Progressive damage analysis in an open hole compression of FRP laminates including fiber kinking and pre-existing damage. (June 2023)

• Record Type: Journal Article
• Authors: Divse, Vishwas; Marla, Deepak; Joshi, Suhas S.
• Is Part Of: Composites. Volume 169(2023)
• Publication Date: 2023-06
• Subjects: Progressive damage -- Fiber kinking -- Open hole compression -- Pre-existing damage; Composite materials -- Periodicals; Manufacturing processes -- Periodicals; Composite materials; Manufacturing processes; Periodicals; 620.11805
• DOI: 10.1016/j.compositesa.2023.107523
• Languages: English
• ISSNs: 1359-835X

Abstract

Abstract: Progressive damage and strength analysis in an open hole compression (OHC) test is crucial in designing components made of fiber reinforced plastics (FRPs) composites. In this work, a continuum damage mechanics based 3D progressive damage model (PDM) incorporating LaRC05 failure criteria is developed. It considers identification of fracture and kink-band plane, shear non-linearity, in-situ strengths, and mixed-mode fracture in the formulation. The proposed PDM is validated through single element tests and applied to analyze the damage patterns and strength during open hole compression (OHC) tests of three different layups with two different hole sizes and a pre-existing damage. It is observed that fiber kinking in 0 ∘ plies initiates at the location of maximum in-plane shear stress and propagates due to a combined action of longitudinal compression and in-plane shear stress. The final failure of 0 ∘ dominant and quasi-isotropic layup is governed by fiber kinking in 0 ∘ plies, whereas matrix cracking in ± 4 5 ∘ plies is responsible for the final failure of the shear dominant layup. Furthermore, OHC strength of the layups reduces by 15%–20%, when the hole size increases from 6.35 mm to 9.525 mm. In the presence of a pre-existing kink or matrix damage, fiber kinking initiates at locations of the pre-existing damage. A small pre-existing damage reduces OHC strength of the layups up to 7%–11%. The results predicted by the model match well with the experimental results in the literature.