OpenFOAM not writing to Time 1

blockMesh.log (2.4 KB)
snappyHexMesh.log (31.2 KB)

Hello All,

I have attached logs for blockMesh and snappyhexMesh. I see in the snappyHexMesh that it does not go into refinement phase at all. At every iteration, the number of non-orthogonal cells is 0. This is not possible, as the geometry is not orthogonal. As expected, when I view the mesh after running snappyHexMesh, I only see the blockMesh and buildings are simply missing. I have tried a very simple box orthogonal and still the problem persists.

Can somebody share a solution? (470.2 KB)

1 Like

From a quick scan of the log it seems that you might have snap to false so only the castellated mesh is generated. Try enabling it so Time 2 can be generated and saved.

Not sure why nothing is there though since there should be something at time one as well. Need to find time and test it.


Yes @TheodorosGalanos,
Thanks a lot for a quick response.
I am trying exactly that right now. SnappyHexMesh is running. Will post the results.

Unfortunately, it did not work @TheodorosGalanos.
blockMesh.log (2.4 KB)
snappyHexMesh.log (13.6 KB)

Your log indicates that no refinement happens at all (look at the Surface Refinement Iteration 0 section).

Usually this would happen due to either no refinemet levels being added or a misnamed patch type. BF should avoid that and OF would show a warning anyways at the beginning. Is it possible there’s a 0 in refinement levels?

Wait I just saw it. You might be reaching your cell limit with blockMesh. In the same section it says:

Selected for refinement : 0 cells (out of 14452788)

How many are you allowing in total cause that’s already 14.5million

Yes. I kept refinement level 0,0 for the geometry. The globalRefineLevel input on snappyHexMesh I kept as it is. No input there.

maxGlobalCells I kept 250,00,000

Well if you put 0 level in the geometry no refinement will happen at al, which means that when your total cells allowed are less than the blockMesh then no geometry will be modeled.

Tbh I’ve never tried having ALL input geometries at 0, not sure if that automatically means no refinement. It could but I’m guessing it doesn’t. If your total is 2.5m, your blockMesh is currently overshooting that by 12.5m

I think I will have to use a refinement region in that case and will also have to change the default wind tunnel parameters provided by Butterfly.

I have successfully run a simulaton with everything at refinement 0.

The maxglobalcells are 25M I believe.

Then simply add a 1,1 refinement level to see what happens. Dont need to see all meshing, just follow the refinement phase.

By adding a refinement region, I could run it Thanks a lot for all the tips @TheodorosGalanos. Always appreciated.