TY - JOUR
T1 - Reliability of wall shear stress estimations of the flow around a wall-mounted cylinder
AU - Schanderl, Wolfgang
AU - Manhart, Michael
N1 - Publisher Copyright:
© 2016 Elsevier Ltd.
PY - 2016/4/10
Y1 - 2016/4/10
N2 - The flow field and wall shear stresses around a wall-mounted cylinder for a Reynolds number of ReD=39000, based on the diameter of the cylinder and the bulk velocity, was investigated applying Large-Eddy Simulation (LES). We used a Finite-Volume method on a Cartesian grid with staggered arrangement of the variables. The curved surface of the cylinder has been approximated by a conservative second order Immersed Boundary method.We carefully validate our simulation in terms of grid resolution, inflow condition and contribution of subgrid scale stresses. Around the cylinder, local grid refinement provides spacings of 7.5 wall units in horizontal and 1.9 wall unit in vertical direction based on the the wall shear stress of the oncoming flow. Thus, the contributions of the modeled subgrid scale stresses remain small compared to other stresses and the flow variables converge with grid resolution.We demonstrate that the flow structure and predicted wall shear stresses strongly depend on the inflow condition, which has to take into account turbulent fluctuations and the secondary flow, if present. A maximum amplification factor of the time-averaged wall shear stress of 12.0 with respect to the one in the oncoming flow field shows up in the lateral front of the cylinder. Instantaneous wall shear stresses however reach a maximum amplification factor of up to 40 in the wake of the cylinder.
AB - The flow field and wall shear stresses around a wall-mounted cylinder for a Reynolds number of ReD=39000, based on the diameter of the cylinder and the bulk velocity, was investigated applying Large-Eddy Simulation (LES). We used a Finite-Volume method on a Cartesian grid with staggered arrangement of the variables. The curved surface of the cylinder has been approximated by a conservative second order Immersed Boundary method.We carefully validate our simulation in terms of grid resolution, inflow condition and contribution of subgrid scale stresses. Around the cylinder, local grid refinement provides spacings of 7.5 wall units in horizontal and 1.9 wall unit in vertical direction based on the the wall shear stress of the oncoming flow. Thus, the contributions of the modeled subgrid scale stresses remain small compared to other stresses and the flow variables converge with grid resolution.We demonstrate that the flow structure and predicted wall shear stresses strongly depend on the inflow condition, which has to take into account turbulent fluctuations and the secondary flow, if present. A maximum amplification factor of the time-averaged wall shear stress of 12.0 with respect to the one in the oncoming flow field shows up in the lateral front of the cylinder. Instantaneous wall shear stresses however reach a maximum amplification factor of up to 40 in the wake of the cylinder.
KW - Boundary layer
KW - Horseshoe vortex
KW - Large-Eddy Simulation
KW - Wall shear stress
KW - Wall-mounted cylinder
UR - http://www.scopus.com/inward/record.url?scp=84957543138&partnerID=8YFLogxK
U2 - 10.1016/j.compfluid.2016.01.002
DO - 10.1016/j.compfluid.2016.01.002
M3 - Article
AN - SCOPUS:84957543138
SN - 0045-7930
VL - 128
SP - 16
EP - 29
JO - Computers and Fluids
JF - Computers and Fluids
ER -