• No results found

Combustion characteristics of multicomponent fuel-air mixtures

N/A
N/A
Protected

Academic year: 2022

Share "Combustion characteristics of multicomponent fuel-air mixtures"

Copied!
16
0
0

Loading.... (view fulltext now)

Full text

(1)

COMBUSTION CHARACTERISTICS OF

MULTICOMPONENT FUEL-AIR MIXTURES

by

KESHAVA MURTHY P.V.

Department of Mechanical Engineering

submitted

in fulfillment of the requirements of the degree of DOCTOR OF PHILOSOPHY

to the

INDIAN INSTITUTE OF TECHNOLOGY, DELHI NEW DELHI, INDIA

May - 2005

(2)

a

I. 1.

LV:

31

\, :-? A.14'Y

4se.

aloILEZ.1.39...

j••••

(3)

Dedicated to

my dear Parents and Teachers

(4)

r. Anjan Rai Associate Professor.

Dept Mechanical Engineering.

Indian Institute of Technology Delhi

CERTIFICATE

The thesis entitled "Combustion Characteristics of Multicomponent Fuel-Air Mixtures" being submitted by Mr. Keshava Murthy T.V. to the Indian Institute of Technology Delhi for award of the degree of Doctor of Philosophy, is a record of original bonafide research work carried out by him. He has worked under our guidance and supervision, and has fulfilled the requirements for the submission of this thesis, which has attained the standard required for a Ph.D. degree of this Institute.

The results represented in this thesis have not been submitted elsewhere for the award of any degree or diploma.

C9 g../-') Dr. M.R. Ravi Associate Professor,

Dept Mechanical Engineering, Indian Institute of Technology Delhi

(5)

ACKNOWLEDGEMENTS

How can I write my gratitude,

However much like Shakespeare I might write.

Above all else, you've shaped my attitude, Nurturing me with discipline and light.

Knowledge is the least of what you taught, Yet that least at least prepared my head.

Out of your heart I've learned the things I ought, Underscoring words you never sa

i

d.

It was a great journey which lasted for six years. Even though at times looked long, when I look back now, it seems like a short and crisp tenure filled with all round development. Many many people were involved in this work and I like to thank all those unknown people.

My heartfelt gratitude to my supervisor Dr. M.R. Ravi, who shaped me to the present form. He has been a great teacher, who not only taught me the technical complexities of the subject, but also a spiritual teacher. His great ocean of positiveness submerged all my negativities. Words fall short to express his wisdom, his knowledge and his simplicity.

I thank him with all my heart for supervising this work from the front.

I am grateful to my supervisor Dr. Anjan Ray for his unending enthusiasm to explore

• the complexities of combustion and flames. His hard work and perseverance has been inspiring me to the core. His advice, suggestions and timely help at various stages of this work helped me in a great way to accomplish the objectives.

I am also thankful to Prof. R.R. Gaur, Prof. L.M. Das, Dr. P.M.V. Subbarao and Dr. Sangeeta Kohli for their constant monitoring and guidance.

Prof. P.L. Dhar was of great inspiration to me. Even though I never got a chance to interact with him, I was fortunate enough to learn the simple laws of universal human values from his lectures. I learnt from watching him walking in the corridors, during his lectures and from his lifestyle.

My heart is filled with a wave of overwhelming gratitude for the unlimited facilities given to me to carry out this dissertation at the Fuels and Combustion Laboratory of

(6)

ii IIT,Delhi. My sincere thanks to all the staff of the laboratory, especially, Mr. Ram Prasad, Mr. Sultan Singh Negi, Mr. Kuldeep Singh, Mr. Prathap Sing Negi, Mr.

Ayodya Prasad for their skillful work during the fabrication and experimentation. My sincere acknowledgements are due to Mr. Pandey, who lifted those heavy gas cylinders along with me from the ground floor to the lab. I thank him for his selfless devotion for duty.

I am greatly indebted to the Saint of Strings, the maestro of Indian Classical Music, my guru, Pt. Bhajan Sopori who colored my dreams. His divine smile always lifted my spirits. I also thank Shri. Aditya Biswal for introducing me to Indian Classical Music.

It was Vaidya who made me laugh at times of boredom, Subhash who filled me with joy at times of guilt, Prathap who gave all round help when it was required. It was wonderful being with Vaidya. Discussions with him at coffee shop, hostel mess and corridors ranged from CFD & Combustion to Philosophy and the Absolute Truth of life.

I thank Vaidya, Subhash, Prathap, Velamati and all my friends for all the support and help. Many many thanks to my colleagues at National Power Training Institute for being with me always.

My sincere thanks to Prof. C.J. Rallis, Prof. Ishwar K. Puri, Prof. B.N. Raghu- nandan, Prof. Egolfopoulos, Prof. Sheppard, Prof. Rogg, Prof. L.P.H. de Goey and Prof. Zevenbergen for their help during the tenure of this work. Their technical advice by e-mails and/or in personal meetings helped this dissertation to take a proper shape.

I owe everything to my Parents and I have no words to express my gratitude to them for their never ending encouragement, support and love. I also thank my family members, Leela, Indira, Ramesh, Mukund, Anu, Praju and Chintu. Many many thanks to Jyothi, who stud by my side day-in and day-out, at times of joy and sorrow. She had the wisdom to accept me with all my commitments. I thank her from the bottom of my heart. A special thanks to Amritavarshini for filling me with joy towards the end of the thesis defence.

Keshava Murthy T.V.

November 9, 2005

(7)

Hi

ABSTRACT

The generation and utilization of low calorific value gaseous(LCVG) fuels finds a prominent position in fuels and energy research. LCVGs are commonly composed of, car- bon monoxide, hydrogen and to some extent, methane. The present investigation involves an LCVG, called producer gas, with a composition of 22%H2, 22%CO3 4%CH4. 10%CO2 and 42%N2. Producer gas is the outcome of gasification of biomass, the composition of which depends on the type of biomass used in the gasifier. The above composition is considered as a standard composition in the present work. The present work investigates the combustion aspects of this multi-component fuel, from a fundamental point of view, by focussing on the premixed combustion of this fuel. The work involves investigations on laminar burning velocity and flame structure of this fuel-air mixture. The wide range of transport properties and burning velocities of the component fuel gases of producer gas makes the investigations even more interesting.

From the extensive literature review, it was noted that laminar burning velocity is the only combustion characteristic which is an intrinsic property of the fuel and all others such as minimum ignition energy, quenching distance and flammability limits. instru- ment specific or influenced by external parameters. A comprehensive data base of lami- nar burning velocity of producer gas-air mixture as a function of pressure, temperature, equivalence ratio and fuel composition still doesn't exist even though a lot of effort has gone in the determination of burning velocity of simple fuels such as methane and hy- drogen. Most of the recent sophisticated experimental methodologies were accurate ones but limited to determining laminar burning velocities at ambient conditions of pressures and temperatures only. Computational techniques were limited by unavailable accurate transport properties and comprehensive chemical kinetics, even for single fuels. Out of the many experimental techniques, constant volume bomb, if used carefully, can he a versatile technique since it gives burning velocity as a function of pressure and tempera- ture with minimum effort, but with a limited accuracy due to various reasons. Since the

(8)

iv prime objective of the present investigation was to develop a comprehensive data base of laminar burning velocity of producer gas, a constant volume bomb was decided to be de- veloped. Since the producer gas was investigated for laminar burning velocity and flame structures for the first time, it was decided to develop a simple experimental facility such as an orifice burner with direct photography initially, so as to get a qualitative estimate of laminar burning velocity and also an overview of the gross flame surface structure.

Experimental facilities developed in the present investigation included an orifice burner with accessories and safety equipments such as, fuel-air mixing tube, flame arrestor, flame arrestor testing unit, flow control panel with rotameters and rotameter calibration unit.

Since the experiments were fundamental in nature, synthetic producer gas was prepared instead of drawing producer gas directly from the gasifier. For this purpose, a fuel mix- ing chamber was fabricated along with a manifold system to mix the component gases in predetermined percentage composition by partial pressure method. A gas chromato- graph was used in order to cross check the composition of fuel attained in the mixing chamber. For more exhaustive experimental investigations, a spherical constant volume bomb experimental facility was developed along with the fuel-air mixture proportioning system and the associated instrumentation. The reaction vessel was a spherical vessel of about 1 litre capacity. made of AISI 204 Stainless Steel, with an inside diameter of 124 mm and a thickness of 25 mm, designed to withstand 100 bar pressure. The mixture proportioning system consisted of a small manifold connected to the bomb, which was in turn connected to 6 gas cylinders corresponding to 5 gas components of producer gas and a zero-air cylinder for reactant preparation in the reaction vessel by partial pressure method. The pressure-time history in the bomb during combustion was measured by a piezo-electric pressure transducer along with a charge meter. An oscilloscope was used for reading and recording the pressure-time trace measured from the pressure transducer.

The methodology used in orifice burner experiments was the premixed cone angle measurement from direct photography. The methodology used in constant volume bomb

(9)

experiments utilizes pressure-time trace from the experiments to calculate flame radius and laminar burning velocity assuming adiabatic compression of the unburned gases by the propagating spherical flame. Both the experimental facilities were validated for methane-air flames, for comparing the results of laminar burning velocity obtained from the present investigations with the published literature. In addition to methane. experi- ments were also conducted in the constant volume bomb to compare the present results with that of published literature for a binary fuel mixture and a multi-component fuel-air mixture.

Experiments with producer gas-air mixtures in orifice burner showed various flame phenomena due to the effects of preferential diffusion and a severe negative stretch expe- rienced by the conical flame at its tip. Polyhedral flames, which are the manifestation of cellular flames in burners was observed at an equivalence ratio of less than about 0.8. This behavior established a hydrogen like characteristic of producer gas. Other flame phenom- ena observed were open tip flames, flames with weak tips and flames with bright orange luminous sheaths with respect to equivalence ratios. Two correlations were developed for producer gas-air mixtures which show the pressure and temperature dependency. one applicable up to a pressure of 10 bars and the other applicable up to a pressure of 30 bar, both correlations applicable up to a temperature of 475 K. The correlation coefficients were given for the tested equivalence ratio range of 0.9 to 1.4. Experiments were also conducted in constant volume bomb to investigate the effect of variation of individual fuel components, viz., hydrogen, carbon monoxide and methane on the laminar burning velocity of standard producer gas. Correlations were accordingly developed for all the cases of fuel compositions. Laminar burning velocities of producer gas-air mixtures were seen to be very sensitive to hydrogen concentration in the fuel.

Computational investigations were undertaken to quantify the effect of flame stretch and preferential diffusion on laminar burning velocity and to determine unstretched burn- ing velocities. The computational simulations of outwardly propagating spherical flames

(10)

vi were carried out using the solution of the unsteady, one-dimensional governing equations incorporated in the laminar flame computer code called RUN 1DL. This code was used to determine unstretched burning velocities from spherical flames and also to analyze flame structures. Another flame code, PREMIX was utilized to investigate freely propa- gating planar flames for the analysis of flame structures of planar flames at zero stretch rates. A detailed C1 chemistry mechanism with 97 reactions and 16 stable species was utilized for computations. The computational results showed very little effect of flame stretch on the laminar burning velocity of producer gas-air flame probably due to the excess diluent present in the fuel itself, thereby altering the transport property of the mixture through an increase in the Lewis number thereby stabilizing the flame against flame stretch and preferential diffusion. At equivalence ratio below 0.8, cellular flames were indicated from the computations similar to the noted cellular formation in orifice burner and constant volume bomb experiments. The comparison of laminar burning ve- locity obtained from computations and experiments for all the cases of fuel compoSitions tested was good. Computational investigations in fact showed that use of constant vol- ume bomb without any corrections for for stretch for the present producer gas mixtures does not involve errors in the results due to the insensitivity of this multi-component fuel mixture to the combined effects of stretch and preferential diffusion in the tested range of equivalence ratio from 0.9 to 1.4. From these experimental investigations with a back up from computations laminar burning velocities of producer gas-air mixtures were mapped for a pressure range of 1-30 bar, temperature range of 300-475 K, equivalence ratio range of 0.9-1.4 and various fuel compositions. Since these parameters are close to conditions encountered in IC engines, the present results can be directly applied to such high pressure/temperature applications.

(11)

Contents

Acknowledgements

Abstract iii

List of Figures xii

List of Tables xxii

Nomenclature xxvi

1 Introduction 1

1.1 Significance and Overview 1

1.2 Background 3

1.2.1 Flame Structure 3

1.2.2 Flame Thickness 5

1.2.3 Flame Stretch 7

1.2.4 Effects of Flame Stretch on Flame Structure 10

1.2.5 Flame Front Instability 11

1.2.6 Laminar Burning Velocity 12

1.3 Motivation for the Present Work 15

vii

(12)

CONTENTS viii

1.4 Specific Objectives 15

1.5 Organization of the Thesis 16

2 Literature Review 18

2.1 Combustion Characteristics of Mixtures 18

2.1.1 Minimum Ignition Energy 18

2.1.2 Quenching Distance 22

2.1.3 Flammability Limits 25

2.1.4 Laminar Burning Velocity 28

2.2 Experimental Investigations of Laminar Burning Velocity and Flame Struc-

ture 40

2.2.1 Experimental Techniques 40

2.2.2 Data Reduction Methodology 49

2.2.3 Fuel Variations 57

2.2.4 Parametric Variation 68

2.2.5 Discussion of Accuracies and Uncertainties 70 2.2.6 Determination of other Combustion Characteristics 72 2.3 Computational Determination of Laminar Burning Velocity 73 2.4 Status of Combustion Research at Indian Institute of Technology, Delhi . 75

2.5 Discussion 77

2.6 Investigation Strategy Adopted 80

3 Experimental Facilities Developed 82

3.1 Orifice Burner Experimental Facility 83

(13)

CONTENTS ix

3.1.1 Orifice Burner 83

3.1.2 Fuel-Air Mixing Tube 84

3.1.3 Flame Arrestor 86

3.1.4 Flame Arrestor Testing unit 87

3.1.5 Control Panel with Rotameters and Rotameter Calibration Unit . 88

3.2 Fuel Mixing Chamber with Manifold 90

3.3 Spherical Constant Volume Bomb Experimental Facility 92

3.3.1 General Requirements 92

3.3.2 Spherical Combustion Chamber 93

3.3.3 Initial Mixture Composition, State of the Reactant and Propor-

tioning System 96

3.3.4 Ignition System 100

3.3.5 Recording of Pressure-time History of Combustion 100

3.4 Closing Remarks 103

4 Experimental and Computational Methodology 104

4.1 Background 104

4.2 Experimental Methodology 107

4.2.1 Orifice Burner Experiments 107

4.2.2 Constant Volume Bomb Experiments 109

4.3 Validation Experiments and Demonstration of Calculation Methodology . 122 4.3.1 Validation of Orifice Burner Experiments 123 4.3.2 Validation of Constant Volume Bomb Experiments 123

4.4 Computational Methodology 149

(14)

CONTENTS x

4.4.1 Features of RUN 1DL Code 151 4.4.2 Features of PREMIX Code 155 4.5 Validation for Methane-Air Mixtures 157 4.5.1 Flame Stability and Evolution 158 4.5.2 Burning Velocity/Stretch Interactions 159

4.5.3 Markstein Numbers 162

4.5.4 Unstretched Laminar Burning Velocities 163

4.6 Concluding Remarks 166

5 Results and Discussion -

Experimental

167

5.1 Orifice Burner Experiments 167

5.1.1 Polyhedral Flames 168

5.1.2 Open tip flames with a luminous tail 171

5.1.3 Weak tip 172

5.1.4 Smooth cone with uniform luminosity 174 5.2 Constant Volume Bomb Experiments 179 5.2.1 Experiments with CO — CH4 Mixture 181 5.2.2 Experiments with Beech Wood Gas 183 5.2.3 Experiments with Producer Gas 190

5.3 Concluding Remarks 221

6 Results and Discussion -

Computational

224

6.1 Beech Wood Gas-Air Flames 224

6.1.1 Flame Stability and Evolution 225

(15)

CONTENTS xi

6.1.2 Burning Velocity/Stretch Interactions 225

6.1.3 Markstein Numbers 227

6.1.4 Unstretched Laminar Burning Velocities 929

6.2 Producer Gas-Air Flames 231

6.2.1 Flame Stability and Evolution 232

6.2.2 Burning Velocity/Stretch Interactions 233

6.2.3 Markstein Numbers 235

6.2.4 Unstretched Laminar Burning Velocities 235 6.2.5 Flame Structure/Stretch Interaction 237 6.3 Effect of Individual Fuel Components on the laminar burning velocity of

Standard Producer Gas 241

6.3.1 Effect of Hydrogen Variation 249

6.3.2 Effect of Carbon Monoxide Variation 245

6.3.3 Effect of Methane Variation 248

6.4 Comparison of Computational and Experimental Results 253 6.4.1 Comparison of Laminar Burning Velocity of Producer Gas-Air Mix-

ture 253

6.4.2 Effect of Hydrogen on Burning Velocity of Producer Gas 255 6.4.3 Effect of Carbon Monoxide on Burning Velocity of Producer Gas 255 6.4.4 Effect of Methane on Burning Velocity of Producer Gas 257

6.5 Producer Gas as a Fuel 257

6.6 Closing Remarks 259

7 Conclusions 262

(16)

CONTENTS xii

7.1 Conclusions 262

7.2 Scope for Future Work 266

Bibliography 268

A Properties of Producer Gas 290

B Uncertainty Analysis 291

B.1 Uncertainty in Burning Velocity 293

B.2 Uncertainty in A/F Ratio 296

C Repeatability and Standard Deviation 298 C.1 Repeatability of Pressure-time record in Constant Volume Bomb 298 C.2 Standard Deviation Calculation in Constant Volume Bomb 299 C.3 Repeatability and Standard Deviation of Gas Chromatograph Tests . . 300

D Mixing Time Calculation 303

E Rotameter Calibration Facility 306

F Curve Fit for Methane-Air Mixtures 308

G Curve Fit for Producer Gas-Air Mixtures 310

List of Publications 314

Brief Biodata of the Author 315

References

Related documents

The combustion, performance and emission characteristics of a single cylinder, four stroke, air-cooled, direct injection diesel engine having a power output of 4.4 kW at a

The effect of the gas flow rate, the moisture of the piles of coal, the humidity of the air and particle size on the spontaneous combustion characteristics of coal samples were

The first attempt to use this technique was made by Schmalzreid [31] using gas mixtures and air as a reference electrode, he reported that at lOOO'C and between 10~*^ and

In the present work, constant pressure outwardly propagating spherical flame method is used to experimentally evaluate the LBV and burned gas Markstein length (L b ) of various

In the mixture-diluent approach, the experiments were conducted considering the composition of the premixed combustible mixture as y% diluent + (100 - y)% (H2 + CO + 02). The

The values of electron iiicAility in air in presenuo of a magnetic field varying from 0 to 200 Gauss and over a wide range of pressure have been computed from

In the case of the illm ninaliug gas and air the pressure inside the discharge tube goes on increasing up to a certain value after the discharge is started

tion of a particular insulating liquid is an effective agent in determining its dielectric strength. H e also found that the effect of temperature and pressure