[Rotidpor]

Hello everyone.


I'm very new to openfoam and I'm trying to use the buoyantSimpleFoam application. My first cases ran very well and I obtained reasonable results.


I'm currently trying to use again cfmesh and buoyantSimpleFoam on another geometry but I have some troubles doing so.


After a few iterations, my epsilon starts to blow up where there are sharp edges (see screenshot). Around sharp edges, the mesh (especially the BLs) becomes badly distorted and I think that is why epsilon blows up towards ridiculous values.


So my questions are :
- Is there a way to avoid BL distortion around sharp edges ? I used the surfaceFeatureEdges utility for my mesh.
- Can I do something with buoyantSimpleFoam to stop epsilon from blowing up ?


My first cases contained sharp edges and lots of 90° angles but everything went well. I used relaxation, first order discretization schemes and speed ramp up for the begining of the calculation.



Any input will be greatly appreciated. Thanks.

[Rotidpor]

Hello everyone.


I finaly found my mistake. I hadn't changed the initialization value of the turbulent viscosity. I calculated the adequate values for k, epsilon and nut according to https://www.cfd-online.com/Wiki/Turb...ary_conditions and now my case is running smoothly.


I didn't think the underestimation of nut by a factor ~50-100 would be enough to crash my simulation.