Mesh Error

[cristiano.oliveira]

Hi,

I have the problem with create mesh in the IV 2010 SP3.

Follow part.

[henderh]

Hi Cristiano,

  Could you attach a different part?  I cannot figure out how to display any solid geometry (iProperties of volume / area are zero).  See image attachment.

  Alternately, if you could attach the full assembly files that was Substituted (or shrinkwrapped) that would be even better.

Thanks, -Hugh

[cristiano.oliveira]

Please, pull down the "End of Part".

Thanks.

[BMiller63]

Hugh,

you can try this one, obtained from unchecking the Sub option, but I'm not sure its got everything still.

[BMiller63]

Cristiano,

I think we need the assembly file no?

see the file I attached, it was looking for the assembly listed.

if we do need that you might try suppressing the link to the assembly and then posting the substitute, I think that'll provide all of the geometery.

[henderh]

Ahh...of course the EoP marker.  I see the solid geometry now with the first part posted.  There are a series of unconnected rings.  Are they supposed to be connected?

I'm meshing it now in R2010_SP3.

Thanks, -Hugh

[cristiano.oliveira]

Ok Hugh,

Follow the SAT File.

Or send me FTP for the full assembly.

Thanks in advance.

[henderh]

Hi Cristiano,

  Unfortunately, the model has trouble meshing in R2010, R2011 and the latest in-development R2012.  I've logged the issue as 1372656.  Development is already looking into why it fails to mesh and is working towards a solution.

  A couple of observations is the model is very thin compared to the overall size.  With the sat model provided, R2010 does a "thin model check" and deems it too thin to mesh.  In R2011, although we made the thin model check not so restrictive,  the mesh still fails.  However, the first model (just the rings) wouldn't mesh in R2010 but R2011 can successfully mesh the model without error.  We made great improvements in our R2011 mesher compared to R2010 and we're making even more improvements for R2012.

  I thought that we could take advantage of symmetry and try meshing 1/2 (or even 1/12) of the model.  Depending on the loading conditions, sometimes we can split away a potion of the model and add frictionless constraints to the faces created by the cut.  This will cut down on the meshing and solving times.  However, the model is not perfectly symmetric (see image).

  By combining the assembly into a single part could made of a "artificial" problem for our mesher.  There is a chance that by meshing each part individually in an assembly, our mesher will succeed.  If it is an assembly in real-life this is the preferred way of simulating in Stress Analysis.

  If you could send me an email, we can set up an FTP to see if the assembly will mesh, and it will give extra information for Development to address the issue.

  Thank you for bringing this issue to our attention and providing the datasets.  It will help us in improving the robustness of our mesher.

Sincerely, -Hugh

hughDOThendersonATautodeskDOTcom

[henderh]

Hi Cristiano,

  We appreciate you sending us the assembly version.

  The assembly still has issues meshing in R2010_SP3, but it meshes successfully without error in R2011 (image attached).  If possible, could you could upgrade to R2011?

  The part version of assembly still has issues meshing in R2011 and Development is looking into how we can fix the issue in an upcoming service pack or release.

  Thanks again for your time and patience.

Sincerely, -Hugh