| Two linear shrinkage models, the residual strain (RS) and corrected in-mold residual stress (CRIMS) models, were evaluated for their ability to predict the shrinkage of 30% glass-filled nylon-6,6 and poly(butylene terephthalate) (PBT). Both models overpredicted the in-flow and near gate cross-flow linear shrinkage and underpredicted the cross-flow shrinkage at the end of fill. The difference between measured and predicted shrinkage, however, was same with the glass-filled material as with an unfilled polycarbonate used in a previous study. These linear shrinkage predictions were insensitive to the viscosity models, juncture loss coefficients, solidification temperatures, input flow rates, and input switchover positions, but were affected by changes in packing pressure. Volumetric shrinkage was far more sensitive to processing conditions, simulation inputs, and changes in the analysis type. Although fiber orientation predictions have shown reasonable agreement with experimental results, these shrinkage results suggest that fiber orientation has little effect on the overall shrinkage predictions for fiber filled materials. |
