Abstract:Accurate analysis of composite delamination using interface elements relies on having sufficient elements within a softening region known as the cohesive zone ahead of a crack tip. The present study highlights the limitations of existing formulae used to predict numerical cohesive zone length and demonstrates modifications necessary for improved accuracy. Clarification is also provided regarding the minimum number of interface elements within the cohesive zone. Finally, appropriate values of numerical interfacial strength are examined. The results presented will aid the application of mesh design techniques that both preserve numerical accuracy, whilst minimising computational expense. Where applicable, subscripts I, II and m are used to denote properties under mode I, mode II and mixed mode loading respectively. Keywords

A cohesive zone interface element degradation law is presented for analyzing delamination crack propagation under cyclic loading. Development of the law is based on a detailed study of the numerical cohesive zone and the extraction of strain energy release rate from this zone, enabling a direct link with experimental Paris Law data. The law is implemented using three dimensional interface elements within the explicit finite element code LS-Dyna. Validation is presented by way of application to composite material fatigue fracture toughness tests; Double Cantilever Beam for Mode I, End Notch Flexure for mode II and Mixed Mode Bending for the mixed mode case. In all cases good agreement with experimental data available in the open literature and/or theoretical solutions was obtained.

This paper presents a detailed study of the influence of maximum interfacial stress on interface element analyses for composites delamination. The development of the non-linear cohesive zone ahead of a crack tip is analysed with respect to length, stress distribution and mode ratio. The energy absorbed by interface elements is compared with the crack tip strain energy release rate from fracture mechanics analyses. These studies are performed initially on standard fracture toughness specimens, where mode-ratio is fixed by the applied displacement constraints. Results show close agreement with linear elastic fracture mechanics solutions. A simple ply drop specimen is then modelled, where the mode ratio is not constrained by the boundary conditions, and results are compared with the Virtual Crack Closure Technique. In this case maximum interfacial stress has a far greater influence on the numerical results, due to its significant influence on cohesive zone length, mode ratio and energy absorbed.

No abstract

No abstract