s t e e l u n i ve r s i t y.

o r g 
Continuous Casting Simulation, version 1.51. User Guide
1 Introduction and Disclaimer.............................................................................2
2 About this Version ..........................................................................................2
3 Introduction to Continuous Casting..................................................................3
4 Simulation Objectives .....................................................................................4
5 Plant Layout and Description ...........................................................................4
5.1 Dimensions of the Casting Machines .......................................................................... 4
6 Simulation Options .........................................................................................5
6.1 User Levels ....................................................................................................................5
6.1.1 University Student Level ..................................................................................5
6.1.2 Steel Industry Works Technical Level .............................................................5
6.2 Steel Grades................................................................................................................. 6
6.2.1 Crack Sensitive Grades ...................................................................................7
6.2.2 Sticker Sensitive Grades..................................................................................7
6.3 Soft Reduction Level .....................................................................................................7
6.4 Casting Speed and Secondary Cooling Rate.............................................................. 8
6.5 Mold Oscillation Settings .............................................................................................. 9
6.5.1 Settings ............................................................................................................ 9
6.5.2 Oscillation Marks ............................................................................................10
6.6 Mold powder ................................................................................................................ 11
6.6.1 Important Parameters..................................................................................... 11
6.7 Ladle Ordering ............................................................................................................. 13
6.7.1 Time ................................................................................................................ 13
6.7.2 Temperature ...................................................................................................14
6.7.3 Calculation of Liquidus Temperature.............................................................14
6.8 Review of Choices....................................................................................................... 15
7 Running the Simulation................................................................................. 15
7.1 Starting the Cast.......................................................................................................... 15
7.2 Ladle Change .............................................................................................................. 15
7.3 Steel Cleanness...........................................................................................................16
7.4 Strain Analysis Model for Slab Casting Machine .......................................................16
7.4.1 Estimation of Internal Cracking...................................................................... 17
7.4.2 Estimation of Surface Cracking .....................................................................18
7.5 Avoiding Breakout .......................................................................................................19
8 User Interface.............................................................................................. 20
8.1 Simulation Controls .................................................................................................... 20
8.1.1 Simulation Rate ............................................................................................. 20
8.1.2 Ladle Turret ................................................................................................... 20
8.1.3 Ladle .............................................................................................................. 20
8.1.4 Tundish .......................................................................................................... 20
8.1.5 Strand ............................................................................................................ 20
8.1.6 Change SEN (Works Technical Only) ...........................................................21
8.1.7 EMS (Only for Bloom and Billet Caster) ........................................................21
8.2 Casting Information .....................................................................................................21
8.2.1 View Event Log (Key E) .................................................................................21
8.2.2 View Flows (Key F) ........................................................................................21
8.2.3 Show/HIde Inner Rolls (KEY H).....................................................................21
8.2.4 View Level of Steel (Key L) ............................................................................21
8.2.5 View Quality (Key Q) ......................................................................................21
8.2.6 View Temperature (Key T) .............................................................................21
8.2.7 Close Casting Iinformation Dialog Box (Key X).............................................21
8.3 Simulation Results...................................................................................................... 22
9 References ..................................................................................................23
steeluniversity.org Continuous Casting User Manual

1 Introduction and Disclaimer

This document has been prepared as a user guide to the continuous casting simulation,
available at http://www.steeluniversity.org/. The interactive simulation has been
designed as an educational and training tool for both students of ferrous metallurgy and
for steel industry employees.
The information contained both in this document and within the associated website is
provided in good faith but no warranty, representation, statement or undertaking is given
either regarding such information or regarding any information in any other website
connected with this website through any hypertext or other links (including any warranty,
representation, statement or undertaking that any information or the use of any such
information either in this website or any other website complies with any local or national
laws or the requirements of any regulatory or statutory bodies) and warranty, representation,
statement or undertaking whatsoever that may be implied by statute, custom or otherwise is
hereby expressly excluded. The use of any information in this document is entirely at the risk
of the user. Under no circumstances shall the International Iron and Steel Institute, The
University of Liverpool or their partners be liable for any costs, losses, expenses or damages
(whether direct or indirect, consequential, special, economic or financial including any losses
of profits) whatsoever that may be incurred through the use of any information contained in
this document.
Nothing contained in this document shall be deemed to be either any advice of a technical or
financial nature to act or not to act in any way.

2 About this Version

This is the first public release of this simulation, intended for evaluation purposes. The full
version of the continuous casting simulation will be available online at the end of July 2005.

Changes since version 1.41

• The underlying model has been improved and the quality graph changed accordingly.
• Changed mold powder costs.
• Too low a mold powder consumption will now result in a breakout because excessive
friction between mold and strand causes the strand to stop.
• The underlying model is now updated when the EMS state is changed.
• The soft reduction level can now be changed during casting.
Changes since version 0.36
• SEN changes now take 15 seconds instead of 30 seconds.
• Soft reduction level is now chosen after casting speed, cooling rate and mold oscillation
settings.
• Choice of mold powder will now influence production cost and the chances for successful
casting.
• It is now possible to toggle between hidden or shown inner strand rolls.
Changes since version 0.23
• All four steel grades are now available.
• Both "Works Technical" and "University Student" user levels are available.
• Solidification of metal in a vessel can now occur.
• Ladle state of repair, nozzle clogging and delay of ladle arrival has been added for "Works
Technical" level.
Limitations
• All strands behave identically, which is of course not the case in real life.

© 2005 The University of Liverpool 2

steeluniversity.org Continuous Casting User Manual

• The tundish temperature at the start of cast is based on a simplified model to simulate
pre-heating of the tundish.
• Strand surface temperature and strand wall thickness has been pre-calculated using a
finite element model for the different combinations of casting speeds, cooling rates and
steel grades.
• Misaligned rolls can be adjusted during casting which is absolutely not the case in real
life.
• To optimize the overall performance of the simulation, all of the underlying model
calculations are executed sequentially. However, one implication is that the mold level
decreases as the simulation rate is increased and vice versa. Be sure to maintain a high
mold level before using a high simulation rate.

3 Introduction to Continuous Casting

Continuous casting of steel is a process in which liquid steel is continuously solidified into a
strand of metal. Depending on the dimensions of the strand, these semi-finished products are
called slabs, blooms or billets. The process was invented in the 1950s in an attempt to increase
the productivity of steel production. Previously only ingot casting was available which still has
its benefits and advantages but does not always meet the productivity demands. Since then,
continuous casting has been developed further to improve on yield, quality and cost efficiency.
Liquid steel is supplied to the continuous caster from the secondary steelmaking shop. The
ladle is usually delivered by crane and positioned into a ladle turret, which subsequently
rotates the ladle into the casting position. A slidegate in the bottom of the ladle is opened to
allow the liquid steel to flow via a protective shroud into a tundish, a vessel that acts as a
buffer between the ladle and the mold. As the tundish fills, stopper rods are raised in order to
allow the casting of steel into a set of water-cooled copper molds below the tundish.
Solidification begins at the mold walls and the steel is withdrawn from the mold by a dummy
bar. As it leaves the mold, the strand of steel requires a sufficiently thick solid shell to carry
the weight of the liquid steel that it contains, i.e. the ferrostatic pressure.
Throughout the entire casting process, the mold oscillates vertically in order to separate the
solidified steel from the copper mold. This separation is further enhanced by introducing a
mold powder into the mold.
The strand is withdrawn from the mold by a set of rolls which guide the steel through an arc
until the strand is horizontal. The rolls have to be positioned close enough together to avoid
bulging or breaking of the thin shell.
As the steel leaves the mold, it has only a thin solidified shell which needs further cooling to
complete the solidification process. This is achieved in the so-called secondary cooling zone,
in which a system of water sprays situated between the rolls are used to deliver a fine water
mist onto the steel surface. At this point, the steel, solidified shell and liquid center, is called
the strand.
After the strand has been straightened and has fully solidified, it is torch-cut to pre-
determined product lengths. These are either discharged to a storage area or to the hot rolling
mill.

Table 3-1. Summary of different components in the continuous casting process.

Component Primary Task Secondary Task
Ladle Transport and hold the liquid steel Facilitate inclusion removal
Position full ladles over the tundish Free the cranes for higher
Ladle Turret
and remove empty ones productivity
Act as a buffer between ladle and
Tundish Facilitate inclusion removal
mold
Mold Cool down the liquid steel to form a

© 2005 The University of Liverpool 3

steeluniversity.org Continuous Casting User Manual

solidified shell
Further cool the strand to fully
Strand System
solidified and straighten the strand

4 Simulation Objectives
The aim of the simulation is to successfully sequence cast three ladles meeting the specified
criteria of surface quality, internal quality and inclusion content.
You should also aim to minimize the cost of the whole operation.

5 Plant Layout and Description

Figure 5-1 Screenshot showing the plant layout used in the simulation. Two ladles are positioned in the
ladle turret which turns to position the ladle over the tundish.

The plant in the simulation is laid out as shown in Figure 5-1. At the start of the simulation,
one full ladle is positioned over the tundish.

5.1 Dimensions of the Casting Machines

The simulation includes three different casting machines that are used for casting four
different steel grades. These are slab, bloom and billet caster. Properties of the different
casting machines are listed in the table below.

© 2005 The University of Liverpool 4

steeluniversity.org Continuous Casting User Manual

Table 5-1 Table of casting machine properties

Type Slab Bloom Billet

Linepipe steel
Steel grade(s) Construction steel Engineering steel
ULC steel
Ladle size / metric ton 250 100 100
Radius / m 9 12 8
Number of strands 2 5 6
Casting speed / m min-1 1.0-2.0 1.2 - 1.8 3-5
Cross section dimension / mm 1200 × 230 250 × 250 130 × 130

Typical use flat products, i.e. plate, long products, i.e. long products, i.e.
sheet, coils bars, beams, bars, channels,
pilings wires
Roll spacing, section I / mm 202 (35 rolls in 45°)
Roll spacing, section II / mm 283 (25 rolls in 45°)
R56=9, R57=11.3, R58=15,
Radii at bending/straightening / m
R59=22.6, R60=45.2

6 Simulation Options
Before you start the simulation, it is important that you plan ahead. The first thing to do is to
choose a target casting speed that allows the steel to be cast in such a manner that all quality
criteria are met. Secondly, the mold oscillation settings are important to ensure a good
enough surface quality. Finally, the temperature of the liquid steel and the arrival of ladle two
and three need to be planned accordingly.
This section presents the key underlying scientific theories and relationships that are required
in order to successfully complete the simulation. In no way is it designed to be a
comprehensive treatment of continuous casting theory and practice – for this, the user is
directed to other excellent publications.

6.1 User Levels

The simulation has been developed for use by two different user groups:

• University students of metallurgy, materials science and other engineering

disciplines.

• Steel industry works technical.

6.1.1 UNIVERSITY STUDENT LEVEL

At this level the user will be expected to approach the problem scientifically, using the relevant
thermodynamic and kinetic theories to make decisions on the various processing options.
For example, the user will need to decide which combination of casting speed and secondary
cooling rate that will yield a good quality strand.
At this level there will be no operational problems to overcome and casting will be relatively
straightforward.

6.1.2 STEEL INDUSTRY WORKS TECHNICAL LEVEL

At this level, you will be expected to approach the problem scientifically. However, you may
also experience a range of operational problems that require you to make adjustments to your
planning and use your experience to make rapid decisions.
Typical examples of the operational problems you might encounter are delays to the time of
ladle arrival, nozzle clogging, and different states of repair for the ladles in use.

© 2005 The University of Liverpool 5

steeluniversity.org Continuous Casting User Manual

To simulate that ladles are being worn down there are three different states of repair; good,
acceptable and poor. The cooling rate of liquid steel in the ladle is affected by the state of
repair and the corresponding values are 0.50, 0.75 and 1.00 °C min-1.
Ladles may be delayed by up to 10 minutes. Remember to compare the ordered delivery time
with the estimated arrival time, which is shown in text above the ladle turret station after the
simulation starts.
You will also have to monitor the state of the submerged entry nozzles (SEN) to determine
when these need changing due to "nozzle clogging". Nozzle clogging is the progressive build-
up of inclusions in the nozzle during casting. It reduces the flow rate from the tundish into the
mold due to the reduced area of flow and eventually the SENs have to be changed. To
minimize the rate of nozzle clogging, you should maintain a high liquid steel level in the
tundish at all times (this gives inclusions time to float to the surface). The only way to detect
clogged nozzles is by monitoring the flow rate from the tundish. When it is no longer possible
to maintain the required flow rate for the target casting speed, you will need to change the
nozzles. This is achieved by:

• Decrease the casting speed

• Stop the flow from the tundish.

• Press the "Change SEN" button. This costs $200 per strand and it will take 15
seconds to change.

• Re-start the flow from the tundish to refill the mold.

• When the mold is refilled, return to the target casting speed

NOTE: It is essential to avoid emptying the mold during the nozzle

change. It may be necessary to further decrease the casting speed if
the mold level is running low.

6.2 Steel Grades

The simulation includes a number of different steel grades to illustrate a range of different
processing options.
The general-purpose construction steel grade is a crack sensitive relatively undemanding
steel grade that is recommended for the novice user. Construction steel is cast using a bloom
casting machine with the cross section dimensions being 250 × 250 mm. The inclusion level
can be moderate without suffering any quality problems.
The TiNb ultra-low carbon steel is a sticker sensitive steel grade used for automotive body
parts with a carbon specification of less than 0.0035 %C in order to optimize formability. This
steel is cast in a slab casting machine with the cross section dimension 1200 × 230 mm. To
meet the cleanness requirements of this steel grade it is of the utmost importance that the
inclusions levels are very low.
The linepipe steel for gas distribution is a very demanding grade as the combination of high
strength and high fracture toughness having extremely low levels of impurities (S, P, H, O and
N) and inclusions. Together with the ultra-low carbon steel grade, this steel has got the
highest demands on having very low inclusion levels and both steel grades are cast using the
slab casting machine with a cross section of 1200 × 230 mm. Depending on composition, this
grade can be either crack sensitive (peritectic) or sticker sensitive (hypo-peritectic).
The engineering steel is a heat-treatable low alloy grade, which is cast at high speeds in a
billet caster using cross section of 130 × 130 mm.

© 2005 The University of Liverpool 6

steeluniversity.org Continuous Casting User Manual

6.2.1 CRACK SENSITIVE GRADES

Steel grades in continuous casting are divided into two subgroups: cracking and sticker
sensitive grades.
Cracking (longitudinal cracks) is a serious problem in medium carbon steels (0.06 – 0.18 %C).
There is a 4 % mismatch between the thermal shrinkage coefficients for δ-ferrite and
austenite. This results in stress in the shell and stress release comes through longitudinal
cracking of the steel shell. The usually adopted strategy involves the reduction of the stresses
by keeping the thickness of the shell to a minimum. This is achieved by reducing the
horizontal heat transfer by increasing the thickness of the solid layer and the crystallinity of
the solid slag layer.

6.2.2 STICKER SENSITIVE GRADES

In contrast, sticker breakouts occur when the shell is not strong enough to withstand the
ferrostatic pressure and the liquid steel pours out of the shell. The strategy adopted here is to
build a thicker shell and this is achieved by increasing the horizontal heat flux by decreasing
the thickness and increasing the glassy fraction of the solid slag layer.

Table 6-1 Table of compositions for steel grades available in the simulation.
Construction steel TiNb ULC steel for Linepipe steel Engineering steel
car bodies
C 0.1450 0.0030 0.0700 0.4150
Si 0.2000 0.2100 0.1800 0.4000
Mn 1.4000 0.7500 1.0500 0.7500
P <0.0250 0.0650 <0.0120 0.0350
S <0.0200 <0.0120 <0.0030 0.0350
Cr <0.1000 <0.0500 <0.0600 1.0500
Al 0.0350 0.0450 0.0300 0.0225
B <0.0005 0.0030 <0.0050 0.0050
Ni <0.1500 <0.0800 <0.0500 0.3000
Nb 0.0500 0.0200 0.0150 0.0000
Ti <0.0100 0.0300 <0.0100 0.0000
V <0.0100 - <0.0100 0.0100
Mo <0.0400 <0.0100 <0.0100 0.2250
As - <0.0010 - 0.0000
Ca - - <0.0050 0.0000
N <0.0050 <0.0040 <0.0045 0.0050
H <0.0005 <0.0005 <0.0002 0.0002
O <0.0010 <0.0005 <0.0007 0.0005

6.3 Soft Reduction Level

Soft reduction is only available during slab casting and it is used for reducing the severity of
center segregation. For soft reduction to have any effect, casting speed and secondary cooling
rate must be chosen so that the metallurgical length, i.e. the length to which steel is still liquid
in the center, is in the soft reduction zone.
A drawing of the soft reduction zone is shown in Figure 6-1. The soft reduction takes place
over two segments each with a length of 2 meters. Each segment consists of 5 rolls and the roll
pitch is about 400 mm. The reduction zone is positioned at a strand length between 27 and 31
meters when casting Linepipe steel and between 21 to 24 meters when casting Ultra-low
carbon steel.

© 2005 The University of Liverpool 7

steeluniversity.org Continuous Casting User Manual

Metallurgical length
Length of soft reduction Amount of
Roll pitch reduction

Figure 6-1 Graphical representation of the soft reduction zone

In the simulation, three different amounts of reduction can be selected. These are soft,
medium and hard with corresponding strand thickness reduction of 2.4 mm, 6.0 mm and
10.8 mm, respectively. The same conditions apply to both ultra-low carbon steel (ULC) and
line-pipe steel (LPS). Since the position of the soft reduction cannot be changed, there are
only a few combinations of casting speed and secondary cooling rate which lead to an optimal
condition of soft reduction.
This option is only available when casting a steel grade in the slab caster.

6.4 Casting Speed and Secondary Cooling Rate

Choosing the right combination of casting speed and secondary cooling rate is of the utmost
importance. This choice will influence many different parameters during casting and is one of
the key choices for getting a good quality cast. One parameter that is directly influenced by
this choice is the metallurgical length, the distance from the mold at which the strand
becomes totally solid.

Liquid Metallurgical length

(measured along center line)

Solid

Figure 6-2 Diagram illustrating metallurgical length

The metallurgical length is a complex function of steel composition, casting speed, cooling
rate and strand dimensions, the calculation of which is beyond the scope of this simulation. To
help you make informed decisions about your casting parameters, the tables below are
provided.
The different types of casting machines have different possible casting speeds and cooling
rates, see below for tables of metallurgical lengths depending on these parameters.

© 2005 The University of Liverpool 8

steeluniversity.org Continuous Casting User Manual

Table 6-2 Metallurgical lengths for construction steel cast in the bloom caster, 250 × 250 mm.
Cooling Rate / Casting Speed / m min-1
kg water per kg steel 1.2 1.4 1.6 1.8
0.3 22.48 26.55 30.43 34.65
0.4 21.78 25.57 29.10 33.12
0.5 20.96 24.43 27.55 31.26
0.6 20.04 23.17 25.57 29.22

Table 6-3 Metallurgical lengths for ultra-low carbon steel cast in the slab caster, 1200 × 230 mm.
Cooling Rate / Casting Speed / m min-1
kg water per kg steel 1.0 1.2 1.4 1.6 1.8 2.0
0.4 19.03 23.06 27.23 31.55 36.06 40.73
0.5 18.30 22.16 26.16 30.30 34.62 39.10
0.6 17.67 21.38 25.23 29.22 33.36 37.70
0.7 17.11 20.70 24.43 28.30 32.28 36.47
0.8 16.63 20.10 23.70 27.46 31.35 35.40

Table 6-4 Metallurgical lengths for linepipe steel cast in the slab caster, 1200 × 230 mm.
Cooling Rate / Casting Speed / m min-1
kg water per kg steel 1.0 1.2 1.4 1.6 1.8 2.0
0.4 20.17 24.50 28.98 33.65 38.55 43.56
0.5 19.40 23.56 27.86 32.35 37.02 41.87
0.6 18.75 22.74 26.88 31.20 35.70 40.37
0.7 18.17 22.02 26.04 30.21 34.56 39.10
0.8 17.65 21.40 25.30 29.33 33.57 37.97

Table 6-5 Metallurgical lengths for engineering steel cast in the billet caster, 130 × 130 mm.
Cooling Rate / Casting Speed / m min-1
kg water per kg steel 3.0 4.0 5.0
0.8 17.20 22.40 28.00
0.9 16.70 21.53 26.83
1.0 16.20 20.73 25.66
1.1 15.70 19.86 24.42
1.2 15.20 19.06 23.33

6.5 Mold Oscillation Settings

An oscillating mold is used primarily to reduce the friction between the mold plate and the
strand shell. This is facilitated by the induced flow of mold powder from the meniscus down
the gap between the strand shell and the mold plates.

6.5.1 SETTINGS
Stroke, S [mm]: Normally, the stroke ranges between 3 and 10 mm. By increasing the
stroke, the negative strip time (see below) increases proportionally. Hence, the depth of
oscillation marks and the consumption of mold powder also increase.
Frequency, f [min-1]: Customary hydraulic mold oscillators realize frequencies between
100 and 250 cycles per minute. By increasing the frequency, the negative strip time decreases,
hence, the depth of oscillation marks and mold powder consumption decrease as well.
Negative strip time, tN [s]: The negative strip time is the period where the downward
velocity of the mold is higher than the casting speed, as given by:

60 1000 vcast
tN = arccos [s] 6-1
π f π fS

© 2005 The University of Liverpool 9

steeluniversity.org Continuous Casting User Manual

where:

f = frequency, min-1
S = stroke, mm
vcast = casting speed, m min-1

Oscillation mark depth, d [mm]: While oscillating the mold is a necessity for continuous
casting it also decreases surface quality due to so called oscillation marks. The surface of
continuous castings is characterized by the presence of oscillation marks that form
periodically at the meniscus due to mold reciprocation. They have an important influence on
the surface quality because they are often the source for transverse cracks.
Oscillation mark depth depends on the chosen mold powder, oscillation stroke, oscillation
frequency and casting speed. A regression of values from literature [1] yields:

d = 0.065 ⋅ 1.145 S ⋅ (200 ⋅ 0.9 S ) t N [mm] 6-2

where:

tN = negative strip time, s

6.5.2 OSCILLATION MARKS

Figure 6-3 shows the formation mechanism for oscillation marks. The top of the figure shows
the mold position varying with time. The formation mechanism of oscillation marks is
outlined in the bottom part of the figure. The negative strip time (hatched areas) is the main
influencing factor for the formation of oscillation marks. Increasing negative strip time is
accompanied with increasing depth of oscillation marks.

Figure 6-3 Formation of oscillation marks [2]

© 2005 The University of Liverpool 10

steeluniversity.org Continuous Casting User Manual

In order to minimize the depth of the oscillation marks it is essential to properly optimize the
oscillation settings. The negative strip time should be as close to 0.11 s as possible combined
with a stroke that results in the smallest possible oscillation mark depth.
Note that the maximum acceptable oscillation mark depth is 0.25 mm for ultra-low carbon
steel while a depth of 0.60 mm is acceptable for the other steel grades.

6.6 Mold powder

Mold powder is a synthetic slag which is continuously fed onto the liquid pool surface during
casting. The powder melts and flows down between the mold walls and the strand shell.
Choosing the right mold powder is a critical choice to ensure a good enough surface quality of
the cast material. The chosen powder primarily influences oscillation mark depth and mold
powder consumption.
The function of casting powders is to:

• Act as a lubricant between strand and mold

• Improve heat transfer from strand to mold

• Provide thermal insulation of the top surface of the molten pool

• Protect liquid steel against reoxidation

• Absorb inclusions that rise to the metal surface

Figure 6-4 shows the general disposition of a powder in the continuous casting mold. Mold
powder is added to the top of the liquid steel in the mold. The powder melts and infiltrates the
mold/strand gap at the meniscus. This infiltration is the key process in continuous casting
because it is necessary to ensure both good lubrication and a uniform heat transfer between
the strand and the mold.

Figure 6-4 Function of mold powder [3]

6.6.1 IMPORTANT PARAMETERS

Mold powder consumption depends not only on the chosen type of mold powder but also
on the oscillation settings and casting speed. The consumption is measured in mass per unit
area of strand surface, e.g. kg m-2. Since the molten mold powder is pumped by the oscillating

© 2005 The University of Liverpool 11

steeluniversity.org Continuous Casting User Manual

movement of the mold into the mold/strand gap the oscillation settings have an essential
influence on the mold powder consumption.
There is a great variation within literature of fitted relations for mold powder consumption. In
this simulation the following expression is used to calculate mold powder consumption:

1.7 × t N
Q= [kg/m²] 6-3
η × vcast

where

Q = mold powder consumption per unit area, kg m-2

tN = negative strip time, s
η = mold powder viscosity, Pa s
vcast = casting speed, m min-1

Too low a mold powder consumption rate will cause sticking between the strand and the
mold, eventually resulting in a breakout. To avoid this, the powder consumption rate should
be above 0.30 kg m-2 except for engineering steel where 0.15 kg m-2 is adequate.
One of the most important properties of a mold powder is break temperature. It is defined
as the threshold temperature at which the powder's viscosity increases dramatically, i.e. the
point where liquid lubrication starts to break down.
Figure 6-5 shows how the break temperature varies with different casting speeds. A crack
sensitive grade should be cast using casting powder A or B to provide as good conditions as
possible, while sticker sensitive grades should be cast using powder type C or D.
1 250

Crack sensitiv e grades

1 200
A B v c <1 m min-1
Break Temperature / °C

1 1 50
v c =1 -1 .40 m min-1
C
1 1 00

v c >1 .40 m min-1

1 050
E D
Sticker sensitiv e grades
1 000

950
0 0.05 0.1 0.1 5 0.2 0.25 0.3

V iscosity / Pas (at 1 300 °C)

Figure 6-5 Break temperature and viscosity of the mold powder in relation to casting speed [4]

Table 6-6 contains material property data and costs for the five different types of mold
powders that can be used.

© 2005 The University of Liverpool 12

steeluniversity.org Continuous Casting User Manual

1250

Crack sensitive grades

1200
A B vc<1 m min-1
Break Temperature / °C

1150
vc=1-1.40 m min-1
C
1100

vc>1.40 m min-1
1050
E D
Sticker sensitive grades
1000

950
0 0.05 0.1 0.15 0.2 0.25 0.3

Viscosity / Pas (at 1300 °C)

Figure 6-5 Break temperature and viscosity of the mold powder in relation to casting speed [4]

Table 6-6 Material properties of available mold powders.

Powder Viscosity Break temperature Cost Purpose
/ Pa s / °C / $ per kg
A 0.12 1170 0.40 Used for crack sensitive
B 0.21 1190 0.35 grades
C 0.19 1130 0.45 Used for sticker sensitive
D 0.10 1050 0.50 grades
E 0.03 1050 0.55 Used for high speed casting

6.7 Ladle Ordering

The objective of the simulation is to sequence cast three ladles. The first ladle is in place over
the tundish when the simulation begins, but the other two will arrive at a later point. You can
choose the arrival temperature for all three ladles and the estimated arrival time for the last
two ladles. This gives ample opportunity to optimize time-temperature control to achieve the
right casting conditions in the mold.
Estimated arrival time is input as the number of minutes passed after the simulation
begins and the estimated arrival temperature is input as °C.
Remember that the steel loses temperature over time due to heat losses. It is assumed that the
temperature loss for the ladle is 0.5 °C min-1.

6.7.1 TIME
The time it takes to empty a ladle determines how long you should allow between ladles, e.g.
adjust the arrival time of ladle 2 so that ladle 1 is emptied just before or after ladle 2 arrives.
The emptying time depends on cross sectional area of the mold/strand, the number of strands
per tundish and casting speed.
The volume of material cast per strand per minute is given by:

V& = w ⋅ t ⋅ vc [m3 min-1] 6-4

© 2005 The University of Liverpool 13

steeluniversity.org Continuous Casting User Manual

where:

w = strand width, m
t = thickness of the strand, m
vc = casting speed, m min-1

Therefore the mass of material per minute for the tundish is given by:

M& T = n ⋅ ρ liq ⋅ w ⋅ t ⋅ vc [kg min-1] 6-5

where:

n = number of strands
ρliq = liquid steel density, 7400 kg m-3

Under steady state casting conditions (i.e. constant vc) the time to drain a ladle to a given level
of steel will be given by:

mladle mladle
τ= = [min] 6-6
M& T n ⋅ ρ liq ⋅ w ⋅ t ⋅ vc

where:
mladle = mass of liquid steel to be teemed from the ladle, kg. Note that teeming
automatically stops when slag is detected at the slidegate, typically when
the steel level reaches 5 %.

Example
You are casting a linepipe steel using a 1.5 by 0.2 m cross section twin strand slab casting
machine. The casting speed is 1.8 m min-1 and the caster is supplied via 200 tonne ladles.
Calculate the time to teem a ladle at steady state assuming that teeming stops at a level of
5 %.

200 000 × 0.95

τ= = 23.8 [min]
2 × 7400 × 1.5 × 0.2 × 1.8

6.7.2 TEMPERATURE
In order that the steel has the optimum temperature in the mold, it is important that the
ladles are ordered with the correct temperature. For University Student level the liquid steel
in the ladle cools at 0.5 °C min-1 but for the Works Technical level the cooling rate depends on
the state of the ladle and will vary between 0.5 and 1.0 °C min-1.
By carefully calculating the overall time from start of simulation to the time when the ladle is
emptied, the temperature loss is possible to compute. Subsequently, the necessary steel
temperature at arrival can be calculated.

6.7.3 CALCULATION OF LIQUIDUS TEMPERATURE

It is imperative to prevent the liquid steel temperature falling below the liquidus temperature
(i.e. the temperature at which the steel begins to solidify). The liquidus temperature, Tliq, is
dependent on composition and can be calculated from the following equation [5]:

© 2005 The University of Liverpool 14

steeluniversity.org Continuous Casting User Manual

Tliq = 1537 - 78%C - 7.6%Si - 4.9%Mn - 34.4%P – 38%S

In practise it is necessary to keep the steel temperature slightly higher than the liquidus
temperature, due to temperature variations within the steel (i.e. edges and corners tend to
have a lower temperature). The difference between liquidus temperature and actual
temperature is called superheat. Make sure that the superheat always is above 10 °C to avoid
freezing.
Increasing superheat reduces the shell thickness of the strand due to the extra heat energy
that has to be extracted by the mold. If the shell at any part of the strand is too thin to support
its own weight there will be a breakout. Maximum superheat is 50 °C for slab casting
machines and 60 °C for bloom and billet casters.

6.8 Review of Choices

The last screen before the simulation starts allows you to review the choices that you have
made. After pressing 'next', the simulation starts and you cannot go back and change these
choices without restarting the simulation.

7 Running the Simulation

Having selected the different settings for your continuous casting operation it is time to start
casting. The aim is to control the flow of liquid metal from ladle to tundish to mold so that the
selected casting speed can be maintained and a good quality casting achieved. You will also
need to exchange ladles, check the rolls for any excessive misalignment and cut the strand into
semi-finished products.

7.1 Starting the Cast

The first step is to start teeming the ladle. Open the slidegate to increase the flow rate of steel
from the ladle to the tundish. This can either be done by clicking on the up or down arrow or
by directly entering a number in the flow rate controller labeled "Ladle flowrate".
After reaching a sufficient buffer level of steel in the tundish, raise the stopper rods to increase
the flow from the tundish to the mold. This can be done by done either clicking on or entering
a number in the flow rate controller labeled "Tundish flowrate".
Wait until the mold level is high enough (preferably over 70 %), then start casting by choosing
a relevant casting speed.
You will need to balance the flow between ladle, tundish, and molds to ensure that the levels
are sufficiently high at all times. Typically you should aim to maintain an 80-90 % level in
both tundish (see Section 7.3) and molds (to avoid breakout, see Section 7.5). Clearly however
you do not want to overfill either of the vessels.

7.2 Ladle Change

Subsequent ladles are automatically lowered into the turret. At works technical level there
may be delays in delivery of up to 10 minutes so be prepared to counteract this.
Make sure to stop the flow from the ladle before attempting to rotate the new ladle in place
over the tundish.
The level in the tundish will inevitably decrease while the ladles are being exchanged, so make
sure you have a sufficiently high level in the tundish beforehand. You will need to start
teeming the new ladle at a higher rate in order to replenish the tundish to its target level after
a completed ladle exchange.

© 2005 The University of Liverpool 15

steeluniversity.org Continuous Casting User Manual

7.3 Steel Cleanness

Certain applications, such as linepipes for oil and gas distribution require very 'clean' steels –
i.e. with very low levels of oxide and sulfide inclusions, since these can act as crack initiation
sites. The chemistry of oxide and sulfide formation and subsequent removal during secondary
steelmaking is extremely complex and the subject of ongoing research. For more
comprehensive information it is kindly suggested that articles and books about these subjects
are consulted.
In the simulation you will aim to achieve 'moderate', 'low' or 'very low' levels of inclusions
depending on the chosen steel grade. Various factors influence the end-level of inclusions. The
level of inclusions at arrival to the casting station is assumed to be appropriate to the level that
is needed in order to achieve a clean enough steel. For example, if you are casting engineering
steel the inclusion level does not need to be very low, it is sufficient for the purpose to have a
low level of inclusions so the cleanness of the steel in the provided ladle will be low from the
start. This does not however mean that the casting will automatically succeed.
The inclusion level of the steel can be kept at its present level or even be slightly lowered by
using the tundish as a buffer. This allows for removal of inclusions to the walls of the tundish
and to the slag layer on top of the liquid steel. Thus, having a long residence time in the
tundish is very important for casting as clean steel as possible.

7.4 Strain Analysis Model for Slab Casting Machine

For the ULC steel and the linepipe steel, a uniform casting machine is assumed. Figure 7-1
shows a schematic drawing of the slab caster. The strand guide is curved from the mold all the
way down to the end of the straightening section. The curvature is divided into two zones with
35 and 25 rolls, respectively.

roll-pitch l1
H(i)

202mm
αi = i*45/35

Radius of curvature: R=9000mm

i-te roll
roll-pitch l2=283mm
Zone I (45°):
35 rolls

Zone II (45°):
25 rolls

Figure 7-1 Schematic drawing of the slab caster

In the following section the theoretical background on internal cracking and surface cracking
will be given together with the working equations from which the simulation calculates these
phenomena.

© 2005 The University of Liverpool 16

steeluniversity.org Continuous Casting User Manual

7.4.1 ESTIMATION OF INTERNAL CRACKING

To estimate the possibility of internal cracking the strain on the solidification front is
compared with a critical strain. Therefore, the strain on the solidifying front caused by the
process on each roll can be calculated as follows [6-8].
The tensile strains at the solidifying front caused by bulging, bending, straightening and
misalignment of supporting rolls are calculated using the following empirical equations. The
strain caused by bending and straightening is given by:

⎛d ⎞ 1 1
ε BS = 100 ⋅ ⎜ − S⎟⋅ − 7-1
⎝2 ⎠ Rn −1 Rn

where:

d = slab thickness, mm
S = shell thickness, mm
Rn-1and Rn = radii of roll number n-1 and n, mm

Both the bending and straightening take place with a multi point (five-point) method. Figure
7-2 shows the five point straightening method with the assumed radii. The bending method is
the same as the straightening with identical radii.

R56 = 9000 mm

R57 = 11300 mm
R58 = 15000 mm
R59 = 22600 mm
R60 = 45200 mm

Figure 7-2 Five point straightening method

To calculate the bulging strain εB (%), a typical empirical formula can be used:

0.101972 ⋅ P ⋅ l 3
εB = ⋅ 100 7-2
3800 ⋅ S 3

where:

S = solidifying shell thickness, mm

P = the static pressure of liquid steel, N mm-2
l = the roll pitch, mm

The strain due to roll misalignment εM (%) can be evaluated from following equation:

© 2005 The University of Liverpool 17

steeluniversity.org Continuous Casting User Manual

3⋅ S ⋅δM
ε M = 1.15 ⋅ ⋅ 100 7-3
l2

where:

δM = the roll misalignment amount, mm

Finally, the total strain at the solidifying front εintern during continuous casting of slab is
considered to be given by a sum of strains caused by bending/straightening, bulging and roll
misalignment as:

ε intern = ε BS + ε B + ε M 7-4

When the total strain exceeds the critical strain, internal cracks will be formed. The critical
strain depends on the steel composition and the strain rate [8]. The construction and
engineering steel grades can withstand a critical strain of about 1%, while linepipe and ultra-
low carbon steels have a critical strain of roughly 2%.
Internal crack formation will be a limiting factor in increasing the casting speed. The demands
on LPS and ULC steel are such that failing internal quality results in a downgrading of the
slab.

7.4.2 ESTIMATION OF SURFACE CRACKING

To estimate the surface quality in slab casting of the ultra-low carbon and linepipe steel, it is
assumed that only transverse cracking can occur [9]. The surface strain εsurf arising during
continuous casting is considered to be given by a sum of strains caused by
bending/straightening εBS, roll misalignment εM, bulging of solidifying shell εB and thermal
contraction εth:

ε surf = ε BS + ε M + ε B + ε th 7-5

The strain on the surface caused by bending/straightening can be approximated by:

d 1 1
ε BS = 100 ⋅ ⋅ − 7-6
2 Rn −1 Rn

where:

d = slab thickness, m
R = strand radius, m
N = roll number

As already described (Figure 7-2) the bending and straightening takes place with a five point
method. The strain due to the roll misalignment can be evaluated from the change of radius
caused by the deviation from the original position of any roll as:

d 1 1
ε M = 100 ⋅ ⋅ − 7-7
2 R0 R d

© 2005 The University of Liverpool 18

steeluniversity.org Continuous Casting User Manual

where:

R0 = radius of original position, mm

Rd = deviated position of the roll, mm

The surface strain due to bulging of solidifying shell is assumed to be equal to the strain at the
solidifying front due to bulging and therefore, can be calculated with equations (7-2) and (7-
3). The thermal strain is calculated as a product of thermal expansion coefficient α and
temperature difference ∆T:

ε th = α ⋅ ∆T ⋅ 100 7-8

To calculate the surface strain due to bulging it is assumed that the same equation to calculate
the strain at the solidification front (Equation 7-2) can be used.

A volume element on the surface of the strand travels through the total continuous casting
process and therefore, a total accumulated surface strain ε surf
tot
must be calculated, where n is
the number of rolls.

n
ε surf
tot
= ∑ ε surf (i ) 7-9
i

If the total surface strain exceeds the critical strain, surface cracks will be formed. The critical
strain depends on steel composition, depth of oscillation marks and surface temperature in
the straightening zone [9]. Nb-alloyed steels (linepipe and construction steel) are very prone
to surface crack formation. The surface temperature should not fall below 1050 °C for linepipe
steel and not below 1100 °C for the construction steel. Furthermore, oscillation marks should
be as shallow as possible. A depth of below 0.2 mm will result in higher critical strain. Slabs
and blooms with surface cracks have to be scarfed. The cost for scarfing amounts to 3 % of the
production costs. ULC slabs with surface cracks will be scarfed and downgraded because of
the high surface finish requirements.

7.5 Avoiding Breakout

A breakout will occur if the ferrostatic pressure exceeds the strength of the strand shell. It is
avoided by ensuring that the shell thickness at any given point in the strand is sufficient to
carry the weight of the liquid steel above this point.
Since the probability of breakout increases with a decreasing shell thickness, it is very
important to maintain a high steel level in the mold and to have as low superheat as possible.
This allows the shell to solidify to an adequate thickness before the steel leaves the mold.
Given that a thin shell might break under the pressure of the liquid steel, oscillation marks
should be kept as shallow as possible.
Depending on its precise composition, the linepipe steel grade can either be peritectic or
hypo-peritectic. Peritectic compositions are very crack sensitive, and so require more careful
casting to avoid cracks forming and propagating through the shell to cause breakout.
The following equation for carbon equivalent can be used to determine if the steel is peritectic
or not. Peritectic steel will have a carbon equivalent over 0.1 wt-%C [10].

C eq = %C - 0.1%Si + 0.04%Mn - 0.04%Cr + 0.1%Ni - 0.1%Mo 7-10

© 2005 The University of Liverpool 19

steeluniversity.org Continuous Casting User Manual

Using the correct mold powder for the selected steel grade is also very important. Selecting a
mold powder that is optimized for the wrong type of steel increases the probability of a
breakout. Also, the mold powder level must at all times be kept sufficiently high to maintain a
full lubrication between strand and mold.
In summary, avoid breakout by having:

• a mold level between 80 and 90 %

• a low superheat

• shallow oscillation marks

• the correct mold powder

8 User Interface
This section describes the basic 'mechanics' of running the simulation, e.g. how to move the
ladle, how to make alloy additions, how to control the various pieces of equipment, etc.

8.1 Simulation Controls

8.1.1 SIMULATION RATE
The simulation can be run at a range of different speeds between ×1 and ×32. The rate can be
changed at any time during the simulation. Highest recommended speed is ×16. This can be
done by clicking on the arrows on the so called numeric stepper to increase or decrease the
simulation rate with an increment of 1. Another way is to double-click inside the field with the
current simulation rate to select the current number, delete it, enter the new simulation rate
and then press "Enter".

8.1.2 LADLE TURRET

The ladle turret can be rotated by pressing the button labeled "Rotate". The turret cannot be
rotated as long as a ladle is missing in the turret or if the turret is currently rotating. Neither
can it be rotated if the ladle slidegate is open. If you are having difficulties to get the ladle
turret to rotate, please make sure that these three criteria are being met.

8.1.3 LADLE
Flow rate from the ladle to the tundish is controlled by using the numeric stepper labeled
"Ladle flowrate". The stepper is operated like the simulation rate controller. The flow rate
from the ladle is controlled to a precision of 100 kg min-1.

8.1.4 TUNDISH
The liquid steel flow from the tundish is controlled by using the numeric stepper labeled
"Tundish flowrate". The stepper is operated like the simulation rate controller. The flow rate
from the tundish to the mold is controlled to a precision of 25 kg min-1.

8.1.5 STRAND
Casting speed is controlled by selecting one of the choices listed in the "Casting Speed" drop-
down box. The choices pre-fixed with '*' are used for starting the cast. Please note that it is
only after choosing a valid casting speed, i.e. a casting speed that is not marked with '*', that
the cast strand can meet any of the quality criteria.
Misaligned rolls change color to indicate which roll pair that has become misaligned. To view
how big the misalignment is and how much it would cost to repair this, move the mouse over
the colored rolls. To align the rolls, i.e. repair the roll pair, simply click on the colored roll. The
repair cost will automatically be added to your total operational cost.

© 2005 The University of Liverpool 20

steeluniversity.org Continuous Casting User Manual

NOTE: In reality, misaligned rolls are repaired between castings,

not during.

8.1.6 CHANGE SEN (WORKS TECHNICAL ONLY)

SENs can be changed by pressing the button "Change SEN". SENs can only be changed when
the flow from the tundish has been stopped. It takes 15 seconds to change the SENs and it also
adds $200 per strand to the total cost.

8.1.7 EMS (ONLY FOR BLOOM AND BILLET CASTER)

Electro-magnetic stirring (EMS) can be used while using the bloom or billet caster. Using the
EMS decreases the segregation in the strand, thus improving the internal quality. If you are
having difficulties meeting the segregation quality requirement, try casting with EMS on.
Clicking on the button "EMS" will turn EMS on or off depending on the current state. The
button border is highlighted when EMS is on.

8.1.8 SOFT REDUCTION (ONLY FOR SLAB CASTER)

The amount of soft reduction used during casting can be changed by using the "Soft
reduction" drop-down box located next to the soft reduction zone marked as dark red. It is
possible to choose between off, low, medium and high.

8.2 Casting Information

It is possible to view detailed information about the casting during and after the simulation.
The following views are available by pressing the relevant key.

8.2.1 VIEW EVENT LOG (KEY E)

The event log keeps a chronological record of all the major events including some simulation
settings. This is useful for keeping track of what you have done and what has occurred so far
during the simulation. It is also very useful in helping you analyze your results at the end of
the simulation, as the log will often contain clues as to why you passed or failed the different
criteria.

8.2.2 VIEW FLOWS (KEY F)

Pressing 'F' shows a graph of the liquid steel flows from ladle to tundish and from tundish to
mold.

8.2.3 SHOW/HIDE INNER ROLLS (KEY H)

Pressing 'H' will toggle between shown and hidden inner strand rolls. This is most useful for
being able to see the whole strand whilst casting ULC or LPS steel in the slab caster.

8.2.4 VIEW LEVEL OF STEEL (KEY L)

Pressing 'L' shows how the level of liquid steel has changed in the ladle and in the tundish.

8.2.5 VIEW QUALITY (KEY Q)

Pressing 'Q' shows a graphical representation of the strand such as it has been cast. Good and
bad areas are marked and key figures about the strand are also displayed. This choice is only
available after completing the simulation.

8.2.6 VIEW TEMPERATURE (KEY T)

Pressing 'T' shows the variation of temperature over time in the ladle and in the tundish. This
choice is only available after completing the simulation.

8.2.7 CLOSE CASTING IINFORMATION DIALOG BOX (KEY X)

Pressing 'X' will close the casting information dialogue box.

© 2005 The University of Liverpool 21

steeluniversity.org Continuous Casting User Manual

8.3 Simulation Results

When the last steel has been cast and the strand is finished, the simulation will end and the
results of the casting operation will be displayed. Four key figures are shown immediately and
then you also have the possibility to further investigate the success or failure of the casting by
looking in further detail at one of the five detailed views. The key figures include:
Total Length of the Cast, expressed in meters.
Length Meeting Quality Criteria, expressed in both meters and in %.
Total Operating Cost, expressed in $, which includes the hourly operating cost and
additions for repairing misaligned rolls, taking temperature measurements, etc.
Cost per Metric Ton, which is the total operating cost divided by mass of cast steel meeting
the quality criteria.
The detailed views include information about temperature, level and flow variations in the
ladle and tundish as well as the event log and the quality log. These views are intended to help
in analyzing the casting operation in order to find where problems might come from and give
ideas on where casting conditions might be improved upon.
The quality log show quality variations in the cast strand. The first cast material is therefore
at x = 0 in the diagram and the last cast material is at the right hand edge. There are five
categories of failure for the strand:

• Internal Cracking

• Surface Cracking

• Center Segregation

• Inclusion Content

• Oscillation Marks
The diagram shows the strand quality using these five criteria. Shaded areas indicate where
quality requirements have not been met. Such defects have different impact for different
steels. Some surface defects can be removed by scarfing, other defects cause downgrading of
the steel and in the worst case the whole length will be scrapped. Lengths without any defects
are always adequate for their intended purpose. See Table 8-1 for which countermeasures that
will be taken depending on which defects that are present in the cast steel.

Table 8-1 Quality defects and their countermeasures.

Steel Scarfe Downgrade Scrap

CON Surface cracking Any two defects More than two defects

Surface cracking or
ULC Any defect More than two defects
Oscillation marks
Internal cracking or
Surface cracking or
LPS Center segregation or More than two defects
Oscillation marks
Inclusion content

ENG - Any two defects More than two defects

Downgrading a cut length will reduce the profit by 20 %, while scrapping reduces profit by
80 %. The cost for scarfing a cut length is about 2.5 % of the cost for the steel grade. To
improve cost effectiveness it is therefore very important to try to cast a strand without defects.

© 2005 The University of Liverpool 22

steeluniversity.org Continuous Casting User Manual

Internal cracking and surface cracking is decided by the strains and stresses in the
strand during casting. Possible measures to prevent crack formation are optimizing mold
powder and mold oscillation to result in an oscillation mark depth < 0.2 mm and to provide
good machine maintenance regarding misaligned rolls.
Center segregation can be reduced by choosing a combination of casting speed and
secondary cooling rate so that the point of final solidification is well within the soft reduction
zone. Having done so, the next step of optimization is to increase the soft reduction level
further to achieve a greater thickness reduction.
Inclusion content can be lowered by making sure that the residence time for liquid steel in
the tundish is as long as possible.
Oscillation marks are decided by the oscillation settings that are chosen before the
simulation is started. A failure here means that these settings must be optimized further to
result in smaller oscillation marks.

9 References
1) E Schürmann et al.: Einfluss der Kokillenoszillation auf die Oberflächenqualität von
Stranggussbrammen, Stahl und Eisen, 1986, vol. 106, pp. 1196-1201
2) H Tomono: Elements of oscillation mark formation and their effect on transverse fine
cracks in continuous casting of steel, Doctor Thesis, EPF Lausanne, 1979
3) AISE, The Making, Shaping and Treating of Steel, Casting Volume CD, AISE, 2003
4) Normanton et al: VAI´s 8th Continuous Casting Conference, 2000, Linz, Austria
5) T Kawawa: Report of 6th Meeting on Solidification of Steel, No. 6-III-9, Japan 1973
6) Y Morita et al.: Strain analysis on internal cracks in continuously cast steel slab, The
Sumitomo Search, 1985, vol. 30, pp. 19-30
7) Z K Han and B Liu: Prediction and Analysis of Internal Cracks in Continuous Cast
Slabs by Mathematical Models, ISIJ International, 2001, vol. 41, pp. 1473-1480
8) Y M Won et al.: A New Criterion for Internal Crack Formation in Continuously Cast
Steels, Met. Mat. Trans B, 31B (2000), 779-794
9) M Suzuki et al.: Simulation of transverse crack formation on continuously cast
peritectic medium carbon steels slabs, Steel Research, 1999, no. 70, pp. 412-419
10) M Wolf: Initial Solidification and Strand Surface Quality of Peritectic Steels in
Continuous Casting vol. 9, Iron- and Steel Society, Warrendale, USA, 1997
11) G Arth et al.: Mould powder consumption in continuous casting of steel, Bachelor
Thesis, Department of Metallurgy, University of Leoben, 2004
12) H Steinrück et al : Modeling for fluid flow in continuous casting, Berg- und
Hüttenmännische Monatshefte, Austria, Leoben, 1996, vol. 141, no. 9, pp. 399-403,
ISSN: 0005-8912

© 2005 The University of Liverpool 23