Warning on grid generation of imported STEP model

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Warning on grid generation of imported STEP model

randress
After starting a new problem (3D, DC Conduction) the following STEP file was imported using
Geometry-->Import Geometry-->STEP Format ...

B1S1S2m_v1.step

Upon grid generation with the default setting "70", the following warning was displayed:



Warning: gobj_decompose: could not group edges.

How does this affect the imported geometry? What am I being warned about that could affect the following FEA workflow?

After grid generation the model file was saved:

gridGenWarn.fea

Thanks,
Randal
Reply | Threaded
Open this post in threaded view
|

Re: Warning on grid generation of imported STEP model

Precise Simulation
Administrator
The geometry engine has been updated to account for units in STEP and IGES CAD files which might help.

Note that your geometry may be difficult to mesh due to high anisotrotpy (very thin shell compared to the rest of the dimensions) and if you might have to try to mesh it externally/manually in Gmsh and confirm that you indeed can get the mesh you want.
Reply | Threaded
Open this post in threaded view
|

Re: Warning on grid generation of imported STEP model

randress
Precise Simulation wrote
The geometry engine has been updated to account for units in STEP and IGES CAD files which might help.
As indeed it does! Imports as meters now and no warning on grid generation.  I am not afraid of warnings that I understand, but if I don't, they make me uncomfortable :-)

Precise Simulation wrote
Note that your geometry may be difficult to mesh due to high anisotrotpy (very thin shell compared to the rest of the dimensions) and if you might have to try to mesh it externally/manually in Gmsh and confirm that you indeed can get the mesh you want.
Understood.  I should say that I do not expect to need to push this thin shell geometry to its inevitably "fatal" limits.   The two uses I have in mind for such relatively small geometry juxtaposed to larger geometry would allow for me to adjust the "thinness" so that the simulation serves only as an example that does not need to be carried to its practical/actual (very thin) limits by me ... one of which limits would be even down to molecular dimensions - obviously beyond the scope of my abilities or that of FEATool.  

So having it work as it does now should be quite sufficient. :-)

Kind regards,
Randal