Injection Barrel/Nozzle/Mold-Cavity Scientific Real-Time Sensing and Molding Quality Monitoring for Different Polymer-Material Processes

Kai-Fu Liew, Hsin‐Shu Peng, Po‐Wei Huang, Weijie SuSensors 2022 · DOI

TL;DR

This paper introduces a real-time sensing system for injection molding that monitors melt pressure across the barrel, nozzle, and mold cavity. By correlating these pressure readings with a calculated viscosity index, the researchers demonstrate a method to effectively monitor and control product quality (weight and shrinkage) for different polymer types, particularly highlighting the distinct responses of semi-crystalline (PP) and amorphous (PS) polymers.

What Problem Does This Paper Address?

Injection molding is a complex, non-linear process where numerous factors influence the final product's quality. Achieving consistent quality, especially in long production runs, is challenging due to variations in plastic materials and process parameters. Traditional methods often rely on post-production inspection, which is inefficient. The paper addresses the need for a real-time, in-situ monitoring system that can predict and control product quality by understanding the dynamic behavior of the molten polymer during the injection process. Specifically, it aims to bridge the gap between melt characteristics (like viscosity) and final product attributes (like weight and shrinkage) by leveraging multi-point pressure sensing.

Methods

The core of this research is the implementation of a real-time sensing system within an injection molding machine. This system utilizes pressure sensors strategically placed in the injection barrel, nozzle, and mold cavity. These sensors continuously collect melt pressure data throughout the injection and packing stages.

A key methodological contribution is the calculation of a viscosity index (VI) derived from the melt pressure and time data, specifically integrating the pressure profile from the point the melt contacts the sensor to the point of V/P switch-over. This VI serves as a proxy for the material's rheological behavior under processing conditions.

The study also employed a 3D optical measurement system (3D-OMS) to precisely measure the dimensional accuracy and roundness (shrinkage characteristics) of the molded parts.

The experimental design involved systematically varying several controllable injection molding parameters (screw speed, back pressure, melt temperature, injection speed, and V/P switch-over position) while keeping others constant, allowing for the analysis of their impact on melt pressure, viscosity index, and product quality.

Datasets

The study utilized two types of polymers with a controlled melt flow index (MFI) of 14.5 ± 0.5 g/10 min:

  • Polypropylene (PP): A semi-crystalline polymer.
  • Polystyrene (PS): An amorphous polymer.

The material properties (viscosity, P-V-T data, etc.) were sourced from the Moldex3D Information Library.

The experimental data collected comprised:

  • Real-time melt pressure readings from barrel, nozzle, and cavity sensors.
  • Calculated viscosity index values.
  • Measured sample weights.
  • Measured roundness (shrinkage) of the molded parts using a 3D-OMS.

No new benchmark datasets were introduced; rather, the study generated experimental data under controlled conditions.

Key Insights and Ideas

  • Multi-point sensing is crucial: Monitoring pressure at the barrel, nozzle, and cavity provides a more comprehensive understanding of melt behavior than a single sensor location. The distinct pressure profiles at each location reveal different aspects of the injection process.
  • Viscosity Index as a quality predictor: The calculated viscosity index, derived from real-time pressure data, shows a strong correlation with critical product quality attributes like weight and roundness. This offers a non-destructive, in-situ method for quality assessment.
  • Polymer type matters significantly: Amorphous polymers (PS) exhibit more pronounced changes in melt pressure and viscosity index compared to semi-crystalline polymers (PP) under similar processing conditions. This highlights the need for material-specific monitoring strategies.
  • Process parameter sensitivity: Different process parameters have varying impacts on melt behavior and product quality. For instance, screw speed and back pressure significantly influence melt pressure and viscosity, while melt temperature has a more pronounced effect on PS flow characteristics.
  • Shrinkage is complex: Shrinkage (roundness) is influenced by multiple factors, including material type and process parameters. Semi-crystalline polymers like PP show greater sensitivity to certain parameters (e.g., back pressure) regarding shrinkage.

Key Results

  • The study successfully demonstrated that real-time melt pressure measurements from the barrel, nozzle, and mold cavity, when combined with a calculated viscosity index, can effectively monitor and predict injection product weight and shrinkage variations.
  • For materials with similar MFI, amorphous polymers (PS) showed significantly larger variations in pressure and viscosity index compared to semi-crystalline polymers (PP).
  • The melt pressure-peak for PP was measured at 294 bar, while for PS it was 526 bar under standard conditions, with corresponding viscosity indices of 13.17 bar-s for PP and 33.18 bar-s for PS.
  • Increasing screw speed led to higher melt pressure-peak and viscosity index, with a slight decrease in weight and a significant increase in roundness for PP.
  • Increasing back pressure also increased melt pressure-peak and weight, and significantly increased roundness, particularly for PP.
  • Increasing melt temperature decreased melt pressure-peak and viscosity index for both materials, with PS being more affected.
  • Increasing injection speed generally decreased melt pressure-peak and viscosity index.
  • The V/P switch-over position was found to be critical for adjusting melt pressure-peak, viscosity index, and product weight.

Limitations and Future Work

Limitations: * The study focused on a specific sample geometry (100 mm diameter, 4 mm thickness round-shaped sample) and two specific polymer types. The findings may not directly translate to all geometries or polymer families without further validation. * While the viscosity index is a useful proxy, it's a calculated value based on pressure and time, not a direct rheological measurement. * The research acknowledges that factors like gate location and type can also influence quality, but these were not the primary focus of the parameter variation experiments.

Future Work: * The authors suggest that real-time melt pressure-peak and viscosity index information collected from sensors can serve as a characteristic value for adaptive process control in scientific injection molding. * Further research could explore a wider range of materials, geometries, and processing conditions to generalize the findings. * Integrating this sensing system with advanced control algorithms for closed-loop adaptive process control is a logical next step. * Investigating the direct correlation between the calculated viscosity index and actual rheological measurements under processing conditions would strengthen the findings.

Summary updated

Abstract

Scientific injection molding technologies involve the integration and collaboration of cyber-physical systems and smart manufacturing. In order to achieve adaptive process control and production optimization, injection molding systems with real-time sensing have gradually become the development- and application-trend of smart injection molding. At the same time, this technology is a highly non-linear process in which many factors affect the product quality during long-run fabrication processes. Therefore, in order to grasp changes in the characteristics of plastic materials and product quality monitoring, the injection process has become an important research topic. We installed sensors in the molding machine (injection barrel, nozzle, and mold-cavity) to collect the melting pressure and used different materials (semi-crystalline and amorphous polymer; the melting-fill-index (MFI) is unified to 14.5 ± 0.5 g/10 min) to explore the influences of melting pressure variation and its viscosity index on the quality characteristics of molded products. The experiment reveals that a combination of barrel, nozzle, and mold-cavity sensing on the melt-pressure trend-based injection process-control incorporated with viscosity index monitoring can confirm the weight and shrinkage variation of the injection product. At the same time, the pressure and viscosity index value measured and calculated during the melt-filling of two materials with similar MI resulted in significant variations in the amorphous polymer. This study showed the possibility of mastering and controlling the rheology (barrel position) and shrinkage properties of polymers and successful application in various product-quality monitoring platforms.
BibTeX
@inproceedings{liew2022injection,
  author = {Kai-Fu Liew and Hsin‐Shu Peng and Po‐Wei Huang and Weijie Su},
  title = {{Injection Barrel/Nozzle/Mold-Cavity Scientific Real-Time Sensing and Molding Quality Monitoring for Different Polymer-Material Processes}},
  booktitle = {Sensors},
  year = {2022},
  doi = {10.3390/s22134792},
}
Figure