[Roversella]

Hello and good evening,

I am encountering an annoying issue while creating ACMI interfaces. Particularly, when the createBaffle utility calculates the faceAreaWeight, I get a bad points error.

--> FOAM FATAL ERROR: 
Bad points:(0.0488523 -0.0409136 0.0101859) (0.0488523 -0.0409136 0.0101859) (0.0488523 -0.0409136 0.0101859)

    From function void Foam::plane::calcPntAndVec(const point&, const point&, const point&)
    in file meshes/primitiveShapes/plane/plane.C at line 90.

FOAM aborting

What I got is that the calcPntAndVec function (plane.C at 90) is trying to evaluate the normal of the triangle made of three points that are actually coincident. So the utility complains and I cannot have my ACMI built.

In the attachment, the boundary mesh and the bad-overlapping points depicted in red. The mesh is ok (high non-orthogonality warnings at the most).

Any ideas on how to fix this?

Many thanks in advance.

Nicola

[wyldckat]

Greetings Nicola,

Just in case the report https://bugs.openfoam.org/view.php?id=2860 - is ever deleted, here is part of what I wrote there:

I'm sorry to say that your report does not follow the rules for this issue tracker: https://bugs.openfoam.org/rules.php

Namely:
- A clear description of the problem to help others who encounter a similar issue in the future.
- Evidence that user error is not the cause of the issue.

Having all of the actual information placed elsewhere makes this issue pretty much useless if the two links are ever deleted.

Either way, running "checkMesh -allTopology -allGeometry" reveals that this mesh has a lot of problems, not just a "high non-orthogonality warnings at the most". Attached is the log file that shows the output of the aforementioned command.

The log file is in attachment here as well.

Essentially if the createBaffles utility has problems with a triangle that is defined by 3 identical points, then clearly something very weird happened during the generation of the mesh.
I don't know how you've generated this very interesting mesh (looks like a lobular pump), but I suspect that the existing two mesh regions (checkMesh mentions this) is the main reason why it was not possible to create the contact baffles, given that if they are not connected, then there is nothing to separate.

Best regards,
Bruno

[Roversella]

Dear Bruno,

Many thanks for your reply, and sorry if the bug report did not follow the issue tracker rules.
Actually, the problem is not that I have two separate regions ( which is indeed required for building ACMI interfaces, see Sliding interface- ACMI #20). This is probably related to the way we are developing to account for the motion.

I will let you know if I manage to fix the problem.

Regards,

Nicola