engine Simulations with GGI instead of Sliding Interface
Does anyone have any experience doing enginge simulations with topological changes, replacing the sliding interface by a ggi? The reason I am trying to do this is, I am having some troubles with the sliding interface, and it seemed to me that the ggi was a bit more robust. I got it working for a 3D case, but just setting up the valve bottom, so there is nothing above the valve bottom, as if it were a wall moving down. Now I am trying a 2D case with the complete valve, it runs well for one iteration, and on the second one it crashes and says:
--> FOAM FATAL ERROR:
Problem with patch-to zone addressing: some patch faces not found in interpolation zone
From function void ggiPolyPatch::calcZoneAddressing() const
in file meshes/polyMesh/polyPatches/constraint/ggi/ggiPolyPatch.C at line 77.
Has any one faced something like this? is it even possible?
Ok don't mind me, silly me I was running a 2D case, and I put the same name for the ggi patch to the left and the right of the valve. Of course this would be no problem in 3D because both are connected (valve is round), but in 2D it does't work. So now I set them to different patches (valveCurtainPort_Left, valveCurtainPort_Right, valveCurtainCyl_Left, and valveCurtainCyl_Right) and it works fine :D
Ok so it turns out there really is a problem here :S
So it did run, but with no attachDetach faces. But then if I set up the attachDetach faces, by the second iteration the GGI doesn't work. Could it be that the change in boundary because of the attachDetach face somehow affects the GGI??
I also want to perform a 3D engine simulation. I am stuck at defining the sliding interfaces. would you mind sharing your case with me?
|All times are GMT -4. The time now is 10:25.|