<%BANNER%>

Implementation and Experimental Characterization of a Novel, Fuel-Flexible, Laboratoryscale, Premixed Combustor for Preh...

Permanent Link: http://ufdc.ufl.edu/UFE0044714/00001

Material Information

Title: Implementation and Experimental Characterization of a Novel, Fuel-Flexible, Laboratoryscale, Premixed Combustor for Preheated Gas
Physical Description: 1 online resource (74 p.)
Language: english
Creator: Brissonneau, Julien
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: annular -- burner -- combustion -- combustor -- composition -- flameless -- flexible -- fuel -- lean -- nitric -- nox -- oxide -- premixed -- reactor -- stable -- syngas -- variable
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Aerospace Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The challenging, clean, plentiful and full of promise biomass energy sector motivates a global awareness in burning greatly variable fuels in gas turbine combustors. Existing combustor stability and low emission burner technologies are designed for a given fuel composition. The current study aims at implementing a fuel flexible combustion system which can be used for kinetic investigations, as well as for low purity fuels combustion. A premixed burner previously designed based on the concept of a forced recirculation by jet impingement and strong stirring has been implemented and calibrated. A numerical investigation of the combustion performance had been previously performed on a fully detailed model to define an operative range of interest. The entire system had been designed to provide this range of operation based on detailed heat transfer cooling and gas dynamics injector models. This operative range has been observed with an experimental approach and differences between the previous model and the experimental observations have been quantified, analyzed and explained. Optical and physical diagnostics capabilities have been implemented in order to provide insight into the reactor chemistry. This allows the characterization of the combustor performance. The research documented in this current paper is part of a larger research effort to investigate kinetics of alternative fuel combustion on the one hand, and of the kinetic effects of CO2 and H2O in flameless combustion on the other hand. The long term goal of such investigations is to create a variable geometry combustor with optimized operating configurations allowing perfectly stable flameless combustion for a range of variable fuel compositions. The operating principles are expected to be relevant to the design of conventional gas turbine combustors, as well as the combustion system of novel semi-closed cycle engines.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Julien Brissonneau.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Hahn, David W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044714:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044714/00001

Material Information

Title: Implementation and Experimental Characterization of a Novel, Fuel-Flexible, Laboratoryscale, Premixed Combustor for Preheated Gas
Physical Description: 1 online resource (74 p.)
Language: english
Creator: Brissonneau, Julien
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: annular -- burner -- combustion -- combustor -- composition -- flameless -- flexible -- fuel -- lean -- nitric -- nox -- oxide -- premixed -- reactor -- stable -- syngas -- variable
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Aerospace Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The challenging, clean, plentiful and full of promise biomass energy sector motivates a global awareness in burning greatly variable fuels in gas turbine combustors. Existing combustor stability and low emission burner technologies are designed for a given fuel composition. The current study aims at implementing a fuel flexible combustion system which can be used for kinetic investigations, as well as for low purity fuels combustion. A premixed burner previously designed based on the concept of a forced recirculation by jet impingement and strong stirring has been implemented and calibrated. A numerical investigation of the combustion performance had been previously performed on a fully detailed model to define an operative range of interest. The entire system had been designed to provide this range of operation based on detailed heat transfer cooling and gas dynamics injector models. This operative range has been observed with an experimental approach and differences between the previous model and the experimental observations have been quantified, analyzed and explained. Optical and physical diagnostics capabilities have been implemented in order to provide insight into the reactor chemistry. This allows the characterization of the combustor performance. The research documented in this current paper is part of a larger research effort to investigate kinetics of alternative fuel combustion on the one hand, and of the kinetic effects of CO2 and H2O in flameless combustion on the other hand. The long term goal of such investigations is to create a variable geometry combustor with optimized operating configurations allowing perfectly stable flameless combustion for a range of variable fuel compositions. The operating principles are expected to be relevant to the design of conventional gas turbine combustors, as well as the combustion system of novel semi-closed cycle engines.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Julien Brissonneau.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Hahn, David W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044714:00001


This item has the following downloads:


Full Text

PAGE 1

1 IMPLEMENTATION AND EXPERIMENTAL CHARACTERIZATION OF A NOVEL, FUEL FLEXIBLE, LABORATORYSCALE, PREMIXED COMBUSTOR FOR PREHEATED GAS By JULIEN BRISSONNEAU A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Julien Brissonneau

PAGE 3

3 To my Grand Dad

PAGE 4

4 ACKNOWLEDGMENTS I am extremely thankful to Dr. David W. Hahn, Chairman and Knox T. Millsap s Professor at the Department of Mechanical and Aerospace engineering of University of Florida for his whole advisement during my studies. He has very frequently proved his considerable pedagogic skills, scientific knowledge and natural curiosity. Dr. Hahn always took the time to provide me the support I needed to work in the best conditions. I acknowledge my sincere thanks to Dr. William Lear and Dr. David Mikolaitis for serving as members in my committee. I am very thankful to Benoit Fond, UF alumnus. Hi s good advice and support were very relevant for my project of studies in Florida. I am also graceful to my current and former labmates: Julia, Nathan, Michael, Richard, Sarah, Phil and Kris. They composed a very agreeable team to work with. Our laboratory is definitely a great place for work as well as for discussion and exchange of points of view. Finally, I would like to thank my family for encouraging me to pursue my academic ambitions.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLED GMENTS ................................ ................................ ................................ .. 4 TABLE OF CONTENTS ................................ ................................ ................................ .. 5 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Economical, Fiscal and Geo Political Implications ................................ .................. 14 Nitric Oxides Reduction ................................ ................................ .......................... 16 Synthetic Gas and Fuel Flexibility ................................ ................................ ........... 21 Well Stirred Reactor ................................ ................................ ................................ 26 2 STATEMENT OF SCOPE ................................ ................................ ....................... 28 Burner Specific Geometry ................................ ................................ ....................... 28 Operating System Design ................................ ................................ ....................... 34 3 IMPLEMENTATION ................................ ................................ ................................ 35 The Outside Air Supply Subsystem ................................ ................................ ........ 36 The Inside Air Supply Subsystem ................................ ................................ ........... 38 The Gas S upply Subsystem ................................ ................................ ................... 45 The Computer Interface ................................ ................................ .......................... 45 The Liquid Fuel Supply Subsystem ................................ ................................ ........ 47 The Thermocouples ................................ ................................ ................................ 48 The Gas Analyzer ................................ ................................ ................................ ... 48 4 EXPERIMENTAL CHARACTERIZATION ................................ ............................... 50 Starting Procedure ................................ ................................ ................................ .. 50 Lean Blowout Limit ................................ ................................ ................................ 50 Rich Blowout Limit ................................ ................................ ................................ .. 52 Nitric Oxides Emissions ................................ ................................ .......................... 52 5 DIFFERENCES BETWEEN MODELS AND EXPERIMENTAL RESULTS ............. 56

PAGE 6

6 Reaction Rate ................................ ................................ ................................ ......... 56 Inlet and Reactor Temperature ................................ ................................ ............... 57 Experimental Feedback ................................ ................................ .......................... 61 6 FUTURE WORK ................................ ................................ ................................ ..... 63 APPENDIX A RELEVANT MACHINING DRAWING ................................ ................................ ..... 66 B BIG PICTURE OF THE SUPPLY SYSTEM ................................ ............................ 70 LIST OF REFERENCES ................................ ................................ ............................... 72 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 74

PAGE 7

7 LIST OF TABLES Table page 0 1 Stoichiometry definitions ................................ ................................ ..................... 11 3 1 List of product for the compressed air supply outside the building ..................... 39 3 2 List of flow controllers for the Air and Gas supply ................................ ............... 41 3 3 List of product for the air pre heating system ................................ ..................... 44 3 4 List of materia l used for liquid fuel ................................ ................................ ...... 47

PAGE 8

8 LIST OF FIGURES Figure page 1 1 NOx reducing by exhaust gas recirculation ................................ ........................ 17 1 2 Schematic stable flame ................................ ................................ ...................... 17 1 3 Schematic unstable flame ................................ ................................ .................. 18 1 4 Transition from stable to unstable flame d ue to changes in flow conditions [5] .. 19 1 5 Burner assembly damaged by combustion instability and new burner assembly [6] ................................ ................................ ................................ ....... 20 1 6 Schematic flameless oxidation ................................ ................................ ........... 20 1 7 Single flow passage computational grid for a high speed centrifugal compressor (dim 141x49x33 in the streamwise, spanwise and pitchwise directions) [8] ................................ ................................ ................................ ...... 22 1 8 Calculated and experimental performance map of a a high speed centrifugal compressor ................................ ................................ ................................ ......... 23 2 1 Dependence of H, OH,C3H3,C2H2, benzene and naphthalene concentrations on equivalence ratio for ethylene combustion (1650K, 1 atm) .... 29 2 2 A schematic illustration of soot formation according to the ionic theory [17] ...... 30 2 3 ................................ ................................ ........... 31 2 4 Cuts of B. Fond's combustion chamber ................................ .............................. 32 2 5 Axial Cut of the combustion chamber ................................ ................................ 33 3 1 Picture of the annular combustion chamber from the top ................................ ... 35 3 2 Pictu re of the burner assembly ................................ ................................ ........... 36 3 3 Pressure evolution in the tank at 35 SCFM ................................ ........................ 40 3 4 Picture of the outside air supply system ................................ ............................. 40 3 5 Picture of the flow controllers ................................ ................................ .............. 42 3 6 Picture of the flow controllers ................................ ................................ .............. 43 3 7 Picture of the control box operational ................................ ................................ 44

PAGE 9

9 3 8 Gas supply flexible implementation ................................ ................................ .... 46 3 9 Labview interface ................................ ................................ ................................ 47 3 10 Liquid fuel supply subsystem ................................ ................................ .............. 48 3 11 Gas analyzer ................................ ................................ ................................ ...... 49 4 1 Graph ic of the Lean Limit experimentation ................................ ......................... 51 4 2 Table of the Lean Limit experimentation ................................ ............................. 51 4 3 Graphic of the Lean Limit experimenta tion ................................ ......................... 53 4 4 Table of the Rich Limit experimentation ................................ ............................. 53 4 5 NOx Measurements ................................ ................................ ............................ 54 4 6 NOx Measurements Graphics ................................ ................................ ............ 55 5 1 Adiabatic flame temperature for mixtures of CH4/O2/N2 ................................ .... 58 5 2 Composition at equilibrium for different methane air mixtures at initial temperature of 400K and a constant pressure of 1 atmosphere ......................... 59 5 3 Schematic representation of the 3 flows and the 3 tubes of the burn er's thermal modelisation ................................ ................................ .......................... 61

PAGE 10

10 LIST OF ABBREVIATION S Density (kg.m3) Dynamic viscosity (Pa.s 1) Equivalence ratio G Gas (subscript) Q H eat (J) Heat capacity ratio I Inlet (subscript) L Liquid (subscript) M Mach number M ass flow rate (kg.s 1) NO X N itrogen oxides PAH P olycyclic aromatic hydrocarbons P P ressure (Pa) A S ection (m2) Sonic condition (subscript) Source term (measured in dBA) C P S pecific heat at constant pressure (J.kg 1.K) T T emperature (K) K T hermal conductivity (W.m 1.K 1) R U niversal gas constant (8.314 J.K 1.mol 1) U Velocity(m.s 1) V V olume (m3) W Wall (subscript)

PAGE 11

11 The following definitions may become useful when reading the text. Table 0 1. Stoichiomet ry definitions Name Definition Oxygen requirement Combustion products Conventional equivalence ratio C Effective equivalence ratio CO, Carbon to oxygen ratio CO,

PAGE 12

12 Abstract of Thesi s Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IMPLEMENTATION AND EXPERIMENTAL CHARACTERIZATION OF A NOVEL, FUEL FLEXIBLE, LABORATORYSCALE, PRE MIXED COMBUSTOR FOR PREHEATED GAS By Julien Brissonneau August 2012 Chair: David W. Hahn Major: Aerospace Engineering The challenging, clean, plentiful and full of promise biomass energy sector motiv ates a global awareness in burning greatly variable fuels in gas turbine combustors. Existing combustor stability and low emission burner technologies are designed for a given fuel composition. The current study aims at implementing a fuel flexible combust ion system which can be used for kinetic investigations, as well as for low purity fuels combustion. A premixed burner previously designed based on the concept of a forced recirculation by jet impingement and strong stirring [ 1 ] has been implemented and ca librated. A numerical investigation of the combustion performance had been previously performed on a fully detailed model to define an operative range of interest. The entire system had been designed to provide this range of operation based on detailed hea t transfer cooling and gas dynamics injector models. This operative range has been observed with an experimental approach and differences between the previous model and the experimental observations have been quantified, analyzed and explained.

PAGE 13

13 Optical and physical diagnostics capabilities have been implemented in order to provide insight into the reactor chemistry. This allows the characterization of the combustor performance. The research documented in this current paper is part of a larger research effor t to investigate kinetics of alternative fuel combustion on the one hand, and of the kinetic effects of CO 2 and H 2 O in flameless combustion on the other hand. The long term goal of such investigations is to create a variable geometry combustor with optimiz ed operating configurations allowing perfectly stable flameless combustion for a range of variable fuel compositions. The operating principles are expected to be relevant to the design of conventional gas turbine combustors, as well as the combustion syste m of novel semi closed cycle engine s

PAGE 14

14 CHAPTER 1 INTRODUCTION Economical Fiscal and Geo Political Implications and to reduce the pollutant emissions, superior efficiency and low ratio of pollutant emission over power generation are one of the key drivers in gas turbine development. I n a future resource constrained world, where public opinions are aware of ecological concerns, environmental tax es are expected to become increasingly important. The financial cost of pollution could take diverse forms such as the carbon tax. It is an environmental tax levied on the carbon content of fuels a form of carbon pricing A lot of European countries have already implemented such taxes or more generally, energy taxes. Most environmentally related taxes with implications for greenhouse gas emissions in OECD ( Organization for Economic Co operation and Development ) countries are levied on energy products and motor vehicles rather than on CO2 emissions directly. In 2010, the European Commission considered implementing a pan European minimum tax on pollution permits purchased under the EU ETS ( European Union Greenhous e Gas Emissions Trading System) in which the proposed new tax would be calculated in terms of carbon content rather than volume, so that fuels with high energy concentrations, despite their subsequently high carbon content, will no longer carry the same traditionally low price. These types of environmentally friendly taxes fi nancially motivate gas turbine efficiency improvements for aircraft propulsion systems. In the United States, Cap and trade is an environmental policy tool based on emission trading that is in force. This is a market based approach used to control pollution by providing economic incentives for achievi ng reductions in the emissions of

PAGE 15

15 pollutants A central authority (a governmental body) sets a limit or cap on the amount of a pollutant that may be emitted. The limit or cap is allocated or sold to firms in the form of emissions permits which represent the right to emit or discharge a specific volume of the specified pollutant. Firms are required to hol d a number of permits equivalent to their emissions. The total number of permits cannot exceed the cap, limiting total emissions to that level. Examples of successful cap and trade programs include the nationwide Acid Rain Program and the regional NOx (oxides of nitrogen) Budget Trading Program in the Northeast. The principal pollutants associated with gas turbines are NOx Consequently, these types of environmentally friendly taxes fina ncially motivate gas turbines control strategies for NOx. Those strategies nowadays include water or steam injection and premixed burners as well as the post combustion control by installing a catalytic reduction unit It consists of reacting the NO with i njected ammonia in the presence of a catalyst [ 2 ] Control strategies incorporated within the combustion process often result in reduced combustion efficiency and thus increased emissions of carbon monoxide and unburned hydrocarbons. Consequently, the rema ining goal for the gas turbines designers is to increase the combustion efficiency while reducing the polluta nts formed within the combustor. Furthermore, with the current changes in the fuel sources, burners should now be adapted to alternative fuels besi des decreasing pollutant generation. Indeed, recent modifications in our approach of the fuel sources bring new considerations. Stability in the combustion system is a decisive point on aircraft combustors in which a flame should be controlled over miscell aneous conditions during the flight. The localization and diversity of the alternative fuel source candidates in ground power generation also

PAGE 16

16 requires the development of stable fuel flexible combustion systems. Thus, t he logical gas turbine market expectat ion for new products is to maintain reliability with a low environmental impact and a high level of fuel adaptabilit y. Nitric Oxides Reduction As we explained the emissions abatement is of primary significance for both power generation and aircraft propu lsion systems. Nowadays, combustors use NOx r educing techniques This includes special forms of combustion such as new flameless configurations [ 3 ] which allows us to avoid temperature peaks within stabilized flames in order to mostly suppress NO formation Different mechanisms have been identified in the generation of Nitric oxides such as the Zeldovich [ 4 ] or thermal NOx mechanism. Overall, NOx emissions are strongly linked to the flame temperature. There are three predominantly sources of nitric ox ides f rom combustion processes: prompt NO, fuel NO and thermal NO. The main source is the thermal NO formation, described by the The three principle reactions for this formation are : As indicated in its name, the higher the temperatu re is the more significant the thermal NO formation is. Significant NO emissions can be found if oxygen containing combustion products are exposed to temperatures:

PAGE 17

17 Because of that strong temperature impact, most NO reducing techniques try to remove peak temperatures by keep ing the residence time in high temperature areas low and by avoid ing high oxygen concentration in these zones. Flameless oxidation has been variously termed: dilute combustion, moderate and intense low oxygen dilution (MILD), lean comb ustion, homogeneous combustion and low NOx injection. Figure 1 1 NOx reducing by exhaust gas recirculation Contrary to the combustion in stabilized flames, the combustion at flameless oxidation is temperature and mixture controlled, reached by prec ise flow and temperat ure conditions. A stable flame requires a balance between flow and flame velocity. This is true for premixed as well as diffusion flames. Figure 1 2 Schematic stable flame Stable flames are conceivable over the whole range of combu stion chamber temperature but only for recirculation rates up to 30% For higher recirculation rates Combustion C ooling of combustion products Air Fuel Recirculated exhaust gas

PAGE 18

18 b elow the mixture self ignition temperature th e flame will become unstable lift off and finally blow out. Figure 1 3 Schematic unstable flame C ombu stion instability also referred to as unsteady flow oscillations, is a common problem, and slowed down the development of lean premixed combustors These oscillations may reach sufficient amplitudes to interfere with engine operation, and in extreme cases, lead to failure of the system due to excessive structural vibration and heat transfer to the chamber. The narrow range of stability offered by premixing compared to diffused burners makes lean premixed technology difficult to implement. However, i f the f urnace temperature and the exhaust gas recirculation is sufficiently high, the fuel can react in the very steady, stable form of fl ameless oxidation Emission reductions could be realized through the d ry l ow swirl stabilized combustion [7]. Moreover, the f uel to air ratio in the flame locally impacts the mechanism of soot formation. Thus, large gradients in the composition of the mixture increase the formation of soot precursors. Mixing the air and the pre vaporized liquid fuel upstream of the injector crea tes a more homogenous mixture decreasing significantly the amount of soot precursors formed.

PAGE 19

19 Figure 1 4 Transition from stable to unstable flame due to changes in flow conditions [ 5 ]

PAGE 20

20 Figure 1 5 Burner assembly damaged by combustion instability and new burner assembly [ 6 ] Figure 1 6 Schematic flameless oxidation On the figure 1 6, a schematic representation of flameless oxidation, the gas is supplied axially through the central gas nozzle and the air is injected through concentric arranged nozzl es. Because of the distance of gas and air supply, the reaction could only take place further downstream, when a large amount of exhaust has already mixed in. An additional objective in aircraft propulsion engines development, very conventional, is to scal e down the combustion systems. Basically, we are looking for the highest heat release in the smallest volume. By reducing the weight, we also reduce the overall fuel

PAGE 21

21 consumption. But then we create an extra difficulty for the cooling system. Indeed we have to avoid local temperature peaks in a small volume with high heat release. With our experimental combustor we need to be able to identify a flameless combustion configuration. Relevant characterization measurements for this type of combustion are temperat ure and NOx concentration. Arbitrary, while burning methane in our burner, we will be considering flameless configuration when both measured temperature is less than 1200 C at the exit of the annular space and measured density of NOx particles at the exit of the burner is below 5 ppm. Synthetic Gas and Fuel Flexibility As we mentioned it, the contemporary energy context offers a stimulating force for exploration of alternative sources of fuels such as biomass and coal as fossil fuel supplies decline. An en vironmentally friendly alternative gaseous fuel for internal combustion engine mainly consisting of carbon monoxide (CO) and hydrogen (H2) called syngas is nowadays able to substitute fossil diesel oil in combustion engines. Nevertheless, coal derived synt hesis gas like syngas present combustion characteristics relatively different from deeply examined methane. In addition fuel sources will possibly become progressively local as the diversification of sources becomes a truth, and thus the composition of the syngas is likely to be highly changeable. Hereafter there are durable motivations to develop technologies to enable the safe and efficient use of alternative fuels in aircraft s as in other transportation and power generation engines. The premixing concep t used in dry low NOx systems is particularly sensitive to the composition of the fuel used. We already briefly described how changes in fuel composition impact the emissions and heat release performance

PAGE 22

22 inside the burner, but it also has an influence on t he other components of the fluid path such as the compressor and the turbine. Indeed, increase in air flow rate due to a lower heating value of the burning mixture may affect the compressor performance. This is easily observable such as in the c omputationa l a nalysis of c entrifugal compressor surge c ontrol u sing a ir i njection realized by A Stein, S. Niazi, and L. N. Sankar [ 8 ] They used as a specific confi guration to investigate instabilities a high speed c entrifugal compressor They developed and used a N avier Stokes solver for simulating unsteady viscous fluid flow in turbo machinery components to study fluid dynamic phenomena that lead to instabilities in centrifugal compressors. These studies indicate especially that large fl ow incidence angles, at redu ced flow rates, can cause boundary layer separation near the blade leading edge. Figure 1 7 Single flow passage computational grid for a high speed centrifugal compressor (dim 141x49x33 in the streamwise, spanwise and pitchwise directions) [ 8 ]

PAGE 23

23 Additio nally, the c omput ational a nalysis of this c entrifugal c ompressor gives us an idea of the influence of a change in flow rate over the compressor performance both through a simulation and experiments. This is illustrated in figure 1 8 and should definitely b e considered to fully take advantage of the novel fuel flexible technology burner hypothetically implemented into a gas turbine engine. Figure 1 8 Calculated and experimental performance map of a a high speed centrifugal compressor In the same way, cor rosive products might also be an issue in material wear when designing a gas turbine with a fuel flexible burner. In premixed systems, the oxidation chemistry plays a significant role. As a result, operability and emissions performances are greatly influen ced by the gas composition. In these premixed systems, the mixture of air and fuel is flammable far upstream of the injection point. As a consequence, the

PAGE 24

24 ignition delay has to be sufficiently controlled not to let the ignition arise in an unpredicted area of the combustor. Unexpected ignition could produce potentially critical effects. An even more important critical accident that we have to avoid while implementing our burner is a flashback on the air supply lines. Investigations have been made on the che mistry dynamics in order not to accidently reach any combustion limits (excess of lean or rich combustion) while slightly changing operating conditions. An experimental approach of these investigations is described in the fourth chapter various exotic gas through diverse experimental set ups and compared with computer [1] Datas on the oxidation and pollutant formation kinetics could be extracted from diverse measurements such as flame blowout, ignition delay and laminar flame speed. Shock tube experimentations have been studied broadly and ignition delay relationships have been extracted for several fuels over a wide range of conditions. The ignition delay is relevant to consider in the implementation of the fuel flexible combustor since this delay is significantly impacted by changes in the gas mixture composition. Basically, an auto ignition phenomenon [ 9 ] at an undesired place could cl early damage the combustor. Z. ZhenLong and coworkers have developed correlations for the ignition delay times of H2/air mixture [ 10 ] The hydrogen/air ignition delay times for initial conditions over a wide range of temperatures from 800 to 1600 K, pressu res from 0.1 to 100 atm, and equivalence ratios from 0.2 to 10 were estimated and modeled using the method of high dimensional model representations. These specific auto ignition delays have been very well investigated. D. M. Kalitan, J.D Mertens, M.W. Cro fton and E. L. Petersen have studied ignition

PAGE 25

25 delays of CO/H 2 /air mixtures behind reflected shockwaves at atmospheric conditions and elevated pressures [ 11 ]. They paralleled their experimental measurement s to existing kinetic models. This kind of considera tions appears to be an essential precursor to the use of synthesis gas in power generation plants on a wide scale. Undeniably H 2 and CO have especially low flammability limits and the combustion rates of a synthesis gas mixture are strongly associated wit h the H 2 /CO ratio. Shock tubes experiments results are obviously coming from transient process experiments, but ignition delays can similarly be obtained from constant flow devices and rapid compression measurements machines. The laminar flame speed model for premixed laminar flame combustion contains information on the kinetics of the combustion of the mixture. It also contains information on the diffusivity of its species, and related thermo chemistry [ 12] Natarajan and coworkers burned CH 4 /H 2 /CO mixtur es in Bunsen burner and wall stagnation flame reactors. They achieved flame speed measurements with a charge coupled device camera on the Bunsen flame, and laser Doppler velocimetry on the stagnation reactor. Premixed mixtures went from 5 95% for H2 and CO and up to 40% for CH4, and measurements have been done on preheated mixtures up to 700K, and at pressures ranging from 1 to 5 bars. Measurements could then be compared with some detailed methane oxidation mechanism such as the Gas Research Institute mecha nism version 3 (GRI Mech 3.0). Detailed mechanism provides with kinetic rates of elementary reactions such as thermo chemistry of the species involved. The GRI 3.0 describes a detailed mechanism appropriate to gas natural combustion. It includes the NOx fo rmation and reduction but does not describe soot formation [ 13 ]. In addition, lean

PAGE 26

26 blowout measurements are often done on premixed pre vaporized swirling flame combustor devices. Based on the concept of chemical time limited reaction, the residence time of the reaction at blowout could be linked to the reaction rates and compared with a reactor network model. Zhang et al. studied the effect of the composition of H 2 /CO/CH 2 mixtures on lean blowout in a premixed gas turbine combustor. Nevertheless, a nonhomog eneous reaction zone will produce other combustion regimes than the lean blowout theory such as flamelets into eddy. Well Stirred Reactor Longwell and Weiss designed the first device approaching the well stirred reactor concept in 1953 [14]. In order to s eparate the convective and diffusive physical processes from the chemical reactions; a n appropriate mixing must be reached such as in a well stirred reactor. Overall, t he well stirred reactor concept is largely used as a modeling tool to study reactor beha vior. In this type of reactor, the composition is assumed homogeneous. Consequently, the exhaust mixture composition is the same as the composition in the burner. The assumption of homogeneity is relevant in the case of our reactor since mixing forces gen erated by the velocity jets are significant. The idea is that a residence time of the gases inside the reactor could be extracted which would put together the influence of the incoming flow, the temperature, the pressure inside the reactor and the volume o f the reactor. This residence time could be given by the expression below:

PAGE 27

27 This time represents the delay acceptable for the chemical reaction to progress toward completeness; in this case the maximum reaction rate is preceding blowout. Experimental b lo wout measurements were performed within our burner for rich and lean mixtures and compared to theoretical ones The numerical well stirred reactor model of our burner previously established and described in chapter 5 is considering the burner within a stea dy state with a steady incoming flow and uniform properties over its volume. The modeling equation s used are derivate from the Reynolds Transport Theorem, respectively the continuity equation, the energy equation and the rate laws. They could be expressed as b elow: In chapter 4 we will observe the direct influence of th e residence time over the NOx particles emission when our combustor is running under stoichiometric conditions

PAGE 28

28 CHAPTER 2 STATEMENT OF SCOPE Burner Specific Geometry The geometry of thi [1] allows a sufficient amount of back mixing in order to stabilize the flow and to provide a good homogeneity. It has been designed in order to let the air and fuel flows fully mix before combustion. This is part of the reduction in CO and NOx emissions in new gas turbine combustors strategy introduced before. Along with low CO and NOx emissions, the significant low gradients in fuel/air fraction of the mixture entering the primary zone also prevent t he formation of hot spots responsible for the formation of polycyclic aromatic hydrocarbons (PAH). Studies on premixed laminar flames showed that soot occurred only beyond a highly rich equivalence ratio [ 14 ][ 15 ]. The figure 2 1 illustrates this point, and gives us an illustration of the kind of impact on soot formation a rich equivalence ration could have on the chemistry of combustion. Developments in modeling soot formation and burnout in combustion systems are surveyed [ 16 ] The types of models are div ided up into three classes: empirical, semi empirical and detailed. Empirical models use correlations of experimental data to predict trends in s oot loadings. One of these models, t he ionic theory for soot formation is a theory primarily advocated by Calco te [17] Calcote and coworkers have supported the prominence of a link between ions in flames and soot formation. Excellent arguments and data have been presented showing a strong connection between soot and ions. This is schematically illustrated in Figur e 2 2 The theory s uggests that formation of soot origins from a sequential growth of ions starting with the ion s C 3 H 3 +

PAGE 29

29 Figure 2 1 Dependence of H, OH,C3H3,C2H2, benzene and naphthalene concentrations on equivalence ratio for ethylene combustion (165 0K, 1 atm) The first three reactions of for soot formation have fully been studied The firs t of the three reaction s is reversible and involves the formation of a specific encounter complex sensitive to pressure and ion kinetic e nergy. The second reaction appears to require large amounts of internal energy in the ion in order to

PAGE 30

30 proceed. The third reaction is reversible; nevertheless, in contrast to the initiating reaction, t he C 5 H 3 + ion formed from the [ C 7 H 5 + ]* complex exhibits a much lower reactivity. However, premixing also brought stability issues in homogeneous mixtures, particularly as they operate near lean blowout. Stabilization techniques, such as bluff body and swirlers have been seen as a way to achieve significantly stable operation at low equivalence ratio. Re circulating combustion gases transfer heat to the cool mixture, raising it to the auto ignition temperature. The stabilization performance depends on the time for the cold mixture to ente r the shear layer limiting the recirculation zone, the rate of entrainment of fresh mixture into the recirculation zone, and the residence time of gases inside the recirculation zone. Figure 2 2 A schematic illustration of soot formation according to t he ionic theory [ 17 ]

PAGE 31

31 However, in swirl stabilized flames, vortex precession and breakdown phenomenon would be an important source of combustion instabilities so hat active boundary control and Helmholtz resonators are often employed in current dry low NOx technologies [1] The idea behind the current design is to provide back mixing based on jet impingement and flow split rather than large vortices structures. Figure 2 3 Normal c ut of B. s been elaborated to provide this low velocity recirculation zone. This geometry is shown on Figure 2 3 and Figure 2 4.

PAGE 32

32 Figure 2 4 Cuts of B. Fond's combustion chamber The annular combustor presented here is made of a hollow cylinder liner with a wall on one side and the exhaust on the other side. A smaller tube is mounted and the mixture is injected almost radially somewhere along the length of the liner. The fresh mixture will come in at a relatively important speed into the central tube so that part of it will be entrained near the rear wall and the other part will go in the direction of the exhaust where it will be later diluted with the liner cooling air. This design will provide a cross flow between the injection stream and the combustion products coming from the back of the reactor and so the heat transfer and mixing will be enhanced. In this design two parameters were incorporated to address geometry influences. The inner tube can be replaced, changing considerably the jet impingement configuratio n as well as the combustor volume. In addition, the injectors are cylindrical and incorporate a slight 10 injection angle with respect to the radial direction so they

PAGE 33

33 can be rotated to act on the flow split as show in figure 2 4. The swirl number is defin ed in the equation: S = Where U and U z are respectively the tangential and axial velocities, r i and r o the inner and outer radius of the annular space. Considering the cold gas injection velocity and the combustion product axial velocity, the swirl number could be approximated at 10. However, the high velocity of the cold mixture jet imply a stirred reaction zone more than a swirling flow. In this design, air and fuel are injected at four different points around the circumference and the jets are directed toward the center tube. The number of injectors has been seen as the simple way to provide a good homogeneity and a good control over the reactor composition by opposition to a liner with holes. It revealed stability performance. Figure 2 5 Axial Cut of the combustion chamber

PAGE 34

34 O perating S ystem Design This novel combustion rig is located in the building 241 at the Research energy park of University of Florida. It includes all gases and fuel supply lines and an electric gas exhau st chimney. On its way to the mixing section, air is compressed, stocked, metered, and preheated. The liquid fuel will be pumped, metered, heated and sprayed into the combustion air stream. The mixture of syngas and combustion air is injected in the combus tor primary zone. The flexible gas supply system allows us to simulate exhaust gas (CO2) injected with syngas. Some of the air goes directly to cooling passages (outer and inner flows) before to be mixed with the combustion products. Each single line is mo nitored in real time via Labview. After combustion, the streams are exhausted outside the building. The different systems designed and implemented around the combustion chamber are presented in supply components were d efined with respect to the combustor capacities. The four injection angles were originally chosen horizontal and clockwise as a compromise between a good tangential distribution of the flow and a jet impingement speed.

PAGE 35

35 CHAPTER 3 IMPLEMENTATION The In or der to supply the combustion chamber, various materials have been implemented. The whole system could be described through subsystems such as: The outside air supply system including the compressor, dryer, fliters The inside air supply including v anes, m onitored mass flow controllers, preheaters and premixing systems inside the building The gas supply system including its flexibility and adaptibility aspects, vanes, monitored mass flow controllers, and premixing components The liquid fuel subsystem inclu ding pumps and preheaters The computer interface The thermocouples measuring temperatures of the burner The gas analyzer measuring the composition of the exit gas Figures 3 1 and 3 2 are pictures of the burner assembly. The appendix B represents a schemati c map of the whole implementation. This chapter details the design of these subsystems. These were previously introduced in B. Fond research work [1] Figure 3 1 Picture of the annular combustion chamber from the top

PAGE 36

36 Figure 3 2 P icture of the burn er assembly The O utside A ir S upply S ubsystem The air supply has been designed to feed the combustion chamber with a constant flow whose composition, temperature and pressure are perfectly controlled. As we already mentioned it, at the premixed combustion s ection the fuel and air are mixed. The mixture arriving into the combustion zone is as homogeneous as possible. As also stated before, premixed combustion raised the problem of the presence of a flammable mixture downstream of the injector.

PAGE 37

37 The list below [1] : Delivery of the combustion air flow to the four injectors (2 20g/s) Delivery of the cooling air to the cooling passages (inner and outer) Heating of the injection air to the desired tempera ture (400 700K) Atomization, Vaporization and mixing of the fuel in the air mixture Precise control over the four injector mixture composition. The air supply line starts outside the building with the compressor pumping air to pressurize the outside tank w ith dried air. The maximum pressure the tank could support is 175psi. The compressor has been chosen according to the maximum gas flow needs in the laboratory site. We made the technologic choice to dry the air before letting it go into the tank. This allo ws us to stock already dried air but implied to lose the insurance of having freshly dried air. In order to make sure that stocked air stays dry we implemented a timer activated electric moisture drain valves on the drain exit of the tank. The pressurized air supply is currently available for the whole building; it can now serve for other activities in the laboratory such as for example powering other combustion based experiments. The idea at this point was to work on a duty cycle. Actually, the compressor is turned on immediately when the pressure in the tank drop under a specified charge. When the maximum pressure is reached the compressor is simply turned off. When we are operating the rig, the air flow is drawn from the tank on a continuous basis. Conseq uently, the pressure keeps decreasing in the tank until a specified minimum pressure is reached and then the compressor is turned on. The compressor could be modeled as a volumetric pump which means that the volumetric flow rate is constant and the pressur e determined by the receiving tank pressure [1] The tank could be represented as a constant volume system undergoing change in pressure.

PAGE 38

38 A simulation of the pressure evolution in the tank is shown on Figure 3 3, which shows that the compressor could be t urned on only 8 times when 35 SCFM are drawn to the combustor assuming a 200 gallons tank. An explicit Euler scheme could be used to represent the pressure evolution in a tank described below: Where With V T being the total volume of the in jector and (t) being equal to the volumetric flow rate of the pump when turned on or zero if turned off. The maximum pressure in the tank is set to 180 psi, which is the tank constructor rating and the minimal pressure at 80psi, so that the equipment dow nstream does not see too large of pressure variations. A dryer is used in line with the compressor to remove moisture and cool down to 50F along with two high efficiency filters to remove small particles. The tables 3 1 is a list of the products that were ordered and implemented in order to realize this outside compressor part of the air supply line. The figure 3 4 is a picture of the operational outside air supply system. We can see the Quincy compressor with its included little tank on the left, the drye r in the middle and the main tank on the right. The Inside Air S upply Subsystem The air supply subsystem inside the building has been implemented for an accurate control over the lines (gas and air) as well as for flexibility. This flexibility is

PAGE 39

39 allowing staging and study of dissymmetry influences on stability. This also permits the simulation of injection of recirculated gas into the burner. The four injectors composition are controlled separately for multiple reasons. The first reason is the capacity to run on a staged combustion regime on the long term, where two injectors run rich and two injectors run lean. The second reason is that one can study the effect of a perturbation of one injector on the global reaction regime, especially at low residence tim es. The last reason is the flow of air and fuel being fairly small; it would be difficult to have the composition fairly equal between the four injectors using some manual vanes. Table 3 1 List of product for the compressed air supply outside the build ing Product Supplier Description QT 7.5 7.5.80 Quincy Simplex air reciprocating compressor24.2 CFM at 175psi, mounted on an 80 gal. vertical tank QPHT 50 Quincy High inlet temperature refrigerated air dryer 40F 40 SCFM with aftercooler and coalescing air filter. CPNT 00030 Quincy Standard coalescing filters at 30 SCFM and 100psig, 99% at 0.02 micron and with pressure difference gauge. 5018260 200 Quincy 120 gallons receiver tank 9831K13 McMaster Clog Resistant Timer Activated Electric Moisture Drain Valves

PAGE 40

40 Figure 3 3 P ressure evolution in the tank at 35 SCFM Figure 3 4 Picture of the outside air supply system

PAGE 41

41 The implementation of divers e fuels is considered in this project. The fuels considered are methane, hydrogen and carbon monoxide. We also consider to be running in a diluted regime with mixtures of carbon dioxide, nitrogen and water vapor. The controllers have been chosen so that th eir total number is minimized, and give at the same time the maximum flexibility over the composition of the four injectors. The gas mass flow controllers selected are already calibrated for all the gases that we are using. These are controlled from a comp uter interface and allow various control strategies in order to permit accurate variations in the flow range. The table 3 2 is a list of the flow controllers implemented on this project. The figures 3 5 and 3 6 are picture of the actual flow controllers on ce implemented. Another aspect of the inside air supply subsystem is that the air is preheated. Design specifications for the combustion system included preheating the combustion air to a temperature up to 700K for partial load (7g/s) and 400K for all load s. Table 3 2 List of flow controllers for the Air and Gas supply Fluid Type Flows Flow rates Controller Ref. Quantity Air Cooling central 0 10g/s 500SLPM MC500SLPM 1 Air Cooling annular 0 20g/s 1000SLPM MC1000SLPM 1 Air Combustion reactant 2 30g/s 1500SLPM MC500SLPM 4 Methane Combustion reactant 0 1.5g/s 150SLPM MC50SLPM 2 Methane Combustion reactant 4 0 0.5g/s 50SLPM MC250SLPM 2

PAGE 42

42 Figure 3 5 Picture of the flow controllers Given cost considerations about flexible air lines able to handle such high temperatures, our choice was to use flexible silicone line and to limit as a first step the max preheating temperature of air to 630K. The available supply of air is about 750 SLPM which correspond to about 15g/s. With the four injectors being i ndependently controlled, and the flow controllers not being able to withstand these temperatures, four independent heaters were used. The energy equation could be used to estimate the power required:

PAGE 43

43 The air does not come directly from the dryer as ment ioned before, it comes from the tank. Based on a 293K inlet temperature from the tank, the partial load at 630K requires 592 W and the full load at 400K, 403 W. Figure 3 6 Picture of the flow controllers The heaters are controlled on an ON/OFF basis us ing an auto tunable PID CN132 from Omega. These consider the difference between the set point and the signal of a type K thermocouple at the exit of the heaters. This controller being auto tunable, it sets automatically the 3 constants of the proportional integral derivation action. The table 3 3 is a list of components realizing the preheating function. Figures 3 5 and 3 7 are pictures where one can observe of these components implemented. It is imperative for safety purpose to look at the way we built the pre heating pipes. The thermocouples are

PAGE 44

44 located at the exit of these pipes. This means that if the amount of air flowing through the pre heated air lines is not significant the measured temperature will not be relevant. Thus a dangerous overheating of th e air supply lines could occur without any significant regulation from the PID controller. In order to avoid this situation, we decided to program a safety minimum injected air flow of 4 SLPM per line into the computer interface. Table 3 3. List of produ ct for the a ir pre heating system Quantity Device Supplier Model Description 4 Air Heater Omega AHP 7561 T Type heaters (750W) 4 Thermocouple Omega KQSS 116G 6 Type K thermocouples 4 Heater controller Omega CN132 PID Controllers 4 Controller relay Omeg a SSRL240DC10 Solid state relays Figure 3 7 Picture of the control box operational

PAGE 45

45 The G as S upply S ubsystem We installed flow controllers and triple way vanes such as shown on Figure 3 8 in order to realize the gas supply subsystem. This configurati CH4 such as H2 into the burner, but also to premix these gases in order to inject recomposed synthetic gas for example. It also gives us the possibility to inject CO2 which could be combined with preheated air to simulate exhaust gas recirculation. Given the high flammability of the gas that we are using, we secured the combustor gas supply lines. A kill switch has been installed next to the r ig, it allows us to stop the gas lines by shutting of the mass flow controllers power supply. It comes in addition with easily accessible physical vanes on the side of the rig table and numerical vane on the Labview interface. The Computer Interface The so ftware used to create the computer interface is Labview. The current interface includes elementary safety options. Thus, the stop button acts directly on the gas supply lines to shut down the fuel supply without preventing the air from flowing through the combustor. The air flow controllers command board also includes a lower limit of air flowing to avoid the air preheating system to burn the pipes. Indeed with no air flowing the thermocouple located at the exit of the heating pipes would send a wrong tempe simply melt as described before. The Figure 3 9 shows the labview interface.

PAGE 46

46 Figure 3 8 Gas supply flexible implementation

PAGE 47

47 Figure 3 9 Labview interface The L iquid F uel S upply S ubsystem This subsystem has been realized using peristaltic pumps from Fisher Scientific and temperature c ontrol material listed in tables We can observe the material of the subsystem on Figure 3 1 0 A description of the material used is given Tab le 3 4 Table 3 4 List of material used for liquid fuel Quantity Device Supplier Model Description 4 Flexible Heater for liquid fuel Omega STH051 020 Heavy insulated heater tapes 156W 4 Peristaltic pump Fisher Scientific 13 876 2 Medium flow(0.8 85mL/ min) 2 Thermocouple fuel Omega TMQSS 040G 6 Type T thermocouples 2 Heater controller Omega CN132 PID Controllers 2 Controller relay Omega SSRL240DC10 Solid state relays

PAGE 48

48 Figure 3 10 Liquid fuel supply subsystem The T hermocouples A mobile thermoco uple able to reach the exit of the inner cooling flow, the outer cooling flow and the premixed air and gas flow was implemented. 3 thermocouples were implemented into the outer cooling flow. 2 thermocouples were implemented in contact with the external sur face of the outer tube. The inner cooling flow, the premixed gas/air flow, the outer flow, the inner tube, the outer tube and the rig case tube are shown on Figure 5 3. The G as A nalyzer A mobile gas analyzer has been used during experiments in order to mea sure the gas composition at the exit of out burner. For reasons previously described, we are particularly interested into measuring the NOx concentration. Limitations due to this device had to be taken into consideration, including the maximum temperature limit of 600 C for the probe on a long run. The Figure 3 1 1 shows the gas analyzer used for experimentations.

PAGE 49

49 F igure 3 11 Gas analyzer

PAGE 50

50 CHAPTER 4 EXPERIMENTAL CHARACTERIZATION Starting P rocedure The burner has to be started following a specific proce dure. Indeed, its geometric design does not allow us to start it directly with a stable flame well positioned into the annular space. When we turn on the burner, the flame has to be located at the exit, it implies that the initial ratio of gas injected ove r air injected has to be higher than stoichiometric. In addition, while running under rich conditions, no cooling flow (or extremely low inner cooling flow) is recommended. Once the flame stable at the exit of the burner, we have to wait few minutes to war m up the combustor. After this short delay, we have to slowly change the ratio of gas injected over air injected to come closer from the stoichiometric ratio. We are then able to observe the transition of the flame from the exit to the annular space. Lean B lowout L imit In order to observe clearly the lean operating limit of the burner for CH4, we fixed a total methane flow at 8 SLPM and the injected air flow at the stoichiometric corresponding flow. Then, we slowly increased the total injected air flow unt il the flame blowout. The results of the lean limit experimentation are detailed Figure 4 1 and Figure 4 2

PAGE 51

51 Figure 4 1 Graphic of the Lean Limit experimentation Figure 4 2 Table of the Lean Limit experimentation

PAGE 52

52 Rich B lowout L imit With the i ntention of observing the rich operating limit of the burner for CH4, we fixed a total methane flow at 8 SLPM and the injected air flow at the stoichiometric corresponding flow. Then, we slowly decreased the total injected air flow until the flame blowout. The results of the rich limit experimentation are detailed Figure 4 3 and Table 4 4 On the Figure 4 3 we can observe that the temperature is going down when the operating ratio is close from the blowout one. On the Figure 4 1 it is hard to distinguish an y relevant temperature profile. The reason is that during the experimentation the flame is moving with respect to the thermocouple. Indeed, while running lean the flame is transitioning from the annular space to the exit of the burner while the temperature probe does not move. Nitric Oxides E missions With the purpose of observing the density of NOx at the exit of the burner running stoichiometric, we analyzed the composition of the exit gas. The results of these observations are available in Figure 4 5 and Figure 4 6 We can observe how NOx emissions decrease as the injected flows decrease. This is due to the fact that the mixture residency time into the annular space increases. We can also clearly notice that NOx emissions drop when cooling flows are increa sed

PAGE 53

53 Figure 4 3 Graphic of the Lean Limit experimentation Figure 4 4 Table of the Rich Limit experimentation

PAGE 54

54 Figure 4 5 NOx Measurements

PAGE 55

55 Figure 4 6 NOx Measurements Graphics

PAGE 56

56 CHAPTER 5 DIFFERENCES BETWEEN MODELS AND EXPERIMENTAL RESUL TS Reaction Rate The collision theory and empirical results provide the reaction rate expression for a bimolecular elementary reaction step: in which the rate of reaction is: and k is expressed by: where A and b are constants and E a is the activatio n energy. If the equivalence ratio, flow of the reactant, and temperature varies significantly in the flame, the access to kinetic rates is challenging. The idea for a basic estimation would be to have an homogeneous mixture of vaporized fuel, air, diluent s and products at a controlled temperature all over the reactor volume to facilitate comparison of the product composition (emissions) with a zero dimensional set of inputs (temperature, flow of air, equivalence ratio, dilution ratio, pressure) for a given fuel. The values experimentally obtained could now be compared to computer simulation results based on a chemical mechanism and the well stirred reactor theory. The reaction should be reaction rate limited. An ideal combustor would have its mixture immed iately vaporized and mixed once it goes into the reactor volume. We are close from reaching this objective when using preheated fuel and then spraying it into a

PAGE 57

57 vaporization chamber where it mixes with the air at an high temperature. Within an adequate del ay between the spray and the injection, the liquid fuel totally vaporizes. Then the mixing into the combustor chamber is realized in jet stirred reactor through an important amount of small orifices creating some sonic jets, stirring the reactor vigorously In a gas turbine system, the sonic condition is difficult to reach. Actually, the negative pressure gradients on the compressor liner wall due to a negative relative pressure lead to wall buckling. In effect, the liner is generally thin for weight reason s and the buckling limit is reached before the sonic regime. Inlet and R eactor T emperature The inlet temperature is a significant parameter for the stability of the reactor as well as the flame temperature or reactor temperature in a well stirred system. I ndeed, in a well stirred reactor, the composition is homogeneous and so is the temperature. This is an ideal model; nevertheless, in reality it can depend on heat transfer calculations to predict the limitation in terms of wall cooling and heat losses. Ind eed the higher the flame temperature, the higher the heat transfer rates and the better the cooling should be to maintain the material of the combustor at a reasonable temperature. Considering the benefits of higher temperature, the hotter the inlet mixtur e, the more stable will be the flame. However for an air/fuel mixture in an ideal well stirred reactor, the increase in inlet temperature implies an increase in flame temperature. In theory, within a sufficient residence time, a well stirred reactor reache s the adiabatic flame temperature. Considering the reaction as adiabatic and that chemical equilibrium reached, the entire heat release should raise the mixture to the adiabatic flame temperature. In Figure 5 1, the adiabatic flame temperature is plotted a t constant pressure for CH 4 /O 2 /N 2 mixture for different mixtures and inlet temperatures.

PAGE 58

58 Calculations were done using Cantera, on a GRI3.0 mechanism with fixed enthalpy and pressure by B.Fond [1] For a given oxygen to nitrogen ratio, the adiabatic flame t emperature is maximum at a near stoichiometric mixture (O 2 /CH 4 =2). Indeed, the maximum heat of reaction per unit mass is obtained as almost all of the reactants are consumed to give CO 2 and water. Therefore the flame temperature versus equivalence ratio cu rve is a bell whose top is near the stoichiometric point. The composition of the equilibrium mixture is shown on Figure 5 2. One can observe that for a stoichiometric mixture the products are essentially CO 2 and H 2 O and N 2 and so the heat of reaction is m aximized. Figure 5 1 Adiabatic flame temperature for mixtures of CH4/O2/N2

PAGE 59

59 Figure 5 2 Composition at equilibrium for different methane air mixtures at initial temperature of 400K and a constant pressure of 1 atmosphere On the lean side, one sees th at a good part of the oxygen remains in the products, and so the inert oxygen consumes some of the reaction enthalpy to reach the adiabatic flame temperature. As one goes toward rich mixtures, the concentration of CO which enthalpy of formation is lower th an CO 2 becomes predominant. For this reason, the combustion could be considered as incomplete as not all the CO is completely oxidized in CO 2 following the reaction: The water gas shift equilibrium is also involved on rich combustion leading to the forma tion of H 2 in the products. The reaction involved is: Even though the mass fraction is low, the lightness of hydrogen hides a large molar fraction of 17.2%. As one goes toward a pure oxygen/fuel mixture, there is little or no inert species, and so less h eat is needed to raise them to the final temperature, and therefore the remaining reacting species are heated to a higher temperature.

PAGE 60

60 On the other side, as level of dilution by inert gases such as nitrogen is increased, the adiabatic temperature drops di rectly linked to the thermal effect of dilution. This type of combustion is of particular interest since the reactor is at a reasonable temperature and the gradient between the inlet temperature and the adiabatic flame temperature is fairly small promoting a homogeneous reaction zone. This is one of the ideas behind flameless combustion or moderate and intensive low oxygen diluted combustion. However the kinetic rates are very slow at 400K in a diluted regime, which could not be observed from adiabatic temp erature results. Therefore, the flameless combustion concept is also called high temperature air combustion. The goal is to have a high inlet temperature with a small temperature rise due to combustion. This high inlet temperature is usually obtained by an exhaust gas recirculation as mentioned previously in this thesis. The hot exhaust gases are cooled down to 700 800K mix with the fresh air and are injected back into the combustion chamber. We maximize the heat extracted from the combustion but we lower t he combustion chamber temperature. Most studies performed on well stired reactor based their measurements on an air/fuel mixture inlet temperature of about 400K. We can use our measurements of local temperatures inside the burner to build the mal map which includes temperatures and heat flux. In order to create this thermal model, we can split the whole burner into 6 distinguished areas: the inner cooling flow, the premixed gas/air flow, the outer flow, the inner tube, the outer tube and the ri g case tube. These areas are shown on Figure 5 3.

PAGE 61

61 Figure 5 3 Schematic representation of the 3 flows and the 3 tubes of the burner's thermal modelisation Experimental Feedback It is interesting to compare the numerical investigation as shown Figure 5 2 with our experimentations results to quantify the difference between those and to explain the main causes of our With our numerical investigation of the combustion performance of the r eactor we obtained estimation s of 0.5 and 1.75 for respectively the lean and the rich blowout ratios [ 1 ]. The investigation was performed using GRI 3.0 data and a Python/Cantera program. We experimentally obtained ratios of 0.68 and 1.63. The relative errors could then be estimated at 26% and 7% for respectively the lean and the rich blowout ratios. Th ese differences between experimentation and simulation results may come from the fact that the flame is not fixed with respect to the burner. Actually, the flame is transitioning when app roaching blowout. Indeed, the flame is vertically translating inside the burner, positioning itself at the exit when running too rich and positioning itself at the

PAGE 62

62 bottom of the annular space when running too lean. In addition to changing the blowout ratio s, this transition phenomenon also impact the temperature measurement, since the thermocouples are not moving with respect to the burner. As we can see on the temperature graph Figure 4 1 and Figure 4 phenomena. We could obtain the original curves by using an external observation system such as a camera system positioning the flame and using an expected temperature profile of the flame. We would then have more accurate temperature measurements. In addition, we were not able to perform our experimentations under the exact same conditions as the ones assumed in our model. Indeed, since we were able to hear some combustion noise s while carrying out the measurements we can assume that the injected mixture was not entire ly homogeneous. Without a doubt the noise of the flame was generated by local oscillations of the flame due to local non homogeneity in the flow composition. As mentioned in chapter 1, the well stirred reactor model considers u niform physical and thermo ch emical properties over the combustor These Observations give us some guidelines about future implementation around the whole operating system.

PAGE 63

63 CHAPTER 6 FUTURE WORK The implementation of this novel combustion system offers flexibility for parametric stu dies and constant developments. With the concrete implementations, the effect on the stability performance of very different parameters can be studied. The characterization work of the rig could be completed using gas such as H2 and recomposed synthetic ga s. As explained before, the gas supply lines have been implemented in a flexible way, allowing the injection of various mixtures in the burner. For example, injecting a preheated CO2 based mix could be used to simulate exit gases recirculation. Operating r anges of the burner should also be defined for every identification of flameless combustion configurations for injected gas with different physical properties, thermo che mical properties and oxidation chemistry. The same characterization work should be done with the liquid fuel injections. The main difference with gas is that these lines are preheated. Thus both air and fuel will be is controlling both the liquid fuel and injected air temperatures. Using the gas analyzer available at the solar park, it would be relevant to profile the gas composition at the exit of the burner. This would he lp us to obtain sharper measurements of the CO2, CO and NOx density at the exit and give us a better idea of the efficiency of our reducing NOx efforts. Additionally, we should think about the implementation of a transverse nitrogen guided probe to perform real time sampling toward a GC MS (Gas Chromatograph

PAGE 64

64 Mass spectrometer) or a SMPS (Scanning Mobility Particle Spectrometer). This would enable precise species analysis (GC MS), and particle size distribution (SMPS). Furthermore, we could think of implemen ting a microphone to record sounds coming from the burner. Indeed, having a measurement of the frequency and the intensity of the sounds coming from the burner would be very useful. B Mhlbauer, R Ewert, O Kornow and B Noll worked on Broadband combusti on noise simulation s [ 18 ]. They presented a n umerical broadband combustion noise simulation of open non premixed turbulent jet flames applying the Random Particle Mesh for Combustion Noise ( RPM CN) approach. The RPM CN approach is a hybrid Computational Fl uid Dynamics/Computational Aeroacoustics method for the numerical simulation of turbulent combustion noise, based on a stochastic source reconstruction in the time domain. The combustion noise sources were modeled on the basis of statistical turbulence qua ntities, for example achieved by a Reynolds averaged Navier Stokes (RANS) simulation, using the Random Particle Mesh (RPM) method. RPM generates a statistically stationary fluctuating sound source that satisfies prescribed one and twopoint statistics whic h implicitly specify the acoustic spectrum. Subsequently, the propagation of the combustion noise was computed by the numerical solution of the Linearized Euler Equations Computed radial profiles of the reacting flow field were compared to experimental da ta and discussed. T he acoustic model used in their work is given by :

PAGE 65

65 With the combustion noise source term: Where is the substantial time derivative: A very similar study, using the same models could be done on our combus tion rig. Finally, an advanced reactor model would be useful for understanding of the combustion behavior in the burner. Reacting turbulent flow simulations are a necessary step and analysis have to be performed with the purpose of choosing the best simula tion in terms of computing time, complexity and predictions between different models. A good matching between the kinetic simulation and the experimental results will be essential to offer data on fuel oxidation kinetics.

PAGE 66

66 APPENDIX A RELEVANT MACHINING DR AWING The drawings presented in this Appendix are not necessarily at the scale. The reader should refer to the dimensions noted on the drawings. Only relevant views to the understanding of the geometry are presented here. The complete geometry description [1]

PAGE 67

67

PAGE 68

68

PAGE 69

69

PAGE 70

70 APPENDIX B BIG PICTURE OF THE SUPPLY SYSTEM A schematic plan representing the supply lines can be useful for the understanding of th is thesis. The next page contains such a plan.

PAGE 71

71

PAGE 72

72 LIST OF REFERENCES [1] B Fond, Master Thesis, University of Florida, Gainesville, 2009 [2] R M. Heck, Catalysis Today 53 (1999) 519 523 [3] J.A. Wunning, J.G. Wunning, Prog. Energy Combust. Sci., 23 (1997) 81 94 [4] Zeldovich, J., The oxidation of nitrogen in combustion and explosions, Academy of Science of the USSR. Acta Physiochimica U.R.S.S., ( 1946 ) XXI(4). [5] Ying Huang and Vigor Yang, Combustion and Flame 136 (2004) 383 389 [6] Goy CJ, James SR, Rea S. Progress in Astronautics and Aeronautics ( 2005 ) 210:163 75. [7] Ying Huang, Vigor Yang, Progress in Energy and Combustion Science 35 (2009) 293 364 [8] Alexander Stein, Saeid Niazi, and Lakshmi N. Sankar, Journal of Aircraft ( 2001 ) Vol. 38, No. 3. [9] C. D opazo E. E. O'B rien Acta A stronautica, ( 19 74 ) Vol. 1, pp. 1239 1266. [10] Z. ZhenLong, C. Zheng and C. ShiYi, Correlations for the ignition delay times of hydrogen/air mixtures, Peking University, Beijin, 2009 [11] D. M. Kalitan, J.D Mertens, M.W. Crofton, E. L. Petersen, J. Propulsion and Power, 23 (2005) 1291. [12] J. Natarajan, T. Lieuwen, J. Seitzman, Combust. Flame 151 (2007) 104 119. [13] Q. Zhang, D. R. Noble, T. Lieuwen, Trans. ASME 129 (2007) 688 694. [14] J. P. Longwell, M. A. Weiss, Indust. Eng. Chem. 47 (1955) 1634 1643. [1 5 ] W Tsang, S Manzello, C Stroud, M Donovan, V Babushok, K Smyth, C Frayne, D Lehnert, S McGivern, S Stouffer, V R. Katta, Kinetic Data Base for PAH Reaction and Soot Particle Inception during Combustion, SERDP Project WP 1198, 2007 [16 ] Martin Skov Skjth Rasmussen, PhD dissertation Technical University of Denmark, Kgs. Lyngby 2003 [17 ] Ian M. Kennedy, Prog. Energy Combust. Sci. ( 1997 ) Vol. 23 pp. 95 132. [18 ] Helge Egsgaard, Investigation of the Initial Reactions of the Calcote Mechanism for Soot Form ation, Ris o N a tion a l Laboratory, DK 4000 Roskilde, Denmark 1996

PAGE 73

73 [19 ] Bernd Mhlbauer, Roland Ewert, Oliver Kornow and Berthold Noll A eroacoustics ( 2012 ) volume 11 number 1 pages 1 24

PAGE 74

74 BIOGRAPHICAL SKETCH Julien Pierre Michel Adrien Brissonneau was born in Mont Saint Aignan, France in 1989, the first of a three child family. After two years of preparatory program in mathematics and physics at the Lycee Marie Curie, he passed the entrance examination of Arts et Metiers Paristech where he pursued the 3 years diplome This degree is equivalent to a Master of Science in the French Graduate industrial engineering from 2008 to 2010 and enrolled at the University of Florida in August 2010 to pursue a Master of Science in mechanical engineering as part of a double diploma.