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.

3.5 and 4.1 solver difference?

Please login with a confirmed email address before reporting spam

I have a model, which can be solved in 4.1.

The same setting in 3.5, then the model doesn't converge.

I load the 3.5 version model in 4.1 and add a new study. It converge again.

Who knows the difference between the solvers in 3.5 and 4.1?

Thanks

2 Replies Last Post 9 mars 2012, 23:57 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 9 mars 2012, 21:32 UTC−5
which of the solvers you use? what are the settings? if you are using default, they might be different.
which of the solvers you use? what are the settings? if you are using default, they might be different.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 9 mars 2012, 23:57 UTC−5
You are right. The 3.5 is using the PARDISO, while 4.1 is using MUMPS.

I am coupling electrostatics, Nernst-Planck with Navier-Stokes equation.

When just with electrostatics and nernst-planck, 3.5 can give a correct answer.
Also with just using Navier-Stokes equation, 3.5 is fine too.

However, when I coupled them together, 3.5 didn't converge and said that the error is beyond tolerance.
You are right. The 3.5 is using the PARDISO, while 4.1 is using MUMPS. I am coupling electrostatics, Nernst-Planck with Navier-Stokes equation. When just with electrostatics and nernst-planck, 3.5 can give a correct answer. Also with just using Navier-Stokes equation, 3.5 is fine too. However, when I coupled them together, 3.5 didn't converge and said that the error is beyond tolerance.

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.