Prediction of fatigue limit for unidirectional carbon fibre/epoxy composites

Bent F Sørensen,Stergios Goutianos

doi:10.1088/1757-899x/388/1/012017

Bent F Sørensen, Stergios Goutianos

Open Access

https://doi.org/10.1088/1757-899x/388/1/012017

Copy DOI

Abstract

A micromechanics model is used for the prediction of the fatigue limit of unidirectional carbon fibre/epoxy composite materials. The model is based on the hypothesis that failure of a fibre will result in fibre/matrix debonding of the broken fibre. The associated debond crack tip stress fields will raise the stress in the neighbour fibres as the debond crack tips move along the broken fibre and can thus cause failure of the neighbouring fibres. The fatigue limit is defined from the maximum applied cyclic stress that does not induce failure of any neighbour fibres. Effects of microscale mechanical properties are investigated. The model predicts that the fatigue limit, expressed in terms of stress, increases with fibre volume fraction until 50-60 %, whereafter the fatigue limit decreases with increasing fibre volume fraction. With other parameters held fixed, the fatigue limit increases with increasing interfacial frictional sliding shear stress and with decreasing interfacial fracture energy.

Highlights

Fatigue - material degradation under cyclic loading - is a phenomenon that greatly influences the design limits of composite materials for load-carrying structures
The model predicts that the fatigue limit, expressed in terms of stress, increases with fibre volume fraction until 50-60 %, whereafter the fatigue limit decreases with increasing fibre volume fraction
The parameter study included in this paper suggests that the fatigue limit of unidirectional carbon fibre/epoxy composites can be raised by microstructural optimization

Summary

Introduction

Fatigue - material degradation under cyclic loading - is a phenomenon that greatly influences the design limits of composite materials for load-carrying structures. With a rotation time of about 5 seconds per rotation, 25 years of constant use corresponds to more than 150 million rotations For such high number of load cycles, the use of traditional S-N data may not be attractive, since it is very time consuming and expensive to conduct cyclic experiments and establish lifetimes up to 150 million cycles (or higher). It may be more appropriate to focus on which conditions the material might possess a fatigue limits, i.e., whether a maximum load (or strain) value exists, below which the material would sustain an infinite number of load cycles

Results

Discussion

Conclusion