Geometry too complicated?

About the development of the FEM module/workbench.

Moderator: bernd

User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Wed Apr 08, 2020 4:01 am

thschrader wrote:
Tue Apr 07, 2020 5:25 pm
DE_Michael wrote:
Tue Apr 07, 2020 5:03 pm
Do you have an idea how I could solve the problem with the self-weight boundary condition?
If I remove this constraint completely, I still can't write the input file.
You use 4 different materials with different densities. There should be 4 entries for
self weight. I will try this tomorrow (I am writing from my private pc without FC0.19...)
no there should not. Even if you have four materials. You only need one self weight if you would like to have the self weight for the whole body. See attached file and screen.

multiple-mat-selfweight.FCStd
(132.91 KiB) Downloaded 8 times

Screenshot_20200408_055759.png
Screenshot_20200408_055759.png (228.34 KiB) Viewed 209 times
DE_Michael
Posts: 7
Joined: Mon Apr 06, 2020 4:42 pm
Location: Germany

Re: Geometry too complicated?

Postby DE_Michael » Wed Apr 08, 2020 4:50 pm

Hello Bernd,

thank you for your advice.
bernd wrote:
Wed Apr 08, 2020 4:01 am
Even if you have four materials. You only need one self weight if you would like to have the self weight for the whole body.
That's consistent with my experience.

Anyway, it seems to be (also) related to the materials. Because if I create the entire geometry using one material, everything works without problems within seconds.

Greetings
Michael
thschrader
Posts: 1618
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Geometry too complicated?

Postby thschrader » Thu Apr 09, 2020 6:37 am

@Bernd:
When opening your test file and doubleclicking on mesh, the transform-tool opens (?).
Creating an new mesh with gmsh gives an error.
Remeshing and running with netgen works.

@Michael:
Meshing with netgen works. But writing the inp needs endless time, still running. Lets wait and see... :)
cube_test.JPG
cube_test.JPG (79.8 KiB) Viewed 160 times
endless.JPG
endless.JPG (81.27 KiB) Viewed 160 times
thschrader
Posts: 1618
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Geometry too complicated?

Postby thschrader » Thu Apr 09, 2020 7:13 am

Writing failed. However, Eastern knocks on the door, I will try to remodel the aquarium.
writing_inp.JPG
writing_inp.JPG (41.73 KiB) Viewed 152 times
User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Thu Apr 09, 2020 7:47 am

@tschrader,
this is a pure mesh, no Netgen no Gmesh mesh object. It is similar to a mesh imported by unv mesh file format. You can not edit it with Gui tools ATM. You can distinguish them by the icon. Neither N nor G is a pur mesh object.
User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Thu Apr 09, 2020 7:48 am

but Gmsh should not give an error. I will give it a try too.
User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Thu Apr 09, 2020 9:12 am

thschrader wrote:
Thu Apr 09, 2020 6:37 am
@Bernd:
When opening your test file and doubleclicking on mesh, the transform-tool opens (?).
Creating an new mesh with gmsh gives an error.
...
no problems here on createing a new gmsh mesh object and mesh with gmsh. Might be you have an older version of 0.19 dev. In Februar and March there where some regressions on some FEM objects. They all should be fixed in latest 0.19 dev versions. Would you double check?

cheers bernd
thschrader
Posts: 1618
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Geometry too complicated?

Postby thschrader » Thu Apr 09, 2020 12:58 pm

bernd wrote:
Thu Apr 09, 2020 9:12 am
They all should be fixed in latest 0.19 dev versions. Would you double check?
Updated FC0.19:
Sorry, error remains with test-file.
Running my old steel parts works fine.
done with:
OS: Windows 10 (10.0)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.19.20477 (Git)
Build type: Release
Branch: master
Hash: 7d8ab175a68e476427e6e337fe3099b52da49dc8
Python version: 3.6.8
Qt version: 5.12.1
Coin version: 4.0.0a
OCC version: 7.3.0
Locale: German/Germany (de_DE)
results.JPG
results.JPG (53.02 KiB) Viewed 109 times
User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Thu Apr 09, 2020 1:24 pm

confirmed ...
Error:

Code: Select all

Error: The shapes for the mesh group for the reference shapes of analysis member: None could not be found!
It happens only if group meshing is activated in FEM preferences. I try to have a look at this.
User avatar
bernd
Posts: 9823
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Geometry too complicated?

Postby bernd » Thu Apr 09, 2020 1:47 pm

- https://github.com/FreeCAD/FreeCAD/blob ... 1879-L1881
- https://github.com/FreeCAD/FreeCAD/blob ... 1995-L1998

in both comments it is missing the self weight could have empty references on solid meshes too. The self weight does not have the references attribute but it is not used in FreeCAD. In this case it even creates a problem on group meshing. Load file, activate group meshing, activate the analysis, mesh.

gmsherror.FCStd
(128 KiB) Downloaded 4 times