Double cantilever beam DCB FEM analysis on Abaqus with cohesive elements 1

[rosariocoppolino]

Hi,

i need to perform a double cantilever beam analysis on Abaqus to simulate the MODE I debonding in a carbon fibre composite. Anyone have some text about the procedure that i can follow on Abaqus?

The major problems for me are the definition of cohesive elements on abaqus and how i can connect this elements to the laminas.
Other issues are, for example, if this kind of analysis need a dynamic o static setup.

i attached my file with the setup that i've created.

I hope that anyone can help me!

Thans,

Rosario

 

[Erik Panos Kostson]

These tutorials might be relevant:

https://www.youtube.com/watch?v=uclQHMj2PZY

and

https://www.youtube.com/watch?v=thXyQq-BtTc

Hope this helps.

 

[Erik Panos Kostson]

I would also try to keep it simple. For instance use solid isotropic material and not any complex ones (e.g., laminate used only for shells, probably not for 3D parts)
Perhaps remove all contacts in the first instance, again trying to keep it simple. So in summary keep it simple and add complexity as you gain confidence.

 

[rosariocoppolino]

Thank you EriK,
I've follow your suggestion and perform the analysis with isotropic material.
i'will try to perform the composite material analysis but mi final target is the crash analysis of complex composite material structures. Do you know something on this kind of analysis? I've a good familiarity with static structural and dynamic problem but reach this target is very difficult, so i'm trying to define a route with modellation. Can you suggest me how i can proceed? Have you some source helpful for me?

Thanks,

Rosario

 

[Erik Panos Kostson]

You are more than welcome. Unfortunately though I have zero experience with crash analysis of complex composite structures. I am sure someone will have some feedback - perhaps there might be also some relevant material on the internet.

 

[Erik Panos Kostson]

Hope you have moved forward.

If you model this (battery and structure) like in the model you created (3D elements) then you will generate a very large amount of elements depending on the amount of plies you have and the extent of the structure.

It would probably be easier, if it is a global structure (many elements) that is relatively thin (not a foam/al. core type of composite), to start with conventional shell elements (and composite lay up definition) and then add an appropriate failure criteria (e.g., Tsai Wu,...), in order to model this. In abaqus it should be possible to model the progressive damage and failure of the composite structure (Properties: Damage for FRC-Hashin Damage+Damage Evolution).

So it could be worthwhile to start up with a simpler shell composite model, and see (compare to tests) if that is good enough in the first instance before one looks into more complicated 3D models with cohesive elements.

The next level (before 3D modelling) could also be to look into continuum shell elements (I have never used this), that could incorporate some cohesive failure for detailed analysis of delaminations.
Link