Simulation of the ingot extraction in the continuous casting process

Transcription

1 A R C H I V E S of F O U N D R Y E N G I N E E R I N G Published quarterly as the organ of the Foundry Commission of the Polish Academy of Sciences ISSN ( ) Volume 11 Issue 3/ /3 Simulation of the ingot extraction in the continuous casting process J. Szajnar, W. Sebzda* Foundry Department, Silesian University of Technology, Towarowa 7, Gliwice, Poland *Corresponding author. address: wojciech.sebzda@polsl.pl Received ; accepted in revised form Abstract Cast ingot pulling speed is significantly affecting the nature of the resulting structure and the quality of the outer surface of the ingot. By introducing a variable algorithm for extraction of the ingot we may to some extent control the shape and location of the solid / liquid interface and temperature field in the cross-section of the ingot. The shape of the crystallization front, as well as its position relative to mold plays an important role in the process of continuous casting ingots of grey iron and affects the structure of the casting. In order to verify the impact of an algorithm on the shape and the location of solid / liquid interface, a number of simulations in ANSYS Fluent 12 were made, for determining the shape of crystallization front and temperature distribution on the cross-section of the ingot. Keywords: Continuous Casting, Computer Simulation, Crystallization Front, Grey Cast Iron 1. Introduction Continuous casting is a growing interest in the industry. This is a relatively new technique of casting in the long history of foundry processes. The first attempts to use continuous casting have been carried out in the nineteenth century and were developed to improve the process. Interest in using this method was dictated by the possibility to cast different alloys and by the economic yield obtaining high quality products. Currently, this method is constantly improved and thanks to the use of new technologies and technical capabilities we can better understand this technological process [1]. An effective way to better understanding of this process is the use of experimental stands and numerical computer simulations that bring us closer to this process. By using these research methods we can accelerate the introduction of technical improvements in continuous casting. In this publication, the issue of the computer simulation of continuous casting of cast iron process was presented, whose research is conducted in the Department of Foundry, Silesian University of Technology. On the basis of the existing continuous casting line of cast iron in the Foundry Department, Silesian University of Technology, a model was developed on which the simulations were carried out in ANSYS Fluent Conduct of the study The main assumption of this study was to simulate the ingot pulling process in continuous casting of cast iron in order to determine the location of crystallization front and temperature distribution in the ingot and determining whether the defined extraction speed defined by a specific algorithm is feasible in real conditions. 3. Research Methodology The Foundry Department of the Silesian University of Technology conducts the work on the use of forced convection A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 1 1, I s s u e 3 / ,

2 of liquid metal in the crystallizer mounted in the continuous casting line. The effect of external factors (particularly electromagnetic field) on the solidifying metal was shown, among others in [2 7]. For the purpose of the research, a continuous casting line for 20 mm cast iron rods was designed and built (Fig. 1). Fig. 1. Continuous casting line in the Foundry Department of the Silesian University of Technology Continuous casting process covers many complex phenomena including turbulent multiphase fluid flow, heat transfer and solidification. Convection of molten metal in the crystallizer, and thus shifting and changing of the solid / liquid interface shape have a major impact on the quality of the ingot [2, 6, 7, 14, 15]. Implementation of the work required to set out for a series of simulations involving the following steps: 1. Designing of the crystallizer model in ANSYS Fluent 12 on the basis of actual mold used in the research: identifying the extraction algorithms in order to select the most optimal for use, performing a series of simulations to determine the parameters of the simulation. 2. Conducting the actual experiment on the continuous casting line in the Foundry Department: developing a plan for the real experiment according to the analysis of the results of computer simulation, performing the live experiment on the basis of the research plan. 3. Determination on the basis of simulation and real tests, the correlation between the obtained results. Parameters of the simulation environment: Solver segregated, implicite, two-dimensional, unsteady, and absolute velocity formulation, Viscosity model turbulent k-ε, Materials grey cast iron GJL-250, 1H18N9T steel, graphite, Initial conditions inlet ingot temperature = 1450 C, instantaneous speed of ingot pulling = 0,02 m/s. Setting the specific simulation environment was chosen on the basis of information provided in the works [8 13]. Subsequently, to simulate the models shown in Fig. 2 and 3 ingot pulling algorithms have been proposed. Table 1 compares different times of the forward, backward motion and stopping of the ingot extraction. Table 1. Ingot pulling algorithm Motion Time, s Motion Type A B C forward 10 5 pause 0,5 0,5 only forward movement backward 1 1 pause 0,5 0,5 Before the start of the selected simulation algorithm the ingot was held out in the crystallizer over 5 seconds. Fig. 2. Model I of ingot - crystallizer system used in ANSYS FLUENT 12:1 cooler, 2 crystallizer, 3 Ingot Fig. 3. Model II of ingot - crystallizer system used in ANSYS FLUENT 12:1 cooler, 2 crystallizer, 3 Ingot 94 A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 11, I s s u e 3 / ,

3 4. Results and analysis 18 simulations were preformed, which aimed to determine the temperature field as well as the shape and position of the front a) of crystallization. The paper presents the simulation results for algorithms A, B and C, recognized as the most optimal due to the location of the crystallization front and the temperature at the outlet of the crystallizer. b) Fig. 4. Algorithm A simulation model I (respectivly liquid phase amount and temperature field) a) forward movement after 45 s, b) forward movement after 70 s a) A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 1 1, I s s u e 3 / ,

4 b) Fig. 5. Algorithm B simulation model I (respectivly liquid phase amount and temperature field) a) forward movement after 75 s, b)backward movement after 77 s a) b) Fig. 6. Algorithm C simulation model I (respectivly liquid phase amount and temperature field) a) forward movement after 80 s, b) backward movement after 82 s 96 A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 11, I s s u e 3 / ,

5 a) b) Fig. 7. Algorithm C simulation model II (respectivly liquid phase amount and temperature field) a) forward movement after 80 s, b) backward movement after 82 s a) A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 1 1, I s s u e 3 / ,

6 b) c) d) Fig. 8. Comparison of model I and II for the C algorithm (a,b liquid phase amount, c,d temperature field) a), c) forward movement after 80 s, b), d) backward movement after 82 s In Figures 4 to 6 simulation results are presented in the form of temperature field and the liquid phase amount corresponding to the algorithms A, B and C. It can be said that the algorithm A is not suitable for use in the real process of continuous casting, because it would result in pouring liquid metal from the crystallizer. Simulations allow us to conclude that shortening the forward movement causes a displacement of the solid / liquid interface into the mold (i.e. in the direction of the oven.) As a result of the crystallization front shifting, the temperature at the outlet of the mold was also lowered. 98 A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 11, I s s u e 3 / ,

7 Below Figure 9 shows a view of the external surface of the continuous casting ingot cast at the extraction instantaneous speed of 0.02m / s. Fig. 9. The view of the surface of the ingot cast at the instantaneous extraction speed of 0.02 m/s (frame shows the typical print resulting from the stopping and backward movement of the ingot) In Figures 7 and 8 we can see that the change in model II, which considers the ingot even at a distance of 200 mm from the mold, allowed to eliminate distortions that arise in the model I at the output of the ingot. Because of that in the model II the crystallization front is more stable and the share of liquid phase is reduced in comparison to the whole process in the model I, which can be seen in Figures 8a and 8b. This conclusion could not be reached while analyzing only the temperature field, because the fields are quite similar. It may be noted that in Figure 8c in the model I the temperature of the ingot is higher at the exit of the mold, but this is a small change compared to model II. However, it may be stated that this change in the temperature field can affect the increase of the liquid phase in the model I accordingly to model II. Analyzing the results of simulation and view of the external surface of cast iron rod (Fig. 9) it can be concluded that simulations B and C (Fig. 5 and 6) well reflect the position of solid / liquid interface in a real casting. 5. Summary From the simulations in Ansys Fluent 12 following conclusions can be drawn: the introduction of backward motion significantly affects the shape and position of the solid / liquid interface and temperature field in the ingot cross section, backward movement contributes to the crystallization front shifting in the direction of entry of metal into the crystallizer, the shifting of the crystallization front and temperature field in the direction to the crystallizer entry allows for higher pulling speeds of the ingot. This observation was confirmed by the results of the actual conditions of continuous casting of 20mm cast iron rods. Literature [1] R. D. Pehlke, Computer simulation of solidification processes The evolution of a Technology, Metallurgical and Materials Transactions, vol. 33A, August 2002, pp [2] J. Szajnar, M. Stawarz, T. Wróbel, W. Sebzda, Structure and properties of gray iron casted in the electromagnetic field, Archives of Foundry Engineering, Volume 9, Issue 3/2009, pp [3] J. Szajnar, The influence of selected physical factors on the crystallization process and castings structure, Archives of Foundry Engineering, vol. 9, No. 1M [4] J. Szajnar, The columnar to equiaxed transition at casting solidification with convection forced by rotating magnetic field, Scientific book of Silesian University of Technology, Mechanic, 138, (2001). [5] Li Qiushu, Liu Liqiang, Li Renxing, Hou Xu, Zhai Qijie, Effect of pulse magnetic field on graphite morphology and solidification of grey cast iron, PROCEEDINGS vol.1, 66th World Foundry Congress Istanbul, 2004, pp [6] J. Szajnar, M. Stawarz, T. Wróbel, W. Sebzda, Influence of electromagnetic field on pure metal san alloys structure, Journal of Achievements in Materials and Manufacturing Engineering, vol. 34, No. 1, May 2009, pp [7] J. Szajnar, M. Stawarz, T. Wróbel, W. Sebzda, Influence of electromagnetic field parameters on the morphology of graphite in grey cast iron, Archives of Foundry Engineering, vol. 9, No. 1/2009, pp [8] P. Ramirez-Lopez, R.D. Morales, R. Sanchez-Perez, L.G. Demedices, O. Davila, Structure of Turbulent Flow in a Slab Mold, Metallurgical and Materials Transactions, vol. 36B, December 2005, pp [9] X.K. Lan, J.M. Khodadadi, F. Shen, Evaluation of Six k-ε Turbulence Model Predictions of Flow in a Continuous Casting Billet-Mold Water Model Using Laser Doppler Velocimetry Measurements, Metallurgical and Materials Transactions, vol. 28B, April 1997, pp.321. [10] B. Zhao, B.G. Thomas, S.P. Vanka, R.J. O Malley, Transient Fluid Flow and Superheat Transport in Continuous Casting of Steel Slabs, Metallurgical and Materials Transactions vol. 36B, December 2005, pp [11] M. Reza Aboutalebi, M. Hasan, R.I.L. Guthrie, Coupled Turbulent Flow, Heat, and Solute Transport in Continuous Casting Processes, Metallurgical and Materials Transactions vol. 26B, August 1995, pp [12] H.M. Tensi, C. Xu, R. Rösch, Konvektion bei gerichteter Erstarrung, ȁ nlich dem Stranggiessen, Giessereiforschung 47 Nr. 3, 1995, pp A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 1 1, I s s u e 3 / ,

8 [13] Debasish Chatterjee, Dipak Mazumdar, Sujay Pandit Patil, Physical and Mathematical Modeling of Two-Phase Flows in a Hollow Jet Nozzle, Metallurgical and Materials Transactions vol. 38B, October 2007, pp [14] J. Szajnar, M. Stawarz, T. Wróbel, W. Sebzda, B. Grzesik, M. Stępień: Influence of continuous casting conditions on grey cast iron structure, Archives of Materials and Engineering, Nr 1, vol. 42, 2010, pp [15] J. Szajnar, M. Cholewa, M. Stawarz, T. Wróbel, W. Sebzda: Inoculaction of grey vast iron with use of electromagnetic field, Livarski Vestnik, Nr 3, vol. 57, 2010, pp A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 11, I s s u e 3 / ,