• 한국연소학회:학술대회논문집
• /
• 2014.11a
• /
• pp.81-84
• /
• 2014

A Experimental study on flame extinction behavior was investigated using He curtain flow with counter triple co-flow burner. Buoyancy force was suppressed up to a microgravity level of $10^{-2}-10^{-3}g$ by using He curtain flow. The stability maps were provided with a functional dependency of diluent mole fraction and global strain rate to clarify the differences in flame extinction behavior. The flame extinction curves had C-shapes at various global strain rates. The oscillation and extinction modes were different each other in terms of the global strain rate, and the flames extinction modes could be classified into five modes such as (I) and (II): an extinction through the shrinkage of the outmost edge flame forward the flame center after self-excitation and without self-excitation, respectively, (III): an extinction through rapid advancement of a flame hole while the outmost edge flame is stationary, (IV): self-excitation occurs in the outermost edge flame and the center edge flame and then a donut shaped flame is formed and/or the flame is entirely extinguished, (V): shrinkage of the outermost edge flame without self-excitation followed by shrinkage or survival of the center flame. These oscillation and extinction modes could be identified well to the behavior of edge flame. The result also showed that the edge flame was influenced significantly by the conductive heat losses to the flame center or ambient He curtain flow.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.29 no.9 s.240
• /
• pp.997-1005
• /
• 2005

The extinction characteristics of low strain rate normal gravity (1-g) nonpremixed methane-air flames were studied numerically and experimentally. A time-dependent axisymmetric two-dimensional (2D) model considering buoyancy effects and radiative heat transfer was developed to capture the structure and extinction limits of 1-g flames. One-dimensional (1D) computations were also conducted to provide information on 0-g flames. A 3-step global reaction mechanism was used in both the 1D and 2D computations to predict the measured extinction limit and flame temperature. A specific maximum heat release rate was introduced to quantify the local flame strength and to elucidate the extinction mechanism. Overall fractional contribution by each term in the energy equation to the heat release was evaluated to investigate the multi-dimensional structure and radiative extinction of 1-g flames. Images of flames were taken for comparison with the model calculation undergoing extinction. The two-dimensional numerical model was validated by comparing flame temperature profiles and extinction limits with experiments and ID computation results. The 2D computations yielded insight into the extinction mode and flame structure of 1-g flames. Two combustion regimes depending on the extinction mode were identified. Lateral heat loss effects and multi-dimensional flame structure were also found. At low strain rates of 1-g flame ('Regime A'), the flame is extinguished from the weak outer flame edge, which is attributed to multi-dimensional flame structure and flow field. At high strain rates, ('Regime B'), the flame extinction initiates near the flame centerline due to an increased diluent concentration in reaction zone, which is the same as the extinction mode of 1D flame. These two extinction modes could be clearly explained with the specific maximum heat release rate.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.12
• /
• pp.1557-1566
• /
• 2004

A flame extinction phenomenon is a typical unsteady process in combustion. Flame extinction is characterized by various physical phenomena, such as convection, diffusion, and the production of heat and mass. Flame extinction can be achieved by either increasing the strain rate or curvature, by diluting an inert gas or inhibitor, or by increasing the thermal or radiant energy loss. Though the extinction is an inherently transient process, steady and quasi-steady approaches have been used as useful tools for understanding the flame extinction phenomenon. Recently, unsteady characteristics of flames have been studied by many researchers, and various attempts have been made to understand unsteady flame behavior, by using various extinction processes. Representative parameters for describing flame, such as flame temperature, important species related to reactions, and chemi-luminescence of the flame have been used as criterions of flame extinction. In these works, verification of each parameter and establishing the proper criterions of the extinction has been very important. In this study, a time-dependent flame temperature and an OH radical concentration were measured using optical methods, and the instantaneous change of the flame luminosity was also measured using a high-speed ICCD (HICCD) camera. We compare the unsteady extinction points obtained by three different methods, and we discuss transient characteristics of maximum flame temperature and OH radical distribution near the extinction limit.

• 한국연소학회:학술대회논문집
• /
• 2014.11a
• /
• pp.379-381
• /
• 2014

Influence of fuel concentration gradient was investigated near flame extinction limit in buoyancy-suppressed non-premixed counterflow flame with triple co-flow burner. The use of He curtain flow produced a microgravity level of $10^{-2}-10^{-3}g$ in He-diluted non-premixed counter triple co-flow flame experiments. Flame stability map was presented based on flame extinction and oscillation near extinction limit. The stability map via critical diluent mole fraction with global strain rate was represented by varying outer and inner He-diluted mole fractions. The flame extinction modes could be classified into five: an extinction through the shrinkage of the outmost edge flame forward the flame center with and without self-excitation, respectively ((I) and (II)), an extinction via the rapid expansion of a flame hole while the outmost edge flame is stationary (III), both the outermost and the center edge flames oscillate, and then a donut shaped flame is formed or the flame is entirely extinguished (IV), a shrinkage of the outermost edge flame without self-excitation followed by shrinking or sustain the inner flame (V).

• 한국연소학회:학술대회논문집
• /
• 2002.11a
• /
• pp.25-31
• /
• 2002

For the better understanding of the stability of turbulent combustion, more researches on extinction and re-ignition are needed. Flame interactions in non-premixed flame have also not been greatly researched. We made a hybrid twin jet flame, the combinations of diffusion flame and partially-premixed diffusion flame, in a twin jet counterflow configuration. The extinction limits of a crossed twin jet counterflow have been extended in comparison with those of a one-dimensional counterflow because of flame interactions through heat transfer and joint ownership of various radicals. Besides, we have obtain ignition $Damk\"{o}hler$ number by experimental method without external ignition source using the extinction characteristic in a crossed twin jet counterflow flame. From results, we can identify the hysteresis between extinction and ignition $Damk\"{o}hler$ number in S-curve.

• 한국연소학회:학술대회논문집
• /
• 2001.11a
• /
• pp.111-118
• /
• 2001

Many previous researches on the premixed flame in a tube have treated the unsteady flame behaviors but more detailed and fundamental research has been necessary. The study on the flame stabilization condition in a tube and the unsteady behaviors were carried out in recent years. In this paper, a mean velocity variation larger than the burning velocity was introduced to the stabilized flame for a period longer than the reaction time scale in order to examine the unsteady behavior of flame propagation. Through our previous work it was found that the effects of non-unity Lewis number on the flame extinction was negligible in the extinction by the boundary layer even though they were important in the extinction by the acoustic instability. In this paper we carried out an analytic approach to explain the previous experimental results. It showed that the heat loss, from a flame to the wall, is not a sufficient condition but a required one for the growth of the extinction boundary layer. In addition, the quenching and the flame stretch, under a strong unsteady flow field, are the main causes of the eventual extinction.

• 한국연소학회:학술대회논문집
• /
• 2003.05a
• /
• pp.93-101
• /
• 2003

The extinction of unsteady diffusion flame was experimentally studied in an opposing jet counterflow burner using diluted methane. The stabilized flame was perturbed by linearly varying velocity change that was generated by pistons installed on both sides of the air and fuel stream. As the results, the extinction of unsteady flame is dependent not only on the history of unsteadiness, but also on the initial condition. We found that there are several unsteady effects on the flame extinction. First, the extinction strain rates of unsteady cases are extended well beyond steady state extinction limits. Second, as the slope of the strain rate change increases, the unsteady extinction strain rate becomes larger. Third, the extension of unsteady extinction strain rate becomes smaller as the initial strain rate increases. We also found that the extension of the extinction limit mainly results from the unsteady response of the reaction zone because there is no retardation effect of a mixing layer for our experimental condition.

• 한국연소학회:학술대회논문집
• /
• 2003.12a
• /
• pp.195-200
• /
• 2003

A two-dimensional "twin-jet counterflow" burner has been designed for the better understanding of the stability of turbulent flames. This flow system enables one to systematically investigate various effects on non-premixed flames, including the effects of curvature, negative strain, and non-premixed flame interactions. The objective of this study is comparing characteristics of extinction of non-premixed methane flames with that of non-premixed propane flames investigated previously. The extinction limit of non-premixed methane and propane flames can be extended compare to that for the conventional counterflow non-premixed flame because of the existence of petal shaped flame and have same structure. The hysteresis in transition between the petal shaped flame and the curved two-wing flames could be observed. We could find differences between non-premixed methane flame and non-premixe propane flame such as the position of one wing extinction and the regime of one wing extinction.

• 한국연소학회:학술대회논문집
• /
• 2006.10a
• /
• pp.253-259
• /
• 2006

The effect of DC electric fields on the flame extinction was investigated experimentally in counterflow configurations for the methane/oxygen/nitrogen diffusion flame. The electric fields was applied by connecting the high voltage and ground terminals to the upper and lower burners, respectively. In case of having electric fields, several modes of flame extinction was observed according to the electric field intensity and strain rate defined by the exit velocity. To visualize and characterize the flame structure and intensity, planar LIF technique was adopted for OH radicals. Consequently, several length scales, including the flame width, thickness, and height from the burner tip, were introduced to explain the various flame behaviors and to characterize the flame extinctions. It was found that the variation of flame width and the chemical reaction are strongly related to a critical electric field intensity, thus the various modes of diffusion flame extinction could be observed due to the electric fields.

• Journal of the Korean Society of Combustion
• /
• v.17 no.3
• /
• pp.17-29
• /
• 2012

The effects of strain rate and preferential diffusion of $H_2$ on flame extinction are numerically studied in interacting premixed syngas-air flames with fuel compositions of 50% $H_2$ + 50% CO and 30% $H_2$ + 70% CO. Flame stability diagrams mapping lower and upper limit fuel concentrations at flame extinction as a function of strain rate are examined. Increasing strain rate reduces the boundaries of both flammable lean and rich fuel concentrations and produces a flammable island and subsequently even a point, implying that there exists a limit strain rate over which interacting flame cannot be sustained anymore. Even if effective Lewis numbers are slightly larger than unity on extinction boundaries, the shape of the lean extinction boundary is slanted even at low strain rate, i.e. $a_g=30s^{-1}$ and is more slanted in further increase of strain rate, implying that flame interaction on lean extinction boundary is strong and thus hydrogen (as a deficient reactant) Lewis number much less than unity plays an important role of flame interaction. It is also shown that effects of preferential diffusion of $H_2$ cause flame interaction to be stronger on lean extinction boundaries and weaker on rich extinction boundaries. Detailed analyses are made through the comparison between flame structures with and without the restriction of the diffusivities of $H_2$ and H in symmetric and asymmetric fuel compositions. The reduction of flammable fuel compositions in increase of strain rate suggests that the mechanism of flame extinction is significant conductive heat loss from the stronger flame to ambience.