Fabric compaction and fibre volume fraction evaluation for vacuum-assisted resin infusion modelling
Topic(s) : Manufacturing
Co-authors :
Anastasiia LARIONOVA (DENMARK), Philipp Ulrich HASELBACH (DENMARK), Robert PIERCE (DENMARK)Abstract :
In the wind turbine industry, Vacuum Assisted Resin Infusion (VARI) processes are applied in blade manufacturing to impregnate the dry reinforcement fabrics with liquid resin. In addition to the type of matrix and reinforcement used, the composite material and its properties are defined by the arrangement of the reinforcement (stacking sequence and ply orientation) and its fibre content, often referred to as the fibre volume fraction. The fibre volume fraction is a crucial parameter determining the mechanical properties of the composite, influencing factors such as strength, stiffness, and durability. During manufacturing, the applied out-of-plane force on the reinforcement fabrics produces a compaction effect that can help to reduce voids and optimise the fibre volume fraction. Although the fibre volume fraction can be increased with an increase in pressure, the permeability (resin flow potential) of the compressed fabrics will subsequently decrease, which can result in dry spots and poor manufacturing quality.
Thus, achieving optimal compaction and accurately determining the fibre volume fraction are critical steps in manufacturing fibre-reinforced composite materials. This also becomes crucial for reliable numerical predictions of the filling time and behaviour.
This work aims to investigate different techniques for measuring bag height, estimating fibre volume fraction during manufacturing and quantifying their reliability against traditional destructive characterisation methods. Specifically, a stereo Digital Image Correlation (DIC) system, a blue-light scanner and a laser sensor have been applied simultaneously for a series of repeated small laminate infusions consisting of five layers of biaxial glass fabric.
Height measurements to determine the compression behaviour were obtained before, during and after the VARI process, and their results were compared. Additionally, post-mortem burn-off tests and X-ray scanning were used to validate the fibre volume fraction estimates from beg height measurements.
The gathered experimental data allowed for the development of material models describing the compressed state of the reinforcement layup before infusion and to predict its behaviour in the finite element simulation software Abaqus. The resulting compressed state was exported to the composites process modelling software PAM-RTM to continue with resin infusion simulations. The reliability of the simulation results was assessed by comparing both filling behaviour and filling time for various experimental infusion cases.
Thus, achieving optimal compaction and accurately determining the fibre volume fraction are critical steps in manufacturing fibre-reinforced composite materials. This also becomes crucial for reliable numerical predictions of the filling time and behaviour.
This work aims to investigate different techniques for measuring bag height, estimating fibre volume fraction during manufacturing and quantifying their reliability against traditional destructive characterisation methods. Specifically, a stereo Digital Image Correlation (DIC) system, a blue-light scanner and a laser sensor have been applied simultaneously for a series of repeated small laminate infusions consisting of five layers of biaxial glass fabric.
Height measurements to determine the compression behaviour were obtained before, during and after the VARI process, and their results were compared. Additionally, post-mortem burn-off tests and X-ray scanning were used to validate the fibre volume fraction estimates from beg height measurements.
The gathered experimental data allowed for the development of material models describing the compressed state of the reinforcement layup before infusion and to predict its behaviour in the finite element simulation software Abaqus. The resulting compressed state was exported to the composites process modelling software PAM-RTM to continue with resin infusion simulations. The reliability of the simulation results was assessed by comparing both filling behaviour and filling time for various experimental infusion cases.