Injection molding is a mature technology that has been used for decades; factors including processed raw materials, molds and machines, and the processing parameters can cause significant changes in product quality. Traditionally, researchers have attempted to improve injection molding quality by controlling screw position, injection and packing pressures, and mold and barrel temperatures. However, even when high precision control is applied, the geometry of the molded part tends to vary between different shots. Therefore, further research is needed to properly understand the factors affecting the melt in each cycle so that more effective control strategies can be implemented. In the past, injection molding was a “black box”, so when based on statistical experimental methods, computer-aided simulations or operator experience, the setting of ideal process parameters was often time consuming and limited. Using advanced sensing technology, the understanding of the injection molding process is transformed into a “grey box” that reveals the physical information about the flow behavior of the molten resin in the cavity. Using the process parameter setting data provided by the machine, this study developed a scientific method for optimal parameter adjustment, analyzing and interpreting the injection speed, injection pressure, cavity pressure, and the profile of the injection screw position. In addition, the main parameters for each phase are determined separately, including injection speed/pressure during the mold filling phase, velocity-to-pressure switching point, packing pressure and time. In this study, the IC tray was taken as an example. The experimental results show that the method can effectively reduce the warpage of the IC-tray from 0.67 mm to 0.20 mm. In addition, the parameters profiles obtained by parameter optimization can be applied for continuous mass production and process monitoring.
In-mold and Machine Sensing and Feature Extraction for Optimized IC-tray Manufacturing
