COMPOSITES 2023

Off-axis Strength Predictions for Multidirectional Thin-ply Laminates

  • Mitrou, Anatoli (University of Porto - FEUP)
  • Arteiro, Albertino (University of Porto - FEUP)
  • Reinoso, Jose (University of Sevilla - ETSI)
  • Camanho, Pedro (University of Porto - FEUP)

Please login to view abstract download link

Advances in aeronautical design have increased the use of composite laminates in primary and secondary structures. Thus, the ability to simulate their response and obtain accurate strength predictions, with emphasis being given to notched components that often appear in aeronautical structures (e.g., bolted, riveted parts), is of prominent interest. In this work, focus is attributed to a novel composite system, thin-ply laminates (i.e., plies of thicknesses under 0.1 mm) that have been shown to behave differently during failure when compared to “standard” ones (i.e., made up of plies of thicknesses over 0.1 mm). In the former case, final failure tends to happen in the form of a single fracture plane. From a modelling perspective this allows the use of novel techniques formulated for brittle fracture, such as the phase field method (PF) and Finite fracture mechanics (FFM), to provide an efficient modelling basis following an equivalent single layer (ESL) representation of the composite laminated plate. The feasibility of the application of these methods is evaluated on the basis of off-axis (i.e., referring to loading on a direction that does not coincide with one of the principal axes of orthotropy of the plate) open-hole tension of a multidirectional laminate based on the experimental results of [1]. Results obtained using the anisotropic PF model of [2], which uses a 2nd order structural tensor to account for anisotropic fracture energy, reformulated as in [3] to include specific considerations of the toughness of a composite laminate, are initially presented. A successful prediction of the experimental results both with regards to fracture plane and strength, with a maximum observed error in predicted strength of 4.8%, is achieved. However, implementation does come with its limitations, namely specific considerations regarding the level of anisotropy alongside the definition of numerical parameters, such as the length scale and finite element size, assuming an unknown a priori crack plane, and resulting computational cost of the numerical implementation.