Breakout models in Advanced simulation

[terrors42]

Greetings,
I working through a FEA and wanted to explore doing a break out model. There really is not a tutorial posted on how to do this, but there is a general directions in the help menu. I have created a group of elements and exported all of their displacements to a CSV file. Now I am at the point where I need to make a new FEM that includes the geometry that defines the nodes and elements in the group.I do not know exacty how to do this. Can anyone give me and explanation of how I go about creating this new FEM? Also if anyone has procedure they are willing to pass on for doing a breakout analysis that would be very helpful.

Regards,
Chris
NX6.0.5.3
Windows XP 64-bit

 

[mlamping]

Chris,

I've attached a presentation that covers breakout modeling in NX 6 upwards. It leverages spatial fields based upon tabular data to take the displacement results from the coarse model and apply them to the fine model as enforced displacement constraints. Let me know if you need more information.

The key is understanding how to export results from NX to text or EXCEL files. And you don't want just the results, but also the nodal coordinates that go with the results. Then you create a spatial field based upon XYZ point and displacement data. This presentation talks about those points at a high level with some UI snips from NX.

http://files.engineering.com/getfil...00945d2dcd&file=NX6BreakoutModelingFields.ppt

Regards,
Mark

Mark Lamping
CAE Technical Consultant
Siemens PLM Software

 

[terrors42]

Mark,
Thanks for the response, the presentation was helpful. Do you have recommendation as to how I would break the model apart. I can could always make a copy of the orginal prt file and cut the model apart in the modeling application then create a new FEM, or I could a new idealized model that is partioned then create a new FEM. Is there a prefered way to spilt the model apart for breakout analysis?

Regards,
Chris
NX6.0.5.3
Windows XP 64-bit

 

[mlamping]

Chris,

I think it is best to have the broken out body in the original model as well. So I would generate 2 FEMs (coarse mesh of the entire structure and fine mesh of just the break out body) from the same PRT. That PRT will already have the split in it. That is what I did with the model in the presentation. My models had the following file structure:

1. Master PRT
2. Idealized PRT - with the split body defined for the breakout
3. FEM1 - entire structure coarse meshed
4. FEM2 - fine mesh on just the break out body

Basically, 1 geometry input feeds 2 FEMs. I suggest doing this simply to obtain displacement results on the boundary of the breakout body. It's not required since spatial fields can interpolate to appropriate values, but getting the results on the faces in the first place certainly improves the interpolation results.

Regards,
Mark

Mark Lamping
CAE Technical Consultant
Siemens PLM Software

 

[terrors42]

Good deal Mark,
I think have enough info. to keep playing around with advanced simulation for a while.

Thanks for all the help.

Regards,
Chris
NX6.0.5.3
Windows XP 64-bit