Joint project Thermoset control

 

Development of a control concept to increase process productivity and product quality in
the injection molding of thermosets

 

Motivation

Thermosets offer several significant advantages over thermoplastics. They are characterized
by high temperature and media resistance, good electrical insulation properties and high
mechanical rigidity. Therefore, thermosets are increasingly becoming the focus of technical
series applications. In this context, injection molding presents a key technology for producing
complex molded parts from thermoset material.
However, the production of thermoset moldings poses severe technical challenges. Due to the
thermally irreversible crosslinking reaction, injection molded parts cannot be re-melted after
cross-linking. In addition, the material behavior of the plastic melt depends on the progress of
the cross-linking reaction. The rheological properties of thermosets also vary depending on
the material composition, storage conditions, and storage time.
To ensure a proper as well as reproducible injection molding process, the machine operator
has to adjust the process iteratively based on experience, which is both time-consuming and
prone to errors. This is especially true for highly complex molds. As a result, high personnel
costs, subjective quality results, and usually a high scrap rate due to the iterative procedure
ensue.

 

Project goals and methods

Using model-based control is intended to increase process productivity and product quality in
thermoset injection molding, compensating for the aforementioned disturbances and thus
ensuring reproducible part quality.
In a first step, process variables that significantly affect the molded part quality and that are
suitable for control must be determined. For this purpose, a series of measurements with
different materials and a subsequent sensitivity analysis are carried out in cooperation with
the Institute of Plastics Processing, called IKV, at RWTH Aachen University. Based on the identified
correlations between the process and quality variables, a data-based quality model is
developed that relates the time series of the process variables to a predicted part quality.
The quality model is then used to generate optimal reference curves for the investigated
model-predictive process control. The developed control concept is tested on different molds
and with varying materials to derive a systematic procedure for transferring the control
concept to different mold geometries.

 

Innovations and Perspectives

Research in the field of thermoset injection molding seeks to investigate methods to increase
process productivity and product quality. Doing so is crucial in order to improve the
attractiveness of the material and make its many advantages over thermoplastics
economically viable. A corresponding quality control system does not exist yet and thus has a
high research and market potential.

 

Project partner
RWTH Aachen University, IKV - Institute for Plastics Processing