wrong drag and skin friction for a flatplate test case

kevin_1 · September 7, 2016, 10:25

Hi All,

I created my mesh for a flatplate similar to the flatplate named mesh137x97.su2 with a Y+ around 50. This mesh is created with Gmsh and converted in su2 file. When I compute with my mesh, I have a very low drag around [0.000035;0.00004] and my skin friction is very low too. Moreover, when I visualize the flow.vtk file with Paraview, I have a eddy visosity thicker than the eddy viscosity on the mesh137x97.su2.

It is a problem of the fortran code who's converted the mesh or a problem due to Y+.

I will be so thankful if you can help me
Best Regards

PS : if you need the mesh file or the file which can read by paraview, please give me your email.

tomf · September 8, 2016, 03:32

Hi,

I had a similar result when my y+ value was in the wall function regime, using a y+ around 1 fixed this.

Regards,
Tom

kevin_1 · September 8, 2016, 04:49

Hi,

Yes, i know with a Y+ around 1 or less, I can have good result but my aim is to find a solution for the value of Y+ around 50.
So, if I understand, it is a problem with the size of the first cell in the mesh ?

Best regards
Kevin

tomf · September 8, 2016, 05:11

I know there has been a project to incorporate wall functions, but as far as I know there is no final version yet. So until something like that is integrated in SU2 I do not think you can expect to have good results without near wall resolution of your mesh.

kevin_1 · September 8, 2016, 08:38

I understand.

Thank you for your time.

Best regards

September 7, 2016, 10:25	wrong drag and skin friction for a flatplate test case	#1
kevin_1 New Member Boudigues Kevin Join Date: Sep 2016 Posts: 5 Rep Power: 10	Hi All, I created my mesh for a flatplate similar to the flatplate named mesh137x97.su2 with a Y+ around 50. This mesh is created with Gmsh and converted in su2 file. When I compute with my mesh, I have a very low drag around [0.000035;0.00004] and my skin friction is very low too. Moreover, when I visualize the flow.vtk file with Paraview, I have a eddy visosity thicker than the eddy viscosity on the mesh137x97.su2. It is a problem of the fortran code who's converted the mesh or a problem due to Y+. I will be so thankful if you can help me Best Regards PS : if you need the mesh file or the file which can read by paraview, please give me your email.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
skin friction and pressure Drag	kr63	Tecplot	0	June 5, 2015 02:57
the skin friction coefficient in Fluent	GJ_hit	FLUENT	3	April 15, 2013 08:08

September 8, 2016, 03:32		#2
tomf Senior Member Tom Fahner Join Date: Mar 2009 Location: Breda, Netherlands Posts: 646 Rep Power: 32	Hi, I had a similar result when my y+ value was in the wall function regime, using a y+ around 1 fixed this. Regards, Tom

September 8, 2016, 04:49		#3
kevin_1 New Member Boudigues Kevin Join Date: Sep 2016 Posts: 5 Rep Power: 10	Hi, Yes, i know with a Y+ around 1 or less, I can have good result but my aim is to find a solution for the value of Y+ around 50. So, if I understand, it is a problem with the size of the first cell in the mesh ? Best regards Kevin

September 8, 2016, 05:11		#4
tomf Senior Member Tom Fahner Join Date: Mar 2009 Location: Breda, Netherlands Posts: 646 Rep Power: 32	I know there has been a project to incorporate wall functions, but as far as I know there is no final version yet. So until something like that is integrated in SU2 I do not think you can expect to have good results without near wall resolution of your mesh.

September 8, 2016, 08:38		#5
kevin_1 New Member Boudigues Kevin Join Date: Sep 2016 Posts: 5 Rep Power: 10	I understand. Thank you for your time. Best regards