[FLASH-USERS] Magnetic field boundary condition for NohCylindrical test problem

[Gustavo]

Hello Gregory

If you are using the USM scheme, you also need to enforce the boundary 
conditions in the face variables (basically, the MAG_FACE_VAR in FACEX 
and FACEY structures). If you did not specify BCs for the face variables 
in your implementation, probably the code is using outflow condition for 
them, so the  advancement of the magnetic field is not what you expect.

Best regards

Gustavo

El 28/08/18 a las 06:27, Boyle, Gregory escribió:
> Hi Ernesto,
>
> Thank you very much for your advice!
>
> I would like to enforce a B_theta boundary condition, which should 
> correspond to  KAXIS and MAGZ_VAR as you suggested. This seems 
> consistent with the output of the standard NohCylindrical test problem 
> having only MAGZ_VAR be non-zero.
>
> I tested putting ivar == MAGX_VAR just in case, to no avail. I am now 
> starting to suspect that, by using the unsplit staggered mesh, I 
> perhaps need to enforce a condition on the cell faces.
>
> Any other advice or suggestions is much appreciated!
>
> Cheers,
> Greg.
>
>
> ------------------------------------------------------------------------
> *From: *"Ernesto Zurbriggen" <ezurbriggen at unc.edu.ar>
> *To: *"flash-users" <flash-users at flash.uchicago.edu>
> *Sent: *Friday, 24 August, 2018 16:59:00
> *Subject: *Re: [FLASH-USERS] Magnetic field boundary condition for 
> NohCylindrical test problem
>
> Hi Greg!
> I'm not in the details, but remember that in cylindrical geometry 
> (IAXIS,JAXIS,KAXIS)=(r,z,theta).
>
> >>I am using the Unsplit Staggered Mesh MHD solver, and have included 
> a modified Grid_bcApplyToRegion.F90 that alters the standard "reflect" 
> condition on the HIGH face of IAXIS to:
> So, if you would like to modify the BC on the HIGH face of IAXIS,
>
> >> if(ivar == MAGZ_VAR) regionData(k-i,1:je,1:ke,ivar)= 1.0
> maybe you should put ivar == MAGX_VAR instead.
>
> Best!
>
> Ernesto
>
>
> El vie., 24 ago. 2018 a las 5:48, Boyle, Gregory 
> (<gregory.boyle at desy.de <mailto:gregory.boyle at desy.de>>) escribió:
>
>     Dear FLASH community,
>
>     I am attempting to modify the the 2D NohCylindrical test problem
>     (32.2.4) to include a permanent B_{\theta} condition at rightmost
>     r boundary. This is a first step towards including the effect of a
>     longitudinal current.
>
>     I am using the Unsplit Staggered Mesh MHD solver, and have
>     included a modified Grid_bcApplyToRegion.F90 that alters the
>     standard "reflect" condition on the HIGH face of IAXIS to:
>
>     k = 2*guard+1
>     if(isFace)k=k+1
>     do i = 1,guard
>        regionData(k-i,1:je,1:ke,ivar)= regionData(i,1:je,1:ke,ivar)*sign
>        if(ivar == MAGZ_VAR) regionData(k-i,1:je,1:ke,ivar)= 1.0
>     end do
>
>     However, when I look at the output it seems the condition of
>     setting MAGZ_VAR to 1.0 is being ignored.
>
>     Any suggestions would be greatly appreciated!
>
>     Cheers,
>     Greg.
>
>
>
> -- 
> /Dr. Ernesto Zurbriggen/
> /
> /
> /Consejo Nacional de Investigaciones Científicas y Técnicas 
> //(CONICET)//, Instituto de Astronomía Teórica y Experimental (IATE)./
> /
> /
> /Universidad Nacional de Córdoba //(UNC)//, //Observatorio Astronómico 
> de Córdoba //(OAC)//./
> /
> /
> /Teléfono: +54 0351 4331064-5, interno 222./
> /_Laprida 854, Bº Observatorio (CP 5000), Córdoba, Argentina_.
> /
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20180829/de721d54/attachment-0001.htm>