Critical Plastic Strain

[HarryvL]

zip contains FC-file, SMath-file, force definition pdf
DNV_T_section.zip

I can't get the BooleanFrangements to mesh with GMSH:

Code: Select all

File to load not existing or not readable: /tmp/fcfem_88icrfgd/BooleanFragments_Mesh.unv

[thschrader]

I can't get the BooleanFrangements to mesh with GMSH:

Aaah...sorry.
I used the new DelQuad element option in FC 0.20.1 release for meshing.
Switching back to "Automatic" should work.

DelQuad_elements.JPG (32.68 KiB) Viewed 1170 times

[HarryvL]

Maximum critical strain ratio is 0.85 ... pretty close to cracking:

Screenshot from 2022-09-07 13-34-01.png (161.84 KiB) Viewed 1154 times

[HarryvL]

Well, I made my first pull request https://github.com/FreeCAD/FreeCAD/pull/7467 and must have broken a few rules doing so ... let's see.

While running the unit tests I got 1 unrelated failure. This must have been caused by a previous PR. Where do I report this issue?

Code: Select all

FAIL: test00010 (PathTests.TestRefactoredTestPost.TestRefactoredTestPost)
Test Outputting all arguments.

Traceback (most recent call last):
File "/home/harryvl/FreeCAD/build/Mod/Path/PathTests/TestRefactoredTestPost.py", line 194, in test00010
self.single_compare(
File "/home/harryvl/FreeCAD/build/Mod/Path/PathTests/TestRefactoredTestPost.py", line 99, in single_compare
self.assertEqual(gcode, expected)
AssertionError: 'Argu[259 chars]s the previous line\n --no-axis-modal O[2703 chars]0)\n' != 'Argu[259 chars]s the\n previous line\n[3103 chars]0)\n'
Diff is 6515 characters long. Set self.maxDiff to None to see it.

[HarryvL]

For completeness, the critical strain ratios at beam sections 1 and 2:

Critical Strain Ratio @ Section 1.png (23.28 KiB) Viewed 1060 times

Critical Strain Ratio @ Section 2.png (21.42 KiB) Viewed 1060 times

[HarryvL]

I tried to come up with another practical case displaying high triaxiality (and therefore low ductility).

This is the fictitious case of two intersecting plate girders, with dimension taken from IPE600 and IPE300, respectively.

The material properties are taken from @thschrader's DNV example. The deep plate girder is loaded in uniform tension and the shallow girder is fixed at the ends.

Screenshot from 2022-09-11 12-45-40.png (30.65 KiB) Viewed 1011 times

CCX breaks down at a uniform tension of 356MPa and produces the following critical strain ratio.

Screenshot from 2022-09-11 12-02-25.png (157.22 KiB) Viewed 1011 times

The maximum critical strain aratio at this load is only 0.015. The triaxiality is clearly not high enough to cause low ductility.

Finally a cut view in Paraview:

Triaxiality 2.png (48.2 KiB) Viewed 1011 times

and the FC file for those who want to experiment

Triaxiality.FCStd

(39.39 KiB) Downloaded 52 times

[HarryvL]

Well, I made my first pull request https://github.com/FreeCAD/FreeCAD/pull/7467 and must have broken a few rules doing so ... let's see.

This has been merged now (see https://wiki.freecadweb.org/Release_notes_1.0). Please give it a try. Keep in mind that the result (as all stress and strain related results) is very mesh dependent. This is because CCX extrapolates stresses and strains from integration points to nodal points and this involves averaging over the various elements connected to the node. I therefore recommend to always repeat the analysis with a finer mesh and see if the results properly converge.