I refined the mesh, and reduced the time step, to obtain the convergence, and yes there is a boundary layer method in Elmer.
MHD: HARTMANN PROBLEM 2D
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: MHD: HARTMANN PROBLEM 2D
This is an example of a boundary layer solution
https://github.com/ElmerCSC/elmerfem/tr ... ts/Step_ke
https://github.com/ElmerCSC/elmerfem/tr ... ts/Step_ke
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Hi Kevin,kevinarden wrote: ↑18 May 2021, 21:07 I refined the mesh, and reduced the time step, to obtain the convergence, and yes there is a boundary layer method in Elmer.
resut500.png
ycase.sif
beam.grd
I do not know how I can express my gratitude. I am going to prove it right now. I will let you know.
Regards,
Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Hi Kevin,kevinarden wrote: ↑18 May 2021, 21:07 I refined the mesh, and reduced the time step, to obtain the convergence, and yes there is a boundary layer method in Elmer.
resut500.png
ycase.sif
beam.grd
I have plotted the analytical solution by the equation present in the same report and there is a big difference with respect to the results in Elmer. How is it possible?
-Andrea
- Attachments
-
- analitical_solution.png (29.09 KiB) Viewed 47372 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: MHD: HARTMANN PROBLEM 2D
My results appear to match the report, I would have to see the files used.
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Hi Kevin,kevinarden wrote: ↑19 May 2021, 11:48 My results appear to match the report, I would have to see the files used.
Indeed the results match those of the report, however if you look at page 19, there is the analytic equation. If you solve it, for instance in Matlab or python, you will realize the deep difference.
-Andrea
- Attachments
-
- formula.jpg (25.43 KiB) Viewed 47365 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Hi Kevin,kevinarden wrote: ↑19 May 2021, 11:48 My results appear to match the report, I would have to see the files used.
I share the analytic results achieved by the Hartmann velocity profile equation encountered here https://www.researchgate.net/publicatio ... m_extended (page 16). I have used it and I have obtained the Elmer results.
-Andrea
- Attachments
-
- analitical_solution.png (31.66 KiB) Viewed 47365 times
-
- formula.jpg (25.43 KiB) Viewed 47365 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
-
- Site Admin
- Posts: 4812
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: MHD: HARTMANN PROBLEM 2D
Hi
The curves look awfully similar to me. Seems like a scaling issue. I think the 1st curve has been normalized by the average velocity? The equations are linear so the profile is the most interesting part.
-Peter
The curves look awfully similar to me. Seems like a scaling issue. I think the 1st curve has been normalized by the average velocity? The equations are linear so the profile is the most interesting part.
-Peter
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Hi Peter,
Actually, the velocity is not normalized either in Paraview and in Matlab. If you have any other examples of this problem, could you share them, please? The most interesting part is when I will apply an electric field.
Regards,
Andrea
Actually, the velocity is not normalized either in Paraview and in Matlab. If you have any other examples of this problem, could you share them, please? The most interesting part is when I will apply an electric field.
Regards,
Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
-
- Posts: 176
- Joined: 22 Mar 2021, 18:39
- Antispam: Yes
- Location: POLYTHECNIC UNIVERSITY OF TURIN
Re: MHD: HARTMANN PROBLEM 2D
Good Morning,
I have tried using a velocity of 0.5 m/s instead of 0.0015 m/s like in the previous problem. Unfortunately, the numeric and analytic solutions are quite different. I attached the solution. I am using Hartmann numbers very small to facilitate the convergence and the same case.sif. I attached the results. Thanks in advance for your answers.
Regards,
Andrea
I have tried using a velocity of 0.5 m/s instead of 0.0015 m/s like in the previous problem. Unfortunately, the numeric and analytic solutions are quite different. I attached the solution. I am using Hartmann numbers very small to facilitate the convergence and the same case.sif. I attached the results. Thanks in advance for your answers.
Regards,
Andrea
- Attachments
-
- canale_2D_report_panchall.msh
- (837.61 KiB) Downloaded 135 times
-
- Numeric solution
- M=5.png (15.84 KiB) Viewed 47345 times
-
- analytic solution
- 0.5.png (37.62 KiB) Viewed 47345 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS