Periodicity momentum instead of axial momentum
 

Hi,

I'm new in the community and I've got a qustion about the use of periodic boundary conditions.
I'm trying to simulate a combustion chamber. Since the chamber is rotationally symetric I use just a piece of 4 degrees of the chamber together with periodic BCs. The mesh has 3-4 cell in the tangantial direction.
I'm solving the problem for steady state. Also the problem neither is influenced by the mesh quality (I have done a mesh study already) nor to a certain location in the fluid domain.
Also, by using symetric BCs the simulation is stable.
I suspect the periodic BCs to make the simulation instable.

Problem:

The solution starts to get a angular momentum. Beginning from the outlet it propagats along the axis into the combustion chamber.
The solution then starts to get Mach-numbers of the order of 5-10.

If someone had/has the same problems and/or has a good idea for the solution, please let me know.

Thanks

omez

Hi,

In the real world, what is used to stop the flow rotating? You should put the devices in the model to stop it rotating. If you know the flow does not rotate then use symmetry boundaries and forget about the problems with periodic boundaries. Alternately you could put a directional source term in to damp the rotational component - but I would only do this after a thorough check of your simulation to see what is creating the rotation in the first place (ie, two wrongs don't make a right).

Glenn Horrocks