How to implement an orthotropic material on differnet parts

[ibrahim zawra]

Dear deal.ii Community,

 real problem for graphical board . I used GMSH and identified different material_ids on different parts and made sure it is correct using the function print mesh info.

I wrote the class Material data to substitute the coefficient function. but I tried several things but it didn't work. I hope I can find a help for my problem.

[Wolfgang Bangerth]

On 8/16/21 10:37 AM, ibrahim zawra wrote:
>
>  real problem for graphical board . I used GMSH and identified different
> material_ids on different parts and made sure it is correct using the function
> print mesh info.
> I wrote the class Material data to substitute the coefficient function. but I
> tried several things but it didn't work.

You don't say what specifically you have tried and how or why it didn't work.
But the usual way to do this would be in the assemble_system() function, where
you would do

for (auto &cell : dof_handler.active_cell_iterators())
{
fe_values.reinit (cell);

SymmetricTensor<4,dim> strain_stress_tensor;
if (cell->material_id() == ...)
strain_stress_tensor = ...whatever is appropriate for this material...;
else
strain_stress_tensor = ...whatever is appropriate for other materials..;

for (unsigned int q=...)
for (unsigned int i=...)
for (unsigned int j=...)
cell_matrix(i,j) += ...something that uses the strain stress tensor;

...
}

Best
W.

--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/

[ibrahim zawra]

Thanks a lot that helped, I applied symmetric<2,dim> tensor because it is a heat problem. I also implemented the convection boundary condition. I have a problem in the results, it seems the contact between surfaces is not detected automatically, I want to model such contact.Please I will be grateful for your guidance.

[Wolfgang Bangerth]

On 8/18/21 9:20 AM, ibrahim zawra wrote:
> Thanks a lot that helped, I applied symmetric<2,dim> tensor because it is a
> heat problem. I also implemented the convection boundary condition. I have a
> problem in the results, it seems the contact between surfaces is not detected
> automatically, I want to model such contact.Please I will be grateful for your
> guidance.

You will have to explain more what you are doing, what the symptoms are, what
you have already tried, what worked and what doesn't. We can't see from just
one picture what might possibly be wrong.

[Jean-Paul Pelteret]

Hi Ibrahim,

My guess is that the mesh has been exported without the connectivity between the parts being enforced. So even though the physical vertices of the mesh overlap where you have different parts, there are two or more distinct vertices that lie on top of one another and the “intersecting” parts of the geometry are therefore disconnected from one another. You can normally fix this within the meshing software by using an “Imprint” operation (it might be called something else in your specific software). 

Another approach which might work (depending on which finite elements you are using) is that you can use the AffineConstraints to effectively tie the two overlapping degrees of freedom from each region together. But you’d have to manage this yourself. You can look at 

https://github.com/dealii/dealii/wiki/Frequently-Asked-Questions#how-do-i-get-the-degree-of-freedom-indices-at-vertices

to see how you can get the indices for DoFs with support at vertices, and these functions might also be of use in this regard:

https://dealii.org/developer/doxygen/deal.II/namespaceDoFTools.html#a5514e4f59ea659f63953d62ca429eaff

https://dealii.org/developer/doxygen/deal.II/namespaceDoFTools.html#a06b3c33925c1a1f15de20deda20b4d21

https://dealii.org/developer/doxygen/deal.II/namespaceDoFTools.html#a8b97e816b29ecf963370a9d8b349828f

I hope that this helps,

Jean-Paul

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
--- You received this message because you are subscribed to the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dealii+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/d8166e4d-06e6-7da4-e258-9c936fed2ce8%40colostate.edu.

[ibrahim zawra]

Thanks a lot for your reply, I'm working on a PHD thesis(Rostock university, Germany) in the domain of nonlinear model order reduction. the thesis is attached to a European project  in the field of microelectronics.

the board I attached is a graphics board and has three heat sources, 2 memories and 1 CPU under the heat sink. I should solve the transient heat problem after that as step-26 without adaptive mesh. but I wanted to learn and validate first the static case with step-5.

I also plan in the future to have a transient coupled  thermo-mechanical, I didn't find any step file for such application, any recommendation?

I'm sorry if my question wasn't clear, but as you see in the attached photo the heat sources doesn't influence the other board components which seemed not connected together.

[ibrahim zawra]

Thanks professor, that helped. It works now. For the record I used Boolean fragments (coherence) for GMSH, please find the updated result figure.