get-thread: invalid id (integer) or name (symbol) AND Negative Cell Volume

[pedrolombardiperes]

Hello, everyone,

I am performing a CFD simulation, to analyze acoustic noises from water flow in a valve, whose closing is described by a Dynamic Mesh and UDF.

My mesh has 7.5M elements, and minimum Orthogonal Quality of 0.201. The UDF imposes moving 1mm/s. Domain and zones are reordered using Reverse Cuthill McKee. The solver is Pressure-based, SIMPLE, Least Square Cell Based, Second Order, Second Order Upwind.

For the Dynamic Mesh setup, I can activate Smoothing and Layering, but, when checking Remeshing, this error pops up: get-thread: invalid id (integer) or name (symbol)

I have tried running the simulation without Remeshing, but, as the mesh moves, the cells progressively lose quality and reach null orthogonal quality. Also, I have applied an execute command for /mesh/repair-improve/improve-quality at the end of each time-step, combined with a time-step of 1e-6s, to at least maintain the quality during motion. However, the orthogonality still lowers to 0 and the Negative Cell Volume Detected error pops up.

To identify where the problem was, I generated a mesh with the default sizes (400k cells), from the same SpaceClaim file as before. Proceeding with the same setup conditions, the Remeshing could now be activated, although it did not solve the Negative Cell Volume issue, which continued to show up.

Would you please help me solve this problem? I have been several months into it, and it seems to have no way out.

Thank you so much,

Pedro

[Federico]

Hello, 

could you show some screenshots of the mesh for the dynamic mesh zones and the associated dynamic mesh settings? Layering is used for structured hexahedral mesh zones and Remeshing is generally used for unstructured tetrahedral and prisms mesh zones.

The error message that you are getting is associated to a thread or zone missing. We would need more information to better understand. 

Speculation: Are you using mesh interfaces? If one of these zones vanishes with your mesh motion, it could result in a message like this.

Perhaps the case file has been corrupted. Recreating from clean slate might help.

[pedrolombardiperes]

The mesh is built in unstructured, polyhedral cells. The motion region is the marked in red, whose movement translates 1mm in 1 second, negatively in X axis. The movement is controlled by a .prof motion profile. The Dynamic Mesh settings are found below.

 

Thank you for the rapid response.

[Federico]

It seems like your figures made it as links to your post. Forum policy reasons does not allow me to click on those links. See if you can post them again.

However, based on your response, I can say that polyhedral cells are not supported for Remeshing. So if they are part of the zones requiring remeshing, this will not work. As per my previous response, you will need to have tetrahedral cells in these regions.

[pedrolombardiperes]

Thank you, here are the images.

[pedrolombardiperes]

Regarding the polyhedral mesh, I truly don't know if this is the reason leading to the referred error, since all the meshes I have tried use 100% polyhedra, and the failure in remeshing would only happen for the ones containing Boundary Layers.