thermal expansion

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

Re: thermal expansion

Post by Roland »

Hi,
Finaly I made the geometry and mesh of the plate+air with Gmsh and imported it in Elmer.
And now it works perfectly as showed in the here attached jpg drawing showing the plate deformation of about 10 mm in Z direction.
This means that there must be a mesh element compliance problem when importing from Comsol.
Now that the configurtion with the plate+surrounding air works correctly by coupling HT with LE, I will have to add the square spiral heating inductor and make the whole EM_HT_LE coupling.
My problem is to make such a square spiral inductor with Gmsh(here attached is the one I made with Comsol). Has somebody an idea of how to make such a square spiral tubular inductor?
Thanks again for your help and efforts!
Attachments
spiral_inductor_heating.JPG
spiral_inductor_heating.JPG (100.89 KiB) Viewed 679 times
plate_deformation.JPG
(40.26 KiB) Not downloaded yet
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: thermal expansion

Post by kevinarden »

It would require extrusion along a path, which I can not find in gmsh, Salome has it, and most cad tools, like freecad. From a cad tool you could export a step file and import it into gmsh.
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi Kevin,
Yes that's exactly what I did with Comsol which has such an extrusion ability along a path. As you suggest it I will try to see if this is possible with Freecad or Salome.
Thanks and we keep us informed
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi, I finally made a square spiral inductor with Gmsh which is fed with a 374 Volt(peak) electric potential and which couples on the aluminium plate (which is coated with a thin insulating layer on its top face). I attach the corresponding case.sif file.
The geometry (the plot is clipped to the half of the problem) and nice resulting electric potential repartition is also here attached.
As I explained it already the EM (electromagnetics) is coupled with HT( heat transfer) and with LE (linear solver) in order to get the corresponding thermal expansion of the heated aluminium plate.
Unfortunately the heating of the plate due to the EM Joule heating body force does not work well.
This seems due to a bad calculation of the Joule heating repartition in the aluminium plate as seen in the here attached Joule heating which shows a completely uneven and crazy repartition in the aluminium plate.
I don't know where this erronous calculation of the Joule heating comes from.
It would be very kind if somebody (Peter, Kevin, Rich...?) could take a look on it and help me to fix this problem.
Many thanks in advance for your help!
Roland
Attachments
Joule_heating.PNG
(465.61 KiB) Not downloaded yet
geometry_and_electric_potential.PNG
(163 KiB) Not downloaded yet
case.sif
(7.26 KiB) Downloaded 42 times
Rich_B
Posts: 421
Joined: 24 Aug 2009, 20:18

Re: thermal expansion

Post by Rich_B »

Hello Roland,

Would you be able to upload the geometry?

Maybe just the gmsh geo file, with instructions on how to build the mesh?

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

Re: thermal expansion

Post by Roland »

Hi Rich,
Yes I send you here attached the gmsh geo file (I would like to send you the whole Elmer project but it is a bit more than 1 MiB and the attachement is not accepted...)
I made the geometry just by successive linear and rotation extrusions of the initial square cross section (a square) on the linear parts and on the curved parts with an "increasing pitch" from one turn to the next turn so to get a 6 turns inductor and then I connected the 2 vertical connexion bars for the voltage feeding.
Please keep me informed when you have some news.
Thanks!
Roland
Attachments
Spiral_ind.geo
(9.46 KiB) Downloaded 40 times
Rich_B
Posts: 421
Joined: 24 Aug 2009, 20:18

Re: thermal expansion

Post by Rich_B »

Hello Roland,

Perhaps you could send the ElmerGUI project, by zipping files into individual archives that are under 1 MB in size, and sending them one post at a time?

It's possible that the "completely uneven and crazy repartition" is due to meshing, and having the exact same mesh as you are running will be helpful.

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

Re: thermal expansion

Post by Roland »

Hi Rich,
Many thanks for your helping efforts!
Ok tomorrow I will send you the whole Elmer project by trying to splitt it into several files.
Thanks and see you tomorrow for continuation
Roland
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: thermal expansion

Post by kevinarden »

Github is a good way to share files when the file sizes get large.
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: thermal expansion

Post by Roland »

Hi Rich and Kevin,
As discussed with you yesterday, Rich, here attached are:
1/the Elmer Project (zipped) without the Spiral_ind.geo (which you have already) and without the Spiral_ind_msh mesh file
2/ the Spiral_ind.msh mesh file which is separately attached in a second zipped file.
Very nice if you could take a look and give me your opinion (as you will see the Joule heating has even negative values (!!), which is completely crazy and could come from the mesh, as you said).
Please Kevin, could you just tell me how to upload a big file in the Github (I don't know how to do that...) and also remind me how, assuming that the physical properties are considered as constant, which allows a sequential coupling of the EM,HT and LE physics, it could be possible to solve first the EM and then solve the HT (with the EM_Joule heating beeing the HT body force), and at last the LE (with the thermal expansion coming from the HT temperature)?
Again thanks a lot for your help!
Roland
Attachments
Spiral_ind.msh.zip
(564.36 KiB) Downloaded 40 times
Square_spiral_plate_heating.zip
(592.51 KiB) Downloaded 40 times
Post Reply