[Rebecca513]

Dear all,

Because my tet mesh is too large and runs very slowly in OF, I am trying to use polyDualMesh to convert the tet mesh into a polyhedral mesh.

I am learning this utility by playing with a very simple case, a rectangular flow channel. The tet mesh of the channel fails only one mesh check (small determinant), and works fine giving me results close to the theoretical estimates given by the cubic law. However, after I converted the mesh using 'polyDualMesh 60 -doNotPreserveFaceZone', the mesh failed four tests
high skewness
error in face tets: faces with low quality or negative volume decomposition tets
cells with small determinant found
concave cells found

Then I tried to include the option '-concaveMultiCells'. But the mesh didn't improve at all. I still got all the problems.

I am wondering if there is anything else that I should try when using the polyDualMesh utility to ensure the quality of the mesh generated afterwards.

Truly appreciate any suggestion and comment.

Thank you~

Best,

Hang

[anon_a]

Hey there

I am also in the process of trying to use polyhedral meshes.
According to previous posts, the tetra mesh should be produced based on a Delaunay triangulation to produce good results. Does your original mesh satisfy this restriction?

[Rebecca513]

Hi,

Yes, the tet mesh is generated using Delaunay triangulation. And the tet mesh works well in OF. But when I tried to convert it to polyhedral, it fails several mesh tests in 'checkMesh -allGeometry -allTopology'.

Thank you~

Best,

Hang

[wyldckat]

As future reference, I ended up (partially) answering here: http://www.cfd-online.com/Forums/ope...tml#post378734 post #14

[bigphil]

Hi Rebecca,

Another thing to keep in mind is that polyDualMesh does not like prism boundary layers much. It has a tendency to mess them up and produce wrong oriented faces.

I believe engrid has included a version of polyDualMesh that works much better with prism layers by not converting them. I think it is only in the development version though.

Philip

[kingmaker]

Hello

I am trying to use polyDualMesh for a blower simulation. I am getting an error saying that

--> FOAM FATAL ERROR: Created illegal face 2(4472123 4472125) at position:16171367 when filtering removed points From function polyTopoChange::compact(..) in file polyTopoChange/polyTopoChange/polyTopoChange.C at line 1056. FOAM aborting #0 Foam::error:rintStack(Foam::Ostream&) addr2line failed #1 Foam::error::abort() addr2line failed
then polyDualMesh exits. I am stuck here with this problem. Does any one have an idea why this error happens.. ?

I am using an angle of 40.

Thanks in advance.

[wyldckat]

Greetings Aditya,

Use the following command to first assess the quality of the mesh:

checkMesh -allGeometry -allTopology -constant

If it indicates that there are errors in the mesh, then polyDualMesh cannot perform magic...

Best regards,
Bruno

[openfoammaofnepo]

Dear Bruno,

How are you? I have two questions about the utility of polyDualMesh:

1, from numerical point of view, is there any advantange or disadvantage for us to use polyhedral cells, instead of tetrahedral cells?

2, for this utility, the generated cells have the same faces? or is it possible for different polyhedral cells to have different face numbers?

Thank you so much.

OFFO

Originally Posted by wyldckat

Greetings Aditya,

Use the following command to first assess the quality of the mesh:

checkMesh -allGeometry -allTopology -constant

If it indicates that there are errors in the mesh, then polyDualMesh cannot perform magic...

Best regards,
Bruno

[wyldckat]

Quick answers:

Originally Posted by openfoammaofnepo

1, from numerical point of view, is there any advantange or disadvantage for us to use polyhedral cells, instead of tetrahedral cells?

2, for this utility, the generated cells have the same faces? or is it possible for different polyhedral cells to have different face numbers?

I'll quote myself from this blog post: OpenFOAM: Interesting cases of bad meshes and bad initial conditions

Sometime ago, Symscape published this very interesting blog post: Polyhedral, Tetrahedral, and Hexahedral Mesh Comparison
The conclusion was in it's essence that: polyhedral meshes are awesome and should be used more often. And that even though tetrahedral meshes are excellent for FEM, they are lousy for CFD.

But the thing is that polyhedral meshes can derive from tetrahedral meshes, simply because of the duality of these two kinds of meshes. I won't into detail on this topic, since this is very well explained in a lot of other places, such as Wikipedia: http://en.wikipedia.org/wiki/Dual_polyhedron

[ArpanS]

1. Can polyDualMesh work on a tetrahedral mesh with nonManifoldPoints ?

2. Can polyDualMesh work on a tetrahedral mesh with cells with small determinant ?

3. Is there a version of polyDualMesh which preserves the prism layers around a wall ?

I understand there may not be any direct answers to Questions 1 and 2, just trying to get a feel of the polyDualMesh capability.

Regards
Arpan

[bigphil]

I am not sure I can answer your questions; however, I suggest you try it out and you will find out!