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 The information provided may be out of date.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.
Solver dependency
Posted Aug 10, 2012, 4:01 a.m. EDT Studies & Solvers Version 4.3 1 Reply
Please login with a confirmed email address before reporting spam
Hi,
I'm currently doing a simulation that involves free convection and I've read everything I could find on this forum about the subject. Something important that comes out, is that the solver that should be used is the "fully coupled" one. My solution is however converging with a "segregated" solver, but not with this fully coupled one. My question is the following : Can I trust my results independently of the solver (seeing that when I refine the mesh, the results are pretty much the same) ?
Thank you for the help. Best regards,
Pascal
I'm currently doing a simulation that involves free convection and I've read everything I could find on this forum about the subject. Something important that comes out, is that the solver that should be used is the "fully coupled" one. My solution is however converging with a "segregated" solver, but not with this fully coupled one. My question is the following : Can I trust my results independently of the solver (seeing that when I refine the mesh, the results are pretty much the same) ?
Thank you for the help. Best regards,
Pascal
1 Reply Last Post Aug 11, 2012, 3:23 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
the difference is mainly in the speed and RAM use, a segregated solver usually uses less RAM but takes longer, while a Direct solver usually uses more RAM but can be faster.
Try it out: select the solver node (just above the segregated solver node) and add a direct solver node (right click), that will disable the segregated one, and resolve, and compare the results
If you do this in a new study sequence you can compare the two results and not overwrite ;)
--
Good luck
Ivar
the difference is mainly in the speed and RAM use, a segregated solver usually uses less RAM but takes longer, while a Direct solver usually uses more RAM but can be faster.
Try it out: select the solver node (just above the segregated solver node) and add a direct solver node (right click), that will disable the segregated one, and resolve, and compare the results
If you do this in a new study sequence you can compare the two results and not overwrite ;)
--
Good luck
Ivar