elmer elasticity needed material values

About the development of the FEM module/workbench.

Moderator: bernd

Post Reply
User avatar
bernd
Veteran
Posts: 12849
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland
Contact:

elmer elasticity needed material values

Post by bernd »

For elmer equation elasticity, even without a self weight the Density and Heat expansion coefficient are written to sif file. See https://github.com/FreeCAD/FreeCAD/blob ... if#L56-L57

Are they really needed to be in sif file by elmer solver for a elastic analysis?
raback
Posts: 75
Joined: Fri May 08, 2020 4:02 pm
Location: Finland
Contact:

Re: elmer elasticity needed material values

Post by raback »

I would think that "Poisson Ratio" and "Youngs Modulus" are the minimal required parameters.
RatonLaveur
Posts: 991
Joined: Wed Mar 27, 2019 10:45 am

Re: elmer elasticity needed material values

Post by RatonLaveur »

The way I see it, and I may be wrong, Elmer performs multi-physics "simply" by piling on equations and giving them the correct priorities for solving.

As such, it makes sense to me that solving for Elasticity only still writes all the thermal properties in case .sif file, because in a couple of minutes (I may or may not) add Heat to the mix.

It seems efficient and minimally bloated to write all the material properties in the material card every time.
raback
Posts: 75
Joined: Fri May 08, 2020 4:02 pm
Location: Finland
Contact:

Re: elmer elasticity needed material values

Post by raback »

Yes, also in ElmerGUI the database for given material includes all possible parameters: mechanical, thermal, electric. If they are given, all will be printed to the command file. However, in module-specific menu you can only edit the ones related to the specific equation. To complicate things some of the parameters may be shared. So I would say whatever works for your is fine even if it means writing extra parameters. A perfect solution would certainly make the code more complex.
User avatar
bernd
Veteran
Posts: 12849
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland
Contact:

Re: elmer elasticity needed material values

Post by bernd »

Thanks guys for the input.
raback wrote: Thu Jul 02, 2020 10:31 am A perfect solution would certainly make the code more complex.
thats true.

Neverless I would only write what is really needed. As an experience from CalculiX writer it is better to go this way. The other way is problematic because where to stop. Which value will be written and which not. Just writing everything what is known can lead to endless confusion. Thus writing what is needed is best IMHO. This gets more important on debugging the writer code if there are problems. Someone might search for a problem at a value which is not needed eventually :evil: (happened to me on CalculiX writer some time ago).
RatonLaveur
Posts: 991
Joined: Wed Mar 27, 2019 10:45 am

Re: elmer elasticity needed material values

Post by RatonLaveur »

I see where you're coming from, even as a non coder. But this would mean tagging somehow each property of the material card of FreeCAD to link it to it's preferred equation. And make sure this tagging is carried over in all solvers currently active in FreeCAD. That seems like a lot of work and tremendous opportunity for breaking stuff or interrupting continuity when new tools are explored. :(

Ultimately I don't have a horse in that particular race, but perhaps my somewhat informed opinion can help you chart a path.
User avatar
bernd
Veteran
Posts: 12849
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland
Contact:

Re: elmer elasticity needed material values

Post by bernd »

I will go step by step. Furthermore I will make sure not to break something because I started to extend elmer unit tests. This will involve because of the GSoC which will add even more elmer examples.
User avatar
bernd
Veteran
Posts: 12849
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland
Contact:

Re: elmer elasticity needed material values

Post by bernd »

git commit c3da895a1fbb for elasticity. ATM I will not change this for all other equations.
Post Reply