Keeping solids in groups. 1

[Worldtraveller]

Hopefully somebody can either give me a better way to do this, or suggest a different way of accomplishing what I'm attempting.

I have a thick laminate construction that is too thick to be modeled with plate elements, so it is being modeled with solids stacked in layers. It is a fairly complex curve, so in order to get the solids to mesh, it needs to be cut into some smaller solid sections (we want to use brick elements, not tets). (BTW, this is not necessarily my idea, but I was just brought in on this ongoing project.)

So, being a composite, there are multiple material properties based on ply, weave, etc. Can I either assign material/property values to the solids, so they will maintain those properties when split up, or is there a way to group them, and have them automatically add the new (after slicing) solids to the existing group? The slicing takes multiple actions, so I would prefer to not have to go through and do the same operation to each group, but that appears to be the simplest way to do things at this point.

Thanks for any assistance.

 

[BlasMolero]

Hello!,
My suggestion is to use LAYERS, not groups. I you need to split (slice) solids, the best way (in my opinion) is to isolate the model using layers, this will give you total freedom to manipulate geometry is that layer, allowing to prepare geometry perfectly, you can create layers based in the material. The next step is to prescribe properties to the solids using "MESH > MESH CONTROL > ATTRIBUTES ON SOLIDS", this step is critical, this way you can mesh all solids at the same time, each one with different material properties.

At the end, you can create groups automatically based in properties, yes, this is a good idea for postptocessing, but not for model creation.

Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48011 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/

 

[Worldtraveller]

Thanks. It has been a few years since I've used FeMAP, and I am re-learning all the little tricks I used to know (and discovering some things that are no longer available in version 11).

 

[Worldtraveller]

If I use layers, can I slice the whole model simultaneously and still keep the subsequent new solids in those layers, or do I need to do each layers separately? It wouldn't be too bad in this model, since there are only 4 layers, but that could get tedious if there were many properties I want to keep separated.

Thanks again.

 

[BlasMolero]

Hello!,
Any NEW geometry as resultant of issuing any FEMAP command will go to the active layer only. In any case you can use MOFIDY > LAYER >XX being XX the primitive you want to change.
To understand better your problem upload an instance of your FEMAP model that describes the situation and we can take a look to it.
Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48011 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/

 

[Worldtraveller]

Just getting back to this. For proprietary reasons, I can't upload my model, but I have some additional questions. (I think I'll do the slicing individual work layers, but I'll use the API to try to automate the process.)

As I mentioned I have very thin (~.05) laminate material and want to use solids (hex) to represent the model. I am running into an error when I try to mesh it If I try to mesh with tet elements, it meshes, but if I try to mesh with hex elements, I get:
Solid 127 can not be hex meshed. It must be subdivided further to eliminate holes in multiple faces.
Solid 127 can not be hex meshed. Unable to identify the surfaces for the base and top of the mesh.

The default size settings (set in mesh control) are the same for both instances, just changing from tet to hex.

Attached is a screenshot to give you a general idea of the shape. It is a thin, continuous solid.

 

[BlasMolero]

Hello!,
You need to Split more the geometry using command GEOMETRY > SOLID > SLICE, then you will realice that HEX meshing is posible.
If you don't have experience doing hex meshing in FEMAP, take a look to my videos here, you will learn different techniques:

http://iberisa.wordpress.com/2011/0...allado-con-elementos-hexaedricos-hex-meshing/

http://iberisa.wordpress.com/2011/02/10/mallado-con-elementos-hexaedricos-hex-meshing-ii/

http://iberisa.wordpress.com/2011/10/16/mallado-con-elementos-hexaedricos-hex-meshing-iii/

Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48011 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/

 

[Worldtraveller]

I figured that was the problem. I will try slicing the solid in a few strategic areas and see if that works.

Thanks

 

[Worldtraveller]

Also, I appreciate the tutorials, but my Spanish is no beuno.

Heck, my English isn't that good!

 

[Worldtraveller]

Well, I seem to be making progress...at least I'm getting a different error message.

"Solid 127 can not be hex meshed. Either Meshes on lateral surfaces are not fully mapped or base and top surface meshes do not match."

When I go through and look at the surface mesh, everything looks fine, even the thin edges are nicely lined up, and the mesh on the inner/outer surfaces look perfectly aligned, but there is obviously an error somewhere. I've mapped a mesh onto all the edges/curves of the model to ensure that the inner/outer surface meshes match.

I assume when doing solid HexMesh, FeMap does a surface/edge mesh first, then checks for alignment before making the surface elements into solid hexes? Is there a setting in the GUI that would allow me to loosen the check a little just to see if I can get it create the solid elements? I've tried adjusting the Max Angle tolerance upwards a little (30 is what I used). Any help would be appreciated. It's frustrating being so close to what I'm after and not quite getting the mesher to play nice.

 

[Worldtraveller]

For reference, here's a closeup of the mesh around one of the holes, showing an edge element highlighted.
Everything looks like it's nicely aligned, and the rest of the model has a nice clean almost square mesh to it. Does the relative orientation of the surfaces have an affect on how the surface mesh works? It appears that some of the surface x-y axes are swapped (not top to bottom, but from one to the next on the same inner or outer surface).

 

[BlasMolero]

Hello!,
Post a Little example here and we can see better your problem, and then to help you.

Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48011 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/

 

[Worldtraveller]

I think I see the problem. Based on that graphic you show, my shape is essentially equivalent to a non-round pipe, so I have to make a cut somewhere so Femap can pick an upper and lower surface? That's odd, but I guess it has to do with how the algorithm selects surfaces for meshing. I'll see if there's a convenient place to split my model, since it isn't symmetric.

That'll be my next step. Thanks.

 

[Worldtraveller]

ET wouldn't let me upload (a section) the working file. Is there a size limit? I'll go double check the hints and tricks.