CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] C-Grid around a wing (http://www.cfd-online.com/Forums/ansys-meshing/88014-c-grid-around-wing.html)

bash0r May 5, 2011 10:59

C-Grid around a wing
 
Hello CFD-community,

I'd like to create a C-Grid around a geometry which is similar to a wing.
I tried using the split O-grid feature and selecting diffrent blocks/faces but I dont seem to be able to get a proper mesh for the semi-circle.
The end of the wing is not round, that's why I think a O-grid is not recommended.
The wing is placed inside a channel.

Maybe someone already had a similar problem or someone knows how to deal with this situation.

Thanks in advance!

Best regards.

siw May 6, 2011 02:19

You might want to see these tutorial videos:

http://www.youtube.com/user/ansysinc#p/u/18/tYrbScUH9RE

PSYMN May 6, 2011 09:25

You have an extra complication because the leading edge of your airfoil is aligned with the box opening...

Are you looking at flow around the box also, or just within the box? Can the box be extended a little forward of the airfoil (or the airfoil moved a little further back into the box?

These issues will very much affect what sort of model you end up with.

If your airfoil is just in a rectangular far field domain that extends forward and back (is not limited to the inside of the box), the the link that SIW gave will sort you out. But first you should probably simplify your model to what you are really studying.

If you are modeling flow thru and around the box around the airfoil, then you will need a different method, probably just an HGrid where you collapse the blocks ahead of the airfoil to make a wedge block which would propagate out the sides of the box to the real far field... Still not too hard to do, but you need to decide what you are really trying to capture.

If this box is the full extents (perhaps it is a subcomponent or you plan to array it), the collapse method will also work, but you would just propagate the wedge out to the outer surface of the box, or you could leave a 7 cornered block on the side (turn on the option to turn the 7 noded hexas into pyramids)...

bash0r May 8, 2011 09:29

1 Attachment(s)
Thank you siw. The video is really helpful for meshing airfoils!

@PSYMN:
The flow around the box is not really important.
The flow within and after the box is important. I'm sorry but the geometry cannot be changed. But there is an inlet (see picture).
Which method would you recommend?

PSYMN May 9, 2011 09:58

Is that Green line an actual step in the flow domain or is it just pained on the wall (boco change, without topology change)? That is the important question because you would have "interesting" topology where the volumes meet (where the leading edge meets the side wall with two cusps and a half circle...

bash0r May 9, 2011 10:08

Quote:

Originally Posted by PSYMN (Post 306811)
Is that Green line an actual step in the flow domain or is it just pained on the wall (boco change, without topology change)? That is the important question because you would have "interesting" topology where the volumes meet (where the leading edge meets the side wall with two cusps and a half circle...

The flow enters the pink inlet (that's where I define the inlet boundary condition, but it never gets in touch with the green wall from the left side.
The flow is only able to touch the wall after passing the airfoil.
The flow passing the airfoil then interacts with the not moving fluid on the right side. The flow has to bypass the airfoil to get to the outlet.

PSYMN May 9, 2011 13:11

OK, great. So the green line just represents a boundary layer change.

The blocking process is simple. You can use a CGRID to capture the airfoil. The CGrid should probably start after the converging duct is mostly converged (downstream of the rounds). Everything upstream of that point could just be HGrid.

bash0r May 10, 2011 05:07

Quote:

Originally Posted by PSYMN (Post 306864)
OK, great. So the green line just represents a boundary layer change.

The blocking process is simple. You can use a CGRID to capture the airfoil. The CGrid should probably start after the converging duct is mostly converged (downstream of the rounds). Everything upstream of that point could just be HGrid.

I guess I've managed to create a proper C-Grid.
Thank you!

bash0r May 16, 2011 07:14

1 Attachment(s)
I'd like to have the grid denser in the middle of my domain (right behind the airfoil).

When you look on the airfoil from behind (see picture) the oulet shows a cross where the mesh is really dense.

KateEisenhower July 9, 2015 08:33

Quote:

Originally Posted by siw (Post 306431)
You might want to see these tutorial videos:

http://www.youtube.com/user/ansysinc#p/u/18/tYrbScUH9RE

Hi Siw,

can you provide a working link to these videos. The one you posted links to the main ansysinc page.

Thank you and best regards,

Kate

bash0r July 9, 2015 11:24

Quote:

Originally Posted by KateEisenhower (Post 554650)
Hi Siw,

can you provide a working link to these videos. The one you posted links to the main ansysinc page.

Thank you and best regards,

Kate

This thread is quite old but let me give you the links to the videos:
Part 1: https://www.youtube.com/watch?v=tYrbScUH9RE
Part 2: https://www.youtube.com/watch?v=EknKVAJGEJ8
Part 3: https://www.youtube.com/watch?v=6wQPeBpvwCk


All times are GMT -4. The time now is 18:34.