I’m just getting started using NS Prep and ES for ion trap simulations.
When starting a newly created NS ES job I ran into the following error:
“Error : electrode block (26) not found in mesh blocks … exiting …”
However, the block is clearly there in NS prep (*.cub5) and also contains the correct surface & mesh. Is it possible the abaqus export broke the block<->mesh connection somehow?
I did not have this issue previously. The only difference from previous meshes is that the block ids are not a continuous set of integers as some geometry was deleted in prep to take symmetries into account.
EDIT: I just noticed that the mesh has 25 blocks in total and that NSES does not complain about all electrodes with ids<=25, so this might indeed be connected to block ids not being in a continuous set of integers.
Any idea what might be the issue or how I can debug this?
block all renumber start {a number higher than your block count}
block all renumber start 1
It seems a bit odd to me that this is required, especially as the cubit docs point out that the block id is arbitrary and can be chosen by the user. Could someone from Nullspace please provide some context whether this will stay like this in future and whether it is required, e.g. by the abaqus format?
Unfortunately, this is a current limitation with the ABAQUS mesh format and interface to Nullspace ES. Block numbers do indeed need to be continuous. We’re working to improve this in the future. You can avoid this for now by saving block numbering until the end of your geometry creation process just before exporting the mesh.
Thanks for your use of Nullspace ES and please continue to let us know about any questions or issues you have.