Permeability characterization and microvoid prediction during impregnation of fiber tows in dual-scale fabrics
Date
2005
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Accurate characterization of the permeability of fabrics used as reinforcement materials in liquid composite molding (LCM) processes is necessary to realistically model the flow through these porous preforms. When dual-scale preforms are employed, not just the macro permeability values are of interest; the micro-scale permeability of the fiber tows inherent to the fabric must additionally be determined. These permeability values can then be used in LCM simulations in which standard two- or three-dimensional (2D or 3D) mesh elements are combined with one-dimensional (1D) elements. The 1D elements are attached at each node and represent the fiber tows in preforms that exhibit dual scale porosity. This implementation allows for the interactions between the macro and micro flow, and can predict the saturation of the fiber tows, along with the movement of the macroscopic resin flow front. Consequently, the time it will take to fill the mold and saturate all regions can be accurately predicted and any void formation sites can be forecasted. ☐ A methodology is proposed, which can determine the macro permeability along with a simulation fitting parameter that is closely associated to the micro permeability. The inlet pressure profile of a single 1D constant flow rate resin transfer molding (RTM) experiment is evaluated experimentally and numerically in order to determine the two values. The methodology is validated analytically and in a simulation environment to demonstrate the versatility and limitations, and to reinforce the trends and tendencies. Characterizations for four different fabrics are performed. ☐ Additionally, the percentage and distribution of voids in a composite are evaluated. The voids are usually a function of the fiber preform architecture, the infusion scheme, and the processing method, in addition to the inherent variations in the preform and the hand lay-up technique. For preforms that are woven or stitched, the fiber tows have a much lower permeability as compared to the permeability between the fiber tows. For this reason, processing methodology has converged on letting the resin bleed out of the vent to allow sufficient time for the low permeability fiber tows to fully saturate. The impact of this processing methodology is explored, along with an alternative process in which flow resistance is added. The influence of these modifications is characterized in terms of void content in the composite. Samples are machined from each of the manufactured panels and analyzed using image analysis techniques, so that a relative void content comparison can be made. ☐ The permeability results show that as the fabric layers become less pliable, more deviations result in the macro permeability, since the layers are not able to mesh together as consistently. For all fabric cases, the macro permeability decreases as the micro-permeability parameter increases. For each fabric, as the disparity in the value of the macro permeability and micro-permeability parameter grows, a parameter established as the partially-saturated length additionally increases. ☐ The void distribution results show that if resin is not allowed to bleed, the void content over the length of the part is not uniform; the percentage of voids is much higher near the vent side of the part. When no resistance is used at the vent, the voids increase through the thickness of the part; this trend levels out when resistance is added. When bleeding is allowed and if a resistance at the vent is added, all void levels are reduced.