[FarhanAhmed]

Hi everyone,
I am trying to analyse the Wall shear Coefficient on a blade's surface. But the result which I get in CFD Post is rough(zig zag curves) and it is not smooth. I have attached the images files of both rough Wall shear and smooth wall shear. I want the wall shear to be smooth as in the first image file attached.

I am not able to figure out what exactly is the reason behind it. From reading through the forums, I understand that it might be due to faceted geometry. But there is nothing I do in my geometry which converts my surface from bspline to faceted.

And then there were also suggestions that it may be because the mesh is not capturing the geometry. And by reducing the triangular tolerance, this can be sorted out. I tried reducing the triangular tolerance to e-15 and did the meshing. Still I get the same results. (The standard procedure I follow to mesh is Robust octree -> delete volume elements -> smooth the surface mesh -> quick delauney -> smooth -> prism generation.)

Can someone please shed some light on this issue. Thank you.
Farhan

[taxalian]

Hi Farhan,
At first look on the second picture, it seems that the solution is not coverged or may be their is some flow separation that causes unsteadiness in your solution.

Originally Posted by FarhanAhmed

Hi everyone,
I am trying to analyse the Wall shear Coefficient on a blade's surface. But the result which I get in CFD Post is rough(zig zag curves) and it is not smooth. I have attached the images files of both rough Wall shear and smooth wall shear. I want the wall shear to be smooth as in the first image file attached.

I am not able to figure out what exactly is the reason behind it. From reading through the forums, I understand that it might be due to faceted geometry. But there is nothing I do in my geometry which converts my surface from bspline to faceted.

And then there were also suggestions that it may be because the mesh is not capturing the geometry. And by reducing the triangular tolerance, this can be sorted out. I tried reducing the triangular tolerance to e-15 and did the meshing. Still I get the same results. (The standard procedure I follow to mesh is Robust octree -> delete volume elements -> smooth the surface mesh -> quick delauney -> smooth -> prism generation.)

Can someone please shed some light on this issue. Thank you.
Farhan

[ghorrocks]

It could be that the face is facetted. It could be that the simulation is not converged adequately yet. It could also be many other things.

There is an FAQ on accuracy: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F