Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Simulation of reinforced reactive injection molding with the finite volume method

: Wittemann, Florian; Märtens, Robert; Bernarth, Alexander; Hohberg, Martin; Kärger, Luise; Henning, Frank

Fulltext urn:nbn:de:0011-n-4818973 (1.4 MByte PDF)
MD5 Fingerprint: 9d0b60c3c7493a372132ddc89030d0d8
(CC) by
Created on: 8.2.2018

Journal of composites science 2 (2018), No.1, Art. 5, 16 pp.
ISSN: 2504-477X
Journal Article, Electronic Publication
Fraunhofer ICT ()

The reactive process of reinforced thermoset injection molding significantly influences the mechanical properties of the final composite structure. Therefore, reliable process simulation is crucial to predict the process behavior and relevant process effects. Virtual process design is thus highly important for the composite manufacturing industry for creating high quality parts. Although thermoset injection molding shows a more complex flow behavior, state of the art molding simulation software typically focusses on thermoplastic injection molding. To overcome this gap in virtual process prediction, the present work proposes a finite volume (FV) based simulation method, which models the multiphase flow with phase-dependent boundary conditions. Compared to state-of-the-art Finite-Element-based approaches, Finite-Volume-Method (FVM) provides more adequate multiphase flow modeling by calculating the flow at the cell surfaces with an Eulerian approach. The new method also enables the description of a flow region with partial wall contact. Furthermore, fiber orientation, curing and viscosity models are used to simulate the reinforced reactive injection molding process. The open source Computational-Fluid-Dynamics (CFD) toolbox OpenFOAM is used for implementation. The solver is validated with experimental pressure data recorded during mold filling. Additionally, the simulation results are compared to commercial Finite-Element-Method software. The simulation results of the new FV-based CFD method fit well with the experimental data, showing that FVM has a high potential for modeling reinforced reactive injection molding.