GSoC2020, FEM: extend examples
Moderator: bernd
Forum rules
and Helpful information for the FEM forum
and Helpful information for the FEM forum
Re: GSoC2020, FEM: extend examples
I was going through Elmer's Gui tutorial here and I found out that the capacitance two balls example is actually the tutorial 6 (page number 28) of that pdf, so should't we rename the example to link it with the tutorial documentation?
Or is it too late since it's already in master?
By the way, Elmer has some really cool examples.
And @Bernd, I had a doubt which my sound silly to you.
Why do we define the materials in the examples, whereas in the GUI we can select the pre-defined materials?
Is it because we can't access the pre-defined ones from code?
Or is it too late since it's already in master?
By the way, Elmer has some really cool examples.
And @Bernd, I had a doubt which my sound silly to you.
Why do we define the materials in the examples, whereas in the GUI we can select the pre-defined materials?
Is it because we can't access the pre-defined ones from code?
Re: GSoC2020, FEM: extend examples
@Bernd, another doubt.
In the material_multiple_bendingbeam_fiveboxes and the material_multiple_bendingbeam_fivefaces examples, the third material has no reference.
Judging by symmetry, the first and fifth box/face should have the third material.
Though, even if I assign the third material to them it is not changing the result.
But then why is the material there if it isn't assigned to any box/face in these examples?
Or is that material getting auto assigned to the boxes/faces having no material, cause we usually don't explicitly assign material to the model in other examples.
Even if that is happening, I would prefer having explicit assignments for materials in multimaterial examples, for an easy understanding.
In the material_multiple_bendingbeam_fiveboxes and the material_multiple_bendingbeam_fivefaces examples, the third material has no reference.
Judging by symmetry, the first and fifth box/face should have the third material.
Though, even if I assign the third material to them it is not changing the result.
But then why is the material there if it isn't assigned to any box/face in these examples?
Or is that material getting auto assigned to the boxes/faces having no material, cause we usually don't explicitly assign material to the model in other examples.
Even if that is happening, I would prefer having explicit assignments for materials in multimaterial examples, for an easy understanding.
Re: GSoC2020, FEM: extend examples
I mimicked the "hinged beam" in this thread and created a test case for Elmer. See
https://github.com/ElmerCSC/elmerfem/tr ... BeamHinged
https://github.com/ElmerCSC/elmerfem/tr ... amHinged3D
This approach cannot really do very large displacements since the constraints are linearized. Just wanted to try out that they work...
-Peter
https://github.com/ElmerCSC/elmerfem/tr ... BeamHinged
https://github.com/ElmerCSC/elmerfem/tr ... amHinged3D
This approach cannot really do very large displacements since the constraints are linearized. Just wanted to try out that they work...
-Peter
Re: GSoC2020, FEM: extend examples
On the numbering of Elmer tutorials: We try to have all cases backwards compatible to the extreme but there is no such aim for tutorials numbering. The numbering comes automatically from LaTex and would we write nicer tutorials up front the numbering will change. But once the tutorials have been named I don't think that there is a foreseeable reason to change the name.
-Peter
-Peter
Re: GSoC2020, FEM: extend examples
Based on the description given in the post and simscala validation test, I created the following model but I am getting a vonmises stress of 35.6 MPa:Sudhanshu wrote: ↑Sat Jul 25, 2020 4:54 pmCan I get the FreeCAD file of this one?bernd wrote: ↑Tue Jul 21, 2020 8:25 pm - this very simple example from HoWil for CalculiX and Elmer. the analytical calculation should be added at file top as a coment. https://forum.freecadweb.org/viewtopic. ... 20#p194595
Couldn't find it in that thread.
- Attachments
-
- simple_thermomech.FCStd
- (153.78 KiB) Downloaded 45 times
Re: GSoC2020, FEM: extend examples
Great!raback wrote: ↑Mon Aug 03, 2020 4:21 pm I mimicked the "hinged beam" in this thread and created a test case for Elmer. See
https://github.com/ElmerCSC/elmerfem/tr ... BeamHinged
https://github.com/ElmerCSC/elmerfem/tr ... amHinged3D
This approach cannot really do very large displacements since the constraints are linearized. Just wanted to try out that they work...
Can you share with us the results you got for hinged beam with Elmer?
Re: GSoC2020, FEM: extend examples
In the second attempt, I have changed the material property to match those given on the simscala site and made a finer mesh (with max element size 50mm). Then, to get a better look at the results we see the histogram generated by calculix:Sudhanshu wrote: ↑Tue Aug 04, 2020 3:56 amBased on the description given in the post and simscala validation test, I created the following model but I am getting a vonmises stress of 35.6 MPa:Sudhanshu wrote: ↑Sat Jul 25, 2020 4:54 pmCan I get the FreeCAD file of this one?bernd wrote: ↑Tue Jul 21, 2020 8:25 pm - this very simple example from HoWil for CalculiX and Elmer. the analytical calculation should be added at file top as a coment. https://forum.freecadweb.org/viewtopic. ... 20#p194595
Couldn't find it in that thread.
If I quote the result from simscala:
So from the histogram, for the "whole beam" we will get the von Mises stress as 25 Mpa, which is pretty close.σ
= 24.6 Mpa in the whole beam.
For Elmer, we get a similar result (in MKS):
Judging from the gradient, for the whole beam, the stress should be around 25 only but I don't know how to get the exact value yet.
Re: GSoC2020, FEM: extend examples
If I am right than "6. Constraint fluid boundary" is currently not supported. That's why it is in the sub-Menu under "Constraints without solver". This constraint was originally implemented for the use with CFD which never reached a large audience because CfdOF split off and created a new workbench.Sudhanshu wrote: ↑Mon Jul 27, 2020 4:58 pmHoWil wrote: ↑Mon Jul 27, 2020 3:13 pm5 to 7 should be covered by the "Thermal flow in curved pipe" model mentioned here...
https://forum.freecadweb.org/viewtopic. ... 20#p418076
I think there is no better model to show these constraints in action. Yes this model only works with Elmer.
You are right "5. Constraint body heat source" is not used in the above mentioned example... but it is used in "Tutorial 1" in the https://www.nic.funet.fi/pub/sci/physic ... orials.pdf which is discussed over here https://forum.freecadweb.org/viewtopic. ... 40#p183704; looks like the model has to be built again.....but should not be much work when importing the mesh from the tutorial files. Will do that.
"7. Constraint initial flow velocity" is in the "Tutorial 9" of the GUI-Tutorials the 2D!!!-model can be found attached:
Re: GSoC2020, FEM: extend examples
Here the model for Tutorial1 for the body heat source
- Attachments
-
- Screenshot from 2020-08-07 22-30-42.png (463.41 KiB) Viewed 2394 times
-
- Turorial1.FCStd
- (76.14 KiB) Downloaded 138 times
Re: GSoC2020, FEM: extend examples
great, good stuff HoWil.