Single-screw extruders are used for shaping various plastic products such as pipes, hoses, films, and sheets. The growing demands placed on these machines in terms of productivity, product quality and energy efficiency become increasingly difficult to realise with conventional plasticising screws. One promising solution is the use of so-called wave-dispersion screws: Thanks to their wave-like channel depth profile, they allow for good mixing of the plastic melt at relatively low energy consumption and high production rates. For an optimised design of these screws, however, there has been a lack of calculation models that map the flow conditions in the extruder with sufficient accuracy and provide useful results in a reasonable time frame. As a result, the high potential of wave-dispersion screws has not been fully utilised yet in the plastics processing industry.
Aiming for closing this research gap, an existing calculation approach has been further improved during this research project. This approach maps melt conveying in single-screw extruders via a network of interconnected screw elements, being able to capture the cross-sectional changes in the conveying direction and the flow across the screw flights in wave-dispersion screws. At the same time, the calculation times required are considerably shorter compared to fully three-dimensional flow simulations. To describe the operating behaviour of the individual sections along the screw more accurately, new approximation equations for the pumping capability and the energy input with an extended range of validity were derived. These equations are based on fully three-dimensional computational fluid dynamics simulations that consider the shear-thinning behaviour of polymer melts, cover a wide processing window including pressure-generating and pressure-consuming screw zones, and, for the first time, fully represent the influence of the channel curvature. Both the melt flow in the channel and the leakage flow over the screw flights can be modelled far more realistically by the new equations, which has been proven by an error analysis. In addition, by employing the novel regression models, the approximation of experimentally determined process data by means of network-based calculations is improved.
The fast predictions and the improved accuracy of the network-based calculation tool open up promising opportunities for assessing the operating behaviour of wave-dispersion screws. Applying this tool will accelerate both troubleshooting of running extrusion processes and the development of new screws. The obtained improvements in efficiency and quality will contribute to a more economic and sustainable processing of plastics.
Forschungs-