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.

How to model shell element tied together with solid element?

Please login with a confirmed email address before reporting spam

Hello!

I have a problem regarding how to model the solid elements attached with shell elements on the surface.
For example, a cubic modeled with solid element, fixed left boundary, add face loading on the right boundary,
there is a shell layer attached to the right surface of the cubic, for the shell element, fix four edges on the right surface.

Seems it works if I apply some loading on the solid element surface, since the shell elements and solid elements share same nodes, the shell elements will deform with the solid elements.

But the result seems little bit strange, the deformation, the deformation is rather jaggled on the boundary no matter how fine mesh I have.

Attached the model, could anyone give some suggestions? Thanks a lot.
BR,
Li


4 Replies Last Post 29.09.2010, 07:23 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 10.09.2010, 05:12 GMT-4
Hi

I have no time to llok at your model just now, but do not forget that shell elements have also free rotation DoF's but solid not, so you must link these correctly when you mix solid and shell. You have the new (not yet documened) Rigid connector BC to play with too, or use the antisymmetric strain gradients for teh angles thx, thy, thz

--
Good luck
Ivar
Hi I have no time to llok at your model just now, but do not forget that shell elements have also free rotation DoF's but solid not, so you must link these correctly when you mix solid and shell. You have the new (not yet documened) Rigid connector BC to play with too, or use the antisymmetric strain gradients for teh angles thx, thy, thz -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 10.09.2010, 05:22 GMT-4
Hi Ivar,
Thanks for your answer. I noticed that you have some post of a model about coupling shell and solid element.
That's a very good example. But my problem seems little different with that one, this model, the all the shell elements are attached to the solid elements, it's like an out layer on some solid.
Do you have some example models or documents on this.
Thanks a lot for your help.
/Xiaogai
Hi Ivar, Thanks for your answer. I noticed that you have some post of a model about coupling shell and solid element. That's a very good example. But my problem seems little different with that one, this model, the all the shell elements are attached to the solid elements, it's like an out layer on some solid. Do you have some example models or documents on this. Thanks a lot for your help. /Xiaogai

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 10.09.2010, 05:44 GMT-4
Hi

Well I do not see what is bizarre with your model.

If you take away the graphics refinement and set to "1" and not 4th order, and then use wireframe mesh to see the mesh, you only notice that the shell, being fixed along the external boundary (fixed means absolute fixed, not (only) attached to the bulk model as a rogh shape. Refine your mesh and the roughness will gently decrease, no ?

--
Good luck
Ivar
Hi Well I do not see what is bizarre with your model. If you take away the graphics refinement and set to "1" and not 4th order, and then use wireframe mesh to see the mesh, you only notice that the shell, being fixed along the external boundary (fixed means absolute fixed, not (only) attached to the bulk model as a rogh shape. Refine your mesh and the roughness will gently decrease, no ? -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29.09.2010, 07:23 GMT-4
Hi Ivar!

Thanks a lot for your answer. You were right.
But now I have a new problem regarding shell and solid modeling.

As we can see that in comsol version 4.0a, the shell element is Mindlin Reissner shell element with order from 1st order to 2nd order element.

In the model, I use 3rd order for solid element , 2nd order for shell element. Then I will get a "singularity" problem in the model. But if I tried to change shell element to 1st order, then it works. But it looks strange on the boundary.

My question is, why I got singularity problem if I use 3rd order solid, 2nd order shell,
but it did work if I 3rd order solid, 1st order shell element model?

Is this because of the extra nodes for 2nd order shell doesn't share same with 3rd order solid, while 1st order shell always have same node with 3rd order element?
I attached the model in case I didn't express myself clearly.

Thanks a lot.
Xiaogai
Hi Ivar! Thanks a lot for your answer. You were right. But now I have a new problem regarding shell and solid modeling. As we can see that in comsol version 4.0a, the shell element is Mindlin Reissner shell element with order from 1st order to 2nd order element. In the model, I use 3rd order for solid element , 2nd order for shell element. Then I will get a "singularity" problem in the model. But if I tried to change shell element to 1st order, then it works. But it looks strange on the boundary. My question is, why I got singularity problem if I use 3rd order solid, 2nd order shell, but it did work if I 3rd order solid, 1st order shell element model? Is this because of the extra nodes for 2nd order shell doesn't share same with 3rd order solid, while 1st order shell always have same node with 3rd order element? I attached the model in case I didn't express myself clearly. Thanks a lot. Xiaogai

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.