Reducing setup costs: Tooling force prediction in resin transfer moulding (RTM) and compression RTM

Abstract

Mould tools used for Liquid Composite Moulding (LCM) processes such as Resin Transfer Moulding (RTM) and Compression RTM (CRTM) must withstand significant forces generated by the fluid and the fibrous reinforcement. Prediction of these forces will allow for optimizations in setup costs and time, and maximize the usage of the capabilities of peripheral equipment (such as presses). SimLCM is being developed at the University of Auckland as a generic LCM simulation packaged. It has the capability to predict clamping forces and stress distributions acting on mould tools during complete moulding cycles. Both mixed-elastic and viscoelastic reinforcement compaction models are implemented within the package. A series of experiments, both planar (circular flat plate) and non-planar (truncated pyramid), have been undertaken to validate predictions made using SimLCM. Consideration of both the normal and shear components of the fibre preform compaction stress and the internally generated fluid pressure is required for non-planar geometries. This is especially important to extend the capabilities of SimLCM from rigid tool processes (such as RTM and CRTM) to include flexible tool processes such as RTM Light and Vacuum Assisted RTM (VARTM). A friction-based model is used to account for the shear component of compaction stress. In general, predictions for the planar cases are very good, with the viscoelastic model providing significant improvement over the mixed-elastic model during stress relaxation phases. However, for the non-planar cases presented, the peak force is under predicted during preform compaction, and over predicted during the fluid compression phase of CRTM. The under prediction is potentially due to irregularities with the preforms for the non-planar geometry, and is the subject of ongoing research. The over prediction of the force during the fluid compression phase of CRTM is primarily due to internally generated fluid pressure during fluid compression as a result of significant race tracking during the fluid filling phase This is not well-modeled currently by SimLCM and is the subject of ongoing work.