Session R02: Turbulence: Shear Layers

Dense gas effects on turbulent fluctuations of compressible turbulent shear layer

Supercritical carbon dioxide (SCO2) is widely utilized in numerous current and developing engineering applications. Near the critical conditions, the dense-gas effects of thermodynamics and transport properties are prominent and need to be accounted for. As a canonical configuration, a set of compressible SCO2 planar shear layer flows are investigated using direct numerical simulations that incorporate the equation of state and transport model for real gasses. The growth of shear layers under various conditions across a range of convective Mach numbers and density ratios are studied. The dense-gas effects on flow structures are identified for each condition. The evolutions of turbulence statistics among different flow conditions are compared. The departure function value as the metric of the dense gas effects needs to be considered in addition to the convective Mach number, initial Reynolds number, and Atwood number to characterize the flow conditions used in ideal gas. By considering the departure function, the growth of turbulent density and temperature fluctuations at different freestream temperatures have shown an improved similarity behavior.     

Coherent Motions and Mixing Dynamics of Single-Stream Shear Layers: Insights from 4D-PTV Measurements

The present study addresses the need for comprehensive volumetric measurements within a single-stream shear layer (SSSL), with the aim of advancing our understanding of intrinsic flow dynamics. A 4D-PTV technique is employed to capture the separation of a wall boundary layer and its subsequent development into a self-similar SSSL. The separation region is characterized by viscous domination and a negative production rate, which is further corroborated by a turbulent kinetic energy budget analysis. An evaluation of the statistical flow descriptors shows self-similarity achieved at a significantly shorter distance ≈60θ₀ (θ₀: initial momentum thickness), contrasting with the past reports. Moreover, a comparison of shear layer width, momentum, and vorticity thickness growth rates with literature reports raises intriguing questions about the potential universality of the growth rate scaling for single- and two-stream shear layers. Phase-averaged maps underline the importance of rollup-braid intersection regions as high-production regions. Additionally, the outer edges of coherent regions exhibit peak production, vortex cores favor higher dissipation rates, and the braid region acts as an energy conduit between contiguous roll-ups.   

On the experimental study of 3D turbulent shear layers

Turbulent mixing layers are formed when parallel streams flowing at different speed merge. In the usual laboratory model, the mean flow varies mainly in the directions of the velocity difference and flow convection, with no significant changes in the other direction. However, real flows are three dimensional (3D)−­the incoming flows are not parallel and have not only different speeds but also different directions, thereby giving rise to 3D turbulent shear layers, i.e., skewed mixing layers, the detailed quantitative study of which is missing. Flow skewing in turbulent boundary layers is reported to cause a reduction in Reynolds stresses and drag. We experimentally investigated the mean flow and statistics of skewed mixing layers in a wind tunnel to find out if similar effects prevail. We generated 3D shear layers by skewing the mean flow with turning vanes at the trailing edge of a splitter plate, and used pitot-static tubes and x-wires to probe the flow at different cross-, span- and downstream distances. Preliminary results show that skewed mixing layers, like their 2D counterpart, also spread linearly with downstream distance, and that an approx. homogenous shear develops further downstream. At the resolution of our current measurements, the shear layer thickness and shear rate seem not to depend strongly on skewing of the flow. We seek an explanation for this in the turbulence statistics. Understanding the effect of three dimensionality on mixing layers will help us improve designs of engineering structures.   

Three-dimensional effects in turbulent shear layers

The skewing of mean flow in canonical turbulent boundary layers is known to cause a reduction

in turbulent shear stress and drag, possibly due to the misalignment of turbulent structures.

However, little is known about the impact of flow skewing in turbulent shear layers. Using direct

numerical simulations (DNS), the current study is aimed at investigating the qualitative and

quantitative effects of three-dimensionality in incompressible shear layers. Further, a range of “what if?”

numerical experiments will be performed to isolate the effects of skew in mean flow and

turbulent fluctuations.   

Revisiting Taylor's Hypothesis in Homogeneous Turbulent Shear Flow

Taylor’s Hypothesis of frozen flow has frequently been used to convert temporal experimental measurements into a spatial domain and its validity is of

crucial importance for experimental studies and theoretical investigations. Results from direct numerical simulations are used here to study the

applicability of Taylor’s Hypothesis in homogeneous turbulent shear flow by considering the correlation of the Eulerian acceleration with the convective

acceleration (i.e., the nonlinear term). Using a wavelet-based scale decomposition of the accelerations, their correlations at different scales of motion

are investigated. This approach allows us to revisit Taylor's Hypothesis by examining the cancellation properties of Eulerian and convective accelerations

at different flow scales. The results show that Taylor's hypothesis holds at small scales of the flow as reflected by the anti-alignment of the Eulerian

acceleration and the convective term. Such anti-alignment, however, is not observed at the largest scales of the turbulent motion, indicating that

Taylor's hypothesis does not generally hold for homogeneous turbulent shear flow.   

Mixing in stratified shear turbulence: implications for the Saharan Air Layer

Every year during late spring and summer, the mean easterly winds above Northwestern Africa bring pockets of dust-laden Saharan air to the Americas, causing the Saharan Air Layer (SAL) to form over the tropical North Atlantic. However, gravitational settling of dust particles during their 5-day transatlantic trip in the SAL occurs much slower in reality than current models predict. An explanation for these reduced deposition rates which has been gaining increasing support from models and observations is the occurrence of vertical turbulent mixing within the SAL. Nevertheless, temperature measurements indicate that the SAL interior is usually weakly stratified, so buoyancy effects must be taken into account in order to accurately determine shear-induced turbulent mixing rates. We hypothesize that the initially well-mixed SAL becomes progressively stratified as it moves toward the Americas due to lateral heat fluxes, proportional to the SAL temperature anomaly and resulting from large-scale dynamics. The competition between stratification and turbulent mixing is quantified as the ratio of the corresponding timescales. The impact of this dimensionless parameter on the wind, temperature and dust fields at different Reynolds numbers is assessed via large-eddy simulation (LES) of a stratified shear layer. We aim to link parameters related to the SAL dynamics to mixing rates in stratified shear turbulence, so that turbulence effects on dust fields can be accurately parametrized in large-scale models.