CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   IDDES-SA boundary conditions (http://www.cfd-online.com/Forums/openfoam-solving/94472-iddes-sa-boundary-conditions.html)

 jantheron November 16, 2011 22:13

IDDES-SA boundary conditions

4 Attachment(s)
I am trying to run a simple case of flow over a square cylinder at Re=20,000 using Spalart's DES model implemented in OF2.0.1. Can anyone shed some light on the appropriate boundary conditions to use for k, nuTilda, U, omega and p? I have slip walls with one simple inlet and one outlet (kind of like a wind tunnel).

My grid was generated with gmsh and extruded in the Z direction. It consists of purely hexahedra with maximum orthogonality of 45. It passed all grid checks.

When I run with pisoFoam (keeping Co below 0.4), all the residuals decrease nicely but I can't seem to get rid of some unevenness in the velocity contours - these seem to come from the inlet.

At first I used Gauss linear interpolation but switched to TVD (limitedLinear) which got rid of extreme oscillations in the velocity contours. Yet, these oscilaltions emanating from the inlet corners persist in spite of a very low time step (5e-6 sec). My y+ values are max 14 and average 6.

The domain extends about 25 body diameters around the body, 18 diameters upstream and 40 diameters downstream. I am interested in the effect of shed vorticity on the freestream velocity in the +Y direction (about 20 diameters above the body)

Jan Theron

This is a snippet from my log file:

**********
Time = 0.0035

Courant Number mean: 0.00305846 max: 0.333807
DILUPBiCG: Solving for Ux, Initial residual = 0.0105967, Final residual = 3.03098e-12, No Iterations 2
DILUPBiCG: Solving for Uy, Initial residual = 0.012991, Final residual = 6.97586e-12, No Iterations 2
GAMGPCG: Solving for p, Initial residual = 0.280629, Final residual = 0.00207036, No Iterations 1
time step continuity errors : sum local = 5.72685e-09, global = -2.32239e-10, cumulative = 4.26148e-10
GAMGPCG: Solving for p, Initial residual = 0.0890806, Final residual = 0.00107882, No Iterations 1
time step continuity errors : sum local = 2.82245e-09, global = -3.78255e-11, cumulative = 3.88323e-10
GAMGPCG: Solving for p, Initial residual = 0.0790894, Final residual = 3.05308e-09, No Iterations 8
time step continuity errors : sum local = 7.97931e-15, global = 2.15325e-16, cumulative = 3.88323e-10
DILUPBiCG: Solving for nuTilda, Initial residual = 1.90357e-06, Final residual = 5.79013e-12, No Iterations 1
ExecutionTime = 4890.48 s ClockTime = 5531 s

forces output:
forces(pressure, viscous)((0.00524665 0.00263487 1.15591e-19) (0.000392818 2.52438e-06 -2.2827e-23))
moment(pressure, viscous)((-1.31744e-05 2.62332e-05 -5.77975e-06) (-1.26219e-08 1.96409e-06 1.62711e-08))
local:
forces(pressure, viscous)((0.00524665 0.00263487 1.15591e-19) (0.000392818 2.52438e-06 -2.2827e-23))
moment(pressure, viscous)((-1.31744e-05 2.62332e-05 -5.77975e-06) (-1.26219e-08 1.96409e-06 1.62711e-08))

forceCoeffs output:
Cd = 1.43864
Cl = 0.672805
Cm = -0.0367569

**********

My fvSolution and fvScheme files have been attached, as well as a screenshot of the velocity contours (flow is coming in from the left at 4mps) and one showing a line plot of the velocity across the body and the yplus values.

 All times are GMT -4. The time now is 14:35.