I am trying to analyze compression after impact (CAI) effect on laminated composites. Each lamina is modeled as 3D shell element and interface by cohesive surface interaction. The delamination analysis is working well in impact step. But in the compression step (bottom edge fixed, sides simply supported, top edge displacement loading), response is very strange and unexpected where the laminates tend to buckle immediately even under 0.1% strain.
I ran similar analysis in general - static nonlinear, general- static riks and it works without problem. But I cannot use this since abaqus standard did not allow contact definition with damage parameters. I ran similar analysis in explicit but with 3D elements and it runs fine. I cannot use this one either because it is extremely slow. The contact definition, boundary conditions must be correct.
Is this type of analysis not permitted with 3D shell elements in explicit?
Hello Shreyas,
From a paper I have read that it is possible to apply cohesive elements between shell elements, this paper is: "Davila, Camanho, Turon - Effective simulation of delamination in aeronautical structures using shells and cohesive elements". The paper describes how they implement it in Abaqus. However I did not find information on how to implement cohesive surface interaction between shells.
Currently I'm choosing which method to use to model delaminations in laminated composites for my MSc thesis. And I was wondering why you are choosing to use cohesive surface interaction instead of cohesive elements, do you have a specific reason for that ?
I'm tending to choose cohesive elements, mainly because it seems there is more information on how to implement it in literature.
Kind Regards,
Michaël
Hi Michael
I have read the paper that you mentioned. It is very useful one. I am choosing cohesive surface because it seems to run faster than elements. My analysis worked fine with cohesive surfaces as well as elements for impact analysis but I am getting this strange problem for compression.
I found that another student reported exact same problem on some forums 4 years ago (http://imechanica.org/node/10116). But I did not find the answer for this problem.
To simplify, I just ran the analysis with isotropic material and only tie constraint and it still doesnt work. So it seems that the problem is not really cohesive interaction but some inherent problem with defining any sort of interaction/ constraint with shell elements in abaqus explicit. The analysis works perfectly fine with 3D elements or implicit/ static analysis. But I have to run this with shell elements in explicit because I am simulating compression after impact testing.
Let me know if you come across such problem before.
Thank You
yes, sure you can use cohesive elements between structural or shell elements. add your error or whatever here to be able to understand the problem better. I have used cohesive elements and you need to be more sensational working with them.
Hi Shahram
The analysis runs without any errors but the response is very unexpected. To simplify the problem, now I am running only tie constraint (no cohesive elements, no cohesive surface, no contact definition of any sort) with isotropic laminate. The response with explicit solver is completely different than response with implicit solver. But I have to solve this using explicit. I have attached all input files. Let me know if you have suggestions.
Thank You
- Badges
- Science topic