Physical, mechanical and rheological behaviour of prepreg tapes for part quality and deposition rate improvement in composites automated manufacture

Abstract

The generation of manufacture-induced defects, such as wrinkles formed during the tow steering process is one of the main drawbacks of automated fibre placement (AFP). It is a consequence of the mismatch between the inner and outer radius when a prepreg tape is deposited onto a curved surface or along a curved path. Some studies report that these defects can cause as much as 36% loss of the in-service product’s performance. Hence, they must be minimised or fully avoided. Traditional physical trials are costly and time-consuming. Numerical process simulations could help to drastically reduce part development costs through optimisation in the virtual space. However, the complexity of the tape behaviour (which is anisotropic as well as temperature and rate-dependent) and the dynamic interactions between the different constituents (the prepreg tape, the machine roller and the substrate on which the material is deposited) involved in the process make any attempt at predicting of AFP steering defect extremely challenging. Good understanding of the evolution of the tape behaviour with processing condition, adequate models able to represent the observed behaviour and a robust numerical platform (to be practical the run time of such a platform should be relatively short) for the process are needed to reach this goal.

Based on these considerations, the aim of this PhD thesis was to conduct a thorough study of thermosetting epoxy prepreg’s behaviour under processing condition. Two of the main deformation mechanisms (in-plane shear and tack), which determine wrinkle initiation were comprehensively studied. Recognising there are no standard methods and only a few studies available in the open literature on the behaviour thermosetting prepregs under processing condition, two novel characterisation methods were proposed to investigate in-plane shear and tack behaviour of prepreg under different combination of temperature and deformation rates. Corresponding numerical models were proposed based on the experimental results and were then implemented into an AFP steering model as user material subroutines for the Finite Element (FE) package Abaqus/Explicit. The comparison with experimentally collected AFP data highlight the good predictive capability of the proposed models. The work presented here brings the composite manufacturing community one step closer to virtual manufacturing of the AFP process.

Date of Award	2 Dec 2021
Original language	English
Awarding Institution	University of Bristol
Supervisor	Stephen R Hallett (Supervisor), Dmitry Ivanov (Supervisor), Jonathan P Belnoue (Supervisor) & James Kratz (Supervisor)