Search NASA⌕ Search

SEARCH · Search NASA

Results for “shear layer turbulence”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

LIF measurements of scalar mixing in turbulent shear layers

The structure of shear layer flows at high Reynolds numbers remains a very interesting problem. Straight mixing layers have been studied and yielded information on the probability density function (pdf) of a passive scalar across the layer. Konrad and Koochesfahani & Dimotakis measured the pdf of the mixture fraction for mixing layers of moderate Reynolds numbers, each about 25,000 (Re based on velocity difference and visual thickness). Their measurements showed a 'non-marching' pdf (central hump which is invariant from edge to edge across the layer), a result which is linked to the visualizations of the spanwise Kelvin-Helmholtz (K-H) instability mode, which is the primary instability for plane shear layer flows. A secondary instability mode, the Taylor-Gortler (T-G) instability, which is associated with streamwise vortical structures, has also been observed in shear layers. Image reconstruction by Jimenez et al. and volume renderings by Karasso & Mungal at low Re numbers have demonstrated that the K-H and the T-G instability modes occur simultaneously in a non-mutually destructive way, evidence that supports the quasi two-dimensional aspect of these flows and the non-marching character of the pdf at low Reynolds numbers. At higher Re numbers though, the interaction of these two instability modes is still unclear and may affect the mixing process. In this study, we perform measurements of the concentration pdf of plane mixing layers for different operating conditions. At a speed ratio of r = U(sub 1)/U(sub 2) = 4:1, we examine three Reynolds number cases: Re = 14,000, Re = 31,000, and Re = 62,000. Some other Re number cases' results, not presented in detail, are invoked to explain the behavior of the pdf of the concentration field. A case of r = 2.6:1 at Re = 20,000 is also considered. The planar laser-induced fluorescence technique is used to yield quantitative measurements. The different Re are obtained by changing the velocity magnitudes of the two streams. The question of resolution of these measurements is addressed. In order to investigate the effects of the initial conditions on the development and the structure of the mixing layer, the boundary layer on the high-speed side of the splitter plate is tripped. The average concentration and the average mixed fluid concentration are also calculated to further understand the changes in the shear layer for the different cases examined.

Karasso, Paris S.↗
NASA NTRS1981

A reattaching free shear layer in compressible turbulent flow - A comparison of numerical and experimental results

An investigation of a two-dimensional, free turbulent shear layer reattaching on an inclined surface at Mach 2.92 and at a high Reynolds number is described. The test geometry is specifically designed to isolate the reattachment process of a high-speed separated flow. A numerical solution of the time-dependent, Reynolds-averaged, Navier-Stokes equations for the entire flow field, employing a two-equation eddy viscosity turbulence model, is presented. Detailed comparisons of prediction and experiment are made in the free shear layer, at reattachment, and in the developing boundary layer downstream. These comparisons include mean surface quantities as well as mean and fluctuating flowfield quantities. Although the overall features of this complex flow field are predicted, there are several deficiencies in the numerical solution, particularly in the region downstream of reattachment. Modifications of the turbulence model to correct these deficiencies are discussed.

Horstman, C. C.↗
NASA NTRS1981

A Reattaching Free Shear Layer in Compressible Turbulent Flow: A Comparison of Numerical and Experimental Results

An investigation of the two-dimensional, free turbulent shear layer reattaching on an inclined surface at Mach 2.92 and at a high Reynolds number is described. The test geometry is specifically designed to isolate the reattachment process of a high-speed separated flow. A numerical solution of the time-dependent, Reynolds-averaged, Navier-Stokes equations for the entire flow field, employing a two-equation eddy viscosity turbulence model, is presented. Detailed comparisons of prediction and experiment are made in the free shear layer, at reattachment, and in the developing boundary layer downstream. These comparisons include mean surface quantities as well as mean and fluctuating flow-field quantities. Although the overall features of this complex flow field are predicted, there are several deficiencies in the numerical solution, particularly in the region downstream of reattachment. Modifications of the turbulence model to correct these deficiencies are discussed.

Horstman, C. C.↗
NASA NTRS1974

Calculation of compressible turbulent free shear layers

Rudy and Bushnell's (1972) mixing length theory for the prediction of mean velocity fields in turbulent shear layers is extended to include supersonic free turbulent mixing. The normal momentum equation is coupled with the conventional equations of motion and is solved iteratively. Results show that transverse static pressure variation has very little effect on the mean flow variables in compressible free turbulent shear layers.

Oh, Y. H.↗
NASA NTRS1994

The Roles of Large- and Small-Scale Structures in Turbulent Free Shear Layers

The occurrence of large-scale coherent structures in turbulent free shear flows (especially the planar mixing layer) has been recognized for some time. Indeed, the observation of such structures in mixing layers did much to promote interest in the study of coherent structures in turbulence. It has been widely assumed that the large-scale structures in these flows are responsible for the entrainment of free-stream fluid and the overall growth of the layer, while the small-scale structures provide mixing and dissipation. A model of scalar mixing based on these ideas was proposed for these flows. However, recent experimental and computational evidence suggests that the dominance of the large-scale structures in turbulent mixing layers is not universal. In addition, there is a substantial variation among experiments in several statistical measures of self-similar mixing layers, for example growth rate and velocity variances. To investigate the importance of large-scale structures, several free shear flows (mixing layers and wakes) have been simulated via direct numerical simulation. The simulations are designed to mimic experimental mixing layers in which the splitter plate boundary layers are turbulent. Different levels of two-dimensional forcing are included resulting in large-scale structures of differing strength and importance. These simulations are used to investigate the role of large-scale coherent structures in free shear layers and the effect of these structures on relevant turbulence statistics and scalar mixing. It is found that the statistics and structures in several experiments involving turbulent mixing layers are in better agreement with simulations that do not exhibit dominant large-scale structures than those in which the common mixing layer structures do dominate. It is also found that the level of forcing can have a profound effect on the qualitative and quantitative features of these shear layer, even when they are nominally self-similar.

Moser, Robert D.↗
NASA NTRS1981

Large-eddy simulation of the shear-free turbulent boundary layer

The shear-free turbulent boundary layer is calculated by the large-eddy simulation technique. The filtered Navier-Stokes equations are used; the method of integration employs Fourier expansions in the homogeneous directions and finite differences in the cross-stream direction. Results indicate that the simulation is capable of predicting the primary Reynolds-number effects.

Biringen, S.↗
NASA NTRS1981

Experimental study of combustion in a turbulent free shear layer formed at a rearward facing step

A premixed propane-air flame is stabilized in a turbulent free shear layer formed at a rearward facing step. The mean and rms averages of the turbulent velocity flow field are determined by LDV for both reacting (equivalence ratio 0.57) and nonreacting flows (Reynolds number 15,000-37,000 based on step height). The effect of combustion is to shift the layer toward the recirculation zone and reduce the flame spread. For reacting flow, the growth rate is unchanged except very near the step. The probability density function of the velocity is bimodial near the origin of the reacting layer and single-peaked but often skewed elsewhere. Large-scale structures dominate the reacting shear layer. Measurements of their passing frequency from LDV are consistent with high-speed Schlieren movies of the reacting layer and indicate that the coalescence rate of the eddies in the shear layer is reduced by combustion.

Pitz, R. W.↗
NASA NTRS1975

The measurement of shear stress and total heat flux in a nonadiabatic turbulent hypersonic boundary layer

Turbulent shear stress and direct turbulent total heat-flux measurements have been made across a nonadiabatic, zero pressure gradient, hypersonic boundary layer by using specially designed hot-wire probes free of strain-gauging and wire oscillation. Heat-flux measurements were in reasonably good agreement with values obtained by integrating the energy equation using measured profiles of velocity and temperature. The shear-stress values deduced from the measurements, by assuming zero correlation of velocity and pressure fluctuations, were lower than the values obtained by integrating the momentum equation. Statistical properties of the cross-correlations are similar to corresponding incompressible measurements at approximately the same momentum-thickness Reynolds number.

Mikulla, V.↗
NASA NTRS1974

Calculation of turbulent shear stress in supersonic boundary layer flows

Turbulent shear stress distributions for supersonic boundary layer flows have been computed from experimental mean boundary layer data. The computations have been made by numerically integrating the time averaged continuity and streamwise momentum equations. Distributions have been obtained for flows upstream and downstream of shock-wave-boundary layer interactions and for both two-dimensional and axisymmetric flows. The computed results are compared with recently reported shear stress measurements which were obtained by hot wire anemometer and laser velocimeter techniques.

Sun, C. C.↗
NASA NTRS2000

High-Frequency Excitation of a Plane Wake

In the early 1990's, Glezer and his co-workers at Georgia Tech made a startling discovery. They found that forcing at frequencies too high to directly affect the production scales led to a dramatic alteration in the development of a turbulent shear layer. An experimental study of this phenomenon is presented in Wiltse and Glezer. They used piezoelectric actuators located near the jet exit plane to force the shear layers of a square low-speed jet. The actuators were driven at a high frequency in the Kolmogorov inertial subrange, much higher than the frequencies associated with the large-scale motion (where the turbulent energy is produced and located) but much lower than those associated with the Kolmogorov scale (where the turbulent energy is dissipated). Measurements of the shear-layer turbulence showed that direct excitation of small-scale motion by high-frequency forcing led to an increase in the turbulent dissipation of more than an order of magnitude in the initial region of the shear layer! The turbulent dissipation gradually decreased with downstream distance but remained above the corresponding level for the unforced flow at all locations examined. The high-frequency forcing increased the turbulent kinetic energy in the initial region near the actuators, but the kinetic energy decreased quite rapidly with downstream distance, dropping to levels that were a small fraction of the level for the unforced case. Perhaps most importantly from the present standpoint, the high-frequency forcing significantly decreased the energy in the large-scale motion, increasingly so with downstream distance. Wiltse and Glezer interpreted this behavior as an enhanced transfer of energy from the large scales to the small scales. The initial work by Wiltse and Glezer has expanded into other applications. To explore the potential of high-frequency forcing for active acoustic suppression, in 1998 the first author proposed a set of experiments involving an edge tone shear layer and an open cavity flow. This work was funded by the US Air Force Research Laboratory, and the experiments were developed and executed at Boeing by Raman and Kibens. These experiments involved high-frequency forcing applied to low-speed flows using wedge piezo actuators and powered resonance tubes. The system is simple, open loop, compact, potentially requires little power, and is easily integrated. Dramatic results, such as reductions of 20 dB in spectral peaks and 5-8 dB in overall levels across the entire acoustic spectrum, were obtained in some cases. Sample results are presented. Following this success in low-speed flows, an international cooperative program continuing this work involved transonic experiments in a mid-size facility in the United Kingdom. Similar reductions in noise level were obtained in these transonic experiments. Discussion of this work is given in Raman et at. and Stanek, Raman, Kibens, and Ross. Other experiments at Georgia Tech have shown significant potential of high-frequency forcing in controlling reaction rates in chemically reacting flows.

Cain, Alan B.↗
NASA NTRS1988

Numerical study of turbulent internal shear layer flow in an axisymmetric U-duct

This paper presents a simple algebraic turbulence model developed for internal flows which is based on the vorticity length and magnitude. The length scale is defined using the distance between the maximum and the minimum points of the absolute vorticity. This model is numerically tested in a turbulent internal layer flow through an axisymmetric U-duct with very sharp curvature; the ratio of the boundary layer thickness to the radius of curvature of the duct is of order 1. In this U-duct flow, strong adverse and favorable pressure gradients coexist and interact with each other. Satisfactoy agreement with experimental results is obtained.

Chang, James L. C.↗
NASA NTRS1981

Three-dimensional structures and turbulence closure of the wake developing in a wall shear layer

The turbulent wake interacting with the rotating wall shear layer is investigated analytically and numerically. The turbulent wakes of the rotating blades in a compressor which are interacting with the rotating hub-wall boundary layer are analyzed. A modified version of the closure model of the pressure-strain correlation term in the Reynolds stress transport equation is developed to predict the effect of rotation, which is appreciable for the present flow because the thick hub-wall boundary layer is interacting with the rotor wake. It is noted that the Poisson type equation for the pressure-strain correlation has an extra rotation term when the entire flow field is rotating. This extra rotation term is modeled to accommodate the effect of rotation. In addition, the standard correction for the wall effect is incorporated for the utilized Reynolds stress closure model. The rotation-modified Reynolds stress closure model is used to predict the present flow, and the predictions are compared with the experimental data. The experimental data reveal that the characteristics of the three-dimensional turbulent wake interacting with the wall shear layer are considerably altered by the effects of the wall and the rotation. These features are predicted with good accuracy by the turbulence closure model developed.

Hah, C.↗
NASA NTRS1993

CFD validation of subsonic turbulent planar shear layers

The primary objective of the present study is to assess the limitations and capabilities of RPLUS in predicting the entrainment, mixing, and burning characteristics of a high subsonic planar shear layer; these computations are performed in conjunction with an experiment being conducted at NASA Lewis Research Center. Turbulence is modeled by a two-equation k-epsilon closure modified for compressibility effects. Combustion is modeled by two different models: one in which a finite-rate laminar chemistry model is used for H2-O2 oxidation, and the other based on a composition joint pdf approach to account for the turbulence-chemistry interaction effects. Between upwind and centered differencing schemes, minimal differences are observed for the flow properties of a nonreacting shear layer case studied. In comparison with experimental data, computed growth rates are underpredicted in both reacting and nonreacting cases of a shear layer, but more severely for the reacting flows. However, the qualitative agreement between computation and experiment is reasonably good for the other observable characteristics, and indicative of potential for further improvements.

Lai, H. T.↗
NASA NTRS1993

Computation of turbulent flows over backward and forward-facing steps using a near-wall Reynolds stress model

Separation and reattachment of turbulent shear layers is observed in many important engineering applications, yet it is poorly understood. This has motivated many studies on understanding and predicting the processes of separation and reattachment of turbulent shear layers. Both of the situations in which separation is induced by adverse pressure gradient, or by discontinuities of geometry, have attracted attention of turbulence model developers. Formulation of turbulence closure models to describe the essential features of separated turbulent flows accurately is still a formidable task. Computations of separated flows associated with sharp-edged bluff bodies are described. For the past two decades, the backward-facing step flow, the simplest separated flow, has been a popular test case for turbulence models. Detailed studies on the performance of many turbulence models, including two equation turbulence models and Reynolds stress models, for flows over steps can be found in the papers by Thangam & Speziale and Lasher & Taulbee). These studies indicate that almost all the existing turbulence models fail to accurately predict many important features of back step flow such as reattachment length, recovery rate of the redeveloping boundary layers downstream of the reattachment point, streamlines near the reattachment point, and the skin friction coefficient. The main objectives are to calculate flows over backward and forward-facing steps using the NRSM and to make use of the newest DNS data for detailed comparison. This will give insights for possible improvements of the turbulence model.

Ko, Sung HO↗