Host Institution: The University of Nottingham
Start Date: 1st April, 2019
Duration: 6 months
Lead Investigator: Shuai Chen
Tubular composite components are fundamental elements of complex composite structures, and are widely used in industry, such as aircraft keel beams, helicopter frames and automotive chassis frames. Numerous manufacturing techniques have been developed to produce composite tubes, including filament winding, pultrusion and braiding. The challenge is how to make an axially curved tubular preform with a variable section shape, and in many cases this requires a highly labour-intensive fabrication using prepreg mouldings and adhesive bonding. Although braiding can currently be employed to directly produce complex 3D tubular components, this is based on using individual mandrels with robotic control. In this process, the material deposition rate can be high, but process time is significantly affected by the time taken to start and finish the braid on each component and by variations in the local geometry, which slows the axial feed rate. It is far more efficient to produce straight tubes of constant cross-section. The Feasibility study introduces a method for forming complex curved beams with variable cross-section from continuous braided feedstock – offering a step-change in manufacturing rate. This requires the development of a predictive model for forming of tubular composite preforms and an understanding of the applicable geometric parameters of the braided feedstock and the supporting deformable core. This technique extends the capability of making continuous, jointless composite frames for improving structural integrity and overcomes constraints imposed by using straight tubular parts. The introduction of tube forming brings to light an affordable and fully-automated process with high-volume production to make high-quality tubular composite structures. The novelty in this study is the development of a numerical modelling tool for pre-forming tubular structures, requiring innovative steps in fabric forming simulation where existing models can only address fabric sheet forming issues. Essential to the development of the manufacturing process, the model will be used to inform the range of workable process parameters for these complex anisotropic materials such as maximum change in cross-section and shape and minimum bend radius.
In this research, braiding is employed to produce a basic prototype of the tubular preform followed by an automated forming process to finalise the geometry of the complex preform as shown in Figure 1. Prior to forming, a pre-braided sock is slipped over a crushable core or an inflatable bladder in a standard format. A set of matched tools is employed to form the preform the assembly into a designated shape. During forming, the inflatable bladder can be used to conform the preform to the cavity surface by pressurising inside the tube. Alternatively, a crushable core can be designed to resist unwanted deformation. Finally, the curvature of the axis and the variation of section shape are realised from this forming process.
The aim of this research is to develop a digital twin to assist with process design and optimisation for defect-free fabric preforms by stamping braided sleeves into desired complex configurations, enabling the proposed step change. The main objectives are (1) to develop an explicit FE model of the braid process, (2) to develop an explicit FE model of the forming process, and (3) to understand the primary factors of producing a defect-free component. This research fits within the Hub priority areas of both “High rate deposition and rapid processing technologies” and “Design for manufacture via validated simulation”.