Large-Eddy Simulation of Wake and Boundary Layer Interactions Behind a Circular Cylinder

Abstract

Large-eddy simulations (LESs) of flow past a circular cylinder in the vicinity of a flat plate have been carried out for three different gap-to-diameter (G/D) ratios of 0.25, 0.5, and 1.0 (where G signifies the gap between the flat plate and the cylinder, and D signifies the cylinder diameter) following the experiment of Price et al. (2002, “Flow Visualization Around a Circular Cylinder Near to a Plane Wall,” J. Fluids Struct., 16, pp. 175–191). The flow visualization along with turbulent statistics are presented for a Reynolds number of Re=1440 (based on D and the inlet free-stream velocity U∞). The three-dimensional time-dependent, incompressible Navier–Stokes equations are solved using a symmetry-preserving finite-difference scheme of second-order spatial and temporal accuracy. The immersed-boundary method is employed to impose the no-slip boundary condition at the cylinder surface. An attempt is made to understand the physics of flow involving interactions of shear layers shed from the cylinder and the wall boundary layer. Present LES reveals the shear layer instability and formation of small-scale eddies apart from their mutual interactions with the boundary layer. It has been observed that G/D ratio has a large influence on the modification of wake dynamics and evolution of the wall boundary layer. For a low gap ratio, it is difficult to identify the boundary layer because of its strong interactions with the shear layers; however, a rapid transition to turbulence of the boundary layer, which is similar to bypass transition, is observed for a large gap ratio.

Initial growth of a disturbance in a boundary layer influenced by a circular cylinder wake

“…The trajectory of the secondary vortex core, which is defined as the centroid of the λ ci concentration region with a threshold of 0.7, is nearly parallel to the wall. This differs from the G/D = 1.0 case of Sarkar & Sarkar (2009), where both the secondary vortex and the lower wake vortex gradually move away from the wall, suggesting that the direct lift-up induction from the wake vortices onto the underlying secondary vortex is negligible here. We note that the trajectory of the phase-averaged secondary vortex begins at about x/D = 3, that is, the secondary vortex appears there in a statistical sense.…”

“…Pan et al 2008;Sarkar & Sarkar 2009;Mandal & Dey 2011), secondary spanwise vortical structures are also observed here. The phase-averaged λ ci fields in the side-view plane during one vortex-shedding period are presented in figure 3.…”

“…This suggests that the boundary layer edge cannot be determined by the traditional 0.99U ∞ criterion. Similar to Sarkar & Sarkar (2009), the height of the local maximum of U between the wall and the wake deficit region is used to indicate the boundary layer edge. By this method, the boundary layer thickness δ at x/D = 2.5 is about 0.59D, whereas it increases to 1.15D at x/D = 15.…”

Evolution of Lagrangian coherent structures in a cylinder-wake disturbed flat plate boundary layer

“…Further downstream, the profile near the wall surpasses that of the Blasius profile, indicating a boundary layer transition. As no uniform outer flow can be observed due to the existence of the wake, we adopt a definition of boundary layer thickness used in the studies of Sarkar & Sarkar (2009) and He et al (2013). In this definition, the height of the maximum streamwise velocity between the wall and the wake deficit region is the boundary layer edge (as indicated by the horizontal dashed lines for the profiles at x/D = 5 and 25).…”

Experimental and CFD prediction of heat transfer and friction factor characteristics in cross flow tube bank with integral splitter plate

“…For the bare cylinder arrangement, the higher magnitude of the vortex formation takes place. The plots for the streamline contours for bare cylinder is similar in terms of separation point, and length of the bubble formation to that of the unbounded flow for flow over single cylinder in case of LES simulation at Re = 3900 by -Breuer [34], and Sarkar [35]. In the case of splitter plate attachment smaller size of vortex with nearly symmetric recirculation length and flow pattern takes place on either side of the splitter plate.…”