Project C3: LIMMCAST: Modelling of steel casting

PI: Sven Eckert (HZDR)
Partners: HZDR, TUBAF

1. Scientific case for the project

1.1 Background

The continuous casting process is used to provide more than 90% of the worldwide steel production, because the high mass flow rates this technology provides a striking productivity. The energy efficiency and the profit are directly coupled with the casting speed, but the enhancement of the casting speed may lead to violent flow conditions in the mould. Most of the problems in continuous casting affecting the steel quality are associated with an improperly conditioned mould flow. Small improvements in the flow pattern can achieve therefore large effects in terms of quality assurance and energy savings. Magnetic fields are a powerful tool to control the melt flow within the continuous casting process. AC magnetic fields are employed as electromagnetic stirrers (EMS) for a better homogenization of the melt and for a promotion of the double-roll flow pattern in the mould, which is supposed to decrease many slab defects [1]. The utilisation of DC magnetic fields as electromagnetic brake (EMBR) is considered as a suitable approach to ensure high quality slabs at high casting speeds. Although various EMBR designs have already been adopted for industrial use since more than 20 years [2], the impact of a DC magnetic field on such highly turbulent and complex flows is a complicated phenomenon and has not been fully understood yet. Contrary to the usual expectations, static magnetic fields may even destabilize liquid metal flows. Respective indications have been found in diverse liquid metal experiments on laboratory scale [3,6]. The impact of an EMBR on the melt flow in the continuous casting mould was addressed by many numerical studies [4,5] considering various magnetic field configurations or examining the influence of variations of different casting parameters on the magnetic field effect. Most of the previous calculations rely on RANS turbulence models and report a plain suppression of the mean flow and the turbulent fluctuations as well. The application of a DC magnetic field on the flow of any electrically fluid exerts an electromagnetic damping effect, which produces a redistribution of the turbulent energy by a selective damping of turbulent structures depending on their spatial alignment. This anisotropic Joule dissipation is not covered by conventional turbulence models. Appropriate numerical simulations regarding the magnetic field impact need an extension of the respective turbulence models as well as an experimental data base for code validation. Previous activities mainly employed water models, which are, however, of limited value compared to the real steel flow and cannot provide any useful information with respect to the interplay between the melt flow and the magnetic field applied.

1.2 Most important goals of the planned work

The project will be focused on investigations of the fluid flow in the continuous casting process comprising both single- and multiphase flows occurring in the tundish, submerged entry nozzle (SEN) and the mould. The main goal consists in the development of suitable technologies for an efficient flow control by means of external magnetic fields. An essential prerequisite is the significant improvement of the knowledge about the magnetic field impact on the turbulent steel flow. A substantial deficit has to be noticed regarding experimental data of the velocity field in real liquid metal systems being suitable for the validation of numerical codes. Therefore, flow measurements in low melting point alloys will be a mainstay of the current project. The flow measurements at the LIMMCAST facility at HZDR will rely on measuring techniques like the ultrasound Doppler Velocimetry (UDV) or the Contactless Inductive Flow Tomography (CIFT). A further improvement and a better adaption of these techniques with respect to the specific needs at LIMMCAST will be aspired within this project. Moreover, new approaches to detect two-phase flows both in the mould and the SEN should be implemented. A close collaboration with the YIG “Measurement Techniques” is planned in this respect.
An experimental data base should be established concerning both DC and AC magnetic field applications. The experiments using DC magnetic fields for flow control should verify the conditions for an assured magnetic braking effect. Critical field configurations and casting parameters causing a risk for flow destabilization have to be identified. The research will lead to proposals for an efficient flow control by optimised magnetic field configurations.
Further investigations will concern the effect of magnetic stirrers. The potential of AC magnetic fields regarding flow stabilizing and homogenization should be validated in the mould. Rotational mould stirring is considered for billet casting, whereas slab geometries will be equipped with various combinations of travelling fields.
The magnetic damping of turbulent flows causes an anisotropic dissipation term, the so-called Joule dissipation, which produces a redistribution of the turbulent energy by a selective damping of turbulent scales. This mechanism acts as a local source of momentum, vorticity and kinetic energy and is not covered by conventional turbulence models. The project C3 shall contribute to an adequate modelling of the turbulent flow in the continuous casting process under the impact of a strong DC magnetic field. The numerical simulations going along with the experiments are supposed to accomplish an up-scaling of the experimental results gained from the LIMMCAST mock-up towards real industrial facilities.

2. Existing competencies and infrastructure

Model experiments using low melting point metals are employed at HZDR for investigations of fluid flow and related transport processes. In particular, the so-called LIMMCAST (LIquid Metal Model for continuous CASTing of steel) programme has been established to investigate the steel flow in tundish, nozzle and mould, which comprises a large facility operating with SnBi alloy and small-scale facilities using GaInSn at room temperature [7]. Specific competence exists at HZDR in developing and applying specific diagnostic techniques, in particular UDV, CIFT and local conductance anemometers, for measuring the fluid flow in the LIMMCAST models [8,9]. Previous flow measurements at a small-scale model showed that a DC magnetic field does not lead inevitably to a smooth reduction of the velocity fluctuations all over the mould region. Actually, under certain conditions the application of a horizontal magnetic field may even cause intense, non-steady and non-isotropic flow structures [6]. Further experiences exist in experimental investigations considering the peculiar features of quasi-two-dimensional turbulence [10].
The experiments at the LIMMCAST facility will be accompanied by numerical calculations performed by TUBAF. The research group of Prof. Schwarze is known for their competence in simulating metallurgical processes. Among others, the flows in different parts of continuous casting machines (tundish, SEN and mould) have been studied in detail by numerical simulations [11-13]. A major topic of the research is the formulation of suitable turbulence models for the flows in these devices. Within the investigations, the first rigorous analysis of the coherent oscillations of the large-scale eddy structures in the upper part of the mould was performed [13]. In addition, a numerical model of the hydrodynamic flow in the Mini-LIMMCAST facility at HZDR has been presented recently [14]. At present, the employed numerical models for the metallurgical flows are upgraded in order to describe the complete thermo-fluiddynamics with solidification in casting machines correctly.

3. Resource planning and Budget Justification

Within the considered project an experienced scientist will be needed at HZDR, who has to contribute to the design of the experimental setup and the appropriate installation of the measuring techniques. This person will perform the measurements and will be responsible for data acquisition and analysis. The numerical simulations at TUBAF will be conducted by a PhD student, who will be supported by the numerical teams from both TUBAF and HZDR

Links: Close relations exist to projects A3, A4, A5, B5 and the YIG.

References

[1] S. Kunstreich, 2003, Electromagnetic stirring for continuous casting. Rev. Met., Vol. 11, 395-408
[2] J. Nagal, K.I. Suzuki, S. Kojima, S. Kollberg, 1984, Steel flow control in a high-speed continuous slab caster using an electromagnetic brake. Iron Steel Eng., Vol. 61, 41-47.
[3] U. Burr, L. Barleon, P. Jochmann, A. Tsinober, 2003, Magnetohydrodynamic convection in a vertical slot with horizontal magnetic field, J. Fluid Mech., Vol. 475, 108-113.
[4] K. Takatani, K. Nakai, N. Kasai, T. Watanabe, H. Nakajima, 1989, Analysis of Heat Transfer and Fluid Flow in the Continuous Casting Mold with Electromagnetic Brake. ISIJ Int. 29, 1063-1068.
[5] K. Cukierski, B.G. Thomas, 2008, Flow Control with Local Electromagnetic Braking in Continuous Casting of Steel Slabs. Metall. Mater. Trans. B, Vol. 39B, 94-107.
[6] K. Timmel, S. Eckert, G. Gerbeth, 2011, Experimental investigation of the flow in a continuous casting mould under the influence of a transverse DC magnetic field. Metall. Mater. Trans. 42B, 68-80.
[7] K. Timmel, S. Eckert, G. Gerbeth, F. Stefani, T. Wondrak, 2010, Experimental modeling of the continuous casting process of steel using low melting point metal alloys - the LIMMCAST program. ISIJ International, Vol. 50, 1134-1141.
[8] T. Wondrak, V. Galindo, G. Gerbeth, T. Gundrum, F. Stefani, K. Timmel, 2010, Contactless inductive flow tomography for a model of continuous steel casting. Meas. Sci. Techn. 21, 045402.
[9] T. Wondrak, S. Eckert, G. Gerbeth, K. Klotsche, F. Stefani, K. Timmel, A. Peyton, N. Terzija, W. Yin, 2011, Combined electromagnetic tomography for determining two-phase flow characteristics in the submerged entry nozzle and in the mould of a continuous-casting model. Metall. Mater. Trans. B, Vol. 24B, 1201-1210.
[10] S. Eckert, G. Gerbeth, W. Witke, H. Langenbrunner, 2001, MHD Turbulence Measurements in a Sodium Channel Flow Exposed to a Transverse Magnetic Field. Int. J. Heat Fluid Flow, Vol. 22, 358-364.
[11] R. Schwarze, H. Chaves, Ch. Brücker, 2009, Investigation of the gas-liquid flow in a stopper-rod controlled SEN. steel res. int, Vol. 80, 834-840.
[12] R. Schwarze, J. Klostermann, Ch. Brücker, 2008, Experimental and numerical investigations of a turbulent round jet into a cavity, Int. J. Heat Fluid Flow, Vol.. 29, 1688-1698.
[13] R.Schwarze, 2006, Unsteady RANS simulation of oscillating mold flows, Int. J. Num. Meth. Fluid, Vol. 52, 883–902.
[14] R. Schwarze, A. Maiwald, K. Timmel, G. Gerbeth, 2009, Comparison of numerical and experimental flow data from a liquid-metal model of a continuous casting mold. Proc. 6. International Conference on Electromagnetic Processing of Materials, Dresden.