Convergence vs finite element density

General discussion about Elmer
Post Reply
campo
Posts: 13
Joined: 04 Mar 2016, 14:35
Antispam: Yes

Convergence vs finite element density

Post by campo »

I´ve made a simple simulation of flow through a sheet (2D) with navier-stokes equation and porous resistivity in material and the results are very accurate. However this happens when the mesh is coarse, when I make smaller elements convergence becomes more difficult and results are bad.

Can someone explain why?
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Convergence vs finite element density

Post by raback »

Hi

Hard to say with this info:

Is this transient or steady state?
Are you talking about convergence of linear or nonlinear system?

-Peter
campo
Posts: 13
Joined: 04 Mar 2016, 14:35
Antispam: Yes

Re: Convergence vs finite element density

Post by campo »

Hi Peter

Thanks for your reply and sorry for the low info, I thought it was a general question not so hard to answer. It is steady state with convergence of nonlinear system. My previous thought was that a denser mesh was hard to compute but easier for convergence, but now I realize that it's not that simple. In fact I can resolve with more rigid boundaries and lots of iterations, but wanted to understand why is more difficult and what density must use.

Thanks again

- Jon
Post Reply