New parabolized Navier-Stokes code for three-dimensional chemically reacting flows
Explore the source record for details and available documents.
SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Research has been conducted for a number of years at the NASA Langley Research Center to develop a supersonic combustion ramjet (scramjet) capable of propelling a vehicle at hypersonic speeds in the atmosphere or beyond. Recently, that research has been directed toward the optimization of the scramjet combustor, and in particular the efficiency of fuel-air mixing and reaction in the engine. This paper describes a study of fuel-air mixing and reaction in a supersonic flow field, and discusses several techniques that were applied for enhancing the mixing processes and overal combustion efficiency in the flow. Based on the results of the study, an alternate fuel injector configuration was computationally designed, and that configuration significantly increased the amount of fuel-air mixing and combustion over a given combustor length that was achieved.
Explore the source record for details and available documents.
The two-dimensional, elliptic Navier-Stokes equations are used to investigate supersonic flows with finite-rate chemistry and radiation for hydrogen-air systems. The chemistry source terms in the species equation is treated implicitly to alleviate the stiffness associated with fast reactions. The explicit, unsplit MacCormack finite-difference scheme is used to advance the governing equations in time, until convergence is achieved. The specific problem considered is the premixed flow in a channel with a ten-degree compression ramp. Three different chemistry models are used, accounting for increasing number of reactions and participating species. Two chemistry models assure nitrogen as inert, while the third model accounts for nitrogen reactions and NO(x) formation. The tangent slab approximation is used in the radiative flux formulation. A pseudo-gray model is used to represent the absorption-emission characteristics of the participating species. Results obtained for specific conditions indicate that the radiative interactions vary substantially, depending on reactions involving HO2 and NO species and that this can have a significant influence on the flowfield.
An approximate axisymmetric method was developed which can reliably calculate fully viscous hypersonic flows over blunt nosed bodies. By substituting Maslen's second order pressure expression for the normal momentum equation, a simplified form of the viscous shock layer (VSL) equations is obtained. This approach can solve both the subsonic and supersonic regions of the shock layer without a starting solution for the shock shape. The approach is applicable to perfect gas, equilibrium, and nonequilibrium flowfields. Since the method is fully viscous, the problems associated with a boundary layer solution with an inviscid layer solution are avoided. This procedure is significantly faster than the parabolized Navier-Stokes (PNS) or VSL solvers and would be useful in a preliminary design environment. Problems associated with a previously developed approximate VSL technique are addressed before extending the method to nonequilibrium calculations. Perfect gas (laminar and turbulent), equilibrium, and nonequilibrium solutions were generated for airflows over several analytic body shapes. Surface heat transfer, skin friction, and pressure predictions are comparable to VSL results. In addition, computed heating rates are in good agreement with experimental data. The present technique generates its own shock shape as part of its solution, and therefore could be used to provide more accurate initial shock shapes for higher order procedures which require starting solutions.
Turbulent features in a shock tube's contact surface with the reversible reaction 2NO2 to N2O4 are explored. Turbulent densities and velocities show an overall power-law dependence in the fluctuation spectra. At fixed reactant percentages, turbulence increases with increasing Reynolds number. At fixed Reynolds numbers, increasing the reactant percentages causes a reduction in turbulence and an increase in the relative strength of discrete high-frequency components in the density fluctuation spectra. The reliability of conventional chemical modeling for this system apparently decreases with increasing turbulence in a manner suggestive of reaction rate distortion, consistent with theoretical speculation.
Explore the source record for details and available documents.
An implicit, time-dependent approach is taken to solve the inviscid conservative equations in a vertex-based finite-volume formulation. The convective fluxes are estimated by a modified Riemann solver, and higher-order accuracy is obtained by means of the MUSCL. Implicit techniques are adopted to deal with the equation stiffness as a result of the clustered grid and chemical source term. The real-gas Roe matrix and the ADI and LU schemes have been tested on the sample problems: hypersonic flow over blunt configurations, such as the forebody of a double ellipse, and the near wake of a conic aerobrake. Temperature distribution and standoff distance are validated against the shock-fitting, central-differencing results. Equilibrium temperature may be recovered by artificially increasing the chemical rates at the wall. The present method satisfactorily resolves the flow structure, including the strong expansion behind the flat base. The relative merits and performance of the real-gas TVD and implicit schemes are discussed briefly.
UV spontaneous vibrational Raman scattering and laser-induced predissociative fluorescence (LIPF) are combined and applied to a supersonic flame. For the first time, simultaneous measurements of temperature, major species (H2, O2, N2, H2O), and minor species (OH) concentrations are obtained with a 'single' excimer laser in a supersonic-lifted hydrogen-air diffusion flame. In the supersonic flame, a small amount of reaction occurs upstream of the lifted flame base, due to shock wave interactions and mixing with hot vitiated air. The strong turbulent mixing and high total enthalpy fluctuations lead to nonequilibrium values of temperature, and major and minor species concentrations. Combustion occurs farther downstream of the lifted region where slow three-body recombination reactions result in superequilibrium OH concentrations that depress the temperatures below their equilibrium values. Farther downstream, ambient air entrainment contaminates flame properties.
A second-order method is developed to correct a radiative transfer analysis for possible local thermodynamic nonequilibrium (LTNE) effects. It uses a two-species excitation model for nitrogen with chemical raction rates obtained from the Kunc and Soon (1989) atomic transition method. A vibrational energy equation is added to calculate a third temperature, which describes the average vibrational energy state of all the diatomic species. A new diffusional model is developed to improve the calculation of the diffusional fluxes of mass and energy, and second-order radiative correction factors are developed for a two-step excitation model for atomic nitrogen. Radiation reaching the wall with the second-order LTNE model shows a greater IR-line contribution and less UV-line center absorption than the first-order one.
Pursuant to a NASA-Langley development program for a scramjet HST propulsion system entailing the optimization of the scramjet combustor's fuel-air mixing and reaction characteristics, a numerical study has been conducted of the candidate parallel fuel injectors. Attention is given to a method for flow mixing-process and combustion-efficiency enhancement in which a supersonic circular hydrogen jet coflows with a supersonic air stream. When enhanced by a planar oblique shock, the injector configuration exhibited a substantial degree of induced vorticity in the fuel stream which increased mixing and chemical reaction rates, relative to the unshocked configuration. The resulting heat release was effective in breaking down the stable hydrogen vortex pair that had inhibited more extensive fuel-air mixing.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
A specific reactive flow configuration, the interaction of a detonation wave with convected homogeneous isotropic weak turbulence (which can be constructed by a Fourier synthesis of small amplitude shear waves) is addressed. The effect of chemical heat release on the rms fluctuations downstream of the detonation is presented as a function of Mach number. In addition, for the particular case of the von Karman spectrum, the one dimensional power spectra of these flow quantities is given.
Results of the numerical simulation of 15 degree downstream helium injection into a unconfined Mach 6 airstream are presented. Both details of downstream mixing and mean flow are in good agreement with experimental data. Results of the numerical simulation of similar hydrogen injection into a high enthalpy (Mach 17) confined Mach 6 airstream are presented with favorable comparison to experimental plume structure and wall data. The low enthalpy inflow from the unconfined case was then provided to the high enthalpy geometry in order to study the feasibility of using low enthalpy simulations of mixing for scramjet flight performance estimation. Results indicate that the mixing is substantially lower for the high enthalpy case but can be shown to appropriately scale using the inflow velocity. Production and decay of axial vorticity, cross-flow velocities, and the mean-flow velocities of these confined flows are then related and discussed to illustrate the effect of residence time on jet mixing.
The effects of compressibility, chemical reaction exothermicity and non-equilibrium chemical modeling in a combusting plane mixing layer were investigated by means of two-dimensional model free numerical simulations. It was shown that increased compressibility generally had a stabilizing effect, resulting in reduced mixing and chemical reaction conversion rate. The appearance of 'eddy shocklets' in the flow was observed at high convective Mach numbers. Reaction exothermicity was found to enhance mixing at the initial stages of the layer's growth, but had a stabilizing effect at later times. Calculations were performed for a constant rate chemical rate kinetics model and an Arrhenius type kinetics prototype. The Arrhenius model was found to cause a greater temperature increase due to reaction than the constant kinetics model. This had the same stabilizing effect as increasing the exothermicity of the reaction. Localized flame quenching was also observed when the Zeldovich number was relatively large.
A pilot scale hypersonic wind tunnel operating on pure iodine vapor has been designed and tested. The wind tunnel operates intermittently with a run phase lasting approximately 20 minutes. Successful recirculation of the iodine used during the run phase has been achieved but can be improved. Relevant issues regarding the full scale facility's design and operation, and the use of iodine as a working gas are discussed. Continuous wave laser induced fluorescence was used to monitor number densities within the plume flowfield, while pulsed laser induced fluorescence was used in an initial attempt to measure vibrational energy state population distributions. Preliminary nozzle flow calculations based on finite rate chemistry are presented.
A new two-temperature dissociation model for flows undergoing compression is derived from kinetic theory. The model minimizes uncertainties associated with the two-temperature model of Park. The effects of the model on AOTV type flowfields are examined and compared with the Park model. Calculations are carried out for flows with and without ionization. When considering flows with ionization, a four temperature model is employed. For Fire II conditions, the assumption of equilibrium between the vibrational and electron-electronic temperatures is somewhat poor. A similar statement holds for the translational and rotational temperatures. These trends are consistent with results obtained using the direct simulation Monte Carlo method.