Formation of voids due to transitions in permeability and cavity diameter during resin injection processes

The specific mechanical properties of fiber reinforced composite components are unmatched, considering their low weight. To optimize the lightweight potential of fiber reinforced composites, fiber volume contents have to be maximized and imperfections must be eliminated. However, during the production of fiber reinforced composite laminates via resin injection processes, the formation of microscopic voids is nearly inevitable. Even low amounts of imperfections can cause significant deteriorations in the mechanical properties of the material. To reduce the number of voids inside composite components, understanding the formation and transport of voids is essential. Numerous renowned models describe said formation of voids in dependence of local flow front conditions during the impregnation of textile preforms with thermoset resins. State of the art are models emphasizing the formation of meso-and microvoids in dependence of the modified capillary number. These models show plausible correlations when applied to unidirectional preforms or fabrics with constant permeability along the direction of flow. However, the formation of voids remains to be investigated at points of transitioning permeability, such as alterations in the setup of layers or abruptly changing cavity diameters. To expand the applicability of the existing models onto preforms with local changes in permeability, an experimental setup for gradually increasing cavity diameter and varying layer setup is introduced. A planar mold with three increasing levels of cavity height is used to induce changes in permeability. The rate of change in permeability is controlled by defining the slope between each level. In this study, injection pressure as well as flow front velocity were optically traced, and material data was measured. Resulting local void volume contents were quantified by calcination. It is demonstrated how alterations of diameter and layup take effect on the resulting local porosity. The observed impact on void formation is put in context to changes in tow permeability due to local differences in fiber volume content. By including the slope dependent rate of change in tow permeability into the existing model for calculation of void formation by GUEROULT ET AL., the accuracy of the model can be increased. Comparing the unaltered model to experimental results, the deviations between calculations and measurements were diminished when using the newly introduced factors. Although the error of p