Notifications
Clear all

To model fracture

4 Posts
2 Users
0 Reactions
1,678 Views
Massoud
Posts: 23
Topic starter
(@mass)
Paid Intern
Joined: 4 years ago

Below You can see my structure which has been cut its 1/4 to only see inside it. It is exerted axial compression force to crush.

I have to have fracture at the connecting of rings with inner and outer tubes due to shear and/or tensile stress( Fig 2) . What does come to your minds to model this fracture?

 

My approach is using Contact_TIED_SURFACE_TO_SURFACE_FAILURE according the Fig3 and Fig4.

Am I on the right track?

forum
3 Replies




Negative Volume
Posts: 665
Admin
(@negativevolume)
CEO
Joined: 6 years ago

Hi @mass

You won't be able to use that card as those mesh connections share nodes currently. You could do what you are proposing by splitting off the connecting rings into their own part and then disconnecting them from the outer and inner rings. If you do this, you can also try *Contact_automatic_surface_to_surface_tiebreak. However, both of these contact based failure methods will only impact the interface between the outer and inner connection points, not the material itself.

If you wanted to define failure in the material as a whole, you could use *Mat_Add_Erosion, *Mat_Add_Damage_Gissmo, or *Mat_Add_Damage_Diem. Each of these would require you having a critical strain value/curve for failure, not stress. 

Reply
2 Replies
Massoud
(@mass)
Joined: 4 years ago

Paid Intern
Posts: 23

Thank you @negativevolume. I know what you mean. I don't have shared nodes. Actually after design structure I copy elements of connecting rings to the another part then I move twin tubes to this part; so I have different coincident nodes (picture below).

But new news is that my supervisor changed his mind suddenly. He asked me to use shell element instead of solid. So I am doing all things from the beginning. Will I have the same story with shell element you think?

 Screenshot 2
Reply
Negative Volume
Admin
(@negativevolume)
Joined: 6 years ago

CEO
Posts: 665

@mass yeah it should be the same with shell elements. To make this into a model with 2D elements, you could find the faces of the current mesh, which would give you the outer shell mesh, and then you could delete the new side 2D elements and pick either the inner or outer surface elements to offset half the thickness to get a mid plane mesh. You will just need to double check everything at the end against the current 3D mesh to ensure that the 2D mesh is in the mid plane.

Let me know if you more explanation.

Reply







Share: