Two‐phase flows simulations in a space‐time framework for injection molding applications: Comparison with experimental results

Abstract High‐resolution numerical simulations of the injection molding process are crucial for precise process and part behavior prediction and tool fabrication. However, the complex flow behavior with high shear rates, rapid cooling as well as the nonlinear material properties post many challenges, and efficient high‐resolution numerical analysis is still subject to research. This study presents a simplified numerical description of the polymer flow and cooling during the injection phase, aiming for efficient and yet accurate simulations of the process. The in‐house finite‐element solver XNS is used to simulate the polymer and air flows, as a two‐phase flow, using the level‐set method. The injected polymer is characterized as a shear‐thinning flow using the non‐Newtonian Cross‐Williams‐Landel‐Ferry rheology model. To reduce the computational time, other polymer model effects such as a precise crystallization kinetics model during the injection phase have been neglected as a result of the findings of a previous study. Heading towards more efficient computations, we use a space‐time discretization, which has a higher convergence rate than commonly used time discretizations and can allow future mesh refinement in time, particularly convenient for our application. The results of the simulation are compared to the propagation of the flow front in an injection molding process of a plate part. The flow front is traced in‐situ by three aligned infrared sensors that detect the temperature increase from the mold wall. The time of flow front propagation between two sensors is in good agreement with experimental results. The peak temperature is calculated lower as compared to the experiment; however the temperature gradients are in good agreement validating the numerical model presented here.