AnsweredAssumed Answered

"Material not defined for one or more shells" problem

Question asked by Unspecified Unspecified on Feb 26, 2013
Latest reply on Mar 1, 2013 by Mikael Martinsson

Like • Show 0 Likes0
Comment • 5

As stated in the title i have a problem with static simulation in SolidWorks 2011. I'm doing this for an assignement (I'm a student). This is a frame for an agricultural device, made out of two rectangular and two circular beams and sheet metal. I made midsurfaces for all the parts and deleted the bodys. In the Study i defined shells from midsurfaces. I am able to mesh the assembly, i also defined the material (alloy steel) but when i try to run a basic load and fixture example it plots a warning: "Material not specified to one or more shells". I have no idea what is the case with this as all the parts have material defined in the simulation tree.

I also attached the assembly in this post. I hope someone has the solution because this is pretty important to me.

Attachments

okvir_mid_surf.zip908.7 KB

No one else has this question

Outcomes

Mikael Martinsson Feb 27, 2013 3:26 AM

Hi.

I got it to run when I switched to standard mesh, global size 50 mm.

But you have more work to do to run this problem. Surfaces have gaps between them and some also interfere. Contact conditions need to be added manually. (see results below)

Actions

Unspecified Unspecified @ Mikael Martinsson on Feb 27, 2013 9:26 AM

Thank you very very much Mikael. I tried the simulation again with removed beams (only the middle part of the frame remaining). I defined contasts manually but still, the top two parts with three holes each penetrate the top shell (as seen on the picture bellow). Would anyone know the solution for this?.

To answer your question about the gaps: I believe you mean the gaps between the shells and rectangular beams. They are there because the shell of the rectangular beam itself is generated using midsurface of the beam and is therefore narower and lower then the actual beam (by 2,5 mm in each direction). I can alter the measurements of the holes for beams and get the edge to surface contact, i was just worried i would change my frame too much. I also read on Solidworks help page that you can mesh and simulate shells with gaps inbetween and the program would somehow still bond them together?

The other question i have is: i used midsurface generated shells for this simulation. How important is shell thickness and how do this shells represent the cross section

I posted the simulation results bellow and also the new file.

characteristics of sheet metal. Orgininal thickness of metal is 15mm.

Attachments

okvir_mid_surf_brez_cevi.zip3.6 MB

Actions

Mikael Martinsson @ Unspecified Unspecified on Feb 27, 2013 3:06 PM

Hi.

To my knowledge you can not have gaps between shell elements and still get bonded contact from the global contact condition.

This is what I would do (and did in attached file)

Save the assembly as a new part file, with the selection "all components". Now you have a single part with all surfaces and solids.

Extend, trim and knit surfaces togehter so they are in contact (close all gaps from mid surfaces)

Then set up and run your simulation. With good contact between surfaces, the global contact condition will work.

Here is the displacement result (exaggregated).

I included the 4 solid parts in the simulation. Thickness of shells 15 mm and alloy steel all over.

My file is attached to this message.

Deformations are small, but the load is only 100 N/m^2 in your example.

Attachments

okvir_mid_surf FEM Model.zip322.5 KB

Actions

Unspecified Unspecified @ Mikael Martinsson on Feb 28, 2013 3:54 PM

Mikael i owe a huge beer and you also get 100000000 internet points .

I am repairing the model and will try all of your advices. Sadly i cannot open the file you send me because I'm using the 2011 version. But it doesn0t matter i think i know what i did wrong.

Thank you very much again.

Actions

Mikael Martinsson @ Unspecified Unspecified on Mar 1, 2013 12:47 AM

Hi.

Ok, didn't notice the "old version" message when I saved the file.

Attached you'll find a parasolid v15 with the correct surfaces and solids that should work in 2011.

Attachments

okvir_mid_surf FEM Model.zip42.2 KB

Actions

5 Replies

Mikael Martinsson Feb 27, 2013 3:26 AM
Mark Correct
Correct Answer
Hi.
I got it to run when I switched to standard mesh, global size 50 mm.

But you have more work to do to run this problem. Surfaces have gaps between them and some also interfere. Contact conditions need to be added manually. (see results below)
Like • Show 0 Likes0
Actions
- Unspecified Unspecified @ Mikael Martinsson on Feb 27, 2013 9:26 AM
  Mark Correct
  Correct Answer
  Thank you very very much Mikael. I tried the simulation again with removed beams (only the middle part of the frame remaining). I defined contasts manually but still, the top two parts with three holes each penetrate the top shell (as seen on the picture bellow). Would anyone know the solution for this?.
  
  To answer your question about the gaps: I believe you mean the gaps between the shells and rectangular beams. They are there because the shell of the rectangular beam itself is generated using midsurface of the beam and is therefore narower and lower then the actual beam (by 2,5 mm in each direction). I can alter the measurements of the holes for beams and get the edge to surface contact, i was just worried i would change my frame too much. I also read on Solidworks help page that you can mesh and simulate shells with gaps inbetween and the program would somehow still bond them together?
  
  The other question i have is: i used midsurface generated shells for this simulation. How important is shell thickness and how do this shells represent the cross section
  
  I posted the simulation results bellow and also the new file.
  characteristics of sheet metal. Orgininal thickness of metal is 15mm.
  Attachments
  - okvir_mid_surf_brez_cevi.zip3.6 MB
  Like • Show 0 Likes0
  Actions
  - Mikael Martinsson @ Unspecified Unspecified on Feb 27, 2013 3:06 PM
    Mark Correct
    Correct Answer
    Hi.
    To my knowledge you can not have gaps between shell elements and still get bonded contact from the global contact condition.
    This is what I would do (and did in attached file)
    
    Save the assembly as a new part file, with the selection "all components". Now you have a single part with all surfaces and solids.
    Extend, trim and knit surfaces togehter so they are in contact (close all gaps from mid surfaces)
    Then set up and run your simulation. With good contact between surfaces, the global contact condition will work.
    
    Here is the displacement result (exaggregated).
    I included the 4 solid parts in the simulation. Thickness of shells 15 mm and alloy steel all over.
    
    My file is attached to this message.
    Deformations are small, but the load is only 100 N/m^2 in your example.
    Attachments
    okvir_mid_surf FEM Model.zip322.5 KB
    Like • Show 0 Likes0
    Actions
    - Unspecified Unspecified @ Mikael Martinsson on Feb 28, 2013 3:54 PM
      Mark Correct
      Correct Answer
      Mikael i owe a huge beer and you also get 100000000 internet points .
      I am repairing the model and will try all of your advices. Sadly i cannot open the file you send me because I'm using the 2011 version. But it doesn0t matter i think i know what i did wrong.
      
      Thank you very much again.
      Like • Show 0 Likes0
      Actions
      - Mikael Martinsson @ Unspecified Unspecified on Mar 1, 2013 12:47 AM
        Mark Correct
        Correct Answer
        Hi.
        Ok, didn't notice the "old version" message when I saved the file.
        Attached you'll find a parasolid v15 with the correct surfaces and solids that should work in 2011.
        Attachments
        okvir_mid_surf FEM Model.zip42.2 KB
        Like • Show 0 Likes0
        Actions

"Material not defined for one or more shells" problem

Attachments

Outcomes

Attachments

Attachments

Attachments

Related Content

Recommended Content