<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Dear all, <br>
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
I have been working for a time on an MHD project that normally dies due to spurious velocities in boundaries. I simplified the problem and the hydrodynamic version works well so I think maybe the divergence free condition is violated in boundaries and it produces
the inflow you can see in the attached images. The colourmap shows the velocity magnitude and the vectors the magnetic field and velocity respectively.<br>
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<img size="151154" contenttype="image/png" style="-moz-user-select: none; width: 454.641px; height: 412.071px;" unselectable="on" tabindex="-1" data-imgsize="custom" width="454.64059365643084" height="412.0706669861749" data-outlook-trace="F:1|T:1" src="cid:6cd9e2c1-5ec3-436a-abbe-e8cc1543eb97"><img size="154748" contenttype="image/png" style="-moz-user-select: none; width: 454.219px; height: 411.69px;" unselectable="on" tabindex="-1" data-imgsize="custom" width="454.2187864881577" height="411.69011020492803" data-outlook-trace="F:1|T:1" src="cid:bd2e7906-ded8-44b2-be8a-f9561b2b08a8"></div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
I am using AMR, with an HLLD solver, CT for keep divB=0, and the OUTFLOW boundary condition for the lateral sides. I have used some extrapolations of the magnetic field in the boundaries that delay the problem but not for so long.</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Now, I am trying to write an outflow for MAGY_VAR and determine MAGX_VAR to enforce divB=0 in guard cells. Basically, I want to try: [Bx(i+1)-Bx(i)]/Dx + [By(i+1)-By(i)]/Dy = 0. The problem is with faces, if I am in case gridDataStruct == FACEX, can I call
By field in FACEY? I have not found how.<br>
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
I would appreciate any idea about this or other way to solve this problem, thanks in advance!
<br>
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Regards, <br>
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Abril Sahade<br>
</div>
</body>
</html>