Vibration-Based Monitoring of Gas Stirring in Steelmaking Ladles: Effects of Fluid Properties and Operational Parameters

Abstract

Gas stirring is widely used in Ladle Metallurgy to enhance chemical reactions and the removal of Non-Metallic Inclusions (NMIs) prior to the production of high-quality steel and high-niched alloys. Although gas stirring is of major importance, accurate techniques have yet to be developed for improved monitoring and control of the stirring process in steelmaking ladles. Indirect measurements such as vibration measurements have shown an increased potential for monitoring and control of the stirring process. Vibration measurements at laboratory scale as well as at industrial scale ladles are valuable for the development of process monitoring and control tools. However, most of the existing research has relied on laboratory scale models using water as a modelling fluid, whose physical properties differ significantly from those of molten steel. The physical properties of the involved fluids have been shown to affect significantly the characteristics of the two-phase flows.Therefore, investigations using mechanistic models filled with fluids that more closely resemble the properties of molten steel are essential for better understanding e.g. the vibrations in steelmaking ladles.

In this work, vibration measurements were performed during argon injection in two different modelling fluids: Sn–40 wt% Bi alloy at 200°C and water at room temperature, utilizing the same experimental ladle. The experiments were conducted at the LIMMCAST laboratory facilities at the Helmholtz-Zentrum Dresden-Rossendorf in Germany. Additionally, vibration measurements were carried out in an industrial steelmaking ladle during vacuum degassing (VD) at Uddeholms AB, Hagfors, Sweden.

At the LIMMCAST testing facility, vibration measurements were conducted for argon injection in the liquid domain through different types of stirring plugs as an orifice plug, porous plug and slit plug. The measurements were performed at ambient pressure for both Sn–40 wt% Bi and water, as well as at reduced pressure on the top free surface of the liquid Sn–40 wt% Bi. The relationship between vibrational amplitude and flow rate differed between the two examined fluids, despite utilizing the same measuring setup and experimental ladle. The experiments revealed distinct characteristic frequency ranges corresponding to various bubble evolution events from bubble formation at the stirring plugs to rise and collapse at the top free surface of the liquids.Variations in the physical properties of water resulted in modification of the size and number of produced bubbles, shifting the vibrational frequencies to higher values. Furthermore, bubble shape affects vibrational characteristics where elongated bubbles from slit plugs generate higher-frequency vibrations due to increased shape oscillations and bubble-liquid interaction. Slit width and number influence plume dynamics where narrower, more numerous slits enhance bubble frequency and the resulting turbulence, raising the vibrational amplitude. Reduced pressure on the top free surface of the liquid metal results in the formation of larger bubbles, shifting vibrational frequencies to lower values and increasing the amplitude.

Vibration measurements in an industrial ladle showed good agreement with those obtained at LIMMCAST filled with Sn–40 wt% Bi, both in the pattern of vibrational amplitude increase with rising flow rate and in the observed vibrational frequencies. Analysis of low-frequency ranges (100–300 Hz) in steel making ladles provided insights into bubble formation and gas injection conditions. In contrast, high-frequency ranges (900–1600 Hz) offered information about stirring dynamics and top surface movements. Vibration measurements provide an effective means to assess mixing conditions in ladles during VD and can aid in developing stirring monitoring models. While IR cameras capture temperature distribution and surface movement characteristics, limitations exist in detecting subtle flow phenomena such as soft stirring. Vibration-based models could complement IR monitoring by assessing even soft stirring, enabling improved process control.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-365378