D. Sumner

Flow-pattern identification for two staggered circular cylinders in cross-flow

The flow around two circular cylinders of equal diameter, arranged in a staggered configuration, was investigated using flow visualization and particle image velocimetry for centre-to-centre pitch ratio P / D = 1[ratio ]0 to 5.0 and angle of incidence. α = 0° to 90°. Experiments were conducted within the low subcritical Reynolds number regime, from Re = 850 to 1900. Nine flow patterns were identified, and processes of shear layer reattachment, induced separation, vortex pairing and synchronization, and vortex impingement, were observed. New insight was gained into previously published Strouhal number data, by considering the flow patterns involved. The study revealed that vortex shedding frequencies are more properly associated with individual shear layers than with individual cylinders; more specifically, the two shear layers from the downstream cylinder often shed vortices at different frequencies.

Fluid forces on kayak paddle blades of different design

Three kayak paddle blades of different design (Conventional, Norwegian, Turbo) were tested in a low-speed wind tunnel at a maximum chord Reynolds number of Re = 2.2–2.7 × 105 (corresponding to speed through water of ≈1 m/s). The mean drag force and side force acting on each blade were measured, as the yaw and pitch angles were varied. The results were compared with those recorded for a finite rectangular flat plate of similar area and aspect ratio. For zero pitch angle of the blades, the results indicate that the drag coefficient was mostly independent of the blade design as the yaw angle was varied between ± 20°, with only the Norwegian blade design displaying a marginally higher drag coefficient than either of the other two blades or the flat plate. Increasing the pitch angle to 30°, while maintaining the yaw angle at zero, resulted in a 23% reduction of the drag coefficient for the flat plate, but only a 15% reduction of the drag coefficients for the three blades. For all designs, the drag coefficient reduction followed a simple cosine relationship as the pitch angle or yaw angle was increased. The wind tunnel experiments revealed that the side force coefficients for all three paddle blade designs were entirely independent of the blade design and were indistinguishable from those recorded for a flat plate. In summary, the study showed that the nondimensional force coefficients are largely independent of the paddle blade design.

Some vortex-shedding characteristics of the staggered configuration of circular cylinders

A pair of circular cylinders of equal diameter arranged in the staggered configuration was tested in the subcritical Reynolds number regime at centre-to-centre pitch ratios of P/D=2.0 and 2.5. For each pitch ratio, the incidence angle was varied in small steps in the range α=0–90°. Vortex-shedding frequency measurements were made behind the upstream and downstream cylinders as the incidence angle, α, was varied. At small incidence angles, 2°<α<15°, the Strouhal number data were unreliable, since the peaks in the power spectra were either absent or broad banded. In this same range of incidence angle, the downstream cylinder experiences a maximum inward-directed lift force and a minimum drag force. The results indicate that the flow mechanisms responsible for the changes in the aerodynamic forces have an important influence on the vortex-shedding activity also.

The effect of surface roughness on the turbulence structure of a plane wall jet

In this paper, an experimental investigation of the turbulence characteristics of a plane wall jet over smooth and rough surfaces, using laser Doppler anemometry (LDA), is reported. The Reynolds number based on the slot height and exit velocity of the jet was approximately Re = 7500. A 36-grit sheet was used as the rough surface, creating a transitionally rough flow regime (44<ks+<70). Both inner and outer scales were used to analyze the effects of surface roughness on the Reynolds stress profiles. Comparisons between the present results and other LDA and hot-wire anemometry studies for a smooth surface indicate a similar behavior for the Reynolds stress profiles. However, the magnitudes of the peak values of the Reynolds stress were higher than in most previous studies due to the lower slot Reynolds number. The present results indicate that surface roughness does not appear to significantly modify the Reynolds stress profiles in the outer region of the jet except for a reduction in the level. In contrast...

An Experimental Investigation of Aspect Ratio and Incidence Angle Effects for the Flow Around Surface-Mounted Finite-Height Square Prisms

The flow around a surface-mounted finite-height square prism was investigated using a low-speed wind tunnel. The experiments were conducted at a Reynolds number of Re = 7.3 × 104 for prism aspect ratios of AR = 3, 5, 7, 9, and 11 and incidence angles from α = 0 deg to 45 deg. The thickness of the boundary layer on the ground plane relative to the side length was δ/D = 1.5. Measurements of the vortex shedding frequency were made with a single-component hot-wire probe, and measurements of the mean drag and lift forces were obtained with a force balance. For all aspect ratios and incidence angles, the mean drag coefficient and Strouhal number were lower than those of an infinite prism, while the mean lift coefficient was of nearly similar magnitude. As the aspect ratio was increased from AR = 3 to 11, the force coefficients and Strouhal number slowly approached the infinite-square-prism data. The mean drag coefficient and Strouhal number for the finite prism were less sensitive to changes in incidence angle compared to the infinite square prism. The critical incidence angle, corresponding to minimum mean drag coefficient, minimum (most negative) mean lift coefficient, and maximum Strouhal number, shifted to a higher incidence angle compared to the infinite square prism, with values ranging from αcritical = 15 deg to 18 deg; this shift was greatest for the prisms of higher aspect ratio. The behavior of the force coefficients and Strouhal number for the prism of AR = 3 was distinct from the other prisms (with lower values of mean drag coefficient and mean lift coefficient magnitude, and a different Strouhal number trend), suggesting the critical aspect ratio was between AR = 5 and AR = 3 in these experiments. In the wall-normal direction, the power spectra for AR = 11 and 9 tended to have weaker and/or more broad-banded vortex shedding peaks near the ground plane and near the free end at α = 0 deg and 15 deg. For AR = 7 to 3, well-defined vortex shedding peaks were detected along the entire height of the prisms. For AR = 11 and 9, at α = 30 deg and 45 deg, vortex shedding peaks were absent in the power spectra in the upper part of the wake.

Closely Spaced Circular Cylinders in Cross-Flow and a Universal Wake Number

To investigate the effectiveness of a universal wake number for groups of closely spaced bluff bodes, staggered cylinder configurations with center-to-center pitch ratios of P/D =1.125 and 1.25, and incidence angles from α=0 deg-90 deg, were tested in the subcritical Reynolds number regime. The aerodynamic forces, base pressure, and vortex shedding frequencies were measured for the upstream and downstream cylinders, and were found to be strongly dependent on the incidence angle and small changes in the flow pattern. The Griffin number was found to be an appropriate universal wake number for the closely spaced staggered cylinders, based on the total drag force acting on the two cylinders, and the average base pressure for the two cylinders

A Visualization Study of Fluid-Structure Interaction between a Circular Cylinder and a Channel Bed

Flow visualization was used to study the fluid-structure interaction between a circular cylinder and a shallow turbulent open channel flow. The Reynolds number ranged from ReD= 1500–4400 based on the cylinder diameter, and from ReH=7,800−27,600 based on the channel hydraulic radius. The cylinder was mounted vertically on the channel bed and the flow depth-to-cylinder-diameter ratio was varied fromd/D=7.0−11.7. Tests were carried out over smooth and rough beds, with the rough beds being either permeable or impermeable. The study showed that the horseshoe vortex forming at the cylinder-bed junction affects many of the flow structures, including the mode of vortex shedding, the shear layer dynamics, the vortex formation length, and the width of the near-wake region. The influence of the horseshoe vortex can be recognized throughout the depth of flow; however, its influence decreases with an increase in distance from the channel bed. It was also possible to discern that the bed roughness resulted in a change to the above interaction and the permeability of the bed resulted in additional changes.

Incomplete similarity of a plane turbulent wall jet on smooth and transitionally rough surfaces

This study assesses the hypothesis of incomplete similarity for a plane turbulent wall jet on smooth and transitionally rough surfaces. Typically, a wall jet is considered to consist of two regions: an inner layer and an outer layer. The degree to which these two regions reach equilibrium with each other and interact to produce the property of self-similarity remains an open question. In this study, the analysis of the outer and inner regions indicates that each region is characterised by a half-width which exhibits its own distinct dependence on the streamwise distance x from the slot, and a single self-similar structure for both regions does not exist. More specifically, the inner and outer layers of the wall jet exhibit different scaling laws, which results in two self-similar mean velocity profiles, both of which retain a dependence on the slot height H. As such, incomplete similarity of the wall jet on smooth and transitionally rough surfaces is confirmed by this study. In addition, comparison of the experimental results for the transitionally rough surface with the smooth wall case indicates that the surface roughness modifies the development of the mean velocity profile in both the inner and outer regions, although the effect on the outer region is relatively small and close to the experimental uncertainty.

Numerical Simulation of the Flow Around a Surface-Mounted Finite Square Cylinder

Large eddy simulation (LES) was used to study the unsteady turbulent flow over a surface-mounted square cylinder of finite height, for cylinder aspect ratios of AR = 3 and 5, at a Reynolds number of Re = 500. Employing a semi-implicit fractional step method for momentum and a pressure correction scheme for mass conservation, a collocated finite-volume code was used to solve the Navier-Stokes equations in three dimensions. A multi-grid scheme was employed to accelerate the pressure solver. The cylinder geometry was created on a Cartesian mesh using internal boundary conditions. The motivation for the study was to better understand the complex wake structure of the finite cylinder and the influence of the aspect ratio. The numerical results were consistent with the observations from experiments documented in the literature. For the case of AR = 3, the wake was dominated by downwash effects, where as for more slender cylinder, AR = 5, some upwash effects were noticeable closer to the ground plane.Copyright

Testing three-dimensional bluff-body models in a low-speed two-dimensional adaptive wall test section

Thin, sharp-edged disk models were evaluated in a low-speed two-dimensional adaptive flexible wall test section to determine the optimum adaptive wall testing environment for three-dimensional bluff-body models, by providing model testing recommendations for nominal solid blockage ratio and model span ratio. Drag coefficient measurements obtained under straight wall and adapted wall conditions showed that for a two-dimensional adaptive wall test section, the model span ratio imposes a more severe restriction upon model size than does the nominal solid blockage ratio. Minimum wall interference conditions were achieved with adapted walls for nominal solid blockage ratios less than 3 percent and model span ratios less than 21 percent, independent of the nominal test section aspect ratio, based on favorable comparison with previously-published experimental data. Data obtained under straight wall conditions confirmed that wall interference effects can only be neglected in conventional, straight-walled test sections for solid blockage ratios less than 0.5 percent and model span ratios less than 10 percent.