Turbulent flow of liquid steel and argon bubbles in slide-gate tundish nozzles: Part I. model development and validation

  • PDF / 4,313,056 Bytes
  • 15 Pages / 612 x 792 pts (letter) Page_size
  • 66 Downloads / 194 Views

DOWNLOAD

REPORT


I. INTRODUCTION

TUNDISH nozzle geometry is one of the few variables that is both very influential on the continuous casting process and relatively inexpensive to change. Slide gates are commonly used to control the flow rate, but their off-center throttling generates asymmetry. Argon injection into the nozzle is widely employed to reduce nozzle clogging. These variables all affect flow in the nozzle and, subsequently, in the mold. Poor flow in the mold can cause many quality problems.[1,2,3] There is, thus, great incentive to understand quantitatively how these variables affect the flow pattern in the nozzle and the jet characteristics exiting the outlet ports, as a step toward optimizing steel quality. Previous modeling studies of flow in nozzles have focused on single-phase flow. Hershey and co-workers[4,6] and Thomas and Najjar[5] assessed the accuracy of two-dimensional (2D) and three-dimensional (3-D) finite-element simulations of single-phase flow in a bifurcated submerged entry nozzle (SEN) through comparison with velocity measurements and water modeling observations. They demonstrated the reasonable accuracy of separating the nozzle and mold calculations and using 2-D simulations for some symmetrical flows. Their work was later extended[6] to perform an extensive parametric study of single-phase symmetrical flow in the nozzle. Wang[3] employed a 3-D finite-element single-phase model of a complete tundish nozzle (including the upper tundish nozzle (UTN), the slide gate, and the SEN) to confirm the asymmetrical flow caused by the slide gate. Yao[7] used a finite-volume method to model flow through the SEN and the mold together. Recently, Sivaramakrishnan[8] modeled transient flow in the HUA BAI, Senior Research Engineer, is with the Dow Chemical Company, Freeport, TX 77541. BRIAN G. THOMAS, Professor, is with the Department of Mechanical and Industrial Engineering, University of Illinois at Urbana–Champaign, Urbana, IL 61801. Manuscript submitted May 26, 2000.

METALLURGICAL AND MATERIALS TRANSACTIONS B

mold using large-eddy simulation and showed the importance of accurate modeling of the flow from the nozzle in achieving a good match with particle-image velocimetry (PIV) measurements. Experimental works have reported the importance of twophase flow in nozzles when argon is injected. Tsai[9] observed the partial vacuum pressure at the upper portion of an SEN in the water experiments and found that proper argon injection might avoid the vacuum pressure and, hence, reduce the air aspiration. Heaslip et al.[10] performed water model experiments to investigate the use of injected gas to carry alloying elements into the liquid. Burty et al.[2] observed a flowpattern transition from dispersed bubbly flow to “annular” flow where the gas and liquid separate. A criterion for this transition was developed based on water model experiments through stopper-rod nozzles that depend on both the gas flow rate and liquid flow. Sjo¨stro¨m et al.[11] performed an experimental study of argon injection and the aspiration of air