Showing papers on "Diffusion flame published in 1982"

Theory of Laminar Flames

Abstract: 1. Governing equations of combustion 2. The premixed plane flame 3. Perturbations: SVFs and NEFs 4. Steady burning of a linear condensate 5. Unsteady burning of a linear condensate 6. Spherical diffusion flames 7. Cylindrical and spherical premixed flames 8. Multidimensional theory of premixed flames 9. Burners 10. Effects of shear and strain 11. Stability 12. Ignition and explosion.

Internal group combustion of liquid droplets

Abstract: A quasi-steady internal group combustion model of a spherical droplet cloud has been developed to assess the effects of the collective behavior of fuel droplets on combustion characteristics and cloud structures. Analytical solutions are obtained for the predictions of the distributions of the temperature, concentrations of fuel vapor, and oxidizer in strongly and weakly interacting zones. Overall burning rate under internal group combustion mode at various flame penetration is also obtained. Numerical analysis is carried out to determine the combustion behavior of droplet clouds at some selected group combustion numbers. It was found that for a cloud of n-butylbenzene droplets, the group envelope flame is stabilized on the boundary of the droplet cloud for a group combustion number of 1.36. As the group combustion number decreases, the envelope flame penetrates into the droplet cloud and divides the cloud into two zones; a strongly interacting zone located inside the group envelope flame and a weakly interacting zone established between the envelope flame and the boundary of the cloud. In the strongly interacting zone, the droplet vaporizes and the vapor produced is consumed at the group envelope flame. The droplets in a weakly interacting zone burn with an envelope flame surrounding eachmore » droplet. When the group combustion number decreases to 0.1, the group envelope flame degenerates into a point flame at the center of the droplet cloud.« less

An experimental study on extinction and stability of stretched premixed flames

Abstract: Law et al. (1981) and Ishizuka et al. (1982) have experimentally investigated the effects of flame stretch, preferential diffusion, and downstream heat loss on the extinction and stability of propane/air flames. The obtained results suggest that in the case of rich propane/air mixtures downstream heat loss, in addition to flame stretch, is needed for flame extinction. In the case of lean mixtures, flames can be extinguished by flame stretch alone. The data obtained in connection with the present study provide convincing evidence regarding the correctness of the previous results on the nature of flame extinction due to stretch. It is found that, in accordance with theoretical predictions, extinction by stretch alone is possible only when there is a deficiency regarding the less mobile reactant.

Prediction of turbulent jet diffusion flame lift-off using a pdf transport equation

Abstract: A recently developed theory for the stability of turbulent diffusion flames is evaluated topredict lift-off of jet flames. A fundamental quantity of the theory is the scalar dissipation rate, conditioned at stoichiometric mixture. This quantity appears in the transport equation for the probability density function (pdf) of the conserved scalar. It can be evaluated on the basis of an a-priori assumption about the pdf. For two assumptions about the pdf the conditioned scalar dissipation rate in a turbulent round jet is calculated and compared to the one obtained under the assumption of statistical independence between the scalar and the scalar dissipation rate. Comparison with measured lift-off data for methane flames show a good agreement for small but discrepancies for large lift-off heights. The effects of turbulence inhomogenities on the results are discussed.

Comparison between CARS and corrected thermocouple temperature measurements in a diffusion flame.

Abstract: Coherent anti-Stokes Raman spectroscopy (CARS) has been used to obtain radial temperature profiles in an axisymmetric methane diffusion flame. Temperatures were obtained from analysis of background-free nitrogen Q- and O-branch spectra. The spectra were analyzed with a nonlinear least-squares CARS fitting program and compared to measurements from radiation- and conduction-corrected thermocouples. Excellent agreement was obtained in regions of relatively constant temperature, whereas improved CARS spatial resolution was required to obtain agreement near steep temperature gradients.

Structure and extinction of near-limit flames in a stagnation flow

Abstract: Experimental studies of the structure and extinction of near-limit premixed flames in astagnation flow were made using counterflow twin flames established in the forward stagnation region of a porous cylinder. The stagnation surface between the two flames can be expected to be adiabatic and noncatalytic wall surface. Near-limit rich- and lean-methane/air and propane/air flames were used in the experiment. The structure of the twin flames, the flame temperature, the distance between the two flame zones, and the concentrations of reactants on the stagnation surface were measured, and the extinction mechanism is discussed. Two distinct modes of flame extinction exist in the stagnation flow field: one is flameextinction which occurs close to the stagnation surface due to incomplete combustion and the other is flame extinction which occurs at a finite distance from the stagnation surface due to flame stretch. The Lewis number of the deficient reactant in the premixed combustible gases (fuel in the lean mixture and oxygen in the rich mixture) is responsible for the existence of these two distinct modes of flame extinction. These results strongly support the theoretical predictions by Sivashinsky and by Sato and Tsuji. The extinction of a flame in which the diffusion coefficient of the excess reactant is muchlarger than that of the deficient reactant is also affected by dilution of the reaction zone by the excess reactant with the larger diffusion coefficient.

An experimental study on stability and combustion characteristics of an excess enthalpy flame

Abstract: A further experimental study was conducted on the excess enthalpy flame system proposedby Takeno and Sato to burn mixtures of low heat content. In the light of previous experimental findings an entirely new burner was designed to reduce the heat loss as much as possible. A bundle of ceramic tubes was used as a combustion tube and four perforated ceramic plates were placed upstream and downstream of the tube to reduce the radiative heat loss from the heated tube. The stability and combustion characteristics of the burner were studied to explore to what extent the range of flame stability was extended below the normal lean limit. It was found that the flame stabilized ahead of, in, or behind, the combustion tube, depending on flow rate and equivalence ratio. The stability limits for the respective flames extended below the normal limit. The observed temperature distributions revealed that the heat recirculation through the perforated plates, as well as the combustion tube, played an important role in the flame stabilization. General characteristics of the proposed flame system are discussed on the basis of the experimental findings obtained so far.

Effects of lewis number on extinction behavior of premixed flames in a stagnation flow

Abstract: The combined effects of the velocity gradient and the Lewis number of deficient reactantsin the premixed gas on the behavior and extinction of a premixed flame in a stagnation flow were studied experimentally. The flat twin flames established in the stagnation region of the two dimensional opposed flow of the same mixture were used in order to make an adiabatic and noncatalytic stagnation surface. The flame extinction behavior observed is purely attributed to the flow-transport properties-chemical reaction interaction. The mixture employed was hydrogen, methane, propane, or butane with air. The flame temperature and the distance between the two flames were measured for the wide variations of the fuel concentration and the velocity gradient. As the stagnation velocity gradient was increased by increasing the opposed flow velocity, the flat two flames approached each other, and finally the extinction occurred abruptly. By observing the flame behavior, the extinction process of the flame was classified into two patterns corresponding to the ranges of the Lewis number Le of the deficient reactant in the premixed gas (fuel in the lean mixture and oxygen in the rich mixture). Its ranges are Le 1. Under the condition of Le 1 (lean hydrogen or methane-air and rich propane or butane-air), as the velocity gradient is increased, the flame temperature increases and reaches a maximum point at a certain value of the velocity gradient, beyond which the flame temperature decreases. The flame is extinguished close to the stagnation surface or in contact with the stagnation surface. The results of this experiment are in good agreement with the theoretical predictions.

On Stability of Premixed Flames In Stagnation - Point Flow

Abstract: A quantitative description of flame stabilization in stagnation-point flow is proposed. Asymptotic and stability analyses are made for a flame model where the density of the gas is assumed to be constant and the reaction zone is assumed to be narrow and concentrated over the flame front. It is shown that, if blowing is sufficiently strong, the corrugations disappear and a plane flame results. The phenomena cannot be fully described by means of classical linear stability analysis.

Entrainment and Flame Geometry of Fire Plumes

Abstract: This study concerns the flame structure and fire plume entrainment of natural gas diffusion flames on 0.10, 0.19 and 0.50 m. diameter burners. The heat release rates ranged from 10 kW to 200 kW. Flame heights based on high speed photography and eye averages show a transition in the dependence of flame height on a dimensionless heat addition parameter around unity. For flames taller than three burner diameters, the initial diameter of the fire does not affect the length of these flames whereas for short flames initial geometry becomes important. Another prominent feature of these flames is the presence of large scale ring vortex-like structures which are formed close to the burner surface more or less regularly. It is found that these structures are responsible for the fluctuations of the flame top. Entrainment measurements spanned heights starting very close to the burner surface to distances about six times the average flame heights. Experiments indicate the presence of three regions; a region close to the burner surface where plume entrainment rates are independent of the fuel flow (or heat release) rates; a far field region above the flame top, where a simple point source model correlates the data reasonably well; and a not so well-defined intermediate region where entrainment seems to be similar to that of a turbulent flame with plume-like characteristics. It is also found that the disturbances in the ambient atmosphere will greatly enhance the fire plume entrainment. Finally, a theoretical study of a steady, buoyant, diffusion flame indicated the importance of the puffing in the entrainment process.

Steady State Diffusion Flame Structure with Lewis Number Variations

Abstract: The effects of non-unity Lewis numbers of fuel and oxidizer on the structure of diffusion flames are studied for a variety of flow configurations. By using a generalized formation for the three-dimensional flame-sheet combustion, it is shown that the flame-sheet temperature is increased/decreased with decreasing/increasing Lewis number of either reactant, and that blowing reduces/aggravates these non-unity Lewis number effects if it is parallel/opposing the direction of diffusion. Ignition and extinction analyses are also conducted, yielding explicit criteria governing the occurrence of these events. The falsification of empirically-determined activation energy with the unity Lewis number assumption is emphasized.

Controlling Mechanisms of Flame Spread

Abstract: Recent advances in the experimental study of the mechanisms controlling the spread of flames over the surface of combustible solids are summarized in this work. The heat transfer and gas phase chemical kinetic aspects of the flame spread process are addressed separately for the spread of flames in oxidizing flows that oppose or concur with the direction of propagation. The realization that in most practical situations, the spread of fire in opposed flows occurs at near extinction or non-propagation conditions is particularly significant. Under these circumstances, gas phase chemical kinetics plays a critical role and it must be considered if realistic descriptions of the flame spread process are attempted. In the concurrent mode of flame spread, heat transfer from the flame to the unburnt fuel appears to be the primary controlling mechanism. Although gas phase chemical kinetics is unimportant in the flame spread process, it is important in the establishment and extension of the diffusion flame that generates the spread process. The current experimental observations, although still in need of further verification, provide insight toward the development of accurate descriptions of the flame spread process.

Asymptotic Theory of Diffusion-Flame Extinction with Radiant Loss from the Flame Zone

Abstract: Laminar diffusion flames in counterfiow configurations such as stagnation-point boundary layers are analyzed by methods of matched asymptotic expansions with large parameters being the temperature sensitivities of the rates of chemical heat generation and radiant heat loss. Formulas are derived defining critical conditions for flame extinction, including influences of radiant loss.

On the Opening of Premixed Bunsen Flame Tips

Abstract: The local extinction of Bunsen flame tip and edges of hydrocarbon/air premixtures has been experimentally investigated using a variety of burners. Results show that, while for both rich propane/air and butane/air mixtures tip opening occurs at a constant fuel equivalence ratio of 1.44 and is therefore independent of the intensity, uniformity, and configuration of the approach flow, for rich methane/air flames burning is intensified at the tip and therefore opening is not possible. These results substantiate the concept and dominance of the diffusional stratification mechanism in causing extinction, and clarify the theoretical predictions on the possible opening of two-dimensional flame wedges.

Flame ionization control of a partially premixed gas burner with regulated secondary air

Abstract: A gas burning furnace has a burner in which the premixed fuel/air ratio of the burned gas in the burner flame is measured by a flame rod sensing the ionization current. The flame rod is connected to a fuel/air controller which controls the gas and primary air to the burner to maintain the maximum flame ionization current which results in an excessive amount of gas to the burner. The secondary air to the combustion chamber is proportionally controlled to add sufficient oxidant to the gas combustion in the combustion chamber to ensure complete combustion. Other properties of the flame or combustion products can also be used.

The mechanisms of lean limit extinguishment of an upward and downward propagating flame in a standard flammability tube

Abstract: The mechanisms by which lean limit upward and downward propagating flames extinguish in a standard flammability tube (51mm ID, 1.8m long) were deduced using schlieren and direct light photography and temperature measurements and by observing the behavior of the flame under transient loading. The lean methane-air system was studied. The upward propagating flame takes the shape of a stable hemispherical cap attached to a trailing skirt. This flame always extinguishes first at the holding region on the axis of the tube and a failure wave subsequently washes down the flame until extinguishment is complete. The extinguishment occurs because the flame is stretched at the tip and heat loss to the walls is not important to this extinguishment. On the other hand the extinguishment of a downward propagating flame is a multistep process. First heat loss to the walls causes the flame to extinguish near the walls and a smaller residual flame is left in the center of the tube. Once this happens, differential buoyancy of the hotter central gases and the cooler surrounding gases forces the cooler gases to move ahead of the flame. Finally, the downward propagating flame is observed to rise slightly. At this point the flame extinguishes completely.

The group combustion of a spherical cloud of monodisperse fuel droplets

Abstract: The vaporization and combustion of a spherical, uniform monodisperse cloud of fuel drops in equilibrium with a quiescent atmosphere is considered. For typical fuel sprays in which the drops are separated by five to ten drop diameters only the droplets within a thin inwardly propagating vaporization wave at the edge of the cloud will vaporize. In analogy with single drop theory the cloud radius is found to decrease acording to a “d2 law,” although with a modified vaporization constant for both purely vaporizing and burning clouds. The cloud interior in all cases remains in saturated, non vaporizing equilibrium at a temperature determined by ambient conditions. Burning occurs at a spherical diffusion flame front outside the cloud. The flame radius to cloud radius ratio is found to be constant and independent of the cloud radius and the droplet radius and number density within the cloud. An external ignition temperature is determined by considering the bifurcation of steady state solutions for single step, irreversible Arrhenius kinetics. At ignition the cloud interior reaches a new equilibrium which is still too cool and rich for single drop ignition. Under conditions of the present analysis a fuel cloud, if it burns at all, will do so with an external diffusion flame.

Modeling Calculations for HMX Composite Propellants

Abstract: HMX (cyclotetramethyline tetranitramine) composite propellants burn at rates significantly lower than HMX monopropellant. To model the behavior of these propellants, a new model was developed within the framework of the Beckstead-Derr-Price (BDP) modeling approach. A time-averaging approach has been developed assuming that the propellant burns through alternate layers of binder and oxidizer at significantly different rates. The model has been compared in detail with experimental results from 17 HMX/HTPB propellants. Both the data and the model show that there is only a small dependence of rate on particle size. The model predicts that the rates of HMX/HTPB propellants will converge with increasing solids loadings, and that above -85% solids there is very little change in rate for varying formulations. The interpretation of the data using the model indicates three predominant mechanisms leading to the peculiar characteristics of HMX propellants. First, the HMX binder diffusion flame is an energy poor flame that robs energy from the products that would otherwise result from the monopropellant flame. Second, there appears to be a significant ignition delay time associated with large particles that impedes the overall rate. Third, the binder rate appears to be very significant. The model indicates that changing the rate using conventional catalysis approaches would be very difficult, since the rate is more dependent on binder decomposition characteristics than on the oxidizer.

Development and optimisation of atom cells for sensitive coupled gas chromatography-flame atomic-absorption spectrometry

Abstract: The development of four novel atom cells for the determination of volatile organometallic compounds by coupled gas chromatography-flame atomic-absorption spectrometry is described. Tetraalkyllead compounds provided a model system in the optimisation of the four atom cells by the variable step-size simplex method. The effects of the various parameters on analytical performance are discussed. In the most sensitive system presented the effluent from the chromatograph was fed to a small hydrogen diffusion flame and the atoms from this flame were swept into a flame-heated ceramic tube. This simple, readily demountable arrangement enjoys the advantages of continuous operation associated with flames but, because of the relatively long atomic residence times, gave detection limits of 17 pg for both tetraethyl- and tetramethyllead. These limits are superior to any previously reported for a gas chromatographic-atomic-absorption spectrometric technique, including those employing electrothermal atomisation.

Velocity and Temperature Fluctuations in a Flat Plate Boundary Layer Diffusion Flame

Abstract: Measurements of mean value and turbulent intensity of fluctuating velocity as well as fluctuating temperature have been made on a turbulent boundary layer diffusion flame over a flat plate. Results show that the turbulence production occurs along the flame zone. This behavior is reflected in the high gradient of mean velocity. That is, the variation of velocity fluctuation due to the flame is primarily attributed to the variation of the mean velocity profile. Therefore it can be concluded that an ordinary shear-generated turbulence plays an important role on turbulence in the whole region of the boundary layer. Results also show that the temperature fluctuation is attributed to the gradient of mean temperature in a turbulent flow. At the same time, the temperature fluctuation in the air stream side is intensified due to the penetration of room temperature gases into the burnt gas region.

Flame modelling and burning velocity measurement

Abstract: Strictly, the term “burning velocity” is only meaningful in relation to planar flame frontsin one-dimensional flow systems. On the other hand, burning velocity measurements are obtained from flames which are frequently curved, and in systems where the flow diverges to a greater or lesser extent on approach to the reaction zone. Thus, in order to extract from an experimental arrangement a burning velocity which can validly be used in one-dimensional flame modelling, it becomes necessary to define an appropriate reference plane in the flame, and to estimate the gaseous mass flux (or some equivalent) at this plane. A computational method of defining such a plane is suggested, and has been used on a range of methane-air and hydrogen-air flames. Resulting reference plane temperatures and positions are presented. They almost coincide with the maximum reaction rates in the flames, and for practical purposes they may be identified in the case of methane flames with the leading edge of the luminous zone. The implications of the findings are discussed for both stationary and non-stationary flamemethods of measuring burning velocities. For mixtures initially at room temperature and atmospheric pressure, laminar burning velocities are presented graphically for a range of methane-air flames not too far from stoichiometric. The “best” value for the stoichiometric methane-air flame turns out to be around 37 or 38 cm s −1 .