CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Native Meshers: snappyHexMesh and Others (http://www.cfd-online.com/Forums/openfoam-meshing-snappyhexmesh/)
-   -   sHM, 2D, foils, sharp edge (http://www.cfd-online.com/Forums/openfoam-meshing-snappyhexmesh/115282-shm-2d-foils-sharp-edge.html)

skeptik March 27, 2013 11:47

sHM, 2D, foils, sharp edge
 
2 Attachment(s)
Hi all!

I had some problems with my airfoil 2D mesh.

First one was in bl at the tip of the foil. The layers just collapsed without connection.
Attachment 20189
I just reduced the number of surface smooth normals and this problem was solved.

Another one was in layers at the tail
Attachment 20190
I did not succeed. May be there is no way to solve this problem (maxConcave was 180 and featureAngle 90).

May be somebody would say something about it or suggest any idea about sHM and geometry with sharp edges?

MadsR April 2, 2013 11:38

I don't know if this is applicable to your case (especially since it seems that you have a very sharp TE), but you could try my 2D airfoil mesher which generates a blockMesh file ready for OpenFOAM.

http://hvirvel.dk/airfoilmesher/

/Mads

skeptik April 3, 2013 03:27

Quote:

Originally Posted by MadsR (Post 417864)
I don't know if this is applicable to your case (especially since it seems that you have a very sharp TE), but you could try my 2D airfoil mesher which generates a blockMesh file ready for OpenFOAM.

http://hvirvel.dk/airfoilmesher/

/Mads

:) Thanks a lot!

I tried this mesher (is it your?) several times, but i had problems using it. May be i was due to airfoil profile data.

So for my case i developed blockMesh, with C-mesh suitable well to my test case. But sHM is just for my interest. May be little bit later i will find a solution.

MadsR April 3, 2013 05:40

Yeah it is mine, trouble is if you have sharp trailing edge. My generator creates o-meshes and expect some bluntness at the TE. If you just click "submit" without submitting any airfoil-data-file, it will default to a mesh for a blunt-TE NACA0015. There is an example airfoil-data-file on the page too.

Mads

haakon April 3, 2013 05:52

If you want to try I have made a C-mesh generator here: https://www.hpc.ntnu.no/display/hpc/...l+Calculations (see bottom of page). It can work with sharp trailing edges. Currently it generates NACA 4-digit airfoil meshes, but as long as you can get a table of coordinates into the script, it will mesh any airfoil with some small modifications.

However, it is not "production quality". If I were to use it in any serious work, I would have made some close-to-surface blocks that I would have fitted perfectly to the airfoil to get a more non-orthogonal mesh close to the surface.

skeptik April 3, 2013 07:14

2 Attachment(s)
Quote:

Originally Posted by MadsR (Post 418017)
Yeah it is mine, trouble is if you have sharp trailing edge. My generator creates o-meshes and expect some bluntness at the TE. If you just click "submit" without submitting any airfoil-data-file, it will default to a mesh for a blunt-TE NACA0015. There is an example airfoil-data-file on the page too.

Mads

I checked your script again and i got nothing.
Profile data is here (there's no sharp edge).

Attachment 20383

blockMesh file generated by your script here

Attachment 20384

I used default values in your script, except domain radius. I set 5 meters (profile data was in y/c, x/c coordinates)

skeptik April 3, 2013 07:22

Quote:

Originally Posted by haakon (Post 418022)
If you want to try I have made a C-mesh generator here: https://www.hpc.ntnu.no/display/hpc/...l+Calculations (see bottom of page). It can work with sharp trailing edges. Currently it generates NACA 4-digit airfoil meshes, but as long as you can get a table of coordinates into the script, it will mesh any airfoil with some small modifications.

However, it is not "production quality". If I were to use it in any serious work, I would have made some close-to-surface blocks that I would have fitted perfectly to the airfoil to get a more non-orthogonal mesh close to the surface.

Thanks.

In my C-mesh i have similar blocking structure. It's quite good for simulations. Close-to-surface blocks is a good idea, cause i had problems with this blocking structure. It was solved by adjusting of the edge grading. Some additional blocks could make mesh better definitely but it needs time.

sh.d June 1, 2013 10:08

help
 
hi skeptik
i want to create a c-mesh in blockmesh but i cant
please can you help me ?

skeptik June 2, 2013 06:31

Hi

what is your problem?

sh.d June 2, 2013 11:01

1 Attachment(s)
hi
i want to creat a c-meh in blockmesh
please please can you help me?
there is this error:
Create time

Creating block mesh from
"/home/shirin/OpenFOAM/shirin-2.1.x/run/tutorials/mesh/snapp/block/constant/polyMesh/blockMeshDict"
Creating curved edges
Creating topology blocks
Creating topology patches

Check topology
#0 Foam::error::printStack(Foam::Ostream&) in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/lib/libOpenFOAM.so"
#1 Foam::sigSegv::sigHandler(int) in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/lib/libOpenFOAM.so"
#2 in "/lib64/libc.so.6"
#3 Foam::blockMesh::checkBlockMesh(Foam::polyMesh const&) const in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/lib/libblockMesh.so"
#4 Foam::blockMesh::createTopology(Foam::IOdictionary const&, Foam::word const&) in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/lib/libblockMesh.so"
#5 Foam::blockMesh::blockMesh(Foam::IOdictionary const&, Foam::word const&) in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/lib/libblockMesh.so"
#6
in "/opt/OpenFOAM/OpenFOAM-2.1.x/platforms/linux64Gcc46DPOpt/bin/blockMesh"
#7 __libc_start_main in "/lib64/libc.so.6"
#8
at /home/abuild/rpmbuild/BUILD/glibc-2.14.1/csu/../sysdeps/x86_64/elf/start.S:116
Segmentation fault

sh.d June 2, 2013 12:06

2 Attachment(s)
hi
i can create c-mesh but the block is not correct
please help me

skeptik June 2, 2013 16:05

Quote:

Originally Posted by sh.d (Post 431459)
hi
i can create c-mesh but the block is not correct
please help me

it seems that you have extra blocks at the tip zone.
you did arc edges, that's right, but there left straigth edges of another block, i see

MadsR June 3, 2013 04:45

@Ilya, my simplistic online blockMesh-preprocessor can't handle negative values for the x-coordinate. If you change those three points where x is negative in your airfoil geometry it seems to work.

Just move it a bit out of the positive x-axis to the smallest x-coordinate is zero.

Mads

skeptik June 3, 2013 04:53

@Mads
Now it's clear!

Thank you very much. I'll try it little bit later. I think your tool would be very useful.

skeptik June 6, 2013 09:08

Quote:

Originally Posted by MadsR (Post 431573)
@Ilya, my simplistic online blockMesh-preprocessor can't handle negative values for the x-coordinate. If you change those three points where x is negative in your airfoil geometry it seems to work.

Just move it a bit out of the positive x-axis to the smallest x-coordinate is zero.

Mads

It works fine.
Thank you for your script. It saves time.

sh.d August 2, 2013 11:30

rotate
 
Hi dear forumer
i want rotate airfoil in different attack angle but fix the domain and it dont rotate
Can anyone help me?

skeptik August 7, 2013 03:16

answer
 
Quote:

Originally Posted by sh.d (Post 443515)
Hi dear forumer
i want rotate airfoil in different attack angle but fix the domain and it dont rotate
Can anyone help me?

look for 'surfaceTransformPoints'. i think it could help you.

MadsR August 7, 2013 04:20

@skeptik: Um, did I understand you wrong?: Just change the inlet velocities.

Mads

skeptik August 7, 2013 07:15

rotate
 
Quote:

Originally Posted by MadsR (Post 444275)
@skeptik: Um, did I understand you wrong?: Just change the inlet velocities.

Mads

Of course, it's reliable way to change flow direction. But if you have a wing in tube, better to change surface position. So it depends from case.

MadsR August 7, 2013 08:48

Oh I didn't realise you had your wing in a tube. That's a bit unusual :)

Mads


All times are GMT -4. The time now is 14:13.