[umer.chaudrey]

Dear Foamers,

I am a new user of OpenFOAM. I have been trying to observe transient flow behaviour through a geometry which is like a large cube (very thin) represents air at atmospheric pressure in a kind of thin chamber. There is a very small square cross section pipe in the center, which is given a time varying boundary condition for pressure. From atmospheric to about 3000Pa the pressure drops within 3 seconds linearly.

I have a few queries.

1 - the relation of Courant number has a velocity variable in it. Under this kind of Boundary condition, when the initial velocity is zero. How do I assume my MaxCo.

2 - With adjustable timestep and a max Co of 0.5, my simulation at a normal quality of mesh is taking a lot of time. Similarly how do I assume values for epsilon & k when I do not know velocity. The relations given in the OpenFOAM documents have velocity.

3 - I did obtain converged solutions, but when I increased the quality of my mesh manually by having more number of cells. The pimple solution did not converge for the same values, but a lesser value of deltat.

4 - My values of pressure are fluctuating big time with manual change in mesh refinement. Infact two similar mesh produced different results with everything same, but with the only difference in deltat.

I am using discretizer for generating my grid (it is a simple geometry) and using rhoPimpleFoam solver, with RAS model (k-epsilon).

Can anyone please lock at these queries and advise me. I will really appreciae the help. I can send my files and/or some screenshots if requried. Thanks

Umer

[umer.chaudrey]

Okay, atleast can anyone guide me how to assume k and epsilon values for my specific case when there is no velocity boundary condition. for MaxCo, as I saw in other threads I believe for accuracy it should be 0.5.

Can anyone please guide me for k and epsilon? Thank you, I will appreciate the guidance.

Umer