1. Ladle refining furnace
1. Introduction 6. Sampling
2. Basic Metallurgy 7. Operative Practice
5. Alloys and Slag builders
CVS MAKINA - LF TRAINING - D.E. 1
CVS MAKINA - LF TRAINING - D.E. 2
The LF represents a reality in a large number of steel production lines since
its origins date back to the era in which ladle furnace were equipped with
slide gates and porous plugs.
Thanks to the possibility of treating steel in the ladle, the melting furnace is
no longer subject to long refining times and its production capacities have
been accelerated while rendering time management much more flexible with
respects to the CCM.
In recent years all the above has contributed, thanks to a proper use of this
machine, to a significant improvement, not only form a production point of
view but also from a quality and economic point of view thus achieving
CVS MAKINA - LF TRAINING - D.E. 3
Some theoretical indications with extensive operating description
provided here below will show how this equipment, if use correctly, can
provide a decisive support to three main points, namely:
• Cost reduction
CVS MAKINA - LF TRAINING - D.E. 4
The LF is designed to ensure the following main purpose:
1. Easier secondary metallurgy operations
2. Optimal and through homogenisation process
3. Accurate analyses for the CCM
4. Improved castability in the CCM
5. Achievement of cleaning steel
6. Reduction of the EAF consumption figures
7. Reduction of the EAF operating times
8. Increase in the productivity of the whole line EAF-LF-CCM
CVS MAKINA - LF TRAINING - D.E. 5
1. Easier secondary metallurgy operations
The possibility of decreasing the S content level down to very values
(ex.0.003%) is of basic importance. This would occur under ideal conditions.
However the Key to a correct DeS process (normally reaching a delta ranging
from 50% to 80%, between the S value during the tapping phase and the final
S value of the CCM tundish) is the compliance with the three basic parameters,
• no slag is tapped from the EAF into the Ladle
•the value of the free oxygen is lower than 30 ppm
•during tapping and subsequently in the LF a slag is formed with a ratio higher
Provided that the slag is properly prepared (basic and sufficiently fluid) the
second positive effect is obtained. In fact, under this condition, many oxide and
a large number of inclusion rise to the surface of the fluid slag thus leaving the
steel very clean.
CVS MAKINA - LF TRAINING - D.E. 6
2. Optimal and through Homogenisation Process
Homogenisation is a very important process and it is achieved trough porous
plugs, the number of which depends on the ladle size (one for ladle up to 80t,
and two for larger ladles).
The homogenisation guarantee the following:
• High and regular speeds for chemical combinations
• Elimination of temperature pick-up
• Very representative samples for analysis
• Acceleration of inclusion decantation
• Required physical contact between steel and slag
CVS MAKINA - LF TRAINING - D.E. 7
3. Accurate Analyses for the CCM
In order to obtain such result, it is recommended to observe very carefully the
•make sure the the stirring is correctly carried out at all times and that the
adjusted of the argon and nitrogen flow rates and pressures are such that a
steel ‘’bare eye’’ achieved with a diameter not exceeding 10-12 inches with
respect to the slag
•Always carry out the first analysis two minutes after beginning to heat the steel
with electric arc when the slag is sufficiently liquid
•When sampling the steel for the analysis (see above) check the level of free
oxygen in the steel
•Keep under control the operations by following the corresponding steel making
CVS MAKINA - LF TRAINING - D.E. 8
4. Improvement Castability in the CCM
A considerable increase in the steel castability can be achieved provided that:
• the tundish and nozzle are sufficiently hot, with the ladle having necessarily
received the steel from the EAF at a standard temperature during tapping
• the heating conditions are the same as those of the tundish without tapping
slag from the furnace
• all other basic requirements are complied with
The above result can be achieved provided that:
•the steel temperature is very constant time
•the steel is much cleaner and free from macro inclusion
•the steel is chemically free from considerable quantities of oxides
CVS MAKINA - LF TRAINING - D.E. 9
5. Achievement of cleaner Steel
The simple fact that the treatment time, depending on the steel grade, lasts
from a minimum of 25-30 min to a standard of 50-60 min, obviously leads to the
decantation of impurities.
This phenomena, that occurs due to the difference of specific weights, may be
considerably accelerated, provided that the use is correctly, by varying the
•temperature changes above 80°C as compared with the Tliq
•calibrated stirring according to the current position, as to the specific operating
•use of correct slag viscosity
CVS MAKINA - LF TRAINING - D.E. 10
6. Reduction of the EAF consumption figures
7. Reduction of the EAF operating times
The decisive factor is that, by means of this system, the melting furnace is
released from all the aspects relevant to the so-called refining process. As
results,being just a melting machine, its functions will be the creation of a fixed
matrix as far as analyses and temperature are concerned, without obviously
pouring slag into the ladle.
By saving time and being able to tap at relatively low temperatures, it follows
that for the EAF there will be a considerable saving of :
real time, power supply, electrode and refractory consumption
Several times, by acting on the rate and on the repeatability, for long CCM
sequences (from 5 up to 50) in progressive scale, the EAF performance can
reach optimal levels thus reducing the relevant costs and improving
performance levels to unexpected but actual results.
CVS MAKINA - LF TRAINING - D.E. 11
8. Increase in the Productivity of the whole Line (EAF-LF-CCM)
Assuming to have a secondary metallurgy station it becomes easier to manage
programs, times,rate and conditions by gradually optimising them and, in the
vent of any delays of one of the production units, to use the LF as a buffer,
which is able to cope with the problems caused by the CCM as well as EAF.
Thanks to this natural flexible operation mode, sequence casting becomes
easier with consequent increase in the productivity and reduction of the
Another major factor, often considered as an effect, consists in the
improvement of the quality level which, in the rolling process and in the other
finishing phases, in addition to favouring high productivity levels, it considerable
lowers the quantity of rejects.
Substantially, this is a very strategic aspect that may easily result into positive
returns as far as sponsoring is concerned but also as regards immediate
CVS MAKINA - LF TRAINING - D.E. 12
7. Basic Metallurgy
LIQUIDUS AND SOLIDUS TEMPERATURE.
Pure Fe melts at 1536°C.
If the liquid iron contains other elements ,it will no longer have a fixed melting
point, but will rather melt within a solidification interval,determined by a solidus
and liquidus temperature. The solidus temperature is of limited importance to
us whereas the liquidus temperature is all the more important..
We use the T.liq as a reference level on top of which we add a certain over
temperature for defining the optimum casting temperature.
Fe is typically alloyed with Carbon, which changes the melting point from pure
iron to a liquidus line.
CVS MAKINA - LF TRAINING - D.E. 13
CVS MAKINA - LF TRAINING - D.E. 14
8. Basic Metallurgy
Our method of calculation considers 1536.6°C as the solidification
temperature of pure iron,
because of the presence of alloyed and residual elements in the steel,
which have the effect of decreasing the temperature,
it is necessary to take into consideration the chemical composition of the steel
in order to compute its liquidus temperature,
Our logarithm, widely used for calculation, is based on statistically evaluated
results of metallurgical laboratories.
CVS MAKINA - LF TRAINING - D.E. 15
Tliq = 1536,6 - (%C * Z) - (%E * W)
Tliq = liquidus temperature
C% Z E W
0.06-0.10 89 Si 8 %C = carbon content
0.11 -0.50 88 Mn 5
0.51-0.60 86 P 30 Z = multiplying factor for
0.61-0.70 84 S 25 carbon
0.71-0.80 83 Cr 1.5
0.81-0.90 82 Ni 4 %E = content of elements
Cu 5 listed in the table
W = multiplying factor for the
Al 5.1 listed elements
CVS MAKINA - LF TRAINING - D.E. 16
9. Basic Metallurgy
nowadays ,this instrument is very fast in supplying the steel analysis for all the
necessary and required elements, its principle is based on the basic
assumption that each chemical element has its individual specter, with precise
bands (similar to those found on a supermarket label) whilst the relevant
percentage is determined in relation to the proportional light emissions. Within
two minutes the sequences of the desired and required elements will be
directly displayed on the video screen and is generally based on the following
C Mn Si P S Cu Ni Mo Sn Al As Ti V Nb W Ca B Co Te
of course, the calibration and the reading curves, will be a function of the
required analysis and of the usual percentages of certain elements.
All this will be carried out through tests of samples with known chemical
analysis to obtain the conversion factors.
CVS MAKINA - LF TRAINING - D.E. 17
Finally the L.F. Operator will obtain some figures that should be interpreted as
• 0.01% is equivalent to one centesimal point,therefore saying that c=18 is the
same as c=0.18% or mn =67 is the same as saying mn =0.67% .This system
is valid for the following elements c.Mn.Si.Cu.Ni.Cr. W. Mo. And co.
• 0.001 is equivalent to one millesimal point ,therefore saying that p=29 is the
same as saying p is 0.029,or al,=55 is the same as al=0.055.This system
applies to the following characteristics:::;p.S.Al.Sn.As.Ti.Nb.B.V.Ca.Pb.Te
• O2..N2 and H2 as already stated ,one ppm (i.e. one part per million) is
equivalent to 0.0001% reference is made to carbon pick -up.
CVS MAKINA - LF TRAINING - D.E. 18
10. Basic metallurgy
Secondary metallurgy is divided into two steps.
• Operating steps
The basic steps are as follows
1. Preventing slag from leaving the primary melting units.
2. Mixing and homogenisation.
4. Charging of additives.
5. Vacuum treatment..
6. Shrouding of the steel flow in the ladle and Tundish.
7. Electromagnetic stirring during continuous casting.
CVS MAKINA - LF TRAINING - D.E. 19
1. Stirring gas treatment with porous plugs, lances or with the aid of
2. Hopper fed alloying elements ,deoxidation agents or slag forming agents.
3. Lance injected solids
4. Vacuum degassing with the aid of various techniques
5. Spooling in of cored wire
6. Ladle furnace with electric arc
The various methods and processes of secondary metallurgy are frequently
combinable, giving rise to precisely defined sequences ,for the production of
special steel grades and for the compliance within defined tolerances of the
alloying elements or additions
CVS MAKINA - LF TRAINING - D.E. 20
11. Basic Metallurgy
Slag is not an undesired by-product of refining.
On the contrary, it is an indispensable prerequisite for it.
The slag layer functions like a reservoir which can absorb and retain a tramp
element like S, but also in case the process is not properly controlled, return
the element to the bath and ruin the efforts of refining.
Refining slags normally have CaO as the major constituent,and such slags
are referred to as basic slags, in exceptional cases Acid slags with SiO2 as
the main component are used , mainly for the removal of manganese from
low manganese steels..
CVS MAKINA - LF TRAINING - D.E. 21
The reasons for preventing the primary furnace slag from entering the ladle
furnace are :
• minimizing the phosphorus reversion into the steel
• minimizing the amount of oxygen entering the ladle.
CVS MAKINA - LF TRAINING - D.E. 22
12. Basic Metallurgy
LIMITING THE LINING WEAR.
The primary furnace slag contains more or less phosphorus pentoxide P2O5,
bound to the lime as calcium phosphate.
The P content depends on the raw material used in the primary furnace.
Clean scrap and /or reduced pellets ( sponge iron made from pure ore ) give
hardly any phosphorus in the steel (or in the slag) made in the arc furnace.
The content of phosphorus pentoxide in the slag can vary from less than 1% to
more than 20%.
The acceptable amount of slag entering the ladle furnace is thus, amongst other
things,dependent on the phosphorus level in the slag
CVS MAKINA - LF TRAINING - D.E. 23
The crude steel leaving the primary furnace is in general oxidized. The
oxygen content being a function of the carbon level.
Since there is a relationship between the the oxygen content in the steel and
in the slag respectively, this means that the furnace slag can be more or less
aggressive to the refractory lining .
The oxygen exists in the slag as iron oxide Fe0, fluxes, MgO and CaO which
are the main constituents of the refractory lining .
The more of this slag coming into the ladle furnace, the greater is thus the
risk of lining wear. This type of slag also has a negative effect on the yield of
added alloys, especially those having a strong affinity for oxygen,such as Si,
Al, Ti, etc.
It is important in most cases to minimize the quantity of furnace slag entering
the ladle furnace.
CVS MAKINA - LF TRAINING - D.E. 24
13. Basic Metallurgy
On the other hand there are examples where large slag volumes are
desirable in the ladle furnace, such as when desulphurising.
This is in general fulfilled by adding adequate amounts of lime.
If however, the P content in the primary furnace slag is sufficiently below the
critical level, it is possible to let this slag ,or part of it ,remain in the ladle
furnace to take part in the subsequent refining (Desulphurization) which will
take place there.
In practice the slag during primary melting , for instance in the arc furnace is
allowed to run off during the final stages of the melting. At this stage the slag
usually contains a high iron oxide content
CVS MAKINA - LF TRAINING - D.E. 25
After tapping into the ladle ,the steel is normally pre-deoxidized but it still
contains a high oxygen content ,and the oxygen potential of the slag is very
high,.In order to order to prevent the primary furnace slag from entering the
ladle furnace there are three possible ways to deslag:
1. DESLAGGING OF THE PRIMARY FURNACE BEFORE TAPPING.
2. DESLAGGING OF THE LADLE BEFORE STARTING TREATMENT..
The method used mostly today is,is to deslag before treatment,this can be
done by tilting the ladle,.simply running off via a slag spout,or by the most
popular method of using a slag rake
CVS MAKINA - LF TRAINING - D.E. 26
14. Basic Metallurgy
Modern ladle cars ,have been fitted with a hydraulic system making it
possible,to tilt part of the ladle car,and with the use of the slag rake make it
very easy to deslag. A slag spout is necessary in order to protect the ladle
flange and to prevent the build up of slag on the brim of the ladle. Which would
jeopardize a satisfactory seal between the roofs and the ladle furnace..
The spout can be attached by a hydraulically operated device or placed on
and removed from the ladle with the help of a crane.
CVS MAKINA - LF TRAINING - D.E. 27
Refining slags are composed of the main constituents from the Ca0-SiO2-
Al2O3 - MgO system.
CaO comes from the added lime and the slag should normally be saturated
SiO2 and Al2O3 come from the primary furnace slag,deoxidation products
and,to some extent from the refractory wear, Al2O3 is also an important flux in
many synthetic slags.
The MgO content normally comes from the dissolution of MgO in the ladle
and roof refractories.
CVS MAKINA - LF TRAINING - D.E. 28
15. Basic Metallurgy
MgO may be added as pure material or as dolomitic lime for reducing
ladle lining wear,Its content is normally 6 to 8% in refining slags..
Addition of 5 to 10% of CaF2 may be needed for fluxing the slag.
FeO, MnO and Cr2O3 appear, as “tramp” constituents and their presence
is detrimental for the steel quality.Their sum should not exceed 1.0 to
S appears as CaS and is picked up from the metal.
P is practically absent in ladle refining slags,due to the low oxygen
TiO2 may appear as a result of dissolution of bauxite refractories.It is also
detrimental for steel quality.
CVS MAKINA - LF TRAINING - D.E. 29
Refining slags may have a different composition depending upon whether
Si-killed,or Al -killed ( Si free) steels are being produced.
Si -killed steels that shall be continuously cast into billets require that the
slag,have a low Al2O3 ,content in order to avoid ,tundish nozzle clogging.
Al -killed ,Si free steels require that the SiO2 content be low so as to
minimize the Si pick-up from the slag by the metal
CaO 56-62% CaO 56-62%
SiO2 6-10% SiO2 15-20%
Al2O 3 20-25% Al 2O3 5-8%
FeO+MnO+Cr2O3 <2% FeO+MnO+Cr2O3 <2
MgO 6-8% MgO 6-8%
S 0.3-2.0% S 0.3-1%
CVS MAKINA - LF TRAINING - D.E. 30
16. Basic Metallurgy
CVS MAKINA - LF TRAINING - D.E. 31
Slags for steel refining must have not only the right chemical character but
also the right liquidus temperature, so that they are readily liquid but not
overly aggressive at steelmaking temperatures.
Slags have liquidus and solidus temperatures just like metals.In contrast
to metals however, it is much more complicated to predict these
temperatures as a function of the chemical composition, because there is
a much stronger interaction between the slag constituents than between
Even if a slag is liquid,it may have such a high viscosity,that further
addition of flux is required.Generally speaking ,additions of Al2O3 or CaF2
are used to regulate the fluidity of slags on a rather empirical basis.
CVS MAKINA - LF TRAINING - D.E. 32
17. Basic Metallurgy
The heat capacity for the aforementioned slags is 0.41kwh/(tonne,K) and
practically independent of temperature within the 1500 - 1700 °C range.
The density of refining slags can be calculated with fair precision on the
basis of empirical models.Densities for the two aforementioned slags are
in the 2.7 to 2.8 kg/dm3 range for temperatures within the 1500 to
CVS MAKINA - LF TRAINING - D.E. 33
S te e l D e n sity
A s F u n c tio n o f T e m p e ra tu re
0 .1 2 % C , 0 .2 5 % S i, 0 .7 5 % M n
D en s ity, kg /d m 3
6 .9 5
6 .9 0
6 .8 5
1 ,5 5 0 1 ,6 0 0 1 ,6 5 0 1 ,7 0 0
T em p eratu re, C
CVS MAKINA - LF TRAINING - D.E. 34
18. Basic Metallurgy
The single most important numerical index which summarizes the
composition of a slag and serves as a single parameter with which its
chemical character can be correlated, is the (BASICITY INDEX.)
Basicity is not a univocal notion at least 14 different definitions have been put
forward over the years.
The simplest and most widespread definition is the one by Herty in the 1920,s
called V -RATIO and defined as :
V-ratio =(WT-% Ca O ) / ( WT-% SiO2 )
The V-ratio ignores the effects of the other basic (FeO,MnO, MgO, 0 and acid
( Fe2O3,Al2O3,P2O5) OXIDES.More complex definitions abound but do not
add any further substance to the discussion that follows.We shall therefore
limit ourselves to the use of V ratio for basicity.
CVS MAKINA - LF TRAINING - D.E. 35
The V ratio cannot be successfully used for accurately predicting
desulphurisation capacity but is merely and expedient way of roughly
characterizing a slag as regards its refining capacity
The V ratio in refining slags lies normally in the 1.8 to 3.5 range.
CVS MAKINA - LF TRAINING - D.E. 36
19. Basic Metallurgy
One of the most important tasks of refining slag is to absorb and hold S as
Slag property to hold S is called Sulphide Capacity and it is defined as
Cs = Ks * (%S) * [aO ] / (fS * [%S ] where log Ks = (935/T - 1.375)
Cs can be calculated from (%S) content in the slag, oxygen activity [aO ] in
the steel, sulphur concentration [%S ] in the steel, activity coefficient fS and
The expression for sulphide capacity can be witten as the following form that
is called sulphur partition ratio Ls
Ls = (S%)/[S% ] = Cs * fs / (Ks * aO)
and looking the above formula we understand the factors that promote the
CVS MAKINA - LF TRAINING - D.E. 37
Interaction of Carbon on Sulphur
Activity Coefficient f S
As Function of C Content at 0.03 % S, 1600°C
Activ ity Coefficient fS
0 1 2 3 4 5
C Content, %
CVS MAKINA - LF TRAINING - D.E. 38
20. Basic Metallurgy
Refining of steel normally means the removal of one or several elements
from the metal bath.Such refining is brought about by slag metal
reactions,gas-metal reactions or metal bulk reactions.
Refining of the following elements will be discussed: S,H,N,C,O.
In addition fundamental aspects of calcium treatment will be discussed.
Refining of a particular element may be performed by one or several of the
above mentioned reactions.Desulphurisation is a typical slag -metal reaction
whereas deoxidation can be carried out as a metal bulk reaction,slag -metal
reaction ,or even a gas -metal reaction.
CVS MAKINA - LF TRAINING - D.E. 39
Removal of S from steel into the slag is conventionally illustrate by the
[S] + CaO CaS + [O]
To achieve a good desulphurisation it is first of all necessary that three
fundamental conditions be satisfied.
1. Basic Slag.
2. High Temperature (60-80deg greater than delta T sl).
3. De -oxidation of the bath(aO2 less than 40ppm).
The stirring of the bath and the reactivity of the slag are important to speed
up the formation of Ca Sulphide.
CVS MAKINA - LF TRAINING - D.E. 40
21. Basic metallurgy
Desulphurisation is mathematically represented by the formula
S final = S initial * e -ks *(A/V)* t
A = bath area m 2
V = metal volume m 3
Ks = desulphurisation rate constant m/min
t = process time
Ks values depending on the stirring power and we can consider :
Ks in the LF = 0.08 (0.8-1.2 Nl/ton min)
Ks in the VD = 0.20 More details
CVS MAKINA - LF TRAINING - D.E. 41
Several elements that occur as tramp elements or alloying elements in the metal bath can move
between slag and metal. From a ladle refining point of view, the most important element in this
respect is S, and refining of it will therefore be described in rather detail.
The partition of an element between slag and metal is controlled by the chemical environment,
principally oxygen activity and slag basicity, and can be predicted by use of thermodynamic
CVS MAKINA - LF TRAINING - D.E. 42
22. Basic metallurgy
Removal of S from a metal bath into the slag is conventionally illustrated by the reaction:
S + CaO CaS + O
A basic slag is considered be completely ionised so that we preferably write the desulphurising
S + O2- S2- + O
By writing in this manner, we also confirm the fact that not only CaO but also MgO, FeO, Na2O
and other oxides to a varying degree contribute the oxygen ions necessary for the reaction.
We have already defined an expression for the S partition between slag and metal (see section on
sulphide capacity). Now we want to predict the final sulphur content in the metal, however, for
which purpose we must know the specific slag weight, i e kgs of slag per ton of metal, slag
sulphide capacity, the initial concentration of S in the slagbuilders and in the metal and, finally, the
S activity coefficient in the metal.
We assume that the total amount of S remains constant, i e that what has left the metal will be
found in the slag.
CVS MAKINA - LF TRAINING - D.E. 43
It is then possible to derive the expression:
[ %S ]*W ( )
+ %S0 *W slag
[ %S ]=
W steel +LS *Wslag
where ‘W’ denotes weight.
At this point, we substitute L S for C S * fS / (K S * a O) as described earlier in the discussion about
Sulphide Capacity. That brings us to the expression for final, equilibrium S content in the metal :
%S0 *W steel %S0 *W slag
W steel CS* fS / KS* a O *W slag
The activity aO can easily be calculated or measured in metals of very varying composition; fS is
likewise easily calculated using Wagner’s formalism; C S can be calculated from slag analysis and
temperature. Weights of metal and slag are already known.
We therefore have all necessary information for calculating the maximum possible extent of
desulphurisation. The figure shows an example with equilibrium sulphur content in the metal as
function of slag V-ratio.
CVS MAKINA - LF TRAINING - D.E. 44
23. Basic metallurgy
It is evident that the V-ratio must be at least greater than 2 in order that a good desulphurisation be
Referring back to the phase diagram CaO - SiO2 - MgO - Al2O3, the best desulphurisation results
are attained with slags which are just saturated with CaO.
The fact that a considerable desulphurisation is possible does not necessarily mean that a low
final S content will actually be reached. The exchange reaction between metal and slag takes time
and is mainly influenced by the stirring and mixing intensity.
Desulphurisation is mathematically represented by the well-known exponential function
CVS MAKINA - LF TRAINING - D.E. 45
kS * A / V * t
%S % S0 %S * e %S
where %S = S content at time ‘t’
%S0 = initial S content in metal
%S = S content after infinite time (equilibrium content)
kS = desulphurisation rate constant, m/min
A = stagnant metal-slag interfacial area, m 2
V = metal volume, m 3
t = desulphurisation time, min.
In the above function, we know everything except for the rate constant kS . The rate constant can
be approximated from the gas stirring flowrate, temperature and ambient pressure but such
models are not accurate. Measurement is actually the best method of determining the rate
constant for various operational cases such as heating in ladle furnace and treatment under
CVS MAKINA - LF TRAINING - D.E. 46
24. Basic metallurgy
R a te o f D e s u lp h u ris a t io n in L F a n d V D
a t In itia l S 0 .0 3 0 % , E q u ilib riu m S 0 .0 0 8 %
S C o n te n t, %
0 .0 3 5
LF VD E q u il i b ri u m
0 .0 3 0 k S 0 .0 5 k S 0 .2 0 0 .0 0 8 %
0 .0 2 5
0 .0 2 0
0 .0 1 5
0 .0 1 0
0 .0 0 5
0 10 20 30 40 50 60 70
D e s u lp h u r is a tio n T im e , m in
CVS MAKINA - LF TRAINING - D.E. 47
Evaporation of Metallic Constituents
Metallic constituents of a steel bath have vapour pressures that can relatively easy be calculated
as described previously.
If we let gas bubbles pass through the melt, they will therefore pick up evaporated constituents,
theoretically up to the saturation point at which the partial pressure reaches equilibrium with the
activity of the element in the liquid metal. Kinetic constraints may prevent the bubbles from
becoming saturated as they rise, however.
The metallic vapours condense and, generally, oxidise above the metal bath to form what we
simply experience as dust.
CVS MAKINA - LF TRAINING - D.E. 48
25. Basic metallurgy
Vacuum treatment and electric arc heating are two well-known cases of generous evaporation and
dust formation, although for different reasons.
Vacuum treatment generates substantial quantities of dust mainly because the inert gas bubbles
passing through the metal expand strongly due to the low pressure above the metal bath and
therefore pick up much evaporated solutes.
Electric arc heating generates much dust, principally because the temperature of the electric arcs
is extremely high which induces a lot of evaporation.
Zinc and lead have very high vapour pressure at steelmaking temperature and may therefore be
found in large quantities in dust. Such dust must be disposed of under safe conditions so as not to
Manganese has got a comparatively high vapour pressure. Elevated levels of MnO are therefore
always found in dust from secondary metallurgy installations. Vacuum treatment may produce a lot
of pyrophoric, condensed Mn particles in the suction line.
CVS MAKINA - LF TRAINING - D.E. 49
Examination of dust from ABS in
the VD dust separator
E v a p o r a tio n o f M e ta ls Element Content, wt-%
a n d D u s t G e n e r a tio n Fe 4-6
O2 Mn 10-15
D ust Cr 0-0.3
Fe Pb 2-6
T Mn pM n
Cr MgO 15-35
S tirr in g G a s
CVS MAKINA - LF TRAINING - D.E. 50
26. Basic metallurgy
C alc ulated E v aporation of Elem ents as
F unction of P ressure During Vacuum D egassing
1600 C , 1 % M n, 0.05 % A l, 25 m in utes
Ev aporated A m ount, kg/ton
0 .1 0 Mn
0 .0 8 Fe
0 .0 6 Al
0 .0 4 T o tal
0 .0 2
0 1 2 3 4 5 6 7 8 9 10
P res s ure, torr
CVS MAKINA - LF TRAINING - D.E. 51
FLUSHING OF DISSOLVED GASES BY GAS PURGING
The fact that the partial pressure of a dissolved gas depends upon its
activity in the metal, and hence upon its concentration, can be taken
of for flushing a dissolved gas out of the metal.
When the metal bath is flushed by an inert gas , for example Argon, the
atoms of the dissolved gas(-es) will recombine at the inert gas bubble
surface and escape into the bubbles.There, the dissolved gas builds up a
partial pressure which approaches equilibrium with the activity of the
dissolved solute. Taking nitrogen for example
2[N ] 2N N2
CVS MAKINA - LF TRAINING - D.E. 52
27. Basic metallurgy
The volumes of purging gas and dissolved gas in a bubble are proportional to
their respective partial pressures so that
Vdiss/Vpurge =Pdiss/ Ppurge
furthermore ,the total pressure is equal to the sum of the partial pressures of
dissolved gas(es) and purging gas according to
P=Pdiss +Ppurge Flushing of Dissolved
Gases by G as Purging
Layer Of S urfactant
2N p N2
T 2H p H2
CVS MAKINA - LF TRAINING - D.E. 53
Deoxidation is the process by which the content of dissolved oxygen is
reduced in the metal.One distinguishes between three different techniques
for bringing about deoxidation
1. PRECIPITATION DEOXIDATION
2. TOP SLAG DEOXIDATION.
3. VACUUM CARBON DEOXIDATION.
CVS MAKINA - LF TRAINING - D.E. 54
28. Basic metallurgy
PRECIPITATION DEOXIDATION - REMOVAL OF SLAG INCLUSIONS.
Precipitation deoxidation is by far the most common deoxidation method.
As the name suggests,a deoxidiser normally AL or Si is added to the
metal bath. Upon addition and mixing, the dissolved oxygen will react
with the added d.the combination of oxidizer so as to produce oxides
which precipitate in the melt.
The precipitation deoxidation in itself does not reduce the oxygen content
but just transforms the dissolved oxygen into oxidic oxygen. Next ,the
oxidic oxygen particles must be removed from the metal bath by flotation.
Expressed in the simplest way,the respective reactions are,
Al+O2 Al2O3 and Si+ O2 SiO2
CVS MAKINA - LF TRAINING - D.E. 55
The dissolved oxygen content which results from precipitation
deoxidation depends upon the value of the equilibrium constant and upon
the activities of the reactants and the reaction products.
The dissolved oxygen content, is what really concerns us for the sake of
cleanness of the metal. Al2O3 appears as clusters of small, solid, sharp-
edged particles in the metal.
They are easily floated by gas stirring and induction stirring, Si is rarely
used as a single deoxidiser but rather in combination with manganese.
CVS MAKINA - LF TRAINING - D.E. 56
29. Basic metallurgy
The combination of Si and Mn more efficiently reduces the oxygen content
than either one of the elements alone.
The reason is as follows, when using only silicon as a deoxidiser, the
deoxidisation product is pure SiO2 the activity which is equal to unity.
This SiO2, is solid up to 1713°C.
Furthermore the precipitated particles do not stick together like Al2O3
particles but appear as discrete particles in the metal.They are therefore not
easy to flotate after precipitation.
When adding both Si and Mn to the metal bath,both elements will be
engaged in the deoxidation so that the reactions will run in parallel.
Si +2O >>>>>SiO2
Mn + O >>>>>MnO
CVS MAKINA - LF TRAINING - D.E. 57
The reaction products SiO2 and MnO attract one another strongly because
SiO2 is of acid character whereas MnO is of strongly basic.A manganese
silicate will therefore form, in which the activities of SiO2 and MnO are
D is s o lv e d O x y g e n C o n te n t in
S i( -M n )- k ille d S t e e l a t 1 6 0 0 ° C
C 0 .1 2 , M n /S i= 3 . 0
100 S i o n ly
50 S i- M n
0 0.05 0 .1 0 .1 5 0.2 0.25 0 .3
S i C o n te n t, %
CVS MAKINA - LF TRAINING - D.E. 58
30. Basic metallurgy
One normally aims at an Mn/Si ratio after deoxidation of >3.
This ensures that the manganese formed silicate is liquid at
Precipitation deoxidation generates slag inclusions ( in most cases
AL2O3 or Mn -silicates) that must be removed by flotation from the
The picture illustrates schematically how a high, initial dissolved oxygen
content drops much faster than the total oxygen content.
The difference between total oxygen content and dissolved oxygen
content is the oxidic oxygen tied to the deoxidizer in the precipitates.
CVS MAKINA - LF TRAINING - D.E. 59
Floatation of Inclusions
Addition of Deoxidiser at Time = 0
40 Oxide + Dissolved Oxygen
0 5 10 15 20 25 30
CVS MAKINA - LF TRAINING - D.E. 60
31. Basic metallurgy
.The metal bath is always exposed to reoxidation,which generates new
Such reoxidation is caused by FeO, MnO and Cr2O3 in the top slag
refractories. It is also caused by direct oxidation with atmospheric oxygen
coming into the metal bath via an open purging eye or via the teeming
CVS MAKINA - LF TRAINING - D.E. 61
CVS MAKINA - LF TRAINING - D.E. 62
32. Basic metallurgy
CALCIUM TREATED STEELS
The aim of calcium treatment is in most cases to improve castability, by
modifying the sharp edged, solid Al2O3 particles into spherical liquid calcium
If reoxidation is allowed to take place after calcium treatment, new
unmodified Al2O3 particles will be generated and they will again obstruct the
casting. Precisely for this reason, Ca treatment must ,with few exceptions, be
the very last ladle refining operation, after which the steel must be cast
The stirring power and stirring type are of decisive importance for the final
cleanliness of the steel.
CVS MAKINA - LF TRAINING - D.E. 63
Refining by calcium treatment means that the metal bath is treated with
metallic calcium. This treatment is done to improve the castability and /or
improve the mechanical properties like isotropy with the steel product.
Several forms of calcium can be used, for example SiCa ,Ca pellets,and
CaFe. Whatever the form of Ca used it will immediately vaporize in contact
with the steel, because the vapour pressure of pure Ca is 1.92atm at
Its solubility at 1600 °C and 1atm is only 300 ppm.
Castability is improved by the reaction between Ca and Al2O3 particles in
the steel according to:
3Ca + Al2O3 >>>>>> 3CaO +2Al.
CVS MAKINA - LF TRAINING - D.E. 64
33. A l 2 O 3 -C a O
CVS MAKINA - LF TRAINING - D.E. 65
Al2O3 particles are solid and sharp edged at steelmaking temperatures. They are
prone to accumulating in the tundish nozzles and obstructing the steel flow.
By reaction with the formed CaO, the the Al2O3 particles will form liquid spherical
Ca -aluminate particles which flow through the nozzles without obstructing them.
x CaO + yAl2O3 >>>>>>>>>>(CaO)x (Al2O3)y.
Since Ca has such a strong affinity to oxygen and sulphur it can if supplied in
sufficient quantity,react and transform all oxides and sulphides existing in the metal
bath..Sulphides will be transformed to CaS whereas other oxides will be
transformed to CaO and release Mn ,Si etc back to the metal.
CVS MAKINA - LF TRAINING - D.E. 66
34. Basic metallurgy
Modification of Non-Metallic
Inclusions by Ca Treatment
Before Ca Treatment After Ca Treatment
Al2 O 3
Al2 O 3 +MnO Calcic
+SiO 2 globular
CVS MAKINA - LF TRAINING - D.E. 67
C h a n g e s in In c lu s io n C o m p o s itio n
D u rin g In je ctio n o f C a a t 1 6 0 0 °C
S teel: 0.1 C,0 .2 S i,0 .4 5 M n , 0 .0 3 A l, 0 .0 2 S
% A l2 O 3 in A lum inates, % C aS in In clus ions
1 60 0 °C
L iq u idu s
A lu m ina in
c alc ium a lu m ina tes , %
40 C aS in
in clu s ion s
La titu d e
20 inc lu sio ns
fully liqu id
0 20 40 60 80 100
Injec ted C a Q uantity, A rbitrary U nits
CVS MAKINA - LF TRAINING - D.E. 68
35. Basic metallurgy
The amount of Ca injected into the steel shall be such that the composition of the
Ca -alluminate falls roughly within the 3(CaO)(AL2O3) to (CaO)(AL2O3) range.
This ensures that the formed aluminates are liquid at steelmaking temperature.
Too little Ca causes insufficient modification and solid inclusions.
Too much Ca generates solid inclusions as well. As the inclusions become
increasingly rich in CaO a point is reached at which Ca starts reacting with the S to
create CaS according to
Ca + S>>>CaS.
This CaS is solid and will obstruct casting nozzles just like Al2O3 particles.
It is therefore important that the S content be reduced by another method of
desulphurization before Ca treatment in order that CaS will not be formed by direct
reaction with Ca.
CVS MAKINA - LF TRAINING - D.E. 69
As a rule of thumb, the S content should be below 0.010 % before
The consumption of Ca for modification of slag inclusions is 0.15.to
0.35 kg.ton (expressed as pure calcium).
Due to its high vapour pressure and its very limited solubility in
steel, practically all Ca in the steel after Ca treatment appears as
oxide or sulphide.
No significant amount of dissolved Ca remains as a buffer for
modifying any AL that may be deoxidized after complete Ca
CVS MAKINA - LF TRAINING - D.E. 70
36. Basic metallurgy
If the steel for one reason or another has become deoxidized after
Ca treatment , it will have to be Ca treated again .
Pneumatic injection of Ca ensures that all Ca reaches to the bottom
layer of steel in the ladle.
In case wire feeding is employed for Ca feeding, the wire diameter
and injecting velocity must be selected so as to ensure that the wire
rests the bottom of the ladle before melting and releasing this Ca
Ca supply rate should not exceed about 80gr/tonne,min in order that
the bath agitation be contained .
CVS MAKINA - LF TRAINING - D.E. 71
C a S i W i r e F e e d r a te a s F u n c t io n o f S t e e l W e ig h t
f o r 1 2 m m W ir e D i a m e te r
F e e d r a t e , m / m in
200 O d e r m a th
D ia 1 2 m m
0 50 100 150 200 250
S t e e l W e ig h t , to n
CVS MAKINA - LF TRAINING - D.E. 72
The electric arc burns between electrode and metal bath at a tempe rature of 4000 - 8000°C. At this
high tem perature the gas in the arc become s ionised and transform s into a plasm a. The high
plasma temperature promotes an efficient energy transfer between arc and bath under each
The electrom agnetic fields in each phase G eneration of Hot-Spot W ear
exe rt a strong influence on the arcs so that A rc D eflection - Hot Particle Jet
they deflect from vertical orientation and
incline towards the ladle wall during ea ch half
cycle. The higher the electrode current and
the closer the grouping of the electrodes, the
stronge r the deflection .
Electrodynamics forces cause the plasma to
have a very high internal pressure. The arc
therefore exerts a pressure on the m etal bath
and creates a dimple.
The electrodynamics force s in the electric arcs
generate a gas stream with a ve locity of about
1500 m/s along the arc.
The velocity is particularly high during the half
cycle when the electrode has negative
CVS MAKINA - LF TRAINING - D.E. 73
The electric arc length is approximately
L arc = Varc (Volt) - 35 (mm)
35 = anode -cathode voltage drop(constant), used for accelerating and
decelerating electron and ions.
E le c t r ic A r c L e n g t h a s F u n c t io n o f A r c V o lt a g e
A rc L eng th, m m
G a s S ti r r in g
20 In d u c t i ve S ti r r i n g
40 60 80 100 12 0 140
A r c V o lt a g e , V
CVS MAKINA - LF TRAINING - D.E. 74
Power input into a ladle furnace is normally limited by the refractory wear
The actual refractory wear rate is a very complex function of,among other
factors the following :
Arc power input Arc Power Input Limitation
Transmission of Heat to Metal Bath
Refractory type and quality
Metal and slag temperature.
Thermal profile in the refractory lining.
Arc Power Slag-Metal
Metal and slag composition. Interface Level
Type and power of stirring
Ladle furnace geometry.
CVS MAKINA - LF TRAINING - D.E. 75
The electric arcs are immersed in a slag which absorbs a large part of the heat
and transfers it to the metal bath.The slag is superheated 50--150k in relation to
the metal temperature.
The metal bath actually cools the slag.
The metal bath must be stirred by gas or inductive stirring so that the transferred
heat becomes homogeneously distributed.
A practical and reasonable approach to dimensioning the electric power supply is
to make it proportional to the heat transferring area.=Bath area
CVS MAKINA - LF TRAINING - D.E. 76
Experience shows that an arc power input of up to about 2mw/2 bath area can be
transferred to the metal bath and is tolerable as a dimensioning criterion for the
The arc power is not measured but can be calculated.
As a typical value,the arc power is about 84 - 88% of the active power being
supplied by the transformer.
The remaining 12% are lost in resistive heating of the electrical conductors and the
The heating rate as a function of the active power (ie.[Arc power]/0.88) is shown .
CVS MAKINA - LF TRAINING - D.E. 77
LF H eating Rate vs. Specific Power Input
H eating R ate, K/m in
W ate rc. roof
5 °C/m in
4 °C/m in
0 0.05 0 .1 0.15 0.2 0 .2 5
Ac tive Power, MW /tonne
Experience shows that an arc power input of up to about 2 MW/m2 bath area can be transferred to
the metal bath and is tolerable as dimensioning criterion for the power supply.
CVS MAKINA - LF TRAINING - D.E. 78
The following operational aspects of arc heating will be discussed
1. Slag composition and thickness
2. Stirring of metal
3. Carbon pick -up
4. Electrode wear
5. Pick up of gases from the atmosphere
6. Refractory wear
CVS MAKINA - LF TRAINING - D.E. 79
The electric arc is a plasma ,the stability of which depends upon the presence
of a conductive and insulating slag layer.
High arc stability is essential for attaining,high average power input and this is
one of the many reasons why a slag layer must always be present during arc
Metallurgical refining slag are in most cases of basic character i.e.
%CaO / %SiO2 > 2
Such slag have sufficient electrical conductivity.
The more acid slag ,the more pronounced the the tendency of the electrodes to
jump up and down.
CVS MAKINA - LF TRAINING - D.E. 80
If the slag layer becomes excessively thick,say 150 to 200mm ,the slag layer
may become superheated 1800 to 1900 °C ,and heating of the metal bath
virtually cease despite continuous power supply.
In such a case CaO in the slag will react with the graphite in the electrodes
to form calcium carbide according to the strong endothermic reaction
CaO+3C+q>>>>>CaC2 +CO.The carbide dissolves on the slag ,and may
later react with the humidity according to,
CaC2+H2O>>>>>>C2H2+CaO.To form acetylene,which is explosive.Care
should be taken that the slag thickness is properly controlled so as to avoid
carbide formation ..
CVS MAKINA - LF TRAINING - D.E. 81
The metal bath picks up carbon from the electrodes.
The extent of this carbon pick-up depends on the following factors,.
• Oxygen activity in the metal bath.
• Metal bath stirring method.
• Arc length.
Experience shows that carbon pick-up by a rimming steel,(oxygen activity
>150ppm) is practically negligible because carbon from the electrodes
reacts with dissolved oxygen in the metal bath to form CO.
The situation is different in steels with lower oxygen activity.
CVS MAKINA - LF TRAINING - D.E. 82
Absorbed by the metal.The electrode tip wear is visible as a cavity in
the electrode at the electric arc foot-point.
Fluctuations are provoked by stirring of the metal bath. Gas stirring
provokes a rather unsteady surface whereas inductive stirring ,if
installed provokes a standing w aeve of rather constant profile. The
incidence of metal droplets coming into contact with the electrodes is
greater with gas stirring than with inductive stirring and the carbon pick-
up is higher.The arc length is obviously of decisive importance for the
carbon pick-up.A study on carbon pick -up as a function of arc length
shows that the arc length for gas stirred ladle furnaces must be about
6.5cm in order that carbon pick-up will not exceed 2.0.To 2.5 ppm
min.The corresponding figure for inductive stirred ladles is rather 5.0
CVS MAKINA - LF TRAINING - D.E. 83
C pick-up during heating
(ppm /m in)
10 Ind. stir
0 20 40 60 80 100
Arc leng th [m m ]
CVS MAKINA - LF TRAINING - D.E. 84
Electrode wear is generally described as a combination of tip wear and side
The electrode wear is roughly proportional to the energy consumption in the
The total wear,expressed as kg of graphite per tonne of steel,can be
estimated, to 0.8 to 1.2% of the energy consumption,expressed as kwh/tonne,
thus corresponds to a graphite consumption of 0.48 to 0.72 kg/tonne.
CVS MAKINA - LF TRAINING - D.E. 85
Nominal diameter Lenght Weight per electrode Nipples-Kg and dimension AxL
1500 312 Kg 15 215.90-304.80
400 1800 376 Kg 20,5 241.30-338.60
2100 445 Kg
Nominal diameter Lenght Weight per electrode Nipples-Kg and dimension AxL
1500 324 Kg 222 T4 15,9 222.25X304.80
400 1800 397 Kg 222T4L 17,9 222.25X355.60
2100 456 Kg
* weight wthout nipple
Ucar AGX 250-450 diameter mm
bulk density 1.65-1.72 t/m3 kg/dm3 g/cm3
specific resistance 4.8-6.0 microohm * m
flexural strenght 9.0-12.5 Mpa
tensile strenght 6.0-8.5 MPa
coefficient ot thermal expansion WG 0.5-1.0 microm/°Cm(30-100 °C)
porosity 20-24 %
thermal conductivity 210-280 W/m °C
real density - all fully graphitized material have a real density of about 2.20 to 2.25
CVS MAKINA - LF TRAINING - D.E. 86
Provide premium grade electrodes are being used, electrode consumption is mainly dependent
on steelmaking process conditions. Current level, power on time, oxidizing electrode surface area
and tons per hour are the most important parameters influecing electrode consumption.
Empiriacally developed formula are avalilable to calculate the amount of the two continuous
electrode consumption process.
Typical average oxidation and sublimitation rates are :
RsubAC = 0.0135 kg/kA hr - per one electrode tip
Rox = 8 Kg/m2 hr - for the oxidinziing cone of water spray cooled electrodes.
Calculation of continuous graphite consumption (sublimation)
I = current per phase (KA)
2 tpo = power on (hrs)
C tip = Rsub * ( I * tpo) / ( 0.454 * P ) [Kg/ton]
P = productivity (tons/heat)
Calculation of continuous graphite consumption (oxidation)
A = oxidizingelectrode surface area (m2)
ttap = tap to tap (hrs)
side = Rox * ( A * ttap) / P [Kg/ton]
P = productivity (tons/heat)
CVS MAKINA - LF TRAINING - D.E. 87
Specific electrodes consumption performed at Alpha Steel
Guarante electrode consumption 0.012 Kg/kWh
Electrode consumption obtained 0.007 Kg/kWh
Initial weight (Kg) Final weight (Kg) delta W (Kg)
Electrode column 1 908 760 148
Electrode column 2 908 796 112
Electrode column 3 908 780 128
Total weight 1-2-3 2724 2336 388
Total Electric consumption after 18th heats 56800 Kwh
S.E.C obtained after 18th heats 0,007 kg/kWh
Initial weight (Kg) Final weight (Kg) delta W (Kg)
Electrode column 1 908 698 210
Electrode column 2 908 740 168
Electrode column 3 908 735 173
Total weight 1-2-3 2724 2173 551
Total Electric consumption after 28th heats 80700 Kwh
S.E.C obtained after 28th heats 0,007 kg/kWh
Initial weight (Kg) Final weight (Kg) delta W (Kg)
Electrode column 1 908 601 307
Electrode column 2 908 647 261
Electrode column 3 908 641 267
Total weight 1-2-3 2724 1889 835
Total Electric consumption after 39th heats 113800 Kwh
S.E.C obtained after 39th heats 0,007 kg/kWh
CVS MAKINA - LF TRAINING - D.E. 88
There are numerous factors which affect the final nitrogen level
•long arc versus short arc
•foaming slag practice
CVS MAKINA - LF TRAINING - D.E. 89
Electric arc heating is potentially cause a substansial pick-up of
atmospheric gases, notably nitrogen
N2 2N 2 [N]
The arc caused the N2 molecule to dissociate to atomic N or ions.
Nitrogen removal is also affected by surface actives elements O
CVS MAKINA - LF TRAINING - D.E. 90
P ic k - u p o f N it r o g e n a s F u n c t io n
o f S la g T h ic k n e s s
N it r o g e n P i c k - u p , p p m
0 20 40 60 80 100 120 140 160 180 200
S la g T h ic k n es s , m m
CVS MAKINA - LF TRAINING - D.E. 91
Pick-up o f Hy dro ge n in LF after V ac uu m
D ega ssing as Fun ction o f He ating T im e
H ydro ge n C o nte nt, p pm
0 20 40 60 80
H ea tin g Tim e, m in s
H ea t W e ig ht 17 2 tonn es
CVS MAKINA - LF TRAINING - D.E. 92
The refractory wear rate in the ladle,can be characterised as being of two
• General slag line wear
It is caused by the gradual dissolution of the slag line, refractory, mainly
MgO in the slag.
• Hot-spot wear
The extent of this wear mainly depends on
• slag composition
• working temperature
• poor stirring
• electrode regulation
CVS MAKINA - LF TRAINING - D.E. 93
M g O S o lu b ilit y a s F u n c t io n o f
S la g B a s ic ity a t 1 5 % A l2 O 3 a n d 1 6 0 0 ° C
M gO C o n te n t i n S la g , %
1 2 3 4 5 6 7 8
S l a g B a s i c i t y , % C a O / % S iO 2
CVS MAKINA - LF TRAINING - D.E. 94
Where does the electric energy go ?
• Heating of the steel Utilisation of Electric Energy, kWh/tonne
• Fusing of the slag builders. 100 ton LF, Steelgrade 20CrNi4
8.4 Melting Alloys 9.2
• Heating of the slag. Fusing Slagbuilders
• Melting of the alloys.
• Resistive losses in power
supply systems and electrodes.
• Losses to refractory, Resisitive Losses 16.8
roof, off gas and ambient.
The pie chart shows an example of electric energy distribution including
CVS MAKINA - LF TRAINING - D.E. 95
Selection of operating point for the ladle furnace.
The ladle furnace operator can set independent variables only, which define
the so called operating point of a ladle furnace..
1. Secondary voltage.
2. Electrode current.
The choice of operating point will then decide what values will be obtained of
the independent variables active, and reactive power, power factor heating
CVS MAKINA - LF TRAINING - D.E. 96
We shall here discuss how suitable operating pints are selected for a given
ladle furnace,the parameters which must be optimised are:
• Electrode current
• Active and apparent power
• Arc power
• Power factor
• Arc length.
CVS MAKINA - LF TRAINING - D.E. 97
Ex am ple O perating P oints for Ladle Furnace
Power, A rc Power, MW ,
Reac t. P ower, MV Ar,
Power F actor
Arc Length, cm
14 1 .4
12 1 .2
8 0 .8
6 0 .6
4 0 .4
2 0 .2
0 10 20 30 40
E lectrode Current, kA
A ctive P ow er A rc P owe r Reac tiv e Po we r Arc Leng th
Po we r Fa ctor
MW MW M VA r cm
CVS MAKINA - LF TRAINING - D.E. 98
Graphite electrodes can accept a certain maximum current density without
Higher current density may provoke cracking phenomena and superheating of
the electrode joints.
The maximum current densities recommend by the suppliers take into
account that a reduction of the electrode section takes place due to side
oxidation during use.
CVS MAKINA - LF TRAINING - D.E. 99
M A X IM U M E L E C T R O D E C U R R E N T
F O R V A R IO U S E L E C T R O D E G R A D E S
M a x im u m c u rre n t, k A
250 300 350 400 450 500 550 600
E le c t ro d e D ia m e te r, m m
S ta n d a r d Im p re g n a te d
CVS MAKINA - LF TRAINING - D.E. 100
Active and apparent power
The maximum active power is limited by the maximum permissible arc
Higher power will lead to rapid destruction of the refractory lining in the
Even though the maximum power with respect to tolerable refractory wear
rate is not exceeded.
One must also check that the transformer itself is not loaded too much.
CVS MAKINA - LF TRAINING - D.E. 101
The maximum arc power is limited by the refractory wear rate in the hot spots
and in the slag line.
The maximum permissible specific arc power is in the order of 2 MW/m 2 metal
CVS MAKINA - LF TRAINING - D.E. 102
Refractory wearing index is a parameter that define the maximum
arc power that can be supplied to the metal without damage the
RWI has been calculated by using Schwabe formula
RWI = P arc * (V arc -35 ) / (3 a 2 )
a = bath diameter - pitch circle diameter - electrode diameter
CVS MAKINA - LF TRAINING - D.E. 103
The actual power of the power factor cos gives us an idea about the arc
cos is the angular displacement between voltage and current
if cos =1 V and A run in parallel without displacement. In such a case there
would not exist any voltage to re -ignite the arc after the current had passed
through zero during each half cycle.
It has to be a displacement between voltage and current such that the voltage
runs before the current and provides sufficient re-ignition.
Only in that case we will obtain an uninterrupted stable arc and high average
CVS MAKINA - LF TRAINING - D.E. 104
P h a s e D is p l a c e m e n t B e tw e e n C u rr e n t
a n d V o lta g e a t c o s = 0 .8 0
R e la t iv e V a lu e O f V o lta g e A n d C u rr e n t
V o lta g e
C ur ren t
- 0 .5
0 90 180 270 360
C y c l e A n g le , d e g r e e s
S u f f ic ie n t vo lt ag e f or reig n itin g th e elec tric
a rce n th e c u rre n t p a s s e s th r ou g h z ero .
CVS MAKINA - LF TRAINING - D.E. 105
Experience shows that a power factor of cosp = 0.78 to 0.80 is ideal for ladle
furnace operation .
With a liquid slag of the proper composition it is possible to operate up to
0.90 without problems.
There is no purely technically motivated lower limit for the power factor.
Since P = S*cosp it follows that S = P / cosp.
The transformer must be dimensioned on the basis of the required active
CVS MAKINA - LF TRAINING - D.E. 106
The arc length is chosen with respect to the carbon pick -up.
In the ladle that used stirring gas arc length is chosen within the range 6 - 9cm.
Arc length depends on the arc voltage by the empirical formula
L arc = V arc - 35
35 = is anode-cathode voltage drop for accelerating and decelerating electrons and ions
Apparent power S 17 + 20% MVA
Active power P < 14.2 MW
Arc power p arc < 12.1 MW
Arc length 6 - 9 cm
CVS MAKINA - LF TRAINING - D.E. 107
Limitations imposed by the electric power supply
The utility company supplying electric power to a ladle furnace,generally
imposes certain restrictions as to the furnace load, typically .
• Flicker < 1.0 Pst95
• Limitations on harmonics generation
• Cosp > 0.95
In contrast to the EAF ,the ladle furnace constitutes a very stable load so that
flicker and harmonic generation are of no practical concerns, since a ladle
furnace operates at a cos p of 0.70.-0.85 a power factor compensation
equipment is often requested for raising cosp to 0.95 on the medium voltage
line. Such equipment typically consists of a capacitor bank ,designed as a
filter for the third and if necessary for reasons of limiting the voltage increase
when switching in the filter.Additionally for the 5th harmonic see the line
CVS MAKINA - LF TRAINING - D.E. 108
Schematic Representation of Ladle Furnace Power
Supply with Power Factor Compensation
High Voltage Line
Medium Voltage Line, cos 0.95
(small furnaces only)
cos 0.70-0.85 Furnace
Filter 3rd Filter 5th
CVS MAKINA - LF TRAINING - D.E. 109
Stirring and mixing are absolutely indispensable for efficient steel refining.
Stirring refers to the velocity with which a volume or element moves in a
G as Stirring in Ladle
Stirring Gas Supply
CVS MAKINA - LF TRAINING - D.E. 110
Stirring is required for reactions to take place within the steel bulk such as the
flotation of non metallic inclusions.
It is also required for reactions between the metal and the atmosphere,for
example during vacuum treatment.
Mixing between metal and slag is additionally required for mass exchange
between metal and slag,such as for desulphurisation.
CVS MAKINA - LF TRAINING - D.E. 111
Purging Plugs with Random
or Directed Porosity
Slots in dense material
porous for directed porosity
Directed porosity is generated by producing narrow slots or channels in
an otherwise dense plug.
Ores slots or channels must be so narrow that the steel surface tension
prevents steel from flowing back into them.which means that a
diameter/gap in the order of 0.1 to 0.5mm.
CVS MAKINA - LF TRAINING - D.E. 112
A bottom purging plug is in most designs a conical, refractory brick with a
gas tight sheet steel casing, the plug has an appropriate gas permeability
that the inert gas can be purged through it.
Plugs have random porosity or directed porosity.
Random porosity is created by a special size distribution of the refractory
grains constituting the plug.
CVS MAKINA - LF TRAINING - D.E. 113
Purging Plug Set W ith Bayonet And G ate Fitting
W ell block Plug Sleeve
CVS MAKINA - LF TRAINING - D.E. 114
Inert G as Flow ra te as Function of
Pressure Difference in P urging P lugs
G as Flowrate, Nl/m in
900 M any slots - high porosity
300 Few slots - low porosity
0 1 2 3 4 5 6 7
Pressure Difference, bar
P roperties determ ined at room tem perature
CVS MAKINA - LF TRAINING - D.E. 115
TOP LANCE STIRRING.
Top stirring lances are usually of monolithic design with 1 to 3 gas exit nozzles.
They are made usually of high alumina castable on a bauxite or corundum
Only such high grade materials stand a chance to survive more than one heat
in the ladle furnace.
CVS MAKINA - LF TRAINING - D.E. 116
Inductive stirring of a metal bath works as an electric motor in which the
stirrer itself is the stator and the metal bath the rotor.
A current of up to 1350A in the inductive stirrer generates a very strong
travelling electromagnetic field.
The metal velocity pattern is more uniform with inductive stirring than with
gas stirring which ensures rapid homogenisation of the temperature and the
The lack of ripples on the metal surface minimises carbon pickup during arc
heating but gives poor mixing between slag and metal and therefore poor
CVS MAKINA - LF TRAINING - D.E. 117
CHARACTERISATION OF STIRRING.
This phenomena is expressed mathematically according to (sundberg).as:
= 371* Qgas *Tl * [( 1-Ta/Tl) + in (P 1/P2)] W/ton
where Qgas = gas flowrate Nm3/sec.
G as S tirrin g P o w er A s F u n ctio n
Tl =metal temperature K. O f M e ta l B at h H e ig ht
Ta = gas entrance temperature K B ath H eig ht, m
P1 =total pressure at ladle bottom
P2 = pressure above bath surface 2 .5
0 10 20 30 40
S tirring P o w er, W /to nn e
G a s flo w r a te 1 .5 N l /m in,to n , 1 6 0 0 °C
CVS MAKINA - LF TRAINING - D.E. 118
60. Alloys and Slag buiders
Additions of ferro alloys affects metal bath temperature.
The cooling coefficient is expressed as temperature drop upon the
additional of one kg of material into one tonne of steel.
The energy needed to bring a ferroalloy into solution in the metal bath
depends on the following steps :
• heating from room temperature to the melting point of material
• melting the material
• bringing the material from its melting point to the metal temperature
• dissolving the material into the metal bath
CVS MAKINA - LF TRAINING - D.E. 119
Alloys and Slag buiders
material is lost in various ways, the losses are principally of three kinds
• losses due to handling and transportation
• air oxidation
• losses of alloying elements by reaction in the metal and in the slag
One alloying material for example FeSiMn contains several alloying
elements as Si, Mn, Fe, etc
CVS MAKINA - LF TRAINING - D.E. 120
61. Alloys and Slag buiders
Material Stirring G as
h = (Material Yield) * (Air Oxidation Factor)
* (Chemical Yield)
CVS MAKINA - LF TRAINING - D.E. 121
Alloys and Slag buiders
Ferroalloys added to the metal are heated, melted and mixed.
Heating and melting of a alloy is called dissolution and after the mixing
phase is homogenisation.
D iss o lu tio n A n d H o m o g en is a t io n
O f F err o a llo y s
F orm ation of
s teel "s h ell"
H om og en i-
s atio n
M e ltin g of
p ar tic le s
S ti rrin g G a s
CVS MAKINA - LF TRAINING - D.E. 122
62. Alloys and Slag buiders
The time required to complete the three operation, described before,
depends on the following factors :
• quantities of ferroalloys
• quality and grain size of ferroalloys R ate O f D issolutio n O f H e av y Alloys
Sche m atic D iagram
• steel temperature D iss olution T im e, m in
• heating 10 poor stirring
• slag quality
S m all particles
0 20 40 60 80 100
Superheating of M etal Bath, °C
CVS MAKINA - LF TRAINING - D.E. 123
Alloys and Slag buiders
Chill factor Density Melting
K/Kg,t t/m3 range °C
FeSi -0,49 3,70 1,60 1265->1330
FeMnHC 2,74 7,30 4,10 1150->1265
Mn metal 2,30
FeSiMn 1,70 6,10 3,00 1075->1320
FeP 2,59 3,90
FeCrHC 2,49 7,10 3,85 1340->1450
FeCrLC 1,94 7,30 3,90 1600->1650
Cr metal 2,15 7,20 3,90 1830
FeS 1,91 3,00
FeNi 1,30 8,90 3,60 1454
FeMo 60 1,08 9,00 4,60 1800->1900
FeNb 1,10 8,10 4,80 1530->1580
FeV 60 1,81 6,90 3,84 1550->1600
FeV 80 1,77 6,40 3,62 1690->1770
FeTi 70 1,23 5,40 3,51 1070->1135
Cu 2,00 8,90 1080
Al pigs -1,80 2,80 1,44 657
Al w ire -1,80 2,40 657
Al gran. 1,60 657
CVS MAKINA - LF TRAINING - D.E. 124
63. Alloys and Slag buiders
TYPICAL ANALYSIS %
C Si Mn P S Cr Ni Co Mo Cu Nb V Al Ti N W
FeSi 0,10 77,00 0,40 0,05 0,02 0,40
0,10 77,00 0,30 0,05 0,02 0,30 1,50
0,10 77,00 0,30 0,04 0,02 0,20 0,10 0,05
0,10 65,00 0,40 0,05 0,02 0,40 2,50
0,20 45,00 0,60 0,05 0,02 0,50 2,00
FeMnHC 7,00 6,00 78,00 0,05 0,02
FeMnMC 2,00 3,00 88,00 0,10 0,02
FeMnLC 0,50 1,80 90,00 0,05 0,02
FeSiMn 0,50 25,00 65,00 0,05 0,02
FeP 3,50 5,10 25,00 2,50
FeCrHC 7,00 1,50 0,03 0,06 65,00
FeCrLC 0,05 1,50 0,03 0,02 67,00 0,20
Cr metal 0,03 0,20 0,02 0,02 99,00
FeS 48,0 1,10
FeNi 0,02 0,00 0,00 99,80
FeMo 60 0,05 0,80 0,05 0,10 60,00
FeW 0,10 0,80 0,20 0,03 0,02 6,00 0,1 4,00 80,00
FeNb 0,10 2,00 0,08 0,08 67,00 1,50 0,80
FeNb w ire 65,00
FeV 60 0,20 1,50 0,05 0,05 60,00 2,00
FeV 80 0,20 1,50 0,05 0,05 80,00 2,00
FeTi 70 0,20 0,50 0,10 0,05 0,05 5,00 70,00
Al pigs 0,15 0,50 99,70 0,04
Al gran. 1,20 0,50 96,00
CaSi 1,00 62,00 1,50
CVS MAKINA - LF TRAINING - D.E. 125
Alloys and Slag buiders
TYPICAL ANALYSIS % Chill Density Bulk Melting Size
SLAGBUILDER CaO SiO2 Al2O3 MgO CaF2 FeO MnO S P H2O CO2 Factor Density Range
°C/(kg/t) t/m3 t/m3 °C mm
Lime 86 1,8 6 0,06 0,1 2 3 3,1 1,0 3->15
70 2 22 0,06 0,1 2 2 3,1 1,0 3->15
Fluorspar 20 75 0,3 3,1 1,3 3->15
4 92 0,15 0,1
5 92 0,2 0,1
7 90 0,3 0,2
Alumina 0,1 0,25 96,5
Calcium aluminate 50 3 42 1,5 1,7 5->10
CVS MAKINA - LF TRAINING - D.E. 126
The steelmaking process time cycle is became shorter year by year
A a consequence new technology for the immediate determination
of various properties are being developed. Industrially applied
methods are available for direct reading of :
• oxygen activity
• hydrogen content
• nitrogen content
• carbon content
CVS MAKINA - LF TRAINING - D.E. 127
A good sampling method exhibits the following characteristics
• the sample is representative of the entire melt
• the sampling procedure can be mechanized
• sample preparation, if any, is simple and rapid
• high reliability and reproducibility with sampling and analyzing
• sampling+transport+preparation+analysis+reporting time is short
• cost is reasonable
CVS MAKINA - LF TRAINING - D.E. 128
Probe for metal samples
900 mm Mould
Quartz glass tube, 60*6 mm
Single cap, 0.6 mm thick
CVS MAKINA - LF TRAINING - D.E. 129
L F A utom a tic S a m pli ng E qu ip m e nt
The automatic sampling equipment
consists of a support structure with
moving sampling lance holder. The
lance immersion depth and immersion
time is automatically controlled.
CVS MAKINA - LF TRAINING - D.E. 130
Oxygen Activity Probe
Silica tube with
reference mixture Solid electrolyte tube
Slag cap ZrO2; CaO stabilized
cardboard and steel
CVS MAKINA - LF TRAINING - D.E. 131
"Bofors" Mould for Total
Sampling Total Oxygen
Content in Ladle Oxygen Content Sampling
W ooden cover
Sample Sample pin turned
from solid sample
Al-wire for killing
CVS MAKINA - LF TRAINING - D.E. 132
Hydris is a method for
direct reading of
hydrogen in the steel .
The hydrogen is
absorbed by the
nitrogen when it passes
as bubbles trough the
steel inside the bell.
The nitrogen becomes
saturated with hydrogen
after it has passed
several times through
CVS MAKINA - LF TRAINING - D.E. 133
Glass Pipette for C o p p e r M o u ld
ID 3.5+0.2 mm
OD 6-7 mm
CVS MAKINA - LF TRAINING - D.E. 134
68. Operative practice
This chapter describes general guidelines for Standard Operating practice
for CVS MAKINA ladle furnace.
The following operation will be discussed :
• treatment sequence
• temperature control strategy
• alloying strategy
• slag making strategy
CVS MAKINA - LF TRAINING - D.E. 135
The total time cycle for a ladle furnace ca be divided as follows :
T total = T process + T transfer + T buffer + T random + T idle
T process = actual time of refining steps
T transfer = ladle transfer time between primary furnace and LF
T buffer = waiting for the CCM as a buffer
T random = it does not need explication
T idle = idle time between heats
CVS MAKINA - LF TRAINING - D.E. 136
69. Operative practice
Operations time Cumulative time
Tapping 3 3 The ladle stay in the ladle furnace
Ladle transfer from EAF to LF 4 7 for about 40-45 min but this time
Lalde transfer under LF roof 1 8
Argon coupling-stirring control 2 10 can be reduced to 35-40 min if the
Slagbuilders additions 1 11 steel can be processed allows it.
Heating and stirring 3 14
Temperature and sampling 1 15 We have to understand that
Heating and stirring (fluxes 6 21
additions) process time sometimes is longer
Ferroalloys additions 2 23 due to human errors and not to
Heating and sstirring (fluxes 5 28
additions) machine failure.
Temperature and sampling 1 29
Heating and stirring (fluxes 5 34
Ferroalloys additions 2 36
Heating and stirring 5 41
Temperature and sampling 1 42
Heating and stirring 4 46
Trimming additions 2 48
Stirring for inclusion removal 3 51
Lalde transfer to CCM or ingot 5 56
Total residence time in LF 43
CVS MAKINA - LF TRAINING - D.E. 137
Schematic Ladle Refining Cycle
Operative practice Temperature, C
Flux Tliq + 70/80°C
Schematic heat cycle is
used as a reference for
this discussion 30-35°C
Tliq before tapping
Tliq after tapping
Tliq after alloying
Heating1 Heating 2 Trim-
A T A A
CVS MAKINA - LF TRAINING - D.E. 138
T T T
70. Operative practice
The general guiding are the following :
• C is added at tapping in such a quantity that the remaining adjustment
of C can be made by using high C alloys ( very difficult to control )
• Mn and Si ratio must be respected
• Si is added at tapping to 0.2 %
• Al is added for deoxidation and analysis
• in tapping phase the analysis must be 80% of the final
• lime quantity according to 1% rule
CVS MAKINA - LF TRAINING - D.E. 139
Slag making strategy
CaO 56-62% CaO 56-62%
SiO2 6-10% SiO2 15-20%
Al2O 3 20-25% Al 2O3 5-8%
FeO+MnO+Cr2O3 <2% FeO+MnO+Cr2O3 <2
MgO 6-8% MgO 6-8%
S 0.3-2.0% S 0.3-1%
CVS MAKINA - LF TRAINING - D.E. 140
71. Operative practice
LF operation Specific flowrate Flowrate for 100 ton
Sampling 0.5-1 Nl/ton, min 50-100 Nl/min
Heating 1.0-2.0 Nl/ton,min 100-200 Nl/min
2.0-3.0 Nl/ton,min 200-300 Nl/min
>3.0 Nl/ton,min > 300 Nl/min
<1 Nl/ton,min < 100 Nl/min
CVS MAKINA - LF TRAINING - D.E. 141
Example of Ladle furnace operations.
1. On arrival at the l.F,connect Ar Hose line and start stirring with a 15/20 cm
eye, take temperature and slag off if required,move the carriage to the roof
position and lower the roof from the local box on the platform.
2. Apply the power (the tap will be according to the initial
temperature).Increase stirring eye to 30 cm and add 1st 100 kg CaO reduce
stirring after addition.
3. After 4-5 min take sample,,check slag condition for viscosity.Take
temperature if slag thick add caf2 or other on a10-1 ratio.
4. On receipt of analysis .Calculate alloy additions ,add additions and 2nd
100kgs CaO increase stirring before making the additions(if level two is in
operation alloys will be calculated automatically from sample analysis.
CVS MAKINA - LF TRAINING - D.E. 142
72. Operative practice
5.After 4-6 min (depending on amount of alloy addition) take sample and
temperature ,check slag condition.
6. Adjust temperature whilst awaiting sample result to correct temperature
for casting,v/d treatment ,etc.
7.On receipt of analysis result,make trim additions(if necessary)this
includes al additions .
8.If al spec high and ladle is going direct to CCM or ingot casting make
CaSi wire feeding for inclusion control.
9.If the ladle is proceeding to the v.D,keep the al ,or if si killed to the top of
the range to allow for the drop during the next process.
10. If temperature and analysis are ok ,dispatch ladle to casting or VD.
CVS MAKINA - LF TRAINING - D.E. 143
Slag from the EAF in the ladle
1. Before starting to heat , put 50kgs al balls (size 2p3mm ) over the slag
2. Heating with electric power for 2 minutes
Add 400kgs CaO
3. Heating for 4 minutes
4. Oxygen sampling
5. If oxygen free greater than 80 ppm,add 40kgs al and 200kg CaO oxygen
sample. Repeat unti oxygen free is less than 80 ppm
6. Temperature and analysis sampling
7. Start standard LF Treatment
CVS MAKINA - LF TRAINING - D.E. 144
73. Operative practice
Ladle porous plugs not working.
1. Transfer ladle immediately to the ladle furnace.
2. Maximum electric power for 5 minutes.
3.Prepare one empty ladle ready to receive the heat.
4. Try to open the plugs with the max gas pressure (2 or 3 times)
5. If the porous plugs open ,start standard l.F. Treatment.
6.If the plugs remain closed make the following :
Pour the steel into the empty ladle.
Give max electric power for 15 minutes.
Then start standard LF Treatment.
CVS MAKINA - LF TRAINING - D.E. 145