Learning Forum

[Yalr550]

Hi , 

I am performing a non linear bending test on a light weight steel (0.95mm thk) studs, using a two point load acturaor placed at equal spacing from the mid span. It is a combined stud (face to face channel), with batten plates between them to form the combined structural system.

I have used a displacement load of 30mm vertically, and 10 minimum number load steps. The model seems to converge several load steps, until the mesh tends to become distortional and the model can not converge. 

I have assinged the contacts between the battens to the studs flanges as a frictonless, load cells to stud web as rough. The screws between the batten and the plates are modelled and contact is assigned between those elements. 

 

 

Is there any way to make this converge, or if I have assigned something wrong in the boundary condition?

Attached the exported file (wbpz). 

https://uoa-my.sharepoint.com/:u:/g/personal/yalr550_uoa_auckland_ac_nz/EcuJMG-J-AFEmOlyBlgcbTABzXrsSgjEcds-i1_N8YlSlw?e=OpERS5

 

 

[peteroznewman]

Hi,

I opened your model and have a suggestion. All parts in the structure have a thin wall compared to the length and width. The best practice to simulate this type of structure is to use SpaceClaim or your CAD system to replace every solid body with a midsurface. Mechanical will mesh with shell elements instead of solid elements. This type of mesh will have a much easier time converging and provide a much more accurate evaluation of stress in the structure.

Good Luck!
Peter