|
[Sponsors] |
June 20, 2003, 12:40 |
shadow boundaries?
|
#1 |
Guest
Posts: n/a
|
Hi, there,
I build a mesh with GMABIT2.0.4, which contains a solid block surrouding by the fluid area (2D). When exporting the mesh, I found some extra boundaries generated, with "-shadow" suffix for their names. These seem to duplicate the surface of the solid block. What are those boundaries? How should I treat them? I seem not able to delete them. Should I specify the BC for them in FLUENT? Many thanks, jx |
|
June 20, 2003, 13:03 |
Re: shadow boundaries?
|
#2 |
Guest
Posts: n/a
|
Sorry, I just searched the forum and found some explanations for these shadow boundaries.
Now I have one thing not sure about: What is the point to specify a different BC for a shadow'ed face from its original one, since physically they are the same? Also, do I have to explicitly specify the same BCs for both the original face and its shadow, if I don't have special need to use different BCs for them? Thanks, jx |
|
June 20, 2003, 22:14 |
Re: shadow boundaries?
|
#3 |
Guest
Posts: n/a
|
I think that is because fluent calculate the solution at cell (cubic) center. The wall is located on the face, so there are must 2 cell centers in each side of the face. One is inside of fluid and one is inside of solid. Actually there should be no value calculated on the wall.
|
|
June 21, 2003, 04:52 |
Re: shadow boundaries?
|
#4 |
Guest
Posts: n/a
|
Thanks for your time,lee.
Sure I understand the calculation is cell center based. But that does not necessarily mean the actual BC should also be based on cell centers. I am not sure how fluent specify the BCs, but when I do my own coding, I convert the mathematical restraint from faces to the cell center (by interpolations), for both sides of a boundary where appropriate. So essentially the BCs are specified right on the faces, just practically they are implemented on the adjacent cell centers. -- this way, it does not make sense to specify different BCs, in fluent, for those walls and their shadows. Do I understand this correctly? (or, again, what is the point to specify different BCs for a wall and its shadow?) Someone enlightens me, please? jx |
|
June 22, 2003, 04:44 |
Re: shadow boundaries?
|
#5 |
Guest
Posts: n/a
|
Jx what lee said it is correct. I would like to add this to it. Shadow wall will get created only at junction of fluid and solid. If you open that wall in BC panel in Fluent you can see adjacent cell zone (either flid or solid.) As you might be knowing in CHT problems temp on each side (same wall but on fluid or solid face) will be different. This shadow option help us to get it. It can be used for other special cases by removing coupled option and specifying temp or heat flux to either side if you know it. I hope your doubt is clear now. Rahul
|
|
June 24, 2003, 12:44 |
Re: shadow boundaries?
|
#6 |
Guest
Posts: n/a
|
You can also have a shadow zone created when a wall of zero thickness splits a fluid zone. So in this case, you have two fluid zones on either side of the zero thickness wall. In such a problem you may want to have different boundary conditions on each side of that wall. For example, you may want to use a different emissitivity on each side in a radiation problem.
|
|
June 29, 2003, 04:44 |
Re: shadow boundaries?
|
#7 |
Guest
Posts: n/a
|
Thanks all for your responses. I get some understanding now.
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[ICEM] Shadow walls in Fluent. ICEM meshes vs Workbench | aarvay | ANSYS Meshing & Geometry | 11 | January 12, 2017 12:51 |
Setting Flow/Pressure Boundaries in Floworks | Eran | FloEFD, FloWorks & FloTHERM | 3 | August 11, 2009 04:23 |
periodic boundaries - flow through a net | PK | FLUENT | 0 | July 12, 2007 11:58 |
Periodic Boundaries in GAMBIT!! | swetha | FLUENT | 1 | November 26, 2006 22:02 |
mass flux correction at outflow boundaries | Subhra Datta | Main CFD Forum | 2 | November 24, 2003 13:11 |