Hi @Johannes_S
So right now it seems like the adapters are in the correct state, so now lets make sure that the tutorial case you are trying to run is in fact what you intend to run.
Can you pull the latest develop branch of the tutorials and then change the relevant CalculiX files and do a git diff and post the output here? That way we can be sure that only the files that you intended to change are being touched.
Hi @IshaanDesai
Here are the git diff results for the CalculiX folder:
— a/./flap.inp
+++ b/home/johannes/tutorials/perpendicular-flap/solid-calculix-test/flap.inp
@@ -8,7 +8,7 @@
3000
*SOLID SECTION, Elset=Eall, Material=EL
*STEP, INC=1000000
-*DYNAMIC, ALPHA=0.0, DIRECT
+*DYNAMIC, ALPHA=0.0
1.E-2, 5.0
*BOUNDARY
Nfix1, 1, 3
Seems like I’ve added a DIRECT.
The results for the OpenFOAM folder are in a .txt file because there were a lot of spaceholders like this:
// ************************************************************************* //
I’ve cut most of them out and copied the meaningful outputs:
OpenFOAM_diff.txt (3.5 KB)
I’ve tried to run the simulation with the exchanged CacluliX files in the new tutorial repository, but it fails like the other ones I tried before.
I’m clueless right now, maybe I have to start over again and create a new Mesh…
Hi @Johannes_S
Alright the changes to OpenFOAM seem to be non-intrusive in terms of case setup. From the git diff above it seems you have not added the DIRECT keyword in the CalculiX configuration but in fact removed it. Can you try your case again by adding it again?
I’ve tried to run the simulation with the exchanged CacluliX files in the new tutorial repository, but it fails like the other ones I tried before.
I do not really understand what you mean here? Which files did you exactly exchange?
Hi @IshaanDesai
I’ve exchanged the all.msh, fix_beam.nam and the interface_beam.nam files. So the ones that belong to the new mesh.
Because of the unchanged flap.inp the keyword DIRECT was already added in my simulation run.
Hi @Johannes_S
To be absolutely sure about your issue, I once again copied over all of your CalculiX mesh files and ran the simulation again with the develop and 2d3d_revamp branches of the CalculiX adapter. The develop branch shows the exact error that you present above. The 2d3d_revamp branch simulations runs as expected right until the end. I really do not know what more could be going wrong from your side as I am unable to see the difference in what we both are running.
I would ask you to carefully check if you are really using the develop branch of the tutorials repository and only replacing the CalculiX mesh files that you really intend to. If that is the case then lets also make sure that you are using the latest preCICE version.
Hi @IshaanDesai
Great news!
The simulation runs fine with the 2d3d_revamp adapter and the develop branch. In fact I still used the old tutorial because I’m not that into GitHub… my bad.
But this wasn’t the problem at all.
In all the hustle I didn’t installed ccx_2.19 correctly (sometimes CalculiX can be quite tricky to install). So the simulation ran on ccx_2.16. Therefore the 2d3d_revamp adapter wasn’t used.
Now everything works just fine.
I really apprecite your patience with this fiddling.
I will mark the post where you’ve sent me the link to the 2d3d_revamp adapter as the solution to this topic.
@Johannes_S great news!
As a final note, please be on the look out that at some point the 2d3d_revamp branch will be merged to develop of the CalculiX adapter, so you will have to switch then to be in sync with the latest features deployed on the CalculiX adapter.
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.