Learning Forum

[ae22b001]

Hi, I'm new to CFD and I'm trying to create a 2D mesh in ICEM. Can someone please guide me on how to create this type of mesh? (The one on the right) Thank you!

[Rob]

That's boundary adaption. So don't use inflation and once in the Fluent solver you refine the mesh. I'd use Boundary registers. 

[ae22b001]

Hi, thanks for the reply.
I think that works, but could you maybe tell me how to do it in ICEM? I'd like to have more control over the sizing.
In this thread, they explained how I might be able to do it in Ansys Meshing using Multipart (Link to Ansys Forum Thread), could you tell me how to do a similar thing in ICEM? 
Thanks

[Rob]

It's a coarser base mesh that is adapted in the Fluent solver. 

[ae22b001]

in the thread i linked, they gave this solution to create the mesh in workbench meshing, is there a similar method in icem?

[Rob]

You would need to set all of the sizes manually and would finish up with a nonconformal mesh. It can be done but it's a lot of work and unnecessary zone linking in the solver. 

[ae22b001]

I tried it in Ansys meshing itself, In the geometry designer, I split the face into multiple parts. Then i meshed the faces separately. The only problem is that now when I open it in fluent, the interfaces between the faces cannot be set as interior. What should I do 

[Rob]

Yes, that's why I said to use a single coarse mesh and adapt. You need to look at nonconformal interfaces: you can't have an interior with the model you have. 

[ae22b001]

Got it! Thanks for your help

[ae22b001]

Hi so i tried to refine an existing mesh near the wall to reduce its ascept ratio. I wanted to split the cells into 2 to reduce its aspect ratio. To do that i set the minimum edge length to a value that is greater than the height of the cell. I expected it to only be refined in the other direction. However, it still splits the cell into 4, retaining its Aspect ratio. I split the zone into 2, to see a comparison of before vs after. 
How do I get it to split into 2 cells and not 4? Is that possible?

[Rob]

Adaption doesn't work like that. You can increase the wall adjacent cell with anisotropic inflation but not do what you want.