A study of boundary layer–wake interaction in shallow open channel flows

This paper describes an experimental investigation of the interaction between the boundary layer on a horizontal floor of a shallow open channel flow and the wake of a thin flat plate mounted vertically on the floor of the channel. The nominal thickness of the flat plate was limited to 2 mm in order to minimize the effect of the flume side walls on the generated wake. Two flat plate chord to thickness ratios (10 and 25) and two depths of flow (50 and 80 mm) were considered. The boundary layer thickness of the approaching flow was comparable with the depth of flow. The recovery of the boundary layer is then studied by observing the characteristics of the velocity profile downstream of the flat plate along the wake axis. The results indicate that the recovery process is slow, and that it is clearly non-monotonic. When compared with the approaching flow, the streamwise turbulence intensity values increase in the near-wake region followed by a gradual but slow decrease with increasing axial distance. Neither mean nor higher-order moments indicate a complete recovery even at large distances from the wake generator. The present results also indicate that the inner region appears to develop more quickly than the outer flow. Since the development of the quasi-two-dimensional wake is also of interest, velocity measurements are also presented along the wake cross-section. These velocity profiles indicate that the wake effects are still prevalent at 200 plate widths downstream of the wake generator. Through a proper choice of scaling variables, the mean velocity profiles across the wake can be collapsed onto a single curve, indicating a sense of similarity. Received: 23 September 1999/Accepted: 30 August 2000     

GROUND EFFECT OF FLOW AROUND AN ELLIPTIC CYLINDER IN A TURBULENT BOUNDARY LAYER

The flow characteristics around an elliptic cylinder with an axis ratio of AR=2 located near a flat plate were investigated experimentally. The elliptic cylinder was embedded in a turbulent boundary layer whose thickness is larger than the cylinder height. For comparison, the same experiment was carried out for a circular cylinder having the same vertical height. The Reynolds number based on the height of the cylinder cross-section was 14000. The pressure distributions on the cylinder surface and on the flat plate were measured for various gap distances between the cylinder and the plate. The wake velocity profiles behind the cylinder were measured using hot-wire anemometry. In the near-wake region, the vortices are shed regularly only when the gap ratio is greater than the critical value of G/B=0·4. The critical gap ratio is larger than that of a circular cylinder. The variation of surface pressure distributions on the elliptic cylinder with respect to the gap ratio is much smaller than that on the circular cylinder. This trend is more evident on the upper surface than the lower one. The surface pressures on the flat plate recover faster than those for the case of the circular cylinder at downstream locations. As the gap ratio increases, the drag coefficient of the cylinder itself increases, but the lift coefficient decreases. For all gap ratios tested in this study, the drag coefficient of the elliptic cylinder is about half that of the circular cylinder. The ground effect of the cylinder at small gap ratio constrains the flow passing through the gap, and restricts the vortex shedding from the cylinder, especially in the lower side of the cylinder wake. This constraint effect is more severe for the elliptic cylinder, compared to the circular cylinder. The wake region behind the elliptic cylinder is relatively small and the velocity profiles tend to approach rapidly to those of a flat plate boundary layer     

Turbulent heat transfer in a flat plate boundary layer disturbed by a cylinder

Augmentation of heat transfer from a flat plate using a turbulence promoter has been studied. A circular cylinder 8 mm in diameter was placed in the turbulent boundary layer detached from the flat plate. It was located parallel to the plate and perpendicular to the flow direction. Clearance, c, between the cylinder and the flat plate was varied in nine steps: c=0, 1, 2, 3, 4, 6, 11, 20 and 29.5 mm. Measurements were made of the local heat transfer coefficients, mean velocity profiles, turbulence intensity profiles, static pressure and skin friction. Experimental results showed that the heat transfer deterioration which occurs just downstream of the cylinder at c=0 mm can be removed by displacing the cylinder a small distance from the wall. The improvement in heat transfer is mainly due to the unsteadiness of the recirculating flow on the plate and the effect of intense turbulence arriving at the near wall region from the lower shear layer of the cylinder wake. Heat transfer augmentation is most effective when c=4 mm and becomes less effective when c is increased more than 6 mm. The enhancement disappears far downstream from the cylinder.     

A boundary layer analysis of combined heat and mass transfer by natural convection from a concentrated source in a saturated porous medium

A boundary layer analysis was carried out to investigate the coupled phenomena of heat and mass transfer by natural convection from concentrated heat and mass sources embedded in saturated porous media. Both line and point source problems were treated. The boundary layer equations based on Darcy's law and Boussinesq approximation were solved by means of similarity transformation to obtain the details of velocity, temperature and concentration distributions above a concentrated heat source. Two important parameters, namely the Lewis number Le and the buoyancy ratioN were identified to conduct a series of numerical integrations. For the case of small Le, a substance diffuses further away from the plume centerline, such that the mass transfer influences both velocity and temperature profiles over a wide range. For large Le, on the other hand, the substance diffuses within a narrow range along the centerline. Naturally, the influence of mass transfer is limited to the level of the centerline velocity, so that a peaky velocity profile appears for positiveN whereas a velocity defect emerges along the centerline for negativeN. For such cases of large Le, the temperature profiles are found to be fairly insensitive to Le.     

Absolute and convective instabilities of two-dimensional bluff body wakes in ground effect

Local and global instabilities are investigated of wakes of general two-dimensional bluff bodies placed near and parallel to a plane boundary or ground. A spatio-temporal linear stability analysis is first applied to a four-parameter family of local wake profiles to investigate the fundamental local stability characteristics of the wake in ground effect. The analysis shows significant dependencies of the stability characteristics of the wake on the distance from the wake centreline to the ground (normalised by the wake width), and also on the velocity ratio of the near- and far-ground sides of the wake. The analysis is then compared with earlier experiments on a circular cylinder to examine, according to the transition scenario of the steep global modes, the streamwise variation of the local stability characteristics of the wake in ground effect. The comparison indicates that the near wake region of the cylinder changes from being absolutely unstable to being convectively unstable when the cylinder comes down into the near-ground range in which the von Kármán-type vortex shedding from the cylinder is suppressed, being qualitatively consistent with the transition scenario for general wake-type flows. A possible explanation is also given for the counter-intuitive relation between the thickness of the boundary layer on the ground and the critical gap distance for the cessation of the von Kármán-type vortex shedding in ground effect.     

Theoretical analysis of laminar film condensation in a rotating cylinder with a scraper

The rimming film condensation on the inside wall of a rotating cylinder with a scraper is analyzed. The whole cylinder is divided into two regions, one is the so-called boundary layer region where the radial velocity of the condensate is much smaller than the peripheric velocity so that the boundary layer theory is assumed to be valid; the other is the scraper region where because of the disturbance of the scraper the boundary layer theory does not apply. The boundary layer integral method in the boundary layer region coupling with the integral momentum theorem across the scraper region provides a method to determine the velocity, temperature, and film thickness distributions, and heat transfer coefficients. An extensive discussion about the previous models is given. The sublayer flow rate constancy principle and the variability principle of the boundary layer thickness (therefore the interface velocity) at the scraper position with respect to the rotational speed are proposed. The present model greatly improved the prediction of the average heat transfer coefficient. Received on 5 January 1998     

Flow and heat transfer over a backward-facing step with a cylinder mounted near its top corner

A study was made to see if it is possible to enhance the heat transfer in the downstream region of a backward-facing step, where heat transfer is normally deteriorated, by the insertion of a cylinder near the top corner of the step. Cylinder size and streamwise position of the cylinder were kept constant but the cross-stream position of the cylinder was changed in three steps. Results of the heat transfer experiment, flow visualization, and measurement of the averaged and fluctuating flow fields were reported. When the cylinder was mounted at a position, a little higher than the top surface of the step, a jet-like flow pattern emerged in the averaged velocity profile beneath the cylinder and the recirculating flow was intensified. Therefore, the velocity of recirculating flow near the wall is increased at some streamwise positions. Additionally, the velocity fluctuation was intensified not only in the shear layer between the jet-like flow and the recirculating flow regions but also in the near wall region, resulting in the effective augmentation of heat transfer in this case. Therefore, it is concluded that the mounting of a cylinder is effective in the enhancement of deteriorated heat transfer in the recirculating flow region, if its is mounted in a proper position.     

A bypass wake induced laminar/turbulent transition

The process of laminar to turbulent transition induced by a von Karman vortex street wake, was studied for the case of a flat plate boundary layer. The boundary layer developed under zero pressure gradient conditions. The vortex street was generated by a cylinder positioned in the free stream. An X-type hot-wire probe located in the boundary layer, measured the streamwise and normal to the wall velocity components. The measurements covered two areas; the region of transition onset and development and the region where the wake and the boundary layer merged producing a turbulent flow. The evolution of Reynolds stresses and rms-values of velocity fluctuations along the transition region are presented and discussed. From the profiles of the Reynolds stress and the mean velocity profile, a ‘negative' energy production region along the transition region, was identified. A quadrant splitting analysis was applied to the instantaneous Reynolds stress signals. The contributions of the elementary coherent structures to the total Reynolds stress were evaluated, for several x-positions of the near wall region. Distinct regions in the streamwise and normal to the wall directions were identified during the transition.     

Stabilization of cylinder wakes in shallow water flows by means of roughness elements: an experimental study

Shallow wakes that occur in a wide range of natural flows tend to generate instabilities that develop into large, 2D coherent structures (2DCS). We present the results of an experimental study to stabilize shallow wakes by local, enhanced bottom roughness. Two successful stabilization strategies are compared to a base case of an unsteady bubble wake. First, localized bed roughness is placed in the lateral shear layers near the shoulders of the cylinder. Second, a local roughness element is placed at the end of the recirculation bubble, in the downstream region where large-scale vortices would normally shed. Dye visualization is used to assess the qualitative behavior of the wake, and two-component laser Doppler velocimetry (LDV) measurements are made to measure the Reynolds stress distributions and time-averaged velocity profiles. In both stabilization cases, a minimum patch size of the enhanced roughness elements is required for stabilization, which depends on the momentum thickness of the shear layers and the locations of enhanced Reynolds shear stresses. The main effect of the wake stabilization is a reduction in momentum exchange with the ambient flow due to damping of the large 2DCS. This reduction in eddy diffusivity results in a narrower wake and a slower decay of the centerline velocity deficit with downstream distance compared to the base case of an unsteady bubble wake.     

Active Control of a Circular Cylinder Flow at Transitional Reynolds Numbers

Active and passive control of flow around a circular cylinder, at transitional Reynolds numbers was investigated experimentally by measuring cylinder surface pressures and wake velocity profiles. Two- and three-dimensional passive boundary layer tripping was considered and periodic active control using piezo-fluidic actuators was introduced from a two-dimensional slot that was nearly tangential to the cylinder surface. The slot location was varied circumferentially by rotating the cylinder and this facilitated either upstream- or downstream-directed actuation using sinusoidal or modulated wave-forms. Separation was controlled by two distinct methods, namely: by forcing laminar-turbulent transition when applied at relatively small angles (30–60°) from the forward stagnation point; and by directly forcing the separated shear-layer at larger angles. In the latter case, actuation produced the largest load changes when it was introduced at approximately 90° from the forward stagnation point. When the forcing frequency was close to the natural vortex-shedding frequency, the two frequencies “locked-in” creating clear and persistent structures. These were examined and categorized. The “lock-in” effect lowered the base pressure and increased the form-drag whereas delaying separation from the cylinder did the opposite.     

Flow of a particulate suspension in the wake of a circular cylinder

Measurements are reported for the average local particulate velocity and concentration distributions in the wake of a cylinder immersed in a stream containing a polydisperse aerosol. The wake centerline defects and transverse distributions were determined for both parameters. It was found that the particulate centerline defect persists a considerable distance downstream of the cylinder before fully developed conditions are satisfied. Transverse particulates and gaseous velocity distributions assume a Gaussian profile at the same point downstream. The charge-to-mass ratio throughout the wake region was equal to the free stream value for all experimental conditions and was of such a magnitude to permit the electrostatic effects to be neglected in the governing equations. Gaseous and particulate transport properties were identified in the wake.     

Effect of velocity ratio on the streamwise vortex structures in the wake of a stack

The time-averaged velocity and streamwise vorticity fields within the wake of a stack were investigated in a low-speed wind tunnel using a seven-hole pressure probe. The experiments were conducted at a Reynolds number, based on the stack external diameter, of Re_D=2.3×10⁴. The stack, of aspect ratio AR=9, was mounted normal to a ground plane and was partially immersed in a flat-plate turbulent boundary layer, where the ratio of the boundary layer thickness to the stack height was δ/H≈0.5. The jet-to-cross-flow velocity ratio was varied from R=0 to 3, which covered the downwash, crosswind-dominated and jet-dominated flow regimes. In the downwash and crosswind-dominated flow regimes, two pairs of counter-rotating streamwise vortex structures were identified within the stack wake. The tip vortex pair located close to the free end of the stack, and the base vortex pair located close to the ground plane within the flat-plate boundary layer, were similar to those found in the wake of a finite circular cylinder, and were associated with the upwash and downwash flow fields within the stack wake, respectively. In the jet-dominated flow regime, a third pair of streamwise vortex structures was observed, referred to as the jet-wake vortex pair, which occurred within the jet-wake region above the free end of the stack. The jet-wake vortex pair had the same orientation as the base vortex pair and was associated with the jet rise. The peak vorticity and strength of the streamwise vortex structures were functions of the jet-to-cross-flow velocity ratio. For the tip vortex structures, their peak vorticity and strength reduced as the jet-to-cross-flow velocity ratio increased.     

Investigation of a compressible turbulent boundary layer and a region of separation at M=5 by a laser Doppler anemometer

Results are cited of an experimental investigation of the structure of a compressible turbulent boundary layer on a thermally insulated cylinder placed longitudinally in the flow. The experiments were conducted at M=5 and Re_x10⁷. In order to establish a longitudinal positive pressure gradient and a region of separation at the end of the cylinder, a tailpiece in the shape of an axisymmetric isentropic compression surface, or conical flaps with various half angles, were mounted. Profiles of the longitudinal velocity component were measured using p_o' and t_o probes, and also using a laser Doppler anemometer (LDA) with a Fabry-Perot interferometer. In the absence of a longitudinal pressure gradient the velocity profiles measured by the different methods were in satisfactory agreement among themselves and with the results of calculations. In the presence of a longitudinal positive pressure gradient, the velocity profiles become less filled and the static pressure, calculated according to the results of measurements of the velocity with the aid of the LDA and the pressure p_o', varied across the thickness of the boundary layer. In the separated region, the recirculating velocity of the flow was measured with the aid of the LDA.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 2, pp. 175–178, March–April, 1976.     

Momentum and heat transport in a finite-length cylinder wake

This paper reports an experimental study of turbulent momentum and heat transport in the wake of a wall-mounted finite-length square cylinder, with its length-to-width ratio L/d = 3–7. The cylinder was slightly heated so that heat produced could be considered as a passive scalar. A moveable three-wire probe (a combination of an X-wire and a cold wire) was used to measure velocity and temperature fluctuations at a Reynolds number of 7,300 based on d and the free-stream velocity. Measurements were performed at 10 and 20d downstream of the cylinder at various spanwise locations. Results indicate that L/d has a pronounced effect on Reynolds stresses, temperature variance and heat fluxes. The downwash flow from the free end of the cylinder acts to suppress spanwise vortices and, along with the upwash flow from the cylinder base, makes the finite-length cylinder wake highly three-dimensional. Reynolds stresses, especially the lateral normal stress, are significantly reduced as a result of suppressed spanwise vortices at a small L/d. The downwash flow acts to separate the two rows of spanwise vortices further apart from the wake centerline, resulting in a twin-peak distribution in temperature variance. While the downwash flow entrains high-speed fluid into the wake, responsible for a small deficit in the time-averaged streamwise velocity near the free end, it does not alter appreciably the distribution of time-averaged temperature. It has been found that the downwash flow gives rise to a counter-gradient transport of momentum about the central region of the wake near the free end of the cylinder, though such a counter-gradient transport does not occur for heat transport.     

Fluid flow and heat transfer around circular cylinder – flat and curved plates combinations

Experimental studies were carried out to investigate the fluid flow and heat transfer around a heated circular cylinder which was placed at various distances of a wall boundary with different geometries (flat or curved plate) with subcritical Reynolds number ranging from 3.5×10³ to 10⁴. The effects of plate geometry (aspect ratio: W|H=1.0,1.5 and 2.0, and rim angle, φ=0°,60°,90°, and 120°) and gap ratio, (G|D=0.0,0.86,2.0,7.0,10.0) on the fluid flow and heat transfer characteristics (static pressure around cylinder surface, wake width, base pressure, pressure drag coefficients, velocity distribution, and both local and mean Nusselt numbers) were presented. Also flow visualization was carried out to illustrate the flow patterns around the cylinder at various gap ratios (G|D). It was found that the heat transfer and fluid flow characteristics are dependent on the plate geometry at all tested gap ratios, except for G|D=7.0 and 10.0, they are independent of the plate geometry.     

Wake structure of a finite circular cylinder of small aspect ratio

The wake of a finite circular cylinder of small aspect ratio was studied with a seven-hole probe and thermal anemometry. The cylinder was mounted normal to a ground plane and was partially immersed in a turbulent boundary layer. The time-averaged velocity and streamwise vorticity fields showed the development of the tip vortex structures, the extent of the near-wake recirculation zone, the downwash phenomenon and base vortex structures within the boundary layer. The wake structure and power spectra were similar for cylinder aspect ratios of 5 to 9, but a distinctly different behaviour was observed for an aspect ratio of 3.     

Application of a Virtual-Boundary Method for the Numerical Study of Oscillations Developing Behind a Cylinder Near A Plane Wall

The conditions of onset and the character of the oscillations developing behind a circular cylinder located above a plane wall (screen) in a flow with a velocity profile of the boundary layer type are studied numerically. The dependence of the critical Reynolds number (at which a steady flow regime in the wake behind the cylinder is replaced by an oscillatory regime) on the cylinder-wall gap and the free-stream boundary layer thickness is found.     

Forced Convection Heat Transfer from a Finite-Height Cylinder

This paper presents a large eddy simulation of forced convection heat transfer in the flow around a surface-mounted finite-height circular cylinder. The study was carried out for a cylinder with height-to-diameter ratio of 2.5, a Reynolds number based on the cylinder diameter of 44 000 and a Prandtl number of 1. Only the surface of the cylinder is heated while the bottom wall and the inflow are kept at a lower fixed temperature. The approach flow boundary layer had a thickness of about 10% of the cylinder height. Local and averaged heat transfer coefficients are presented. The heat transfer coefficient is strongly affected by the free-end of the cylinder. As a result of the flow over the top being downwashed behind the cylinder, a vortex-shedding process does not occur in the upper part, leading to a lower value of the local heat transfer coefficient in that region. In the lower region, vortex-shedding takes place leading to higher values of the local heat transfer coefficient. The circumferentially averaged heat transfer coefficient is 20 % higher near the ground than near the top of the cylinder. The spreading and dilution of the mean temperature field in the wake of the cylinder are also discussed.     

Heat transfer on a cylinder in accelerated slip flow

The axisymmetric laminar boundary layer flow along the entire length of a semi-infinite stationary cylinder under an accelerated free-stream is investigated. Considering flow at reduced dimensions, the boundary layer equations are developed with the conventional no-slip boundary condition for tangential velocity and temperature replaced by a linear slip-jump boundary condition. Asymptotic series solutions are obtained for the heat transfer coefficient in terms of the Nusselt number. These solutions correspond to prescribed values of the momentum and temperature slip coefficients and the index of acceleration. Heat transfer at both small and large axial distances is determined in the form of series solutions; whereas at intermediate distances, exact and interpolated numerical solutions are obtained. Using these results, the heat transfer along the entire cylinder wall is evaluated in terms of the parameters of acceleration and slip.     

Measurement in a Zero-Pressure Gradient Turbulent Boundary Layer with Forced Thermal Convection

A subsonic zero-pressure gradient turbulent boundary layer developing on a uniformly heated surface at a Reynolds number in the range of 3, 560?≤?Re _θ ?≤ 5,360 was investigated. Particle-image velocimetry measurements were performed at various positions in the streamwise direction for several wind-tunnel speeds and for different wall excess temperatures to show the thermal convection effects to expand the boundary-layer thickness δ _0.99 and to enlarge the turbulence intensities in the log-law and wake region. The mean velocity profiles are found to be self-preserving. The inclination of large-scale ramp-like vortex packets increases to higher characteristic angles, i.e., the mean angles are enlarged by approximately 5–10°. Hairpin-like vortex structures originating from the near-wall region seem to undergo higher climbing rates in the wall-normal direction causing the above mentioned significant changes in the boundary-layer thickness δ _0.99 and the strongly increased distributions of turbulence intensities in the wake region of the boundary layer. Changes in the distributions of the skewness and flatness of the probability density function (PDF) of the streamwise fluctuations corroborate these findings. The two-point correlation distribution of the streamwise velocity fluctuations R _uu is increased for wall distances y/δ _0.99?=?0.1 to y/δ _0.99?=?0.75 indicating the existence of coherent structures in higher regions of the boundary layer.