The Influence of Processing and Fluid Parameters on Injection Molding Flow.

Injection molding is one of the most common and most important methods of processing thermoplastic polymers. Analysis of injection molding is difficult because of the non-Newtonian flow of the polymer melt. This non-Newtonian flow also influences the physical properties of the solidified part. An optical access mold was designed and built to allow viewing of the melt during the molding of a rectangular plaque. Particle image velocimetry (PIV) was used to measure the midplane velocity field of STYRON 615 APR during mold filling. The gate pressure, flow front shapes and front propagation velocities were also measured. These measurements were compared to Moldflow simulations. Moldflow was found to be accurate for the bulk of the flow, with an average vector orientation error of 1.7 degrees. The ratio of cavity-averaged PIV velocity magnitudes to Moldflow average magnitudes had a mean of 0.998 and a standard deviation of 0.25. This deviation was due to problem areas near the cavity walls and the flow front. Moldflow also had difficulty predicting realistic flow front shapes. A power-law model of the flow was developed and was used to predict the gate pressure. The model and the measured gate pressure were used to estimate the power-law parameters of the melt. The power-law exponent was found to be 0.4063, within 0.4% of the value measured by capillary rheometer. The power-law coefficient was found to be 2659 Pa-s^n, within 9.5% of the measured value. Velocity fields during packing were also measured. The random component of the flow was initially 10-20% and grew to 60-80% after 2.5 seconds of packing. The random component increased as the packing pressure was decreased. The mass flow during packing was modeled analytically as a combination of compressibility and solidification effects. Two limiting cases of the model bracketed the experimental data within 20%. The parameters of the Poincare optically equivalent model of the residual strain field in the solid parts were measured with a polariscope. Retardation and primary and secondary axis orientations were measured. The stress field was two-dimensional near the edges of the parts and three-dimensional far from the edges.PhDMechanical EngineeringUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/62419/1/tbress_1.pd