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 The information provided may be out of date.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.

Non-convergence P2-P1 lagrange elements

Please login with a confirmed email address before reporting spam

Hi,

I am doing some CFD on a laminar case, and it appears that my study converges well when I use P1+P1 elements, but it doesn't converge at all when using P2+P1 elements!

Any clue on why this could happen? It doesn't happen when using simpler geometries...

3 Replies Last Post 28 lug 2011, 14:01 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 lug 2011, 23:03 GMT-4
I have the same problem! Any luck yet?

Sirisha
I have the same problem! Any luck yet? Sirisha

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 lug 2011, 04:55 GMT-4
Unfortunately not....

I tested it on a simpler geometry and it also caused some problems. When reducing the Reynolds number it's better and I get convergence in P2-P1, but that doesn't really solve the problem...

Let me know if you are more lucky than I !

Guillermo
Unfortunately not.... I tested it on a simpler geometry and it also caused some problems. When reducing the Reynolds number it's better and I get convergence in P2-P1, but that doesn't really solve the problem... Let me know if you are more lucky than I ! Guillermo

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 lug 2011, 14:01 GMT-4
Based on version 4.2 manual, P2-P1 is recommended for creeping flows. I am still using version 4.1 but I refer to the new manual because it has more details than the old manual.

My problem is that, with P2-P1 elements and a direct solver, I am able to achieve convergence with a coarse mesh. However when I use P2-P1 elements and a direct solver with a finer mesh, then I am unable to achieve the convergence.

Also, when I use the coarse mesh, I am running it using the COMSOL GUI on my desktop but when I use the fine mesh, I use batch mode and run it on a server. Not sure if that is making a difference. I cannot run the fine mesh on the desktop because the problem needs more memory than what my desktop can provide.

I am hoping someone has an answer for this problem!

Thanks
Sirisha
Based on version 4.2 manual, P2-P1 is recommended for creeping flows. I am still using version 4.1 but I refer to the new manual because it has more details than the old manual. My problem is that, with P2-P1 elements and a direct solver, I am able to achieve convergence with a coarse mesh. However when I use P2-P1 elements and a direct solver with a finer mesh, then I am unable to achieve the convergence. Also, when I use the coarse mesh, I am running it using the COMSOL GUI on my desktop but when I use the fine mesh, I use batch mode and run it on a server. Not sure if that is making a difference. I cannot run the fine mesh on the desktop because the problem needs more memory than what my desktop can provide. I am hoping someone has an answer for this problem! Thanks Sirisha

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.