Convergence trouble for a MHD problem

Numerical methods and mathematical models of Elmer
Post Reply
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Convergence trouble for a MHD problem

Post by Andrea_P »

Hi,
I have implemented this MHD problem following the case published in GitHUB https://github.com/ElmerCSC/elmerfem/tr ... ntHartmann.
Unfortunately I don't manage to understand why it doesn't convergence with this Velocity Profile:

Code: Select all

Initial Condition 1

  ! Pressure = 0.0 
  Surface = Variable Coordinate 1
     Real MATC "-tx+h*h1" ! preaviously 0.0575originaly 0.2
  Velocity 1 = 0
      
  Velocity 2 = Variable Coordinate 1
      Real MATC "if (tx<=h*h1) ((A*sinh(Ha*B*tx/h1)-cosh(Ha*B*tx/h1)*(-1+1/(Ha^2*B^2))+1/(Ha^2*B^2))*Up); else 0;"

End
Indeed, previously, I had used another velocity profile with the same geometry and case.sif and the simulation was running without problems. The mesh is at this link: https://we.tl/t-ibUaT9s9bs. Thanks in advance for your help and answers.
Regards,
Andrea
Attachments
case.sif
(6.35 KiB) Downloaded 119 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
kevinarden
Posts: 2221
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Convergence trouble for a MHD problem

Post by kevinarden »

I was informed that the FlowSolver, stokes does not like a complete do nothing conditions on the inlet and outlet, so I made velocity 1 at the intlet and outlet (tangential to the in and out) and it is converging.
Attachments
velocity.png
(17.1 KiB) Not downloaded yet
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: Convergence trouble for a MHD problem

Post by Andrea_P »

kevinarden wrote: 01 Jul 2021, 23:45 I was informed that the FlowSolver, stokes does not like a complete do nothing conditions on the inlet and outlet, so I made velocity 1 at the intlet and outlet (tangential to the in and out) and it is converging.
Hi Kevin,
Could you share please your case.sif ?
-Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
kevinarden
Posts: 2221
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Convergence trouble for a MHD problem

Post by kevinarden »

attached
Attachments
case.sif
(6.36 KiB) Downloaded 122 times
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: Convergence trouble for a MHD problem

Post by Andrea_P »

kevinarden wrote: 02 Jul 2021, 12:29attached
Thanks Kevin, I run it right now,
-Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: Convergence trouble for a MHD problem

Post by Andrea_P »

kevinarden wrote: 02 Jul 2021, 12:29attached
Hi,
I have tried, but it doesn't converge unfortunately. I attach the image. The velocity profile in the initial condition should be kept during the time in the interface,namely from 0 to h*h1 along x.
-Andrea
Attachments
Screenshot at 2021-07-02 17-09-49.png
(81.93 KiB) Not downloaded yet
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
Post Reply