Cancel
Showing results for 
Search instead for 
Did you mean: 

Persitance of mesh size attributes on curves

Genius
Genius

I have a really hard time updating FEMAP models with design changes. The following things are blocking my progress:

 

  1. You can't flip a surface normal if the surface is within a body
  2. You often cannot remove a surface from a body before exploding it or recovering a nonmanifold add.
  3. When you explode a solid or recover a nonmanifold add to try to replace and edit surfaces, all attributes are lost (including property type, and most importantly mesh sizes on edges)
  4. Geometry is often suddenly corrupted without warning.
  5. Projecting curves (from the meshing toolbox) deletes mesh size attributes on all of the face's curves, even if they are not touched by the projection.

This third point is so so bad. But all the points mentioned above combine to completely interrupt the updating process. Let's say I need to remove a face, or add a big fillet between two faces. Either of these totally simple changes mean that I must start from scratch and redefine all mesh sizes on all edges (because I'm forced to explode or recover nonmamnifold add)! This is fine for academic models, but we work with models which have literally thousands of faces, some of which are buried deep within the structure and are very difficult to notice and correct.

 

Am I missing something? Or is this just how it is?

5 REPLIES

Re: Persitance of mesh size attributes on curves

Siemens Phenom Siemens Phenom
Siemens Phenom

In response to your posting:

 

1) The ability to reverse the normal of a surface on the body would invalidate the body as a true solid.

 

2) You can copy a surface from a body and you can choose to Match Original Mesh Sizes, Loads, Constraints...

 

2016-03-02_20-46-28.jpg

 

3) Can you post or email an example model where this occurs as this is not the way Femap has been designed.

 

4) Did you check the original geometry upon import?  I strongly recommend that when you import any geometry, regardless of the source, that you run the Geometry -> Solid -> Cleanup command on the imported geometry after importing it.

 

5) Can you post or email an example model where this occurs as this is not the way Femap has been designed.

 

Best Regards,
Chip Fricke
Principal Applications Engineer - Femap Product Development

Re: Persitance of mesh size attributes on curves

Siemens Phenom Siemens Phenom
Siemens Phenom

What version are you using?

 

Regarding Item 2, in v11.2.x we added the ability to delete individual surfaces from any Parasolid body.  In the upcoming v11.3, we also added Geometry Surface Extract, that will pull the surfaces out of the solid by themselves, all the surfaces selected at the same time are connected to each other, but their connection to everything else is severed.

 

We will look at the other issues and see if there are any workarounds or future development projects that could make this better.

 

 

 

Re: Persitance of mesh size attributes on curves

Genius
Genius

Using FEMAP 11.2.1.

 

For point 5 I've attached a model (and unexpected result).

 

1. Open model junk.modfem

2. Project curves 1999 & 2000 on face 2218

3. Note that now Face 2293 has suddenly changed mesh size along it's edges.

 

The result is shown in junk2.modfem. This loss of mesh size causes a lot of problems when updating (e.g. trimming) old geometry.

 

Thank you for your time!

 

Re: Persitance of mesh size attributes on curves

Genius
Genius

Hello,

 

I replied to the other comment, not sure if you get notified of that so I'm replying here too. Please see that response. Thanks for your time!

 

Also, I think part of my face extraction/deletion problem was that I had a non-structural mass applied to the faces I was trying to delete. Thus they were undeletable. I then removed the NSM, but still no luck, then rebuild model and they were then deletable.

 

Also, one further request to the dev team would be: I once copied a face which was part of a NSM region (total mass on area, smeared mass on the whole model). This copied the entire NSM region onto the single copied face. I then had twice as much NSM as expected, with half of it applied to a single face. I later noticed the error but it could have easily gone unnoticed. I'd request that when a face is copied out of a NSM region with total mass on faces, that the NSM not be copied, or at least some warning be given that it is being copied (although it makes no sense to do so).

 

-Kava

Re: Persitance of mesh size attributes on curves

Siemens Phenom Siemens Phenom
Siemens Phenom

Kava,

 

We have made some changes to the upcoming FEMAP v11.3 that may help you significantly in meshing parts like the one you posted.  Is it ok if I make a video using your model/part describing these changes and post it here?

 

Mark.