Tong-Miin Liou

Downward flame spread over a thick PMMA slab in an opposed flow environment: experiment and modeling

Abstract This work investigates experimentally and theoretically the downward spread of a flame over a thick polymethylmethacrylate (PMMA) slab with an opposed flow of air. Simulation results, using an unsteady combustion model with mixed convection, indicate that the ignition delay time increases with a decreasing opposed-flow temperature or increasing velocity. The ignition delay time is nearly constant at a low opposed flow velocity, i.e., u ∞ ≤ 30 cm/s . Experiments were conducted at three different opposed flow temperatures and velocities, namely, T i = 313, 333 , and 353 K and u ∞ = 40, 70, and 100 cm/s , respectively. Measurements included the flame-spread rate and temperature distributions, using thermocouples and laser-holographic interferometry. The qualitative trends of the flame-spread rate and thermal boundary layer thickness, as obtained experimentally and from numerical predictions, were identical. For a quantitative comparison, the predicted and experimental flame-spread rates correlated well with each other, except at the lowest velocity ( u ∞ = 40 cm/s) . The discrepancies between the measured and predicted thermal boundary layer thicknesses decreased with an increasing flow velocity. The quantitative agreement with a high velocity indicates that the spread of an opposed flame is mainly controlled by the flame front, whereas the discrepancies at low flow rates demonstrate the importance of radiation, the finite length of the fuel, and also three-dimensional effects, which were not considered in the model. The temperature profiles around the flame front measured by interferometric photographs indicate a recirculation flow there, as predicted by the simulation.

Transition Characteristics of Flowfield in a Simulated Solid-Rocket Motor

The e ow characteristics in a two-dimensional porous-walled duct simulating a solid-propellant rocket motor are numerically computed to investigate the effects of viscosity, compressibility, and ine ow turbulence ( w) on the e ow transitions. The e nite volume technique is used to solve the time-dependent compressible Navier‐ Stokes equations with a subgrid-scale turbulence model, and the numerical e uxes are computed using a modie ed Godunov scheme. In addition to computed axial mean velocity and turbulence intensity proe les, the axial variations of skin friction coefe cient and the transverse location of peak turbulence intensity are used to identify the mean-e ow transition and turbulence-intensity transition, respectively. In particular, a new way of identifying turbulence-intensity transition by the use of the power spectrum of velocity e uctuations is presented for the e rst time in the present study. The minimum centerline Mach number for the onset of mean-e ow transition is obtained as the compressibility is considered alone. The critical values of w for the onset of turbulence-intensity transition and mean-velocity transition advance as well as for the concurrence and delay between the two transitions are also determined to illustrate why some researchers could observe only a single transition whereas others observed two transitions.

Large-eddy simulations of turbulent reacting flows in a chamber with gaseous ethylene injecting through the porous wall

Abstract Large-eddy simulations were performed to study the turbulent reacting flows in a simulated solid-fuel combustion chamber. The time-dependent axisymmetric compressible conservation equations were solved directly without using subgrid-scale turbulence models. The combustion process considered was a one-step, irreversible, and infinitely fast chemical reaction and the pyrolizing solid fuel was simulated by gaseous ethylene injected through a porous wall for a practical range of fuel blowing velocity encountered in solid-fuel combustion chambers for the first time. The numerical code used the finite-volume technique which involved alternating in time the second-order, explicit MacCormacks and Godunovs methods. Characteristic-based boundary conditions were applied on inflow and outflow boundaries, which allowed outlet boundary conditions to be nonzero gradients and, in turn, a practical length of computational domain to be realized. The effects of combustion on the large-scale unsteady flow structure and the mean flameholder recirculation zone were documented in terms of the density contours, vorticity dynamics, streamlines, mean-velocity vector fields, temperature profiles, flame position, and fuel blowing velocity. A comparison of the distributions of instantaneous and mean mass fractions of reactants shows that the present method appropriately reveals the effects of large-scale turbulent motions on combustion. Furthermore, the present large-eddy simulations have achieved a significant improvement in predicting the mean effective reattachment length over the previous calculations incorporating with turbulence models. The physical insight regarding the decrease of the mean effective reattachment length with combustion was also addressed.

Compressibility effects and mixing enhancement in turbulent free shear flows

Computational simulations have been performed to study compressible, spatially developing turbulent free shear layers with various velocity regimes—subsonic/subsonic, supersonic/subsonic, and supersonic/supersonic— for convective Mach numbers in the range of 0.14-1.28. The numerical code used the finite volume technique and a modified Godunovs scheme. The computed results for the supersonic/subsonic case are first compared with experimental axial mean-velocity profiles, vorticity thickness, and turbulence parameters. Mixing layers with various velocity regimes are then calculated to investigate compressibility effects on the evolution of large-scale structures through the flow visualization of the vorticity field, growth rate of the vorticity thickness, and vorticity dynamics analysis. Various forcing frequencies are applied at the inflow boundary to examine mixing enhancement for free shear layers with higher convective Mach numbers. It is found for the first time that the growth rate of supersonic/supersonic free shear layers increases markedly when the forced layers move up and down with time instead of forming vortex roll-up and pairing.

Numerical visualization and residence time determination of turbulent reacting duct flow with mass bleed and a backstep on one wall

A numerical experiment was performed to study turbulent reacting duct flows with a backstep and mass bleed on one wall. The time-dependent two-dimensional compressible conservation equations were solved with a subgrid-scale turbulence model. The combustion process considered was a one-step, irreversible, and infinitely fast chemical reaction. The numerical code used the finite-volume technique, which involved alternating in time the second-order, explicit MacCormacks and modified Godunovs schemes. Computed mean-velocity profiles are compared with existing experimental data, and good agreement is attained. Various numerical visualization techniques are implemented to reveal the evolution of large-scale vortical structures, temperature and species contour maps, and paths of fuel and air particles. The numerical time lines experimental allows the particle residence time in the flame-holding recirculation zone to be determined directly by the Lagrangian method through residence time probability density function for the first time. Moreover, a compact exponential expression for the particle number decay rate is provided to correlate the numerical data. A comparison of the Eulerian method previously used by all the researchers to indirectly find the residence time with the present Lagrangian method is further performed. An insight into the size of the detecting area and its relation to the uncertainty associated with the residence time determination using the Eulerian method are then provided.

Flammability Limits and Probability Density Functions in Simulated Solid-Fuel Ramjet Combustors

A numerical analysis was performed to study turbulent reacting e ows in solid-fuel ramjet combustors. The time-dependent axisymmetric compressible conservation equations were solved with a subgrid-scale turbulence model. The combustion process considered was a one-step, irreversible, and ine nitely fast chemical reaction. The numerical code used the e nite volume technique, which involved alternating in time the second-order, explicit MacCormack’ s and modie ed Godunov’ s schemes. Computed temperature proe les are compared with existing experimental data and the previous calculations, incorporating the k-e turbulence model and two-layer near-wall treatment. The superiority of the present predictions is demonstrated. Flammability limits in solid-fuel ramjet combustors have been determined using the characteristic exhaust velocity. Moreover, a compact correlation relating the minimum step height to the combustor diameter, air mass e ow rate, and equivalence ratio for sustained combustion is provided for practical design reference. Probability density function (PDF) proe les of temperature and mixture fraction are also analyzed to understand the thermal structure of turbulent e ames, and to provide information for examining the applicability of the conserved-scalar approach incorporated PDF models to turbulentreacting e ows.

Numerical Visualization of Vortical Structures in a Spatially Evolving Turbulent Compressible Mixing Layer

Large-eddy simulations are performed to numerically visualize the generation of streamwise vortical structures and its interaction with spanwisely rolled-up coherent vortical structure during the spatial development of a turbulent supersonic/subsonic mixing layer at convective Mach numberMc=0.51. Time-dependent three-dimensional compressible conservation equations were solved with a subgrid-scale turbulence model. The numerical code used the finite-volume technique, which adopted alternately in temporal discretization the second-order, explicit MacCormack’s and modified Godunov’s schemes. Both transverse and spanwise perturbations were imposed initially for promoting the formation of spanwise rollers and counter-rotating streamwise vortices, respectively. Numerical visualizations are presented in terms of time-sequence isopressure surfaces and vorticity contours of spanwise and streamwise components. The results show that the spatial growth of three-dimensional vortical structures, in particular, the formation of chain-link-fence type structures, is adequately captured by the present computations. Vorticity dynamics is further analyzed, for the first time, to identify the dominant roles played by the convection effect followed by the vortex stretching effect on affecting the evolution of streamwise and spanwise vortical structures, respectively, forMc<0.6.

Heat Transfer and Fluid Flow in a Square Duct With 12 Different Shaped Vortex Generators

Detailed local Nusselt number distributions, streamwise mean flow patterns and cross-sectional secondary flow patterns, and friction factors in the first pass of a sharp turn two-pass square channel with various configurations of longitudinal vortex generator arranged on one wall were measured using transient liquid crystal thermography, laser-Doppler velocimetry, and pressure transducer probing, respectively. The Reynolds number, based on channel hydraulic diameter and bulk mean velocity, was fixed at 1.2 × 104. The vortex generator height-to-hydraulic diameter ratio and pitch-to-height ratio were 0.12 and 10, respectively. Comparisons in terms of heat transfer augmentation and uniformity and friction loss are first performed on 12 configurations of longitudinal vortex generator. The fluid dynamic mechanisms and wall confinement relevant to heat transfer enhancement are then documented for three-selected vortex generator models. In addition, the differences in fluid flow and heat transfer characteristics between a single vortex generator and a vortex generator array are addressed for the delta wing 1 U and 45° V U models which provide better thermal performance. The direction and strength of the secondary flow with respect to the heat transfer wall are found to be the most important fluid dynamic factors affecting the heat transfer promotion through the channel wall, followed by the convective mean velocity, and then the turbulent kinetic energy. Furthermore, the effects of the two-dimensional heat conduction near the vortex generator edge and unseen heat transfer areas on the Nusselt number estimation are documented in detail.Copyright

Excited Turbulent Reacting Flows in a Chamber With Gaseous Fuels Injecting Through Porous Walls

A numerical analysis was performed to study excited turbulent reacting flows in a combustion chamber consisting of gaseous fuels injecting through porous walls. The time-dependent axisymmetric compressible conservation equations were solved directly without using subgrid-scale turbulence models. The combustion process considered was a one-step, irreversible, and infinitely fast chemical reaction. The numerical code used the finite-volume technique, which involved alternating in time the second-order, explicit MacCormack’s and modified Godunov’s schemes. Excitations with various forcing frequencies were applied to the air inflow boundary to investigate their effects on the large-scale unsteady flow structures, vorticity dynamics, mean flame position, mean temperature distribution, and characteristic exhaust velocity.From the numerical flow visualization of the instantaneous vorticity field and the associated vorticity dynamics analysis, a rationale for the augmentation or diminution of organized vortical structures under external excitations was provided. It is found that an excitation with the fundamental frequency attains a larger high temperature recirculation zone with a higher peak temperature and hence a better flame-holding condition. An application of the active control by forcing the air inlet velocity with the fundamental frequency at various amplitudes to an original non-sustained combustion case indicates that the combustion can be sustained if an appropriate exciting amplitude is selected.Copyright