[massimo.marro]

Hello,

I'm trying to simulate the velocity field for a heated cavity with buoyantSimpleFoam and a k-epsilon turbulence model. I'm using OpenFOAM-2.0.1.
I copied the characteristics of the solver from the directory "tutorial/heatTransfer/buoyantSimpleFoma/hotroom/system".

At the inlet a turbulent boundary layer is imposed, with fixed profiles of velocity, epsilon and k.
At the outlet zeroGradient for velocity, k, epsilon. On the walls, the boundary conditions are no slip condition for the velocity and compressible::wallFunction for k and epsilon. At the top fixed values of velocity, k and epsilon are imposed.
The boundary condition for the pressure is buoyantPressure everywhere.

If the inlet values of k and epsilon are low, pressure instabilities rise at the inlet and the simulation crashes.

Thanks for your help,
Massimo

[al_pr]

Hello Massimo,

can you post your boundary condition files for k, epsilon and pressure? I think this would help us to find an answer. Could you also post your fvSchemes and fvSolution file?

I found some threads of this topic in this forum. Many try to increase simply epsilon to make the simulation more stable.


Regards,
Alex

[massimo.marro]

Hello Alex,

thank you for your answer.
I have posted the boundary condition files for k, epsilon and p_rgh. The inlet condition for k is an experimental profile of a boundary layer.
The boundary condition for p are $internalField.

I have also posted the fvSchemes and fvSolution files.

If the inlet and top values of epsilon are "too" low, large pressure instabilities rise at the inlet.

Regards,
Massimo

epsilon.gz

k.gz

p_rgh.gz

fvSchemes.gz

fvSolution.gz