CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Boundary Conditions for External Flow (https://www.cfd-online.com/Forums/openfoam-solving/58735-boundary-conditions-external-flow.html)

 raytracer July 14, 2008 07:23

I am solving for a flow past a

I am solving for a flow past a cylinder in external flow in 2D using the icoFoam solver. It consists of a circle in a rectangular domain.

I'm using the following boundary conditions:
1. Left side of the rectangle is the inlet with fixedValue velocity and zeroGradient pressure.
2. Right side of the rectangle is the outlet with fixedValue pressure and zeroGradient velocity.
3. fixedValue for velocity and zeroGradient pressure at the surface of the cylinder.

I have the following doubts:
1. Are the above conditions I've used correct?
2. I don't know what conditions to use for the top and bottom sides of the rectangle. Should I treat them as outlets as well?
3. What would be the difference between specifying the walls of the cylinder as 'wall' or 'patch'? Does the difference come into play only in turbulence simulations?

Thanks for the help.

 pbo July 14, 2008 08:06

1/correct 2/if inlet velocity

1/correct
2/if inlet velocity is parallel to bottom and top planes, you can use the 'slip' BC on these (however make sure bottom and top planes are far enough from the cylinder for this BC to be physically correct for external flow, otherwise you will spuriously accelerate the fluid over the cylinder).
3/It shouldn't make any difference for a laminar solver whether you use wall or patch in the boundary file

 raytracer July 14, 2008 08:41

Hey thanks a lot, Patrick.

Hey thanks a lot, Patrick.

And also, Would 'slip' be a better condition to use than zeroGradient for both pressure and velocity? If so, why - does it not unnecessarily remove the normal component?

 pbo July 14, 2008 10:13

I don't know what is better.

I don't know what is better.
The slip BC may stabilize the computation more than the other, try both settings, and see what happens.
All in all, if both BCs work fine for your case, you should obtain very similar solutions (for top and bottom planes far enough from the cylinder).

 raytracer July 14, 2008 12:18

Oh okay. Thank you very much!

Oh okay. Thank you very much!

 All times are GMT -4. The time now is 15:58.