Demonstration of preventing catastrophic compressive failure of carbon fibre-based composite components under 3-point bending through lay-up configuration control.
     Topic(s) : Material and Structural Behavior - Simulation & Testing

    Co-authors​ :

     Ellis HILL (UNITED KINGDOM), Meisam JALALVAND , Andrew HAMILTON  

    Abstract :
    Catastrophic compressive failure in carbon fibre-based composite materials is a common and undesirable outcome in many applications. This research demonstrates the opportunity to mitigate this outcome at the component level through lay-up configuration control. The literature has highlighted methods of preventing unwanted catastrophic failure at the coupon level through hybridisation and use of thin-ply composite prepreg. However, the higher component level is yet to be explored for such opportunities. A further necessity for research is due to the failure mechanisms at the component level occasionally differing from those employed for gradual failure outcomes in coupon specimens (e.g. closed cross-sections preventing delamination due to the lack of free edges). To address this, simple composite tubular structures were tested to study different failure mechanisms that may be utilised for more gradual failure outcomes at this level when loaded under 3-point bending. The experimental conditions and specimen geometry were chosen to represent common structures found in typical carbon fibre-based composite applications (e.g. bicycle handlebar). The first specimen group to be tested represented the common configuration of choice for such applications; whereby, the lay-up consisted of multiple plies of unidirectional carbon fibre prepreg, sandwiched between two woven weave plies. The outcome of this configuration was catastrophic compressive failure near the region under external loading, resulting in the complete loss of its load-carrying capabilities. The second group's configuration was designed using simple Euler-Bernoulli beam theory and finite element simulation to initiate shear failure instead. The concluded lay-up consisted of only one woven weave ply at the inner-most layer, surrounded by multiple unidirectional plies at the outer-most layers with fibres directed along the length of the tube. Both groups were of equivalent cross-sectional area. The outcome for this case was successful in achieving shear failure, resulting in matrix cracking along only half the specimen’s length at its mid-plane. This allowed for the materials' fibres to remain largely intact, providing further reinforcement post-initial failure. Specifically, at first-ply failure, the load dropped by a maximum of 80% across all specimens within this group. After this, the specimens continued to increase their load-carrying capabilities to hold more than 40% of their maximum load. The final failure was a combined outcome of multiple fractures in different locations which were observed using the results of X-Ray CT scans. Furthermore, there appears to be no trade-off between the stiffness and strength of the specimens with the non-linear failure performance. The newly proposed configuration for inducing shear failure showed no significant change in strength, an increase in stiffness by 14% and an increase in the total energy absorbed during loading by up to 48%.