Analysis of in-mold shrinkage measurement for amorphous and semicrystalline polymers using a multivariate sensor

The continuously growing need for quality assurance and systems integration in plastics manufacturing, promoted by Industry 4.0 technologies as well as interest in sustainability and recycling, creates significant opportunities for advanced multivariate in-mold sensing. This work describes the design of a multivariate shrinkage sensor (MVSS) incorporating a spring-biased pin with a digital linear displacement transducer to measure in-mold shrinkage directly. The multivariate sensor also incorporates a piezoelectric ring for cavity pressure measurement and an infrared detector for melt and mold temperature acquisition. The combined use of the sensor signals allows real-time process monitoring and prediction of the molded part dimensional quality. A design of experiments (DOE) was used to validate the sensor functionality for an amorphous high-impact polystyrene (HIPS) and a semicrystalline polypropylene (PP). The results indicated that the root mean square error of the predicted thicknesses was 6.3 microns (i.e., 0.21%) for a regression model based on the DOE factor settings, 4.8 microns (i.e., 0.16%) based on traditional cavity pressure and temperature data, and 3.4 microns (i.e., 0.11%) based on just the available MVSS data. The modeled main effects highlight the different shrinkage behavior of HIPS and PP indicating the need for in-mold shrinkage data.