Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Shear Layers”

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.

At least 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

Scalar mixing in the supersonic shear layer

Experiments were conducted in a two-stream planar mixing layer facility at convective Mach numbers of 0.28 and 0.62. Mie scattering from condensed alcohol droplets and planar laser-induced fluorescence (PLIF) of nitric oxide were used for flow visualization in both the side and plan views. The PLIF signals were approximately proportional to mixture fraction and were used to generate statistical quantities. Visualizations using both the Mie scattering and PLIF indicate the structure is essentially two-dimensional at Mc = 0.28 and three-dimensional at Mc = 0.62. Perspective renderings of side view images show the structures are streamwise ramped at Mc = 0.28 and cross-stream ramped at Mc = 0.62. This difference appears to be associated with decreasing streamwise structure spacings at the higher Mc condition. The statistical analysis suggests that with increasing compressibility, the scalar fluctuations are smaller, and the fraction of mixed fluid is higher.

Clemens, N. T.↗
NASA NTRS1991

Phase decorrelation of coherent structures in a free shear layer

The vortices near the origin of an initially laminar mixing layer have a single frequency with a well-defined phase; i.e., there is little phase jitter. Further downstream, however, the phase jitter increases suddenly. Even when the flow is forced, this same transition is observed. The forcing partially loses its influence because of the decorrelation of the phase between the forcing signal and the passing coherent structures. In the present investigation, this phenomenon is documented and the physical mechanism responsible for the phase decorrelation is identified.

Ho, Chih-Ming↗
NASA NTRS1990

A fractal transition in the two dimensional shear layer

The dependence of product generation with the Peclet and Reynolds number in a numerically simulated, reacting, two dimensional, temporally growing mixing layer is used to compute the fractal dimension of passive scalar interfaces. A transition from a low dimension of 4/3 to a higher one of 5/3 is identified and shown to be associated to the kinematic distortion on the flow field during the first pairing interaction. It is suggested that the structures responsible for this transition are non-deterministic, non-random, inhomogeneous fractals. Only the large scales are involved. No further transition is found for Reynolds numbers up to 20,000.

Jimenez, Javier↗
NASA NTRS1992

Mixing in straight shear layers

Planar laser-induced fluorescence measurements were performed in a liquid plane mixing layer to extract the probability density function (pdf) of the mixture fraction of a passive scalar across the layer. Three Reynolds number (Re) cases were studied, 10,000, 33,000 and 90,000, with Re based on velocity difference and visual thickness. The results show that a non-marching pdf (central hump invariant from edge to edge of the layer) exists for Re = 10,000 but that a marching type pdf characterizes the Re = 33,000 and Re = 90,000 cases. For all cases, a broad range of mixture fraction values is found at each location across the layer. Streamwise and spanwise ramps across the layer, and structure-to-structure variation were observed and are believed to be responsible for the above behavior of the composition field. Tripping the boundary layer on the high-speed side of the splitter plate for each of the above three cases resulted in increased three-dimensionality and a change in the composition field. Average and average mixed fluid compositions are reported for all cases.

Karasso, P. S.↗
NASA NTRS1992

The eigenvalue spectrum of the Rayleigh equation for a plane shear layer

The eigenvalue spectrum of the Rayleigh equation is examined using three different solution techniques. In particular, a simple second-order finite difference scheme and two spectral methods, the Chebyshev tau and Chebyshev collocation methods, are used to discretize the equation. All of the approximation methods are shown to be capable of predicting the discrete spectrum as well as the continuous spectrum associated with the critical point singularity for the Rayleigh equation. The global eigenvalue methods considered here provide an efficient way of obtaining either an approximation to the complete eigenvalue spectrum or initial guesses for a local shooting procedure for the discrete part of the spectrum.

Liou, William W.-W.↗
NASA NTRS1993

Parametric resonant triad interactions in a free shear layer

We investigate the weakly nonlinear evolution of a triad of nearly-neutral modes superimposed on a mixing layer with velocity profile u bar equals Um + tanh y. The perturbation consists of a plane wave and a pair of oblique waves each inclined at approximately 60 degrees to the mean flow direction. Because the evolution occurs on a relatively fast time scale, the critical layer dynamics dominate the process and the amplitude evolution of the oblique waves is governed by an integro-differential equation. The long-time solution of this equation predicts very rapid (exponential of an exponential) amplification and we discuss the pertinence of this result to vortex pairing phenomena in mixing layers.

Mallier, R.↗
NASA NTRS2000

Generation and Radiation of Acoustic Waves from a 2-D Shear Layer

A parallel numerical simulation of the radiation of sound from an acoustic source inside a 2-D jet is presented in this paper. This basic benchmark problem is used as a test case for scattering problems that are presently being solved by using the Impedance Mismatch Method (IMM). In this technique, a solid body in the domain is represented by setting the acoustic impedance of each medium, encountered by a wave, to a different value. This impedance discrepancy results in reflected and scattered waves with appropriate amplitudes. The great advantage of the use of this method is that no modifications to a simple Cartesian grid need to be made for complicated geometry bodies. Thus, high order finite difference schemes may be applied simply to all parts of the domain. In the IMM, the total perturbation field is split into incident and scattered fields. The incident pressure is assumed to be known and the equivalent sources for the scattered field are associated with the presence of the scattering body (through the impedance mismatch) and the propagation of the incident field through a non-uniform flow. An earlier version of the technique could only handle uniform flow in the vicinity of the source and at the outflow boundary. Scattering problems in non-uniform mean flow are of great practical importance (for example, scattering from a high lift device in a non-uniform mean flow or the effects of a fuselage boundary layer). The solution to this benchmark problem, which has an acoustic wave propagating through a non-uniform mean flow, serves as a test case for the extensions of the IMM technique.

Agarwal, Anurag↗
NASA NTRS1994

Effects of Heat Release on a Two-Dimensional Reacting Shear Layer

The diffusion controlled binary reaction between initially segregated reactants in a two-dimensional low Mach number mixing layers is studied via numerical simulation. The stoichiometric ratio of the reactants is chosen to be much larger than one, as is typical of hydrocarbon flames in air. This results is a flame that is offset from the main vortical region of the mixing layer. In agreement with experimental observations, the flame remains surprisingly uncontorted during the flow evolution and is not entrained into the mixing layer. The effect of the heat release of the flame on the evolution of the mixing layer is thus similar to the effect of a difference in free-stream density between the two sides of the layer. The resulting baroclinic torque inhibits the familiar rolup and pairing of mixing layer vortices common in constant density flows. This also contributes to the layers inability to entrain the flame. The increase in viscosity caused by the heating of the flame reduces the effective Reynolds number of the flow. But, contrary to what has commonly been suggested, this is not the major reason for the inhibition of the usual large-scale mixing layer structures.

Moser, Robert↗
NASA NTRS2006

Development of Comprehensive Reduced Kinetic Models for Supersonic Reacting Shear Layer Simulations

Large-scale simulations of multi-dimensional unsteady turbulent reacting flows with detailed chemistry and transport can be computationally extremely intensive even on distributed computing architectures. With the development of suitable reduced chemical kinetic models, the number of scalar variables to be integrated can be decreased, leading to a significant reduction in the computational time required for the simulation with limited loss of accuracy in the results. A general MATLAB-based automated mechanism reduction procedure is presented to reduce any complex starting mechanism (detailed or skeletal) with minimal human intervention. Based on the application of the quasi steady-state (QSS) approximation for certain chemical species and on the elimination of the fast reaction rates in the mechanism, several comprehensive reduced models, capable of handling different fuels such as C2H4, CH4 and H2, have been developed and thoroughly tested for several combustion problems (ignition, propagation and extinction) and physical conditions (reactant compositions, temperatures, and pressures). A key feature of the present reduction procedure is the explicit solution of the concentrations of the QSS species, needed for the evaluation of the elementary reaction rates. In contrast, previous approaches relied on an implicit solution due to the strong coupling between QSS species, requiring computationally expensive inner iterations. A novel algorithm, based on the definition of a QSS species coupling matrix, is presented to (i) introduce appropriate truncations to the QSS algebraic relations and (ii) identify the optimal sequence for the explicit solution of the concentration of the QSS species. With the automatic generation of the relevant source code, the resulting reduced models can be readily implemented into numerical codes.

Zambon, A. C.↗
NASA NTRS2013

Low Velocity Difference Thermal Shear Layer Mixing Rate Measurements

Current CFD modeling techniques are known to do a poor job of predicting the mixing rate and persistence of slot film flow in co-annular flowing ducts with relatively small velocity differences but large thermal gradients. A co-annular test was devised to empirically determine the mixing rate of slot film flow in a constant area circular duct (D approx. 1ft, L approx. 10ft). The axial rate of wall heat-up is a sensitive measure of the mixing rate of the two flows. The inflow conditions were varied to simulate a variety of conditions characteristic of moderate by-pass ratio engines. A series of air temperature measurements near the duct wall provided a straightforward means to measure the axial temperature distribution and thus infer the mixing rate. This data provides a characterization of the slot film mixing rates encountered in typical jet engine environments. The experimental geometry and entrance conditions, along with the sensitivity of the results as the entrance conditions vary, make this a good test for turbulence models in a regime important to modern air-breathing propulsion research and development.

Bush, Robert H.↗
NASA NTRS2015

Comparison of High-Order and Low-Order Methods for Large-Eddy Simulation of a Compressible Shear Layer

The objective of this work is to compare a high-order solver with a low-order solver for performing Large-Eddy Simulations (LES) of a compressible mixing layer. The high-order method is the Wave-Resolving LES (WRLES) solver employing a Dispersion Relation Preserving (DRP) scheme. The low-order solver is the Wind-US code, which employs the second-order Roe Physical scheme. Both solvers are used to perform LES of the turbulent mixing between two supersonic streams at a convective Mach number of 0.46. The high-order and low-order methods are evaluated at two different levels of grid resolution. For a fine grid resolution, the low-order method produces a very similar solution to the highorder method. At this fine resolution the effects of numerical scheme, subgrid scale modeling, and filtering were found to be negligible. Both methods predict turbulent stresses that are in reasonable agreement with experimental data. However, when the grid resolution is coarsened, the difference between the two solvers becomes apparent. The low-order method deviates from experimental results when the resolution is no longer adequate. The high-order DRP solution shows minimal grid dependence. The effects of subgrid scale modeling and spatial filtering were found to be negligible at both resolutions. For the high-order solver on the fine mesh, a parametric study of the spanwise width was conducted to determine its effect on solution accuracy. An insufficient spanwise width was found to impose an artificial spanwise mode and limit the resolved spanwise modes. We estimate that the spanwise depth needs to be 2.5 times larger than the largest coherent structures to capture the largest spanwise mode and accurately predict turbulent mixing.

large eddy simulations↗
NASA NTRS2015

Comparison of High-Order and Low-Order Methods for Large-Eddy Simulation of a Compressible Shear Layer

The objective of this work is to compare a high-order solver with a low-order solver for performing large-eddy simulations (LES) of a compressible mixing layer. The high-order method is the Wave-Resolving LES (WRLES) solver employing a Dispersion Relation Preserving (DRP) scheme. The low-order solver is the Wind-US code, which employs the second-order Roe Physical scheme. Both solvers are used to perform LES of the turbulent mixing between two supersonic streams at a convective Mach number of 0.46. The high-order and low-order methods are evaluated at two different levels of grid resolution. For a fine grid resolution, the low-order method produces a very similar solution to the high-order method. At this fine resolution the effects of numerical scheme, subgrid scale modeling, and filtering were found to be negligible. Both methods predict turbulent stresses that are in reasonable agreement with experimental data. However, when the grid resolution is coarsened, the difference between the two solvers becomes apparent. The low-order method deviates from experimental results when the resolution is no longer adequate. The high-order DRP solution shows minimal grid dependence. The effects of subgrid scale modeling and spatial filtering were found to be negligible at both resolutions. For the high-order solver on the fine mesh, a parametric study of the spanwise width was conducted to determine its effect on solution accuracy. An insufficient spanwise width was found to impose an artificial spanwise mode and limit the resolved spanwise modes. We estimate that the spanwise depth needs to be 2.5 times larger than the largest coherent structures to capture the largest spanwise mode and accurately predict turbulent mixing.

large eddy simulations↗