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.

Comsol 4.3b vs 5.3

Please login with a confirmed email address before reporting spam

I have recently updated comsol to 5.3 version, and since I have already some results obtained using 4.3b I tried to re-built the same model using 5.3 and re-run the simulation. It is so weird that I have got different résults known that I haven't changed anything. Does anyone have any clue what was the reason ?


1 Reply Last Post Oct 1, 2017, 11:33 p.m. EDT
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 6 years ago Oct 1, 2017, 11:33 p.m. EDT
Updated: 6 years ago Oct 1, 2017, 11:35 p.m. EDT

There are a large number of possible reasons. You should, of course, check to see that in your new version, you have represented the problem in the same way (and that it was not mangled during input or conversion). That said, all finite element models are only approximations to reality. Version 5.3 and 4.3b differ in some ways that (obviously) can impact the nature and accuracy of those approximations. But probably more important is how you (not Comsol) actually set up the problem. Your particular set-up may be more valid (i.e., a better approximation) in one version than the other. Without any additional info, I would trust the newer version a bit more... but only a little bit more! And if there are big differences in results, then your problem set-up is probably a rather poor one, in either one or both versions of the code. It could be that you made poor choices in meshing, solver settings, boundary conditions, constraints, etc. So, if having an accurate solution to your problem is of some importance, you will need to study it carefully, and to not trust either model too much, until you better understand what is going on. Good luck.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
There are a large number of possible reasons. You should, of course, check to see that in your new version, you have represented the problem in the same way (and that it was not mangled during input or conversion). That said, all finite element models are only approximations to reality. Version 5.3 and 4.3b differ in some ways that (obviously) can impact the nature and accuracy of those approximations. But probably more important is how you (not Comsol) actually set up the problem. Your particular set-up may be more valid (i.e., a better approximation) in one version than the other. Without any additional info, I would trust the newer version a bit more... but only a little bit more! And if there are big differences in results, then your problem set-up is probably a rather poor one, in either one or both versions of the code. It could be that you made poor choices in meshing, solver settings, boundary conditions, constraints, etc. So, if having an accurate solution to your problem is of some importance, you will need to study it carefully, and to not trust either model too much, until you better understand what is going on. Good luck.

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.