o z o C J 5 2 5 i •J z o M D Q O z M z a D ffi 154) FIG. 5.8 O CENTRAL • MID-RADIUS A OUTSIDE 2 0 0 0 4 0 0 0 SWIRL INTENSITY, N , RPM E F F E C T OF SWIRL ON t 1 0 - 9 0 600 8 0 0 T IME, t , mS F I G . 5 . 9 EFFECT OF SPARK LOCATION ON BURNING DURATION ( 0 RPM) TIME, t , mS FIG. 5.10 EFFECT OF SPARK LOCATION ON BURNING DURATION (2000 RPM 800 0 10 20 30 40 TIME, t , mS F I G . 5.11 EFFECT OF SPARK LOCATION ON BURNING DURATION (4000 RPM) 800 0 5 10 15 20 TIME, t , mS F I G . 5.12 EFFECT OF SPARK LOCATION ON BURNING DURATION (6000 RPM) 800 TIME, t , mS F I G . 5 . 1 3 P R E S S U R E C U R V E S FOR C E N T R A L I G N I T I O N : 2 0 0 0 R P M to 800 TIME, t , mS FIG. 5.14 PRESSURE CURVES FOR C E N T R A L I G N I T I O N : 4 0 0 0 RPM 800 T TIME, t, mS F I G . 5 . 1 5 P R E S S U R E C U R V E S F O R C E N T R A L I G N I T I O N : 6 0 0 0 R P M i—1 to oo 800 TIME, t , mS F I G , 5 . 1 6 P R E S S U R E C U R V E S F O R O U T S I D E I G N I T I O N : 6 0 0 0 R P M 800 0 5 10 15 20 TIME, t , mS FIG. 5.17 PRESSURE CURVES FOR MID-RADIUS IGNITION: 6000 RPM (HSM1) 800 T IME, t , mS FIG. 5.18 PRESSURE CURVES FOR MID-RADIUS IGNITION: 6000 RPM (HSM2) 800 7 0 0 -600 500 400 3 0 0 -200 100 O SD t ~ 10 • 0 [ c ~ ] — i — i — i — i — r MID-RADIUS 1 1 1 1 1 — OUTS i — i — i — i — i — i — i — r CENTRAL SPARK LOCATION, r/R. IG. 5.19 EFFECT OF SPARK LOCATION ON STANDARD DEVIATION: 0 RPM CO o z o C J w tf) o CJ a 1/3 z o »-l H < > w c Q < o z < H CO 800 700-600 500 400-300 200-100-Q S D tio-go O SD E -10 0 • 0 c i—r CENTRAL T 1 1 1 1 1 1 1 1 1 1 T MID-RADIUS "i—r i — i — OUTSIDE SPARK LOCATION, r / R . F I G . 5 .20 EFFECT OF SPARK LOCATION ON STANDARD DEVIATION: 2000 RPM 8 0 0 7 0 0 6 0 0 5 0 0 H 4 0 0 H 100 3 0 0 - K 200H T — i — i — i — | — i — r MID-RADIUS i 1 r OUTSIDE SPARK LOCATION, r/R. FIG. 5.21 EFFECT OF SPARK LOCATION ON STANDARD DEVIATION: 4000 RPM 800 700-600 T 1 r C E N T R A L 1 1 1 1 r M I D - R A D I U S i r O U T S I D E S P A R K L O C A T I O N , r / R . F I G . 5 . 2 2 E F F E C T O F S P A R K L O C A T I O N ON S T A N D A R D D E V I A T I O N : 6 0 0 0 RPM to o o CJ w CO • J 2 o g g 2 2 8 30 25 -20 -15-10 5-• HSM2 — o • t l o O ^ 0 - 9 0 / / N N [ HSM1 0 . 0 0 . 5 1 .0 1 .5 2 . 0 2 . 5 ( 3.0 3 . 5 F I G . 5.23 • 1200 to D 2 O U IOOO-w o (X u £ 800 I © in l-P Q 10 600-O 400-1 M < w 200-| Q a < o s CO 0 . 0 F I G . 5.24 SPARK G A P , G , MILLIMETERS E F F E C T OF SPARK GAP ON BURNING DURATION O 0 - HSM1 HSM2 0 . 5 1 .0 1 .5 2 . 0 2 . 5 3.0 3 . 5 SPARK GAP, G , MILLIMETERS E F F E C T OF SPARK GAP ON CYCLIC VARIATION BRIDGE ON/OFF GAS EXHAUST PORT ZERO ADJUSTMENT — ROTAMETERS TEMPERATURE ^ ® ® ® ® ® © ® ® ® ® ® • • • • • • ® ® ® ® o o o o COLOMN A O B O • ® ® ® f® ® " " " \® ® ® THERMAL CONDUCTIVITY ^ ® ® \ ) ® PROGRAMMER OVEN F I G . A . l GAS CHROMATOGRAPH MOSLEY RECORDER (INTEGRATOR NOT SHOWN) 8 AREA PERCENT F I G . A . 2 CHROMATOGRAPH CALIBRATION CURVE: ETHANE (C 2 Hg) to AREA PERCENT F I G . A . 3 CHROMATOGRAPH CALIBRATION CURVE: PROPANE (C 3 Hg) -400 C H A R G E , Q , p C FIG. C . l PRESSURE TRANSDUCER CALIBRATION CURVE Table 4.1 EQUIPMENT SWITCH SETTINGS 1) Oscilloscope TRIGGER: EXT, D.C, negative slope, NORM, PRE-TRIG = ZERO. TIME PER POINT: 5 to 20 microseconds. CHANNEL A: OFF CHANNEL B: 4 v o l t s per h a l f scale, D.C, input connected to (+), zero set at -4 v o l t s . STORAGE: STORE and LIVE turned ON. EXPANSION: AUTO CENTER ON, DISPLAY = Y/T, HORIZONTAL AND VERTICAL are turned OFF. NUMERICS: GRID and RESET are turned OFF. FUNCTION: DATA MOVE and SUBTRACT are turned OFF. Internal switch settings for RS-232c: closed open 1 X 2 X 3 X 4 X 5 X 6 X 7 X 8 X Depending on combustion duration. The minimum time which displayed the pressure r i s e to peak pressure was chosen. ** i . e . , the pressure trace started at the bottom l e f t corner of the screen. 2) Charge Amplifier Transducer S e n s i t i v i t y : 1.32 pC or mV/per Mechanical Unit. Transducer S e n s i t i v i t y Range: 10-110, scale = 1 Mechanical Unit Per Volt . Mode: Charge Mode and Ca l i b r a t i o n . Time Constant: medium. Zero Set with oscilloscope. Reset lever i n OPERATE po s i t i o n for experiments. F i l t e r : f c = 180 Hz. 1 4 6 T a b l e 4.2 TEST OF " R i a h i " B u r n i n g D u r a t i o n  ^ „ t 0 - 1 0 t 0 - 5 0 t 0 - 9 0 t 1 0 - 9 0 T e s t 1 Man. 8.429 0 > 2 % 11.62 Q < 1 7 % 13.33 0 > 4 5 % 4.900 0 > 5 % Comp. 8.446 11.64 13.39 4.925 T e s t 2 Man. 12.45 0 > 3 % 15.37 0 # 1 9 % 17.21 0 < 5 8 % 4.753 1 - 5 % Comp. 12.49 15.40 17.31 4.824 T e s t 3 Man. 11.96 0 > 3 3 % 14.84 n < 3 3 % 16.54 Q > 6 7 % 4.583 1 > 3 % Comp. 12.00 14.89 16.65 4.643 T e s t 4 Man. 11.77 0 > 5 1 % 14.57 Q > 4 % 16.54 0 > 7 9 % 4.583 0 > 2 % Comp. 11.83 14.63 16.41 4.573 S t a n d a r d D e v i a t i o n T e s t 1 S D t Q _ 1 0 S D t 0 _ 5 Q S D t Q _ 9 0 S D t 1 0 _ g o Man. 0.297 3 < ? % 0.306 1 - 6 % 0.326 4 . 2 9 % 0.185 Q > 5 % Comp. 0.308 0.311 0.340 0.184 T e s t 2 Man. 0.420 5 > 2 % 0.431 0 . 2 3 % 0.462 2 > 4 % 0.161 4 > 9 % Comp 0.398 0.430 0.451 0.153 T e s t 3. Man. 0.388 0 > 4 9 % 0.521 2 > 6 9 % 0.529 0 > 7 6 % 0.192 6 > 3 % Comp. 0.407 0.535 0.533 0.180 T e s t 4 Man. Comp. 0.186 '-' 0.269 ' 0.318 — ^ " 0.209 0.193 3 < 6 % 0.287 6 > 3 % 0.316 Q < 6 % 0.229 8 > ? < (Man. = m a n u a l c a l c u l a t i o n . ) (Comp. = c o m p u t e r c o m p u t a t i o n . ) ( U n d e r l i n e d numbers r e p r e s e n t p e r c e n t d i f f e r e n c e b e t w e e n m a n u a l c a l c u l a t i o n a n d c o m p u t e r c o m p u t a t i o n . ) 147 Table 5.1 SUMMARY OF BURNING DURATIONS FOR MID-AXIS SPARK LOCATIONS t10 NSC 14.4 NSM 13.9 NSO 15. 8 LSC 14.5 LSM 10.8 LSO 8.8 MSC 13.2 MSM 8.5 MSO 7.7 HSC 11.8 HSM1 7.8 HSO 8.4 HSM2* 8.1 t50 NSC 23.7 NSM 23.3 NSO 30.0 LSC 20.6 LSM 15.8 LSO 13.0 MSC 17.5 MSM 11. 6 MSO 11.1 HSC 14.6 HSM1 10.1 HSO 11.6 HSM2 10.4 ^ 0 NSC 29.4 NSM 33.3 NSO 42.4 LSC 22.6 LSM 18.3 LSO 15.8 MSC 19.1 MSM 13.7 MSO 13 . 2 HSC 16.4 HSM1 11. 8 HSO 13 .4 HSM2* 12.1 t10 -90 NSC 15. 0 NSM 19.4 NSO 26.6 LSC 8.1 LSM 7.5 LSO 7.0 MSC 5.9 MSM 5.2 MSO 5.4 HSC 4.6 HSM1 4.1 HSO 4 . 9 HSM2* , 4.0 ( A l l numbers represent time i n milliseconds.) (Underlined numbers represent shortest burning durations for given burning time type and swi r l l e v e l , eg., for ^10-90 and no-swirl (NS), have 15.0.) repeated data with same i n i t i a l conditions as HSM1 Table 5.2 SUMMARY OF STANDARD DEVIATIONS  FOR MID-AXIS SPARK LOCATIONS SD t 1 Q NSC 90 LSC 160 MSC 280 HSC 190 NSM 160 LSM 360 MSM 180 HSM1 740 HSM2* 240 NSO 260 LSO 220 MSO 500 HSO 310 SD t 5 0 NSC 170 LSC 270 MSC 280 HSC 280 NSM 410 LSM 4 50 MSM 220 HSM1 970 HSM2 260 NSO 390 LSO 290 MSO 650 HSO 310 SD t 9 Q NSC 240 LSC 240 MSC 380 HSC 330 NSM 600 LSM 400 MSM 250 HSM1 920 HSM2* 320 NSO 570 LSO 380 MSO 560 HSO 340 a u u10-90 NSC 170 LSC 150 MSC 260 HSC 220 NSM 52 0 LSM 190 MSM 190 HSM1 280 HSM2 3 00 NSO 320 LSO 350 MSO 160 HSO 180 ( A l l numbers represent time i n microseconds.) (Underlined numbers represent l e a s t standard deviation for a given standard deviation and sw i r l l e v e l , eg., for SD t-, 0_ gp and no swi r l (NS) , NSC i s le a s t at 170.) repeated data with same i n i t i a l conditions as HSM1 149 Table 5.3 COMPARISON OF BURNING DURATIONS FOR MID-PLANE AND END WALL SPARK LOCATIONS -10 -50 -90 -10-90 No Swirl (0 RPM) Base End Wall 13.9 17.8 23.3 28.3 33.3 36.9 19.4 19.1 Low Swirl (2000 RPM) Base 10.8 15.8 18.3 7.5 End Wall 13.9 18.9 21.3 7.4 Medium Swirl (4000 RPM) Base 8.5 11.6 13.7 5.2 End Wall 11.4 14.7 16.4 5.0 High Swirl (6000 RPM) Base 7.8 10.1 11.8 4.1 Base2 8.1 10.4 12.1 4.0 End Wall 10.4 13.4 15.1 4.7 ( A l l numbers represent time i n milliseconds.) (Underlined numbers represent shortest duration case for a given type of burning time and swi r l l e v e l , eg, t 1 0 and High-Swirl has a time of 7.8.) ("Base" i s mid-plane and mid-radius.) ^"End Wall" i s 2.0 mm from end wall and mid-radius.) Repeated t e s t of same i n i t i a l conditions as HSM1. 150 Table 5.4 EFFECT OF SPARK GAP ON BURNING DURATION GAP t 10 t 50 t 90 t 10-90 0.5.mm 2.0.mm^  2.0 mm 3.5.mm 9.3 7.8 8.1 7.9 11.9 10.1 10.4 10.5 13.4 11.8 12.1 12 .1 4.1 4.1 4.0 4.2 (Numbers represent time i n milliseconds.) ( I n i t i a l conditions are mid-radius, mid-axis i g n i t i o n l o c a t i o n with high swirl.) (Underlined numbers represent shortest burning time for a given type of burning duration, eg., for t 1 Q the 2.0 mm gap y i e l d s the lea s t burning time at 7.8.) repeated t e s t of same i n i t i a l conditions as HSM1 Table 5.5 EFFECT OF SPARK GAP ON CYCLIC VARIATION GAP SD t 10 SD t 50 SD t 90 SD t 10-9 0.5.mm 2.0.mm 2.0.mm 3.5.mm 740 740 240 170 1060 970  260  280 1070 920 320 430 370 280 300 310 (Numbers represent time i n microseconds.) ( I n i t i a l conditions are mid-radius, mid-axis i g n i t i o n l o c a t i o n with high swirl.) (Underlined numbers are those that define c y c l i c variation.) * . repeated t e s t of same i n i t i a l conditions as HSM1 152 Table 5.6 EFFECT OF "FLAME HOLDERS" ON BURNING DURATION -10 Low Swirl (2000 RPM) -50 -90 -10-90 Base FH1 FH2 FH3 FH4 10.8 10.8 9.8 10. 0 10.4 15.8 15.9 14.8 15.0 15.3 18. 3 18. 6 17. 6 17.8 18.1 7.5 7.8 7.9 7.8 7.7 Medium Swirl (4000 RPM) Base FH1 FH2 FH3 FH4 8.5 8.6 9.3 8.9 11.6 11.6 12.4 11.9 13.7 13.6 14.2 13.8 5.2 5.0 4.9 4.9 High Swirl (6000 RPM) Base Base2 FH1 FH2 FH3 FH4 ** 7.8 8.1 9.9 10. 0 10.1 10.4 12.7 12.7 11.8 12 .1 14.2 14.3 4.1 4.0 4 . 3 4.3 ( A l l numbers represent time i n milliseconds.) (Base i s mid-plane and mid-radius.) (Underlined numbers represent shortest time for a) given type of burning duration and s w i r l l e v e l , eg., for t-^ Q and High Swirl, shortest time i s base at 7.8.) More than f i v e m i s s f i r e s . Repeated t e s t of same i n i t i a l conditions as HSM1. 153 Table 5.7 EFFECT OF "FLAME HOLDERS" ON CYCLIC VARIATION SD t 1 ( ) SD t 5 Q SD t g o SD t 1 0 _ 9 0 Low Swirl (2000 RPM) Base FH1 FH2 FH3 FH4 360 270 190 330 320 450 500 430 540 700 400 500 380 470 640 190 380 340 230 360 Medium Swirl (4000 RPM) Base FH1 FH2 FH3 FH4 180 440 410 400 220 440 480 530 250 570 670 710 190 320 440 420 High Swirl (6000 RPM) Base Base** FH1 FH2 FH3 FH4 740 240 1120 1660 970 260 1340 1860 920 320 1400 1800 280 300 430 300 ( A l l numbers represent time i n microseconds.) (Base i s mid-plane and mid-radius.) (Underlined numbers represent le a s t c y c l i c v a r i a t i o n for a given type of c y c l i c v a r i a t i o n and s w i r l number, eg., SD t 1 0 and High Swirl i s le a s t for base** case at 240.) More than f i v e m i s s f i r e s . Repeated t e s t of same i n i t i a l conditions AS HSM1. Table A.l OVEN TEMPERATURES AND RETENTION TIMES Component Temperature Approximate Time CH4 35°C 4 minutes C 2H 2 130°C 6 minutes C 3H 8 130°C 8 minutes Table C.1 RANGE CAPACITOR VALUES 155 "Transd. Sens. Range" 0.1-1.1 1-11 10-110 100-1.Ik l k - I l k LOk-nok Range 1,2,5 * 10 4 10 3 10 2 10 1 1 0 _ 1 mech. 1,2,5 * 10 3 10 2 10 1 IO" 1 10~ 2 units 1,2,5 * 10 2 10 1 IO" 1 10" 2 10~ 3 per v o l t 1,2,5 * 10 1 IO" 1 10" 2 10~ 3 IO" 4 These columns give the value of Cg 156 BIBLIOGRAPHY Damkohler, G. Z. Electroicheinie angewandte phys. chem.. 1940, 46, p. 601. (English t r a n s l a t i o n , NACA TM 1112 1947.) Dyer, T. M. Characterization of one- and two-dimensional homogeneous combustion phenomena i n a constant volume bomb. Society of Automotive Engineers, 1979, paper S. A. E. 790353. Groff, E. G. & Sinnamon, J . F. The ef f e c t s of i g n i t i o n l o c a t i o n i n a sw i r l f i e l d on homogeneous-charge combustion. Society of Automotive Engineers, 1982, paper S. A. E. 821221. Grupta, A. K., L i l l e y , D. G., & Syred, N. Swirl flows. Tunbridge Wells, England: Abacus Press, 1984. Hanson, R. J . & Thomas, A. Flame development i n s w i r l i n g flows i n closed vessels. The Combustion I n s t i t i u t e , 1984, p. 257-279. Hempson, J . G. G. The automobile engine 1920-1950. Society of Automotive Engineers. 1976, paper S. A. E. 760605. Hung, J . The e f f e c t s of propane or ethane additives on laminar burning v e l o c i t y of methane-air mixtures. Ma. A. Sc. Thesis, Vancouver: The University of B r i t i s h Columbia, 1986. Inoue, T., Nakanishi, K., Noguchi, H., & Iguchi, S. The r o l e of swi r l i n combustion of the SI engine. Hamberg: FISTA 18th International Congress, VDl-Berichte Nr. 370, 1980. Lewis, B. & von Elbe, G. Combustion, flames, and explosions of gases. New York: Academic Press, 1961. Mattavi, James N. The att r i b u t e s of fas t burning rates i n engines. Society of Automotive Engineers, 1980, paper S. A. E. 800920. Mayo, J . The e f f e c t s of engine design parameters on combustion rate i n spark-ignited engines. Society of Automotive Engineers, 1975, paper S. A. E. 750355. Nagao, A. & Tanaka, K. The e f f e c t of s w i r l control on combustion improvement of spark i g n i t i o n engine. I n s t i t u t e of Mechanical Engineers, 1983, paper IMech E C54/83. Nagayama, I., Araki, Y., & Iioka, Y. E f f e c t s of s w i r l and squish on S.I. engine combustion and emission. Society  of Automotive Engineers, 1977, paper S. A. E. 770217. Wakuri, Y., Kido, H., Ono, S., Nakashima, K., & Murase, E. Influences of s w i r l and turbulence on the burning v e l o c i t y i n an engine cylinder. B u l l i t i n of the Japan  Society of Mechanical Engineers, 1981, 24., No. 188. Witze, P. 0. & V i l c h i s , F. R. Stroboscopic l a s e r shadowgraph study of the e f f e c t of s w i r l on homogeneous combustion i n a spark-ignited engine. Society of Automotive  Engineers, 1981, paper S. A. E. 810226. Witze, P. 0. The e f f e c t of spark location on combustion i n a va r i a b l e s w i r l engine. Society of Mechanical Engineers, 1982, paper S. A. E. 820044. 158 Zawadzki, A. & J a r o s i n s k i , J . Laminarization of flames i n ro t a t i n g flow. Combustion Science and Technology, 1983 35, p. 1-13. 159 Appendix A Gas Chromatograph Operation The gas chromatograph used for mixture strength measurement throughout t h i s work i s a Hewlett Packard model 5750b Research Gas Chromatograph. I t i s located i n the Environmental Engineering Laboratory i n the C i v i l Engineering department under the care of Ms. Paula Parkinson who supplied h e l p f u l advice. The following i s a simple outline on how the chromatograph works. The unknown sample, injected into a port i n the machine, enters a t h i n , long tube c a l l e d a column. This column separates the hydrocarbons and a i r . According to Hung (1986, p. 94): Each component i s i d e n t i f i e d by the retention time, time for the component to go through the column, which i s a function of column temperature. The component then goes through a wheatstone bridge detector. The change i n resistance due to the presence of the component produces a peak on the recorder. The area under t h i s peak i s proportional to the volume injected. Thus, volume percent can be determined from the area percent. This was the procedure for mixture analysis: 1. C a r r i e r gas flow i n the oven as shown i n F i g . A . l was v e r i f i e d with the soap bubble detector. The detector was connected to the exhaust port located on the upper l e f t hand side of the oven. The rubber bulb of the detector was squeezed a few times u n t i l a bubble was caught by the c a r r i e r gas. There are two circumferential l i n e s on the glass tube d i r e c t l y above the rubber bulb which represent a volume of 3 ml. The flow rate was adjusted so that a bubble would pass 160 between these two lines in approximately ten seconds, i.e., about 20 ml in one minute. When testing methane mixtures, this flow rate may have been too fast, resulting in short retention times. In this case, the carrier gas flow rate was reduced. Short retention times were indicated by the overlapping of consecutive output peaks or by RT values separated by less than about 0.8 minutes. 2. The bridge current in machine #2, the oven, and the other electronic panels were turned ON and allowed at least three hours of warm up time. Normally the chromatograph was always l e f t ON so that one only had to check i f there was s t i l l power to the various parts. 3. Since some higher order gases take more time through the column, oven temperatures were set according to Table A.l. These temperatures were not c r i t i c a l and only affected the time for the gases to pass through the chromatograph. Recall that i t i s the integrals of the output of the machine that are used for measurements and not directly the flow-through time. If the retention time for methane was s t i l l too short, the oven doors were opened and the oven fan turn ON. This reduced the oven temperature from the closed-oven-door minimum temperature of 50°C to room temperature. 4. A mixture sample was prepared in a v i a l . F i r s t the v i a l was evacuated and flushed with the unknown mixture a few times. For accurate measurements, the bottled unknown mixture had to be completely homogeneous by remaining untouched for a few days. The f i n a l pressure i n the v i a l was l e f t higher than atmospheric so that i f there was any leakage through the v i a l ' s rubber seal, the mixture would leak outward and not be d i l u t e d by ambient atmosphere. 5. The integrator power was turned ON and the ZERO adjusted at the bridge. Recently, a shorter column was i n s t a l l e d i n the chromatograph r e s u l t i n g i n increased zero s e n s i t i v i t y . The mixture measurement r e s u l t s do not seem to depend too much on the zero se t t i n g since i t was the i n t e g r a l area comparison of the two gases that was measured. The chart was set to AUTO and the STOP TIMER to a few more minutes than the time l i s t e d i n Table A . l . 6. Next, a one m i l l i l i t e r sampling syringe was flushed a few times with some of the unknown r a t i o mixture. Precisely one m i l l i l i t e r of mixture at atmospheric pressure was injected i n the B column i n the oven. 7.Immediately a f t e r i n j e c t i o n , the START/STOP button was pressed on the integrator. 8. The r e s u l t i n g d i g i t a l area percents were converted to volume percents with the c a l i b r a t i o n curves found i n Figs. 3.18, A. 2, and A. 3. The same sample was tested a few times u n t i l the operator was s a t i s f i e d that the r e s u l t s were consistent and accurate. While the r e s u l t s fluctuated, the r e s u l t s i n d i c a t i n g the r i c h e s t mixture were considered most r e l i a b l e since conceivably only a i r from the ambient atmosphere and not f u e l could leak into the t e s t sample. 9. When finishe d , the power to the integrator was turned 1 6 2 OFF. I f the operator was not c a r e f u l , the sample contained i n the syringe may not have been representative of the mixture i n the v i a l . But with c a r e f u l syringe technique, error could be minimized. The syringe used was a one m i l l i l i t r e syringe. Using a larger syringe would have resulted i n a l e s s accurate sample s i z e . I t was important that the syringe be flushed out a few times before the f i n a l sample was taken. When the sample was taken, more than one m i l l i l i t r e was drawn into the syringe. The excess mixture was slowly pushed out from the syringe when the needle was removed from the v i a l . This ensured that the syringe and needle only contained unknown mixture with no d i l u t i n g a i r and that the pressure i n the syringe was atmospheric without d i l u t i n g a i r entering to compensate the pressure. The sample was taken j u s t before i t was needed so that there was l i t t l e time for d i l u t i o n within the syringe needle. When the sample was injected into the gas chromatograph, the syringe plunger pressed with a smooth and continuous manner, i . e . , the plunger was not stopped and started again. The i n j e c t i o n was made f a i r l y quick and immediately a f t e r the plunger stroke was completed, the integrator was started. I f the r e s u l t s showed inconsistencies l i k e too many retention time peaks, or very small retention times, the syringe technique was at f a u l t . For further information, consult: McNair, H. M. & B o n e l l i , E.J. Basic gas chromatography. 1 6 3 Georgetown, Canada: Varian Aerograph, 1968. (This book i s available i n the Environmental Engineering Laboratory i n the C i v i l Engineering Department.) 164 Appendix B MIXTURE PREPARATION CALCULATIONS For methane and a i r i n a chemically balanced r a t i o , the reaction equation i s : CH 4 + 2 0 2 —> 1 C0 2 + 2 H 20 and the t o t a l container pressure i s the sum of the reactant p a r t i a l pressures: 1 mole CH 4 + 2*( 1 + 3.76 ) moles of a i r = P^ot Thus the p a r t i a l pressure r a t i o n of methane to a i r i s : £meth = i = — 1 — P t o t 1 + 2*( 1 + 3.76 ) 10.52 Bottle pressure may not exceed 17.3 MPa so precombustion pressure ( i . e . , the pressure before accidental combustion) i n the b o t t l e may not exceed 17.3 / 7 = 2.47 MPa. 2.00 MPa was chosen f o r P m a x . The p a r t i a l pressure of methane i n the b o t t l e was: 2000 = 190 kPa 10.52 or: 190 * 7.5006 = 1426 mm of Hg abs. since: 1 kPa = 7.5006 mm of Hg @ 0°C. The evacuated b o t t l e was f i l l e d with methane to 190 kPa or 1426 mm of Hg abs. Bottle temperature was allowed to return to ambient and then the b o t t l e was f i l l e d with dry a i r to 2000 kPa. The mixture was analyzed and the process i t e r a t e d u n t i l the desired r a t i o and accuracy was achieved. 165 A p p e n d i x C P r e s s u r e T r a n s d u c e r C a l i b r a t i o n The i n d e n t e d p a r a g r a p h s b e l o w w e r e o b t a i n e d f r o m t h e m a i n t e n a n c e m a n u a l f o r t h e R i c a r d o H y d r a s i n g l e c y l i n d e r r e s e a r c h e n g i n e . I n t r o d u c t i o n P i e z o - e l e c t r i c p r e s s u r e t r a n s d u c e r s h a v e b e e n i n u s e f o r some y e a r s a s a m e t h o d o f m e a s u r i n g r e c e n t l y t h a t t h e y h a v e b e e n a c c e p t e d a s a m e t h o d o f m e a s u r e m e n t whose a c c u r a c y a p p r o a c h e s t h a t o f t h e w e l l e s t a b l i s h e d b a l a n c e d d i s c i n d i c a t o r . A c u r r e n t e x p e r i m e n t a l p r o j e c t (2) r e q u i r e d a h i g h a c c u r a c y p r e s s u r e s i g n a l , a n d , t h e r e f o r e , a p r o c e d u r e f o r c a l i b r a t i n g t h e s e t r a n s d u c e r s was d e v i s e d , a n d i s d e s c r i b e d i n t h i s n o t e . I n o r d e r t o a c h i e v e h i g h a c c u r a c y w i t h t h e s e t r a n s d u c e r s i t i s n e c e s s a r y t o c l e a n a n d r e c a l i b r a t e t h e m a t f r e q u e n t i n t e r v a l s . A l t h o u g h t h e t r a n s d u c e r s h a v e o n l y d y n a m i c r e s p o n s e a n d c a n n o t , t h e r e f o r e , be c a l i b r a t e d s t a t i c a l l y , i t i s i n f a c t p o s s i b l e t o c a l i b r a t e t h e m q u a s i - s t a t i c a l l y on a d e a d w e i g h t t e s t e r p r o v i d e d s u i t a b l e p r e c a u t i o n s a r e t a k e n . The m e t h o d o f c a l i b r a t i o n , u s e d w h e r e a c c u r a t e m e a s u r e m e n t s a r e r e q u i r e d , i s d e s c r i b e d i n t h i s n o t e ; f o r l o w a c c u r a c y w o r k l e s s p r e c a u t i o n s n e e d t o b e t a k e n . c y l i n d e r p r e s s u r e s ( 1 ) . H o w e v e r , i t i s o n l y 1 6 6 Method of Ca l i b r a t i o n On receipt of a new transducer i t i s desirable to c a l i b r a t e i t before use, because several instances have been found when the new c a l i b r a t i o n d i d not agree exactly with the manufacture's figures. Thereafter the transducer should be recalibrated, cleaned and re c a l i b r a t e d again at in t e r v a l s of f i f t e e n hours. I f q u a l i t a t i v e pressure measurements only are required, the amplifier used can be a simple uncalibrated FET amplifier such as the Cussons P4543, but i f accurate quantitative r e s u l t s are required then a more sophisticated uni t must be used, such as the K i s t l e r 5001. This amplifier has a very high input resistance (1014 ohms) when set to the long time constant, and i s therefore suitable f o r q u a s i - s t a t i c measurements provided the leakage i n the transducer and leads i s kept low. This can be achieved by soaking the transducer i n benzyl alcohol and baking i n an oven at 150°C f o r an hour, and ensuring the leads are clean and dry. The K i s t l e r 5001 amplifier i s designed to be simple to use, and the philosophy i s successful provided that the user does not mind what siz e the output signal i s between zero and the amplifier saturation l e v e l at ten v o l t s . However, i f some other l i m i t a t i o n i s set (such as the Histomat A.D.C. 167 saturation l i m i t s at + two volts) then the following equation w i l l be found useful : Q v = ( i ) Cg * k where; V = amplifier output i n v o l t s , Cg = value of range capacitor i n pF, k = s e t t i n g of ten turn potentiometer on the amplifier expressed as a part of unity (e.g. .808), and Q = charge produced by transducer, pC. The manufactures recommend s e t t i n g the ten turn potentiometer to the value of the transducer s e n s i t i v i t y , but use of the above rule obviates t h i s when required. The value of the range capacitor i s set by the "Range" switch i n combination with the "Transd. Sens. Range" indicator and the values are given i n Table C.1. To use Table C.1 the procedure i s as follows: a) Set up the controls "Range" and "Transd. Sens." to give the required signal s i z e . b) Find the column corresponding to the p o s i t i o n of the "Transd. Sens. Range" Indicator, and the l i n e corresponding to the s e t t i n g on the "Range" switch. The value of the range capacitor i n pF i s now given on the same l i n e i n the two l e f t hand columns. 1 6 8 As an example, i f the "Transd. Sens. Range" indicator i s set at 1-11 and the "Range" switch at 500 mech. units/volt, then the value of the range capacitor i s 5000 pF. The procedure for calibration i s as follows: 1) Soak and bake the transducer as described above, unless the transducer has been used and the calibration i s required in i t s dirty state. 2) Check the leads and clean i f necessary (note: the recommended cleaner i s Freon TF solvent MS-180 by the Miller Stephenson Chemical Co. Inc.); the leads and amplifier should be the same as those used in the installation (small variations have been found between different 5001 amplifiers), and must be of the special low-static variety. 3) Assemble the transducer onto the inner sleeve and assemble the inner sleeve into the outer sleeve. To avoid the necessity of removing the outer sleeve from the engine, a spare outer sleeve may be purchased for this purpose. Install the transducer in i t s sleeve in the deadweight tester, tighten to normal torque, connect up a l l leads, switch on and leave the amplifier and D.V.M. to warm up for at least one hour. The leads should be arranged so that they cannot be disturbed during calibration. 4) Switch the amplifier to "reset", i f necessary, adjust the zero by removing the case to expose the 169 adjusting screw. Set the time constant to "long". With ambient pressure applying at the transducer, switch the amplifier 'reset' hold for a few seconds, and switch to operate. Apply the maximum required pressure to the transducer, and adjust the range switch and the "transd. sens." potentiometer to give an output voltage of just under ten v o l t s , or al t e r n a t e l y set the amplifier controls to the settings normally used i n the i n s t a l l a t i o n . The output voltage should now be steady, i f not, return to stage (1) . The range switch and "transd. sens." potentiometer should not be readjusted again during the c a l i b r a t i o n . 5) Unload the transducer, reset again and reload back to the peak pressure for t h i r t y seconds. I f the reading has changed by less than 0.5% i n t h i s time, the the leakage rate i s s u f f i c i e n t l y low to make a qu a s i - s t a t i c c a l i b r a t i o n of the transducer, otherwise, return to stage (1). Return to ambient pressure. I f the signal does not return to within 0.5% of the peak signal from the zero, return to stage (1) . 6) Dividing the pressure range into about ten segments, load the transducer to each pressure, returning i n between to ambient pressure and re-se t t i n g the amplifier. A p l o t i s made of the charge produced, i n pC, verses pressure as t h i s i s done, 170 c a l c u l a t i n g from equation (1). 7) I f the pl o t i s a reasonably s t r a i g h t l i n e ( i . e . within + 1%) draw the best s t r a i g h t l i n e f i t to the points and the transducer s e n s i t i v i t y i s the gradient of the l i n e . I f not, the transducer should be re-soaked, baked and ca l i b r a t e d again. I f the repeat i s s t i l l not straight, the transducer i s fa u l t y and should be returned to the manufacturer. 8) I f there i s a l i m i t a t i o n on the s i z e of the output signal when i n use, then the amplifier controls "Range" and "Transd. Sens." can be used to adjust the output signal to the desired magnitude on the oscilloscope, and knowing the transducer s e n s i t i v i t y equation (1) can be used to calcu l a t e the s i z e of the pressure s i g n a l . I f there i s no such l i m i t a t i o n then the amplifier can be used i n the way i t was designed to be, which i s that the "Transd. Sens." control i s set to the value of the transducer s e n s i t i v i t y , and the range switch then gives the magnitude of the output i n mechanical units per v o l t (the mechanical units being the same as those i n the c a l i b r a t i o n figure; for example, i f the c a l i b r a t i o n i s 14.1 pC/bar, then the output shown w i l l be i n ba r / v o l t ) . Conclusion By regular and car e f u l c a l i b r a t i o n , the better p i e z o - e l e c t r i c pressure transducers can be expected 171 to give r e s u l t s with and accuracy of better than + 2%, not including the error i n e s t a b l i s h i n g the absolute pressure l e v e l . Lancaster Kreiger and Lienesch (3) give r e s u l t s showing the e f f e c t of both pressure and phasing errors on the accuracy of the various types of c a l c u l a t i o n normally c a r r i e d out with cylinder pressure diagrams. These show that such pressure accuracy i s acceptable under most circumstances. I t i s recommended that the method of c a l i b r a t i o n described i n t h i s note be used for a l l p i e z o - e l e c t r i c transducers c a l i b r a t i o n s c a r r i e d out here i n the future. References 1) DP18797 "Piezo E l e c t r i c Pressure Transducers for Engine Indicating" J.G.G. Hempson and F.A. Putland, Oct. 1974 2) DP20744 "The I n i t i a l Report of a study of the e f f e c t s of varying the Rate of Burning of the charge i n a Petrol Engine" R.H. Thring The above procedure was followed as c l o s e l y as possible but the transducer never had the low decay rate that the above procedure s p e c i f i e d . The author s p e c i f i e d a decay rate no le s s than 0.5 percent drop i n t h i r t y seconds. The pressure transducer, used i n t h i s project, had a decay rate of about 0.5 percent i n twenty one seconds and so was cleaned and baked as prescribed above. The next t e s t revealed that the decay rate now was even worse at 0.5 percent i n about ten seconds, 172 so again the transducer was cleaned and baked. The decay time dropped further to about 0.5 percent i n eight seconds. For the l a s t attempt, ethel alcohol was used for cleaning r e s u l t i n g i n no improvement. According to the procedures above, the leads and a l l connections must be cleaned with a freon cleaning spray. This had no e f f e c t on decay rate. The transducer cable was also dried i n the oven for one hour at 150°C r e s u l t i n g i n no improvement. A technician at K i s t l e r recommended observation of the d r i f t f i r s t with only the charge amplifier, then with the charge amplifier and the transducer cable, and then, f i n a l l y , with the charge amplifier, cable, and transducer to determine which component caused the most amount of d r i f t . This t e s t indicated that the transducer was the cause of the d r i f t . At t h i s point the transducer s t i l l was not c a l i b r a t e d . The transducer was f i n a l l y c a l i b r a t e d by p l o t t i n g , on the oscilloscope, the transducer voltage verses time. The r e s u l t i n g l i n e was l i n e a r l y extrapolated back to time = zero since an exponential decay can be modeled as a st r a i g h t l i n e for time much smaller than the decay time constant. This time = zero voltage and known pressure were used to c a l i b r a t e the transducer. Although, admittedly, t h i s was a crude method, the r e s u l t i n g c a l i b r a t i o n curve correlated very well with the manufacture's c a l i b r a t i o n curve. F i g . C.l i s the c a l i b r a t i o n curve from the manufacturer. The question remained whether the transducer was s t i l l r e l i a b l e since i t had a higher decay rate than that s p e c i f i e d 173 f o r q u a s i - s t a t i c c a l i b r a t i o n . Recall that the time of combustion most important for t h i s project was the time from i g n i t i o n to maximum pressure, which, for most cases, was under t h i r t y milliseconds. During t h i s time, the decay of the transducer was imperceptible, although, during combustion t e s t s , the time constant was set to MEDIUM and not LONG as for qu a s i - s t a t i c c a l i b r a t i o n . In conclusion, the pressure transducer c a l i b r a t i o n curve was the same as the c a l i b r a t i o n given by the manufacture with the transducer. The method used to v e r i f y t h i s was unorthodox, but appears to be v a l i d . 174 Appendix D EQUIPMENT SPECIFICATIONS 1) BAROMETER: Precision Thermometer and Instrument Co. Type: 0.250 inch I.D. F o r t i n Type Mercury Column barometer accurate to .3 mm Hg -2.3 mm Hg (at 740 mm Hg indicated pressure) +0.3 mm Hg (at 740 mm Hg indicated pressure) -2 mm Hg * e.g. subtract 2 mm Hg from indicated pressure reading to correct for the e f f e c t s of room temperature and l a t i t u d e Accuracy: Correction for temperature (at 19°C): Correction for l a t i t u d e (at 50° l a t i t u d e ) : E f f e c t i v e correction: * 2) CHARGE AMPLIFIER: K i s t l e r Instrument Corporation Model: 5004 Type: Dual mode amplifier Scale settings, 12 steps 1,2,5 sequence, for transducer s e n s i t i v i t i e s 0.01 to 0.11 pC/ MU or mV/MU* MU/V 0.1 to 1.1 pC/ MU or mV/MU MU/V 1.0 to 11 pC/ MU or mV/MU MU/V 10 to 110 pC/ MU or mV/MU MU/V 100 to 1,100 pC/ MU or mV/MU MU/V Output voltage and current V&mA 100 to 500,000 10 to 50,000 1 to 5,000 .1 to 500 .01 to 50 + 10 & < + 5 175 Output impedance Input cable, i n s u l a t i o n resistance (charge mode) Frequency response, with standard 5311 f i l t e r (-3dB)** Time constant (depending on selected range)"Long" "Medium" "Short" Amplitude l i n e a r i t y Accuracy of ranges (charge mode) of two most s e n s i t i v e ranges C a l i b r a t i o n input referred to charge input Noise at output; 10 Hz to 330 kHz std. 18 0 kHz f i l t e r d i a l 10-00 & 1-00 Noise input (cable) D r i f t (due to leakage current) Operating temperature range Connectors input & output power Power ohms 100 Teraohms 100 kHz s s s %FS0 pC/mV m Vrms P crms/ n F pC/s °C Type Type VAC Hz & VA MU = mechanical unit (e.g.: p s i , l b , g, etc.) ** Optional f i l t e r s : low pass, series 5311 & 5313 notch, series 5312 0 to 180 1,000-100,000 1 to 5000 0.01 to 50 <+0.05 <+l <+5 1 + 0.5% 0.1 & 1 0.01 < +0.03 0 to 50 BNC neg. 3 pin with gnd 100 to 130 50 to 60 & 8 176 3) COMBUSTION CHAMBER APPARATUS: Ron J . P i e r i k Model: A-1 Type: Multiple s w i r l , spark location, L/D combustion chamber Swirl rates: Spark locations: Valve l i f t s : Valve shapes: L/D r a t i o s : p i n i r a n 9 e :  Chamber Size: diameter: length: Maximum pressure Material: body: end- and side-plates: valve and pu l l r o d : spark probes: Fan Speed: Material: body and impeller: shaft: rpm r/R 1/L mm bar 0 to 6000 0 to 1 0.5* 0 to 7 no-swirl and s w i r l 0.5 0 to 10 mm mm bar 100 50 100 *** 316 s t a i n l e s s s t e e l polycarbonate 4340 s t e e l drawn s t e e l wire (1 mm dia.) rpm 0 to 23,500 aluminum 316 s t a i n l e s s s t e e l 177 Optional equipment: quartz end plates for laser doppler work air heater for variable T^n^ 1/L from 0 to 1 i s possible with special end plates L/D from 0 to 2 is possible again with special end plates *** For high pressures, the optical plates should be replaced with steel plates. 4) COMPUTER: Compaq Computer Corporation Model: Compaq personal computer Memory: 640 K RAM 5) FAN MOTOR: Sears Robuck Co. Model: 758 17860 Type: Sears Wet and Dry Home-N-Shop Vacuum Input: 120 Volts 10 Amps. 6) GAS CHROMATOGRAPH: Hewlett Packard Company Model: 5750b Type: Research gas chromatograph (The specifications for this instrument are too numerous and complex to l i s t here. For more information, please consult the Environmental Engineering laboratory in the C i v i l Engineering department. This laboratory has trained technicians and a small library of manuals and other reference sources.) 178 7) HAND HELD TACHOMETER: Model: Type: Reading Range: Accuracy: 6 - 5000 rpm: 5000 - 30,000 rpm Display: Memory System: System Control: Detection: E f f e c t i v e distance: Light source: Sensor: Re f l e c t i o n mark: Up date gate time: Batteries, included: Size: L i f e : Low voltage in d i c a t o r Operating temperature: Shimpo DT-205 Battery operated, hand held, computer c i r c u i t r y controlled, non contact type tachometer. 6 - 30,000 rpm 1 rpm 2 rpm 5 d i g i t 9mm high LCD Last reading displayed for 2 minutes a f t e r instrument removal. Four i n t e r -mediate readings, l a s t , maximum, and minimum readings stored i n memory auto-matically or by se l e c t i o n . Single chip C-MOS micro-computer Up to 3 feet Incandescent l i g h t Photo-transistor Single tab r e f l e c t i v e tape T y p i c a l l y 1 second 4 1.5 V AA Averages 6 hours continuous use "B" fla s h i n g display 0° to 45°C 179 Construction: Die-cast aluminum housing 8) HIGH-SPEED CAMERA: Red Lake Laboratories, Inc. Photo Instrumentation Equipment Model: Hycam K20S4E Type: High speed, rotating prism, 16 mm motion picture camera Available frame formats: Full frame, .5 frame, .25 frame Available voltages: 115 and 23 0 volts Film capacity: Up to 400 feet Speed Range: 20 to 11,000 Frames per second Resolution: Center-vertical and horizontal: 68 lines/mm ( f u l l frame) Edges-vertical and horizontal: 56 lines/mm ( f u l l frame) Solid state electronic frame rate control to + 1% in regulated speed ranges. Automatic servo brake which prevents film over-run and makes possible high speed stop/starts. Adjustable drag brake on supply spindle for smooth low-speed frame rates. Through-the-lens viewing with direct upright view finder and ground glass focusing gate. Event synchronizer with either normally open or normally closed output. Dual timing lights provide timing pulses in both sides of film. 180 End-of-film cutoff switch. Viscous clutch for smooth s t a r t s . "C" mount lens mount accommodating wide angle ar any f o c a l length "C" mounting lens. 9) IGNITION DELAY TIMER: Donald P. Bysouth Model: A-2 Type: Resistor/capacitor c o n t r o l l e d 555 timer Time range: ms .978 - 99.3 0 Repeatability @ 23°C: ms within < 0.001 Input voltage: V 5-18 (12 normally) Trigger by external microswitch 10) IGNITION SYSTEM: Heath Company Model: CP-1060 Type: Capacitive discharge i g n i t i o n system Nominal Input 12-14 v o l t battery. Maximum Input 18 v o l t s DC. Trigger Source Ign i t i o n points or generator capable of switching 4 0 ohm load to ground within 1 v o l t . Output Pulse Duration 0.6 ms at low voltage and rpm. T y p i c a l : 8.5 v o l t s with b a l l a s t r e s i s t o r i n system. 0.4 ms at normal voltage and rpm. i T y p i c a l : 12-14 v o l t s at 500-4000 rpm. 181 Maximum Allowable Ignition Point Contact Resistance Minimum Allowable Ignition Point Shunting Resistance Typical Open C i r c u i t Voltage With Standard C o i l 0.2 ms at high voltage and rpm. Typ i c a l : 14.5 v o l t s at 6000 rpm. 7 ohms. 100 ohms. Cranking, 8-V input: >26 KV. 500 rpm, 12-V input; >38 KV. 6,000 rpm, 14 V input; >40 KV. 11) LASER: Spectra-Physics Model: 165-00 Type: 1.5 watt Argon Wavelength: Beam diameter at l / e 2 points: Beam Divergence: Cavity length: Cavity configuration: Bore material: Resonator construction: Long term output power s t a b i l i with power s t a b i l i z e r ON: 514.5 nm 1.5 mm 0.5 m i l l i r a d i a n s 1 meter with two mirrors, 1.05 with 1 mirror and 1 prism, long radius BeO Three longitudinal quartz rods heat shielded and thermally coupled. + 0.5 % over 10 hours. 182 with power s t a b i l i z e r OFF: Noise with power s t a b i l i z e r ON 10 Hz to 2 MHz: with power s t a b i l i z e r Off 10 Hz to 2 MHz: Warmup with power s t a b i l i z e r ON: with power s t a b i l i z e r Off: Output p o l a r i z a t i o n : Mirror and prism changes: p o s i t i o n Tube e x c i t a t i o n : Input power requirements: Water flow required: Weight Head: Power supply: + 3 % a f t e r 30 minutes warmup 0.2 % rms t y p i c a l . 0.5 to 1 % rms t y p i c a l . < 1 second to within 95 % of preset l e v e l . <+0.5 % a f t e r 30 minutes. >75 % at turn-on. >95 % at 3 0 minutes, v e r t i c a l . By snap-in p o s i t i v e mirror holders. Current regulated D-C. Regulated to better than +0.5 %. 190 - 225 V, 3 phase, 35 amps per l i n e . 2.2 gallons per minute minimum at 2 5 p s i . Water temperature 35°C maximum. 18.2 kg. (40 lbs.) 31.8 kg. (70 lbs.) 12) LENS: (unknown source) Type: converging, one side f l a t , other spherical Diameter: 300 mm Focal length: 1000 mm Central thickness: 26 mm Edge thickness: 5 mm 13) MOTOR VOLTAGE CONTROL: Staco Energy Products Co. Model: 3PN1010 Type: Variable transformer Input: 120 Volts 50/60 Hz Output: 0 - 120/140 Volts 10 amps. 14) OSCILLOSCOPE: Nicolet Instrument Canada Inc. Model: 3091 Type: D i g i t a l storage oscilloscope Analog Inputs: Two, d i f f e r e n t i a l Input Coupling: AC/DC/GND Input Impedance: 1 Megohm//47 pF Input Range (FS): + 100 mV to + 40 V Safe Overload: + 10V to + 40V Range: 400V ± IV to +400mV Range: 300V Analog Bandwidth: + IV to + 40V Range: 300 kHz + lOOmV to + 400mV Range: 150 kHz 184 Common Mode Rejection Ratio: Common Mode Voltage Range: Noise, RMS: Linearity: Accuracy: Crosstalk: Digital ADC Resolution: Channel Sampling: Maximum Sampling Rate: Minimum Sampling Rate: Sampling Time Uncertainty: Display Memory: Buffer Memory: Trigger Modes: Source: Slope: Coupling: Pre-trigger Delay: Outputs Digital: 90 dB 225% F.S. 0.03%, F.S. 0.1% F.S. 0.2% F.S. < -60 dB 12 bits Simultaneous 1 sec/point (1 MHz) 200 sec/point + 1 ns at 1 MHz 4000 points/channel, 12 bits 4000 points/channel, 12 bits Auto or Normal Ch A, Ch B, or Ext +, -, or Dual AC or DC Cursor selected up to 100% sweep time RS-232C, up to 19,200 baud 1 8 5 15) PRESSURE GAUGE: Heise Model: CMM-8035 Type: Temperature compensating bourdon tube with 3 00 mm face Range: 0 - 760 mm Hg 0 - 760 mm Hg Vacuum 16) PRESSURE TRANSDUCER: K i s t l e r Instrument Corporation Model: 6121a Type: Quartz pressure transducer, miniature p r o f i l e Range bar 0...250 Calibrated p a r t i a l range bar 0...25 bar 0... 2,5 Overload bar 350 S e n s i t i v i t y pc/bar -13.2 Natural frequency kHz >55 Frequency response + 1% kHz 6 L i n e a r i t y % FSO <+ 1,0 Hysteresis % FSO <1,0 Acceleration s e n s i t i v i t y A x i a l bar/g <0,003 Transverse bar/g <0,0002 Shock resistance g 2000 Thermal s e n s i t i v i t y s h i f t 20...100°C % <+0,5 20...350°C % <±3 186 200+ 50"C Calibrated i n range Operating temperature range Transient temperature error (propane flame intermittent on front, 10 Hz) Insulation: at 20°C at 350°C Ground i n s u l a t i o n Weight UC bar Teraohms ohms ohms g <±1 20...350 -196..350 <0,02 >10 >10 >10 9,5 10 8 APPENDIX E: DATANIC LISTING 10 PRINT " Program DATAQ inputs data from N i c o l e t o s c i l l o s c o p e . " 20 PRINT " . " 30 PRINT 40 PRINT 50 PRINT "This program reads one screen f u l l of i n f o r m a t i o n s t o r e d on the" fctO PRINT " N i c o l e t O s c i l l o s c o p e and reproduces i t on t h i s monitor. " 70 PRINT " I f requested, the data can be s t o r e d on a f l o p p y d i s k i n d r i v e B. 80 PRINT 90 DEFINT A 100 DIM A(4100) 110 DIM N(8) 120 PRINT "In order to t r a n s f e r data, check that the o s c i l l o s c o p e i s i n " 130 PRINT "the 'STORE' mode." 140 PRINT "Press 'P' to open communications, then press the RS232 button" 150 PRINT "on the o s c i l l o s c o p e a f t e r the tone." 160 PRINT "Communications remain open u n t i l the DATA ACQUISITION COMPLETE" 170 PRINT "message appears ( t h i s takes approx. 2 minutes)." 180 PRINT 190 IVF-0 200 PRINT 210 PRINT " TAKE DATA,ENTER P AT KEYBOARD," 220 PRINT " THEN PRESS RS-232 ON NICOLET AFTER TONE." 230 PRINT 240 PRINT 250 B$-INPUT$(1) 260 IF B$-"p" THEN 280 270 IF B$-"P" THEN 280 ELSE 250 280 PRINT "COMMUNICATIONS OPEN..." 290 ' Open l i n e t o ramdrive C through f i l e #2 300 ' Open communications f i l e to RS232 p o r t ( f i l e #1), 310 ' 9600 baud, no p a r i t y check, 7 data b i t s , ignore CTS and DSR , 320 OPEN "COM1:9600,S,7,1,CS,DS" AS #1 330 ' Input character s t r i n g from communications b u f f e r . . . 340 ' Write c h a r a c t e r s t r i n g to ramdrive ( f i l e #2)... 350 ' P r i n t character s t r i n g s on screen... 360 SOUND 880,25 370 INPUT #1, I :PRINT I 380 FOR J-3 TO 4002 390 A(J)-VAL(INPUT$(5,1)) 400 NEXT J 410 A$-INPUT$(1,1) 420 FOR L - l TO 8 430 N(L)-VAL(INPUT$(5,1)) 440 NEXT L 450 PRINT 460 CLOSE #1 470 PRINT "DATA ACQUISITION COMPLETE" 480 SOUND 880,10 490 PRINT 500 HZER0-N(4) 510 V1-(N(5)-5)*10 A(N(6)-12) 520 H1-(N(7)-5)*10 A(N(8)-12) 530 HEND-H1*4000 540 HMID-HEND/2 550 VPOS-V1*2000 560 VPOS2-VPOS/2 570 VNEG—VPOS 580 VNEG2-VNEG/2 590 PRINT 600 ANS$-"Y" 610 TTL$-" NICOLET OSCILLOSCOPE RECORD " 620 YAX$-"VOLTS" 630 XAX$-"TIME (SEC.)" 640 IF ANS$-"y" GOTO 720 650 IF ANS$-"Y" GOTO 720 660 PRINT 670 INPUT; "ENTER TITLE ", TTL$ 680 PRINT 690 INPUT; "X-AXIS UNITS ", XAX$ 700 PRINT 710 INPUT; "Y-AXIS UNITS ", YAX$ 720 KEY OFF 730 SCREEN 2,,0,0 740 LINE (100,90)-(600,90) 750 LINE (100,1)-(600,1) 760 LINE (100,180)-(600,180) 770 LINE (100,1)-(100,180) 780 LINE (600,1)-(600,180) 790 LINE (100,45)-(105,45) 800 LINE (600,45)-(595,45) 810 LINE (600,135)-(595,135) 820 LINE (100,135)-(105,135) 830 LINE (350,180)-(350,176) 840 LOCATE 12,8,0 :PRINT "0.00"; 850 LOCATE 1,6,0 :PRINT USING "###.##";VPOS; 860 LOCATE 6,6,0 :PRINT USING "###.##";VPOS2; 870 LOCATE 23,6,0 :PRINT USING "###.##";VNEG; 880 LOCATE 18,6,0 :PRINT USING "###.##";VNEG2; 890 LOCATE 25,52,0 :PRINT XAX$; 900 LOCATE 9,1,0 :PRINT YAX$; 910 LOCATE 1,28,0 :PRINT TTL$; 920 LOCATE 24,12,0 :PRINT "0.0"; 930 LOCATE 24,42,0 :PRINT HMID; 940 LOCATE 24,73,0 :PRINT HEND; 950 LOCATE 1,1,0 960 FOR X-100 TO 600 970 Y-90-A((X-99)*8-5)/22.75 980 PSET(X.Y) 990 NEXT X 1000 INPUT; ST$ 1010 SCREEN 0 1020 PRINT 1030 PRINT "THE DATA ARE BEING STORED ON DISKETTE IN DRIVE B" 1035 PRINT "DO YOU WANT TO STORE DATA IN ASCII (A) OR BINARY (B)? 1036 INPUT " PLEASE INPUT A OR B ",ANS$ 1040 INPUT "Name of the f i l e ",INF$ 1050 VD$-"B:"+INF$ 1060 HW$-"B:UM+INF$ 1065 IF ANS$-"A" GOTO 1132 1066 IF ANS$-"a" GOTO 1132 1070 OPEN "R", #3, VD$, 2 1080 FIELD #3, 2 AS X$ 1090 FOR 1-3 TO 4002 1100 LSET X$-MKI$(A(I)) 1110 PUT #3 1120 NEXT I 1130 CLOSE #3 1131 GOTO 1139 1132 OPEN VD$ FOR OUTPUT AS #3 1134 FOR 1-3 TO 4002 1136 WRITE #3, A(I) 1137 NEXT I 1138 CLOSE #3 1139 IF IVF-1 GOTO 1200 1140 OPEN HW$ FOR OUTPUT AS #3 1160 WRITE #3,VI 1190 CLOSE #3 1200 IVF-IVF+1 1210 PRINT "Number of stored data files: ",IVF 1220 SOUND 880,10 1230 PRINT 1240 INPUT; "INPUT ANOTHER DATA SET (Y/N)"; AN$ 1250 IF AN$-"N" GOTO 1280 1260 IF AN$-"n" GOTO 1280 1270 GOTO 200 1280 PRINT :PRINT "PROGRAM ENDED" 1290 END 1131 GOTO 1139 1132 OPEN VD$ FOR OUTPUT AS #3 1134 FOR 1-3 TO 4002 1136 WRITE #3, A(I) 1137 NEXT I 1138 CLOSE #3 1139 IF IVF-1 GOTO 1200 1140 OPEN HW$ FOR OUTPUT AS #3 1160 WRITE #3,VI 1190 CLOSE #3 1200 IVF-IVF+1 1210 PRINT "Number of stored data files: ",IVF 190 1220 SOUND 880,10 1230 PRINT 1240 INPUT; "INPUT ANOTHER DATA SET (Y/N)"; AN$ 1250 I F AN$-"N" GOTO 1280 1260 I F AN$-"n" GOTO 1280 1270 GOTO 200 1280 PRINT :PRINT "PROGRAM ENDED" 1290 END A P P E N D I X F : R I A H I L I S T I N G C* THE PROGRAM WAS DESIGNED BY A.RIAHI. * C* MARCH/1987 * C* * C** A A AAA* **************** *"***• A AAAAA*************AAAAAAAAAAA*AAAAA C INTEGER NO CHARACTER*1 ANS INTEGER DATAl(1000),NODATA,NOFILE,NEWNO COMMON/SMOOTH/PSMOTH(27,1000),TSMOTH(1000) COMMON/DATA/NODATA,DATAl,NOFILE COMMON/BURN/T10(27),T50(27),T90(27) COMMON JMAX(27) COMMON/ENS/PREAVG(1000).PREDEV(IOOO) COMMON/NEW/PNEW(1000),TNEW(27,1000) COMMON/AVERG/TIMEAV(1000).TIMEDV(IOOO),NEWNO COMMON/TAVG/T10AVG,T50AVG,T90AVG COMMON/TSTDEV/T10DEV,T50DEV,T90DEV C C DO 500 1-1,27 J.MAX(I)-0 500 CONTINUE OPEN (UNIT-1,FILE-'TERMINAL') 1 CALL CLS WRITE(1,10) 10 FORMAT ( ' *** A A* A A A A A A A"A"A'A A A A * A A A A A A A A A A A A A A * A A A A A A A A A A A A A A A ' 1//'* YOU HAVE THE FOLLOWING OPTIONS: 1//'*1. READ IN MAXIMUM OF 30 PRESURE DATA FILES IN 1 '* BINARY FORM OBTAINED USING DATANIC. l / / ' * 2 . OBTAIN THE ENSEMBLE AVERAGE AND STANDARD 1 '* DEVIATION OF PRESSURE RECORDS l / / ' * 3 . OBTAIN THE TIME FOR 10% , 50% OR 90% BURNING 1 '* RATE AND THEIR AVERAGE AND ST_DEV l / / ' * 4 . OBTAIN THE ENSEMBLE AVERAGE AND STANDARD 1 '* DEVIATION IN THE TIME DOMAIN FOR A GIVEN 1 '* SET OF PRESSURES l / / ' * 5 . EXIT THE PROGRAM 1//'* PLEASE INPUT 1 OR 2 etc 1 > A A *********** **AAAA'AAAAAAAAAAAAAAAAAAAA *•*•**•* * * * * i READ(l,*)NO GO TO (100,200,300,400,999),NO 100 CALL READPR GO TO 2 200 CALL AVPRES(NODATA,PSMOTH,PREAVG,PREDEV,NOFILE) GO TO 2 300 CALL PERBUR GO TO 2 400 CALL BURNRT * ' 1 *' t / *' t * ' 1 / *' t *' , / *' t *' f / *' f / *' t *' . *' , / •*») 2 CONTINUE CALL CLS WRITE(1,20) 20 FORMAT(//' DO YOU WANT TO STORE THESE VALUES ?') READ(1,3)ANS 3 FORMAT(A) IF(ANS.EQ.'Y'.OR.ANS.EQ.»y') THEN CALL STORE(NO) END IF WRITE(1,21) 21 FORMAT (' DO YOU WANT TO DO ANY THING ELSE ?') READ(1,3) ANS IF (ANS.EQ.'Y'.OR.ANS.EQ.'y') GO TO 1 999 CLOSE (UNIT-1) STOP END C C SUBROUTINE READPR C C* * C* IN THIS ROUTINE THE PRESSURE RECORDS OBTAINED BY DATANIC* C* ARE READ . THESE VALUES ARE ORIGINALLY STORED IN BINARY * C* FORM .THEY ARE READ IN BINARY FORM AND THEN TRANSFERD IN* C* TO ASCII FORM USING THE ROUTINE WRITTEN BY LEPP.G * C * * * * * * V r A * A * * * * * * * * * * * * * * * * * * * * * * * * * * * c CHARACTER*1 FINAME(12),BLANK CHARACTER*1 F4(10),ANS CHARACTER*12 INFILE(27),BLANK1 CHARACTER*13 SCFILE INTEGER DATAl(1000),NODATA,NOFILE,NO,IFLAG REAL SAMPFR,SCALE C0MM0N/SM00TH/PSM0TH(27,1000).TSMOTH(IOOO) COMMON/DATA/NODATA,DATAl,NOFILE COMMON/FILE/INFILE DATA BLANK/' '/ DATA BLANK1/' '/ DATA F4/'0' ,'1','2','3','4','5','6\'7','8\'9'/ DO 11 1-1,12 11 FINAME(I)-BLANK DO 9 1-1,27 9 INFILE(I)-BLANK1 CALL CLS WRITE(l.lO) 10 FORMAT (' *HHHh»r*********^ > 1 '* PRESSURE RECORDS ARE READ IN THIS PART OF *' 1 '* THE PROGRAM *' 1 ' * * ^ * * * * * * * * * * * * * * * * * * * * * * * * * r * * * * * * * * * * * * * * * * * ' 1 'YOU CAN READ A MAXIMUM OF 27 FILES. ',/ 193 1 'PLEASE INPUT THE TOTAL NUMBER OF DATA FILES TO BE READ') READ(1,*)NOFILE CALL CLS WRITE (1,20) 20 FORMAT('THE FILES TO BE READ CAN HAVE NAMES UP TO 8 ',/ 1 'CHARACTER LONG egB:DATA.l ',// 1 ' ^ H ^ * • • * * * ^ • ^ ^ r > f r > » ^ > | / 1 '* TO REDUCE THE OPERATOR EFFORT TO INPUT THE NAME OF ALL*',/ 1 '* DATA FILES, THE PROGRAM IS DESIGNED TO AUTOMATICALLY *',/ 1 '* READ ALL THE FILES WHICH HAVE THE SAME NAME BUT *',/ 1 '* DIFFERENT VERSION NUMBERS (INCREMENTED BY 1) *',/ 1/'* EX DATA.1 , DATA.2 et c . *',// 1 '* PLEASE INPUT THE NAME OF THE FIRST DATA FILE WITHOUT *',/ 1 '* THE VERSION NUMBER eg ; B:ME *',/ 1 ' T / H H H H H ^ * * * I H ^ * ^ * * * * * * ^ ' ) READ (1,30,END-101)(FINAME(I),1-1,6) 30 FORMAT (6A) 101 NLENGT-I NLNTP1-I+1 NLNTP2-I+2 IVER-2 IVER1-0 DO 100 I-l.NOFILE FINAME(NLENGT)-' . ' FINAME(NLNTP1)-F4(IVER) IF (I.GT.9) THEN FINAME(NLNTP2)-F4(IVER1) ENDIF L~l DO 102 J-1.NLNTP2 INFILE(I)(L:J)-FINAME(J) L-L+l 102 CONTINUE IF(I.LT.9)THEN IVER-IVER+1 ELSE IVER1- IVER1+1 IVER-1 END IF 100 CONTINUE IF (FINAME(1).EQ.'B'.OR.FINAME(1).EQ.'b') THEN SCFILE(1:2)-INFILE(1)(1:2) SCFILE(3:3)-'U' SCFILE(4:)-INFILE(l)(3:) ELSE SCFILE(1:1)-'U' SCFILE(2:)-INFILE(l)(1:) END IF OPEN(UNIT-6,FILE-SCFILE,FORM-'FORMATTED',STATUS-'OLD') READ(6,*)SCALE CLOSE(UNIT-6) CALL CLS WRITE(1,32) 32 FORMAT (/' DO YOU WANT TO READ ASCII (A) FILE OR'/, 1 ' BINARY (B) FILE '/, 1 'INPUT A OR B') READ (1,34)ANS 34 FORMAT (A) WRITE(1,31) 31 FORMAT(' ENTER THE N TH POINT TO BE READ',/ 1 ' eg : EVERY 4 TH POINT OR EVERY 10TH POINT e t c ' / , 1 ' MAXIMUM OF 1000 POINTS') READ(1,*)JSTEP WRITE(1,33) 33 FORMAT(/'INPUT THE SAMPLE_FREQUENCY') READ(1,*)SAMPFR WRITE(1,35) 35 FORMAT(//'HOW MANY DATA FILES ARE STORED IN ONE DISKET ?') READ(1,*)IFLAG DO 104 I-l,NOFILE NO-I CALL BINARY(INFILE(I),ANS,JSTEP.NO) IF (I.EQ.l)THEN NDATM1-NODATA-1 TSMOTH(1)-0. DO 103 J-l.NDATMl 103 TSMOTH(J+l)-REAL(J)*JSTEP/SAMPFR END IF DO 105 J-l.NODATA 105 PSM0TH(I,J)-SCALE*DATA1(J) IF (I.EQ.IFLAG.AND.I.LT.NOFILE) THEN WRITE (1,36)1 36 FORMAT (14,' FILES IN DISKET NO.1 HAS BEEN READ '/, 1 ' PLEASE CHANGE YOUR DISKET AND PRESS C TO CONTINUE') READ(1,37)ANS 37 FORMAT (A) END IF 104 CONTINUE RETURN END C C SUBROUTINE BINARY(ALAK,ANS,ISTEP,NO) C C THIS ROUTINE READS UNFORMATTED DIRECT ACCESS FILE * C* DESIGNED BY A.RIAHI * C***************** * * ** ********************************** c C CHARACTER* 1 ANS CHARACTER*2 IDATA 195 INTEGER*2 BDAT,ADAT INTEGER DATAl(lOOO),NODATA,NOFILE CHARACTER*12 ALAK COMMON/DATA/NODATA,DATA1,NOFILE CALL CLS WRITE(1,10)ALAK 10 FORMAT(' FILE'.4X.A.2X,'IS FOUND AND BEING READ '//) IF (ANS.EQ.'B'.OR.ANS.EQ.'b') THEN DO 100 I-l,4000,ISTEP IF(I.EQ.l) THEN J - l ELSE J- J + l ENDIF OPEN(UNIT-2.FILE-ALAK,FORM-'UNFORMATTED'.ACCESS-'DIRECT', 1 RECL-2,STATUS-'OLD') READ(2,REC-I)IDATA CLOSE (UNIT-2) BDAT-ICHAR(IDATA(:1))+ICHAR(IDATA(2:))*256 DATAl(J)-BDAT 100 CONTINUE IF (NO.EQ.l) THEN NODATA-J END IF ELSE OPEN (UNIT-2,FILE-ALAK,FORM-'FORMATTED',STATUS-'OLD') DO 201 1-1,3999 READ (2,*) AD AT IF (I.EQ.l) THEN IF (ADAT.LT.O)THEN ADAT-NINT(REAL(ADAT)/2 5 6.) END IF J - l DATAl(J)-ADAT J-J+l END IF IF (MOD(I,ISTEP).EQ.O) THEN IF (ADAT.LT.O) THEN ADAT-NINT(REAL(ADAT)/256.) END IF DATAl(J)-ADAT J-J+l END IF 201 CONTINUE IF (NO.EQ.l) THEN NODATA-J-1 END IF CLOSE (UNIT-2) END IF WRITE(1,30)ALAK 30 FORMAT(' FILE',4X,A,2X,'WAS READ SUCCESSFULLY') RETURN END C C SUBROUTINE AVPRES(NODATA,ALAK1,ALAK2,ALAK3,NOFILE) C C C******************************^ C* IN THIS ROUTINE THE AVERAGE PRESURE AND THE ST_DEV IS CALCULATED * C****************************************************^ c DIMENSION ALAK1(27,1000).ALAK2(NODATA).ALAK3(NODATA) CALL CLS WRITE(1,80) 80 FORMAT(//' THE ENSEMBEL AVERAGE AND THE CORESPONDING '/. . 1 'ST_DEV IS CALCULATED ') DO 10 I-l,NODATA SUM1-0. DO 20 J-l,NOFILE SUM1-SUM1+ALAK1(J,I) 20 CONTINUE ALAK2(I)-SUMl/FLOAT(NOFILE) 10 CONTINUE DO 40 J-l,NODATA SIG1-0. DO 30 I-l,NOFILE SIG1-SIG1+(ALAK1(I,J)-ALAK2(J))**2 30 CONTINUE ALAK3(J)-SQRT(SIG1/FL0AT(N0FILE)) 40 CONTINUE RETURN END C C SUBROUTINE PERBUR C C********************************************************* C* 10% , 50% , 90% BURNING RATE TIME ARE CALCULATED HERE * C*********************************************************** c c CHARACTER*1 ANS REAL DP10,DP50,DP90,T10AVG,T50AVG,T90AVG,T10DEV, 1 T50DEV,T90DEV,TIME INTEGER NODATA,DATAl(1000),NOFILE,L.N COMMON/SMOOTH/PSMOTH(27,1000).TSMOTH(IOOO) COMMON/DATA/NODATA,DATAl,NOFILE COMMON JMAX(27) COMMON/BURN/T10(27),T50(27),T90(27) COMMON/TAVG/T10AVG,T50AVG,T90AVG COMMON/TSTDEV/T10DEV,T50DEV,T90DEV CALL CLS WRITE (1,1) I FORMAT(' * * * * * * * * * * * * * * * * * * * * A A***A A A A A A *A * * * * * * * * * * * * * * * * * * ' > / 1 '* THE TIME FOR 10% , 50% AND 90% BURNING RATE ARE *',/ 1 '* CALCULATED HERE *'/, 1 '* *'/, 1 '* MAXIMUM PRESSURE IN EACH FILE IS FIRST OBTAINED *'/. 1 '* PLEASE WAIT *'/, 1 ' * * * * * * * * * * * * * * r A * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ' ) DO 10 1-1,NOFILE L-0 N-I CALL MAXPRE(N,NODATA,L) JMAX(I)-L 10 CONTINUE I I WRITE(1,80) 80 FORMAT(/' YOU CAN FIND EITHER OF THE FOLLOWINGS'/, 1 ' 1. TIME TO BURN 10% OF THE MIXTURE.'/. 1 ' 2. TIME TO BURN 50% OF THE MIXTURE.*/, 1 ' 3. TIME TO BURN 90% OF THE MIXTURE.'/, 1 /'PLEASE INPUT YOUR CHOICE ') READ(l,*)NO WRITE (1,79) 79 FORMAT (/'CALCULATION IN PROGRESS ') DO 20 1-1,NOFILE L-JMAX(I) N-I DP-PSMOTH(I,L)-PSMOTH(1,1) GO TO (100,200,300),NO 100 DP-.1*DP GO TO 101 200 DP-.5*DP GO TO 101 300 DP-.9*DP 101 CALL TINPOL(L,N,DP,TIME) GO TO (110,210,310),NO 110 T10(I)-TIME GO TO 20 210 T50(I)-TIME GO TO 20 310 T90(I)-TIME 20 CONTINUE WRITE(1,81) 81 FORMAT(//' DO YOU WANT THE AVERAGE AND ST_DEV OF THE CALCULATED', 1 ' TIME ?') READ(1,3)ANS 3 FORMAT(A) IF (ANS.EQ.'N'.OR.ANS.EQ.'n') GO TO 901 GO TO (1000,1010,1020),NO 1000 CALL BUSTAT(NOFILE,T10.T10AVG.T10DEV) GO TO 901 1010 CALL BUSTAT(NOFILE,T50,T50AVG,T50DEV) GO TO 901 1020 CALL BUSTAT(NOFILE,T90.T90AVG.T90DEV) 901 WRITE(1,802) 802 FORMAT(//'DO YOU WANT TO CALCULATE ANOTHER BURNING RATE TIME ?') READ(1,3)ANS CALL CLS IF (ANS.EQ.'y'.OR.ANS.EQ.'Y') GO TO 11 RETURN END C C SUBROUTINE TINPOL(L,N,DP,TIME) C C TIME INTERPOLATION IS DONE HERE * C C COMMON/SMOOTH/F-SMOTH (27,1000), TSMOTH (1000 ) REAL DP,TIME INTEGER L.N KLO-l KHI-L 1 IF(KHI-KLO.GT.l) THEN K-(KHI+KL0)/2 IF (PSMOTH(N,K).GT.DP) THEN KHI-K ELSE KLO-K END I F GO TO 1 END IF H-(PSMOTH(N,KHI)-PSMOTH(N,KLO)) A-(PSMOTH(N,KHI)-DP)/H B-(DP-PSMOTH(N,KLO))/H TIME-A*TSMOTH(KLO)+B*TSMOTH(KHI) RETURN END C C SUBROUTINE BUSTAT(NOFILE,ALAK,ALAVG,ALASD) C C************************************** C* AVERAGE AND ST_DEV OF THE BURNING TIME IS CALCULATED HERE* C*************************************** C REAL ALAVG,ALASD DIMENSION ALAK(NOFILE) CALL CLS WRITE (1,10) 10 FORMAT (/'AVERAGE AND ST_DEV IS CALCULATED ') SUM1-0. DO 20 I-l,NOFILE SUM1-SUM1+ALAK(I) 20 CONTINUE ALAVG-SUM1/FLOAT(NOFILE) SIG1-0. DO 30 I-l,NOFILE SIG1-SIG1+(ALAK(I)-ALAVG)**2 ALASD-SQRT(SIGl/FLOAT(NOFILE)) RETURN END C C SUBROUTINE BURNRT C C*************** * ************************************* C* THE MEAN AND ST_DEV IN THE TIME DOMAIN FOR A GIVEN* C* SET OF PRESSURES ARE CALCULATED HERE * C C INTEGER DATAl(1000),NODATA,NOFILE,NEWNO COMMON/SMOOTH/PSMOTH(27., 1000) .TSMOTH(IOOO) COMMON/DATA/NODATA,DATAl,NOFILE COMMON JMAX(27) COMMON/NEW/PNEW(1000),TNEW(27,1000) COMMON/AVERG/TIMEAV(1000),TIMEDV(1000),NEWNO CALL CLS IF (JMAX(l).EQ.O) THEN WRITE(1,8) 8 FORMAT (' MAXIMUM PRESSURE HAS TO BE OBTAINED ',/ 1 ' PLEASE BE PATIENT ') DO 9 I-l,NOFILE L-O N-I CALL MAXPRE(N,NODATA,L) JMAX(I)-L 9 CONTINUE END I F CALL CLS WRITE(l.lO) 10 FORMAT(' TO OBTAIN THE MEAN AND ST_DEV IN THE TIME DOMAIN',/ 1 ' YOU NEED TO SPECIFY A RANGE OF PRESSURES. THE ',/ 1 'MAXIMUM PRESSURE IN EACH OF THE FILES PROCESSED ARE:'//) Ll-JMAX(l) L2-JMAX(NOFILE) WRITE (l,*)((PSMOTH(I,L),L-L1,L2),I-l.NOFILE) C l l FORMAT (2X.F6.4) WRITE (1,12) 12 FORMAT(//,' WHICH ONE OF THE MAXIMUM PRESSURE IS LOWEST',/ 200 1 ***** PLEASE INPUT 1 OR 2 OR 3 etc') READ(1,*)N WRITE(1,13) 13 FORMAT(//'PLEASE SPECIFY A SUITABLE NUMBER OF PRESSURE POINTS',/ 1 ' FOR WHICH THE CORRESPONDING TIME CAN BE CALCULATED .',/ 1 ' eg: 100 OR 1000 POINTS (MAXIMUM NO - 1000)',/ 1 'THE MEAN AND ST_DEV IN THE TIME DOMAIN ' , 1 /'CAN THEN BE OBTAINED.') READ (1,*) NEWNO NL-JMAX(N) DP-PSMOTH(N,NL)/NEWNO DO 30 I-l,NEWNO PNEW(I)-I*DP 30 CONTINUE T-0. WRITE (1,14) 14 FORMAT (//' THE BURNING TIMES FOR THE GIVEN PRESSURES ARE BEING', 1 ' CALCULATED ') DO 40 I-l,NOFILE DO 50 J-l,NEWNO N l - I DPl-PNEW(J) NLl-JMAX(I) CALL TINPOL (NL1,N1,DP1,T) TNEW(I,J)-T 50 CONTINUE 40 CONTINUE WRITE (1,15) 15 FORMAT(//' THE MEAN AND ST_DEV IN TIME DOMAIN IS CALCULATED ') CALL AVPRES(NEWNO,TNEW,TIMEAV,TIMEDV,NOFILE) RETURN END C C SUBROUTINE STORE(NO) C C**************************************************************** C* DIFFERENT SETS OF OUT PUT FILES ARE PREPARED HERE FOR LATER * C* PLOTTING AND OBSERVATION * C**************************** * * ********************************* c c CHARACTER*10 OUTFIL,FILPRE(27) INTEGER NUMBER,NUM1,DATAl(1000),NOFILE,NODATA CHARACTER*1 ANS COMMON/SMOOTH/PSMOTH(27,1000),TSMOTH(1000) COMMON/DATA/NODATA,DATAl,NOFILE COMMON/BURN/T10(27),T50(27),T90(27) COMMON/ENS/PREAVG(1000),PREDEV(1000) COMMON/AVERG/TIMEAV(1000),TIMEDV(1000),NEWNO COMMON/TAVG/T10AVG,T50AVG,T90AVG 201 COMMON/TSTDEV/T10DEV,T50DEV,T90DEV COMMON/NEW/PNEW(1000),TNEW(27,1000) GO TO (100,200,300,400),NO 100 CALL CLS WRITE(1,11) 11 FORMAT(' IN THIS PART OF THE PROGRAM DATA FILES OF PRESSURE VS', 1 ' TIME ARE PREPARED.'//) WRITE(1,12)NOFILE 12 FORMAT(/4X,14,2X,'DATA FILES HAVE BEEN READ .»,/ 1 'DO YOU WANT TO STORE ANY OF THESE IN ASCII FORMAT ?'/) READ(1,1)ANS 1 FORMAT(A) IF (ANS.EQ.'n'.OR.ANS.EQ.'N') RETURN WRITE (1,13) 13 FORMAT(' HOW MANY FILES DO YOU WANT TO MAKE IN ASCII FORM ?'/) READ(1,*) NUMBER WRITE(1,14)NODATA.NOFILE 14 FORMAT( ' THERE ARE',2X,14,IX,' DATA POINTS IN EACH OF',14,/ 1 'FILES READ. DO YOU WANT TO STORE ALL THESE POINTS ?'/) READ(1,1)ANS IF (ANS.EQ.'Y'.OR.ANS.EQ.'y') GO TO 101 WRITE (1,15) 15 FORMAT (' HOW MANY OF VALUES IN EACH FILE DO YOU', 1 ' WANT TO STORE ?»/) READ(1,*) NUM1 ASTEP-FLOAT(NODATA)/FLOAT(NUM1) ISTEP-NINT(ASTEP) GO TO 105 101 ISTEP-1 105 CALL CLS DO 102 1-1,NUMBER WRITE (1,16) 16 FORMAT(' INPUT THE NAME OF OUT PUT FILE') READ(1,1)FILPRE(I) WRITE (1,17)1 17 FORMAT(//' THE PRESSURE FILE',14,' IS BEING PREPARED ') OPEN(UNIT-3,FILE-FILPRE(I).FORM-'FORMATTED',STATUS-'NEW') DO 103 J-l,NODATA,ISTEP WRITE(3,18)PSMOTH(I,J),TSMOTH(J) 18 FORMAT(E12.4,2X,E12.4) 103 CONTINUE CLOSE (UNIT-3) 102 CONTINUE GO TO 999 200 CALL CLS WRITE(1,201) 201 FORMAT('******MEAN AND ST_DEV ARE BOTH STORED IN THE SAME FILE'/. 1 ******* FIRST VALUE- MEAN SECOND-ST_DEV THIRD-TIME') WRITE (1,203) 203 FORMAT(/' INPUT THE NAME OF THE OUT PUT FILE'/) READ(l,l)OUTFIL OPEN (UNIT-3 , FILE-OUTFIL, FORM-' FORMATTED' , STATUS-' NEW' ) WRITE(1,204)NODATA 204 FORMAT( ' THERE ARE ',14,' MEAN PRESSURE VALUES '/, 1 ' DO YOU WANT TO STORE ALL OF THEM ?'/) READ(1,1)ANS IF (ANS.EQ. 'Y' .OR.ANS.EQ. 'y') GO TO 210 WRITE (1,205) 205 FORMAT(' HOW MANY OF THESE VALUES DO YOU WANT TO STORE ?'/) READ(1,*)NUM1 ASTEP-FLOAT (NODATA) /FLOAT (NUM1) ISTEP-NINT(ASTEP) GO TO 211 210 ISTEP-1 211 WRITE(1,208)OUTFIL 208 FORMAT(//' THE FILE ' ,A10,'IS BEING PREPARED ') DO 220 I-l,NODATA,ISTEP WRITE(3,207)PREAVG(I).PREDEV(I),TSMOTH(I) 207 FORMAT(E12.4,2X,E12.4,2X,E12.4) 220 CONTINUE CLOSE (UNIT—3) GO TO 999 300 CALL CLS 360 WRITE(1,302) 302 FORMAT(//' YOU CAN :'/, 1 ' 1 . STORE THE TIME FOR 10% BURNING,',/ 1 ' 2 . STORE THE TIME FOR 50% BURNING,',/ 1 ' 3 . STORE THE TIME FOR 90% BURNING,',/ 1 ' 4 . STORE THE CORRESPONDING AVERAGES AND ST_DEV,'/, 1 ' 5. TO LEAVE THIS ROUTINE.'//, 1 ' PLEASE INPUT YOUR CHOICE 1 OR 2 etc') READ(1,*) NUMBER IF (NUMBER.EQ.5) GO TO 999 WRITE(1,303) 303 FORMAT (//'INPUT THE NAME OF THE OUT PUT FILE') READ(l,l)OUTFIL OPEN (UNIT-3,FILE-OUTFIL,FORM-'FORMATTED',STATUS-'NEW') WRITE(1,304)0UTFIL 304 FORMAT (//'THE FILE ',A10,' IS BEING PREPARED '/) GO TO (310,320,330,340).NUMBER 310 CALL SORT(NOFILE,T10) DO 311 I-l,NOFILE WRITE(3,305)T10(I) 305 FORMAT(E12.4) 311 CONTINUE GO TO 350 320 CALL SORT(NOFILE,T50) DO 321 I-l,NOFILE WRITE(3,305)T50(I) 321 CONTINUE GO TO 350 330 CALL SORT (NOFILE, T90) DO 331 1-1,NOFILE WRITE(3,305)T90(I) 331 CONTINUE GO TO 350 340 WRITE(3,308)T10AVG.T50AVG.T90AVG 308 FORMAT(' THE BURNING RATE TIME IN ASCENDING ORDER ARE '/. 1 ' ( i e 10,50,90):'/E12.4,/E12.4./E12.4/) WRITE(3,309)T10DEV,T50DEV,T90DEV 309 FORMAT( /'THE CORRESPONDING ST_DEV ARE (10,50,90):'/. 1 E12.4,/E12.4,/E12.4) 350 CLOSE (UNIT-3) GO TO 360 400 CALL CLS WRITE(1,402) 402 FORMAT(/'PLEASE INPUT THE NAME OF THE OUT PUT FILE'/) READ(l,l)OUTFIL OPEN(UNIT-3,FILE-OUTFIL,FORM-'FORMATTED',STATUS-'NEW') WRITE(1,403)NEWNO 403 FORMAT(/'THERE ARE ',14,' DATA POINTS FOR THE MEAN BURNING' 1 ' TIME'/'DO YOU WANT TO STORE ALL OF THEM ?'/) READ(1,1)ANS IF (ANS.EQ.'Y'.OR.ANS.EQ.'y') GO TO 410 WRITE (1,404) 404 FORMAT(/'HOW MANY OF THEM DO YOU WANT TO STORE ?'/) READ(1,*)NUMBER ASTEP-FLOAT(NEWNO)/FLOAT(NUMBER) ISTEP-NINT(ASTEP) GO TO 411 410 ISTEP-1 411 WRITE (1,405)OUTFIL 405 FORMAT( ' THE FILE ',A10,' IS BEING PREPARED'/, 1 ' THE FIRST VALUE -AVERAGE SECOND VALUE-ST_DEV ', 1 ' THIRD-PRESSURE'.//'PLEASE BE PATIENT ') DO 420 1-1,NEWNO,ISTEP WRITE(3,406)TIMEAV(I).TIMEDV(I),PNEW(I) 406 FORMAT(3(F12.8,2X)) 420 CONTINUE CLOSE (UNIT-3) 999 CALL CLS RETURN END C C SUBROUTINE SORT(N.RA) C C********************************** C* THE DATA IS SORTED IN ASCENDING ORDER * C************************************* C REAL RA(N) L-N/2+1 IR-N 10 CONTINUE IF(L.GT.l) THEN L - L - l RRA-RA(L) ELSE RRA-RA(IR) RA(IR)-RA(1) IR-IR-1 IF(IR.EQ.l) THEN RA(1)-RRA GO TO 999 END IF ENDIF I-L J-L+L 20 IF (J.LE.IR) THEN IF (J.LT.IR) THEN IF (RA(J).LT.RA(J+1)) J - J + l ENDIF IF (RRA.LT.RA(J)) THEN RA(I)-RA(J) I - J J-J+J ELSE J-IR+1 ENDIF GO TO 20 ENDIF RA(I)-RRA GO TO 10 999 RETURN END C C SUBROUTINE MAXPRE(N,NODATA,L) C C*********************************** C* MAXIMUM PRESSURE IS OBTAINED HERE * C*********************** C C REAL ALAK(IOOO) INTEGER N,NODATA,L COMMON/SMOOTH/PSMOTH(27,1000).TSMOTH(IOOO) DO 10 1-1,NODATA 10 ALAK(I)-PSMOTH(N,I) CALL SORT(NODATA,ALAK) DO 20 1-1,NODATA L-I I F (ALAK(NODATA).EQ.PSMOTH(N,I)) GO TO 11 20 CONTINUE 11 RETURN END 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097192/manifest

Comment

Related Items