Helped Needed: Error in Discrete Geometry

Altair Forum User

Hi All,

 

Let me explain my Problem.

 

I have two geometries (say A and B ) I created mesh individually for both of them in ICEM. Then in ICEM, I brought both meshes together and saved the .uns file.

 

I read it in AcuConsole and it showed me the Mesh.

 

While converting the Mesh to Discrete Geometry, I got the following error.

 

ACUSIM: There are <20218> intersections in the input mesh

ACUSIM: *** ASSERTION in Function <agsGetNumSimRegions> File <agsUtils.cpp> Line <1375>
ACUSIM: *** Simulation model is null

 

I have tried almost everything, but I dont quite understand why the Mesh is read correctly and why while converting to Discrete Geometry, it gives and error.

 

Any help on this would be very appreciated.

 

Regards,

Juzer

ydigit

It is advisable to import direct geometry (Parasolid, ACIS etc Formats) in AcuConsole if you want to mesh with AcuConsole.

If you already have a volume mesh from ICEM (or HyperMesh), why would you want to convert it discrete geometry? 

 

TIP: If you have two geometries, one of the methods which generally works well is to import them in solidThinking Inspire and export as a Parasolid file (that can be imported/meshed in AcuConsole).

Altair Forum User

Thanks for your reply.

 

Thanks for the tips. But I am not sure if my Company has license for solidThinking Inspire.

 

Since there is Problem with interesction of elements (while converting it to discrete geom), I would expect my Simulation to Crash? Or will it?

 

I imported the .stl file also. But same error.

 

This error is only not there when I Import the .uns (mesh) file from ICEM.

 

Not quite sure what's going on.

 

Regards,

Juzer

ydigit

solidThinking Inspire also comes as part of HyperWorks and is an extra installation. Getting a parasolid geometry imported into AcuConsole would be the preferred option.

 

If mesh from ICEM is imported successfully, the volume mesh should generally run with AcuSolve. Have you tried running this mesh?

 

Working with discrete geometries should be the last option.If you are getting intersection problems while extracting discrete geometry (from ICEM mesh), it is unlikely that you would be able to create a volume mehs.

Altair Forum User

Hi,

 

yes, running the mesh with AcuSolve would be my next Task.

 

I am only worried about the Interface Definition between the two Volume (A and B )

 

I have one question regarding the Interface Definition.

 

Should there be a surface defined (separately) on each of the Volumes? Or should be a one common surface between A and B which is defined as Interface?

 

I have used CFX before, and I know that in CFX each Volume Needs to have it's own surface.

 

Thanks a lot for your prompt replies. Much appreciated.

 

Regards,

Juzer

ydigit

Are A and B both solids  or fluid or solid-fluid? Could you describe your case setup?

Depending on what you want to simulate, it may or may not be necessary to have separate surfaces for each volume.

It would be necessary to have separate surfaces (split nodes) if you want to simulate moving meshes.

Altair Forum User

Altair Forum User said:

Are A and B both solids  or fluid or solid-fluid? Could you describe your case setup?

Depending on what you want to simulate, it may or may not be necessary to have separate surfaces for each volume.

It would be necessary to have separate surfaces (split nodes) if you want to simulate moving meshes.

 

Sorry the late reply.

 

A and B, both are fluids. I am solving a moving mesh Motion Problem.

 

A is stationary and B is rotating.

 

How do I define the Interface between them?

 

Regards,

Juzer

ydigit

Workshop 4 Sliding Mesh will get you started.

You can have one interface surface and then split the nodes, as described in the tutorial.

 

http://training.altairuniversity.com/analysis/cfd/cfd-acusolve-workshops/