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.

Singular matrix error

Please login with a confirmed email address before reporting spam

I have a complicated model I am trying to import from a Cad file. I kept getting errors with the model and I have distilled the problem to a local area of the model. I made a sub model which is attached.

This model does not run properly. Can anyone tell me why? Where is the problem with the geometry?

The model is a surface model, with shell elements.


2 Replies Last Post 19.07.2010, 09:16 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17.07.2010, 17:51 GMT-4
Hi

I do not know why but I removed and replaced your fixed and edge load, then remeshed it finer and in my 4.0a it solved OK

I then added the fins and set "form union" and it solved again (had some convergence warnings though)

It could be that edge load and fixing are limit on half circles, some points have very "tangeant loads" that could lead to singularities, one should ensure that the BC are complete w.r.t shell needs you have 6 variables

Good luck
Ivar
Hi I do not know why but I removed and replaced your fixed and edge load, then remeshed it finer and in my 4.0a it solved OK I then added the fins and set "form union" and it solved again (had some convergence warnings though) It could be that edge load and fixing are limit on half circles, some points have very "tangeant loads" that could lead to singularities, one should ensure that the BC are complete w.r.t shell needs you have 6 variables Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 19.07.2010, 09:16 GMT-4
Ivar

Thanks for taking the time to look at this.

While I appreciate that you can solve this problem by changing BCs and load points, and by changing the mesh size, the problem can be solved.

However there is a fundamental problem here in that
1) COMSOL provides no useful help
2) COMSOL provides no useful error messages
3) the occurance of this issue appears random and occurs in other shell element models I am working with.

Does this problem have a physical basis in the geometry definition?
Is this related to some unusal formulation of the shell element (Is it a 6 DOF shell or a 5 DOF shell?)
Am I just being stupid (that's usally the correct answer)
Ivar Thanks for taking the time to look at this. While I appreciate that you can solve this problem by changing BCs and load points, and by changing the mesh size, the problem can be solved. However there is a fundamental problem here in that 1) COMSOL provides no useful help 2) COMSOL provides no useful error messages 3) the occurance of this issue appears random and occurs in other shell element models I am working with. Does this problem have a physical basis in the geometry definition? Is this related to some unusal formulation of the shell element (Is it a 6 DOF shell or a 5 DOF shell?) Am I just being stupid (that's usally the correct answer)

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.