Abstract
In this paper, we present results from a computational fluid dynamics (CFD) model for the mixing process used to disperse synthetic fibers in wet-lay process. We used CFD software, FLUENT, together with the MIXSIM user interface to accurately model the impeller geometry. A multiple reference frame (MRF) model and standard k-e turbulence model were used to model the problem. After obtaining a converged solution for the mixing tank with water, a discrete phase model was constructed by injecting spherical particles into the flow. A mixing tank with baffles and a centrally located impeller was used in experiments. PET fibers (1.5 denier, 6.35, 12.7, and 38.7 mm) at a concentration of 0.01% were mixed in water for the study. In regions behind the baffles, where the model predicted higher concentration of particles, experimental results showed a 34% higher concentration relative to the region in the high turbulence zone near the center. Instantaneous sheets were formed by rapidly dipping and removing a flat wire mesh strainer into the tank at two different locations to assess the state of dispersion in the tank. The sheets were transferred onto a blotting paper and examined under a microscope to count defects. Results show that the number of rope defects was 43% higher in sheets drawn from the region behind the baffles than in the sheets drawn from regions near the center of the tank. Changing baffles from a rectangular to a triangular cross section decreased the number of rope defects, but increased the number of log defects in the sample sheets at the same location. The CFD model can be used to optimize mixing tank design for wet lay fiber dispersion. The model provides further insight into the mixing process by predicting the effect of changes in design parameters on dispersion quality.