Unsteady Fluid Flows in the Slab Mold Using Anticlogging Nozzles

Abstract

The characterization of the fluid flow of liquid steel in a slab mold, using two nozzle designs under unclogged and clogged conditions, is performed using physical and mathematical simulations. Nozzle A, with an expanding and contracting geometry, yields larger sub-meniscus experimental velocities than nozzle B, with internal flow deflectors. The numerical predictions indicate quick time-changing velocity profiles in the submeniscus region between the mold’s narrow face and the nozzles. The flow deflectors in nozzle B have two effects; the high dissipation rate of kinetic energy in the upper-half length induces lower velocities in the ports than nozzle A. The neutralization of the biased flow caused by the sliding gate allows a balanced fluid through the ports. According to the results, nozzle A yields velocity profiles in the sub-meniscus region with larger standard deviations than nozzle B, leading to an unstable bath surface. The clogged nozzles produced biased-asymmetrical flow patterns in the mold, finding approximated matchings between numerical predictions and experimental measurements. The internal protrusions of the deposits lead to covariance losses of the bath surface wave heights. The use of internal deflectors helped to decrease the amount of clog material in nozzle B.

Characterization of the Turbulent Fluid Flow Structures in a Slab Mold with Experimental Measurements and Different Turbulence Models

“…The right-side, upper roll flow yields the highest magnitudes in the velocity scale. Although, as reported in another work, this model can predict well appreciable velocity changes in the mold after longer periods [41]. The DES model predicts more dispersed discharging jets than the RKE model, matching more closely the experimental observations of the jets revealed through the injection of a red dye tracer, as seen in Figures 5a-5d.…”

Physical Modeling and Mathematical Modeling Using the Scale‐Adaptive Simulation Model of Nozzle Design Effects on the Flow Structure in a Slab Mold

“…These time‐averaged velocity fields yield four vortical flows developed by the double roll flows located at each side of the nozzle. Although the model describes changes in the flow patterns (flow turnovers with periods of ≈300 s for both nozzles), [ 41 ] including the oscillations of the discharging jets with time, it cannot provide information about the fine structure of these flows. In contrast, the SAS model yields a fine structure demonstrating the existence of multiple and small vortex flows, especially when using nozzle B.…”

“…Physically, the slow decreasing correlations mean that the flow patterns change after times of the order of 360–370 s. The k–ε realizable model correctly predicted these changes at the mentioned frequencies. [ 41 ]…”

“…The velocity fields in the central-symmetric planes of the mold, using nozzles A and B, calculated through the k-ε realizable URANS and the SAS models, are in Figure 6a-d [41] including the oscillations of the discharging jets with time, it cannot provide information about the fine structure of these flows. In contrast, the SAS model yields a fine structure demonstrating the existence of multiple and small vortex flows, especially when using nozzle B.…”

“…Physically, the slow decreasing correlations mean that the flow patterns change after times of the order of 360-370 s. The k-ε realizable model correctly predicted these changes at the mentioned frequencies. [41] Figure 14a-c shows Nozzle A's power spectra density (PSD) for the three points mentioned along the imaginary line; see Figure 2. The power spectra density is defined as follows…”