SEARCH · Search NASA
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.
Investigation of the Mach reflection associated with a steady two-dimensional overexpanded nozzle flow. The conditions of occurrence of triple point shock configuration are described, and Mach reflection configurations for two values of the incident shock are presented, along with the Mach stem height variation for different Mach numbers under conditions of different pressure ratios.
An implicit, finite volume code for solving two dimensional, compressible turbulent flows is described. Second order upwind differencing of the inviscid terms of the equations is used to enhance stability and accuracy. A diagonal form of the implicit algorithm is used to improve efficiency. Several zero and two equation turbulence models are incorporated to study their impact on overall flow modeling accuracy. Applications to external and internal flows are discussed.
Numerical predictions of the Mars Science Laboratory (MSL) reaction control system (RCS) jets interacting with a Mach 10 hypersonic flow are compared to experimental nitric oxide (NO) planar laser-induced fluorescence (PLIF) data. The steady Reynolds Averaged Navier Stokes (RANS) equations using the Baldwin-Barth one-equation turbulence model were solved using the OVERFLOW code. The experimental PLIF data used for comparison consists of qualitative two-dimensional visualization images, qualitative reconstructed three-dimensional flow structures, and quantitative two-dimensional distributions of streamwise velocity. Through modeling of the PLIF signal equation, computational flow images (CFI) were produced and directly compared to the qualitative PLIF data. Post processing of the experimental and simulation data enabled the jet trajectory to be extracted for a more quantitative comparison. The two-dimensional velocity fields were reconstructed through interpolation of a series of single-component velocity profiles. Each distribution of single-component velocity was obtained using molecular tagging velocimetry (MTV). After validating the numerical model, the numerical solution was further examined to gain insight into hypersonic jet-in-crossflow interaction. Future NO-PLIF experiments are proposed based on this analysis.
The numerical solution of the Euler or Navier-Stokes equations by Lagrangian vortex methods is discussed. The mathematical background is presented and includes the relationship with traditional point-vortex studies, convergence to smooth solutions of the Euler equations, and the essential differences between two and three-dimensional cases. The difficulties in extending the method to viscous or compressible flows are explained. Two-dimensional flows around bluff bodies are emphasized. Robustness of the method and the assessment of accuracy, vortex-core profiles, time-marching schemes, numerical dissipation, and efficient programming are treated. Operation counts for unbounded and periodic flows are given, and two algorithms designed to speed up the calculations are described.
Incompressible laminar boundary layer development over two-dimensional symmetric body in presence of shear flow
An investigation was made of the flow downstream from a "two-dimensional" grid formed of parallel rods. In both two and three dimensional jet fields there is a critical range of grid density below which the downstream flow is stable and above which it is unstable. The flow can be completely stabilized by means of an adequate lateral contraction beginning immediately after the grid or by use of a fine-mesh damping screen parallel to the grid plane and within a definite range of positions downstream from the grid.
This paper describes numerical simulations of self-excited oscillations in a two-dimensional transonic diffuser flow obtained by solving the Navier-Stokes equations with a two-equation turbulence model. Comparisons were made between the computational results and experimental data. For the mean flowfields, the agreement between computation and experiment is good for the wall pressures, shock location, and the separation and reattachment points. However, the thickness of the computed recirculation zone is about 50 percent of the measured thickness. For the fluctuating flowfields, a great deal of qualitative similarity exists between the computation and experiment; however, the predicted oscillation frequency is about 50 percent higher than the measured value. The formation of a succession of downstream-traveling counter-rotating vortices, as seen experimentally, is also vividly displayed in the numerical results.
Displacement thickness for two-dimensional or axisymmetric compressible boundary layer with blowing
Spark shadowgraphs, high-speed schlieren movies, and the tracing of individual eddies were used to study a turbulent reacting shear layer in a premixed propane/air flow. The flow took place in a two-dimensional combustor, and the flame was stabilized behind a rearward facing streamlined step. The data show that in the velocity range 7.5 to 22.5 m/sec the mixing layer is dominated by Brown-Roshko type large coherent structures in both reacting and nonreacting flows. Information on the eddies, the flame, and the distribution of gas concentrations is presented.
Numerical predictions of the Mars Science Laboratory reaction control system jets interacting with a Mach 10 hypersonic flow are compared to experimental nitric oxide planar laser-induced fluorescence data. The steady Reynolds Averaged Navier Stokes equations using the Baldwin-Barth one-equation turbulence model were solved using the OVERFLOW code. The experimental fluorescence data used for comparison consists of qualitative two-dimensional visualization images, qualitative reconstructed three-dimensional flow structures, and quantitative two-dimensional distributions of streamwise velocity. Through modeling of the fluorescence signal equation, computational flow images were produced and directly compared to the qualitative fluorescence data.
The effects of nonequilibrium chemistry and three dimensionality on the stability characteristics of hypersonic flows are discussed. In two-dimensional (2-D) and axisymmetric flows, the inclusion of chemistry causes a shift of the second mode of Mack to lower frequencies. This is found to be due to the increase in size of the region of relative supersonic flow because of the lower speeds of sound in the relatively cooler boundary layers. Although this shift in frequency is present in both the equilibrium and nonequilibrium air results, the equilibrium approximation predicts modes which are not observed in the nonequilibrium calculations (for the flight conditions considered). These modes are superpositions of incoming and outgoing unstable disturbances which travel supersonically relative to the boundary-layer edge velocity. Such solutions are possible because of the finite shock stand-off distance. Their corresponding wall-normal profiles exhibit an oscillatory behavior in the inviscid region between the boundary-layer edge and the bow shock. For the examination of three-dimensional (3-D) effects, a rotating cone is used as a model of a swept wing. An increase of stagnation temperature is found to be only slightly stabilizing. The correlation of transition location (N = 9) with parameters describing the crossflow profile is discussed. Transition location does not correlate with the traditional crossflow Reynolds number. A new parameter that appears to correlate for boundary-layer flow was found. A verification with experiments on a yawed cone is provided.
A linear, two-dimensional model of a transonic plasma flow in equilibrium is constructed and given an explicit solution in the form of a complex Laplace integral. The solution indicates that the transonic state can be solved as an elliptic boundary value problem, as is done in the numerical code FLOW [Guazzotto et al., Phys. Plasmas 11, 604 (2004)]. Moreover, the presence of a hyperbolic region does not necessarily imply the presence of a discontinuity or any other singularity of the solution.
Inverse solution of two dimensional gas dynamic flow fields of rotational or irrotational character
A method was developed which calculates two-dimensional, transonic, viscous flow in ducts. The finite volume, time-marching formulation is used to obtain steady flow solutions of the Reynolds-averaged form of the Navier-Stokes equations. The entire calculation is performed in the physical domain. This paper investigates the introduction of a new formulation of the energy equation which gives improved transient behavior as the calculation converges. The effect of variable Prandtl number on the temperature distribution through the boundary layer is also investigated. A turbulent boundary layer in an adverse pressure gradient (M = 0.55) is used to demonstrate the improved transient temperature distribution obtained when the new formulation of the energy equation is used. A flat plate turbulent boundary layer with a supersonic free-stream Mach number of 2.8 is used to investigate the effect of Prandtl number on the distribution of properties through the boundary layer. The computed total temperature distribution and recovery factor agree well with the measurements when a variable Prandtl number is used through the boundary layer.
A method was developed which calculates two-dimensional, transonic, viscous flow in ducts. The finite volume, time-marching formulation is used to obtain steady flow solutions of the Reynolds-averaged form of the Navier-Stokes equations. The entire calculation is performed in the physical domain. This paper investigates the introduction of a new formulation of the energy equation which gives improved transient behavior as the calculation converges. The effect of variable Prandtl number on the temperature distribution through the boundary layer is also investigated. A turbulent boundary layer in an adverse pressure gradient (M = 0.55) is used to demonstrate the improved transient temperature distribution obtained when the new formulation of the energy equation is used. A flat plate turbulent boundary layer with a supersonic free-stream Mach number of 2.8 is used to investigate the effect of Prandtl number on the distribution of properties through the boundary layer. The computed total temperature distribution and recovery factor agree well with the measurements when a variable Prandtl number is used through the boundary layer.
A research program was undertaken to develop a new experimental technique commonly known as particle image displacement velocity (PIVD) to measure an instantaneous two dimensional velocity field in a selected plane of flow field. This technique was successfully developed and applied to the study of several aerodynamic problems. A detailed description of the technique and a broad review of all the research activity carried out in this field are reported. A list of technical publications is also provided. The application of PIDV to unsteady flows with large scale structures is demonstrated in a study of the temporal evolution of the flow past an impulsively started circular cylinder. The instantaneous two dimensional flow in the transition region of a rectangular air jet was measured using PIDV and the details are presented. This experiment clearly demonstrates the PIDV capability in the measurement of turbulent flows. Preliminary experiments were also conducted to measure the instantaneous flow over a circular bump in a transonic flow. Several other experiments now routinely use PIDV as a non-intrustive measurement technique to obtain instantaneous two dimensional velocity fields.
An algorithm is presented for unsteady two-dimensional incompressible Navier-Stokes calculations. This algorithm is based on the fourth order partial differential equation for incompressible fluid flow which uses the streamfunction as the only dependent variable. The algorithm is second order accurate in both time and space. It uses a multigrid solver at each time step. It is extremely efficient with respect to the use of both CPU time and physical memory. It is extremely robust with respect to Reynolds number.
An algorithm is presented for unsteady two-dimensional incompressible Navier-Stokes calculations. This algorithm is based on the fourth order partial differential equation for incompressible fluid flow which uses the streamfunction as the only dependent variable. The algorithm is second order accurate in both time and space. It uses a multigrid solver at each time step. It is extremely efficient with respect to the use of both CPU time and physical memory. It is extremely robust with respect to Reynolds number.