I'm currently working on low velocity impact and decided to investigate the influence of damage patch size with 3D elements and the rest of the lamina with conventional shells.
I'm using COH3D8 to simulate delamination and C3D8R for the composite but i'm always looking at not physical delamination at the patch boundaries.
How can i avoid this delamination?
Is it a problem related to the cohesive elements failure?
Hello Mr Ghesmati-Kucheki,
First of all, thank you for your answer. I'll share more details about the model i'm working on. I'm indeed using cohesive elements to simulate delamination using a traction separation based response with damage initiation of quadractic nominal stress criterion and BK for the damage evolution.
The results with merely 3D elements are as expected however the aim of my research is not to propose a patch repair using 3D elements but rather reduce the overall CPU time of a LVI by using 3D elements only in the damaged area and shell elements to simulate the undamaged area and the overall composite response. The general results using solid/shell abaqus coupling are good but i'm still looking at unrealistic delamination at the solid/shell interface.
For the interested one, i'm currently following an approach based on Sun and Hallet work "Barely Visible Impact Damage in Scaled Composite Laminates:
Experiments and Numerical Simulations" where i'm using a 3D purely elastic transitional interface with no damage criteria out of the centered one where i'm implementing a VUMAT based on Puck's criteria. Something similiar but not exactly as shown on the picture attached.
DB
Can you please share the vumat code. I am working on low velocity impact on composite laminates.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Inorganic Materials
- Metals
- Alloys
- Steel