CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Deciding the mesh element size in CFD (http://www.cfd-online.com/Forums/fluent/105577-deciding-mesh-element-size-cfd.html)

saisanthoshm88 August 2, 2012 18:26

Deciding the mesh element size in CFD
 
Could some one please let me know how to choose / decide on the element size for meshing in a CFD analysis

cfd seeker August 3, 2012 10:00

Quote:

Originally Posted by saisanthoshm88 (Post 375128)
Could some one please let me know how to choose / decide on the element size for meshing in a CFD analysis

it depends what is the problem in hand, what you are trying to simulate and what results you are looking for? can you tell what problem you are solving?

saisanthoshm88 August 5, 2012 06:41

My intrest is to know some generic guidelines to be followed while choosing the element size for CFD meshing.

I'd really appreciate it if you could quote some examples to explain it

Thanking you for the quick reply

cfd seeker August 5, 2012 10:59

Quote:

Originally Posted by saisanthoshm88 (Post 375430)
My intrest is to know some generic guidelines to be followed while choosing the element size for CFD meshing.

I'd really appreciate it if you could quote some examples to explain it

Thanking you for the quick reply

General guideline is "mesh size should be small enough to capture all flow features properly and large enough to ensure that overall mesh size is within your computational resources" e.g if you are solving an external aerodynamic problem then upstream the size does not really matter but downstream in order to capture wake properly mesh size should be very small as compared to upstream. also if you want to capture wing tip vortices then again size in the span wise direction should be very small. On the airfoil or wing if the flow is fully turbulent then normal spacing in the vicinity of leading edge is enough but if the flow is transitional then node clustering at the leading edge is required to properly capture transition.
In summary I would like to say that it is the physics of the problem which defines how small or how large is the element size in your mesh.

Touré August 6, 2012 03:34

This is the problem of the convergence versus the meshing size. There is no formula for that. You should do several simulations decreasing the size of the elements until your residuals don't change a lot between two types of meshing. Then, you get the good meshing for your problem. Your elements should be as small as possible to catch the boundary layer for example. For the boundary layer you decide how many layers you want and you can also approximate the height of the boundary layer.


All times are GMT -4. The time now is 07:08.