[NX/NASTRAN] different result with iges import or NX part.

[mitchellmarine]

Dear all,

I'm quite new with NX/NASTRAN(7.5) package but I have a small FEA background.

I'm doing some practical testing to learn how to use NX with NASTRAN.

I'm getting bogged on a simple I beam made with 2D shell elements.

Simulation is quite simple : I beam with pivot ends and tranversal force. It's just bending !

When I draw my beam with an other CAD software(Rhino) and import IGES in NX simulation, everything run ok, hand calculation give same result.

But when I draw my beam with NX (I tried everything in a part or assembly, same result), with the same way of preprocessing, result are stranges with different stress on top or bottom (near 20% difference!).

Thanks for your help

Regards

Mitch

 

[mitchellmarine]

preview

http://files.engineering.com/getfil...8db6-c15445a4fe03&file=nx_beam_2D_element.JPG

 

[BlasMolero]

Hello!,
I saw the picture, but not clear differences I found between the two images, when oly numbers appear in one of them.

In summary, not any difference should exist to create the geometry native in NX or imported as IGES, in the second case make sure you have your surfaces "stitched", if not surely NX AdvSim WILL NOT MERGE nodes between coincidence curves, so make sure to stitch surfaces in the idealization part. You can look also for merged nodes in the FEM environment.

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/

 

[mitchellmarine]

Hey mates,

Scall is the same for both pics.

The big difference we can see on pics is that with IGS import, bending case is clear with egual tensil/compression on external face (both deep red, around 9.5MPa)

But with NX drawn geometry top external face got around 9.5MPa while bottom face got only 7MPa. What is quite strange for pure bending?

Thanks for your help.

Regards