Previous session | Next session

Session Cg - Near Surface Turbulence.
ORAL session, Sunday, November 23
307, Moscone Center

[Cg.01] Interaction of turbulent shear flows with a free surface

Patrice Maheo, Morteza Gharib (Graduate Aeronautical Laboratories, California Institute of Technology)

In this study, we aim to quantify the differences between turbulent free-surface shear flows and their fully-submerged counterparts. Turbulent wakes and shear layers generated by a sharp two-dimensional surface-piercing splitter plate were investigated experimentally. Digital particle image velocimetry (DPIV) was used to obtain instantaneous two-dimensional velocity measurements in image planes throughout these flows. Large data sets of flow images were acquired to obtain well-converged Reynolds-averaged quantities and to monitor the flow dynamics. For the submerged cases, these measurements show good agreement with published data for two-dimensional shear flows. In the absence of mean streamwise vortices, fully-submerged shear flows have two-dimensional characteristics in the mean. However, near the surface, two weak counter-rotating streamwise vortices, whose presence was unveiled by time averaging of flow-normal cross-sections, widen the surface flow to a depth comparable to their size and contribute to a velocity and vorticity field that is highly three-dimensional in the mean.

[Cg.02] The study of Large Eddy Simulation models using Digital Particle Image Velocimetry

Bradley Dooley, Morteza Gharib (Graduate Aeronautical Laboratories, California Institute of Technology)

Results are presented of an experimental investigation of turbulent bluff body wakes in the presence of a free surface. These data are then applied to several models used in Large Eddy Simulation (LES) for evaluation purposes.

Measurements have been performed using digital particle image velocimetry (DPIV) to obtain the in-plane velocity fields in free-surface-parallel and freestream-normal planes at several depths, ranging from very near-surface to far into the bulk flow. Two thousand image pairs were processed at each measurement location, allowing us to extract both mean and fluctuating (rms) values for in-plane velocity components and thus surface-normal vorticity. To obtain sufficient spatial resolution, measurements at each depth were taken using several overlapping camera views, which can be combined into a mosaic of the entire working section.

As our data gives spatial resolution down to the order of the Taylor microscale, we utilize it for a posteriori studies of LES techniques. By coarsely filtering the experimental data and then attempting to recover the smaller scales through modeling, the apparent realism of various subgrid scale models is studied.

[Cg.03] Grid Turbulence Interaction with a Free Surface

Haitao Huang (Graduate Aeronautical Laboratories of California Institute of Technology)

The goal of the present study is to understand the interaction between turbulence and free-surface. The digital particle image velocimetry (DPIV) is used in the experiment as it is able to provide velocity field information. Turbulence is generated with a regular square-grid with a wire diameter d_w = 0.2 mm and with a mesh distance of d_m = 12.7 mm. Reynolds number based on the mesh distance is Re_m = 3200 and 2300. Preliminary result shows that the turbulence intensity in this free-surface layer is much stronger than that in the bulk. The thickness of this free-surface layer reminds almost constant within an investigation range of x/d_m = 20. Turbulence interaction with the free-surface with and without the presence of the free-surface layer is the main focus of the present study. With the presence of the free-surface layer, the interaction of the turbulence with the free-surface is strong, resulting in a quick decay of the turbulence in the bulk. Without the presence of the free-surface layer, the interaction of the turbulence with the free-surface becomes weaker, showing a slower decay of the turbulence in the bulk.

[Cg.04] Experimental Study of the Wake Behind a Surface-Piercing Cylinder for a Clean and Contaminated Free Surface

Amy Warncke Lang, Morteza Gharib (Caltech)

This experimental investigation into the nature of flows at a free surface, is to study the

effects of surfactants on the wake of a surface-piercing cylinder. The purpose is to better

understand the process of vorticity generation and conversion at a free surface due to the

absence or presence of surfactants, which can result in shear stresses at the free surface

thus altering the boundary condition. The wake behind a surface piercing cylinder is

studied using Digital particle Image Velocimetry (DPIV) to map the velocity and vorticity

field for three orthogonal cross- sections of the flow. In addition, shadowgraph and a new

surface mapping technique were used to visualize and measure the free surface. It was

found that, depending on the surface condition, the connection of the shedding vortex

filaments at the free surface was greatly altered with the propensity for surface tension

gradients to redirect the vorticity near the free surface to that of the surface-parallel

component.

[Cg.05] On The Correlation Of The Free-Surface Roughness With The Near-Surface Turbulence

Dana Dabiri (GALCIT)

Free-surface turbulence is understood through the free-surface deformation or roughness. By decomposition of surface velocity, pressure and surface deformation into roughness and scattering components, the subsurface flow of the roughness problem behaves similar to the flow near a free-slip wall (Dommermuth, 1993). For low Froude number

cases, the free-surface elevation is hydrostatically balanced with the

components of the pressure that is induced by the near-surface vortical or turbulent flow field.

Since roughness is proportional to pressure, the spectrum of roughness should behave as S(\kappa) \propto \epsilon^4/3F\!r^4\kappa^-7/3, where \epsilon is the turbulent dissipation rate, \kappa is the wave number, and \kappa^-7/3 is related to the pressure in the inertial range (Monin amp; Yaglom, 1975). This therefore provides a theoretical basis for the prediction of the pressure/velocity correlation from the measurement of the free-surface roughness. We have developed a real-time free-surface elevation mapping technique, capable of providing a global mapping of the free-surface deformation. This technique, combined with DPIV, provides a powerful tool to study free-surface turbulence by allowing measurements of \eta ' u', \eta ' v ',and dw/dz. We will discuss the trends and results from grid-generated turbulence. Sponsored by Office of Naval Research (N00014-92-J-1618).

abstract.

[Cg.06] Turbulent Shear Flow near A Free Surface

Lian Shen, Xiang Zhang, Dick K.P. Yue (MIT), George S. Triantafyllou (Levich Institute)

Direct numerical simulation is used to examine free-surface turbulence (FST) over a two-dimensional mean shear. It is found that the shear flow FST is characterized by coherent vortex structures which are hairpin shaped. As a hairpin structure approaches the free surface, the ``head'' is dissipated quickly in the surface layer; while the two ``legs'' are connected perpendicularly to the free surface and decay at a laminar rate. The coherence and persistence of connected normal vorticity is the most prominent surface signature of FST. Analysis of Reynolds-stress evolution shows the significance of inter-component energy transfer in the shear flow FST. In the underlying bulk flow, turbulent kinetic energy is extracted from the mean shear flow to the streamwise velocity component first, then transfered from the latter to the span-wise and vertical velocity components. Near the free surface, a considerable amount of energy is transfered from the vertical velocity component to the horizontal ones. The difference between FST and its conventional approximation, free-slip-wall turbulence, is elucidated by comparing the inter-component energy transfer during the splat and anti-splat processes at the free surface. We find a considerable reduction of the pressure-strain correlation at the free surface relative to that at a free-slip wall even for very small Froude numbers. This is explained in terms of the different behaviors of the pressure field at the surface.

[Cg.07] On the Generation of Vorticity at a Free Surface

Thomas Lundgren (University of Minnesota), Petros Koumoutsakos (CTR, NASA/Ames; IFD, ETH, Zurich Switzerland)

An algorithm for the generation of vorticity at a viscous free surface is described. The scheme is a free surface analogue of Lighthill's strategy for determining the vorticity flux at solid boundaries. In this method the zero shear stress and pressure boundary conditions are transformed into a boundary integral formulation suitable for the velocity-vorticity description of the flow. Vorticity is generated at free surfaces when there is flow past regions of surface curvature. It is shown that vorticity is conserved in a vortex dynamics formulation of free surface viscous flows. Vorticity which flows from the fluid is gained by a vortex sheet along the surface; the integral of vorticity over the entire fluid region plus the integral of ``surface vorticity'' over the free surface remains constant. The implications of the present algorithm as a tool for numerical calculations are discussed.

[Cg.08] Free Surface Velocity and Elevation Measurement Techniques as = Applied to the Interaction of a Vortex Pair with a Contaiminated = Interface.

J. P. Mortzheim, J. D. Gayton, A. Hirsa (Rensselaer Poly. Inst.)

A technique derived from digital particle image velocimetry (DPIV) is developed to measure the velocity field along boundary-fitted coordinates for free surface flows with non-zero Froude number. This method is especially useful in high Reynolds number flows at contaminated air/water interfaces where large velocity gradients are present. In the present technique, the location of the free surface is first determined by cross-correlating a pair of adjacent interrogation windows (atop of each other with one inverted) which are marched down the image until auto-correlation is achieved. At this point, the two respective windows appear as one image due to the total internal reflection at the free surface. This procedure is continued until the surface elevation profile is obtained throughout the field. Velocity in the near surface region is subsequently determined using DPIV performed along the boundary-fitted coordinates, as compared to the conventional orthonormal grid. This method was applied to the interaction of a vortex pair with a surfactant-influenced interface, with Re =3D 23,000 and Fr =3D 0.32. Maximum surface elevations of order 1 mm were measured with a resolution of order 10 microns. The application of this technique for obtaining the various velocity derivatives is discussed, including the direct measurement of surface dilatational viscosity.

[Cg.09] Nonlinear Wave Radiation and Diffraction by a Near-Surface Body,

P. Ananthakrishnan (Florida Atlantic University)

Physics of surface-wave and rigid-body interactions is of importance in naval architecture, in that a good understanding of wave-body interactions is necessary for the design of hulls with minimum ship-motion and resistance characteristics. Particular topics of contemporary research such as generation of spray and breaking waves by a surface ship and control of ship motion in high seas are however highly nonlinear, rendering analysis a challenging task. Using a robust numerical algorithm developed for analyzing fully nonlinear free-surface flow in a viscous fluid (see P. Ananthakrishnan, Three-dimensional wave-body interactions in a viscous fluid, Proc.~of ISOPE'97 Conference, Hawaii), we have investigated diffraction and radiation of waves by floating and submerged rigid bodies. In the numerical model, the Navier-Stokes equations subject to exact free-surface and body boundary conditions are solved in primitive variables using a fractional-step finite-difference method which is implemented using curvilinear coordinates. Approximate conditions are however used to model the open boundary and the movement of the contact line. Results presented shed light to a better understanding of generation and ensuing spatial-temporal evolution of vortices under the influence of a free surface, vortical and potential components of hydrodynamics forces, symmetry-breaking in the case of large-amplitude oscillations, generation and damping of super-harmonic waves, and parameter ranges in which effect of viscosity is significant.

[Cg.10] Thermal Signature of Turbulence at a Free Surface

R. A. Handler, J. R. Saylor, R. I. Leighton (Naval Research Laboratory)

Numerical simulations of turbulence in an open channel have been performed at a Prandtl number of 2 and Reynolds number based on friction velocity of 180. Both constant flux and constant temperature boundary conditions have been imposed at the free surface. Correlations of the surface temperature and surface heat flux with the velocity and vorticity fields beneath the surface, along with a topological analysis of the surface flow, were completed. The results show a clear signature which appears to be associated with hairpin vortex structures impacting the surface. In agreement with experiments, these structures appear to be the principal surface renewal agents for these flows.

Part C of program listing