CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   Airfoil Meshing (http://www.cfd-online.com/Forums/star-ccm/107829-airfoil-meshing.html)

PLP3A_ENSIAME October 7, 2012 10:26

Airfoil Meshing
 
Hello every body I'm new in this forum and I hope you will have a solution or informations about my problem. I'm trying to mesh an airfoil on STARCCM+ and it's a SELIG 4083 profile. So first I defined a volume around my profile in CATIA like a "C-grid" with a free stream and then I imported the STP in STARCCM+. My problem occures during the meshing. I have few problem with my different volumetric control. I created five blocks around my profile to be more accurate but when I mesh, the mesh considers only the smallest or the biggest volumetric control. Is there any interferences between each volumetric control or not ? Then the other problem is at the trailing edge of the profile because the trailing edge is sharp I have a bad mesh on the boudary layer and I don't know how to improve the mesh here. Maybe I can try to put a cylinder as a volumetric control.
If someone can bring me some help or if someone have ever meshed an airfoil in STARCCM+ his help is welcomed !

siara817 October 8, 2012 03:41

in the continua, click on mesh. In the property window, select preregion option. Then generate mesh, and see if it is solved.
good luck

PLP3A_ENSIAME October 8, 2012 06:28

Hello,

I made the change you asked me and the result is not better as you can see on the pictures following. I still have the problem with my different volumetric control and then I don't have a good boundary layer at the leading edge of my profile.
Any idea ?

Thank you for your answer and your help.

Below you can see first the mesh with the option you asked me to change and then without this option.

http://img4.hostingpics.net/thumbs/m...eadingedge.png
http://img4.hostingpics.net/thumbs/m...dingedgeWC.png
http://img4.hostingpics.net/thumbs/m...ailingedge.png
http://img4.hostingpics.net/thumbs/m...lingedgeWC.png

And here you can see what is the real problem, I have 5 different volumetric control but the mesh considers only one of them.
http://img4.hostingpics.net/thumbs/m...Globalview.png
http://img4.hostingpics.net/thumbs/m...ricControl.png

siara817 October 8, 2012 07:16

About the quality of the mesh, you can set small values for the airfoil. Try once without prism layer and one time with prism layer.

PLP3A_ENSIAME October 8, 2012 08:27

Hi,

I tried to mesh first without Prism Layer Mesher in the smallest volumetric control and then I tried with the Prism Layer Mesher in the smallest volumetric control and here is the difference with the change.

http://img4.hostingpics.net/thumbs/m...yerTrimmer.png
http://img4.hostingpics.net/thumbs/m...yerTrimmer.png
http://img4.hostingpics.net/thumbs/m...yerTrimmer.png

NB : this trial is made with a "Slow" default growth rate.

So I think it's better now but I still have a problem in the mesh. The problem is that now I have a segmentation of my profile. How can I improve this to stick to my real profile ? I decrease my base size from 0.5 to 0.4 and it's better but ahead my profile a small base size is useless. Any idea ?

Thank you :)

PLP3A_ENSIAME October 8, 2012 08:35

With 0.3 in my base size :

http://img4.hostingpics.net/thumbs/m...Trimmerbis.png

PLP3A_ENSIAME October 8, 2012 10:31

This is what I have with the version 7.04 of STARCCM+. Maybe there are problem of the displayer in STARCCM+ 4.04 :

http://img4.hostingpics.net/thumbs/m...MeshScene2.png

http://img4.hostingpics.net/thumbs/m...ne2attaque.png

http://img4.hostingpics.net/thumbs/m...cene2fuite.png

abdul099 October 8, 2012 14:44

Reduce the surface size at the airfoil boundary.
When setting volumetric controls, CCM+ will only consider the smalles value within a given area, so when a volumetric control is completely covered another one with the same or smaller custom sizes, the smaller vc will have no effect.

By the way, I suspect your prism layer thickness to be way too high. Try a smaller thickness.

lava12005 October 8, 2012 21:26

I agree with Abdul099, you should reduce the surface size of the airfoil.

Btw is your leading edge airfoil doesn't looks like a curve at all. You can try put a volumetric control on the leading edge only and put smaller surface size. Then generate the surface mesh first to see whether you get the airfoil that you want (provided the CAD file is correct)

PLP3A_ENSIAME October 9, 2012 05:38

Hello,

I followed your advices and I reduced the surface size on my airfoil thanks to "custom surface size". I decrease the relative minimum size from 25 to 10 and it's visibly better at the leading edge but I lose accuracy at the traileng edge because I have cells too long compared with the boudary thikness as you can see on pictures. I will try with only on VC on the Leading edge and one on the trailing edge to see if there is any difference.

http://img15.hostingpics.net/thumbs/...eonairfoil.png

http://img15.hostingpics.net/thumbs/...eonairfoil.png

PLP3A_ENSIAME October 9, 2012 08:19

Re,

I tried a new thing : I didn't put any VC at the leading edge, I reduced only the surface size of the airfoil and I obtained a very good surface mesh which follow my geometry profile. At the trailing edge, I tried to put a VC and I obtained a mesh like you can see on picture but (tell me if I'm wrong) it's not a very good mesh for the trailing edge and I really don't know how to handle a shaped geometry with the meshing. I need someone throws light on this problem.

Thank you

http://img15.hostingpics.net/thumbs/...viewSimple.png

http://img15.hostingpics.net/thumbs/...edgeSimple.png

http://img15.hostingpics.net/thumbs/...edgeSimple.png

siara817 October 9, 2012 14:10

Do you use prism layer ?

PLP3A_ENSIAME October 9, 2012 15:38

Yes I use prism layer in my mesh and in my VC

siara817 October 10, 2012 02:49

Try one time not to use prism layer, instead reduce surface size and also reduce the growth rate. Then see what will be the mesh situation. I hope it work

abdul099 October 10, 2012 18:29

Cut the trailing edge. It's easier to mesh when the TE is not a sharp edge. And I still think your prism layer thickness is much too high.
And maybe you can think about the aspect ratio of your near wall cells...


All times are GMT -4. The time now is 03:18.