UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mould taper, heat transfer and spray cooling in high speed continuous casting Fu, Junlong 2001

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2001-0196.pdf [ 9.26MB ]
Metadata
JSON: 831-1.0090040.json
JSON-LD: 831-1.0090040-ld.json
RDF/XML (Pretty): 831-1.0090040-rdf.xml
RDF/JSON: 831-1.0090040-rdf.json
Turtle: 831-1.0090040-turtle.txt
N-Triples: 831-1.0090040-rdf-ntriples.txt
Original Record: 831-1.0090040-source.json
Full Text
831-1.0090040-fulltext.txt
Citation
831-1.0090040.ris

Full Text

MOULD TAPER, HEAT TRANSFER AND SPRAY COOLING IN HIGH SPEED CONTINUOUS CASTING B Y JUNLONG FU B. A. Sc. University of Science and Technology Beijing, 1984 M. A. Sc. University of Science and Technology Beijing, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF METALS AND MATERIALS ENGINEERING We accept this thesis as confirming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A March 2001 © Junlong Fu, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada Date AT'1 ^ 1 DE-6 (2/88) ABSTRACT Competition around the world and market expectations are driving the steel industry to improve billet quality and lower costs by increasing casting speed of continuous casters. The demand for high casting speeds can be achieved by increasing mould length and by making upgrades to the secondary cooling system. A plant trial was conducted at Co- Steel Lasco for the purpose of assessing the performance of a 1016 mm long mould having a parabolic taper on the high speed casting of 6 grades of steel billets. The mould was instrumented with 52 thermocouples. The billet section size was 178x127 mm, cast at speeds in the range of 2.2-2.9 m/min. Additionally, three linear variable displacement transducers were installed on the mould wall to monitor the mould oscillation. Billet samples were collected for several operating conditions. A n inverse heat conduction model was used to calculate mould heat fluxes from measured mould wall temperatures. Existing mathematical models were employed to investigate mould-billet interaction and the adequacy of the mould taper. It was shown that steel carbon content and casting speed had significant effects on heat flux in the mould. Heat flux was highest for medium carbon steel followed by high carbon steel; low carbon steel had the least amount of heat transfer. Increases in casting speed consistently led to heat flux increases, although not proportionately. Off-corner internal cracks and narrow face concavity were noted on all billet samples. It was demonstrated that the taper was too tight for the low carbon grades which caused squeezing of the shell by the mould and was responsible for the off-corner internal cracks and the narrow face concavity of the billets. For the medium and high carbon billets the ii mould taper was inadequate especially in the lower part of the mould. Here, it was likely that bulging of the broad face and corner rotation gave rise to longitudinal depressions on the narrow face and off-corner internal cracks. Optimum tapers were recommended respectively for low carbon, medium carbon and high carbon steels. Spray cooling systems for three companies were also investigated. Billet surface temperature was calculated with a mathematical model; the effect of surface reheating between mould and spray cooling, between different spray cooling zones and between spray cooling and radiation as well as the billet liquid pool depth were examined It was shown that midway cracks can result from the billet surface reheating in the secondary cooling zone. Additional spray zones and a longer spray chamber were shown to decrease the occurrences of surface reheating and thereby mitigating midway cracking. Finally, recommendations were made for a spray cooling chamber appropriate for high speed casting. in Table of Contents ABSTRACT ii LIST OF TABLES ix LIST OF FIGURES x LIST OF SYMBOLS xv ACKNOWLEDGEMENTS xv i i . CHAPTER 1 - INTRODUCTION l CHAPTER 2 - LITERATURE REVIEW 4 2.1 Heat Transfer in the Mould 6 2.1.1 Mould with Different Lubricants 7 2.1.2 Influence of Superheat 8 2.1.3 Influence of Steel Grade 9 2.1.4 Influence of Casting Speed 10 2.2 Billet Shrinkage and Mould Taper in High Speed Casting 11 2.2.1 Billet Shrinkage in the Mould 11 2.2.2 Mould Taper 11 2.2.3 Mould Taper for High Speed Continuous Casting 13 iv 2.3 Secondary Cooling Zone 14 2.3.1 Water Spray and Air-mist Spray 14 2.3.2 Factors to Influence the Heat Transfer of Spray Cooling Zone 15 2.3.3 Billet Surface Temperature and Reheating 17 2.4 Defects in Continuous Casting 18 2.4.1 Mechanical Properties of the Steel at High Temperature 20 2.4.2 Defects Related to the Mould 23 2.4.2.1 Oscillation Marks 23 2.4.2.2 Rhomboidit 24 2.4.2.3 Off-corner Internal Cracks 26 2.4.3 Defects Related to the Secondary Cooling Zone 27 2.4.3.1 Midway Cracks 27 2.4.3.2 Diagonal Cracks 28 CHAPTER 3 - SCOPE AND OBJECTIVES 30 CHAPTER 4 - INDUSTRIAL PLANT TRIAL 32 4.1 Caster Details and Nominal Operating Practice 32 4.2 Mould Temperature Measurement 33 4.3 Metal Level and Casting Speed 34 4.4 Mechanical Signals 34 4.5 Billet Samples 34 v CHAPTER 5 - RESULTS OF PLANT TRIALS 41 5.1 Billet Quality Evaluation 41 5.1.1 Surface Quality 41 5.1.2 Off-corner Internal Cracks 41 5.1.3 Off-squareness 41 5.1.4 Midway Cracks 42 5.1.5 Oscillation Mark Depth 42 5.2 Oscillator Performance 42 5.3 Mould Temperature 43 5.3.1 Mould Temperature Response 43 5.3.2 Time-average Mould Temperature Distribution 43 CHAPTER 6 - MODELLING OF MOULD HEAT TRANSFER 64 6.1 The Mould Heat Transfer Model 64 6.2 Heat Flux Calculation 67 6.2.1 Broad and Narrow Face Heat Flux Profiles 67 6.2.2 Comparison of Lasco Data with Results from Other Plants 68 6.3 Influence of Process Factors on Mould Heat Flux 69 6.3.1 The Influence of Steel Carbon Content on Mould Heat Flux 69 6.3.2 The Influence of Casting Speed on Mould Heat Transfer 69 6.3.3 The Influence of Cooling Water Velocity on Mould Heat Transfer. 70 6.3.4 The Influence of Oil Flow Rate on Mould Heat Transfer 70 6.3.5 The Influence of Superheat on Mould Heat Transfer 70 vi 6.3.6 The Influence of Metal Level on Mould Heat Transfer 70 6.4 Difference Between Heat Fluxes Predicted by Model and by Cooling Water 71 CHAPTER 7 - SHRINGKAGE CALCULATION AND MOULD TAPER DESIGN 86 7.1 Mathematical Model of Billet Shrinkage 86 7.2 Shell Growth and Billet Surface Temperature 87 7.3 Billet Shrinkage 88 7.4 Mould Behavior and Billet Quality 89 7.4.1 Mould Taper and Related Defects 89 7.4.2 Mould Heat Transfer, Metal Level Fluctuations and Off-squareness 90 7.5 Taper Design 91 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING 109 8.1 Plant Data 109 8.2 The Spray Heat Transfer Model 110 8.3 Results and Analysis 112 8.3.1 Shell Thickness and Surface Temperature 112 8.3.2 Surface Reheating 112 8.3.3 Metallurgical Length 113 8.3.4 Spray-related Defects 114 8.4 Sprays design 116 vii CHAPTER 9 - CONCLUSION 128 REFERENCES 131 viii LIST OF TABLES Table 4.1 Casting machine specifications 36 Table 4.2 Mould details 36 Table 4.3 Parabolic mould taper 37 Table 4.4 Summary of heat chemistry 38 Table 4.5 Casting parameter changes and billet sampling 40 Table 5.1 Billet surface evaluation summary 45 Table 5.2 Time-average temperature and standard deviation on inside curved wall 60 Table 5.3 Time-average temperature and standard deviation on left straight wall 60 Table 5.4 Time-average temperature and standard deviation on outside curved wall 61 Table 5.5 Time-average temperature and standard deviation on right straight wall 61 Table 6.1 Heat flux analysis 75 Table 7.1 Shrinkage analysis 97 Table 7.2 Predicted mould tapers 108 Table 8.1a Supplied and calculated data of spray system from three companies 118 Table 8. l b (con't) Supplied and calculated data of spray system from three companies 118 Table 8.2 Surface reheating and depth of liquid pool at different plants 120 Table 8.3 Alta Steel spray cooling design 124 ix LIST OF FIGURES Fig. 2.1 Three zones of heat removal in a continuous casting machine 5 Fig. 2.2 Heat transfer profile over the mould length 7 Fig. 2.3 Affect of carbon content on mould heat flux 9 Fig. 2.4 Heat flux down the length of the mould for various casting speeds 10 Fig. 2.5 Influence of surface temperature on spray heat-transfer coefficient, h 16 Fig. 2.6 Influence of water flux on spray heat transfer coefficient 17 Fig. 2.7 Schematic drawing of strand cast section showing different types of cracks 20 Fig. 2.8 Scanning electron micrograph of surface of a crack formed near the solidus temperature. The inclusions are sulfides 22 Fig. 4.1 Thermocouple layout (all faces) 39 Fig. 5.1 Photograph of longitudinal depression on a billet sample from heat 664, medium carbon steel 46 Fig. 5.2 Lasco Steel Trial December 1998, narrow face concavity and midway cracks (High carbon steel, 1.99m/min, heat 665) 47 Fig. 5.3 Lasco Steel Trial December 1998, midway and off-corner internal cracks (Low carbon steel, 2.59m/min, heat 643) 48 Fig. 5.4 The influence of steel carbon contents on off-squareness 49 Fig. 5.5 Oscillation marks for low, medium and high carbon steel 50 Fig. 5.6 The measured mould velocity calculated from the measured mould displacement for normal casting practice of a 0.2 % C steel cast at 2.16 m/min for heat 668 51 Fig. 5.7 Mould thermal response of thermocouples located on the inside of the curved wall for heat containing 0.155 pet. carbon during casting speed change at approximately 1480 s 52 x Fig. 5.8 Mould thermal response of thermocouples located on the inside of the curved wall for heat containing 0.155 pet. carbon during casting speed change at approximately 1480 s 53 Fig. 5.9 Mould thermal response of thermocouples located on the left straight wall for heat containing 0.155 pet. carbon during a casting speed change at approximately 1480 s 54 Fig. 5.10 Mould thermal response of thermocouples located on the left straight wall for heat containing 0.155 pet. carbon during a casting speed change at approximately 1480 s 55 Fig. 5.11 Time-averaged mould temperature distribution on the inside curved wall for heat 642 containing 0.155 pet. carbon 56 Fig. 5.12 Time-averaged mould temperature distribution on the outside curved wall for heat 642 containing 0.155 pet. carbon 57 Fig. 5.13 Time-averaged mould temperature distribution on the left straight wall for heat 642 containing 0.155 pet. carbon 58 Fig. 5.14 Time-averaged mould temperature distribution on the right straight wall for heat 642 containing 0.155 pet. carbon 59 Fig. 5.15 Time-averaged mould temperature distribution on the inside curved wall for heats containing carbon levels of 0.157 and 0.412 pet. respectively 62 Fig. 5.16 Time-averaged mould temperature distribution on the inside curved wall for heat 664 containing 0.412 pet. carbon for casting speeds of 1.85 and 2.29 m/min respectively 63 Fig. 6.1 Schematic diagram of the midface longitudinal section of the mould wall 65 Fig. 6.2 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.155 pet. (heat 642) 72 Fig. 6.3 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.412 pet. (heat 664) 73 Fig. 6.4 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.767 pet. (heat 665) 74 Fig. 6.5 Comparison of axial mould heat flux profiles at three companies for high carbon steels 76 xi Fig. 6.6 Comparison of axial mould heat flux profiles at four companies for medium carbon steels 77 Fig. 6.7 Comparison of axial mould heat flux profiles at two companies for low carbon steels 78 Fig. 6.8 The influence of steel carbon content of average mould heat Transfer 79 Fig. 6.9 The influence of casting speed on axial mould heat flux profiles for a high carbon steel (heat 665) 80 Fig 6.10 The influence of mould water velocity on axial mould heat flux profiles (heat 645) 81 Fig 6.11 The influence of mould oil flow rate on axial mould heat flux Profiles (heat 644) 82 Fig. 6.12 The influence of superheat on axial mould heat flux profiles 83 Fig. 6.13 The influence of metal level location on axial mould heat flux profiles on the outside curved wall (heat 685) 84 Fig. 6.14 Graph showing the match between predicted heat extraction rate in the mould and the heat extracted by the mould cooling water 85 Fig. 7.1 Mesh used for modeling one quarter of a transverse section of a billet 93 Fig. 7.2 Predicted billet surface temperature and shell thickness profiles for heat 642 containing 0.155 pet. carbon 94 Fig. 7.3 Predicted billet surface temperature and shell thickness profiles for heat 664 containing 0.412 pet. carbon 95 Fig. 7.4 Predicted billet surface temperature and shell thickness profiles for heat 665 containing 0.767 pet. carbon 96 Fig. 7.5 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.155 pet.(heat 642) 98 Fig. 7.6 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.183 pet.(heat 643) 99 xii Fig. 7.7 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.412 pet.(heat 664) 100 Fig. 7.8 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.767 pet.(heat 665) 101 Fig. 7.9 Schematic diagram showing the development of concavity on the narrow face due to excessive squeezing at the corners 102 Fig. 7.10 Schematic diagram showing the formation of off-corner internal cracks due to bulging at the broad face and rotation of the corners 103 Fig. 7.11 The influence of casting speed on off-squareness 104 Fig. 7.12 The relationship between off-squareness and temperature difference of adjacent faces of the mould wall at 950 mm below the top of the mould 105 Fig. 7.13 The relationship between off-squareness and standard deviation of the thermocouple near the meniscus 106 Fig. 7.14 Calculated mould tapers for the narrow face of the 5 x 7 inch mould 107 Fig. 7.15 Calculated mould tapers for the broad face of the 5 x 7 inch mould 108 Fig. 8.1 Schematic drawing of transverse slice 110 Fig. 8.2 Temperature distribution and shell thickness in Alta continuous casting (low carbon steel, 152*152 mm 2 , casting speed 1.9 m/min) 121 Fig. 8.3 Billet temperature and shell thickness at Alta Steel, high carbon steel, 152x152 mm 2 , casting speed 1.9 m/min 122 Fig. 8.4 Billet temperature and shell thickness at McMaster Steel, low carbon steel, 150x150 mm 2 , casting speed 2.3 m/min 122 Fig. 8.5 Billet temperature and shell thickness at McMaster Steel, high carbon steel, 150x 150 mm 2 , casting speed 2.3 m/min 123 Fig. 8.6 Alta Steel: billet temperature and shell thickness for designed sprays high carbon steel, 203x203 mm 2 , casting speed 1.25 m/min 125 Fig. 8.7 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 203x203 mm 2 , casting speed 1.25 m/min 125 xiii Fig. 8.8 Alta Steel: billet temperature and shell thickness for designed sprays high carbon steel, 152x 152 mm2, casting speed 2.4 m/min 126 Fig. 8.9 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 152*152 mm2, casting speed 2.4 m/min 126 Fig. 8.10 Alta Steel: billet temperature and shell thickness for designed sprays high carbon steel, 120x120 mm2, casting speed 3.6 m/min 127 Fig. 8.11 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 120x120 mm2, casting speed 3.6 m/min 127 xiv LIST OF SYMBOLS Cp specific heat (J kg"1 °C"') Cpw specific heat of water (J kg"1 °C~l) Cpm specific heat of mould (J kg"1 °C"') dw width of cooling-water channel gap (mm) Dh hydraulic diameter (m) fs fraction of solid h heat transfer coefficient (W m"2 °C~l) hr effective heat transfer coefficient due to radiation (W m"2 "C"1) hw heat transfer coefficient at the mould/cooling-water interface (W m"2 °C"1) k thermal conductivity (W m'1 °C"') k m thermal conductivity of mould walls (W m'1 °C"') q heat flux (W m"2) Qfc forced convection heat flux (W m"2) t time (s) tN negative-strip time (s) T a ambient temperature (°C) T s temperature of billet surface (°C) T w temperature of water (°C) V w velocity of cooling water in channel (m/s) W spray water flux (/ m"2 s"1) x, y, z spatial coordinates p density (kg in"3) xv u. viscosity of fluid (N s m"2) o Stefan-Boltzmann constant (5.6703xl0" 8 W m"2 K" 4 ) 8 radiation emissivity Subscripts / fluid / liquid m mould w water xvi ACKNOWLEDGMENTS I would like to express my sincere gratitude to my research supervisor Dr. Indira Samarasekera for giving me an opportunity to work with her and for providing excellent guidance and valuable encouragement throughout the course of my graduate study and this research work. I also wish to thank the University of British Columbia and the Natural Sciences and Engineering Research Council of Canada for the strategic research project of high speed billet casting. The assistance of staff and students of billet casting research group: M r . Neil Walker and M r . Gary T. Lockhart who conducted the industrial plant trial, Baofeng Wang, Cindy Chow, Joungkuil Park, are deeply appreciated. I am also grateful to Mary Jansepar, Joan Kitchen, Dianfeng L i , Zhengdong L iu and Xuzhan Ren for their friendship. xvii CHAPTER 1 -INTRODUCTION CHAPTER 1 - INTRODUCTION Continuous casting of steel has been established worldwide for about thirty years. The process involves pouring of molten steel into a water-cooled copper mould and the continuously withdrawing a partial solidified strand. The oscillation of the mould and a constant supply of lubricant onto the mould wall are required to avoid sticking of the solidifying shell to the copper mould. As soon as the solidified shell is sufficiently thick to contain the liquid steel, the strand leaves the mould and is further cooled by water sprays.'11 The technology of continuous casting always aims to increase the casting speed, that means strand productivity, in conjunction with better internal and surface quality of the cast product.'21 Recently many billet producers are seeking to operate at significantly higher speeds by using longer moulds of the order of 1000 mm and effective secondary cooling. Several machine builders , 2"4 ] have developed specific designs, generally based on a longer mould tube, to achieve high casting speeds with oil or mould powder lubrication. Casting speeds of 3.0 to 4.5 m/min have been achieved in billet castings. '3"5] Furthermore, the casting speed of 5.0 m/min for low carbon steel with low viscosity powder was also reported in a pilot continuous casting machine. ' 5 1 In order to cast high quality billets with high casting speed, heat transfer in the mould and the interaction between the mould and the strand must be emphasized. Heat transfer from molten steel to the walls of the mould is controlled largely by conduction across the air gap which forms as the solidified shell shrinks. To compensate for billet shrinkage, mould walls are 1 CHAPTER 1 - INTRODUCTION tapered inwards; the resulting reduction in the air gap improves the rate of mould heat extraction and decreases the surface temperature of the billet at the mould exit and thereby increases the thickness of the shell. The lack of sufficient taper, additionally, can lead to the formation of off-corner cracks. On the other hand, an excessive taper can cause difficulty in the withdrawal of the strand which promotes mould wear, and in extreme cases, causes the billet to jam in the mould. The quantification of the strand-mould gap is thus a primary step toward defining mould taper. The gap, however is a complex function of several variables such as casting speed, steel grade, etc., which renders it extremely difficult to characterize. Mould taper has gained increasing attention from researchers in the field of billet casting since Dippenaar et al 1 6 1 earliest work, especially for high speed billet casting.' 2 ' 4 1 The calculation and design of mould taper has been made possible in the Billet Casting Research Group in U B C through the development of mathematical models. ' 1 9 ' 2 0 1 Spray cooling is used to extract up to 60 percent of the heat given up on solidification of a continuous cast product at a conventional casting speed and is a key region for the formation of many internal defects.'71 As a result of increasing casting speed, the productivity of the continuous caster and the quality of the products cast depend largely upon the settings chosen for the secondary cooling.' 8 1 Thus, spray cooling needs to be carefully designed and controlled to achieve this effect. Strong spray cooling has been used in high speed continuous casting of billets. The added heat extraction can induce large temperature gradients at the surface of the billets which could adversely affect the internal quality of the products. Additionally, a significant reheating of the billet surface puts the solidification front under tension and creates a risk of midway cracks. Therefore, 2 CHAPTER 1 - INTRODUCTION spray systems must be carefully rearranged and evaluated in order to achieve higher casting speeds. In this study, a plant trial was conducted at Lasco Steel for continuous casting of six carbon grade steel of billets. The axial heat flux profiles of the mould have been calculated using a mathematical model developed at UBC based on temperature data taken from an instrumented mould. The calculated heat flux profiles vary with steel grade, casting speed and other operating parameters. Mould distortion, billet shrinkage and mould tapers for each steel grade have also been quantified and assessed as they impact internal billet quality. Finally, based on these determinations, optimum mould tapers were recommended to the plant. The secondary goal of this work was to understand spray cooling heat transfer in high speed continuous casting. Spray cooling systems in several plants have been investigated. Along with water flux data and an empirical formula, [ 3 8 ] the heat transfer of spray cooling has been calculated using a mathematical model. Both the temperature distribution of the strand and spray-related defects have been analyzed. New spray cooling system was also designed for high speed casting at Alta Steel. 3 CHAPTER 2 - LITERA TURE REVIEW CHAPTER 2 - LITERATURE REVIEW The continuous casting of steel is primarily a heat-extraction process. The conversion of molten steel into a solid semi-finished shape involves the removal of: (a) superheat from the liquid entering the mould from the tundish; (b) the latent heat of fusion at the solidification front as liquid is transformed into solid steel, and (c) sensible heat from the solid shell. These heats, or enthalpies, are extracted by a combination of heat-transfer mechanisms: (a) convection in the liquid pool due to the input of momentum from the tundish stream as well as buoyancy-driven flows; (b) heat conduction down temperature gradients in the solid shell from the hot solidification front to the colder outside surface of the strand; and (c) external heat transfer by conduction, convection and radiation in the three major heat-extraction zones: mould, sprays (plus support rolls for larger sections) and radiation cooling to the surroundings (Fig. 2.1). Because heat transfer is the major phenomenon occurring in continuous casting and also the limiting factor in the operation of a caster, it has been investigated by many researchers [ 1 ( M 8 ] . The depth of the liquid pool obviously cannot exceed the maximum metallurgical length (defined as the distance from the meniscus to the cut-off stand) without cutting into a liquid core. Thus, the casting speed must be limited to allow sufficient time for the heat of solidification to be extracted from the core. The casting speed is also limited by the shell thickness at the exit of the mould, to prevent breakouts. 4 CHAPTER 2 - L1TERA TURE REVIEW ing Fig. 2.1 Three zones of heat removal in a continuous casting machine m Heat transfer not only limits maximum productivity, it also profoundly influences steel quality, particularly with respect to the formation of surface and internal cracks. In part this is because steel, like other metals, expands and contracts during periods of heating or cooling; thus, sudden changes in the temperature gradient through the solid shell, resulting from abrupt changes in surface heat extraction, causes differential thermal 5 CHAPTER 2 - LITERATURE REVIEW expansion and the generation of tensile strains. Depending on the magnitude of the strain relative to the strain-to-fracture of the steel, cracks may form in the solid shell. Considerable research has been done with respect to the relationship between heat extraction and quality[35"37'45"62'. But it should be clear that control of heat transfer, rather than simple maximization of heat extraction, is a key element of good casting practice for the production of quality steel. 2.1 Heat Transfer in the Mould Heat from the solidifying shell is transferred to the mould cooling water via a series of thermal resistances: the air gap separating the mould and the strand, the mould wall, and the mould/cooling interface. Heat extraction from the surface of the shell is governed by the behavior and properties of the gap that forms as the cooling shell shrinks from the mould. The rate limiting process is heat conduction across the gap; and therefore, the gap width strongly affects the steel-to-mould heat flow. The gap width depends not only on shrinkage of shell but also on the ferrostatic pressure which opposes to it. The interaction between shrinkage and bulging of the shell causes the gap width to vary in both the axial and transverse directions. In addition to this complication, the thermal conductivity of the gap depends on the type of material filling it which in turn is a function of the mould lubricant used. In the case of oil, employed for the casting of smaller section sizes, the gap is filled with the gaseous pyrolysis products of the lubricant. However, if mould powders are used, as in the case of bloom and slab casting, the gap contains condensed material that is solid against the mould but may be liquid in contact with the steel. 6 CHAPTER 2 - LITERATURE REVIEW 2.1.1 Mould with Different Lubricants Oil lubricants lubricate the mould by wetting the mould wall; they then partially break down owing to the high temperature and contribute to the atmosphere in the gap. Mould powders, on the other hand, simply melt and wet the steel; the wetting is controlled by interfacial forces. 1 2 2 1 The difference in behavior between the two types of lubricants gives rise to different patterns of heat extraction. <x> <x> T o 2L c d "S< a> T 3 <x> 1.6 u 1.2 1.0 0.8 0.6 0.4 0.2 0 Casting consumables: Oil — Low-melting casting flux — * — High-melting casting flux 100 200 300 400 500 600 700 Distance from mould top edge in mm L 5 Zones 8 Fig. 2.2 Heat transfer profile over the mould length 1 2 1 1 Fig. 2.2 shows the heat extracted per kg of steel in the mould for a 240 mm wide and 700 mm long test liner as a function of the mould length and at a casting speed of 600 mm/min, for lubricants of rapeseed oil, low-melting casting flux and high-melting casting flux. The maximum heat flux in the region of the meniscus is reduced when employing 7 CHAPTER 2 - L1TERA TURE REVIEW high-melting casting flux. Subsequently, the heat flux decreases towards the mould exit, followed by a slight increase in the final zone. The differences in heat flux resulting from the different lubricant types are largest in the meniscus zone and decrease towards the mould exit. For example, near the meniscus the heat extracted in mould with a low-melting casting flux is around 25% greater than in the case of a high-melting flux. Using rapeseed oil for mould lubrication has been found to produce a further overall increase in heat transfer of 40%. A possible reason for the higher upper-mould heat flux with oil is the presence of a hydrogen-rich atmosphere in the gap, due to pyralysis of the oil. The lubrication mechanism in the mould for high speed casting [ 3 1 ] was assumed to be liquid lubrication because the measured frictional force is in good agreement with the values calculated from a liquid lubrication model. The upper limit of casting speed in the continuous casting process with an oscillating mould has been predicted to be 5-8 m/min on the basis of a comparison of the frictional force and the tensile strength of the solidified shell beneath the meniscus. 2.1.2 Influence of Superheat Generally it has been observed that an increase in superheat, within reasonable limits, has a negligible effect on heat transfer in the mould. [ 2 3 1 However, the higher pouring temperature retards solidification in the early stages, thereby reducing the strength of the billet shell. It is also observed that, as the superheat is increased, the shell thickness at the corners decreases. This suggests that high pouring temperatures will give rise to increased breakouts. 8 CHAPTER 2 - LITERATURE REVIEW 2.1.3 Influence of steel grade The effect of steel composition and particularly carbon content on the overall mould heat transfer has been reported from several sources. [ 2 4 ' 2 5 ] These investigations have revealed that the mould heat flux is a minimum when casting a 0.10% C steel, and varies only slightly above 0.25% C in steel, as shown in Fig. 2.3. The effect of carbon content on heat transfer leads to some quality problems being more acute within the carbon range 0.06 to 0.14% C (the peritectic range). E 1800 < if 1600 1 s UOO 2 0,8 C content in % Fig. 2.3 Affect of carbon content on mould heat flux [24] The breakout shells from several castings further revealed that the internal surface of the skin of 0.10% C steel was rippled. The rippling effect decreased progressively with an increase in carbon content; and above 0.40% C the inner surface was relatively smooth. The effect of carbon on the appearance of the outer surface of the billets was found to be the same. 9 CHAPTER 2 - LVTERA PURE REVIEW Grill and Brimacombe [ 2 5 1 have proposed a mechanism based on the 5-y-phase transformation to explain these observations. They have pointed out that, compared to higher carbon steels, 0.10% C steel undergoes the greatest solid-state transformation. Since the transformation is accompanied by a contraction of 0.38%, 0.1% C steel experiences a greater shrinkage than higher carbon steels. Air gaps then may form intermittently because of this enhanced shrinkage and reduce the heat-extraction rate. 2.1.4 Influence of Casting Speed Casting speed also has a marked effect on the distribution and mean heat flux in the mould. Many investigators l l ' 2 4 ] have observed an increase in heat flux with an increase in casting speed, as shown in Fig. 2.4. 200 £ X 3 < LU X 1001 CASTING SPEED V . 1,3 (m/min) V-1.1 Y . I v.o.a 0 100 200 300 400 500 600 700 DISTANCE DOWN MOULD (mm) Fig. 2.4 Heat flux down the length of the mould for various casting speeds [i] The average overall heat-transfer coefficient has also been noted to increase with casting speed. Despite the increase in heat transfer rate with casting speed, it is important 10 CHAPTER 2 - LITERA TURE REVIEW to note that the specific amount of heat extraction, Jkg"1, decreases, resulting in a net decrease in shell thickness. It has been observed that the magnitude of temperature fluctuations in copper mould plates increases with casting speed. 2.2 Billet Shrinkage and Mould Taper in High Speed Casting 2.2.1 Billet Shrinkage in the Mould As previously mentioned, steel cools and shrinks in the mould and an air gap forms between the mould and shell. Heat transfer from molten steel to the walls of the continuous casting mould is largely controlled by the conduction across the air gap. The air gap is a complex function of several variables and its width changes in both the longitudinal and transverse direction which renders it extremely difficult to characterize. Its magnitude is a primary step towards defining mould taper. Conversely, the mould taper also influences the heat transfer and billet shrinkage in the mould. Furthermore, the shrinkage of the billet is affected significantly by the grade of steel being cast, particularly the low carbon grades where the contraction accompanying the solid-state transformation from delta-ferrite to austenite phase must be taken into consideration. 2.2.2 Mould Taper To minimize the air gap and improve heat transfer, moulds are tapered. The inward mould taper which compensates for the shrinkage of the solidifying shell varies from no taper to single taper, double taper and parabolic taper. Excessive taper can cause the billet to bind in the mould; while moderate taper can result in an increase in heat transfer between mould and billet. Although mould taper undoubtedly improves heat transfer and reduces the surface temperature of the billet at the mould exit, excessive taper increases 11 CHAPTER 2 - LITERA TURE REVIEW the resistance to withdrawal and exacerbates mould wear. In the limiting case, if the taper is made too large, the billet can jam in the mould. The conventional mould exhibits a maximum outward bulge near the meniscus. The maximum bulge is located below the meniscus, so that the mould acquires a negative taper above it and a positive taper immediately below. The negative taper at the meniscus can cause defects, such as deep oscillation marks, non-uniform lubrication, and even the possibility of breakout [ 2 6 " 2 9 1 . Several researches [ 2 6 " 2 9 ] have revealed the strong influence of mould taper on the depth and uniformity of oscillation marks on off-corner squareness and off-corner internal cracks. More recent work , 5 9 ' also has shown that mould taper at the meniscus has a profound effect on the local and overall heat extraction from steel, with consequences for mould distortion and billet surface quality. In the first published study of billet mould taper, Dippenaar et al l < 5 ] evaluated mould tapers by estimating billet shrinkage. The two-dimensional transverse heat transfer model originally presented by Brimacombe l 3 0 ] was used to calculate the temperature field in the solidifying billet. The model assumed a constant thermal expansion coefficient. The research concluded that larger gaps formed in the lower region of single-tapered moulds. Calculations, based on axial profiles of measured mould heat extraction, shrinkage of the cooling solid shell and mould distortion, have shown that double taper is desirable. Chandra et al [ 5 9 ] modified the model to include a thermal-expansion coefficient which was a function of temperature and carbon content. His work was particularly important for calculating the shrinkage of peritectic steels. 1 2 CHAPTER 2 - LITERATURE REVIEW 2.2.3 Mould Taper for High Speed Continuous Casting In high speed casting, Danieli Co. 1 2 1 has developed a new technique. The key points of the new technology are a new concept for cooling the mould and a new mould itself. This new mould is called adaptable, because it is able to 'adapt' its original taper to the billet shrinkage. This effect is controlled by the pressure of the mould cooling water. High pressure, larger pressure drop and higher water velocity in the cooling channel are used to improve the cooling conditions. This innovative design allows a substantial increase in the casting speed. Moreover, the mould is able to cast a wide range of steel grades without altering other process parameters. The first design of the adaptable mould ( D A N A M phase 1, with a length of 780 mm and normal thickness of copper tube) permitted the casting of 130x130 mm billets at 4.3 m/min, using open stream pouring. The billet quality results were equivalent to or better than that for conventional continuous casting. The measured thickness of the solid shell along the casting direction was similar to conventional casting at a lower casting speed. It was reported that a thinner adaptable mould ( D A M A M phase2) of 1000 mm length, for speed of 6 m/min, would commence operation in the second half of 1995. But published details of the trial are sparse. For the purpose of improving productivity and billet quality, joint research was conducted on high speed casting with a convex mould. [ 4 1 Beginning in September 1992, a maximum casting speed of 3.5 m/min was achieved. The billet quality was found to be superior which led to the claim that the convex mould was a proven technology. 13 CHAPTER 2 - LITERA TURE REVIEW 2.3 Heat Transfer in Secondary Cooling Zone In the secondary spray zone below the mould, heat transfer by impinging water droplets projected from a nozzle onto the strand involves boiling and the formation of a steam layer against the steel. The water flux depends on the position within the spray pattern, nozzle type, water pressure and the standoff distance. Interestingly, the rate of solidification in the spray zone depends less on the external heat extraction than it does in the mould because the solid shell is considerably thicker. As a result, conduction through the shell becomes the rate limiting process and the primary effect of spray cooling is to alter the temperature distribution through the shell. 2.3.1 Water Spray and Air-mist Spray Beneath the mould of a continuous-casting machine, the moving steel strand is cooled by banks of water sprays. The purpose of the spray cooling is to continue the heat extraction and solidification initiated in the mould without generating tensile stresses of sufficient magnitude to cause shape defects, surface cracks or internal cracks. The water spray pattern impinging on the strand surface should cover as wide an area as possible in order to prevent excessive local cooling in the strand shell. This purpose is served by either employing hollow/full cone nozzles, either of which will provide a square or round spray pattern, or flat spray nozzles | 3 2 ' 3 4 1 . The length of the spray chamber may vary from as little as 0.5 m up to 6 m in the case of billet casters,' while extending to 10 m typically in slab casters, up to 17 m in high-speed slab casters. The sections are usually divided into several smaller zones which can then be individually regulated. Besides typical water spray cooling, an air-water spray has been applied recently. In this system, cooling water is mixed with compressed air in a mixing chamber ahead of 14 CHAPTER 2 - LITERA TURE REVIEW the nozzle, and the mixture emerges from the nozzle as a finely atomized, high-impulse, wide-angled spray. This type of spray cooling is particularly suitable for high-grade steels which are susceptible to cracking. Its more important advantages include a particularly uniform cooling rate, a very wide volume flow control range (1:12), little danger of nozzle clogging and fine water droplets for optimum cooling. l 3 3 ' 3 4 ] 2.3.2 Factors to Influence the Heat Transfer of Spray Cooling Zone Heat extracted from the strand surface by water sprays has been measured by researchers [ 3 4 ' 3 8 ] . Experiments have been conducted employing resistance-heated plates which are spray cooled at steady state. Many spray variables influence the heat extraction rate, q, usually expressed in terms of a heat transfer coefficient, h: cj = hA(Ts-TJ (2-1) where Ts is the surface temperature of the strand and Tw is the temperature of the spray water. The heat transfer coefficient is obtained from experiments or plant measurements. It mainly depends on the water flux. The strand surface temperature level decides the heat transfer mechanism of transition boiling. The effect of variables such as nozzle type, water pressure and nozzle-to-nozzle distance on spray heat transfer can be seen primarily in terms of their effect on the spray water flux; whereas variables like water temperature and steel surface temperature directly affect heat transfer. Under normal continuous casting conditions, in which surface temperatures range between 1300 °F (700 °C) and 2200 °F (1200 °C), surface temperature has only a small effect on the heat transfer coefficient. As shown in Fig. 2.5, taken from the study of Mizikar | 7 ] , increasing surface temperature causes the heat transfer coefficient to decrease slightly. The relative lack of dependence of h on surface temperature, Ts, is characteristic 15 CHAPTER 2 - LITERA TURE REVIEW of the film boiling region of the classical boiling curve. At lower surface temperatures, below 1020 °F (550 °C), the heat transfer coefficient increases sharply as nucleate boiling begins to take effect. SURFACE TEMPERATURE. *F Fig. 2.5 Influence of surface temperature on spray heat-transfer coefficient, h The temperature of the spray water, Tw, does not have a large influence on the heat transfer coefficient The measurements of Sasaki et al [ 3 8 ] have shown that an increase in water temperature from 20 to 60 °C causes h to decrease by only about 14%. A l l studies show that, in the temperature range of interest, the spray-water flux, W, has the largest effect on the heat transfer coefficient. The experimental findings of the different studies are shown in Fig. 2.6. Where it is seen that, with the exception of Mizikar ' s results, the relationship is reported to be nonlinear. 16 CHAPTER 2 - LITERA TURE REVIEW h=W" Most researchers agree that the value of n lies between 0.5 and 1.0. [ 4 0 J (2-2) 10 20 Water flux (gal/ft min) 30 40 50 60 8 V c o o w a. NcioM tt ol o—© Sotoki •> ol O - O Albtrny**"** * - * B o l l « ond Mourtou (lop tproy) 6 - £ 6 o l l t ond Mourtou (bottom sproy) x o-o-o Minlio» ,'(ot 276 liPo) Miiikor^Jot 620 kPo) -Mulltr ond Jtsehor 1 1 • —• Ishiguro tt o l " , _- __, „ H 5 0 0 e o —WOO 0 300 200 g — 100 a. tn Woter (lux. I/mi Fig. 2.6 Influence of water flux on spray heat transfer coefficient[35] 2.3.3 Billet Surface Reheating When the strand passes from a cooling zone with a high heat transfer rate to one with a lower heat transfer rate, the surface temperature of the strand may increase. This is caused by a relaxation of the large thermal gradients created during the high heat transfer period and the subsequent accumulation of enthalpy at the surface of the casting. It is generally agreed in practice that the change in surface temperature of the strand below the mould should be as low as possible in order to avoid internal and surface defects of the billet. Thereby both spray water flux and spray cooling length should be 17 CHAPTER 2 - LITERA TURE REVIEW increased for high speed casting. The reheating effect must be limited, as it causes thermally induced stresses that can result in cracking. Van Drunen and Brimacombe 1 3 7 1 have shown that such a reheating before complete solidification leads to the development of tensile stresses at the solidification front; the strain reached is a function of the amount of reheating and can exceed the fracture strength of the steel at the solidification front. Although the actual reheat amount permitted seems to vary greatly with steel grade and casting practice, some researchers | 8 ' 1 3 ' have suggested reheating limits. Several others [38,39] n a v e c a ] c u i a t e a " t h e thermal stresses of the strand by different stress models. Similarly, cooling rate should be limited. It is proposed ' 3 ? 1 that because the ductility-to-fracture of steel at 1350 °C has been reported to be 0.2-0.3 pet., and because the thermal expansion coefficient of common steel grades is about 0.2 pet. per 100 °C, a surface reheat of 100 °C was likely to produce cracks. Thus Brimacombe 1 3 7 1 recommended a maximum of 50 °C as a design criterion for billet surface reheat. Furthermore for purposes of spray design, he proposed that the surface temperature of the strand in the sprays should be maintained between 1000 °C and 1100 °C to sustain a reasonable solidification rate. 2.4 Defects in Continuous Casting The quality of the continuously cast billet is of great importance for further processing of steel. The formation of cracks which may appear almost anywhere at the surface or in the interior of the billets is one of the most serious quality problems. Surface cracks are oxidized by air and, therefore, cannot reweld during rolling; i f they are not removed by scarfing or grinding, surface cracks will lead to slivers or other types of defects in the rolled product. Internal cracks are less susceptible to oxidation by contact 18 CHAPTER 2 - LITERA TURE REVIEW with air but also can give rise to serious quality problems. Cracks that form just beneath the surface may penetrate to the scale layer depending on the time the steel is at elevated temperature or they may break through to the surface during rolling. Owing to their deleterious influence on quality, cracks in continuously cast steel have been investigated by many researchers I 3 6 , 4 4" 5 2 ' Many of the crack types in continuous casting have been shown in schematic diagrams (Fig. 2.7). The reason for the profusion of crack types lies in the nature of the continuous casting process itself. Continuous casting has achieved widespread popularity because it is capable of extracting heat at a remarkable rate with a combination of mould, sprays and radiant cooling. The rapid cooling, however, results in steep temperature gradients in the solid shell that can change rapidly and generate thermal strains as the shell expands or contracts. In addition, because a semi-solid section is required to move through the machine, it is subjected to a variety of mechanically induced stresses caused by friction in the mould, roll pressure, ferrostatic pressure, machine misalignment, bending and straightening operations. Depending on their magnitude, any of these stresses and strains may result in crack formation. Crack formation depends not only on the operating stresses and strains but also on the mechanical properties of steel at continuous casting temperatures. Combining knowledge of the strain induced in continuous casting and the ductility of steel at casting temperature, we can get a clearer understanding of crack mechanisms which will help to explain the known operating causes of, and solutions to, crack formation. 19 CHAPTER 2 - LITERATURE REVIEW Crocks in continuously cost steel Internal crocks 1 Midway 2 Triple-point 3 Centreline 4 Diagonal 5 Straightening / bending 6 Pinch roll Surface cracks 7 Longitudinal , mid -face 8 Longitudinal , corner 9 Transverse , mid-face 10 Transverse , corner I I S t a r Fig. 2.7 Schematic drawing of strand cast section showing different types of cracks | 3 6 ] 2.4.1 Mechanical Properties of Steel at High Temperature The mechanical properties of steel at elevated temperatures are affected by several variables: temperature, steel chemistry, structure, strain rate and thermal history. Although many efforts 1 4 4 - 4 9 1 have been made, the strength and ductility of steel are 20 CHAPTER 2 - LITERA TURE REVIEW imperfectly known under continuous casting conditions. There are three distinct temperature ranges in which steel has low strength and/or ductility and is, therefore, susceptible to cracking. There is ample evidence to indicate that at temperatures above about 1340 °C both the strength and ductility of steel decline markedly. The low strength and ductility seem to be due to the presence of liquid films in the interdendritic regions which do not freeze until temperatures well below the solidus are reached [ 4 4 ' 5 0 ] . The liquid films apparently contain high levels of sulfur, phosphorus and other elements which have a segregation coefficient less than unity and which concentrate between the growing dendrites. Evidence of liquid films can be seen in Fig. 2.8 which is a scanning electron micrograph of the interior surface of a crack that formed near the solidus temperature. The smoothness of the surface, with no signs of fracture, is indicative of the presence of a liquid film at the time of crack formation. The carbon content also affects the mechanical properties of steel just below the solidus temperature. Morozenskii et al [ 4 7 1 have found that the strain-to-fracture and its plastic component are a minimum for steel containing 0.17 to 0.20 pet carbon. The second zone of low ductility in steel appears in the temperature range 800 to 1200 °C [ 9 ' 4 1 1 . The loss of ductility during cooling below 1200 °C strongly depends on the Mn/S ratio and the thermal history of the steel. Increasing Mn/S increases the ductility of the steel. The influence of thermal history is considerably more complicated. Lankford | 4 6 ] has proposed that low ductility results from the precipitation of liquid droplets of FeS in planar arrays at austenite grain boundaries, which are then paths of easy crack propagation. Steels with Mn/S ratios above 60 are not embrittled because sulfur is tied up in the stable phase, MnS , which precipitates in the matrix, and not 21 CHAPTER 2 - LITERA TURE REVIEW predominantly at grain boundaries. The slower cooling rates improve the ductility even with low Mn/S because manganese then has time to diffuse to the grain boundaries. The third zone of low ductility is in the range of 700 to 900 °C. Generally, low ductility is associated with soluble aluminum in the steel and the precipitation of A1N at grain boundaries | 4 4 - 5 3 1 . l ida et al [ 5 1 ] have reported that A1N precipitation does not proceed appreciably during cooling to as low as 800 °C, but can take place rapidly during heating over the temperature range 700 to 1000 °C. On the basis of these findings, they suggest that repeated cooling and reheating cycles which may occur in the spray chamber could enhance A1N precipitation and lead to low ductility. Owing to the low temperatures involved, the third zone of low ductility is seen to be a factor only in the formation of surface or subsurface cracks. 2 2 CHAPTER 2 - LITERATURE REVIEW 2.4.2 Defect Related to the Mould The mould-related defects such as longitudinal depressions and cracks, transverse cracks and depressions, laps, bleeds, rhomboidity, off-corner internal cracks are the major quality problems encountered in continuously cast steel billets [ 3 6 > 4 4 ' 5 1 J The adverse mould/strand interaction at the meniscus significantly influences these defects. Several studies have focused on the meniscus region to understand the genesis of defects in continuous casting. Surface cracks both transverse and longitudinal, observed at the corner and midface locations in billets originate close to the meniscus [ 3 6 ' 5 1 ] . Defects such as off-corner internal cracks and rhomboidity have been linked to the depth and uniformity of oscillation marks | 2 7 ' 5 6 ' which form at the meniscus. Laps and bleeds in high carbon steels are also thought to be caused by adverse mould/strand interaction at [53,561 meniscus1 . 2.4.2.1 Oscillation Marks Oscillation marks form at the meniscus (the site of initial solidification), because of the mechanical interaction between the mould and strand during the negative strip period. Although oscillation marks have not been considered a quality problem in the past, it is clear | 2 7 - 5 6 ] that they are implicated strongly in the formation of off-squareness, off-corner internal cracks, transverse cracks, bleeds in the mould and breakouts below the mould. In an off-square billet, the oscillation marks are nearly always deeper at one or both of the obtuse-angle corners as compared to the acute-angle corners. The effect of a deep oscillation mark is to locally widen the shell/mould gap, retard heat extraction by the mould and reduce the solidification rate. Thus the shell at obtuse angle corners is thinner and emerges from the mould hotter than the acute-angle corners which, in practice, may 23 CHAPTER 2 - LITERATURE REVIEW be so cold as to appear black. Sprays impinging on the billet below the mold then may pull the billet off-square owing to differential cooling of the hot and cold regions. The influence of non-uniform oscillation marks applies equally to the formation of off-corner internal cracks which generally are located 4-6 mm from the surface and about 15 mm from the corner. There are several factors that influence the depth and uniformity of oscillation marks'271. Reducing negative-strip time down to 0.12 s decreases the depth of oscillation marks and improves the uniformity of the marks between mid-face and off-corner locations. Increasing superheat causes the oscillation marks at the mid-face to become shallow while those at the off-corner remain essentially unaffected. The increasing difference in oscillation mark depth between the two locations helps to explain the dissimilarity in shell growth in the transverse plane and the formation of re-entrant corners. Shallow oscillation marks at the mid-face also are promoted by a large positive mould taper near the meniscus and by a thicker mould wall. More uniform oscillation marks are observed with four-side, compared to two-side constraint of the mould tube and by a large positive mould taper at the meniscus. All these observations can be explained by a mechanism for oscillation mark formation based on the jamming of the mould on the solidifying shell at the meniscus during negative strip and the resultant buckling of the shell. 2.4.2.2 Rhomboidity Rhomboidity could be a result of dimensional instability of the mould tube due to boiling in the cooling channel caused by low cooling water velocity or due to deep and non-uniform oscillation marks. The latter is caused by adverse mechanical interaction of 24 CHAPTER 2 - LITERA TURE REVIEW the mould tube and the billet at meniscus. Factors that influence this are non-uniform constraint of the mould tube at the top, long negative strip-times, inadequate upper taper. Excessive distortion due to poor water quality is also a factor. In an earlier study [ 5 7 1 on rhomboidity in billets, asynchronous boiling in the cooling water channel was identified as an important contributor. A mechanism based on non-uniform cooling around the billet periphery was proposed to explain the formation of these defects in moulds. Another mechanism was proposed to explain the generation of rhomboidity based on oscillation mark formation and non-uniform heat transfer in the mould and the sprays [ 2 9 ] . The problem begins with the formation of deep and non-uniform oscillation marks around the billet periphery. In the vicinity of a deep oscillation mark, the rate of heat removal is low due to a wide mould/strand gap. On the other hand, regions of the billet having shallow oscillation marks experience higher rates of heat extraction. Thus, the presence of non-uniform oscillation marks on the billet surface gives rise to markedly different heat extraction rates around the billet periphery which ultimately leads to a non-uniform solid shell. Thus, the billet exiting the mould, although reasonably square, has a non-uniform solid shell. In the sprays, the colder portions of the strand, having a thicker solid shell, tend to cool faster than the hotter regions because of the effects of unstable boiling; the result is non-uniform shrinkage of the billet and rhomboidity. Also, the metal level fluctuation affects oscillation mark formation, an event closely linked to rhomboidity [ 2 9 - 5 2 1 . Furthermore, since the severity and orientation of rhomboidity change randomly with time, it is possible that the problem has its roots in metal level fluctuations. 25 CHAPTER 2 - LVTERA TURE REVIEW Another interesting aspect of rhomboidity is the effect of steel grade on its severity [ 5 2 ] . The medium carbon grades experience more problems as compared to other grades due to their high mould heat transfer and short freezing range. 2.4.2.3 Off-corner Internal Cracks Internal cracks are sometimes observed near the corners of continuously cast billets. In transverse sections the cracks are found normal to a given face and within 10-20 mm of the corner | 6 1 ' 6 2 ' . The cracks are often associated with a longitudinal surface or off-corner depression on the surface normal to an off-corner crack. This defect may result from bulging of a given face and hinging of the shell about a thin, weak off-corner. As bulging occurs, a hinging action develops near the colder and stronger corners causing off-corner tensile stresses near the solidification front and cracking. Because bulging is affected by mould taper, mould alignment, mould wear and foot-roller guidance of the strand, these variables should affect the formation of off-corner cracks. Since off-corner cracks extend into the upper-spray zones, the extent of cracking may be reduced by increasing the water flux slightly, which reduces bulging and imposes a compressive strain at the solidification front. Off-corner cracks are also associated with rhomboidity and superheat. It has been indicated that internal cracks are more severe with increasing superheat and increasing rhomboidity. Besides, deep oscillation marks at off-corner regions cause local thinning of the shell and contributes to formation of the cracks. In general, the following observations in studies are important in understanding the off-corner internal cracks: 1) A significantly thinner shell at the off-corner position than at the mid-face. 26 CHAPTER 2 - LITERA TURE REVIEW 2) Cracks start to form at about 5.5-12 s from the meniscus level which, in most cases, is equivalent to a distance of from 250-500 mm down the mould; this places the crack initiation event in the lower part of the mould or just below the mould in the upper sprays. 3) An off-corner crack zone sufficiently far from the corner that it has the same temperature as the midface region. 2.4.3 Defects Related to the Secondary Cooling Zone Quality problems can also arise in the spray cooling zone in the continuous casting of steel. Sprays can influence defect formation by generating tensile stresses, by altering the strand temperature and hence, mechanical properties, by influencing the precipitation of second phases which lower steel ductility, and by slightly influencing the solidification rate [ 3 5 1 . Midway cracks and rhomboidity or diagonal cracks may take place in billet casting by unsuitable spray cooling. 2.4.3.1 Midway Cracks It is generally agreed that the excessive secondary cooling and a high casting temperature are operating factors responsible for midway cracks [ 3 5 ' 3 6 ] . In addition, the chemistry of the steel exerts an important influence; sulfur and phosphorus in particular affect the crack susceptibility of steel. Excessive spray cooling is a major factor in the formation of midway cracks because it leads to subsequent reheating of the strand surface which provides the driving force for cracking.'3 7 1 Reheating causes the surface to expand and this imposes a tensile strain in the interior region of the solid shell which is weak and nonductile above about 1340 °C. Viewed in a transverse plane, the tensile strain and stress run parallel to the surface, and 27 CHAPTER 2 - LITERA TURE REVIEW thus, cracks will form perpendicular to the surface, depending on the magnitude of the strain. Strong reheating of the surface occurs whenever the rate of cooling decreases abruptly such as below the spray chamber, between successive spray nozzles or below the bottom of the mould. Obviously, the magnitude of reheating is very dependent on spray cooling conditions. High casting temperature has a large effect on the tendency to form midway cracks because it influences the cast structure. A high casting temperature favors an enlarged columnar zone extending inward from the surface at the expense of the central equiaxed zone. Midway cracks are able to form much more easily between the dendrites in the columnar zone which run perpendicular to the tensile stress, as compared to the equiaxed zone. 2.4.3.2 Diagonal Cracks Diagonal cracks are associated with rhomboidity. This type of crack usually runs between obtuse corners of the rhomboid section 1 3 6 , 5 3 1 . Clearly, diagonal cracks result from distortion of the billet which can arise if two adjacent faces are cooled more rapidly than the other faces in the mould or secondary cooling zone. The contraction of the steel in the vicinity of the colder faces generates a tensile strain, oriented diagonally between these faces. If sufficiently large, the strain causes distortion and cracks to form at right angles to the strain axis that is between the obtuse corners. The cracks form initially in the high temperature zone of low ductility but may grow outward toward the corners depending on the magnitude of the strain. 28 CHAPTER 2 - LITERA TURE REVIEW To eliminate this crack formation, equal cooling on each of the four faces of the billet must be achieved by good alignment between the mould and roller cages and suitable arrangement and maintenance of spray nozzles. 2 9 CHAPTER 3 - SCOPE AND OBJECTIVES CHAPTER 3 - SCOPE AND OBJECTIVES The literature on billet casting indicates that mould and spray cooling are the key to increase casting speed for billets. In high speed casting of billets, more heat must be extracted from the mould and spray cooling zones. Thus, light must be shed on heat transfer in the mould and spray cooling. Furthermore, billet quality is strongly associated with mould behavior, thereforethe interaction between mould and strand under different operating parameters and its effect on the billet quality should be investigated carefully. Since spray cooling also impacts on the generation of billet cracks, suitable secondary cooling should be employed. It is clear that there cannot be a universal mould taper through which all grades of steel can be successfully cast at different casting speed, due to the different shrinkage of the billet in the mould. To examine the variables influencing on mould heat transfer and taper design in high speed casting, plant trials were carried out in which an operating long mould was instrumented with thermocouples and LVDT's . Spray cooling data have been collected in three companies to understand the factors which influence the consequences of spray cooling in high speed casting. With the analysis of the collected plant trial data, the following objectives were formulated: 30 CHAPTER 3 - SCOPE AND OBJECTIVES 1. To study the heat transfer in a parabolic tapered mould for several carbon steel grades and at various operating parameters during high speed casting of large section steel billets. 2. To evaluate the interaction between the parabolic tapered mould and strand during high speed casting of steel 3. To evaluate billet quality in relation to the operating parameters of the caster. 4. To study heat transfer in spray cooling zone and correlate internal billet quality with spray cooling conditions. 5. To design a suitable spray cooling system appropriate for high speed continuous casting of steel billets. 31 CHAPTER 4 - INDUSTRIAL PLANT TRIAL CHAPTER 4 - INDUSTRIAL PLANT TRIAL 4.1 Caster Details and Nominal Operating Practice In December of 1998, an instrumented mould trial was conducted at Co - Steel Lasco at Whitby, Ontario. A retrofitted mould was instrumented with 52 thermocouples, and mould temperature data was recorded for a total of nine heats. The operating parameters of the continuous casting machine at Lasco are presented in Table 4.1 and details of the mould system in Tables 4.2 and 4.3. The section size investigated was 178x127 mm, cast at speeds in the range of 2.2-2.9 m/min, as shown in Table 4.1. The mould was Cu-Cr-Zr with a wall thickness of 13 mm and a length of 1016 mm (see Table 4.2). The tube had a parabolic taper, which varied from 3.42 %/m at 127 mm from the top of the mould to 0.56 %/m at the bottom of the tube, as presented in Table 4.3. Four of the heats monitored were low carbon (0.155-0.159 pet. Carbon) and one each of the following five carbon grades, 0.169, 0.183, 0.204, 0.412 and 0.767 percent. The compositions of the steels cast during the trial are presented in Table 4.4. A total of 16 billet samples were gathered in order to assess the billet quality. Three L V D T ' s were also installed on the mould housing with the objective of determining mould oscillation and negative strip time. The oscillator stroke was typically 9.525 mm while the frequency was set at 2.2 H z for a casting speed of 2.03 m/min, as presented in Table 4.1. The negative strip time is 32 CHAPTER 4 - INDUSTRIAL PLANT TRIAL 0.148 s. Mould water velocity was calculated to be 14 m/s. All heats were cast using Quacast oil. 4.2 Temperature Measurements The instrumentation of billet moulds with thermocouples has been described previously by Brimacombe et al. [ 1 4 ' 2 6 ] A copper mould tube was fitted with 52 single element constantan thermocouples; 13 on each of the four faces at locations shown in Fig. 4.1. The thermocouples were prepared by forming a bead on the end of constantan wires with a TIG welder. Beads were filed flat to 0.3-0.4 mm thick. Heat shrink tubing was applied to the wire to provide electrical isolation away from the bead end. Threaded holes were prepared through the water baffle and halfway into the copper mould wall. The bead side of the wires was inserted into the mould wall to a depth of approximately 6 mm from the hot face and was held in place by threaded copper plugs. Silicone sealant was employed to prevent water leaks through the baffle. The constantan wires were then attached to insulated copper wires in the cooling water plenum. Groups of wires were bunched together and passed through a water-tight seal at the bottom of the mould housing. A copper wire attached to the mould supplied the return current for the single element thermocouples. In addition, four commercially available extrinsic copper-constantan thermocouples were installed - one on the centerline of each of the four faces at the exit of the water channels - in order to measure the outlet temperature of the cooling water. Two similar thermocouples were also placed in the mould housing to measure the bulk inlet and outlet water temperatures. 33 CHAPTER 4 - INDUSTRIAL PLANT TRIAL The thermocouple signals were recorded with the Labtech Notebook (Version 7. 1. 1), a commercially available software package. The data acquisition system consisted of an IBM-PC equipped with a Metrabyte multiplexer DAS-8 board and an EXP-16 expansion interface. This system had a resolution of 0.012 mV or approximately 0.5°C. Thermocouple signals were sampled at a rate of 10Hz for each of the nine heats. 4.3 Metal Level and Casting Speed The metal level and casting speed signals were obtained from the plant as a 4-20mA signal and converted into a 0-20mV output. The signals were recorded at 10 Hz in conjunction with the thermocouple signals. 4.4 Mechanical Signals Three linear variable differential transformers (LVDT's) were placed on the top of the mould housing. In this way, the mould vertical displacement, front-to-back and side-to-side motions were measured. A second data acquisition system was used to acquire these signals at a sampling rate of 100 Hz for 30s intervals during each heat. 4.5 Billet Samples Sixteen billet samples were taken during the tests at steady operating conditions, before or after a change in practice was made (see Table 4.5). Billet samples were transferred to our UBC Lab for evaluation. All samples were shot blasted to remove any oxide layer present prior to examination. The evaluation included a usual inspection for surface defects such as bleeds and laps, oscillation mark appearance, billet surface roughness, longitudinal and transverse depressions, inclusions and pinholes. The oscillation mark depth for all billet samples were measured using the U B C profilometer. The measurements were made at three locations on each of the 34 CHAPTER 4 - INDUSTRIAL PLANT TRIAL straight sides of the billet samples - one along the centerline and two along the off-corner positions, approximately 14 mm from each corner. The dimensional quality of the billet samples was assessed in terms of rhomboidity or off-squareness. Finally the billet samples were sectioned transversely. These sections were ground and etched in order to reveal the presence of any internal defects such as midway and off-corner cracking. The presence of internal cracks and rhomboidity are of particular interest to this study (see below Chapter 7). 35 CHAPTER 4 - INDUSTRIAL PLANT TRIAL Table 4.1 Casting machine specifications Machine Type Curved Mould Mould Length 1016 mm Billet Size 177.8 x 127 mm2 Casting Speed 2.03 m/min Lubricant Quacast Oil Reoxidation Protection No Nominal Meniscus Level 110 mm Oscillation Type Sinusoidal Stroke Length 9.525 mm Oscillation Frequency 2.2 Hz Negative Strip Time 0.148 s Mould Lead 3.21 mm Table 4.2 Mould details Material Cu-Cr-Zr Thickness 13 mm Corner Radius 1/8 in Construction tube Taper parabolic Mould Length 1016 mm Baffle Gap 3/16 in Constraint 4 sided 36 CHAPTER 4 - INDUSTRIAL PLANT TRIAL Table 4.3 Parabolic mould taper Distance F rom Top Distance From Top Taper (in) (mm) (%/m) 0 0 0.00 1 25.4 0.00 2 50.8 0.00 3 76.2 0.00 4 101.6 0.00 5 127 3.42 6 152.4 2.73 7 177.8 2.38 8 203.2 2.10 9 228.6 1.91 10 254 1.78 11 279.4 1.63 12 304.8 1.52 13 330.2 1.43 14 355.6 1.35 15 381 1.27 16 406.4 1.21 17 431.8 1.15 18 457.2 1.10 19 482.6 1.05 20 508 1.01 21 533.4 0.97 22 558.8 0.94 23 584.2 0.90 24 609.6 0.87 25 635 0.84 26 660.4 0.81 27 685.8 0.79 28 711.2 0.76 29 736.6 0.74 30 762 0.72 31 787.4 0.70 32 812.8 0.68 33 838.2 0.66 34 863.6 0.65 35 889 0.63 36 914.4 0.61 37 939.8 0.60 38 965.2 0.59 39 990.6 0.57 40 1016 0.56 37 CHAPTER 4 - INDUSTRIAL PLANT TRIAL Table 4.4 Summary of heat chemistry Element Wt (%) Heat Number 642 643 644 645 664 665 668 685 686 C 0.155 0.183 0.157 0.169 0.412 0.767 0.204 0.159 0.154 M n 0.806 0.756 0.865 0.825 0.756 0.776 0.874 0.672 0.746 P 0.005 0.005 0.007 0.004 0.004 0.007 0.005 0.003 0.003 S 0.023 0.042 0.026 0.028 0.024 0.026 0.032 0.014 0.014 Si 0.214 0.203 0.245 0.220 0.166 0.167 0.221 0.182 0.196 Cu 0.391 0.443 0.354 0.431 0.434 0.486 0.324 0.205 0.315 N i 0.138 0.103 0.124 0.126 0.115 0.143 0.104 0.139 0.114 Cr 0.102 0.085 0.125 0.092 0.088 0.130 0.082 0.330 0.334 M o 0.036 0.028 0.031 0.029 0.029 0.038 0.027 0.031 0.034 Sn 0.013 0.012 0.012 0.013 0.013 0.019 0.011 0.011 0.012 V 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.023 0.028 Zn 0.013 0.014 0.009 0.008 0.010 0.010 0.014 0.010 0.012 A l 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.259 Mn/S 35.0 18.3 33.3 29.5 31.5 29.8 27.3 48.0 53.3 Mn/S i 3.8 3.7 3.5 3.75 4.6 4.6 3.95 3.7 3.8 Cu/Ni 0.36 0.23 0.35 0.29 0.26 0.29 0.32 0.68 0.36 38 CHAPTER 4 - INDUSTRIAL PLANT TRIAL TOP 105 mm 135 mm 150 mm 165 mm 180 mm 195 mm 300 mm 450 mm 600 mm 735 mm 850 mm 900 mm 950 mm Fig. 4.1 Thermocouple layout (all faces). 3 9 CHAPTER 4 - INDUSTRIAL PLANT TRIAL Table 4.5 Casting parameter changes and billet sampling Heat No. C % S/E Practice T i me of Samples 642 0.155 9:00/9:53 C S p : 2.0 A 2.4 m/min (9:28). DT:10 .6 A 11 .7 °C. W . V . : 13.6 m/s 9:25 9:52:22 643 0.183 9:54/10:47 C S p : 2.44 A 2.34 A 2.59 m/min (10:00, 10:11). D T : 11.7 A12.8 °C , W . V . : 13.6 m/s. O i l F low: 60 A 90 ml /min (10:20) 10:14:22 10:42:35 644 0.157 10:48/11:42 CSp: 2.56 A 2.16 A 2.26 m/min (10:49, 11:00). D T : 12.2 A 11.1 A 10.6 °C , W . V . : 13.4 m/s. O i l F low: 90 A 60 ml /min (11:15) 11:05:41 11:34:25 645 0.169 11:43/12:37 W . V . : 14 A 12 m/s D T : 10.6 A12.2 °C (12:00) 12:20:10 664 0.412 10:30/11:24 C S p : 1.83A2.24 m/min (10:54). D T : 12.2A13.3 ° C , W . V . : 13.45 m/s. Nozz le Size: 16.5A18.5 mm. 10:51:54 11:16:02 665 0.767 11:25/12:29 CSp: 2.13 A1.75 (12:00) Nozz le Size: 18.5 A16.5 m m D T : 13.3 A12.2 °C , W . V . : 13.56 m/s. Composit ion Change 12:01:58 * 668 0.204 14:20/15:14 CSp: 2.18 A2.41 (14:31), W . V . : 13.56 m/s. * 685 0.159 9:20/10:25 Meta l Level : 121.9 A142.2 m m (10:11) W . V . : 14.25 m/s, D T : 11.1 A10.6 °C. 9:50:20 10:22:37 686 0.154 10:26/11:23 O.F. : 122 cpm (10:26), 156 (10:35) W . V . : 13.78 m/s, D T : 10.6A11.1 °C. 10:45:00 11:06:05 Note: W . V . - water velocity; D T - temperature difference between outlet and inlet water; CSp - casting speed, * - no time record. O.F. - oscillation frequency, S/E - heat start and end times. 40 CHAPTER 5 - RESULTS OF PLANT TRIALS CHAPTER 5 - RESULTS OF PLANT TRIALS 5.1 Billet Quality Evaluation 5.1.1 Surface Quality A summary of the surface evaluation is presented in Table 5.1. All billets, with the exception of the 0.4 pet carbon grade, had bleeds and laps. Oscillation marks were typically non-parallel and non-linear, which result from metal level turbulence at the meniscus. Two main defects evident on all the sections were longitudinal off-corner depressions on the straight narrow faces and concavity, as shown in Fig. 5.1 and 5.2. The concavity had an average width of 106 mm and was typically 0.7-2.5 mm in depth. 5.1.2 Off-corner Internal Cracks Off-corner internal cracks were found on all billets, which can be seen in Fig. 5.3. On the straight narrow faces, the off-corner internal cracks were located approximately 8.0 mm beneath the surface at the site of longitudinal depressions. The cracks were generally located approximately 19.8 mm off the corner. Cracks were predominant on the straight right narrow face, close to the outside curved wall on all the samples. Many of the samples contained cracks at more than one off-corner location. 5.1.3 Off-squareness Results of an off-squareness measurement, as determined by a difference in the length of the diagonals of the billet samples are presented in Fig. 5.4. Two samples with high 41 CHAPTER 5 - RESULTS OF PLANT TRIALS off-squareness were from a heat cast at the high speeds (2.65-2.75 m/min) and contained 0.18 pet carbon. The two other billet samples cast at corresponding speeds did not exhibit high off-squareness, and had carbon contents of 0.154 and 0.159 pet respectively. 5.1.4 Midway Cracks All billet samples contained midway cracks on the two narrow straight faces, as shown in Fig. 5.2 and 5.3. Occasionally cracks were also seen on the broad curved faces. Crack initiation was typically 12-15 mm and 20-26 mm below the surface with a few exceptions. Crack length was typically 10-30 mm. 5.1.5 Oscillation Mark Depth The oscillation mark appearances for different grades are shown in Fig. 5.5. It is evident that the oscillation marks are deep for all grades in the 0.15-0.2 pet carbon range, with oscillation marks at the mid-face typically of the order of 0.4-0.5 mm. The oscillation marks are shallower for the 0.7 pet carbon grade, and are on average approximately 0.16-0.21 mm at the mid-face. For the 0.4 pet carbon grade, the depths of the oscillation marks are a little deeper at 0.19-0.25 mm. In all cases, the oscillation mark depths are not uniform and vary along the billet length, with some marks twice as deep as others. The pitches of the marks are also not uniform, indicating considerable metal level fluctuations. 5.2 Oscillator Performance Fig. 5.6 shows the mould velocity profiles during casting for a frequency of 129 cpm. It is evident that there is some minor distortion of the mould velocity at the maximum and minimum points. However it is not severe and does not alter the negative strip time significantly. 42 CHAPTER 5 - RESULTS OF PLANT TRIALS 5.3 Mould Temperature 5.3.1 Mould Temperature Response On each face, mould wall temperatures were recorded for twelve of the thirteen installed thermocouples. A sample of the data recorded on the inside curved wall and the left straight wall for 0.155 pet carbon grade (heat 642) during a casting increase, which occurred at approximately 1480 s, is shown from Figs. 5.7 to 5.10. It is evident that temperatures recorded 950 mm below the top of the mould are generally higher than at the 900 mm location on inside curved wall. It is shown that the thermocouples located below the meniscus (110 mm from the top of the mould) recorded a measurable increase in average temperature immediately following the casting speed increase. 5.3.2 Time-average Mould Temperature Distribution The average mould temperature and standard deviation were calculated for each position. The results for heat 642, corresponding to 0.155 pet carbon, are shown in Figs. 5.11 through 5.14. It is evident that the maximum mould temperature is recorded about 25 mm below the meniscus at a location 135 mm below the top of the mould. On the inside and outside curved walls, a minimum temperature was recorded at a location betweenl35 and 200 mm below the top of the mould, whereas for the straight walls the temperature decreased uniformly with no obvious minimum. It was also evident that on the inside and outside curved walls the temperature recorded by the thermocouple located 950 mm below the top of the mould was always higher than at the 900 mm level. This phenomenon was not seen on the straight faces and this difference between the broad curved and narrow straight faces was consistent for all the heats monitored. The average 43 CHAPTER 5 - RESULTS OF PLANT TRIALS temperatures and their standard deviations in each of four faces for heats 644, 686, 664 and 665 are shown in Tables 5.2-5.5. Of all the variables examined, steel carbon content and casting speed had the most significant influence on mould temperature and consequently heat transfer. Figs. 5.15 and 5.16 show the influence of steel carbon content and casting speed on mould temperature monitored on the inside curved wall. It is important to note that the effect of carbon content, for a change of 0.157 to 0.412 pet carbon, on mould temperature is considerably larger than the effect of speed, for a speed change of 1.85 to 2.29 m/min. 44 CHAPTER 5 - RESULTS OF PLANT TRIALS Table 5.1 Billet surface evaluation summary Heat % C #of Billets Bleeds or Laps Irregu-lar Spacing Non-Parallel Non-Linear Rough-ness Depres -sions Inclus-ions Pinholes OIT-Corner Depress-ions L R L R L R L R L R L R L R L R L R 642 0.155 2 2 2 2 2 1 l 2 1 2 2 0 0 0 0 0 0 2 2 643 0.18 2 1 1 2 2 2 2 2 2 2 2 0 0 0 0 l 0 2 2 644 0.16 2 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 1 2 2 645 0.17 ' 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 1 1 1 664 0.41 2 0 0 2 2 2 2 2 2 0 0 0 0 1 2 1 2 2 2 665 0.77 2 1 2 2 2 2 2 2 2 0 0 0 0 1 2 1 2 2 2 668 0.20 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 1 1 685 0.16 2 1 1 2 2 2 2 2 2 2 2 0 0 0 0 0 0 2 2 686 0.15 2 1 1 2 2 2 2 2 2 0 0 1 1 0 0 2 2 Total 16 9 10 15 15 15 15 16 15 12 12 0 0 3 5 4 6 16 16 % Defects 56 63 94 94 94 94 100 94 75 75 0 0 19 31 25 38 100 100 Note: L - left side; R - right side. 45 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.1 Photograph of longitudinal depression on a billet sample from heat 664, medium carbon steel. 46 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.2 Lasco Steel Trial December 1998, narrow face concavity and midway cracks (High carbon steel, 1.99m/min, heat 665) 47 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.3 Lasco Steel Trial December 1998, midway and off-corner internal cracks (Low carbon steel, 2.59m/min, heat 643) 48 CHAPTER 5 - RESULTS OF PLANT TRIALS i 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Carbon (%) Fig. 5.4 The influence of steel carbon contents on off-squareness. 49 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.5 Oscillation marks for low, medium and high carbon steel 50 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.6 The measured mould velocity calculated from the measured mould displacement for normal casting practice of a 0.2 % C steel cast at 2.16 m/min for heat 668. 51 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.7 Mould thermal response of thermocouples located on the inside curved wall for heat containing 0.155 pet carbon during casting speed change at approximately 1480 s. 52 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.8 Mould thermal response of thermocouples located on the inside curved wall for heat containing 0.155 pet carbon during casting speed change at approximately 1480 s. 5 3 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.9 Mould thermal response of thermocouples located on the left straight wall for heat containing 0.155 pet carbon during a casting speed change at approximately 1480 s. 54 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.10 Mould thermal response of thermocouples located on the left straight wall for heat containing 0.155 pet carbon during a casting speed change at approximately 1480 s. 55 CHAPTER 5 - RESULTS OF PLANT TRIALS <0 a. 180 160 4 140 120 E £ 100 8 0 4-6 0 C % = 0 . 1 5 5 , M n % = 0 . 8 0 6 , P % = 0 . 0 0 5 , S % = 0 . 0 2 3 , S i % = 0 . 2 1 4 , C u % = 0 . 3 9 1 , N i % = 0 . 1 3 8 , C r % = 0 . 1 0 2 2 0 0 4 0 0 6 0 0 Thermocouple Position (mm) 8 0 0 1 0 0 0 Fig. 5.11 Time-averaged mould temperature distribution on the inside curved wall for heat 642 containing 0.155 pet carbon. 56 CHAPTER 5 - RESULTS OF PLANT TRIALS 180 160 140 1 120 CL E 100 4-80 60 C%=0.155, Mn%=0.806, P%=0.005, S%=0.023, Si%=0.214, Cu%=0.391, Ni%=0.138, Cr%=0.102 -+-200 400 600 Thermocouple Position (mm) 800 1000 Fig. 5.12 Time-averaged mould temperature distribution on the outside curved wall for heat 642 containing 0.155 pet carbon. 57 CHAPTER 5 - RESULTS OF PLANT TRIALS 180 160 4 140 4 o % 120 k_ cu CL E a> *~ 100 4 80 6 0 C % = 0 . 1 5 5 , M n % = 0 . 8 0 6 , P % = 0 . 0 0 5 , S % = 0 . 0 2 3 , S i % = 0 . 2 1 4 , C u % = 0 . 3 9 1 , N i % = 0 . 1 3 8 , C r % = 0 . 1 0 2 2 0 0 4 0 0 6 0 0 Thermocouple Position (mm) 8 0 0 1 0 0 0 Fig. 5.13 Time-averaged mould temperature distribution on the left straight wall for heat 642 containing 0.155 pet carbon. 58 CHAPTER 5 - RESULTS OF PLANT TRIALS 180 160 4-140 o is 120 0) CL E 100 80 4 60 C%=0.155, Mn%=0.806, P%=0.005, S%=0.023, Si%=0.214, Cu%=0.391, Ni%=0.138, Cr%=0.102 200 400 600 Thermocouple Position (mm) 800 1000 Fig. 5.14 Time-averaged mould temperature distribution on the right straight wall for heat 642 containing 0.155 pet carbon. 59 CHAPTER 5 - RESULTS OF PLANT TRIALS Table 5.2 Time-average temperature and standard deviation on inside curved wall Position (mm) Heat 644 (0.157 %C) . Heat 686 (0.154 % C , CSp=2.53) Heat 668 (0.204 %C) Heat 664 (0.412 %C) Heat 665 (0.767 %C) CSp=2.26 ML=131 ML=111 CSp=2.32 CSp=2.13 CSp=1.99 T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std 105 69.0 5.03 55.2 2.11 64.7 3.53 68.4 5.05 73.3 6.57 68.3 5.55 135 138.2 4.46 81.3 6.49 139.8 6.57 147.6 7.68 161.8 10.88 141.3 15.71 150 108.5 4.49 106.6 6.51 116.8 3.27 121.3 13.40 141.1 3.17 141.6 3.58 165 *** *** *** *** *** *** *** *** 138.2 10.36 137.3 14.11 180 119.5 4.12 121.2 5.34 115.2 3.70 113.2 5.03 156.8 5.31 149.6 6.92 195 118.6 4.14 124.6 5.89 120.7 3.79 107.4 4.17 151.1 4.07 143.8 6.18 300 117.2 4.24 121.5 4.54 120.7 3.46 118.2 3.57 152.7 5.24 148.7 6.86 450 113.8 3.30 123.6 3.87 121.4 3.47 116.9 3.35 131.5 6.96 136.2 5.7 600 100.2 5.54 102.9 3.09 100.7 3.49 103.6 6.22 *** *** *** *** 735 105.3 3.75 112.7 3.16 110.2 3.98 118.3 3.81 121.9 5.12 121.4 4.96 850 89.0 3.41 94.0 2.92 92.0 3.57 102.2 3.88 105.9 4.84 104.7 5.20 900 87.9 3.53 87.1 2.91 85.5 3.23 96.9 3.47 101.3 4.73 98.2 4.74 950 95.2 3.83 93.1 3.54 91.5 4.09 104.2 3.64 109.4 4.92 104.5 4.25 Note: Std - standard deviation; CSp - casting speed (m/min); M L - metal level (mm). Table 5.3 Time-average temperature and standard deviation on left straight wall Position (mm) Heat 644 (0.157 %C) Heat 686 (0.154 % C , CSp=2.53) Heat 668 (0.204 %C) Heat 664 (0.412 %C) Heat 665 (0.767 %C) CSp=2.26 ML=131 ML=111 CSp=2.32 CSp=2.13 CSp=1.99 T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std 105 70.3 5.61 51.4 2.07 63.8 4.10 67.2 4.35 74.2 8.23 64.5 6.08 135 149.1 5.33 78.9 9.89 142.3 9.07 152.1 8.36 177.1 21.69 137.6 18.77 150 141.1 5.32 129.9 13.63 147.0 5.20 150.0 6.74 180.5 5.84 177.5 7.81 165 131.2 5.42 146.9 4.15 135.4 5.40 135.9 6.98 168.2 4.99 167.0 5.16 180 120.4 4.73 114.8 5.32 110.3 4.64 111.9 5.65 154.4 6.01 145.1 5.98 195 119.1 4.42 116.4 6.52 116.0 4.84 116.2 5.41 153.2 5.94 144.6 6.98 300 121.9 3.45 119.8 4.91 122.9 4.91 120.1 4.49 143.8 6.35 140.4 7.01 450 119.1 2.91 123.9 4.11 122.8 4.85 122.9 4.70 138.7 6.75 135.7 6.21 600 102.6 2.72 105.3 3.19 104.3 3.99 106.4 4.43 114.8 5.83 113.2 4.58 735 *** *** *** *** *** *** *** *** *** *** *** 850 87.5 3.17 92.4 2.85 91.3 3.79 94.5 4.86 103.7 5.60 99.52 4.67 900 89.1 3.26 88.7 2.81 87.9 3.72 90.4 5.07 97.5 5.46 93.0 4.33 950 81.5 2.96 87.4 3.20 86.9 3.95 86.9 5.44 96.6 5.82 89.9 4.97 Note: Std - standard deviation; CSp - casting speed (m/min); M L - metal level (mm). 60 CHAPTER 5 - RESULTS OF PLANT TRIALS Table 5.4 Time-average temperature and standard deviation on outside curved wall Position (mm) Heat 644 (0.157 % C ) Heat 686 (0.154 % C , CSp=2.53) Heat 668 (0.204 % C ) Heat 664 (0.412 % C ) Heat 665 (0.767 % C ) CSp=2.26 M L = 1 3 1 ML=111 CSp=2.32 CSp=2.13 CSp=1.99 T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std 105 64.5 4.86 52.7 2.92 60.7 3.52 64.4 4.86 69.6 7.02 64.6 4.78 135 132.3 4.84 76.3 8.27 132.5 6.77 140.5 7.11 153.8 11.73 141.9 14.59 150 115.8 5.13 114.9 10.35 136.5 4.11 127.5 6.12 146.7 7.18 160.1 5.50 165 99.0 4.15 145.7 4.08 134.2 4.94 106.6 6.26 143.4 8.73 154.5 6.92 180 106.6 4.29 124.9 4.48 118.0 4.09 98.6 6.66 132.4 7.57 140.7 6.15 195 119.8 4.88 132.5 6.13 129;0 4.50 105.8 7.61 146.8 8.35 151.0 8.35 300 111.8 3.66 114.9 4.59 115.2 3.94 110.4 4.19 142.1 6.61 145.2 5.99 450 102.9 3.19 113.8 2.64 111.0 3.37 109.4 3.61 122.2 5.16 122.5 4.79 600 108.1 3.32 113.9 2.55 111.3 3.78 114.0 3.85 123.7 5.30 122.6 4.51 735 *** *** *** *** *** *** *** *** *** *** *** *** 850 88.5 4.36 91.1 2.65 89.1 3.42 99.1 3.89 113.3 5.71 101.4 6.66 900 78.4 3.52 83.1 2.62 81.1 3.28 89.2 3.96 93.8 4.53 92.4 3.62 950 93.1 3.87 99.0 3.65 96.6 4.39 111.1 4.28 113.6 5.61 109.3 3.62 Note: Std - standard deviation; CSp - casting speed (m/min); M L - metal level (mm). Table 5.5 Time-average temperature and standard deviation on right straight wall Position (mm) Heat 644 (0.157 % C ) Heat 686 (0.154 % C , CSp=2.53) Heat 668 (0.204 % C ) Heat 664 (0.412 % C ) Heat 665 (0.767 % C ) CSp=2.26 ML=131 ML=111 CSp=2.32 CSp=2.13 CSp=1.99 T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std T,°C Std 105 66.7 4.86 51.5 1.61 60.8 2.62 64.0 3.94 66.1 4.22 63.0 3.42 135 141.1 4.88 77.8 6.23 140.5 8.04 162.5 8.79 158.7 13.26 143.6 15.87 150 133.0 5.84 130.7 10.00 147.7 5.47 147.5 6.70 171.7 5.30 177.8 7.83 165 126.0 5.16 144.7 4.45 131.3 5.45 123.8 6.98 154.3 4.95 160.3 8.87 180 120.2 4.48 145.7 5.70 136.8 6.58 126.5 7.92 153.9 4.11 160.0 7.23 195 119.9 4.81 118.1 6.62 113.2 4.92 112.1 5.65 142.3 5.49 146.7 9.94 300 114.3 3.50 124.8 4.24 126.2 4.65 124.3 4.50 145.0 6.04 144.8 6.54 450 111.8 3.04 119.3 3.08 117.9 3.42 116.4 3.92 130.9 6.49 131.6 6.78 600 101.5 2.57 109.0 3.21 107.9 2.82 108.5 3.45 119.1 5.56 118.9 6.15 735 *** *** *** *** *** *** *** *** *** *** *** *** 850 87.3 2.50 94.8 2.14 93.1 2.95 95.22 3.18 104.8 5.65 101.6 6.72 900 85.0 2.81 93.2 2.47 91.5 3.01 93.23 3.04 101.7 5.50 97.1 6.43 950 87.6 2.93 93.9 2.98 92.7 3.68 95.1 3.00 101.9 6.02 96.9 7.40 Note: Std - standard deviation; CSp - casting speed (m/min); M L - metal level (mm). 61 CHAPTER 5 - RESULTS OF PLANT TRIALS Fig. 5.15 Time-averaged mould temperature distribution on the inside curved wall for heats containing carbon levels of 0.157 and 0.412 pet respectively. 62 CHAPTER 5 - RESULTS OF PLANT TRIALS 60 4-40 -I 1 1 1 1 A 0 200 400 600 800 1000 Distance from the T o p of the Mould (mm) Fig. 5.16 Time-averaged mould temperature distribution on the inside curved wall for heat 664 containing 0.412 pet carbon for casting speeds of 1.85 and 2.29 m/min respectively. 63 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 6.1 The Mould Heat Transfer Model The two-dimensional steady heat transfer model t 5 8 ] was used to simulate heat transfer process occurred in the mould wall. The model calculates heat flux and the temperature distribution in a longitudinal section through the midface of the mould by means of a finite difference method. Fig. 6.1 shows a schematic diagram of the longitudinal section of the mould wall at the midface of the mould wall. The following assumptions were adopted in the formulation of the mathematical model: I. Transverse heat flow in a direction perpendicular to the longitudinal midface plane is negligible. II. Temperature variations due to mould oscillation and metal level fluctuations are ignored, which make it permissible to employ the time-average temperature distribution in the mould. III. The top and bottom surfaces of the longitudinal midface plane are assumed to be adiabatic. IV. The cooling water channel extends to the top and bottom ends of the mould and the cooling water is in turbulent plug-flow. 64 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION V. Temperature dependence of the thermal parameters has not been taken into account, as its effect on the mould heat transfer is negligible. Z = 0 X=0 X=XM X Z = Z F Backing plate/ mould jacket Z = Z M Meniscus Molten steel Cooling water Fig. 6.1 Schematic diagram of the midface longitudinal section of the mould wall1581 65 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION The governing differential equation for heat transfer in the mould wall can be written as following: d ( ,1, d ( dT^ dx Km \ DX ; dz Km \ dz j 0 For plug flow, the heat transfer in the water channel yields: PjJ.cpv (*XT(0, Z) - Tw (z)) = 0 The boundary conditions applied in the model are presented as follows: i. Cold face of mould: x=0, 0<z<Zm ox i i . Top of mould wall: 0<x<Xm, z=0 and bottom of mould wall: 0<x<Xm, z=Zm m dz i i i . Hot face of the mould below the meniscus: x=Xm, Z/<z<Zn -km — = q2{z) ox iv. Hot face of the mould above the meniscus: x=Xm, 0<z<Zf -km%- = h,(zXT(Xm,z)-Ta) dx v. Inlet water temperature: z= Zm T.„ = T (6.1) (6.2) (6.3) (6.4) (6.5) (6.6) w a (6.7) 66 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION The heat transfer between the copper mould wall and the cooling water can be determined by forced convection due to the high cooling water velocity and cold face temperature of the mould. The convection heat transfer coefficient hfC is calculated at the average temperature of water by the following correlation: (hfM = 0.023 V J r \ 0 A cpfMf (6.8) The heat flux Qfc for force convection is calculated as follows: Qfc=hfc(Tw-T(0,z)) (6.9) The mould wall was discretized in the X and Z directions as shown in Fig. 6.1 and finite difference equations were set up for the configuration nodes using the governing equation of heat transfer and boundary conditions. 6.2 Heat Flux Calculations The heat transfer model was employed to calculate mould heat flux distributions from the measured mould temperature data at steady state. The time-averaged axial mould temperature was used as input to the model to determine the axial mould heat flux profile at the centreline of the broad and narrow faces of the mould wall for a range of conditions. 6.2.1 Broad and Narrow Face Heat Flux Profiles Calculated heat flux profiles for the four faces of the mould are shown in Figs. 6.2, 6.3 and 6.4, for heats with carbon contents of 0.15, 0.41 and 0.77 pet respectively. The 67 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION four faces behave differently especially at the meniscus and over most of the mould length. In the case of the 0.15 and 0.41 pet carbon grades, the peak heat flux close to the meniscus is highest on the inside curved faces, whereas the opposite is true for the high carbon grade. The broad faces show greater variations in heat transfer along the length of the mould than either of the narrow faces. Table 6.1 presents the average heat flux on each of the four faces for all the heats monitored during the trial. From this data it is evident that the mean heat flux of the inside and outside broad faces is consistently higher than the average heat extraction on the narrow faces for almost all the heats with the exception of heats 685 and 686; both heats had steel carbon contents of 0.15 pet. 6.2.2 Comparison of Lasco Data with Results from Other Plants The axial mould heat flux profiles obtained for one of the straight walls for Lasco can be compared with data obtained in previous plant trials [ 5 9 ] . Figs. 6.5, 6.6 and 6.7 present comparisons for three steel grades with carbon contents of 0.7, 0.4 and 0.15 pet respectively. For the medium and high carbon grades, it is evident that the heat flux profiles obtained at Lasco are lower than Company C, which had a single taper of 0.4%/m at the meniscus. It has been shown in earlier studies [ 5 9 ] that a shallow meniscus taper, of less than 2.0%/m, such as at Company C, gives rise to high mechanical interaction between the mould and the solidifying shell at the meniscus and consequently high heat transfer. The taper at the meniscus at Lasco was in excess of 3.4%/m, which significantly reduces the mechanical interaction and meniscus heat transfer. Heat transfer data for Company B is lower than that measured at Lasco. Note that the heat transfer profile for the high carbon grade, 0.7 pet, at Lasco is identical to the results obtained at Atlas for high speed billet casting (Fig. 6.5). 68 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 6.3 Influence of Process Factors on Mould Heat Flux 6.3.1 The Influence of Steel Carbon Content on Mould Heat Flux It is well known that steel carbon content has the strongest influence on mould heat transfer, a result which was reconfirmed in this Lasco trial. Fig. 6.8 shows the influence of steel carbon content on the average mould heat flux. Note that the average mould heat flux is a maximum for the 0.412 pet carbon steel, and slightly lower for the 0.77 pet carbon grade. By comparison, the average heat flux is significantly lower for the grades containing carbon contents of less than 0.20 pet, with 0.155 pet carbon heats having the lowest heat flux. Singh and Blazek ( 2 4 ] showed that peritectic steels with carbon contents in the range of 0.08-0.14 pet have the lowest mould heat transfer, and found increasingly higher heat transfer for steel grades with carbon contents above 0.20 pet This study places the transition from low to high heat transfer, at approximately 0.20 pet carbon, although insufficient data in the 0.20-0.40 pet carbon range precludes determination of the exact carbon level at which the transition to higher heat transfer is completed. 6.3.2 The Influence of Casting Speed on Mould Heat Transfer Fig. 6.9 shows the influence of casting speed on axial mould heat flux profiles, for 0.77 pet carbon steel. An increase in speed causes an increase in heat extraction over the entire length of the mould. In this case, a 25% increase in casting speed, from 1.73 to 2.17 m/min for the high carbon grade gave rise to a 11.7% increase in the overall heat transfer. Clearly the increase in heat transfer for a casting speed increase cannot compensate for the reduction in residence time in the mould and subsequently, a disproportional decrease in the average shell thickness of the billet at the exit of the mould. 69 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 6.3.3 The Influence of Cooling Water Velocity on Mould Heat Transfer Mould cooling water velocity has a negligible effect on mould heat transfer, for mould water velocities in the 12-14 m/s range, as can be seen from Fig. 6.10. In this range, boiling in the cooling channel is unlikely and there is no permanent distortion of the mould tube. Thus the heat transfer coefficient at the cooling water interface has a minimal effect on mould heat transfer for cooling water velocities in excess of 12 m/s. 6.3.4 The Influence of Oil Flow Rate on Mould Heat Transfer An increase in the oil flow rate from 60 to 90 ml/min has a negligible effect of mould heat transfer as presented in Fig. 6.11. 6.3.5 The Influence of Superheat on Mould Heat Transfer Increasing superheat gives rise to a small increase in axial mould heat transfer as shown in Fig. 6.12. Higher superheat presumably results in a thinner billet shell which reduces oscillation mark depth and the average dimension of the air gap between the mould and the strand. 6.3.6 The Influence of Metal Level on Mould Heat Transfer Metal level location appears to have some influence on mould heat transfer. Fig. 6.13 shows the effect of changing the metal level position from 111 to 130 mm on mould heat transfer on the outside curved walls. Note that the peak heat flux at the meniscus increases quite significantly. It also can be observed from Table 6.1 that the average heat flux increases from 1693 to 1708 kw/m 2 for heat 685 when the metal level rises from 130 to 111 mm below the top of the mould, although the casting speed decreases from 2.60 to 2.51 m/min. This can be attributed to an increase in the effective heat transfer area of the mould as the metal level rises. 70 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 6.4 Difference Between Heat Fluxes Predicted by Model and by Cooling Water Fig. 6.14 shows a comparison of the average heat extracted by the mould as predicted from the mathematical model versus that calculated from measured inlet and outlet water temperature difference data. The agreement is good, within less than about 12%, which reinforces the validity of the values of heat transfer obtained through the use of the mathematical model. 71 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION C — 0.155 Mn-0.806 S — 0.023 Si — 0.214 Cu—0.391 CSp=2.21 m/min -Inside Lefthand - Outside Righthand 100 200 300 400 500 600 700 Distance from the Top of the Mould (mm) 800 900 1000 Fig. 6.2 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.155 pet. (Heat 642) 72 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION >< 2500 + C — 0.412 Mn-0.756 S — 0.024 Si —0.166 Cu—0.434 CSp=2.13 m/min -Inside Lefthand - Outside - Righthand 100 200 300 400 500 600 700 Distance from the T o p of the Mould (mm) 800 900 1000 Fig. 6.3 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.412 pet. (Heat 664) 73 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION C — 0.767 Mn-0.776 S — 0.026 Si —0.167 Cu—0.486 CSp=1.99 m/min -Inside Lefthand -Outside Righthand 100 200 300 400 500 600 700 Distance from the T o p of the Mould (mm) 800 900 1000 Fig. 6.4 Axial mould heat flux profiles for four faces of the mould for steel carbon content 0.767 pet. (Heat 665) 74 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION Table 6.1 Heat Flux Analysis Heat CSp Chemicals (%) Oil Tundish Temp. Heat Flux (kw/m2) w.v. (°C) No m/min M.L. C Mn S Si Cu O.F. Avg. AT Avg. In Left Out Right 642 2.21 0.155 0.806 0.023 0.214 0.391 1543 42 1589 1609 1564 1596 1575 2.02 1490 1507 1447 1507 1500 2.39 1657 1686 1655 1661 1628 643 2.59 0.183 0.756 0.042 0.203 0.443 60A90 1532 31 1770 1782 1758 1798 1724 2.73 Oil-60 1800 1819 1753 1850 1776 2.72 Oil-90 1846 1856 1841 1888 1799 644 2.26 0.157 0.865 0.026 0.245 0.354 90A60 1542 40 1578 1589 1615 1566 1541 2.26 Oil-60 1565 1570 1612 1586 1523 2.26 Oil-90 1582 1600 1609 1568 1551 645 2.21 0.169 0.825 0.028 0.220 0.431 14A12 1536 36 1598 1621 1587 1614 1551 2.19 W.V.-14 1591 1607 1585 1631 1540 2.22 W.V.-12 1592 1625 1584 1603 1554 664 2.13 0.412 0.756 0.024 0.166 0.434 1538 62 2062 2086 2043 2100 1990 1.85 1947 1997 1928 1990 1873 2.29 2115 2134 2108 2161 2057 665 1.99 0.767 0.776 0.026 0.167 0.486 1513 68 1939 1990 1867 1963 1903 2.17 2020 2071 1951 2054 2005 1.73 1807 1883 1786 1847 1711 668 2.32 0.204 0.874 0.032 0.221 0.324 1538 44 1707 1709 1702 1722 1697 2.19 1668 1671 1707 1659 1670 2.43 1729 1698 1719 1731 1736 685 2.53 0.159 0.672 0.014 0.182 0.205 111A131 1534 28 1698 1693 1702 1688 1714 2.51 M.L.-111 1708 1706 1707 1694 1724 2.60 M.L.-131 1693 1655 1719 1687 1710 686 2.56 0.154 0.746 0.014 0.196 0.315 131A111 1524 18 1636 1615 1654 1628 1658 122A156 2.53 M.L.-131 1631 1609 1613 1648 1653 2.53 M.L.-111 1639 1618 1644 1639 1654 0.F.-122 2.56 O.F.-156 1640 1615 1657 1627 1659 Note: Oil— oil flow rate (ml/min); W.V.— water velocity (m/s); M.L.— metal level (mm); O.F.—oscillation frequency AT - super heat (°C). 75 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 800 700 + 600 500 | f 400 Li-| 300 f 200 100 0 0 COMPANY GRADE (C%) LASCO 0.767 ATLAS 0.72 COMPANY C 0.56 200 400 600 800 Distance from the Top of the Mould (mm) 100 Fig. 6.5 Comparison of axial mould heat flux profiles at three companies for high carbon steels. 76 CHAPTER 6 -MOULD HEA T TRANSFER CALCULA TION C O M P A N Y G R A D E (C%) L A S C O 0.412 C O M P A N Y C 0.45 C O M P A N Y B 0.45 C O M P A N Y E 0.45 100 200 300 400 500 600 700 Distance from the Top of the Mould (mm) 800 900 1000 Fig. 6.6 Comparison of axial mould heat flux profiles at four companies for medium carbon steels. 77 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 4500 4000 4 3500 C O M P A N Y G R A D E (C%) L A S C O 0.155 C O M P A N Y B 0.15 10 15 Time Below Meniscus (s) 20 25 Fig. 6.7 Comparison of axial mould heat flux profiles at two companies for low carbon steels. 78 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 2100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Carbon Content (%) Fig. 6.8 The influence of steel carbon content of average mould heat transfer 79 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 6000 C — 0 . 7 6 7 Mn-0 .776 S — 0.026 Si — 0.167 Cu—0.486 CSp1=2.17 m/min CSp2=1.73 m/min Superheat: 68 °C 200 400 600 800 Distance f rom the T o p of the Mould (mm) 1000 Fig. 6.9 The influence of casting speed on axial mould heat flux profiles for a high carbon steel ( Heat 665). 80 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 4500 4000 3500 — 3000 + CM E 5 x 3 2500 L Z 2000 + -*-» CO iu 1 1500 1000 500 0 0 C — 0.169 Mn-0 .825 S — 0.028 Si — 0.220 Cu—0.431 CSp1=2.19 m/min (V—14 m/s) CSp2=2.22 m/min (V—12 m/s) Superheat: 40 °C 200 400 600 800 Distance from the Top of the Mould (mm) 1000 Fig 6.10 The influence of mould water velocity on axial mould heat flux profdes (Heat 645). 81 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION 3500 C — 0.157 Mn-0 .865 S — 0.026 Si — 0.245 Cu—0.354 CSp1=2.26 m/min (Oil—60) CSp2=2.26 m/min (Oil—90) Superheat: 40 °C 200 400 600 800 Distance from the Top of the Mould (mm) 1000 Fig 6.11 The influence of mould oil flow rate on axial mould heat flux profiles (Heat 644). 82 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION Fig. 6.12 The influence of superheat on axial mould heat flux profiles 83 CHAPTER 6 -MOULD HEA T TRANSFER CALCULA TION 4000 200 400 600 800 Distance from the Top of the Mould (mm) 1000 Fig. 6.13 The influence of metal level location on axial mould heat flux profiles on the outside curved wall (Heat 685). 84 CHAPTER 6 -MOULD HEAT TRANSFER CALCULATION Fig. 6.14 Graph showing the match between predicted heat extraction rate in the mould and the heat extracted by the mould cooling water. 85 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN CHAPTER 7 - SHRINGKAGE CALCULATION AND MOULD TAPER DESIGN A n existing mathematical model of billet shrinkage was employed to calculate billet shell growth and temperature distribution and to predict billet dimensions for the broad and narrow faces. Taking into account mould distortion, these results were employed to obtain ideal cold mould dimensions that would match the predicted billet shrinkage profiles. The predicted cold mould dimensions have been compared with the cold dimensions of the mould used during the trial to assess the adequacy of the taper employed at Lasco. 7.1 Mathematical Model of Billet Shrinkage The Billet Casting Group at U B C has developed a mathematical model ' 5 9 ] to describe the heat transfer in a continuously cast strand and to calculate the shrinkage of the billet as a function of its axial position in the mould. The model is based on the equation for two-dimensional, unsteady-state heat conduction in one quarter of a transverse slice of the strand (shown in Fig. 7.1) as follows: The initial and surface boundary conditions, mathematically expressed, are as follows: d f K ar N d f k ar" dx dx , _j dy 86 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN , = 0, 0 < x < y , 0<y<^, T(x,y) = Tp (7.2) t>0, x = 0, 0<y<~, ~ks^- = q0 (7.3) 2 ox X r)T t>0, y = 0, 0<x<-, -k— = qQ (7.4) 2 Assuming symmetrical heat flow at the center plane we have: X Y ?)T t>0, x = —, 0<y<-, -ks — = 0 (7.5) 2 2 ' dx Y X dT t>0, v = —, 0 < x < — , -k — = 0 (7.6) * 2 2 s By Equation 7.1 was solved, subject to the above initial and boundary conditions, by an alternating direction implicit finite difference method. The initial dimensions of the steel billet were taken as those of the distorted copper mould at the meniscus. The effect of ferro-static pressure was neglected. The mechanical behavior of the solidified shell was also neglected. Neither the strains imposed by the stress field nor creep of the solidified shell is included in the model. The differential coefficient of linear thermal expansion of steel was calculated by computing the amount phases present from the phase diagram. 7.2 Shell Growth and Billet Surface Temperature The calculated shell growth and predicted surface temperature profiles of the billets for heat 642, 664 and 665 are shown in Figs. 7.2, 7.3 and 7.4 respectively. The shell thickness at the exit of the 1016 mm long mould is about 11.5 mm for the 0.15 pet carbon steel, 12.8 mm for the 0.41 pet carbon steel and 11.9 mm for the 0.77 pet carbon steel. The midface temperature of the billet at the mould exit is approximately 1000 °C for the 87 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN low carbon steel, 770 °C for the medium carbon steel (0.412 carbon pet) and 790 °C for the high carbon steel. The temperatures on the broad face at the exit of the mould are about 50 °C lower than those on the narrow face for all heats except for heat 685 and 686, due to the higher heat fluxes on the broad face and corner effects, i.e., the heat fluxes are lower around corner region. Apparently, increasing casting speed results in higher surface temperature. However, it can be seen in table 7.1 that, although the casting speeds of 2.21-2.26 m/min in heats 642, 644 and 645 are about 12% lower than ones (2.51-2.56 m/min) in heats 685 and 686, there are no significant differences in midface temperatures at the exit of the mould due to the fact that the superheats in the first three heats are 16 °C higher than the later two. The higher superheat gives rise to higher surface temperatures of the billet. The corner temperature of the billet at first declines quickly; soon after the formation of the shell, an air gap forms due to the rapid cooling around corner of the strand leading to the low heat extraction rate. Thus, reaching the lowest value of the heat flow, the temperature rebounds. A noticeable result is the reheat of the narrow faces of the billet at a depth of approximately 600 mm below the top of the mould. It is postulated that at the 600 mm level, the narrow face concavity develops and causes a local reduction in heat transfer. 7.3 Billet Shrinkage Hot billet dimensions and those corresponding to an ideal cold mould were calculated from a billet shrinkage model developed by Chadara, et al. [ 5 9 1 The results of the calculations for the steel grades with carbon content of 0.155, 0.183, 0.412 and 0.161% are presented in Figs. 7.5 through 7.8, along with the actual (measured) cold mould dimensions in order to assess the adequacy of the taper employed at Lasco. In all cases it 88 CHAPTER 7 - SHRINKAGE CALCULATION AND MOULD TAPER DESIGN is evident that broad and narrow faces of the billet shrink differently (Table 7.1). It is important to note that for the 0.155 pet carbon steel (Fig. 7.5), binding of the billet and the mould is likely to occur over the entire length of the mould. On the other hand the taper seems to be close to optimum for the 0.183 pet carbon grade (Fig. 7.6), while the taper is clearly inadequate for the 0.412 and 0.767 pet carbon steels (Figs. 7.7 and 7.8). Besides carbon content of the steel, mould heat flux, metal level and casting speed also influence billet shrinkage. When the mould heat flux increases, billet shrinkage increases due to greater thermal contraction resulting in markedly higher taper requirements. 7.4 Mould Behavior and Billet Quality 7.4.1 Mould Taper and Related Defects Recall the longitudinal off-corner depressions, off-corner cracks, and narrow face concavity observed in the billet samples of Lasco plant trial. These defects are formed in the mould and can be linked to inadequate mould tapers (Fig. 7.5). Evidently for the 0.15 pet carbon grades, there is binding on both the narrow and broad faces of the mould over most of the mould length. The binding causes the mould to squeeze the billet at the four corners. Since there is a greater mismatch between the billet and mould dimensions for the broad faces, especially from 200-600 mm below the meniscus, the mould will interact with the billet at the corners on the narrow faces more than at the center, as shown in Fig. 7.9. This could result in the development of concavity observed on the narrow faces. This appears to occur 600 mm below the meniscus, because the big binding just begins at this location, causing the narrow faces to reheat and expand. Consequently, tensile stresses would arise at the solidification front and cause the formation of off-89 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN corner internal cracks, as shown schematically in Fig. 7.9. The average depth of the cracks observed below the surface is approximately 8.0 mm, which coincides with the narrow face shell thickness at approximately 650 mm below the top of the mould and with the maximum reheating of the narrow face shell. The 0.41 and 0.77 pet carbon steel grades shrink considerably more in the mould and there is no binding either on the broad or narrow faces of the mould as shown in Figs. 7.7 and 7.8. However, large gaps open up between the mould and the billet. The broad faces are prone to bulging with increasing distance below the meniscus, because of increasing ferro-static pressure. This could also cause rotation of the narrow face corners as shown schematically in Fig. 7.10, leading to the generation of tensile stresses at the solidification front and off-corner internal cracks, as described in earlier publications [ 6 1 ' 6 2 ] . This event also seems to occur at around the 600-700 mm location below the top of the mould. The accompanying narrow face reheating and dimensional expansion are significantly smaller than for the low carbon grades. 7.4.2 Mould Heat Transfer, Metal Level Fluctuations and Off-squareness In previous studies, [ 5 2 , 5 3 ' it has been shown that off-squareness is the worst in the medium carbon grades (0.18-0.40 pet). High carbon grades are less prone to off-squareness, while off-squareness is minimal in low carbon steels (0.08-0.14 pet). These differences have been explained on the basis that the medium carbon steels have the shortest freezing range and the highest mould heat transfer at the meniscus. Metal level fluctuations result in significant variations in shell thickness around the mould periphery, which is the origin of off-squareness. Low carbon steels have low heat transfer, while high carbon steels have a long freezing range, both of which result in thinner shells at the 90 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN meniscus. Thus metal level fluctuations give rise to minimal variations in shell thickness around the mould perimeter, resulting in lower off-squareness. The findings in this study support these earlier observations [ 5 2 ' 5 3 ] . Fig. 7.11 shows that off-squareness is worse in the medium carbon grades and increase with casting speeds, although the number of data points are not adequate for a rigorous statistical analysis. A correlation between differences in temperature of adjacent faces of the mould at the 950 mm level was observed as shown in Fig.7.12. The large temperature differences are prone to the serious off-squareness of the billets. For the medium carbon steels, there was also limited evidence suggesting that high off-squareness was related to higher fluctuations in meniscus temperature, which is linked to greater metal level turbulence as shown in Fig. 7.13. Meniscus turbulence had less of an effect on off-squareness in the low and high carbon grades. These findings suggest that it would be worthwhile to conduct a long term study to establish whether thermocouples located at the meniscus, and at the bottom of the mould could be used to successfully detect off-squareness quantitatively. This would provide a means of on-line detection of off-sqareness and corrective action could then be taken. 7.5 Taper Design Steel carbon content is the most important factor influencing the mould taper requirements as presented in Figs. 7.14 and Fig. 7.15. These two figures compare the optimum mould taper with the existing taper for the three grades that shrink differently -0.15, 0.41 and 0.77 pet carbon. Fig. 7.14 shows narrow face dimensions (broad face taper) and Fig. 7.15 shows broad face dimensions (narrow face taper). Clearly the existing taper is too severe for the 0.15 pet carbon grade and inadequate for the other two 91 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN grades. The heat flux for the 0.15 pet carbon steel is lowest and hence the billet shrinks less than the other two grades. Furthermore in the upper part of the mould, the narrow face shrinkage is higher than the broad face shrinkage (hence a larger broad face taper is required) than lower down in the mould where the reverse is true. Table 7.2 summarizes the taper requirements calculated for each heat monitored during the trial. 92 CHAPTER 7-SHRINKAGE CALCULATIONAND MOULD TAPERDESIGN 1 y 5 0.1) e o e—— Oj) e ) o o o o ) o o o (ij) o 5 o o o o 3 o o o o 0.1) y= Y/2 x = X/2 Fig. 7.1 Mesh used for modeling one quarter of a transverse section of a billet. [ 5 9 ) 93 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.2 Predicted billet surface temperature and shell thickness profiles for heat 642 containing 0.155 pet carbon. 9 4 CHAPTER 7 - SHRINKAGE CALCULATION AND MOULD TAPER DESIGN Distance from the Top of the Mould (mm) Fig. 7.3 Predicted billet surface temperature and shell thickness profiles for heat 664 containing 0.412 pet carbon. 95 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.4 Predicted billet surface temperature and shell thickness profiles for heat 665 containing 0.767 pet carbon. 96 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN Table 7.1 Shrinkage analysis Heat No 642 643 644 645 664 665 668 685 686 CSp (m/min) 2.21 2.59 2.26 2.21 2.13 1.99 2.32 2.51 2.60 2.56 2.53 Chemicals (%) C 0.155 0.183 0.157 0.169 0.412 0.767 0.204 0.159 0.159 0.154 0.154 M n 0.806 0.756 0.865 0.825 0.756 0.776 0.874 0.672 0.672 0.746 0.746 S 0.023 0.042 0.026 0.028 0.024 0.026 0.032 0.014 0.014 0.014 0.014 Si 0.214 0.203 0.245 0.220 0.166 0.167 0.221 0.182 0.182 0.196 0.196 Cu 0.391 0.443 0.354 0.431 0.434 0.486 0.324 0.205 0.205 0.315 0.315 Parameters CSp 2.02-2.39 Oi l 6 0 -90 Oi l 90-60 W.V. 14-12 CSp 1.85-2.29 CSp 1.73-2.17 CSp 2.19-2.43 M . L . I l l M . L . 130 M . L . 130 M . L . I l l Tundish Temp. (°C) Avg 1543 1532 1542 1536 1538 1513 1538 1534 1534 1524 1524 AT 42 31 40 36 62 68 44 28 28 18 18 Slirinkage Status Narrow B N B B G G B — G N N B B Broard B G B G — B G G G G G G G Shell Thickness (mm) Broad 11.43 11.35 11.33 12.03 12.69 11.84 11.70 ' 11.41 11.10 11.37 11.18 Narrow 11.78 11.38 11.69 12.14 12.64 11.62 11.88 11.83 11.22 11.73 11.24 Billet Midfece Temp. (°C) Broad 1004 958 1030 986 733 761 933 986 1016 1023 1021 Narrow 1056 1026 1063 1042 817 821 994 1003 1025 1033 1043 Heat Flux (kw/m 2) Broad 1603 1790 1578 1618 2092 1977 1716 1691 1671 1622 1628 Narrow 1570 1741 1578 1569 2017 1885 1694 1708 1715 1656 1633 Note: O i l — O i l Flow Rate (ni/min); W . V . — Water Velocity (m/s); M . L . — Metal Level (mm); B — Binding; G — Gap; N — Normal. 97 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 96.2 94.8 -I £ 1 1 1 1 1 1 1 • 1 h-l 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) 66.8 -P 66.7 4 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.5 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.155 pct.(heat 642) 98 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 96.2 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) 66.8 66.7 65.8 -| \1 1 1 j 1 1—• 1 1 1 h1 0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.6 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.183 pct.(heat 643) 99 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 66.8 •g- 66.6 0 66.4 w c OJ 1 66.2 fl) 66 co T5 CO T3 5 65.8 o 65.6 111 mm (M.L.) hot billet dimension cold mould dimension current mould -+-C —0.412 Mn-0.756 S — 0.024 Si —0.166 Cu—0.434 CSp=2.13 m/min Superheat: 62 °C - t -0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.7 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.412 pct.(heat 664) 100 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 96.2 Distance from the Top of the Mould (mm) 66.8 66.7 4-0 100 200 300 400 500 600 700 800 900 1000 Distance from the Top of the Mould (mm) Fig. 7.8 A comparison of mould and billet dimension for the broad and narrow faces of steel with carbon content 0.767 pct.(heat 665) 101 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN Fig. 7.9 Schematic diagram showing the development of concavity on the narrow face due to excessive squeezing at the corners 102 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN O f f Corner Rotation Longi tudinal Depression Inadequate taper Fig. 7.10 Schematic diagram showing the formation of off-corner internal cracks due to bulging at the broad face and rotation of the corners. 103 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN ~ 84-• Low Carbon gHigh Carbon A Medium Carbon 2.1 2.2 2.3 2.4 2.5 Casting Speed (m/min) 2.6 2.7 2.8 Fig. 7.11 The influence of casting speed on off-squareness 12 10 E E, 0) o c V I 5 JO TO C o TO 2 4 «> Low Carbon •H igh Carbon A Medium Carbon 12 17 22 Temperature Difference (°C) 27 32 Fig. 7.12 The relationship between off-squareness and temperature difference of adjacent faces of the mould wall at 950 mm below the top of the mould. 104 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN 12 10 ^ 6 a o • Law Carbcn HHoji Carbon AlVtedumCatcn A A + 2 4 6 8 10 Standard Deviation of Therrral Couple near Meniscus (°C) 12 14 Fig. 7.13 The relationship between off-squareness and standard deviation of the thermocouple near the meniscus. 105 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN Fig. 7.14 Calculated mould tapers for the narrow face of the 5x7 inch mould. 106 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN Fig. 7.15 Calculated mould tapers for the broad face of the 5x7 inch mould. 107 CHAPTER 7 - SHRINKAGE CALCULA TION AND MOULD TAPER DESIGN Table 7.2 Predicted mould tapers Distance from Top of M o u l d (mm) Taper (%/m) or Casting Speed (m/min) H642 H643 H644 H645 H664 H665 H668 H685 H686 Current Narrow 1 0 1 . 6 - 2 0 3 . 2 2.20 2.62 2.30 2.33 3.45 3.11 2.41 2.49 2.38 2.65 2 0 3 . 2 - 4 0 6 . 4 1.21 1.35 1.20 1.55 1.76 1.83 1.34 1.48 1.58 1.51 4 0 6 . 4 - 1016 0.87 0.93 0.80 0.80 1.16 1.07 1.08 0.80 0.65 0.79 Broad 1 0 1 . 6 - 2 0 3 . 2 2.61 3.33 2.76 2.82 4.65 3.37 3.39 3.52 3.40 2.65 2 0 3 . 2 - 4 0 6 . 4 1.34 1.45 1.23 1.38 1.88 1.78 1.62 1.47 1.63 1.51 4 0 6 . 4 - 1016 0.56 0.54 0.51 0.61 0.76 0.66 0.63 0.68 0.56 0.79 C S p (m/min) 2.21 2.59 2.26 2.21 2.13 1.99 2.32 2.51 2.56 1.78/2.29 108 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING Spray cooling systems in three plants have been investigated. The heat transfer during spray cooling has been calculated employing a mathematical model. Both temperature distributions of the strand and spray-related defects have been analyzed. The spray system has also been designed for high speed billet casting at Alta Steel. 8.1 Plant Data Spray cooling data provided by three companies have been analyzed. Parameters, such as casting speed, spray chamber length, number of spray zones, nozzle types, nozzle number, water pressure, nozzle stand off and water flow rate are shown in Tables 8.1a and 8.1b. Tables also include the calculated data, for instance, heat transfer coefficients, water flux and strand surface area of spray in different zones. It should be noted that the heat transfer coefficients were calculated by formula (8-4) [ 3 8 ] and assuming 1000 °C for the billet surface temperature. The spray chamber lengths of casting in three plants vary from 1.2 to 3.04 m. The spray systems usually consist of 2 or 3 zones. However, different nozzles and water flow rate distributions are employed in each zone. In the Stelco McMaster case, the spray chamber is divided into 3 zones; zone 1 consisting of 3 sub-zones and zone 2 and 3 consisting of 2 sub-zones respectively. In view of this design seven zones are used in calculation and analysis. The water fluxes are arranged so as to decrease down the strand 109 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING with the spray zones. The first zone supplies the largest water flux; the last one sprays the smallest flux. 8.2 The Spray Heat Transfer Model A two-dimensional mathematical model is to predict the temperature distribution and shell thickness profiles of the strand. The model is based on the fundamental equation of transient heat conduction, and on empirical data to characterize the complex heat extraction processes at the surface of the strand in the different cooling zones. The primary equation for the processes can be written as follows: dt ox ox dy dy (8-1) y q y (y=Y/2) q x (x=X/2) Fig. 8.1 Schematic drawing of transverse slice The solution of above equation gives the temperature distribution in a transverse slice of steel descending through the casting machine at the same withdrawal rate as the strand. 110 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING A schematic diagram of the slice is shown in Fig. 8.1. The dark part is selected as the calculation region due to symmetrical heat transfer of the strand. The boundary conditions are the heat fluxes qx and qy. Considering symmetrical cooling conditions, they are zero at the symmetrical axis (x=0, or y=0). On the surfaces, they can be expressed in various ways depending on the cooling of the billet. In the mould, the heat fluxes qx and qy can be obtained by measuring the mould temperature in a plant trial or by an empirical formula. In the spray cooling zone: where Ta is the ambient temperature, a is Stefan-Boltzmann constant (5.6703xlCT8 W m"2 K" 4) and s is the radiation emissivity (e =0.8 for steel). , The initial condition is given by specified temperature at the meniscus (corresponding with the tundish temperature). The heat transfer coefficient h in spray cooling was decided by Sasaki's empirical formula ( 3 8 ] , as shown in equation (8-4), and has been applied in the billet spray calculation. The experimental conditions such as billet temperature range, scales of pressure and water flow, are also indicated in the formula. q = h(T-Tw) (8-2) where Tw is the temperature of the cooling water, q represents qx and qy. In the radiation zone: (8-3) 111 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING h = 70SW015TS~L2 + 0.116 (kw/(m2.°C) , l/m2.s) (700<7;<1200oC, \.61<W<A\.61l/m2s, 196<pressure<490 kPa) (8-4) A finite difference method is employed to solve the two-dimensional transient heat conduction equation. 8.3 Results and Analysis 8.3.1 Shell Thickness and Surface Temperature At earlier stage, the shell grows quickly in the mould, a little slower with the spray cooling and then very slowly in the radiation zone as seen in Fig. 8.2. Near the bottom of the liquid pool, the shell thickness increases rapidly due to a diminution in the latent heat of solidification of the steel. Centerline segregation and cracks probably take place there. After a rapid drop in the mould, the midface temperature of the strand keeps in a certain range, usually from 1000 to 1200 °C, with some reheating when the strand goes through the secondary cooling zone. It is evident that the surface temperature during secondary cooling for low carbon steel is about 1100-1200 °C, 100 °C higher than that for high carbon steel, as shown in Figs. 8.2 through 8.5. Besides, billet size, casting speed, mould length and mould taper also impact the billet surface temperature during spray cooling. 8.3.2 Surface Reheating The surface temperature of the strand will increase when the strand passes from a cooling zone with a high heat transfer rate to one with a lower heat transfer rate. This is caused by a relaxation of the large temperature gradients created during the high heat transfer period and subsequent accumulation of enthalpy in the surface of the casting. Table 8.2 shows the correlation of surface reheating with casting speed, billet size, and 112 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRAY COOLING steel grades in different plants. Although McMaster has the largest spray water flow rate (180-202 USg/min.), the maximum surface reheating of the strand is the smallest (104-147 °C) of all three plants. This is mainly due to the number of spray cooling zones (7 sub-zones) employed, which decreases the surface temperature reheating. The longer spray chamber, 3040 mm in McMaster, compared to 1200-2500 mm at the other plants, has also contributed to the lower surface reheating for casting with oil lubrication. The maximum surface reheating is 93-200 °C in Alta, and 84-238 °C in Atlas. It also can be seen from Table 8.2 that the maximum reheating takes place at zone 1 and zone 2 in most cases (sub-zone 3 of zone 1 in McMaster). Thus, the strong spray cooling is required just below the mould. 8.3.3 Metallurgical Length The depth of the liquid pool has a significant influence on the formation of internal cracks and formation of centerline segregation. Due to the rapid drop of center temperature of the strand near the bottom of liquid pool and the low ductility of steel at a temperature close to the solidus, the strand is very susceptible to cracks and centerline segregation. The depth of the liquid pool usually controls the position of the cutting torch. Among various factors to influence the pool depth, casting speed and billet section size perform important functions. The billet metallurgical lengths calculated for the different companies are summarized in Table 8.2. When the casting speed increases from 1.9 m/min, for 152x152 mm 2 billet at Alta, to 2.3 m/min for 150x150 mm 2 billets at McMaster, the pool depths increase about 1 m, from 15.78 to 16.94 m for low carbon steel and 16.20 to 16.90 m for high carbon steel. Similarly, when the casting speed 113 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING changes from 2 .9 to 4 .2 m/min for 1 2 0 x 1 2 0 mm2 billet at Alta and 1 1 0 x 1 1 0 mm2 billet at McMaster, the pool depths increase about 3 m, from 1 6 . 4 9 to 19 .44 m for low carbon steel and 1 7 . 0 0 to 2 0 . 0 9 m for high carbon steel. Comparing the pool depth of 1 2 0 x 1 2 0 (casting speed is 2 .9 m/min.) and 1 5 0 x 1 5 0 mm2 (casting speed is 2.3 m/min.) billets, the pool depths are almost the same; i.e., 17 and 16 .9 m respectively. Pool depth increases with the increasing of section size nearly compensates its increase with casting speed in this case. The distance from the top of the mould to the position of the cutting torch (metallurgical length) is typically about 2 2 m for billet casting. For most cases shown in table 8 .2, the calculated depths of the liquid pool are shorter than this distance. However, the liquid depths of most strands are longer than the distances from the top of the mould to the pinch rolls or unbending point. The mechanical deformation would have an adverse effect upon the quality of the products. It should be indicated that McMaster supplied the currently used spray cooling data for an expecting high casting speed, which causes the calculated liquid depths are a little longer than the metallurgical length. 8.3.4 Spray-related Defects . Several quality problems can arise in the secondary cooling of the continuous casting of steel, including midway cracks and rhomboidity. Midway cracks can be detected in sulfur prints and macroetches of transverse sections (Figs. 5.2 and 5 .3) . They are caused by surface reheating of the strand. The mechanism of crack formation can be briefly described as follows: Surface reheating forces the surface to expand and, in so doing, imposes a tensile strain on the interior, hotter regions of the solidification front, which are weaker and nonductile above 1 3 4 0 °C | 3 6 ] . The tensile strain and stress run parallel to the 114 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING surface, and thus, cracks form perpendicular to the surface, depending on the magnitude of the strain. From the above description, it is clear that midway crack formation can be prevented by minimizing the tensile strain; i.e., surface reheating. Minimization of the tensile strain can be achieved by paying close attention to the design of the spray system to ensure that the rate of cooling does not decrease abruptly between mould and sprays, sprays and radiation cooling or between successive spray zones. Maintenance of the spray system is also an important consideration because clogged or poorly positioned nozzles can cause local reductions in cooling. The other spray-related defect in billets is rhomboidity, although these problems may be traced more frequently to the mould as shown in Fig 5.3. The cause of the rhomboidity is unsymmetrical cooling, which can arise in the upper sprays if water pressure on all four risers is not equal or nozzles are plugged or bent. Thus, if two adjacent faces are being cooled more rapidly than the other faces, the billet contracts to generate a diagonal tensile strain between the colder faces. If the strain is large, the billet distorts and takes on a rhomboid shape with an acute-angle between the colder faces. Rhomboidity formed in the mould may become worse in the spray zone. As stated above, if the surface temperatures at the acute-corner of the billet is below 550 °C, the heat transfer coefficient increases sharply as nucleate boiling begins to take effect. Consequently the temperature at the acute-corner drops quickly including large tensile stresses which menace rhomboidity. It is also conceivable that unsymmetrical spray cooling may shift the strain within the mould to create non-uniform cooling in the mould. 115 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING To minimize rhomboidity, care must be taken to achieve equal cooling on each of the four faces. This requires good alignment between the mould and roller cages and avoidance of plugged or bent spray nozzles in the secondary cooling zone. 8.4 Spray Design The spray chamber of a continuous casting machine must fulfill important thermal requirements if steel is to be cast efficiently with a minimum of internal or external defects. The sprays must remove sufficient heat from the steel to virtually complete solidification of the cast section. The rate at which the heat extraction proceeds is critical to the smooth operation of the process, because undercooling can result in excessively long liquid pools and overcooling can lead to the formation of cracks. The heat extraction in the sprays must also be arranged to achieve a smooth transition of the surface temperature, with a minimum of reheating, as the steel passes from the mould to the sprays and from the sprays to the radiation cooling zones. Water flux distribution and length of the spray chamber can be adjusted to optimize the performance of the sprays. In order to determine the best design for a given casting parameters, Brimacombe'161 set two criteria to define the optimum thermal conditions: minimization of midway crack formation and maintenance of a reasonably high solidification rate. The first condition can be met by minimizing reheating of the surface of the strand either at or below the sprays. The second criterion can be met by maintaining the surface temperature of the strand in the sprays between 1000 and 1100 °C. Alta Steel required increasing the casting speed by about 25%. To do this the spray cooling system must be redesigned. The current spray chamber is 1866.9 mm long and divided into two zones, where zone 1 is 444.5 mm and zone 2 is 1422.4 mm. Under this 116 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING spray system reheating of the billet surface is very large (93 to 200 °C) at or below the sprays. The spray design was carried out employing a mathematical model. In order to smooth the transition of the surface temperature and to achieve a reasonable liquid pool depth for higher the casting speed, a three-zone spray chamber and extensive spray cooling were employed. The new spray chamber is 3289.3 mm long. This new system kept the surface reheating of the billet to less than 100 °C in the spray chamber and radiation zone. Billet temperature distributions and shell thickness were presented in Figs. 8.6 to 8.11. The surface temperatures of the billet were basically between 1000 and 1200 °C, 1000 to 1100 °C for the high carbon grade and 1100 to 1200 °C for the low carbon grade in spray cooling zone for different billet sizes. The liquid pool depths were shorter than the maximum metallurgical length. The different spray zone length, water flux distributions in each zone and surface reheat at sprays and radiation are shown in Table 8.3. 117 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING Table la Supplied and calculated data of spray system from three companies C o m p a n y At las McMas te r Al ta Grades H-Carbon, Stainless LH-Carbon LH-Carbon Lubricant Powder Oil Oil Billet S ize (mm^) 114x114 146x146 254 x203 110x110 150x150 120x120 152x152 203x203 M o u l d Length(mm) 780 813 734 C S p (m/min) 1.9-2.7 1.2-2.8 0.65-0.85 4.2 2.3 2.9 1.9 1 Spray Length (mm) 1200 1200 1200 3040 3040 1866.9 1866.9 1866.9 zonel 200 400 200 38.1 38.1 444.5 444.5 444.5 zone2 1000 800 1000 76 76 1422.4 1422.4 1422.4 zone3 206.5 206.5 zone4 587.7 587.7 zone5 933.45 933.45 zone6 546 546 zone7 648 648 Nozzle Type zonel 6515 Flat 5008 Flat 8008Flat 1/4flat 3/8gg 8.1 1/4gg 10 1/4gg 10 1/4gg 10 zone2 T63.5 Cone T63.5Flat 5004Flat 1/4hh-10 1/4hh 12.5 1/4gg 10 1/4gg 10 1/4gg 10 zone3 1/4hh-10 1/4hh 12.5 zone4 1/4gg 6.5 1/4hh6.5 zone5 1/8gg 5.0 1/8gg 5.0 z0ne6-7 1/8gg 5.0 1/8gg 6.1 Nozzle Number 28 28 24 112 112 48 48 48 zonel 8 12 8 8 8 16 16 16 zone2 20 16 16 8 8 32 32 32 zone3 8 8 zone4 28 28 zone5 36 36 z0ne6-7 24 24 Pressure (psi) 80 80 36 38 40 Stand Off (inch) 4.62 5 6.25 1 1 8 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING Table 8.1b (con ' t ) Supplied and calculated data of spray system from three companies Company Atlas McMaster Alta Grades H-Carbon, Stainless LH-Carbon LH-Carbon Lubricant Powder Oil Oil Billet Size (mm') 114x114 146x146 254 X203 110x110 150x150 120x120 152x152 203x203 Mould 780 813 734 CSp (m/min) 1.9-2.7 1.2-2.8 0.65-0.85 4.2 2.3 2.9 1.9 1 Water Flow (usg) 56-78 42-63 44-58 180.11 202.40 89 92 93 zonel 24-34 18-27 17-22 11.29 16.80 30 31 31 zone2 32-45 24-36 27-36 21.61 26.38 59 61 62 zone3 21.61 26.38 zone4 47.62 47.62 zone5 46.80 46.80 zone6 15.60 19.21 zone7 15.60 19.21 Area (m2) 0.55 0.70 0.61 0.49 1.34 1.82 0.90 1.14 1.51 zonel 0.09 0.23 0.10 0.08 0.02 0.02 0.21 0.27 0.36 zone2 0.46 0.47 0.51 0.41 0.03 0.05 0.68 0.86 1.15 zone3 0.09 0.12 zone4 0.26 0.35 zone5 0.41 0.56 zone6 0.24 0.33 zone7 0.29 0.39 Water Rux(L/m2/S) 7.72 4.77 3.00 2.75 8.49 7.00 6.27 5.11 3.87 zonel 20.06 6.21 6.83 8.55 42.47 46.37 8.87 7.24 5.42 zone2 5.26 4.05 2.24 2.79 40.76 36.49 5.45 4.45 3.39 zone3 15.00 13.43 zone4 11.62 8.52 zone5 7.19 5.27 zone6 4.10 3.70 zone7 3.45 3.12 H. Trans. C (kw/m2/ °C) 0.82 0.57 0.41 0.48 0.88 0.77 0.70 0.60 0.49 zonel 1.69 0.70 0.75 0.89 2.96 3.16 0.91 0.78 0.63 zone2 0.62 0.51 0.33 0.39 2.87 2.64 0.63 0.54 0.44 zone3 1.36 1.25 zone4 1.12 0.89 zone5 0.78 0.62 zone6 0.51 0.47 zone7 0.45 0.42 Heat Flux (kw/m2) 823.99 574.09 405.5 479.7 884.87 765.34 704.27 604.70 490.71 1 1 9 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING Table 8.2 Surface reheating and depth of liquid pool at different plants Plant Steel grade Billet size (mm2) Casting speed (m/min) Water flow, (usg/min) Max surface reheating broad/narrow Location for max reheat (zone) Number ofzone Depth of liquid pool (m) Distance from meniscus to torch /pinch rolls (ni) L C 203x203 1.0 93 131 air 2 12.8 H C 203x203 1.0 93 179 air 2 12.7 Alta L C 152x152 1.9 92 111 1 2 15.8 26.52/ H C 152x152 1.9 92 200 1 2 16.2 8.28 L C 120x120 2.9 89 93 1 2 16.5 H C 120x120 2.9 89 177 1 2 17.0 A 114x114 2.4 67 184 2 2 13.0 M 114x114 1.9 67 216 2 2 9.0 H C 114x114 2.19 67 189 2 2 11.1 A 146x146 1.5 53 238 1 2 12.0 22/9.8 Atlas M 146x146 L5 53 157 1 2 11.3 H C 146x146 1.5 53 182 1 2 11.8 A 254x203 0.85 58 84/91 2 2 15.6 M 254x203 0.85 58 106/115 2 2 13.9 H C 254x203 0.85 58 94/102 2 2 14.9 L C 110x110 4.2 180 104 3 7 19.4 M c - H C 110x110 4.2 180 112 3 7 20.1 16.41/ Master 12.14 L C 150x150 2.3 202 137 3 7 16.9 H C 150x150 2.3 202 147 3 7 16.9 Note: A: austenitic stainless steel; M: martensitic stainless steel; LC: low carbon steel; HC: high carbon steel. 120 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRAY COOLING Distance from the Top of the Mould (mm) Corner Midface • » • - > • center ^ — — Shell Thickness Fig. 8.2 Temperature distribution and shell thickness in Alta continuous casting (low carbon steel, 152x152 mm2, casting speed 1.9 m/min). 121 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING Fig. 8.3 Billet temperature and shell thickness at Alta Steel, high carbon steel, 152x152 mm2, casting speed 1.9 m/min. Distance from the Top of the Mould (mm) | Corner Midface • center — shell thickness | Fig. 8.4 Billet temperature and shell thickness at McMaster Steel, low carbon steel, 150x150 mm2, casting speed 2.3 m/min. 122 CHAPTER 8 - ANALYSIS OF HEAT TRANSFER IN SPRAY COOLING Distance from the Top of the Mould (mm) | Corner -Midface center — Shell Thickness] Fig. 8.5 Billet temperature and shell thickness at McMaster Steel, high carbon steel, 150x150 mm2, casting speed 2.3 m/min. 123 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING Table 8.3 Alta Steel spray cooling design Spray chamber Length (mm) Billet size (mm2) Casting speed (m/min) Water flux (l/m2s) Grade Surface reheat CQ Liquid pool depth (m) Zone 1 4 4 4 . 5 1 2 0 x 1 2 0 3 .6 15 L C 19 HC 7 0 1 5 2 x 1 5 2 2 .4 12.5 L C 2 4 HC 93 2 0 3 x 2 0 3 1.25 7.0 L C 4 4 HC 98 Zone 2 1 4 2 2 . 4 1 2 0 x 1 2 0 3 .6 8.0 L C 4 2 HC 5 2 1 5 2 x 1 5 2 2.4 6.0 L C 6 0 HC 75 2 0 3 x 2 0 3 1.25 3.5 L C 4 8 HC 68 Zone 3 1422 .4 1 2 0 x 1 2 0 3.6 3.5 L C 6 0 HC 6 7 1 5 2 x 1 5 2 2.4 2.5 L C 63 HC 93 2 0 3 x 2 0 3 1.25 1.5 L C 4 9 HC 6 2 Radiation 1 2 0 x 1 2 0 3 .6 L C 73 18 .6 HC 98 19 .2 1 5 2 x 1 5 2 2 .4 L C 64 18.5 HC 93 18.8 2 0 3 x 2 0 3 1.25 L C 4 0 16 .2 HC 7 0 16 .2 Note: HC — high carbon steel; LH — low carbon steel. 124 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRA Y COOLING Fig. 8.6 Alta Steel: billet temperature and shell thickness for designed sprays high carbon steel, 203x203 mm2, casting speed 1.25 m/min. Fig. 8.7 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 203x203 mm2, casting speed 1.25 m/min. 125 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRAY COOLING Fig. 8.9 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 152x152 mm2, casting speed 2.4 m/min. 126 CHAPTER 8 - ANALYSIS OF HEA T TRANSFER IN SPRAY COOLING Dlstance from the Top ofthe Mould (mm) Fig. 8.10 Alta Steel: billet temperature and shell thickness for designed sprays high carbon steel, 120x120 mm2, casting speed 3.6 m/min. D istance fro m the Top of the Mould (mm) Fig. 8.11 Alta Steel: billet temperature and shell thickness for designed sprays low carbon steel, 120x120 mm2, casting speed 3.6 m/min. 127 CHAPTER 9 - SUMMARY AND CONCLUSION CHAPTER 9 - SUMMARY AND CONCLUSION A study was undertaken to analyze the impact of several casting variables on the heat transfer for a long mould having a parabolic taper during casting of large section billets. The interaction between the mould and strand as well as the generation of defects such as off-corner internal cracks and narrow face concavities were examined. Spray cooling heat transfer was also studied for high speed billet casting. Data on mould wall temperature, metal level, casting speed and billet quality were acquired during a plant trial conducted at Co- Steel Lasco. In the plant trial, an operating billet mould was instrumented with an array of thermocouples, together with LVDT's . The water temperature was also measured at the inlet and outlet of the cooling channel. Metal level and casting speed signals were obtained from plant instrumentation. Billet samples for various operating conditions were collected for quality evaluation. Mould heat fluxes were calculated by a heat transfer model from measured mould wall temperatures. Several mathematical models were employed to investigate mould-billet interaction and to design the mould taper. Results from plant measurements, model calculations and billet sample evaluations were utilized to assess the impact of operating conditions on billet quality for 6 carbon steel grades. Spray cooling systems in three plants were also investigated. The temperature distribution of the strand was calculated from a mathematical model. The formation of 128 CHAPTER 9 - SUMMARY AND CONCLUSION spray-related defects, particularly midway cracks, was linked to billet surface reheating. A new spray system was also designed for high speed continuous casting (at Alta Steel). The main conclusions that can be made from this work are as follows: (1) Steel carbon content and casting speed had significant effects on heat flux of the mould. The heat flux was highest for the medium carbon grade (0.412 pet) followed by the high carbon grade (0.76 pet), with the low carbon grades (0.155-0.159 pet) transferring the least amount of heat to the mould. A 25% increase in casting speed led to about 11% increase in the mould heat flux. (2) The most serious defect on all the sections was off-corner internal cracks, located at approximately 8 mm below the surface. All the billets also had concavity on their narrow faces. It was demonstrated that the taper was too tight for the low carbon grades which was responsible for the narrow face concavity and off-corner internal cracks. For the medium (0.412% carbon) and high carbon billets, the mould taper was inadequate, especially in the lower part of the mould. Bulging of the broad face and corner rotation gave rise to longitudinal depressions on the narrow face and off-corner internal cracks. (3) Triple tapered moulds have been designed for low carbon, medium carbon and high carbon steels. Generally, different grades have different taper requirements. Low carbon grades require shallower tapers, while medium and high carbon grades require steeper tapers. (4) Large surface reheating usually occurred after the first zone for the three spray cooling systems studied, which correlated with casting speed, billet size, steel grades. It was determined that additional spray zones and a longer spray 129 CHAPTER 9 - SUMMARY AND CONCLUSION chamber were necessary in order to avoid large surface reheating of the strand (for instance, 7 zones and 3040 m long as used at McMaster). Too great an intensive cooling or a sudden cooling decrease in adjacent zones resulted in large surface reheating which could induce midway cracks at the solidification front of the strand. (5) A new spray system was designed for Alta Steel for a casting speed increase of 25%. Intensive spray cooling, a long spray chamber and three spray zones were employed in the spray design in order to achieve a billet surface temperature between 1000 and 1200 °C in the spray zones and a surface reheating of less than 100 °C. 130 REFERENCES REFERENCES [I] W.R. Irving, Continuous Casting of Steel, The Institute of Materials, 1993. [2] M . Pavlicevic, C. Tercelli and B. Matijasevic, 'High Speed Casting: A New Technology Revolution with D A N A M Mould,' Proceedings of the International Symposium on Near-Net-Shape Casting in the Minimills, Vancouver, British Columbia, Canada, 1995, pp.205-215. [3] R. Bruder, J. Wolf and A. Borowski, 'Investigation and Results of SBQ Billets Cast with Double Speed at Thyssen Stahl A G , Oberhausen,' METEC Proceedings, Vol. 1, 1994, pp. 301-304. [4] M . Tsutsui, M . Shioki, O. Tanno, et at, 'High-Speed Billet Casting with Convex Mould at Himeji Works, TOA Steel Co. Ltd. In Japan,' METEC Proceedings, Vol. 1, 1994, pp. 291-294. [5] K. Takashi, H. Sei, et al, 'Behavior of lubrication and heat transfer in mould at high speed continuous casting,' Tetsu-to-Hagane, Vol. 83, No. 11, Nov. 1997, pp. 701-706. [6] R. J. Dippenaar, I.V. Samarasekera and J.K. Brimacombe, ISS Transactions, Vol.7, 1986, pp. 31-43. [7] E.A. Mizikar, 'Spray Cooling Investigation for Continuous Casting of Billets and Bloom,' Iron and Steel Engineer, Vol. 47, 1970, pp. 53-60. [8] M . Larrecq, J.P. Birat, C. Saguez, and J. Henry, 'Optimization of Casting and Cooling Conditions on Steel Continuous Caster,' 3rd Process Technology Conf, ISS-AIME, Pittsbrugh, PA, 1982, pp. 273-282. [9] J.K. Brimacombe and I.V. Samarasekera, 'Heat Transfer,' Continuous Casting, Vol. 2, ISS-AIME, pp. 1-8. [10] R.I.L. Futhrie, M . Hasan, et al, 'Turbulent Flow, Heat Transfer and Solidification in Thin-strip Casting Machines,' Second Canada - Japan Symposium on Modern Steelmaking and Casting Techniques, 1994, pp 201-220. [II] C. G. Kang, Y . D. Kim, and S. W. Lee, ' A Solidification and Cooling Roll Deformation Analysis Considering Thermal Flow in Twin Roll Strip Continuous Casting Process,' Melt Spinning, Strip Casting and Slab Casting, 1995, pp. 65-86. 131 REFERENCES [12] E.A. Mizikar, 'Mathematical Heat Transfer Model for Solidification of Continuously Cast Steel Slabs', Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 9-16. [13] J.K. Brimacombe, 'Design of Continuous-Casting Machines Based on a Heat Flow Analysis: State-of-the-Art Review,' Can. Metal. Quart., Vol. 15, No. 2 1976, pp. 163-175. [14] I.V. Samarasekera and J.K. Brimacombe, 'Thermal and Mechanical Behaviour of Continuous-Casting Billet Moulds,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 58-72. [15] I.V. Samarasekera and J.K. Brimacombe, 'The Thermal Field in Continuous Casting Moulds,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 45-58. [16] A . M . Ayoub and S.H. Hosam-Eddin, 'Mathematical Modeling of Heat Transfer in Continuous Casting of Steel Billet,' Advances in Continuous Casting Research and Technology, 1992, pp. 97-114. [17] K .C . Mills, P. Grieveson, A. Olusanya, et al, 'Effect of Casting Powder on Heat Transfer in Continuous Casting,' Continuous Casting '85, 1985, pp. 57.1-7. [18] H. Grothe, J.P. Radot, et al, 'Optimized Strand Cooling and Straightening,' Continuous Casting '85, 1985, pp. 68.1-9. [19] S. Chandra, I.V. Samarasekera, J.K. Brimacombe, L A . Bakshi, and B.N. Walker, 'On Line Monitoring of Mould-shell Interaction during Continuous Casting of Steel Billets,' Ironmaking and Steelmaking, Vol. 23, No. 6, 1996, pp. 512-521. [20] C . A . M . Pinheiro, I.V. Samarasekera, J.K. Brimacombe, and B .N . Walker, 'Mould heat transfer and continuously cast billet quality with mould flux lubrication Part A: Mould heat transfer,' Ironmaking and Steelmaking, Vol. 27, No. 1, 2000, pp. 37-53. [21] HansF. Schrewe, Continuous Casting of Steel, 1989, pp. 89-144. [22] C R . Taylor, Metal. Trans., 1975, 6B, pp. 359-375. [23] I.V. Samarasekera and J.K. Brimacombe, 'The Continuous-Casting Mould,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 33-44. [24] S.N. Singh and K.E. Blazek, 'Heat Transfer and Skin Formation in a Continuous Casting Mould as a Function of Steel Carbon Content,' Op. Hearth Proc, Vol. 57, 1974, Atlantic City, pp. 16-38. 132 REFERENCES [25] A. Grill and J.K. Brimacombe, 'Influence of carbon content on rate of heat extraction in the mould of a continuous casting,' Ironmaking and Steelmaking, Vol.3, No.2, 1976, pp.76-79. [26] I.V. Samarasekera, J.K. Brimacombe, R. Bommaraju, 'Mould Behavior and Solidification in the Continuous Casting of Steel Billets 2,' ISS Transactions, Vol.5, 1984, pp. 79-94. [27] I.V. Samarasekera, R. Bommaraju and J.K. Brimacombe, 'Factors Influencing the formation of Oscillation Marks in the Continuous Casting of Steel Billets,' Proceedings of International Conference on Continuous Casting of Steel Billets, 24 t h Conference of Metallurgists, Metallurgical Society of CLM, Vancouver, Canada, 1985, pp.33-58. [28] J.K. Brimacombe et al, 'Mould Behavior and Solidification in the Continuous Casting of Steel Billets. 1 - Industrial Trials,' ISS Transactions, Vol. 5, 1984, pp. 71-78. [29] R. Bommaraju, I.V. Samarasekera, J.K. Brimacombe, 'Mould Behavior of Steel Billets. 3 - Structure, Solidification Bands, Crack Formation and Off-squareness,' ISS Transactions, 1984, Vol. 5, pp. 95-104. [30] J.K. Brimacombe, I.V. Samerasekera, 'Future Trends in the Development of Continuous Casting Moulds,' A publication of the ISS, 1991, pp. 153-160. [31] M . Suzuki, S. Miyahara, T. Kitagawa, et al, 'Effect of Mould Oscillation Curves on Heat Transfer and Lubrication Behaviour in Mold at High Speed Continuous Casting of Steel Slabs,' Tetsu-to-Hagane, Vol. 78, No. 1, 1992, pp. 113-120. [32] HansF. Schrewe, Continuous Casting of Steel, 1987, pp. 52-54. [33] Lauri Holappa, et al, 'Optimization of the Secondary Cooling in the Continuous Casting of Steel Billets,' Proceedings International Symposium on the Continuous Casting of Steel Billets, Vancouver, British Columbia, 1985, pp. 242-259. [34] J. K. Brimacombe, et al, 'Spray Cooling in Continuous Casting,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 105-108. [35] J. K. Brimacombe, P.K. Agarwal, et al, 'Spray Cooling in Continuous Casting of steel,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp.109-123. [36] J. K. Brimacombe and K. Sorimachi, 'Crack Formation in the Continuous Casting of Steel,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 199-214 [37] G. Van Drunen, J. K. Brimacombe, et al, 'Internal Cracks in Strand Cast Billets,' Ironmaking and Steelmaking, Vol. 2, 1975, pp. 125-133. 133 REFERENCES [38] K. Sasaki, Y . Sugitani and M . Kawasaki, 'Heat Transfer in Spray Cooling on Hot Surface,' Tetsu-to-Hagane, Vol. 65, 1979, pp. 90-96. [39] M . Shimada and M . Mitsutsuka, 'On Heat Transfer Coefficient by Forced Water Cooling to Carbon Steel,' Tetsu-to-Hagane, Vol. 52, 1966, pp. 1643. [40] Zhang Keqiang and E.W. Langer, 'Influence of Secondary Cooling on Liquid Pool Depth and Surface Temperature of Strand in Continuous Casting of Steel,' Scand. J. Metallurgy, Vol. 21, No. 6, 1992, pp. 260-265. [41] A. Grill et al, Ironmaking and Steelmaking, No. 3, 1979, pp. 131-140. [42] M . Ei-Bealy, N . Leskinen, and H. Fredriksson, 'Simulation of Cooling Conditions in Secondary Cooling Zones in Continuous Casting Process,' Ironmaking and Steelmaking, Vol. 22, No. 3, 1995, pp. 246-255. [43] R. Bullard, 'Cooling Water System Helps Georgetown Steel Break Continuous Casting Record,' I&SM, August, 1994, pp.23-26. [44] Hirowo G. SUZUKE et al, Transactions of Iron and Steel Institute of Japan, Vol. 22 No. 1, 1982, pp.48-55. [45] L .A . Norstroem and B. Johansson, 'Hot Ductility - A Matter of Strain Rate,' Scand. J. Metall., Vol. 11 (3), pp. 139-142 [46] W T . Lankford,Met. Trans, Vol. 3, 1972, pp.1311-1357. [47] L.I. Morozenskii, O.A. Mitenev, and V.K. Krutikov, Stal, No. 4, 1965, pp. 272-276. [48] F. Weinberg, 'The Ductility of Continuously-Cast Steel Near the Melting Point -Hot Tearing,' Met. Trans. B, Vol. 10B, 1979, pp. 219-227. [49] F. Weinberg, 'The Strength and Ductility of Continuously cast Slabs Above 800 °C; Met. Trans. B, Vol. 10B, 1979, pp. 513-522. [50] C.J. Adams, Open Hearth Proc, Vol. 54, 1971, pp.290-302. [51] Y . Iida, H. Moriwaki, T. Uesa, and Y. Habu, Tetsu-to-Hagane, Vol. 59, 1973, pp. S89. [52] S. Kumar, 'Mould Thermal Response and Formation of Defects in the Continuous Casting of Steel Billets,' Ph.D. thesis, University of British Columbia, Vancouver, Canada, 1996. 134 REFERENCES [53] S. Kumar, I.V. Samarasekera and J.K. Brimacombe, 'Mould Thermal Response and Formation of Defects in the Continuous Casting of Steel Billets, Part 2: Rhomboidity,' I&SM, Dec. 1998, pp.3-18. [54] S. Chandra, J.K. Brimacombe and I.V. Samarasekera,' Mould Strand Interaction in the Continuous Casting of Steel Billets. Part III - Mould Heat Transfer and Taper in Steel Billet Casting Machines,' Ironmaking and Steelmaking, 1993, Vol.20, No.2, pp. 104-112. [55] S. Kumar, J.A. Meech, I.V. Samarasekera and J.K. Brimacombe, 'Knowledge Engineering an Expert System to Diagnose Quality Problems in the Continuous Casting of Steel Billets,' Iron and Steelmaker, 1993, Vol.20, No.9, pp.29-36. [56] J.L. Brendzy, I.A. Bakshi, I.V. Samarasekera, and J.K. Brimacombe, 'Mould-Strand Interaction in Continuous Casting of Steel Billets. Part 2 - Lubrication and Oscillation Mark Formation,' Ironmaking and Steelmaking, 1993, Vol.20, No.2, pp.63-74. [57] I.V. Samarasekera, D.L. Anderson and J.K. Brimacombe, 'The Thermal Distortion of Continuous-Casting Billet Moulds,' Metallurgical Transactions B, 1982, Vol . l3B, pp.91-104. [58] Carlos A . M . Pinheiro, 'Mould Thermal Response, Billet Surface Quality and Mould Flux Behavioor in Billet Casting with Powder Lubrication,' Ph.D Thesis, UBC, 1997. [59] S. Chandra, 'Heat Transfer, Oil Lubrication and Mould Taper in Steel Billets Casting Machines,' Ph.D Thesis, UBC, 1992. [60] J. Szekelyand V. Stanek, Metall. Trans., 1970, 1, pp. 119-126. [61] J.K. Brimacombe, E.B. Hawbolt and F. Weinberg, 'Formation of Off-Corner Internal Cracks in Continuously-Cast Billets,' Continuous Casting, Vol. 2, ISS-AIME, 1984, pp. 229-238. [62] J.K. Brimacombe and K. Sorimachi, Metall. Trans. B, 1977, pp.489. 135 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090040/manifest

Comment

Related Items