thermal expansion

General discussion about Elmer
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi Rich and Kevin,
Thanks to the transfinite gmsh command I finaly got a finer mesh in the plate (see the attached file).
But unfortunately the resulting Joule_heating repartition in the plate is still crazy and completely unevenly spread out (see the attached file which shows even negative Joule_heating values!).
I join also the corresponding Spiral_ind.geo ans Spiral_ind.msh (in the next post) files and also the zip file of the elmer project (in the next post)(after having withdrawn the 2 Spiral_ind.geo ans Spiral_ind.msh files from it).
It would be very nice if you could take a look on it and give your opinions!
We keep in touch and thanks again for your help!
Roland
Attachments
Spiral_ind.geo
(10.17 KiB) Downloaded 34 times
Joule_heating.PNG
(197.95 KiB) Not downloaded yet
Finer_mesh.PNG
(163.8 KiB) Not downloaded yet
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

remaining files here attached...
Attachments
Square_spiral_Elmer_plate_heating.zip
(210.14 KiB) Downloaded 37 times
Spiral_ind.msh.zip
(182.35 KiB) Downloaded 38 times
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: thermal expansion

Post by kevinarden »

Body 3, air, has no initial condition defined, by default it is 0 Kelvin, that is cold air compared to the plate
Rich_B
Posts: 421
Joined: 24 Aug 2009, 20:18

Re: thermal expansion

Post by Rich_B »

Hello Roland,

Gmsh posts this warning when loading the Spiral_ind.geo file:
Warning : Boolean fragments skipped - too few arguments
Apparently the warning is caused by this statement:
BooleanFragments{ Volume{58}; Delete; }{ }
If the above statement is commented out, the warning stops.

When reviewing the results in Paraview, it looks like the spiral inductor is empty, without any interior mesh. Is this what is intended? Looking at the crucible heating tutorial, the inductor interior is meshed, and assigned Air as the material.

Rich.
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi Kevin and Rich,
-When meshing the Spiral_ind.geo in 3D it seems that there is no such "Boolean fragments" error message...
-As the inductor is treated in "impedance boundary conditions" (IBC's conditions) in the EM physics, all the inner parameters (induction, current density etc...) are "brought" to the surface of the inductor, meaning that then the internal part of the inductor needs no more to be meshed. This IBC concept is classical in high frequency EM and saves a lot of meshes.
- Concerning the heat_transfer initial condition in the air there is no need to define one because the heat_transfer ( and the linear elastic physics) are only activated in the plate (aluminium + insulating layer) and not in the air, ok?
What do you think of this Joule_heating result which is completely "crazy" in the plate? This is my main worry and as long this Joule_heating is not correct there is no hope to find a realistic temperature neither a realistic displacement field ok?
We keep us informed.
Roland
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: thermal expansion

Post by kevinarden »

"Concerning the heat_transfer initial condition in the air there is no need to define one because the heat_transfer ( and the linear elastic physics) are only activated in the plate (aluminium + insulating layer) and not in the air, ok?"

if the heat doesn't transfer from the insulating layer to the plate how does it get there? if not through the air? Elmer transfers data through shared nodes of bodies. If the plate and the insulating layer does not share nodes then no data is transferred between them. I assumed it was through the shared nodes of the air mesh, but reviewing the mesh again it does not seem so.

If nodes are not shared by the bodies, then mortar boundary conditions are required to share data between the bodies.
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi Kevin, the aluminum layer and the insulating layer are stuck together like a sandwich ok? Thus the heat transfers by conduction from one to the other. Then on the top and bottom faces the BC (boundaries 4 and 9) is heat convection losses to the surroundings air with the ambient temperature as the reference temperature(293 K). This means that the air itself has not to be activated in the heat transfer physics ok?
If it more clear? If not don't hesitate to tell me what does not seem clear for you ok?
Anyway for the moment, as I specified it, we can skip the HT and LE physics and focus only on the EM for which I get a crazy Joule heating result( including negative values...!) in the aluminum layer, even after having a finer mesh, as you saw it.This is why it would be nice if you could help me to fix this "first step" problem.
Thanks in advance
Roland
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: thermal expansion

Post by kevinarden »

viewtopic.php?t=7794

Just because two bodies are in contact does not mean they transfer heat, they have to share nodes, have mortar BCs, or have radiation turned on.
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Yes you are right, and I just noticed, after zooming the 2 aluminium and insulating layers, that the insulating one has been meshed finer but not the aluminium one (which remains very coarse)! this means that the 2 layers don't share the same nodes as you say, which probably comes from a "fragment" command in gmsh which I forgot yo use properly (I don't master very well gmsh...!)
I will review the gmsh Spiral_ind.geo file and try to get a fine mesh for both layers so that they share same nodes on their common plane.
Many thanks for your help and we keep in touch.
Roland
Rich_B
Posts: 421
Joined: 24 Aug 2009, 20:18

Re: thermal expansion

Post by Rich_B »

Hello Roland,

Try this version, with a small change to Boolean Fragments.

Rich.
Attachments
Spiral10.geo
(9.43 KiB) Downloaded 37 times
Post Reply