86 411-84174804

Technical support and construction service team

Simulation study on liquid steel flow and inclusion removal in bottom-blown gas tundish

Contact Us

Special product design, please send specific data and drawings to our mailbox or form.

Table of Contents

Utilizing a 1:3 water model experiment, we measured the flow field of molten steel within the six-flow tundish employed in billet continuous casting, specifically designed for a 40t tundish. The effects of five different bottom blowing schemes on the flow characteristics and inclusion removal of molten steel in the tundish were studied.

Keywords: tundish; bottom blow; water simulation; inclusions

Bottom air blowing enhances molten steel flow, aiding in non-metallic inclusion removal for cleaner steel. This study investigates the effect of varied bottom blowing locations on tundish flow, using a water model and polystyrene particles to mimic inclusions.

Water simulation experiment

Experimental principle

To ensure both geometric and dynamic similarity between the model and the prototype, we adopt the principle of similarity by utilizing Froude’s quasi-number equality. In actual production, the steel ladle capacity is 90t and the intermediate ladle capacity is 40t. This experiment is based on the 6-stream tundish of billet continuous casting, with a similarity ratio of 1:3. We employ an organic glass model to replicate the tundish, utilizing tap water as a substitute for molten steel and compressed air to mimic argon gas (depicted in Figure 1). During the experiment, we utilize dispersed breathable brick 3.The figure, labeled as Figure 2, provides a clear depiction of the tundish’s structure along with the precise positions for air blowing.

Figure 1 Schematic diagram of water model experimental device

Figure 2. Schematic diagram of the prototype structure of the tundish and the air blowing position in the water simulation experiment​

Experimental methods

Based on the similarity principle, the volume flow rates relate as Qm = λ^5.2Qp for model and prototype. During experimentation, a stabilized tundish flow field is perturbed by pulsing saturated KCl solution through the ladle’s long nozzle. Conductivity probes track temporal conductivity changes at each outlet. Upon reaching a set time, RTD curves are plotted using conductivity meter data for analysis.

Given the tundish’s symmetry, conductivity is measured at outlets 1, 2, and 3 on one side (Fig. 2). The study adopts Ahuja et al.’s model to calculate and analyze volume fractions of the tundish’s dead zone (V4), plug flow (V), and total mixed flow (Vm). Additionally, standard deviation (S) assesses flow characteristic dispersion across outlets.

Six experimental scenarios are explored: (1) Undisturbed tundish (Scheme Y). (2) Air blown at flow injection area ① (Plan I). (3) Air blown between proximal and middle outlets at ② (Plan II). (4) Air blown between middle and far outlets at ③ (Plan III). (5) Plan II with a square turbulator at ④ in the flow injection area (Option IV). (6) Plan III enhanced with a square turbulator in the flow injection area (Plan V).

We utilized polystyrene plastic particles, with a density of 640kg/m³, to mimic inclusions. To measure the removal of these inclusions, we employed a stepwise addition method, utilizing alcohol as a dispersant for the particles. With each injection flow from the ladle, a predetermined quantity of particles was introduced, and these particles were subsequently collected using sieves positioned at the outlets of the tundish’s six flows. The collection process was halted after 1200 seconds, which is double the theoretically calculated residence time. Following the drying of the collected particles, we conducted a statistical analysis to assess the effectiveness of inclusion removal.

Experimental results and analysis

(1) Non-blowing tundish experiment results in Table 1 and Figure 3(a) show a short peak time (avg. 64s) and high conductivity peak, indicating rapid outflow and single flow path causing short-circuit flow.

(2) Dead zone ratio hits 60%, limiting full mixing of old and new steel, resulting in uneven temperatures across streams.

(3) The “peak-shaped” RTD curve indicates brief tundish residence time, hindering inclusion flotation due to its sharp variations.

Table 1 .The residence time and volume fraction of the flow mode of molten steel in the tundish measured experimentally for each scheme

plan nozzle Dead time (tmin)/sPeak time (tmax)/s Actual average dwell time/sFlow pattern/%
piston areadead zonefull mixed zone
No blowingaverage value41642578.3959.0632.55
Ystandard deviation2926184.433.001.43
planaverage value539335611.5943.4444.97
standard deviation5319344.095.471.57
planaverage value3716738816.2438.1045.66
standard deviation2750745.9411.705.91
planaverage value389634010.7445.8443.42
standard deviation2192478.947.482.89
planaverage value3816239916.0535.8148.14
standard deviation2379548.168.670.58
planaverage value468332510.3548.0541.60
standard deviation2560686.8510.894.28

From Table 1, installing the breathable brick at position ① enhances stagnation and peak times versus no air blowing. Average flow times are 53s and 93s, with low standard deviation indicating consistent flow characteristics. The dead zone ratio decreases from 59.06% to 43.44%. Air blown through the brick creates fine bubbles rising upwards, mitigating turbulence from the ladle’s long nozzle and dispersing flow. This lengthens steel residence in the ladle, aiding inclusion flotation, and reduces impact on tundish levels and refractory erosion. Figure 3(b) depicts a reduced peak conductivity and flatter RTD curve trend when air is blown at position ①.

Installing breathable bricks at position ② between near- and mid-flow nozzles (Table 1, Fig. 3c,e) results in flatter RTD curves approaching normal distribution. Peak time and actual residence time significantly increase, with a reduced dead zone ratio of 38.10%. Scheme II doubles the piston area ratio of Scheme Y, with low standard deviation across indicators. Bubbles from the brick evenly stir the steel, improving flow distribution, activating tundish steel, and extending residence time for inclusion flotation. Bubbles also capture inclusions, enhancing collision and removal. Adding a square turbulator in the flow injection area (Plan IV) further reduces peak conductivity, improves outlet synchronization, lowers dead zone to 35.81%, and optimizes tundish flow.

Positioning the breathable brick at location ③, situated between the middle flow and the far flow outlet (refer to Figure 3d and f), significantly enhances the flow characteristics of the far flow, while having minimal influence on the near flow and middle flow. As far as the entire tundish is concerned, the effect is not ideal.

Figure 4 shows better inclusion removal at the far flow outlet compared to near and middle flows. Blowing at position ① reduces turbulence, disperses flow, and extends residence time. High turbulence in the injection zone breaks bubbles into smaller ones, carrying inclusions to the surface. Blowing at position ② obstructs with an air curtain, worsening removal near outlets compared to position ①. Blowing at position ③ shows no significant improvement over non-blowing Plan Y. Adding turbulence devices in the injection area (comparing II vs. V, III vs. V schemes) has minimal impact on inclusion removal when blowing at positions ② and ③.

Figure 3 Residence time distribution (RTD) curve of molten steel in tundish:. Plan Y, no air blowing (a); Air blowing plan-1(b), l(c), III(d), N(e), V(f)

Figure 4 Experimental results of inclusion removal using different solutions


(1) Without air blowing, the prototype tundish exhibits short-circuit flow, high steel flow rates eroding refractory materials and covering agents. Short stagnation and residence times hinder inclusion flotation; high dead zone ratios disrupt composition and temperature uniformity.

(2) Blowing at position ① in the injection area diverts steel flow, prevents short-circuits, reduces turbulence, and extends tundish residence time. High turbulence in the zone breaks bubbles into smaller ones, enhancing inclusion flotation for optimal removal.

(3) Blowing at position ② slightly diminishes inclusion removal near outlets compared to position ①. Blowing at position ③ only improves flow at the far outlet. Adding a turbulator when blowing at ② or ③ does not enhance inclusion removal as effectively as blowing in the injection zone.

LMM YOTAI established in 2007. Our production technology comes from Japanese Yotai. As an experienced and international player in the refractories industry. We have succeeded in expanding both the breadth of its product range and the depth of its services. From raw material selection, refractory portofio & optimization, installation & services & recycle of used refractories on site to further reduce client’s Opex & Capex in refractory consumption per ton steel output, meanwhile improve product quality of client.

Our Product have been supplied to world’s top steel manufacturer Arcelormittal, TATA Steel, EZZ steel etc. We do OEM for Concast and Danieli for a long time


Comprehensive solution of Refractory for steelmaking

Fill out the form below, and we will be in touch shortly.
Contact Information