Abstract

Horizontal Single Belt Casting (HSBC) possesses several advantages over current casting processes (e.g. DC Casting of aluminum, Continuous Casting of steels). These include greatly reduced investment costs, greatly reducing processing steps and therefore the processing time of the near-net-shape product, much improved environmental impacts, with equivalent or improved microstructures. Using an HSBC process, relatively thin strips (3-15 mm thick), with uniform microstructures across the strip thickness, with refined grains, have been obtained directly from liquid state. These cast strip microstructures can be attributed to the higher heat fluxes and enhanced heat transfer coefficients of HSBC, compared to those reported for the conventional casting technologies just noted. Mathematical predictions have also been developed to confirm our pilot scale experiments, using the Computational Fluid Dynamics (CFD) code, Fluent ANSYS v.19 program, to further study and optimize some of the processing parameters. These include liquid metal feeding systems, belt speeds, belt dimensions (length and width), and inlet velocities. Most recently, the distance between the inlet and the inclined plane, termed the “free fall distance” was lowered from 6 to 3 mm, rendering important process improvements. These were attributed to the quality of the falling film, which showed reduced oscillations and suspected entrainment of air. According to recent simulations for various alloy systems (commercial aluminum alloys, TWIP steels, and calcium based Bulk Metallic Glasses), the main purpose of these predictions has been to guide us in our aim to achieve steady, uniform flows, with stable back-menisci, and reduced surface wave formation when a liquid metal flows down onto a moving, horizontal, single belt. We have also shown that transient two-dimensional models can provide adequate information on the temperature evolution, heat fluxes, etc. vs the more granular information provided by three dimensional predictions that enabled us to predict our previous experimental results, close to the entry regions of the caster. The three-dimensional simulations also served our purpose in studying interfacial interactions between the metal and air. However, not unexpectedly, the heat transfer results did not vary significantly with respect to the strip width.  As such, the present work expands on the two-dimensional simulations’ studies of the HSBC process for several aluminum and steel alloys for much longer 5m and 10 m long, endless moving belts, versus the 1.6m long belt, for our proof-of-concept research work. Currently, different boundary conditions for the moving belt are tested to optimize and improve the current simulations, e.g., convective and constant heat flux boundary conditions. The simplified simulations allow one to more readily predict the solidification points and the temperature evolution of the forming strips for different alloys with optimized, but realistic heat transfer conditions. The work also intends to summarize the capabilities of the HSBC process for different alloys, providing expected solidification distances along the cooling belt, cooling rates and temperature profiles for commercial HSBC operations.