Tomographic-PIV measurement of the flow around a zigzag boundary layer trip

Tomographic-PIV was used to measure the boundary layer transition forced by a zigzag trip. The resulting instantaneous three-dimensional velocity distributions are used to quantitatively visualize the flow structures. They reveal undulating spanwise vortices directly behind the trip, which break up into individual arches and then develop into the hairpin-like structures typical of wall-bounded turbulence. Compared to the instantaneous flow structure, the structure of the average velocity field is very different showing streamwise vortices. Such streamwise vortices are often associated with the low-speed streaks occurring in bypass transition flows, but in this case clearly are an artifact of the averaging. Rather, the present streaks in the separated flow region directly behind the trip are resulting from the waviness in the spanwise vortices as introduced by the zigzag trip. Furthermore, these streaks and the separated flow region are observed to be related to a large-scale, spanwise uniform unsteadiness in the flow that contributes significantly to the velocity fluctuations over large downstream distances (up to at least the edge of the present measurement domain).     

Control of vortex formation from a vertical cylinder in shallow water: Effect of localized roughness elements

Vortex formation from a vertical cylinder in shallow water is controlled by placement of a narrow transverse strip of roughness elements on the bed (bottom surface). A technique of high-image-density particle image velocimetry is employed to obtain global, instantaneous representations of the flow patterns, which lead to phase- and time-averaged patterns of streamline topology and Reynolds stress on planes at and above the bed. Near the bed, the overall form of the streamline topology is maintained, even at larger heights of the roughness elements. With increasing height of the elements, the downstream saddle point is further displaced in the streamwise direction. Correspondingly, the streamwise extent of the negative pocket of the streamwise velocity component, i.e., the region of reverse flow along the bed surface, increases substantially in the streamwise direction. The Reynolds stress in the very near-wake, at locations upstream of the roughness elements, is significantly attenuated, even for small height of roughness. This attenuation occurs not only near the bed surface, but also at the midplane of the shallow water wake, and thereby indicates that the consequence of localized roughness is to exert a global influence. In fact, corresponding patterns of instantaneous velocity and vorticity indicate that consistent formation of large-scale vortices in the very near-wake region is attenuated with relatively small surface roughness on the bed. Downstream of the roughness elements, the patterns of Reynolds stress near the bed surface, as well as at the midplane of the water layer, are significantly altered relative to the case of no roughness. Near the bed, highly concentrated patterns of positive and negative Reynolds stress in the absence of roughness give way to lower-level regions of Reynolds stress in the form of alternating concentrations; the particular pattern depends on the height of the roughness elements. At the midplane of the water layer, the Reynolds stress patterns maintain their same overall form, but the extrema of the Reynolds stress concentrations are attenuated in magnitude and are shifted in the downstream direction, with increasing height of the roughness elements. These observations are complemented by patterns of instantaneous velocity and vorticity.     

Large Eddy Simulations and Experiments of Flow Around Finite-Height Cylinders

The paper presents experimental and numerical results for the flow around a surface-mounted circular cylinder at the two height-to-diameter ratios of 2.5 and 5. The Reynolds number based on approach flow velocity and cylinder diameter is 43,000 and 22,000 for these two cases and the boundary layer of the approach flow has a thickness of about 10% of the cylinder height. The experiments comprise both flow visualizations with dye and laser Doppler velocimeter measurements of all mean velocity and fluctuation components. The numerical study is performed by an elaborate large eddy simulation on a staggered Cartesian grid using the immersed boundary method. The instantaneous flow behaviour including the shedding is analysed with information based on animations. For the long cylinder alternating shedding is found to occur over most of the height while for the shorter cylinder the shedding is observed mainly near the ground where it is also mostly alternating but intermittently also symmetrical. The mean-flow behaviour is analysed with the aid of streamlines and contour plots of mean-velocity and fluctuation components in various planes and a detailed comparison of LES and LDV results is provided, showing generally good agreement. The LES with very fine resolution near the free end allow a detailed study of the complex flow in this region with owl-face topology on the end wall previously observed in experiments. Behind the cylinder, the longitudinal recirculation region, the downstream development of tip vortices and the emergence of trailing vortices further downstream are analysed. The sum of the results, together with those from previous studies that were reviewed extensively, provides a comprehensive picture of the very complex flow behaviour.     

On streamwise vortices in Large Eddy Simulations of initially laminar plane mixing layers

This paper details the influence of the nature of imposed inflow fluctuations on Large Eddy Simulations of a spatially developing turbulent mixing layer originating from laminar boundary layers. A simulation with imposed white-noise random fluctuations, commonly used in numerical simulations, produces mean-flow statistics that agree well with reference experimental data. Whilst flow visualisation images show evidence for streamwise vorticity in this simulation, quantitative statistics do not reveal the presence of statistically stationary streamwise vortices. A further simulation that uses physically-correlated inflow fluctuations also produces good mean-flow statistical agreement with reference data. From secondary shear stress contours it can be inferred that this simulation does, however, predict the presence of statistically stationary streamwise vortices. The properties of the streamwise vortices are in good agreement with experimental data. The data presented here indicate that, even for initially laminar conditions, plane mixing layer simulations require accurate physically correlated inflow conditions in order to reproduce the flow features found experimentally.     

Time resolved flow-field measurements of a turbulent mixing layer over a rectangular cavity

High Reynolds number, low Mach number, turbulent shear flow past a rectangular, shallow cavity has been experimentally investigated with the use of dual-camera cinematographic particle image velocimetry (CPIV). The CPIV had a 3 kHz sampling rate, which was sufficient to monitor the time evolution of large-scale vortices as they formed, evolved downstream and impinged on the downstream cavity wall. The time-averaged flow properties (velocity and vorticity fields, streamwise velocity profiles and momentum and vorticity thickness) were in agreement with previous cavity flow studies under similar operating conditions. The time-resolved results show that the separated shear layer quickly rolled-up and formed eddies immediately downstream of the separation point. The vortices convect downstream at approximately half the free-stream speed. Vorticity strength intermittency as the structures approach the downstream edge suggests an increase in the three-dimensionality of the flow. Time-resolved correlations reveal that the in-plane coherence of the vortices decays within 2–3 structure diameters, and quasi-periodic flow features are present with a vortex passage frequency of ~1 kHz. The power spectra of the vertical velocity fluctuations within the shear layer revealed a peak at a non-dimensional frequency corresponding to that predicted using linear, inviscid instability theory.     

Manipulation of the reverse-flow region downstream of a fence by spanwise vortices

The experimental investigation of a turbulent separated flow over a fence is presented. By introducing a periodic disturbance upstream of the separation region in front of the fence, the time averaged length of the separation region downstream of the fence was reduced by as much as 40%. Two types of flow manipulation were applied: an oscillating cross-flow with zero net mass-flux through a spanwise slot in the floor of the test section and a spanwise oriented, oscillating spoiler. The cross-flow was generated by a loudspeaker system connected to a chamber underneath the spanwise slot. Both types of flow manipulation generate spanwise vortices at the fence that convect into the region downstream of the fence where they enhance the mixing in the shear layer and reduce the time mean length of the reverse-flow region downstream of the fence. Velocity profiles phase averaged with respect to the forcing frequency and phase triggered flow visualisations show that the spanwise vortices cause the long reverse-flow region of the unmanipulated flow to break up into separate smaller regions. While the time mean length of the reverse-flow region is reduced in the manipulated case, the length of the region where instantaneous reverse-flow occurs is not changed. The data presented include wall pulsed-wire measurements of the wall shear-stress and its turbulent fluctuations, and LDA measurements of the streamwise and the wall-normal velocity components and turbulent stresses.     

DNS of heat transfer in transitional, accelerated boundary layer flow over a flat plate affected by free-stream fluctuations

Direct numerical simulations (DNS) of flow over and heat transfer from a flat plate affected by free-stream fluctuations were performed. A contoured upper wall was employed to generate a favourable streamwise pressure gradient along a large portion of the flat plate. The free-stream fluctuations originated from a separate LES of isotropic turbulence in a box. In the laminar portions of the accelerating boundary layer flow the formation of streaks was observed to induce an increase in heat transfer by the exchange of hot fluid near the surface of the plate and cold fluid from the free-stream. In the regions where the streamwise pressure gradient was only mildly favourable, intermittent turbulent spots were detected which relaminarised downstream as the streamwise pressure gradient became stronger. The relaminarisation of the turbulent spots was reflected by a slight decrease in the friction coefficient, which converged to its laminar value in the region where the streamwise pressure gradient was strongest.     

Experimental investigation of the unsteady structure of a transitional plane wall jet

 A laminar wall jet undergoing transition is investigated using the particle image velocimetry (PIV) technique. The plane wall jet is issued from a rectangular channel, with the jet-exit velocity profile being parabolic. The Reynolds number, based on the exit mean velocity and the channel width, is 1450. To aid the understanding of the global flow features, laser-sheet/smoke flow visualizations are performed along streamwise, spanwise, and cross-stream directions. Surface pressure measurements are made to correlate the instantaneous vorticity distribution with the surface pressure fluctuations. The instantaneous velocity and vorticity field measurements provide the basis for understanding the formation of the inner-region vortex and the subsequent interactions between the outer-region (free-shear-layer region) and inner-region (boundary-layer region) vortical structures. Results show that under the influence of the free-shear-layer vortex, the local boundary layer becomes detached from the surface and inviscidly unstable, and a vortex is formed in the inner region. Once this vortex has formed, the free-shear-layer vortex and the inner-region vortex form a vortex couple and convect downstream. The mutual interactions between these inner- and outer-region vortical structures dominate the transition process. Farther downstream, the emergence of the three-dimensional structure in the free shear layer initiates complete breakdown of the flow. Received: 8 November 1995/Accepted: 6 November 1996     

Large eddy simulation of gas-particle turbulent channel flow with momentum exchange between the phases

This paper presents results of a large eddy simulation (LES) combined with Lagrangian particle tracking and a point-force approximation for the feedback effect of particles on the downward turbulent gaseous flow in a vertical channel. The LES predictions are compared with the results obtained by direct numerical simulation (DNS) of a finer computational mesh. A parametric study is conducted for particles with two response times in simulations with and without streamwise gravitational settling and elastic, binary interparticle collisions. It is shown that the classical and the dynamic Smagorinsky turbulence models adequately predict the particle-induced changes in the mean streamwise velocity and the Reynolds stresses of the carrier phase for the range of parameters studied. However, the largest discrepancies between the LES and DNS results are found in the cases of particle-laden flows. Conditional sampling of the instantaneous resolved flow fields indicates that the mechanisms by which particles directly oppose the production of momentum and vorticity of the organized fluid motions are also observed in the LES results. However, the geometric features of the near-wall quasistreamwise vortices are overestimated by the use of both turbulence models compared to the DNS predictions.     

Evolution of vortex structure in boundary layer transition induced by roughness elements

The particle image velocimetry (PIV) and hydrogen-bubble visualization technique are used to investigate the flat-plate boundary layer transition induced by an array of roughness elements. The streamwise evolutions of the mean and fluctuation velocity are analyzed, and the critical Reynolds number Re _k,c is determined between 339 and 443 under current experimental setup. The hairpin vortices shed from supercritical roughness elements are visualized by swirling strength, in which two pairs of counter-rotating vortices can be observed: one vortex pair is the manifestation of the neck of the hairpin vortices shed from the top of the roughness; the other vortex pair, which originates from the lower part of the roughness, comes from the streamwise vortices that are constantly perturbed by the hairpin vortex shedding.     

Numerical studies on evolution of secondary streamwise vortices in transitional boundary layers

Formation and evolution of secondary streamwise vortices in the compressible transitional boundary layers over a flat plate are studied using a direct numerical simulation method with high-order accuracy and highly effective non-reflecting characteristic boundary conditions. Generation and development processes of the secondary streamwise vortices in the complicated transitional boundary flow are clearly analyzed based on the of numerical results, and the effects on the formation of the ring-like vortex that is vital to the boundary layer transition are explored. A new mechanism forming the ring-like vortex through the mutual effect of the primary and secondary streamwise vortices is expressed.     

Stability of the laminar boundary layer flow encountering a row of roughness elements: Biglobal stability approach and DNS

The stability of the laminar boundary layer developing on a flat plate in the presence of a periodic row of roughness elements is investigated. A Direct Numerical Simulation is performed to compute the steady flow downstream of the roughness elements, which contains a pair of two counter-rotating streamwise vortices per element, which can be considered as a “pre-streaky” structure. The linear stability of this base flow is analyzed by means of the so-called “biglobal” stability approach. Three-dimensional eigenmodes are found, which are shown to be the continuation of the Tollmien–Schlichting waves present in the case of an unperturbed boundary layer. Moreover, a stabilizing effect due to the roughness-induced vortices is found. A Direct Numerical Simulation of the interaction between a two-dimensional Tollmien–Schlichting wave and the roughness array is also performed. The computed perturbation traveling downstream of the roughness elements is shown to be a linear combination of the biglobal eigenmodes.     

Using digital holographic microscopy for simultaneous measurements of 3D near wall velocity and wall shear stress in a turbulent boundary layer

A digital holographic microscope is used to simultaneously measure the instantaneous 3D flow structure in the inner part of a turbulent boundary layer over a smooth wall, and the spatial distribution of wall shear stresses. The measurements are performed in a fully developed turbulent channel flow within square duct, at a moderately high Reynolds number. The sample volume size is 90 × 145 × 90 wall units, and the spatial resolution of the measurements is 3–8 wall units in streamwise and spanwise directions and one wall unit in the wall-normal direction. The paper describes the data acquisition and analysis procedures, including the particle tracking method and associated method for matching of particle pairs. The uncertainty in velocity is estimated to be better than 1 mm/s, less than 0.05% of the free stream velocity, by comparing the statistics of the normalized velocity divergence to divergence obtained by randomly adding an error of 1 mm/s to the data. Spatial distributions of wall shear stresses are approximated with the least square fit of velocity measurements in the viscous sublayer. Mean flow profiles and statistics of velocity fluctuations agree very well with expectations. Joint probability density distributions of instantaneous spanwise and streamwise wall shear stresses demonstrate the significance of near-wall coherent structures. The near wall 3D flow structures are classified into three groups, the first containing a pair of counter-rotating, quasi streamwise vortices and high streak-like shear stresses; the second group is characterized by multiple streamwise vortices and little variations in wall stress; and the third group has no buffer layer structures.     

The effect of streamwise vortices on the turbulence structure of a separating boundary layer

A high Reynolds number flat plate turbulent boundary layer is investigated in a wind-tunnel experiment. The flow is subjected to an adverse pressure gradient which is strong enough to generate a weak separation bubble. This experimental study attempts to shed some new light on separation control by means of streamwise vortices with emphasize on the change in the boundary layer turbulence structure. In the present case, counter-rotating and initially non-equidistant streamwise vortices become and remain equidistant and confined within the boundary layer, contradictory to the prediction by inviscid theory. The viscous diffusion cause the vortices to grow, the swirling velocity component to decrease and the boundary layer to develop towards a two-dimensional state. At the position of the eliminated separation bubble the following changes in the turbulence structure were observed. The anisotropy state in the near-wall region is unchanged, which indicates that it is determined by the presence of the wall rather than the large scale vortices. However, the turbulence in the outer part of the boundary layer becomes overall more isotropic due to an increased wall-normal mixing and a significantly decreased production of streamwise fluctuations. The turbulent kinetic energy is decreased as a consequence of the latter. Despite the complete change in mean flow, the spatial turbulence structure and the anisotropy state, the process of transfer of turbulent kinetic energy to the spanwise fluctuating component seems to be unchanged. Local regions of anisotropy are strongly connected to maxima in the turbulent production. For example, at spanwise positions in between those of symmetry, the spanwise gradient of the streamwise velocity cause significant production of turbulent fluctuations. Transport of turbulence in the spanwise direction occurs in the same direction as the rotation of the vortices.     

Temporal development of turbulent boundary layers with embedded streamwise vortices

The interaction of streamwise vortices with turbulent boundary layer has been investigated using large-eddy simulation. The initial conditions are a pair of counterrotating Oseen vortices with flow between them directed toward the wall (common-flow-down), superimposed on various instantaneous realizations of a turbulent boundary layer. The time development of the vortices and their interaction with the boundary layer are studied by integrating the filtered Navier-Stokes equations in time. The most important effects of the vortices on the boundary layer are the thinning of the boundary layer between vortices (downwash region) and the thickening of the boundary layer in the upwash region. The vortices first move toward the wall as a result of the self-induced velocity, and then apart from each other because of the image vortices due to the solid wall. The Reynolds stress profiles highlight the highly three-dimensional structure of the turbulent boundary layer modified by the vortices. The presence of significant turbulent activity near the vortex center and in the upwash region suggests that localized instability mechanisms in addition to the convection of turbulent energy by the secondary flow are responsible for this effect. High levels of turbulent kinetic energy and secondary stresses in the vicinity of the vortex center are also observed. The numerical results show good agreement with experimental results.This work was supported by the Office of Naval Research under Grant N00014-89-J-1638. Computer time was supplied by the San Diego Supercomputing Center.     

Distributed Roughness Effects on Transitional and Turbulent Boundary Layers

A numerical investigation is carried out to study the transition of a subsonic boundary layer on a flat plate with roughness elements distributed over the entire surface. Post-transition, the effect of surface roughness on a spatially developing turbulent boundary layer (TBL) is explored. In the transitional regime, the onset of flow transition predicted by the current simulations is in agreement with the experimentally based correlations proposed in the literature. Transition mechanisms are shown to change significantly with the increasing roughness height. Roughness elements that are inside the boundary layer create an elevated shear layer and alternating high and low speed streaks near the wall. Secondary sinuous instabilities on the streaks destabilize the shear layer promoting transition to turbulence. For the roughness topology considered, it is observed that the instability wavelengths are governed by the streamwise and spanwise spacing between the roughness elements. In contrast, the roughness elements that are higher than the boundary layer create turbulent wakes in their lee. The scale of instability is much shorter and transition occurs due to the shedding from the obstacles. Post-transition, in the spatially developing TBL, the velocity defect profiles for both the smooth and rough walls collapsed when non dimensionalized in the outer units. However, when compared to the smooth wall, deviation in the Reynolds stresses are observable in the outer layer; the deviation being higher for the larger roughness elements.     

Bypass transition receptivity modes

The prediction of bypass transition remains an important problem in many engineering applications. This is largely because there is no suitable theoretical model for bypass transition and predictions are made using empirical models. This paper presents numerical results for the receptivity of a zero pressure gradient boundary layer subjected to simple freestream waveforms which are the constituent parts of a turbulent flow field. Significant receptivities are only obtained for a minority of freestream waveforms and these lead to two types of flow structure in the boundary layer. The first type of flow structure is essentially two dimensional in nature and consists of two rows of counter-rotating spanwise vortices and is induced by freestream waves of large normal and spanwise wavelength and streamwise wavelengths approximately equal to the boundary layer thickness. The second type of flow structure are the streamwise streaks frequently observed in flow visualisation experiments. These streaks are induced by freestream waves of long streamwise and normal wavelength and spanwise wavelengths in the range of 14.5-46 θ (1.7-5.4δ). The freestream waves can be formed of velocity components in any direction, however the boundary layer is most receptive to fluctuations that lie in a plane perpendicular to the streamwise direction. The overall receptivity to a full spectrum of waves typical of freestream turbulence is considered and is shown to have similar characteristics to those from experiments.     

Large Eddy Simulations in Turbines: Influence of Roughness and Free-Stream Turbulence

The individual and coupled effects of the incoming free-stream turbulence (FST) and surface roughness on the transition of a separated shear layer over a flat plate is numerically investigated using Large eddy simulation (LES). The upper wall of the test section is inviscid and specifically contoured to impose a streamwise pressure distribution over the flat plate to simulate the suction surface of a low pressure turbine (LPT) blade. The interaction of the streamwise streaks due to FST and roughness with the separated shear layer is captured. The streaks induced by FST are intermittent in nature and the streaks due to roughness are steady for a given topology of the rough surface. Both FST and roughness promoted near-wall mixing. The ‘net effect’ of mixing in the pre-separated region is manifested by a shift the inflection point of the velocity profile towards the wall. This resulted in the upstream shift of the transition point and a significant reduction in the size of the separation bubble. The combined effect of FST and roughness further reduced the size of separation bubble. The streamwise evolution of the boundary layer parameters has been compared for different cases. The potential ‘roughness benefit’ obtained in the case of highly loaded turbine blades in terms of its considerable reduction of profile loss is also shown.     

Numerical and experimental investigation of turbulent characteristics in a drag-reducing flow with surfactant additives

Cetyltrimethyl ammonium chloride (CTAC) surfactant additives, because of their long-life characteristics, can be used as promising drag-reducers in district heating and cooling systems. In the present study we performed both numerical and experimental tests for a 75 ppm CTAC surfactant drag-reducing channel flow. A two-component PIV system was used to measure the instantaneous streamwise and wall-normal velocity components. A Giesekus constitutive equation was adopted to model the extra stress due to the surfactant additives, with the constitutive parameters being determined by well-fitting apparent shear viscosities, as measured by an Advanced Rheometric Expansion System (ARES) rheometer. In the numerical study, we connected the realistic rheological properties with the drag-reduction rate. This is different from previous numerical studies in which the model parameters were set artificially. By performing consistent comparisons between numerical and experimental results, we have obtained an insight into the mechanism of the additive-induced drag-reduction phenomena.

Our simulation showed that the addition of surfactant additives introduces several changes in turbulent flow characteristics: (1) In the viscous sublayer, the mean velocity gradient becomes gentler due to the viscoelastic forces introduced by the additives. The buffer layer becomes expanded and the slope of the velocity profile in the logarithmic layer increases. (2) The locations where the streamwise velocity fluctuation and Reynolds shear stress attain their maximum value shifted from the wall region to the bulk flow region. (3) The root-mean-square velocity fluctuations in the wall-normal direction decrease for the drag-reducing flow. (4) The Reynolds shear stress decreases dramatically and the deficit of the Reynolds shear stress is mainly compensated by the viscoelastic shear stress. (5) The turbulent production becomes much smaller and its peak-value position moves toward the bulk flow region. All of these findings agree qualitatively with experimental measurements.

Regarding flow visualization, the violent streamwise vortices in the near wall region become dramatically suppressed, indicating that the additives weaken the ejection and sweeping motion, and thereby inhibit the generation of turbulence. The reduction in turbulence is accomplished by additive-introduced viscoelastic stress. Surfactant additives have dual effects on frictional drag: (1) introduce viscoelastic shear stress, which increases frictional drag; and (2) dampen the turbulent vortical structures, decrease the turbulent shear stress, and then decrease the frictional drag. Since the second effect is greater than the first one, drag-reduction occurs.     

The effect of boundary layer fluctuations on the streamwise vortex structure in simulated plane turbulent mixing layers

This paper details the influence of the magnitude of imposed inflow fluctuations on Large Eddy Simulations of a spatially developing turbulent mixing layer originating from laminar boundary layers. The fluctuations are physically-correlated, and produced by an inflow generation technique. The imposed high-speed side boundary layer fluctuation magnitude is varied from a low-level, up to a magnitude sufficiently high that the boundary layer can be considered, in a mean sense, as nominally laminar. Cross-plane flow visualisation shows that each simulation contains streamwise vortices in the laminar and turbulent regions of the mixing layer. Statistical analysis of the secondary shear stress reveals that mixing layers originating from boundary layers with low-level fluctuations contain a spatially stationary streamwise structure. Increasing the high-speed side boundary layer fluctuation magnitude leads to a weakening of this stationary streamwise structure, or its removal from the flow entirely. The mixing layer growth rate reduces with increasing initial fluctuation level. These findings are discussed in terms of the available experimental data on mixing layers, and recommendations for both future experimental and numerical research into the mixing layer are made.