Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Boundary layer mesh

Please login with a confirmed email address before reporting spam

Hi!

I am simulating the 3D flow over a building. There is only one small building at the center of the computational domain. My domain size is about 11x21x6 times the height of the building. To capture the flow behavior around the building and the boundary layer closed to the ground, I used the boundary layer mesh. However, I did not understand some things.

1. If I applied the boundary layer mesh to the building's walls and the ground, I found that the whole thickness of the boundary layer mesh is relatively thin to me. Let say it is about 1/16 of the building height and I cannot increase the whole thickness of the layers. I specified the first layer thickness, the stretching ratio and the number of layers. I tried to increase the number of layers and the first layer thickness but I cannot increase the whole thickness of the boundary layer mesh. Please suggest me how to increase it.

2. If I applied the boundary layer mesh to the ground only, I can control the whole thickness of the boundary layer mesh. I can increase it and can decrease it. However, the number of grid elements around the building is too small for me. So I specified the size of the mesh on all walls of the building by using 'Size' of 'Free Tetrahedral' . Then I found that there are two regions of the boundary layer meshes in my domain. One is at the far upstream and far downstream ground of the building. The whole thickness in these regions looks normal. The second region is on the ground around the building. The whole thickness of the boundary layer mesh in this region decrease significantly comparing to the first region. I was wondering about the reason of this. Please explain to me why.

Thank you very much.

Atit


1 Reply Last Post Jun 22, 2012, 5:56 a.m. EDT
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 22, 2012, 5:56 a.m. EDT
Hi

the boundary mesh also adapt to the global thet mesh size, for me, with default boundary mesh layer it does not look too wrong, but you have a general coarse mesh, that for me is too coarse for any turbulence model. (your cell Reynolds number is above 1E5)

If you use the "physics induced mesh size (but you need a lot of RAM then) it should solve more or less OK I believe.

Again I have no benchmark to validate the model against so I cannot tell if its really "good" or not, but a mesh size much greater than the turbulent sizes is in my view not good

Another point, why the 3 lateral side walls, use rather symmetry BC as your wind is not constrained by these walls, at least in laminar this is OK, not fully sure how turbulent mode handles the symmetry w.r.t. open boundaries

--
Good luck
Ivar
Hi the boundary mesh also adapt to the global thet mesh size, for me, with default boundary mesh layer it does not look too wrong, but you have a general coarse mesh, that for me is too coarse for any turbulence model. (your cell Reynolds number is above 1E5) If you use the "physics induced mesh size (but you need a lot of RAM then) it should solve more or less OK I believe. Again I have no benchmark to validate the model against so I cannot tell if its really "good" or not, but a mesh size much greater than the turbulent sizes is in my view not good Another point, why the 3 lateral side walls, use rather symmetry BC as your wind is not constrained by these walls, at least in laminar this is OK, not fully sure how turbulent mode handles the symmetry w.r.t. open boundaries -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.