Simplified model for the tool-part interaction in spring-in of L-shape composite laminates

Abstract

Manufacturing of fibre-reinforced composites is often accompanied by process-induced distortions, primarily due to the anisotropy of the composites constituents (fibres and matrix), their thermo-chemical interactions, and the interaction between the composite and the tooling. Numerical models that account for these factors require extensive experimental material and process characterisation programmes before the models can be effectively used at the design stage. This paper presents experimental measurements of spring-in angles of L-shape IM7/8552 laminates cured on an aluminium mould. The curing process for L-shape laminates was simulated using the Cure Hardening Instantaneously Linear Elastic model. Tool-part interaction was characterised by fitting an analytical model to experimental measurements of warpage of flat laminates and modelled using boundary conditions designed to avoid the need for explicit modelling of the tooling. The spring-in angles predicted by the proposed simulation framework were within $0.5^{\circ }$ of the experimental results for the range of geometries considered. The simulations provided insights into the effects of specimen design (corner radius, flange length, and lay-up) as well as tool-part interaction on the total spring-in angle. It was shown that tool-part interaction significantly contributes to the spring-in angle, particularly in specimens with larger flange lengths.