Effect of Nozzle Port Angle on Transient Flow and Surface Slag Behavior During Continuous Steel-Slab Casting

Abstract

Undesirable flow variations can cause severe instabilities at the interface between liquid mold flux and molten steel across the mold top-region during continuous steel casting, resulting in surface defects in the final products. A three-dimensional Large Eddy Simulation (LES) model using the volume of fluid method for the slag and molten steel phases is validated with plant measurements, and applied to gain new insights into the effects of nozzle port angle on transient flow, top slag/steel interface movement, and slag behavior during continuous slab casting under nominally steady conditions. Upward-angled ports produce a single-roll flow pattern with lower surface velocity, due to rapid momentum dissipation of the spreading jet. However, strong jet wobbling from the port leads to greater interface variations. Severe level drops allow easy entrapment of liquid flux by the solidifying steel shell at the meniscus. Sudden level rises may also be detrimental, leading to overflow of the solidified meniscus region. Downward-angled ports produce a classic double-roll pattern with less jet turbulence and a more stable interface everywhere except near the narrow faces. Finally, the flow patterns, surface velocity, and level predicted from the validated LES model are compared with steady-state standard k-ε model predictions.

Appendix: Spread Angles of Jet Flow

Vertical and horizontal spread angles of the jet flow at the nozzle port outlet are calculated to quantify the flow characteristics leaving the nozzle. The calculation is based on a weighted average of outward flow rate, ignoring the backflow zone where flow enters the port. Figure A1 illustrates the definition of both spread angles.

Fig. A1

Definition of (a) vertical spread angle and (b) horizontal spread angle of jet flow: example of downward-angled nozzle ports

Vertical spread angle, θ_xy,sp is calculated as follows:

$$ \theta_{\text{xy,sp}} = \left| {\theta_{\text{xy}} - \theta_{\text{xy,upper}} } \right| + \left| {\theta_{\text{xy}} - \theta_{\text{xy,lower}} } \right|, $$

(A1)

where θ_xy is the vertical jet angle at the nozzle port outlet surface,[25,46] θ_xy,upper and θ_xy,lower are vertical jet angles calculated from the velocity vectors leaving each region, split into upper and lower regions according to the regions of the port exit surface found above and below the center plane of the jet flow, respectively. Due to the back flow zone located in the upper part at the nozzle port outlet, the two regions are separated, based on a mass flow-rate balance of the jet as follows:

$$ \sum\limits_{i = 1}^{N - n} {\left( {\Delta x} \right)_{i} (\Delta z)_{i} u_{{{\text{mag}},i}} } = \sum\limits_{j = 1}^{n} {\left( {\Delta x} \right)_{j} (\Delta z)_{j} u_{{{\text{mag}},j}} }, $$

(A2)

where i and j are cells in the upper and the lower port regions, respectively, (numbered according to increasing x height when calculating vertical spread angle), N is the total number of cells in the jet region with positive outflow, n is the unknown total number of cells in the lower region to be solved, and u_mag is the velocity magnitude in each cell in the exit plane of the nozzle port.

Horizontal spread angle, θ_yz,sp, is calculated in a similar manner, as follows:

$$ \theta_{\text{yz,sp}} = \left| {\theta_{\text{yz}} - \theta_{\text{yz,IR}} } \right| + \left| {\theta_{\text{yz}} - \theta_{\text{yz,OR}} } \right|, $$

(A3)

where θ_yz is the horizontal jet angle at the nozzle port outlet surface,[25,46] θ_yz,IR and θ_yz,OR are the horizontal jet angles in each ~ half-port region, IR and OR side, respectively.