• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.30 no.2
• /
• pp.91-97
• /
• 2002

Dynamic extinction of solid propellants subjected to rapid pressure drop was studied with the aid of energy equation of condensed phase and flame model in gas phase. It is found that the total residence time($\tau_\gamma$) which measures the residing time of fuel in the reaction zone may play a crucial role in determining the dynamic response of the combustuion to extinction. Residence time was modeled by various combinations of diffusion and chemocal kinetic time scale. Effect of pressure history coupled with chamber volume on the extinction response was also performed and was found that dynamic extinction is more susceptible in a confined chamber than in open geometry. And, dynamic extinction was revealed to be affected profoundly by diffysion time scale rather than chemical kinetic time scale.

• Journal of the Korean Society of Combustion
• /
• v.17 no.3
• /
• pp.9-16
• /
• 2012

Nonlinear dynamics of pulsating instability in radiating counterflow diffusion flames is numerically investigated by imposing Damk$\ddot{o}$hler number perturbation. Stable limit-cycle solutions occur in small ranges of Damk$\ddot{o}$hler numbers past bifurcation point of instability. Period doubling cascade and chaotic behaviors appear just before dynamic extinction occurs. Nonlinear dynamics is also studied when large disturbances are imposed to flames. For weak steady flames, the dynamic extinction range shrinks as the magnitudes of disturbances are increased. However, strong steady flames can overcome relatively large disturbances, thereby the dynamic extinction range extending. Stable limit-cycle behaviors reappears prior to dynamic extinction when the steady flames are strong enough.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.2
• /
• pp.72-79
• /
• 2003

A numerical calculation was conducted to estimate and to elucidate the role of the radiative heat flux from metal particles(Al, $Al_2O_3$) on the dynamic extinction of solid propellant rocket where the rapid depressurization took place. Anon-linear flame modeling implemented by the residence time modeling for metalized propellant was adopted to evaluate conductive heat flux to the propellant surface. The radiative heat feed back was calculated with the aid of a modified comvustion-flow model as well. The calculation results with the propellant of AP:Al:CTPB=76:10:14 had revealed that the radiative heat flux is approximately 5~6% of total flux at the critical depressurization rate regardless of chamber geometry (open or confined chamber). It was also found that the dynamic extinction in open geometry could be predicted at the depressurization rate about 45% larger with radiative heat feedback than without radiation. Thus, it should be claimed that even a small amount of radiative flux 5~6% could produce a big error in predicting the critical depressurization rate of the metalized propellant combustion.

• Journal of the Korean Society of Combustion
• /
• v.13 no.4
• /
• pp.16-25
• /
• 2008

Experiments have been conducted to clarify impacts of curtain flow and velocity ratio on low strain rate flame extinction, and to further display transition of shrinking flame disk to flame-hole. Critical mole fractions at flame extinction are examined in terms of velocity ratio, global strain rate, and nitrogen curtain flow rate. It is shown that multi-dimensional effects at low strain rate flames through global strain rate, velocity ratio, and curtain flowrate dominantly contribute to flame extinction and transition of shrinking flame disk to flame hole. Our concerns are particularly focused on the dynamic behavior of an edge flame in shrinking flame disk.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.6
• /
• pp.688-696
• /
• 2004

Nonlinear dynamic behavior of diffusive-thermal instability in diluted CH$_4$/O$_2$ diffusion flames is numerically investigated by adopting detailed chemistry and transport. Counterflow diffusion flame is adopted as a model flamelet. Particular attention is focused on the pulsating-instability regime, which arises for Lewis numbers greater than unity, and the instability occurs at high strain rate near extinction condition in this flame configuration. Once a steady flame structure is obtained for a prescribed value of initial strain rate, transient solution of the flame is calculated after a finite amount of strain-rate perturbation is imposed on the steady flame. Transient evolution of the flame depends on the initial strain rate and the amount of perturbed strain rate. Basically, the dynamic behaviors can be classified into two types, namely non-oscillatory decaying solution and diverging solution leading to extinction. The peculiar oscillatory solution, which has been found in the previous study adopting one-step chemistry and constant Lewis numbers, is net observed in this study, which is attributed to both convective flow and preferential diffusion effects.

• 한국연소학회:학술대회논문집
• /
• 2004.06a
• /
• pp.95-101
• /
• 2004

Dynamic behavior of diffusive-thermal instability in diluted $CH_4/O_2$ diffusion flames is numerically investigated by adopting detailed chemistry and transport. Counterflow diffusion flame is adopted as a model flamelet. Particular attention is focused on the pulsating-instability regime, which arises for Lewis numbers greater than unity, and the instability occurs at high strain rate near extinction condition in this flame configuration. Once a steady flame structure is obtained for a prescribed value of initial strain rate. transient solution of the flame is calculated after a finite amount of strain-rate perturbation is imposed Oil the steady flame. Transient evolution of the flame depends on the initial strain rate and the amount of perturbed strain rate. Basically, the dynamic behaviors can be classified into two types, namely non-oscillatory decaying solution and diverging solution leading to extinction. The peculiar oscillatory solution. which has been found in the previous study adopting one-step chemistry and constant Lewis numbers, is not observed in this study, which is attributed to both convective flow and preferential diffusion effects.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.9
• /
• pp.1185-1191
• /
• 1999

Radiation-induced oscillatory instability in diffusion flames is numerically investigated with nonlinear dynamics considered. As the simplest flame model, a diffusion flame established in the stagnant mixing layer is employed with optically thin gas-phase radiation and unity Lewis numbers for all species. Attention is focused on the radiation-induced extinction regime, which occurs at large $Damk\ddot{o}hler$ number. Once the steady flame structure is obtained for a prescribed value of the initial $Damk\ddot{o}hler$ number, transient solution of the flame is calculated after a finite amount of the $Damk\ddot{o}hler$-number perturbation is imposed on the steady flame. Transient evolution of the flame exhibits three types of flame-evolution behaviors, namely decaying oscillatory solution, diverging solution to extinction and stable limit-cycle solution. A dynamic extinction boundary is identified for laminar flamelet library.

• 한국연소학회:학술대회논문집
• /
• 2006.10a
• /
• pp.65-72
• /
• 2006

Experiments in methane-air low strain rate counterflow diffusion flames diluted with nitrogen have been conducted to study the behavior of flame extinction and edge flame oscillation in which flame length is less than the burner diameter and thus lateral conduction heat loss in addition to radiative heat loss could be remarkable at low global strain rates. Critical mole fraction at flame extinction is examined with velocity ratio and global strain rate. Onset conditions of edge flame oscillation and flame oscillation modes are also provided with global strain rate and added nitrogen mole fraction to fuel stream (fuel Lewis number). It is seen that flame length is closely relevant to lateral heat loss, and this affects flame extinction and edge flame oscillation considerably. Edge flame oscillations in low strain rate flames are experimentally described well and are categorized into three: a growing oscillation mode, a decaying oscillation mode, and a harmonic oscillation mode. The regime of flame oscillation is also provided at low strain rate flames. Important contribution of lateral heat loss even to edge flame oscillation is clarified.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.30 no.4 s.247
• /
• pp.343-349
• /
• 2006

Systematic experiments in $CH_4/Air$ counterflow diffusion flames diluted with He have been undertaken to study the oscillatory instability in which lateral flame size was less than burner nozzle diameter and thus lateral heat loss could be remarkable at low global strain rate. The oscillatory instability arises for Lewis numbers greater than unity and occurs near extinction condition. The oscillation is the direct outcome from the advancing and retreating edge flame. The dynamic behaviors of extinction in this configuration can be classified into three modes; growing, harmonic and decaying oscillation mode near extinction. As the global strain rate decreases, the amplitude of the oscillation becomes larger. This is caused by the increase of lateral heat loss which can be confirmed by the reduction of lateral flame size. Oscillatory edge flame instabilities at low global strain rate are shown to be closely associated with not only Lewis number but also heat loss (radiation and lateral heat loss).

• 한국연소학회:학술대회논문집
• /
• 2003.05a
• /
• pp.223-227
• /
• 2003

The method of flame hole dynamics is demonstrated as a mean to simulate turbulent flame extinction. The core of the flame hole dynamics involves derivation of a random walk mapping for the flame holes, created by local quenching, between the burning and quenched states provided that the dynamic characteristics of flame edges is known. Then, the random walk mapping is projected to a background turbulent field. The numerical simulations are carried out with the further simplifications of flame string and unconditioned scalar dissipation rate. The simulation results show how the chance of partial quenching is influenced by the crossover scalar dissipation rate. Finally, a list of improvements, necessary to achieve more realistic turbulent flame quenching simulation, are discussed.