You are on page 1of 142

* *

* *
ISSN 1018-5593
European Commission
t e c h n i c a l s t e e l r e s e a r c h
Reduction of iron ores
Te c h n i c a l s t udy i n t o t h e me a n s
of pr ol on gi n g bl a s t fur n a c e c a mpa i gn l i fe
STEEL RESEARCH
European Commission
t e c h n i c a l s t e e l r e s e a r c h
Reduction of iron ores
Te c h n i c a l s t udy i n t o t h e me a n s
of pr ol on gi n g bl a s t fur n a c e c a mpa i gn l i fe
D. Jameson, H. Lungen, D. Lao
Br i t i s h St e e l pi c
9 Albert Embankment
SE1 7SN
United Kingdom
Contract No 7210-ZZ/570
1 June to 1 December 1995
Fi n a l r e por t
Directorate-General
Science, Research and Development
1997 EUR 17247 EN
A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server (http://europa.eu.int.)
LEGAL NOTICE
Neither the European Commission nor any person acting on
behalf of the Commission is responsible for the use which might be made of the
following information
Cataloguing data can be found at the end of this publication
Luxembourg: Office for Official Publications of the European Communities, 1997
ISBN 92-827-9912-3
European Communities, 1997
Reproduction is authorized, except for commercial purposes, provided the source is acknowledged
Printed in Luxembourg
CONTENTS
Page
OBJECTIVE 5
INTRODUCTION 5
3. OPERATIONAL PRACTICES
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
Productivity
Burden
Ti 0
2
Addition
Instrumentation and Control
Tuyere Diameter
Off-blast Periods
Casthouse Practices
Production Rules
References
9
10
17
19
26
26
29
32
33
4. REMEDIAL ACTIONS 77
4.1 Gun/Spray 77
4.2 Enhanced Cooling 78
4.3 Grouting and Welding of Shell 80
4.4 Replacing of Staves and Coolers 81
4.5 Ancillary Equipment 82
4.6 References 88
5. FUTURE DESIGNS 105
5.1 Fourth Generation Staves 105
5.2 Copper Staves 1
5
5.3 Improved Refractories 107
5.4 More Comprehensive and Reliable Instrumentation 110
5.5 Throat Cooling 110
5.6 Improved Hearth and Taphole Design 110
5.7 Furnace Designs and Dimensions 112
5.8 References 118
CONCLUSIONS
133
PROLONGING BLAST FURNACE CAMPAIGN LIFE
British Steel pic
ECSC Agreement No. 7210.ZZ/570
Final Technical Report
1 . OBJECTIVE
The objective of this study was to carry out a comprehensive literature review of techniques used
to extend the campaign life of blast furnaces and identify important aspects of operational
practices, remedial actions and future plant design, to establish cost effective methods of
extending European blast furnace campaign lives.
2. INTRODUCTION
In recent years there has been an increasing requirement to extend blast furnace campaign life:
In order to reduce hot metal costs, by reducing manpower requirements and capital
costs, steel companies have increased the size and reduced the number of their blast
furnaces. In many cases there is no longer a standby furnace to be brought into
operation during blast furnace rebuilds
11
'. Consequently, long campaigns with
minimum reline periods are essential.
The cost of rebuilding or relining a blast furnace can be very high and may represent
a large proportion of the total capital expenditure available to a company. In recent
years, the level of subsidy to the steel industry has fallen, following an increasing
degree of modernisation and privatisation. When funding for capital projects relies
directly on business profits, the cyclical nature in steel demand and prices will have a
major effect on the capital available, and it may be necessary to extend plant life until
improved trading conditions are predicted. In addition, the priority for capital
expenditure has tended to move towards the finishing end to improve product quality
and range, to increase sales and enhance profit directly. Consequently, this results n
a reduction in the proportion of capital available in the primary end.
Techniques to extend blast furnace campaign lives have often been pursued more actively outside
Europe, notably in Japan, where campaign lives have been steadily increasing since the 1970's.
Figure 2.1
(2)
indicates the operational results of Japanese blast furnaces of over 2000 m
3
inner
volume for campaigns completed between 1970 and 1981. Towards the end of that period the
longest campaign was 7 years, with the production per unit volume approaching 5000 t/m
3
. Figure
2.2
(3)
dates from the mid 1980's and shows that there was an increasing number of furnaces,
many then still operating, with a campaign output above the 5000 t/m
3
maximum indicated in
Fig. 2.1. More recent data, from a variety of technical literature, are presented in Fig. 2.3 which
illustrates how the campaign output of Japanese furnaces has increased further, with a figure of
10000t/m
3
being exceeded in 1993. Completed campaign lives are typically 11-13 years, but
these values have been exceeded by furnaces whose campaigns are yet to be completed.
Using information from recent EBFC Blast Furnace Constructional Features compilations, Fig. 2.4
compares the campaign output of European and Japanese furnaces over a similar time period.
Although the best European examples are approaching Japanese levels, in general the campaign
life of European furnaces is significantly lower.
A shorter campaign life may result from a lower level of capital investment, as rebuilding a blast
furnace for a longer campaign life involves additional capital cost. It could also result from a lower
level of development of new equipment and techniques, or from inferior operation. Figure 2.2
suggests that larger furnaces tend to have a slightly higher campaign output per unit volume. This
difference may result from larger furnaces generally being of more modern design or because the
cost of extending the life of a large furnace is lower per tonne of output gained and therefore is
more economic than for a smaller furnace. The viability of an integrated steelworks depends on a
continuous supply of hot metal, which on a works with a small number of large furnaces puts great
importance on long campaign life. Such furnaces are likely to be well instrumented and operated
to achieve this aim. Since Japan tends to have a greater proportion of large furnaces, this could
explain some of the difference between Japanese and European campaign outputs.
Traditionally, the limiting factor to blast furnace campaign life was damage in the lower shaft
13
'.
With improvements in operational practices (e.g. improved burden distribution control), remedial
actions (e.g. stave replacement, gunning, etc.) and improved designs (e.g. improved cooling and
refractories), there are fewer problems in this area. Consequently, other regions which used to
have less effect on furnace life become more critical, e.g. throat armour and upper stack
refractories. Ultimately, as techniques are developed to extend the life of, and repair, the various
regions of the furnace, the most critical region becomes the most difficult region to repair, which is
generally considered to be the furnace hearth.
The techniques for prolongation of blast furnace campaign life will be reported in three categories:
Operational Practices - The control of the blast furnace process has a major effect on
the life of the furnace. The furnace must be operated in a manner to maximise
furnace life, compatible with production requirements. It will often be necessary to
modify operating practices as the campaign progresses and in response to problem
areas, to maximise the furnace life.
Remedial Actions - Once wear or damage that may affect the life of the furnace
becomes evident, engineering repair techniques must be utilised or developed to
maximise campaign life.
Future, Improved Designs - As improved materials and equipment are developed,
these should be incorporated into future rebuilds to extend the life of critical areas of
the furnace, where it is cost effective to do so.
*
E
2
o
> "
ir
(b >
C
C c
S
(
O C

>
E
5.000
4.000
3.000
2.000
-
ft?
oXV

4
^
I
X
\^

o
'

I
/
/ ^

-
rix
ax
o"
^ o
I , 1
3.0 t P 5j0 GJD
Campai gn ( years )
7.0
Plot
O


3


Blow-out
'70 '71
'72 '73
'TU '75
76 '77
78 79
'80 '81
RELATI ON BETWEEN SERVICE LI FE AND PRODUCTI ON
OF BLAST FURNACES HAVI NG 2000 m
3
OR MORE
I NNER VOLUME I N JAPAN
FIG. 2.1
CJ
=3
CU
"ca
1,000 2,000 3,000 4,000
E f f e c t iv e in n e r v o lu m e ( M
3
)
INNER VOLUME vs CUMULATI VE PRODUCTI ON
OF BF BLOWN I N BETWEEN '65 TO ' 80
FIG. 2.2
JAPANESE B.F. OUTPUT PER UNIT VOLUME
Improvements with time
19
CO
10
<
o
H
8
o_

h 6
D
O
2 4
O
| ?
0
.
. " "

,

75 80 85 90 95
END OF CAMPAIGN (Year)
FIG. 2.3
CAMPAIGN PRODUCTION PER UNIT VOLUME
(B.F.s>1800m3T.V.)

co .

3S
co 10
h
i

I '
(D
I *
O
v
(
-
o
) 5 10 15 2
CAMPAIGN LIFE (Years)
0
Japan
o Europe
FIG. 2.4
3. OPERATIONAL PRACTICES
3.1 Pr oduc t i vi t y
The productivity of a blast furnace can have an important effect on campaign life. It is normally
expressed as an index, in tonnes of hot metal per unit furnace volume per day.
High productivity involves an increased throughput of material at higher burden descent rates, with
an increased hearth activity to remove the larger quantity of molten products. The stability of
operation may be affected when the furnace is being driven hard; burden descent may be less
smooth and the melting zone may be higher in the furnace, both of which may affect wall wear.
The increased throughput of molten products may accelerate hearth wear and present more
arduous taphole conditions.
Low productivity may involve extended periods of low blast volume, which may result in reduced
blast penetration and an increased gas flow up the furnace wall, unless suitable modifications to
burden distribution are made. Long production pauses may have a detrimental effect on hearth
condition.
The above factors will be discussed individually, in more detail, in subsequent sections. However,
this brief discussion of productivity serves to illustrate that there are many interacting factors
involved and it is hardly surprising that opinions differ as to the effect of productivity on campaign
life.
Kawasaki Steel's Chiba No. 6 furnace (4500 m
3
Inner Volume) has been operating since 1977,
with a total production exceeding 52 Mt. During such a long campaign, changing steel demand
and energy prices have dictated levels of productivity between 1.5 and 2.3 t/m
3
I.V./d
(4)
. The
average productivity has been below 2 t/m
3
I.V./d and a productivity of 2.2 t/m
3
I.V./d was only
exceeded for two periods: one period of two years in the early stages of the campaign and one of
6 months in the 14th year of the campaign (during No. 5 furnace reline). The reduced production
levels during the majority of the campaign may have contributed towards stable operation and
long life. As a result of the measures taken, a stable, high level of productivity was possible during
the later stages of the campaign, that did not appear to jeopardise the campaign life.
Nisshin Steel's Kure No.1 blast furnace (2650m
3
I.V.), has been operated for over 9 years of its
campaign (1984-1994) at an average productivity of 2.29 t/m
3
1.V./d. The loss of shaft brickwork in
1987 resulted in high heat loads and stave damage but, after stack grouting in 1988 and upper
stack gunning in 1989, the furnace operation has been stable, with 89 consecutive months
operation with productivity exceeding 2.2 t/m
3
1.V./d
(5>
. Figure 3.1 illustrates this productivity and
shows that the furnace stability in terms of slip index was poorer during the early part of the
campaign. After grouting and gunning, and a slight reduction in productivity, a more stable
operation has been achieved for several years.
Nippon Steel's Kimitsu No. 4 blast furnace (4930m
3
I.V.) produced 38 Mt during its last campaign
(1975-1986), at an average productivity of just under 2 t/m
3
I.V./d. Productivity exceeded the
average value for only two years in the middle of the campaign'
6
'.
Sumitomo Metal Industries' Kashima No. 3 blast furnace (5050m
3
I.V.) had produced 48 Mt in
over 13 years, when it was blown out in 1990. The productivity of the furnace was consistently
around 2 t/m
3
I.V./d for the whole of its campaign
171
. Operation remained stable following the
application of repair techniques and operational improvements.
The average productivity of the present campaign of British Steel's Redcar No. 1 blast furnace
(4308m
3
I.V.) is 2.1 t/m
3
1.VVd. During the early years of the campaign, higher productivities were
being achieved'
8
'. Between the third and fifth years of the campaign, some of the hearth sidewall
thermocouples showed a marked rise in temperature. Corrective action was taken to reduce the
temperatures, and hence hearth wear, with a series of operational practice changes which
included a daily productivity limit of 9200 t/d on the 10000 t/d rated furnace. The current campaign
output is over 30 Mt.
BHP's Port Kembla No. 5 blast furnace (3045m
3
I.V.) was blown out in 1991 after making 25.2 Mt
in a campaign lasting over 12 years. The average productivity was less than 1.9 t/m
3
I.V./d,
although productivities up to 2.45 t/m
3
1.V./d were attained'
9
'.
When considering the level of productivity on blast furnaces that have achieved a long campaign
life, it is clear that the furnaces in question have not been operated to their maximum potential for
the majority of the campaign; despite a requirement for higher output whilst other blast furnaces at
the plant are relined and during periods of high demand. The common factor is stable, consistent
operation, with practices employed to monitor and protect the walls and hearth. Such operation is
more easily achieved at production levels below maximum output. However, it is difficult to define
a universal value of productivity index (t/m
3
/day) to achieve this, since the index is affected by
factors other than furnace driving rate, e.g. furnace internal shape, state of refractory wear, local
operating conditions, maintenance periods, etc.
To maximise campaign life, a strategy is required to enable the furnace to be operated in a stable,
controlled manner whilst producing the required output. Many recent rebuilds have involved
increasing the furnace inner volume, not to increase output but to enable production targets to be
met at lower productivity levels and hence offer the potential for more stable operation and longer
campaign life.
Obviously, frequent stoppages reduce the productivity of a blast furnace, but the campaign life is
also reduced due to the excessive number of stop-start operations. The campaign output per unit
volume is reduced disproportionally to the percentage downtime, as shown in Fig. 3.2
(1
', for
Japanese and European blast furnaces. Long campaigns, measured by this criterion, are best
achieved with continuous blast furnace operation without lengthy stoppages.
Short term reductions in productivity may also be required in response to problem areas identified
on the furnace, in order to protect the integrity of the furnace, thereby avoiding a premature
conclusion to the campaign. These will be discussed in more detail in the relevant section.
3.2 Bur de n
3.2.1 Coke Qua l i t y
For stable blast furnace operation at reasonable productivities, good quality coke is essential. It is
one of the most often cited reasons for a poor period of operation. It is such periods of poor,
erratic, often chilled operations that are potentially destructive to the blast furnace lining and
hence to campaign life'
10
'.
Coke must be strong and stabilised, to support the weight of the burden with minimal mechanical
breakdown. It must be sufficiently large and closely sized, with minimal fines, to create a
permeable bed through which liquids can drip down into the hearth without restricting the
ascending gases. A consistent size is required to avoid undesired variations in permeability and to
support the concept of varying coke layer thickness across the furnace radius to control radial gas
10
flow. The coke must be sufficiently unreactive to solution loss, retain its strength under such
conditions, and be low in alkalis to minimise alkali gasification in the raceway, which has a
deleterious effect on coke breakdown and on furnace refractories. A low sulphur content is also
required to minimise hot metal sulphur and desulphurising costs. Coke moisture and carbon
content variations must be controlled to minimise their effect on the thermal state of the process.
At high levels of tuyere hydrocarbon injection, used to reduce ironmaking costs and increase
productivity, there is a corresponding reduction in the proportion of coke charged and
consequently coke quality becomes even more important.
A universal coke quality, for stable operation compatible with long blast furnace life, is difficult to
specify, since not only do different types of operation have different coke requirements but also
physical properties vary according to the point of sampling between the coke ovens and the blast
furnace. A typical specification for a large blast furnace is given in Table 3.1
(10)
.
It is often the case that the coke charged to a blast furnace is sourced from more than one coke
plant. In these cases, a system of blending adequately or charging discretely the different cokes is
essential, as fluctuating proportions of cokes of different properties may result in unstable furnace
conditions
111
'.
Coke in the furnace centre will gradually replace the deadman and the coke in the hearth, which
must remain permeable to allow the liquids to drain across the centre of the hearth. This avoids
excessive peripheral flow of hot metal in the hearth which can result in severe refractory wear at
the base of the sidewall. An increase in the hearth pad centre temperatures has been observed
with an increase in deadman coke size (Figs. 3.3 - 3.5
(10

12,13
'), which indicates increased hearth
centre activity. The aperture size of the coke screens is an important parameter to maintain hearth
permeability. It is usually beneficial to increase the screen size and charge the additional small
coke arising, mixed in with the ferrous burden, away from the furnace centreline. Further details
will be discussed in a later section.
The aim of specifying high quality coke is to ensure that large coke reaches the lower regions of
the furnace. To monitor this objective in the long term, it is advisable occasionally to sample coke
from tuyere level to assess the coke breakdown through the furnace. This is usually carried out
during planned maintenance, often in conjunction with tuyere changes. A large sample of coke is
raked from a tuyere aperture and its properties compared with a sample of the corresponding feed
coke'
14
'. In this way other factors affecting coke size can also be identified. Many operators also
use probes through the tuyere aperture to obtain information on coke properties in the lower
regions of the furnace.
Good, consistent quality coke and the monitoring of both stockline and bosh coke is clearly an
important strategy for long campaign life.
3.2.2 Fe r r ous Bur de n Compos i t i on
Blast furnaces may be operated with a wide variety of ferrous burden components, the main ones
being sinter, pellets and lump iron ore. A variety of fluxes may be used to trim the final chemistry.
Smaller proportions of other iron bearing materials are often used, which may include recovered
scrap, ferrous fines, mill scale, BOF slag, ilmenite, recycled waste or even directly reduced or
granulated iron. The burden of an individual plant usually depends on local factors and on burden
costs.
Integrated steelmaking sites usually have sinter making facilities, resulting in these blast furnaces
operating on a large proportion of sinter, with the balance of the burden consisting mainly of lump
11
ore and/or pellets. This is the case for the long life blast furnaces mentioned in Section 3.1, which
operated with 55-90% sinter'
4
'
6,7

8,9
'. Traditionally, the balance of the burden was mainly pellets,
due to their superior properties. In recent times, more sophisticated blast furnace operation has
enabled increasing proportions of directly charged lump ore to be used, reducing burden costs.
Lump ore has been used at up to 45% of the ferrous burden at British Steel's 14 m Redcar
furnace'
8
'.
Recent environmental concerns on emissions from sinter plants could result in older sinter plants
being forced to close on economic grounds'
15
'. This situation necessitates an increased proportion
of directly charged materials. Operation with a large proportion of pellets and lump ore may also
be necessary when operating a standby blast furnace to satisfy short term increases in steel
demand.
Operations with up to 100% pellets are encountered throughout the world, particularly in North
America and Scandinavia, where indigenous ores suitable for pellet manufacture are plentiful.
Past experience has shown that blast furnaces burdened with a high proportion of pellets suffer
higher heat load variations in the lower stack and bosh'
15,16
', leading to excessive lower stack and
bosh wear and a shorter campaign life'
17)
. One of the main reasons for this has been inadequate
control of burden distribution. Pellets have a much lower angle of repose than sinter or coke and,
when landing on an inclined stockline, tend to roll easily. This may result in a relatively thick
ferrous layer towards the furnace centre which will encourage excessive gas flow at the furnace
wall.
This situation has improved in recent years, with the addition of high density cooling in the lower
shaft and improved burden distribution equipment. BHP's Whyalla No. 2 blast furnace (1543m
3
I.V.) had achieved a campaign output of 7970 t/m
3
I.V. by 1993 on an 80% pellet, 20% lump ore
burden'
18
'. The proportion of pellets was increased from 35-70% at Kobe Steel's Kakogawa No. 2
blast furnace (3850m
3
I.V.) in 1991. Fluctuating lower stave temperatures, increased slipping and
hot metal temperature fluctuations were initially observed, but stable operation was recovered by
burden distribution control with centre coke charging and addition of small coke to the pellets'
19
'.
Hoogovens carried out trials with a 100% pellet burden and found that productivity, fuel rate, metal
quality and heat losses were worse than with the usual 50% sinter burden, although control could
be maintained satisfactorily up to 80% pellets, with the sinter charged in the wall area'
15
'.
An important aspect to consider when selecting individual burden components is their softening
and melting characteristics. The major part of the pressure drop across a blast furnace is in the
region where the ferrous burden is softening, melting and dripping down the coke bed through
which the gases are ascending. A wide melting and softening range may result in an increased
pressure drop and a large cohesive zone root impinging on the lower shaft brickwork, with the
refractories being exposed to high temperatures over a wider area than is desirable. For example,
the reducibility and melting properties of fluxed pellets are superior to acid pellets. Dofasco,
operating with a 100% pellet burden, reported a decrease in mid-stack wall heat losses when
replacing 75% acid pellets with fluxed pellets, resulting from their improved high temperature
properties'
20
'. BHP reported that a change from 80% acid to 80% dolomite-fluxed pellets at
Whyalla No. 2 blast furnace decreased the thermal loading on the lower stack walls and resulted
in improved stability of operation'
21
'. This suggests a lower wall temperature and/or fewer thermal
fluctuations, both of which would help to prolong the life of the shaft brickwork. The melting and
softening properties of a multiple component burden will differ from those of the individual
components and it is desirable to consider softening and melting test data not only of individual
burden constituents but of the proposed mixture, to aid burden selection'
22,23
'.
12
To minimise thermal and chemical variations, a homogeneous burden is desirable. The burden
components should be as intimately mixed as possible. This depends on the number of burden
components and the individual charging system, but it can usually be achieved to a reasonable
degree by selection of storage bunkers and the sequence of material discharge.
Clearly, it is possible to achieve stable furnace operation and long campaign life using different
burdens providing that suitable consistent quality materials, adequate wall cooling capacity and
distribution control are available.
3.2.3 Fe r r ous Bur de n Qua l i t y
To ensure a permeable blast furnace, essential for stable operation, it is important that the ferrous
burden is strong, closely sized and efficiently screened to remove fines. It must not disintegrate
excessively in the stack, which would generate additional fines, and should be sufficiently porous,
reducible and of an appropriate size to allow the material to be adequately reduced by the time it
reaches the softening zone. In this way the cohesive zone will be less restrictive, with less FeO
rich slag, and the thermal load in the lower regions of the furnace will be lower, encouraging
smooth operation. Requirements on the physical and metallurgical properties of sinter and pellets
for efficient operation are discussed in more detail in reference'
24
', and summarised in Table 3.2'
24
'.
The softening and melting properties of the ferrous components have an important effect on blast
furnace operation. Restrictions in the cohesive zone and poor melting characteristics may result in
erratic burden descent, unstable operation and thermal fluctuations; conditions likely to shorten
furnace wall life. Many operators determine the softening and melting properties of ferrous
materials in laboratory tests; details of the British and German versions are given in
references'
22,23
'. There is no standardised softening and melting test and there are many indices
quoted to represent the softening and melting temperatures; start of direct reduction, pressure
drop during melting, quantity of dripped material, etc. British Steel have found that, for production
sinters, the quantity of material retained in the coke bed after dripping ('residual material',
essentially high melting point slag) can be related to blast furnace performance and stability. A low
quantity of residual material is desirable for good furnace operation'
25
'.
3.2.4 Ch e mi s t r y : Al ka l i s , Zi n c
It is well known that alkali metals and zinc have a deleterious effect on the blast furnace process
and refractories, and their input in the burden is maintained at a minimum economic level.
Although generally charged at levels of less than 5 kg/thm (tonne hot metal), by the condensation
of alkali vapour on the descending burden a large recirculating load can build up in the furnace.
This results in increased sinter degradation'
26
' and coke breakdown'
10,14
', and encourages the
formation of wall accretions, all of which can result in irregular burden descent and unstable
furnace operation.
Alkali metal compounds and zinc, in gaseous form, will penetrate cracks and pores in the furnace
wall refractories. The resultant chemical attack and thermal cycling weakens the surface layer of
refractory, which is eventually removed by the descending burden, allowing the process to be
repeated.
Hearth dissections after the end of a campaign have shown that excessive wear may occur at the
base of the sidewall and that a brittle zone is often formed between the shell and the hot face of
the carbon. Zinc and alkalis have often been found at high levels in this brittle zone and various
breakdown mechanisms have been proposed involving these compounds'
1
'. Stress and thermal
cracking in the sidewall will allow gaseous zinc and alkalis to penetrate and be deposited in the
13
pores, which can lead to brick expansion, embrittlement, further swelling and ultimately
destruction of the refractory mass'
16
'. A significant degree of refractory protection from alkalis and
zinc may be achieved if an accretion or skull is frozen on the hot face of the refractory, thereby
protecting the refractory from chemical attack'
16
'. However, zinc and alkalis are not necessarily
involved in the development of a brittle layer, since minimal levels were found in the brittle zone on
Nippon Steel's Muroran No. 3 blast furnace, which was blown out after 1 year and 6 months of its
sixth campaign'
27
'.
The actual level of alkali charged depends on economic and political constraints, with best
practice being around 2 kg/thm. The majority of the alkalis are removed in the slag and the
remainder in the top gas. However, the furnace slag practice, thermal state and burden
distribution play a major role in alkali removal'
25
'. A reduction in slag basicity will increase the
quantity of alkali removed in the slag (Figs. 3.6, 3.7)
(26)
, s will an increase in the thermal level of
the furnace or in the top temperature, by broadening or intensifying the degree of central
working'
28
'. In addition, for a given alkali loading, coke degradation is likely to be greater for
operations with a high rate of tuyere hydrocarbon injection, due to the increased burden residence
time'
29
'.
It is important that the balance of input and output alkali and zinc is monitored and that a furnace
is operated with a thermal and chemical regime compatible with the input level of these elements,
to encourage their removal in the slag and top gas.
3.2.5 Bur de n Di s t r i but i on
Burden distribution is one of the main factors affecting blast furnace campaign life. Not only can it
affect the stability of operation but, by determining the radial gas flow in the furnace, it is one of
the major factors controlling the rate of wear of the furnace walls.
The two types of distribution system that enable sufficient control for high productivity and long
furnace campaign life are the bell-less top using a tiltable rotating chute, most commonly that of
Paul Wurth design, and a bell charging system with movable throat armour.
Primarily, radial gas flow is controlled by the proportion of ferrous burden to coke, since coke is
generally much larger in size. This is most easily achieved by charging the material in discrete
layers and varying the layer thickness across the furnace radius. Protection of the furnace walls is
therefore achieved by increasing the proportion of the ore layer at the wall, which will result in a
reduced quantity of heat removed by the wall cooling system. An example of this is shown in
Fig. 3.8, for Kawasaki's Chiba No. 6 blast furnace'
4
'. However, there is a limit to the proportion of
ferrous material close to the furnace wall, otherwise an inactive layer will form, which may
encourage the formation of wall accretions and allow unprepared burden into the lower regions of
the furnace and increasing tuyere losses. The proportion of coke at the centre of the furnace must
be sufficient to allow stable furnace operation at the desired level of production. A large proportion
of coke creates a relatively permeable region with fewer descending liquids, allowing the use of
maximum blast volume without large fluctuations in blast pressure and erratic burden descent.
The coke at the centre of the furnace replaces the coke in the hearth and a coke rich permeable
centre will encourage a permeable hearth, as shown in Fig. 3.9
(4)
, which relates the liquid flow
across the hearth, as indicated by the hearth pad thermocouples, to the upper shaft probe etaCO
at the furnace centre. The central coke chimney should not be unnecessarily wide, however, or
inefficiency will result, and damage may be incurred to certain parts of the furnace top due to
excessively high heat capacity of the ascending gas.
14
3.2.5.1 Spl i t Si ze Ch a r gi n g
More sophisticated distribution systems permit additional control of burden distribution by utilising
more than one size range of a given material. One of the most commonly used practices is the
charging of fine ferrous materials, often from screenings of the main ferrous burden. Fines are
charged separately in small quantities close to the furnace wall, to give a localised reduction in
permeability and thereby protect the walls, as shown in Fig. 3.10'
30
'. Charging a separate small
batch of finer material will usually reduce the charging capacity of the furnace. At Kawasaki's
Mizushima No. 3 blast furnace, a 3-parallel top hopper system was incorporated at the 1990
rebuild to overcome this problem. Charging of small batches with a bell and movable throat
armour system should cause fewer delays than with a bell-less top due to the reduced discharge
time, as was achieved at NSC's Sakai No. 2 blast furnace (Fig. 3.11'
31
'). It may be possible to
charge small quantities of finer materials to the furnace wall by charging them first into the top
hopper or large bell hopper and using the corresponding initial chute angle or movable throat
armour setting. However, the quantity is limited by the hopper discharge characteristics to that
which will pass through the hopper without mixing with the remainder of the charge. There is also
a financial benefit in using such ferrous fines directly as opposed to returning them to be
re-sintered.
In a similar manner, the ferrous burden may be split into large and small sizes which are then
charged over different parts of the furnace radius to control the radial permeability. Such split
sized sinter charging has been practised at POSCO's Kwangyang furnaces since the mid-1980's,
where the 4-12 mm sinter is charged towards the wall and the 12-50 mm sinter towards the
centre, although the cut size is adjustable for maximum distribution flexibility'
32
'.
3.2.5.2 Coke Nut s
A flexible charging system will allow the use of coke nuts. The size of coke nuts available for
charging will depend on the size and efficiency of the blast furnace coke screens, the company
coke requirements and the marketability of small coke, but is typically in the range 10-30 mm. The
charging of nuts, mixed in the ferrous material and positioned along the mid-radius, often
improves operation by improving reduction efficiency and permeability of the ore layer in the
cohesive zone'
33
'. POSCO reported improved permeability and reduced belly temperatures with
nut coke charging, as shown in Fig. 3.12'
32
'. At NSC's Sakai No. 2 blast furnace, the nut coke was
charged at the wall, sandwiched between the two ore charges (Fig. 3.11), to prevent an inactive
wall region when fine ore was charged at the wall. At Kobe Steel's Kakogawa No. 2 blast furnace,
fitted with movable throat armour and using a burden of up to 70% pellets, coke nuts were added
to the pellets to increase their angle of repose, Fig. 3.13, thereby reducing the proportion of
ferrous material at the furnace centre'
19
'. The replacement of prime coke with coke nuts also
reduces fuel costs.
3.2.5.3 Si ze Se gr e ga t i on
Most charging systems create some degree of size segregation in the input materials. If the initial
material to discharge is finer and the final material is coarser, this characteristic may be utilised to
benefit the radial size distribution, and hence the radial gas flow distribution. This type of
segregation generally occurs on belt charged furnaces rather than on skip charged furnaces and
is more controllable with a bell-less top'
34
'. Suitable modifications may also be added to the
charging system to enhance the desired segregation characteristics, such as segregation control
plates in the top hoppers at Mizushima No. 3 blast furnace, shown in Fig. 3.14'
35
'.
15
Additional radial size segregation may also occur by rolling down an inclined stockline. Size
segregation may also modify the melting and softening characteristics of the burden along the
furnace radius, when one component has a different size range and chemistry.
Some charging systems may result in a circumferential variation in burden distribution. These
variations must be minimised by design or operation. For a parallel hopper bell-less top, this may
be achieved by redesign of the lower chute and downcomer or by periodically alternating the
material type charged from each bunker'
30,36
'. In addition, the starting position of the rotating chute
should be varied for each charge'
35
'. With bell charged furnaces fed by skips, it is essential that the
distributor operates to minimise any circumferential maldistribution, resulting from fines
segregation on the bell. Improvements were made at BHP's Whyalla No. 2 blast furnace when the
operation of the distributor was changed from indexing to continuous rotation'
17
'.
3.2.5.4 Ce n t r e Coke Ch a r gi n g
A large proportion of coke is often required at the furnace centre, to encourage sufficient centre
working for stable operation. This is particularly so at higher productivities and when operating
with high levels of tuyere hydrocarbon injection. However, to operate with entirely coke in the
centre of the furnace is less fuel efficient and techniques have been developed to minimise the
width of this region, by centre coke charging. On a bell-less top, this is achieved by charging a
small batch of coke with the rotating chute fully lowered, resulting in layers as shown in
Figs. 3.15'
32
' and 3.10'
30
'. A furnace with movable throat armour (MTA) has less control of the
layer shape, since all of the falling material has to be deflected by the MTA, which is located near
the periphery of the furnace. Consequently, centre coke layers such as those illustrated are not
normally possible. To overcome this deficiency and enable centre coke charging on MTA charged
furnaces, a technique was developed by Kobe Steel at Kakogawa No. 2 blast furnace, to charge
additional coke to the furnace centre down an inclined chute located between the large bell and
the MTA. The arrangement is shown in Fig. 3.16, and the reductions in furnace pressure drop,
blast pressure fluctuation and slip frequency with increasing quantities of centre charged coke in
Fig. 3.17'
13
'.
A permeable coke bed is necessary in the hearth, to encourage the flow of liquids across the
centre of the hearth and reduce peripheral flow, which can cause excessive sidewall wear. The
coke in the deadman and hearth is gradually replaced by coke from the furnace centre. Centre
coke charging reduces the proportion of ferrous material at the furnace centre and improves
hearth permeability, as indicated by increasing hearth bottom temperatures, Fig. 3.5. Larger,
stabilised coke charged to the centre is likely to further improve hearth permeability.
3.2.5.5 Th r oa t Ar mour Li fe
For long campaign life, it is important to minimise wear on the fixed throat armour caused by the
direct impact of burden materials. Although it is possible to repair the throat armour or incorporate
protection plates, this may involve long maintenance stoppages, which themselves may be
detrimental to furnace life. Consequently, the burden distribution and the stockline height used
should be chosen to avoid such burden impact.
3.2.5.6 Mon i t or i n g
Burden distribution must be monitored regularly to ensure the correct degree of wall protection
and a stable, driving furnace. Changes in furnace operating parameters, e.g. changes in tuyere
hydrocarbon injectant rate or blast volume, may require adjustments to burden distribution. The
effect of burden distribution is monitored with various probes and instruments, which will be
16
discussed in Section 3.4. The use of a predictive mathematical model of burden distribution that
has been calibrated from plant measurements is highly desirable, for assessment of existing
distributions and prediction of future changes.
To maximise the campaign life, it is essential that the charging equipment is capable of accurate
control of burden distribution, that adequate instrumentation is fitted to comprehensively monitor
the resulting furnace operation and that the burden distribution is changed and assessed in a
controlled, technical manner.
3.3 Ti 0
2
Addition
For many furnaces worldwide, samples of the hearth lining at the end of a campaign have been
found to contain titanium bearing deposits. These form a protective layer in eroded regions of the
hearth sidewall, in the salamander and in brick pores and joints. The titanium is usually in the form
of carbonitrides, a solid solution of titanium carbide and titanium nitride. Consequently, practice in
recent decades has involved the introduction of additional titania (Ti0
2
) into the furnace to
promote these protective layers. Three methods of introducing Ti 0
2
are by addition to the burden,
injection at the tuyeres or by addition to taphole clays.
3.3.1 Introduced in Burden
Probably the simplest and most common technique has been by the addition of titaniferrous ores,
commonly ilmenite, to the burden'
38
'. Alternatively Ti 0
2
can be incorporated in sinter, although
usually at low levels. This technique has been more prevalent in Japan, and the Ti 0
2
source is
usually in the form of iron sand. Operators using this technique have experienced sinter quality
problems, with a deterioration in cold and hot strength and a reduction in sinter productivity'
39,40,41
'.
Usinor reported an increasing reduction degradation index (RDI) with increasing Ti content in the
sinter, Fig. 3.18'
39
', which was also experienced at NKK's Fukuyama Works, Fig. 3.19'
40
'. This
figure shows that, although the increase in RDI reached a peak, the sinter productivity continued
to fall, with a reduction of 30% at iron sand additions equivalent to a blast furnace addition rate of
10 kg TiOj/thm. BHP Port Kembla reported a deterioration in sinter quality with ilmenite sand
addition of above 0.5% Ti0
2
'
41>
. It is also possible to incorporate Ti 0
2
during pellet manufacture.
Kobe Steel has practised this with a pellet Ti0
2
of 1.7%, equivalent to a level of 9 kgTiO/thm in a
25% pellet burden'
38
'.
There tend to be two strategies concerning the method of burden Ti 0
2
addition. One is remedial,
commencing Ti 0
2
additions only once high hearth temperatures are observed, indicating hearth
wear. The other takes a preventative approach and adds a small quantity of Ti0
2
continuously,
increasing the addition level if high temperatures are observed. The Ti 0
2
intake for the preventive
approach is generally 3-5 kg/thm, which usually results in up to 0. 1% Ti in the hot metal and
1-1.5% Ti 0
2
in the slag. For remedial action, the Ti 0
2
dosage may be up to 20 kg/thm, at which
level the hot metal may contain up to 0.3% Ti and the slag up to 3.5% Ti0
2
. This creates operating
problems due to high slag viscosity and scaffolding in the runner, and consequently such high
Ti 0
2
levels can only be used for short periods. An increased quantity of titania in granulated or
pelletised blast furnace slag will affect its properties as a cement additive and may decrease its
value and marketability. At high levels of titania addition, steelmaking difficulties may be
experienced. Titanium carbonitrides can also remain in the final steel product, as unwanted
inclusions.
At NKK's Fukuyama No. 5 blast furnace, Ti 0
2
addition was used to overcome localised increases
in sidewall temperature (Fig. 3.20)'
42
'. Ti 0
2
addition was increased up to 20 kg/thm, with a Ti 0
2
balance confirming the retention of up to 1.5 kg/thm. The slag basicity was increased from 1.2 to
17
1.3 to counterbalance the potential slag viscosity problem. At Sumitomo Metal Industries' Kashima
No. 3 blast furnace, Ti0
2
additions were increased to 10-15 kg/thm, after 9 years operation, to
control hearth sidewall temperature (Fig. 3.21)'
43
'. At Wuhan Iron and Steel Co., 2-3 weeks after
charging ilmenite (10.5% Ti0
2
) at up to 10 kgTiOg/thm, the heat load on the hearth staves started
to decrease and after 2-3 months it became stable (Fig. 3.22)
(44
'. In 1981, British Steel's Redcar
No.1 blast furnace reported a decrease of 120C in hearth sidewall temperatures, following an
increase in Ti 0
2
level to 8 kg/thm, which was interpreted as evidence that a protective layer had
been formed on the eroded area'
45
'. At Altos Homos De Mexico, ilmenite was charged at the rate
of 9 kg TiO^thm'
46
' and within two weeks a reduction in hearth temperature of 180C was
observed.
In order to promote the precipitation of Ti(C,N), some operators increase the Ti 0
2
additions before
a shutdown so that the metal remaining in the hearth will be saturated in Ti. As the hearth cools
during the shutdown, this encourages precipitation'
41
'. Others find that the resumption of
production is more difficult at high Ti levels, creating operational problems.
3.3.2 Tuye r e In je c t e d Fi n e s
Several operators have injected fines containing Ti 0
2
through the tuyeres, as an alternative to
Ti 0
2
additions to the burden'
5,47,48
'. Kobe Steel was developing a system at Kakogawa Works from
1984, to increase hot metal viscosity and deposit Ti(C,N) on the hearth sidewall'
47)
. The
advantages quoted include: application at localised positions, reduced cost due to lower Ti 0
2
rate,
good results from short-time injection and unchanged burden properties. When benefits were not
achieved by injection directly above the area of high sidewall temperatures'
49
', model experiments
were carried out to determine the most suitable tuyeres through which to inject, relative to the
position of the damaged area and the taphole in use. Two tuyeres were used for higher injection
rates and to improve protection. The injection position at No. 3 blast furnace (in 1985) is shown in
Fig. 3.23, and the reduction in hearth sidewall temperatures achieved is in Fig. 3.24. The injection
position at No. 2 furnace (in 1991) is shown in Fig. 3.25, and the reduction in hearth sidewall
temperatures achieved is in Fig. 3.26'
49
'. Rutile powder (95% Ti0
2
) was used in the earlier trials,
but latterly ilmenite powder (50% Ti0
2
) was successfully used, to reduce costs. Selection of the
correct injection location is important, with the angle between the damaged portion and the
injection point being 40-90, and the angle between the injection point and the taphole in use
being 80-160
0
'
49
'. Nisshin Steel have injected iron sand at up to 15 kg/thm, claiming a reduction in
hearth brick temperature over a longer period (Fig. 3.27)'
5
'. ILVA injected ilmenite and rutile at
Taranto No. 5 blast furnace, reducing hearth wall temperatures (Figs. 3.28 and 3.29). The
injections took place through a tuyere approximately 150 from the taphole in use to protect an
area 90 from the taphole'
48
'.
3.3.3 Impr e gn a t e d Ta ph ol e Cl a ys
The third method of Ti0
2
addition is by the use of taphole clay containing Ti0
2
. This has been tried
in Japan. Although few details have been published, one such clay tried was tar bonded with
approximately 10% Ti0
2
. Clearly, the titania is bound in the taphole clay in an unreduced form,
and is injected in relatively small quantities. It is questionable whether it will be reduced and
dissolved in the iron in sufficient quantities to be precipitated or whether it would be reduced and
bonded adequately to the hearth sidewall to be of benefit.
18
3.3.4 Suc c e s s a n d Cos t Effe c t i ve n e s s
In a blast furnace, Ti0
2
is partially reduced and is dissolved in the hot metal. The solubility is
greater at higher temperatures. If the Ti in the hot metal is nearing saturation and the refractory
hot face temperature in eroded regions, cracks and pores is lower than the hot metal temperature,
then Ti may be precipitated, as Ti(C,N). The technique is more likely to succeed at higher addition
rates, but there are other factors which may interfere with this basic mechanism, including thermal
state of the hearth, metal/slag chemistry and liquid flow characteristics.
Success in overcoming hearth hot spots with Ti 0
2
additions is varied. Not only will the individual
furnace conditions vary, which will affect the success of the technique, but Ti 0
2
addition is usually
carried out in conjunction with other remedial actions (which are discussed in other sections) such
as reducing productivity, closing tuyeres and improving hearth cooling intensity. For example,
Fig. 3.30'
42
' illustrates the other actions taken at Fukuyama No. 5 blast furnace, although positive
effects are claimed to result from Ti 0
2
addition. The direct effect of Ti 0
2
addition is therefore often
difficult to determine. It is essential to carry out regular, accurate Ti balances to assess the
technique and modify operation to encourage Ti retention. The effect of high rate additions may
even have a detrimental effect on furnace operation, negating any benefits. The delay following
additions, before an observed fall in hearth temperatures, can vary from hours to weeks, as can
be seen in Figs. 3.20 - 3.29.
Once initial hearth wear has stabilised, in the early stages of a campaign, the addition of a small
quantity of Ti 0
2
over a long period of time would seem a sensible preventive measure, depending
on the cost of the source of Ti0
2
. It has been calculated'
38
' that, provided a campaign is extended
for several months by charging ilmenite at a preventive level for 6 years followed by an increased
rate thereafter, it will prove financially beneficial.
The addition of Ti 0
2
for hearth protection should be considered as part of a hearth protection plan
rather than in isolation. Its use should be considered in relation to success achieved by other
techniques, and different methods of its use may be required to achieve the desired benefits.
3.4 In s t r ume n t a t i on a n d Con t r ol
3.4.1 He a r t h Th e r moc oupl e s
Early warning of hearth problem areas is vital to maximise campaign life, and thermocouples
located in the hearth sidewall and in the hearth pad are absolutely essential to monitor hearth
wear. Revised operating practices and actions to protect the hearth will be taken as a result of
increasing hearth temperatures. Hearth pad and sidewall temperatures can also give an indication
of liquid flow in the hearth, Figs. 3.3 - 3.5, an important factor in hearth wear.
Temperatures recorded by thermocouples are influenced by only a small area round the
thermocouple. It is therefore vitally important to locate the thermocouples in the critical wear
areas. Of particular importance are those areas below the tapholes and around the base of the
sidewalls where the so called 'elephant's foot' wear pattern is often found. An adequate number of
thermocouples must be installed, in the best layout to give as complete coverage as is practical.
At several locations, thermocouples should be positioned at two (duplex) or three (triplex) different
depths to allow calculation of the thermal profile in the refractory and hence the thickness of
residual refractory.
19
Sollac's Dunkerque No. 4 blast furnace (14 m hearth) was equipped with 126 thermocouples in
the hearth during its reline (Fig. 3.31 )'
50
'; 26 in the pad on two levels, 72 in the sidewall at three
levels (50% duplex) and 7 round each of the four tapholes. ILVA's Taranto No. 5 blast furnace
(14m hearth) has a total of 116 thermocouples in the hearth, at three different levels'
51
', which was
subsequently increased to a total of 151. Alto Homos de Mexico's No.5 blast furnace (11.2 m
hearth) has 72 thermocouples at six different levels (Fig. 3.32)
(46)
. Chiba No. 6 blast furnace has
81 thermocouples on six levels, Fig. 3.33'
4
'.
The insertion depth of the hearth sidewall thermocouples is important and will depend on whether
a hearth is stave or spray cooled. Thermocouples are usually inserted to the front face of staves
or shell, in front of the conducting ramming between the staves or shell and the carbon/graphite,
and a short distance into the carbon/graphite'
52
'. If the thermocouples are positioned too close to
the hot face, they may deteriorate in the early stages of campaign'
53
'. Ni-Cr/Ni-AI thermocouples
can become poisoned at temperatures over 600C, thereafter indicating too low a temperature'
54
'.
Movement of carbon blocks can nip hearth pad thermocouples, causing false hot junctions or total
failure. These latter problems may be overcome by fitting the thermocouples in sheaths.
Thermocouples should be positioned around the tapholes, to monitor taphole conditions and
operation.
Additional thermocouples are often added part way through a campaign in areas of known
refractory wear, to give a more localised picture of developing problems'
4,42,48
'. Similarly,
thermocouples are often added to repaired areas to monitor the repair.
As an alternative to thermocouples, heat flux probes are currently under development, with 12
installed in the hearth brickwork at Bremen No. 2 blast furnace'
55
'. They consist of 20 fine
thermocouples, mounted on a ceramic carrier which is matched to the thermal conductivity of the
lining. Results indicate that they give more accurate information on refractory wear than
thermocouples, and a spacing of 3 m is adequate as opposed to 1m for thermocouples. Further
details are given in Section 5.4.
3.4.2 Mon i t or He a r t h Cool i n g
Heat flux in the hearth pad or stave cooling water can be determined from the water flow rates
and the difference between inlet and outlet water temperature, using resistance thermometers. It
can be used only to give an indication of the average hearth wear and is therefore of less use than
hearth thermocouples. It is particularly applicable in the later stages of a campaign, following
thermocouple deterioration'
53
'.
Monitoring long term trends in hearth cooling water temperature may give an indication of the
efficiency of the cooling system. Spray cooled hearths may require periodic cleaning of the shell'
42
'
and staves/underhearth cooling may require blast cleaning'
107
' or removal of pipework deposits
using acid, especially if using untreated water.
3.4.3 Fur n a c e Wa l l Con di t i on s
The process conditions at the furnace wall are vital to campaign life. The walls must not be
subjected to high heat loads from an excessive quantity of gas ascending at the wall or
impingement of the melting zone on the wall, which would result in more rapid deterioration of the
refractory and wear of the cooling members. On the other hand the walls must not be so inactive
that large accretions are permitted to form on them, which would prevent smooth burden descent,
20
control of burden distribution and stable blast furnace operation. To monitor wall conditions a
variety of techniques are used.
The most common technique of monitoring the walls is using inwall thermocouples, positioned in
the brickwork, with the tips a short distance back from the hot face to give a good thermal
response. Wall activity is monitored from the temperature level and fluctuations. Thermocouples
are often fitted in the body of staves. Whilst a refractory lining is present, the stave temperatures
have a low value and a damped response. Their temperature rise and increasing fluctuations
indicate when the refractory is lost and thereafter they function as inwall thermocouples.
There should be a good coverage of thermocouples both vertically and circumferentially to monitor
the walls adequately'
6,51
'. Seven levels of thermocouples, each with eight circumferential positions,
is typical. With a large number of thermocouples it is difficult for the operator to monitor the
variation of them all. By using the temperatures at many points, an isothermal map can be
generated, identifying regions of high or low temperatures which may relate to refractory wear,
asymmetrical operation or accretion formation. The dynamic temperature behaviour can also be
utilised to predict the formation or loss and extent of an accretion'
56
"
57
'.
Throat or skin thermocouples are often installed around the periphery, just below the fixed throat
armour. The thermocouple tips are installed level with the hot face of the refractory, to record gas
temperature. These give a direct measure of the gas flow at the wall and are usually unaffected by
deposition of material, unlike inwall thermocouples lower in the stack.
The heat load at the walls, or heat flux, is often calculated from the temperature rise and flow rate
of the shaft cooling water. This usually gives the total heat flux for the stack or stack and bosh,
and is considered a good indication of burden distribution, gas flow and wall wear (e.g. Figs. 3.34,
3.35 and 3.36)'
4,8,58
'. If the cooling system is designed with a series of ring mains serving banks of
a small number of coolers or staves, it is possible to determine the heat load profile down the
shaft, which may relate to different types of operation. British Steel's Port Talbot No. 4 blast
furnace, which is plate cooled, has such a system with 13 zones of heat flux'
59
'.
There are also systems for determining the remaining thickness of wall refractory, besides the
basic method of drilling and measuring. These usually involve embedding sensors in the brickwork
and periodically measuring their remaining length. They can only measure the remaining
thickness of the original brickwork and not the build up of a protective layer on the hot face as can
be inferred from thermocouple readings. At Hoogovens Umuiden Works, lining wear is measured
ultrasonically by ceramic rods incorporated in the refractory. The rods wear with the lining and
their length is measured during furnace stops'
60
'. Other systems used include; time domain
reflectometry, where a pulse is sent down a signal cable, the time taken for the reflection giving
the length of the remaining cable'
53,61
'; a system using a series of fine thermocouples in a single
sheath, the innermost thermocouples failing as the refractory wears back'
62
'; and a similar system
using a series of elements, with lining thickness relating to electrical resistance'
63
'. Infra-red
thermography may also be carried out on the furnace shell to indicate the refractory wear
pattern'
52
'. It is important to monitor the lining thickness as it not only protects the shell directly in
uncooled regions of the periphery or where cooling members have failed, but a worn lining in the
upper stack may result in layer mixing and reduce the control of burden distribution at the wall.
This will ultimately result in increased thermal loads at the wall and affect the stability of operation,
which will accelerate wall wear.
21
3.4.4 Ra di a l Me a s ur i n g Pr obe s
It is accepted worldwide that the use of retractable probes is one of the most important techniques
to monitor and optimise burden distribution, and hence campaign life. Such probes are the only
method of measuring the variation in operating characteristics along the furnace radius, as
opposed to relying solely on wall measurements. They are essentially of two types: overburden
and underburden. A schematic arrangement of such probes on a furnace is shown in Fig. 3.37
(64)
.
Overburden probes have many functions. The simplest type is usually fixed, water cooled and
measures the radial or diametrical top gas temperature profile and, in some instances, the gas
analysis. Most retractable probes measure the stockline layer profile and may be of a mechanical
type, where a weight is lowered to the stockline'
65
' or a non-contact type, using radar'
66
',
microwaves'
64
', lasers'
9
', etc. The radial variation in layer thickness can be measured and the
ore/coke ratio across the radius calculated. This can be related to the charging sequence and
used to calibrate burden distribution models. It is of less value if layer sequences which encourage
the disturbance of a coke layer by the subsequent ferrous material are in use. Top gas velocity
can also be physically determined to measure the quantity of gas flow'
65
', and top gas analysis and
temperature measurement will often be carried out in conjunction with the other functions. Probes
are also used to determine the trajectory of material off the rotating chute or movable throat
armour, for calibration of burden distribution predictive models and to determine the effect of
charging chute wear (Fig. 3.38)'
36
'. For experimental purposes, probes have been constructed to
introduce feeding type vertical probes at different radial positions, to determine temperature and
gas analysis as the burden descends, since the more usual type of vertical probe is only possible
at the furnace periphery.
Underburden, or in-burden, probes sample gas and measure temperature at a number of radial
positions. They are generally positioned in the upper stack, typically 3-6 m below the stockline,
although some furnaces have a lower stack probe which may be 10-15 m below the stockline.
These probes are generally of two types. The consumable type, is typically 50 mm in diameter,
bends with the descending burden and is straightened on withdrawal for subsequent re-use'
36
'. An
example of the other type is the DDS probe, which is oval in section, typically 550 mm by 210 mm,
water cooled and resists bending whilst in the furnace'
66
'.
Since the top gas has to pass from the stockline up one of the four offtakes, the gas flow pattern
begins to distort near the stockline. A large degree of gas mixing then occurs above the burden,
and overburden probes must be positioned close to the stockline, and preferably inclined, to give
acceptable temperature and gas profiles. The consensus of opinion is that upper stack
underburden probes are more sensitive and give superior results to overburden probes. In
addition, fixed overburden probes can be quite large and, depending on the stockline height, can
create a 'shadow' and distort the burden distribution beneath them, which may create
unrepresentative results.
Typical gas efficiency and temperature profiles for a furnace operating at good productivities
compatible with a long campaign life may be of the type shown in Figs. 3.39, 3.40, 3.41,
3.42(6,36,65,67). Many operators use only the temperature profile, which tends to be approximately the
inverse of the gas efficiency profile. Although a direct measure of temperature may be desirable,
particularly near the furnace wall, it may not give as accurate a measure of furnace working,
particularly at the centre. The thermocouple in the tip of a retractable probe takes several minutes
to reach equilibrium after initial insertion, although this time is reduced at subsequent positions.
22
The ideal probe profile for long campaign life at the required productivity is achieved by control of
burden distribution, i.e. by a combination of fines/coarse placement and ore/coke ratio. Since
every furnace has different top geometry, charge material characteristics and material flow
patterns, the ideal profile must be developed for each furnace. The correct charging sequence for
one furnace is not necessarily the best for a similar furnace and the sequence must be optimised
and continually checked and refined as other operating factors change'
36
'.
Probes, especially underburden probes, are therefore essential tools for prolonging blast furnace
campaign life.
3.4.5 He a r t h Mode l s
In recent years, with increasing computing power available, many mathematical and numerical
techniques have been developed to predict blast furnace hearth erosion and liquid flow in the
hearth.
3.4.5.1 Wear Pr ofi l e
Hearth lining wear may be calculated by mathematical model, using temperature measurements
from embedded thermocouples in the hearth bottom and sidewall'
46,52,68
"
77
'. The modelling
techniques generally calculate the isotherms in the hearth, with typically the surface of the lining or
solidified layer being represented by the 1100-1150C isotherm, equivalent to the temperature of
solidified hot metal. The lining thickness will be calculated from the maximum temperatures
achieved at a given point during a campaign and the thickness of solidified layer deduced, should
the temperature at a given point be below its peak value.
For this technique to be accurate, a good coverage of thermocouples is required and their depth
of insertion needs to be known precisely, together with the thermal properties and geometry of the
lining. The accuracy may also be affected by parameters that may change with time, such as the
conductivity of ramming, thermal contact between courses of brickwork and the development of a
brittle zone in the refractory, which may significantly change its conductivity. The use of the
aforementioned heat flux probes may improve the accuracy of lining wear prediction.
Although hearth temperatures alone give a direct indication of hearth wear, this type of model
combines information from the thermocouples, at differing distances from the hot face, to predict
the extent of wear and solidified layers more accurately.
Direct measurement of hearth lining wear is difficult and undesirable since this requires test
borings and embedded sensors through the full refractory thickness. It is of interest to note that a
Romanian and Indian blast furnaces have embedded radioactive sensors in the hearth pad and
sidewall to determine when the hearth wears to that level (Fig. 3.43)'
75,69)
.
3.4.5.2 Li qui d Fl ow
Models to predict the flow of hot metal and slag in the hearth have also been developed'
77
"
81
'.
These involve predicting the configuration of the deadman and hearth permeability to calculate the
hot metal flow direction and velocity, thus enabling the degree of cross hearth and peripheral flow
to be determined. Practical work to back up these models is described and includes physical
models, the use of radioactive tracers introduced through the tuyeres and the interpretation of
hearth pad and sidewall temperatures.
23
The use of such models improves the understanding of the conditions that cause excessive
peripheral flow in the hearth, which lead to wear of the hearth refractory by abrasion and
dissolution, and helps to avoid them by operational changes and improving future designs.
3.4.6 Ar t i fi c i a l In t e l l i ge n c e
The blast furnace process is a complex one, with a large number of process variables. Modern,
well instrumented furnaces have hundreds of sensors which require to be monitored by a
decreasing number of operators. Consequently, computerised systems are being developed to
process the primary information available and give secondary advice to the operators'
4,8,32,57,82,83
'.
This may be based on a set of operating rules, statistical analysis of data, identifying trends that
compare with historical data and use of intelligent techniques such as fuzzy logic'
84
' and neural
networks'
85,86
'. The aim of these systems is to predict deviation from steady operation and to
quantify the change in control parameters required to minimise the deviations in production and
quality. This will result in more stable blast furnace operation, avoiding major operating problems
such as erratic burden descent and chilled conditions, which is a primary requirement for long
campaign life.
3.4.7 Fur n a c e Top Se n s or s
Since the late 1970's, many furnaces have been equipped with infra-red cameras viewing through
windows in the top cone, to measure stockline temperature profile'
87
"
90
'. This technique overcomes
some of the disadvantages of fixed overburden temperature probes: the falling burden is not
scattered as with probes, leading to a more symmetrical burden distribution, and by measuring
material temperature the effects of stockline to probe distance, which can result in gas mixing and
desensitises the temperature profile, are avoided. A further benefit is that the rotation of the
distribution chute in the furnace can be observed. However, these systems are expensive, difficult
to maintain and experience problems in keeping the viewing window clean, due to the moist, dusty
top gas. Problems have been experienced with the dust in the top gas also affecting the
temperature distribution. In addition, a large amount of data processing is required to process the
output, which sometimes proved a problem in the late 1970's, when this technique was
developed. Consequently, these cameras are not a standard fitment and many operators have
abandoned them'
91
' in favour of radial probes.
Some furnaces are equipped with non-contact stockline profile measurement systems installed in
the furnace top cone. SSAB's Lulea No. 2 blast furnace is fitted with a gamma radar type'
92
' and
BHP was, in 1991, developing a top cone mounted laser profilemeter'
9
'. These systems effectively
replace a retractable overburden probe and, although expensive, have the advantage that they
measure over a larger proportion of the stockline than the single radius of a probe.
3.4.8 Th e r mogr a ph y
The use of thermal imaging cameras to detect hot spots, on the furnace shell'
52
', top gas system,
tuyere stocks, stoves, hot blast and bustle mains and other ancillary plant, is important. Not only
does it enable early detection of problem areas and permit their systematic rectification, but it also
helps prevent catastrophic failures, in which the furnace has to be taken off-blast in a sudden
uncontrolled manner followed by an often difficult recovery, which would have a detrimental effect
on campaign life.
24
3.4.9 Le a k De t e c t i on
An efficient system of detecting water leaks into the furnace from tuyeres and other cooling
members is essential. Undetected water leaks may chill the furnace, resulting in erratic operation
and difficult recovery from chilled conditions. Water leakage will directly affect furnace campaign
life if it damages the refractories. Water leaks in lower, hotter regions of the furnace, which are
lined with carbonaceous materials, will inevitably result in oxidation of the refractories. Ratholes in
the hearth refractories may result, which can lead to breakouts. Water leakage may also result in
taphole problems which may disrupt operations.
Tuyere leak detection systems are often used. Following the installation of a graphitic hearth, bosh
and lower stack, Dofasco installed a comprehensive leak detection system to protect its integrity,
incorporating a system of magnetic flowmeters with computer analysis of the differential flows'
93
'.
Another good system of leak detection uses a pressurised closed circuit water system
incorporating make-up tanks; the make-up frequency indicating the severity of a leak. Other
systems involve observation of gas bubbles or dissolved CO content in the water, differential
pressure measurements, etc. Some plants install wall thermocouples beneath the tuyeres to warn
of otherwise undetected water leaks.
A good leak detection system will often warn the operator of a water leak in its early stages,
before an immediate off-blast is required. This gives the opportunity for the leaking member to be
isolated prior to the furnace being taken off in a controlled manner, with reduction in tuyere
hydrocarbon injection and ore/coke ratio adjustments, thereby minimising detrimental effects
resulting from the subsequent stoppage (as detailed in Section 3.6).
3.4.10 Hot Me t a l Qua l i t y
When operating without a protective skull in the hearth, the hearth carbon may be removed by
solution attack of iron and slag. Early carburisation of the iron, before it contacts the hearth
refractory, will minimise such hearth wear.
For early carburisation, an extended period of contact between liquids and coke is required. At a
given productivity, this may be encouraged by a taller dripping zone and deadman, with a higher
cohesive zone. This usually results in an increase in hot metal silicon. Figure 3.44'
94
' shows that
the carbon saturation level decreases with increasing silicon content. As a result, the hot metal will
be closer to saturation at higher silicon levels, for a given furnace size and hot metal temperature.
In addition, an increase in hot metal silicon increases the hot metal liquidus temperature and
thereby reduces its fluidity. This will tend to reduce the flow velocity in the hearth and encourage
the formation of a solidified layer on the hearth refractory.
At lower hot metal temperatures, the carbon saturation level of the iron is lower and will be
achieved earlier. Low hot metal temperature has the added benefit of increased iron viscosity
which will reduce peripheral flow, reducing the tendency to dissolve protective skulls and
penetrate fine cracks and pores.
Higher hot metal silicon and lower hot metal temperature are difficult to achieve together, as a
higher cohesive zone usually results in a warmer furnace, but the overall effect is for the hot metal
entering the hearth to become closer to carbon saturation. A reduction in high top pressure is
likely to result in a slight increase in silicon, without affecting the thermal state of the furnace.
The probability of dissolution of hearth carbon is therefore lower at higher silicon levels.
25
3.5 Tuye r e Di a me t e r
Tuyere diameter is chosen to ensure adequate blast penetration for given operating conditions
and to prevent excessive gas ascending the furnace walls. The selection of tuyere size will
influence the degree of centre working of the furnace and the degree of protection of the bosh and
lower shaft walls. It may be necessary to vary the tuyere diameter around the furnace to ensure
circumferential balance of gas flow, which can be determined given a method of individual blast
flow measurement such as pressure tappings or cast-in venturis in the tuyere stocks. Figure 3.45
shows the variation in tuyere size and blast volume distribution at NKK's Fukuyama No. 5 blast
furnace'
42
'. A similar effect was achieved at NKK's Keihin No. 2 blast furnace, where hot blast
control valves were fitted to individual tuyere stocks'
95
'.
Although tuyere sizes are carefully chosen, a significant increase in diameter is often observed
when a tuyere is changed, particularly when long lives are achieved. This may affect both of the
above factors and it is advantageous in terms of campaign life to change tuyeres after a given
period, not only to minimise the effect of tuyere wear but to reduce the likelihood of water leakage
into the furnace and the number of unscheduled off-blast periods to change failed tuyeres.
Kawasaki Steel have developed an eccentric, high-velocity tuyere with a three layer
plasma-deposited refractory coating and a welded protective layer on the nose. Tuyere failures at
Mizushima Works have reduced to 0.3 tuyeres/million tonnes of hot metal, with service lives in
excess of 1000 days'
96
'.
The diameter of tuyeres directly above the taphole is often reduced, or the tuyeres even closed, to
promote smooth casting and reduce the iron make above the taphole. The effect of closing
taphole tuyeres is illustrated by the trials with hot blast control valves which showed that if those
valves above the taphole were closed when the taphole started to blow, the cast duration could be
extended considerably, with casting rates still greater than the production rate'
95
'.
Tuyere diameter is often reduced locally, in response to high hearth sidewall temperatures, to
reduce the dripping liquids and hearth activity in the problem area'
8,42,46
'. This may be done by the
addition of tuyere inserts or by tuyere replacement. In severe cases, or as a short term
emergency measure, the tuyeres in question may be closed by plugging with clay'
1,51
'. This often
has a rapid effect in reducing the corresponding hearth sidewall temperatures.
3.6 Off-bl a s t Pe r i ods
The number of off-blast periods, particularly unplanned ones, has a major effect on campaign life
in terms of output per unit volume, which is reduced disproportionally to the percentage downtime.
This is illustrated in Fig. 3.2'
1)
, for Japanese and European blast furnaces. Wall damage may
result from an increased degree of wall working at the lower blast volumes encountered whilst
coming off and on blast, cooling and reheating of the refractories or erratic operation during
recovery from the stoppage.
Dofasco's No. 4 blast furnace suffered a hearth breakout following a period of increased downtime
(7.6-14.5%), which increased tuyere losses and contributed to premature ageing of the hearth'
104
'.
It is notable that during the campaign of Kashima No. 3 blast furnace, which produced 48 Mt over
13.5 years, the longest stoppage was of 53 hours, with annual stoppage time of about 200
hours'
1
'.
Some operators indicate that off-blast periods 'rest' the hearth and allow a protective skull to form
or thicken. In fact, taking the furnace off-blast is often an emergency procedure, at later stages of
26
the campaign, when high temperatures are detected within the hearth refractory. An example of
the reduction in sidewall temperature achieved as a result of such a stoppage is shown in
Fig. 3.46<
97
'.
3.6.1 Sh or t St oppa ge s
For planned stoppages, additional coke can be charged several hours in advance, to compensate
for the reduced blast conditions and the heat losses during the stoppage period. This extra fuel in
the lower regions of the furnace assists smooth recovery from the stoppage. It is usual to
decrease or remove tuyere hydrocarbon injection for a stoppage. At high injection rates, there is a
much lower proportion of coke in the furnace, which is consequently less permeable and this may
hinder recovery from the stoppage. In addition, at high injection rates, the furnace would be
markedly fuel deficient during the recovery until the injection was resumed. This may not b until
the blast volume has reached about half of its full rate, when an adequate raceway has formed
and the injectant will be consumed safely. Should there be operating problems in establishing
raceway conditions and returning to the level of blast at which injection is possible, which may
result from cold conditions or tuyere blockages with slag, the furnace would be fuel deficient at a
time when additional fuel was necessary.
In addition to ore/coke ratio compensation, a burden change may be desirable for a stoppage
period. Smaller ferrous components may be removed from the burden to encourage permeability
following the stoppage. High levels of titaniferrous ores may be reduced to avoid problems at
lower hot metal temperatures after the stoppage. The proportion of burden components that
deteriorate when at high temperatures over a long period, such as certain rubble ores prone to
decrepitation, may be reduced in a stoppage burden. In addition, a more acid burden may be
charged, to compensate for higher silicon content in the hot metal during recovery from the
shutdown'
98
'.
During a stoppage, other deleterious factors may occur which will affect the return to full blast
operations. For example, this may include extended periods at reduced blast volume to cast the
furnace dry before the off-blast, an extended stoppage period for a variety of reasons, water
leakage into the furnace during the shutdown, problems during the recovery that may require
several off-blasts (perhaps to rectify blast leaks or charging faults). Under such circumstances, the
undesirable operating conditions may be extended and the additional coke charged may be
inadequate, leading to a less smooth recovery from the stoppage.
To ensure smooth operation and minimise the effect of a stoppage on furnace life, some
operators advocate a slow lead into and out of a planned stoppage; a typical policy being to
control output to 95% on the day before a stoppage and resume at 80%, then 95% output on the
two days following the stoppage. However, this may not prove acceptable to some operators,
under conditions where high output is required.
Unplanned stoppages are undesirable and, if possible, most operators will try to delay taking the
furnace off-blast for long enough to allow a compensated burden to descend to bosh level.
Attempt is usually made to cast the furnace as dry as possible, to avoid getting slag back into the
tuyeres and blowpipes, which may freeze and further prolong the stoppage. This also gives time
to prepare for the repair work due to be carried out, to minimise the time off-blast. To compensate
for the heat lost due to an unplanned stoppage, the tuyere hydrocarbon injection may be
increased after coming back on-blast, providing it is not already at its maximum level.
Production stoppages also occur due to steelmaking problems or during periods of low demand.
These occurrences should be coordinated between iron and steelworks to give advanced warning
27
where possible, and to give the option of a compensated burden. The torpedo ladle fleet or
steelworks' mixers should be used as a buffer for short stoppages. In certain circumstances, when
there is minimal advance warning of a shutdown, the furnace is not dry and there is little empty
ladle capacity, it may be preferable to plate the iron, despite the cost penalty involved.
Stack spray techniques have advanced, enabling the walls to be gunned in a relatively short
stoppage, by blowing the burden down to a low level. Although this allows a large quantity of coke
to be charged at the lower levels of the furnace to aid start-up, there is often difficulty due to the
quantity of rebounded refractory falling into the furnace. Start-up will be easier if a low rebound
material is used and the furnace is blown down to tuyere level, enabling the rebound material to
be raked from the furnace'
99
'. This may be achieved more effectively by the use of T-shaped
sheets of corrugated sheeting (Fig. 3.47), inserted rolled up through the tuyere cooler
apertures'
100
'. The blow-in burden chemistry must also be adjusted to give a slag chemistry that
will enable the residual rebound material to be melted.
There is a difference of opinion as to whether or not cooling water flows should be decreased for
stoppages of greater than a given duration. Some operators prefer reduced flows to maintain
refractory temperatures. The majority prefer the hearth cooling water on full flow to promote a
thicker protective skull'
41
', whilst others who reduce the water flow suggest that by removing less
heat it assists a smooth start-up.
Another factor which may affect the recovery from a stoppage is the removal of an accretion from
the furnace walls, resulting from the additional wall working and erratic burden descent. This will
encourage chilled conditions at a time when they are least desirable. If an accretion is known to
have formed, it is preferable to try and remove it before a long stoppage. A good system of
accretion monitoring will give immediate warning should this occur, to enable thermal
compensation as soon as possible.
3.6.2 Pl a n t Ma i n t e n a n c e
All maintenance work possible should be carried out during production, thereby reducing the
off-blast time necessary. To minimise the duration of a planned stoppage required for engineering
maintenance, good planning and advance preparation are essential. Although these factors are
obvious for economics and to maximise plant output, their long term effect on furnace life is not
always considered.
Preparations should always be in hand for maintenance to be carried out if the furnace comes
off-blast unplanned for other reasons. For instance, if the furnace is off for a tuyere change, it may
be possible for work to be carried out on the charging system. If the furnace is off-blast for
problems at the steelworks, then it may be possible for more extensive maintenance to be
performed. In this way, the total number of stoppages during a furnace campaign can be reduced
and their duration minimised.
Effective engineering maintenance operations will reduce the number of breakdowns which result
in unplanned stoppages. This will involve routine maintenance, regular inspections, periodic
checking of important instrumentation, condition monitoring, e.g. vibration and thermal
monitoring'
101
'. This is most important at later stages of a campaign, as ancillary equipment gets
older and less reliable. An example is the monitoring system for Chiba No. 6 blast furnace
bell-less top, using vibration and displacement sensors to monitor wear in the chute rotation and
tilt, shown in Fig. 3.48'
4
'. Similarly, improved casthouse maintenance techniques may reduce
off-blast time, e.g. extension of the life of the main iron runner on a single taphole furnace will
reduce downtime'
102
'.
28
3.6.3 Lon g St oppa ge s
Depending on the duration of the stoppage, the furnace may be filled with a coke blank and a low
ore/coke burden, or the burden may be blown down to tuyere level. For stoppages of several
weeks or longer, the salamander or bear should be tapped'
1
'. If this is not done, not only will a
considerable amount of process heat be required, during the recovery, to melt it, but it will expand
whilst still solid and create undue stresses on the hearth refractories and shell, shortening their
life. This will be even more important with newer furnaces, where the sump depth has been
increased to reduce peripheral iron flow in the hearth. It is desirable to monitor these stresses with
strain gauges attached to the hearth jacket, Fig. 3.49'
52
', and to develop procedures to minimise
such stresses.
To recover from longer stoppages, when the furnace is in a cold condition, it is necessary to' warm
the hearth and establish an early link between the taphole and the tuyeres to allow liquids to be
removed. This may be done by the use of a blast pipe at the taphole'
99,103
' or the use of an oxy-fuel
lance'
104,105
'. It is important to prevent the oxygen, entering at the taphole, damaging the hearth
carbon, which could directly shorten the campaign life. Recovery from chilled hearth conditions,
following major water ingress during a routine maintenance stop, has been reported to have
resulted in severe hearth erosion'
9
'. Recovery from long shutdowns, with a large quantity of solid
metal in the hearth and an impermeable deadman, may result in excessive peripheral flow in the
hearth with accelerated hearth sidewall erosion'
106,107
'.
3.7 Ca s t h ous e Pr a c t i c e s
Casthouse practices play an important part in controlling liquid flow in the hearth and avoiding
high liquid levels which may impinge on the raceway, affecting blast distribution or even causing
tuyere or blowpipe damage; factors which can affect the stability of operation, result in periods
off-blast and potentially affect campaign life.
3.7.1 Ta ph ol e Le n gt h
With a longer taphole, the molten products are not only drawn from lower in the hearth, but also
from a point closer to the centre of the hearth. This reduces peripheral flow near the taphole and
hence wear of the hearth sidewall. To extend taphole length, it is necessary to increase the
quantity of clay injected over a period of time, to progressively increase the size of the mushroom
on the inside of the furnace, which will also protect the refractory below the taphole. The reduction
in hearth sidewall temperature at Chiba No. 6 blast furnace, after increasing the quantity of
taphole clay, is shown in Fig. 3.50'
108
'. A change in taphole clay resulted in an increase in taphole
length at Kakogawa No. 2 blast furnace, which is shown by a decrease in hearth wall temperature,
Fig. 3.51'
49)
. Figure 3.52 shows how the predicted hearth wall circumferential temperature
distribution, on a plane beneath the taphole, changes with reducing ratio of taphole length to
hearth radius'
79
'. With a short taphole length and casting alternately from widely spaced tapholes,
sidewall temperature fluctuations would increase, potentially increasing refractory erosion. A
temperature trend of hearth sidewall temperature during casting is shown for Kobe No. 3 blast
furnace (Fig. 3.53) and the hearth sidewall temperature distribution for Kakogawa No. 3 blast
furnace (Fig. 3.54), both of which show peak temperatures at 20-50 from the tapholes, in
agreement with the prediction. Figure 3.55 shows an increase in hearth bottom temperature,
corresponding to the decrease in hearth sidewall temperatures achieved when the taphole length
was increased at Kakogawa No. 2 blast furnace, indicating a reduction in peripheral flow and
thereby a reduction in hearth sidewall erosion.
29
High hearth pad temperatures may be experienced due to the loss of a frozen layer and/or hearth
carbon dissolution, whilst hearth sidewall temperatures are satisfactory. At Kakogawa No. 3 blast
furnace, the relationship between hearth pad and hearth wall temperatures was found to be
cyclical as shown in Fig. 3.56'
109
'. In such cases, it may be necessary to shorten the taphole, by
decreasing the quantity of clay injected and possibly by a reduction in the taphole inclination. This
will help reduce hot metal flow near the furnace centre and increase the liquids retained on the
hearth pad'
107
'.
3.7.2 Ta ph ol e Di a me t e r
The taphole diameter necessary to sustain a given productivity will depend on the furnace
parameters, such as the proportion of time casting, top pressure, slag volume, hearth coke size,
liquid viscosities and taphole clay properties. Should' the taphole be too small for a given
production rate, then it will not be possible to cast the furnace dry. If the taphole is too large, fewer
molten products may be removed from the furnace during a cast as the taphole will blow
prematurely, since the liquids above the taphole will be removed before the liquids on the opposite
side of the hearth can descend through the coke bed'
110
'. In both of these cases, the liquid level in
the hearth will remain high and will ultimately affect stable furnace operation. Consequently, the
optimum size of taphole must be arrived at from experience.
When a single taphole is used, the size must be chosen to enable the furnace to be cast dry and
give sufficient time for the clay to cure between casts. On a furnace where alternate tapholes are
used, different sized tapholes may be required under certain operating conditions, to ensure
drainage across the furnace.
On a multi-taphole furnace that is experiencing wear in the hearth pad, it may be desirable to
increase the taphole diameter. This, together with a decrease in taphole length, should reduce the
iron flow across the hearth pad and increase the residual iron in the hearth at the end of cast,
thereby encouraging the formation of a frozen layer on the hearth pad'
107
'.
3.7.3 Soa ki n g Ba r Pr a c t i c e
The use of soaking bars has been practised for several years. A pointed bar or tube is driven into
the freshly injected taphole clay, a few minutes after gunning-up. The characteristics of the clay
are selected to suit the practice. The main benefit is the rapid curing of the taphole by the heat
conducted down the bar from the hotter part of the taphole. This creates a consistent, stronger,
more durable taphole which permits longer casts at a more even casting rate. This results in
improved liquids removal and more stable operation.
3.7.4 Ta ph ol e Cl a ys
The taphole clay properties are vital to blast furnace operations. The clay must set rapidly and
cure fully between casts to create a strong, durable taphole. For soaking bar operation, it must be
sufficiently plastic to allow bar insertion yet be capable of being drilled out without difficulty, to
retain its dimensions and avoid casting delays. The clay must have good adhesion properties, to
build up a strong, permanent structure that will resist the flow of liquids and also protect the hearth
refractories below the taphole. Figure 3.51 illustrates how hearth sidewall thermocouples directly
below the taphole can be used to monitor the properties of taphole clay and the build-up of the
mushroom inside the furnace'
49
'.
30
3.7.5 Numbe r , Pos i t i on a n d Pe r for ma n c e
High productivity can be attained on a medium sized, single taphole blast furnace, e.g. Thyssen's
10.2 m Hamborn No. 9 blast furnace produced an average 4589 t/day in 1994, a productivity of
2.5 t/m
3
W.V./d'
111
'. However, there are advantages when more than one taphole is available, this
being a necessity at higher production levels. Alternate casting from tapholes on opposite sides of
the furnace results in more effective hearth drainage, and also gives a longer period for the
taphole clay to cure fully, resulting in a more durable taphole. The existence of two tapholes
allows major refurbishment of a main iron runner without requiring a period off-blast. If hot spots
occur on the hearth wall of a multi-taphole furnace, it may be possible to use an alternative
taphole which will not encourage peripheral flow in the eroded area. Sidewall wear resulting from
peripheral flow will be spread more evenly around the circumference on a multi-taphole furnace.
For a large, high productivity blast furnace, it is preferable to have four tapholes, allowing an
opposite pair to be operated whilst one runner is repaired and the other is on standby. To even
out sidewall wear and encourage complete hearth drainage, these would ideally be located at 90
intervals. However, this requires individual casthouses and a compromise is usually reached, to
minimise costs, with the tapholes located in pairs within two casthouses, as widely spaced as is
practical. Ideally, two taphole furnaces should have tapholes on opposite sides of the furnace, but
this is not always possible.
3.7.6 Ca s t i n g Fr e que n c y a n d Ra t e
Casting rate is determined by the taphole drill or bar size used, the wear characteristics of the
taphole clay, top pressure, the viscosity of the liquids and the number of tapholes in use. With
modern high performance taphole clays and soaking bar practice, there is a tendency to reduce
the number of casts, which reduces the taphole operating costs. By lowering the casting rate, the
liquid velocities within the hearth will decrease but they continue for a longer period. On a
multi-casthouse furnace, the possibility exists to cast from opposite tapholes simultaneously (lap
casting), providing the taphole clay cures fully in a shorter time than the cast duration and the
manpower and logistics enable it. This technique will reduce the flow velocities in the hearth,
although it is often only used in times of high liquid level or before taking a furnace off-blast.
As explained previously, cast duration can be extended by reducing the tuyere size above the
taphole or by closing the tuyere stock hot blast valves, where fitted, when the taphole starts to
blow. It is also possible on a single taphole furnace to extend cast duration significantly if, once
the taphole starts to blow, a light plug of clay is put in the taphole for a duration of a few minutes
before resuming the cast'
112
'. This presumably allows the liquid surface in the hearth to level out,
whilst undistorted by the blast, thus enabling further casting.
Lengthy casting delays must be avoided at all costs, to minimise disruption to blast furnace
operations. This requires good design and reliable operation of casthouse equipment, good
casthouse practices and well coordinated transport of hot metal ladles.
Some operators use tapping management models to predict high liquid levels in the furnace, but
these are notoriously difficult to calibrate accurately, since a blow at the taphole does not always
signify that the hearth is dry and, with slag pit operation, the quantity of slag removed is difficult to
determine accurately.
31
3.8 Pr oduc t i on Rul e s
3.8.1 Fur n a c e Ope r a t i n g Rul e s
Being a continuous process, the blast furnace is operated by a number of different operators who,
without a set of rules to operate to, would react differently to a given situation. The individual
actions taken would not all be the correct one and, as a result, the process would be more
variable than if the ideal action was taken. The majority of blast furnaces are therefore operated
according to set procedures that have been developed and improved from experience. These
rules cover a wide area, including practical procedures and process control.
To maximise blast furnace stability, it is essential to control accurately both the thermal state and
the aerodynamics of the furnace. Individual plants devise their own rules to control thermal state,
which generally involve the hot metal silicon and temperature as indicators, with the use of top gas
analysis and calculation of the quantity of heat available in the furnace for silica reduction and to
superheat the molten products. Control of thermal state is usually by adjustment of conditions at
the tuyere or by small changes in the quantity of coke charged. Furnace aerodynamics are
monitored by rules relating to furnace pressure drops and burden descent rates, with adjustment
to blast volume, burden distribution or burden properties to achieve stability.
Operating rules are also necessary for non-routine operations, where damage to the furnace may
result from incorrect procedures, for instance in the recovery from chilled hearth conditions, where
damage to refractories could result.
3.8.2 Rul e s Spe c i fi c t o Pr ol on ga t i on of Fur n a c e Li fe
Many operators have a specific set of operating practices for the prolongation of campaign life,
which are in place to minimise damage to or prevent further deterioration of the furnace. As the
hearth is the critical region of the furnace which cannot be repaired without a long shutdown,
these rules or action plans often relate to hearth conditions, Table 3.3'
8
'. Typically, the actions are
defined according to hearth temperatures or refractory thickness, Fig. 3.57'
51
', Table 3.4
<97)
. All of
the actions taken are discussed in Sections 3 and 4 of this study, with the severity of action
depending on the increase in hearth temperature.
32
3.9 References
1. Wilms E., Janz J. , Lungen . ., Peters M., Schmole P.: 'Prolongation of Blast
Furnace Campaigns - factors which affect the life of the furnace hearth', Stahl und
Eisen 113 (1993) Nr. 6, pp 81-92.
2. Morimoto T., Yoshimoto M., Matsumoto T., Ando H.: 'Upgrading Relining of
Mizushima No. 4 Blast Furnace for a Longer Campaign', ISS-AIME Ironmaking
Conference Proceedings ,1982, pp 132-145.
3. Ishikawa Y., Yamamoto T., Abe Y., Okuda .: 'Operation Methods and Improved
Blast Furnace Cooling and Lining Techniques for Prolongation of Campaign Life',
Proceedings of the 1st European Ironmaking Conference , Aachen, 1986, Vol 3, V2.
4. Kobayashi K., Matsumoto T., Yanagisawa K.: 'Technology for Prolonging Campaign
Life of Blast Furnace', Kawasaki Steel Technical Report No. 32, March 1995,
pp9-1 8.
5. Terayama O., Funakoshi T., Hoshikuma Y., Nunomura S., Hirota N.: 'Technical
Improvements to Keep the High Productivity Operation for Long Period at Kure Blast
Furnace', First International Congress of Science and Technology of Ironmaking,
1994, Sendai, ISIJ, pp 456-461.
6. Mori K., Abe Y., Okuda K., Yamaguchi K.: 'Operating Results of Kimitsu No. 4 Blast
Furnace (1st Campaign)', Nippon Steel Technical Report No. 35, October 1987,
PP 1-11-
7. Koike ., Takata K., Kariya J. , Kojima M., Sato K.: 'Extending the Campaign Life of
Kashima No. 3 Blast Furnace', Proceedings of the 2nd European Ironmaking
Conference, Glasgow , 1991. pp 155-167.
8. Campbell D.A.: 'Extending the Redcar Furnace Campaign into the 21st Century',
Steel Times December 1993, pp 505-6.
9. Lathlean D., Ballard P., Edwards K., Piatt J. : 'Ironmaking Developments and Their
Influence on Blast Furnace Rebuilds', Proceedings of the 2nd European Ironmaking
Conference, Glasgow , 1991. pp 77-90.
10. Bowness M. M., Willmers R. R., Parratt J. E., Davies R.J.: 'Coke Quality for
Successful Blast Furnace Operation', Ironmaking and Steelmaking 1987, Vol 14,
No. 4, pp 157-170
1 1 . White W., Charlton K.: 'Coke Quality Requirements in Europe - European
Co-operative Report', Proceedings of the 2nd European Ironmaking Conference,
Glasgow ,1 991 . pp 188-204.
12. Nakajima R., Kishimoto S., Hotta H., Ishii K., Sakurai M., Itagaki S.: 'Behaviour of
Coke Degradation in a Blast Furnace', NKK Technical Review No. 60 (1990), pp 1-8.
13. Uenaka T., Miyatani H., Hori R., Noma F., Shimizu M., Kimura Y., Inaba S.:
'Improvement of Blast Furnace Operation With Center Coke Charging', Iron and
Steelmaker, November 1988, pp 34-39.
33
14. Willmers R. R., Monson J.R., Wilkinson H.C.: 'The Degradation of Coke in the Blast
Furnace', ECSC Contract 7210.AA/8/801-802, 1984/5.
15. Schoone E., Toxopeus H., Vos D.: 'Trials with a 100% Pellet Burden in Blast Furnace
No. 6 at Hoogovens Umuiden', ISS-AIME Ironmaking Conference Proceedings,
1995, pp 465-470.
16. Van Laar J., van Stein Callentis J.E.: 'Blast Furnace Campaign Life and Repairs',
Steel World, September 1991, pp 31-37
17. Ward R., Warren D., Keil R.: 'World Record Lining Life for a 1500 m
3
I.V. Blast
Furnace Burdened at 80% Pellets', ISS-AIME Ironmaking Conference Proceedings
1989, pp 369-374.
18. Ward R.F., Broadbent P.F., Tsalapatis J.: 'The Whyalla No. 2 Blast Furnace Life
Extension With Improved Productivity and Quality', ISS-AIME Ironmaking Conference
Proceedings 1993, pp 93-99.
19. Maki T., Ono R., Goto T., Kiguchi J., Hori R., Kuwano K.: 'High Pellet Ratio Operation
at Kakogawa No. 2 Blast Furnace', ISS-AIME Ironmaking Conference Proceedings
1992, pp 209-213.
20. Aranha B.E.: Operation of Blast Furnaces with 100% Pellets at Dofasco', MSI
Seminar on Sinter and Pellets, Brussels 1988, pp 15-25.
21. Sloan D., Ward R.F.: 'Blast Furnace Performance using Dolomite Fluxed Pellets',
McMaster Symposium No.12, May 1984. Burden Design for the Blast Furnace,
pp 234-249.
22. Clixby G.: 'Influence of Softening and Melting Properties of Burden Materials on Blast
Furnace Operation' Ironmaking and Steelmaking, 1986, Vol.13, No.4, pp169-175.
23. Kortmann H.A., Ritz V.J.: 'Quality Aspects of Blast Furnace Burden Material at
Elevated Temperatures', ISS-AIME Ironmaking Conference Proceedings, 1990,
PP29-43.
24. Lungen .., Poos .: 'Injection of Coal into the Blast Furnace', ECSC Synthesis
Report 7210-ZZ/557, June 1995.
25. Clixby G., Parratt J.E.: 'An Alternative View on Sinter Properties for Stable and
Efficient Blast Furnace Operation', ISS-AIME Ironmaking Conference Proceedings,
1989, pp 459-469.
26. Altpeter W., Beppler E., Gerstenberg ., Kannappel M.: 'Effects of Alkalis in the Blast
Furnace', Stahl und Eisen, 1988, No.17, 22 August 1988, pp 781-790.
27. Fujihara S., Tamura S., Ikeda M., Nakai M.: 'High-Duty Carbon Blocks', Nippon Steel
Technical Report No. 41, April 1989, pp 1-6.
28. Schoone E.E., Steeghs A.G.S., Toxopeus H.L.: 'Burden Distribution at Hoogovens
Umuiden' EBFC Meeting, Taranto, 21-22 September 1995.
34
29. Peters K.H., Mohnkern H., Lungen ..: 'Ways and Means for Low Coke Blast
Furnace Operation', First International Congress of Science and Technology of
Ironmaking, 1994, Sendai, ISIJ, pp 493-499.
30. Hirose S., Sugawara H., Taniyoshi S.: 'Blast Furnace Burden Distribution Control with
3-Parallel Bunker Bell-less Top Using Large Amount of Small Sinter', Kawasaki Steel
Technical Report No. 32, March 1995, pp 3-8.
31 . Shibaike H., Sasaki S., Ogata I., Yasunaga S., Naito M., Hashimoto S.: 'Long-Term
High-Efficiency Operation of Sakai No.2 Blast Furnace (Third Campaign)', Nippon
Steel Technical Report No. 43, October 1989, pp 41-53.
32. Hur .S., Cho B.R., Kim H.D., Eden M.G.: 'Ironmaking Developments at Kwangyang,
POSCO', ISS-AIME Ironmaking Conference Proceedings, 1994, pp 213-222.
33. Isobe M., Sugiyama T., Inaba S.: 'The Reaction Caused by the Mixing of Coke Into
the Ore Layer in a Blast Furnace', Proceedings of the Sixth International Iron and
Steel Congress, 1990, Nagoya, ISIJ, pp 439-446.
34. Kanetsuna T., Akimoto E., Saito S.: 'New Blast Furnace Revamping Technology',
Iron and Steel Engineer, June 1990, pp 32-37.
35. Ohgami M., Uetani T., Senoo Y., Yamazaki M., Fujimori H., Tamura S.: 'Revamping
and Operation of Mizushima No. 3 Blast Furnace', ISS-AIME Ironmaking Conference
Proceedings, 1991, pp 333-340.
36. O' Hanlon J. : 'Blast Furnace Control Philosophy at British Steel', Iron and Steel
Engineer, December 1991, pp 13-19.
37. Uenaka T., Miyatani H., Hori R., Noma F., Shimizu M., Kimura Y., Inaba S.:
'Improvement of the Blast Furnace Operation with the Center Coke Charging at
Kakogawa No. 2 BF', ISS-AIME Ironmaking Conference Proceedings 1988,
pp 589-594.
38. Chao J.T.: 'The Practice of Ilmenite Application in the Blast Furnace', La Revue de
Metallurgie - CIT, October 1989, pp 765-774.
39. Cordier J.A., Lesur F., Graffeuille F.: 'Blast Furnace Burden Optimisation at Usinor -
Dunkirk', McMaster Symposium No.12, May 1984. Burden Design for the Blast
Furnace, pp 43-66
40. Buist D.R., Kuroda K.: Metal Bulletin's 4th International Iron Ore Symposium, Rome,
March 1985.
41 . Alleyn A. G., Munive L.: 'Hearth Behaviour of Port Kembla Blast Furnace',
Proceedings of the International Hearth and Raceway Symposium, 1981,
Wollongong, 11.1-11.22.
42. Yamamoto R., Nakajima R., Koyama Y., Niiya K.: 'Ten Years Life of No. 5 BF in
Fukuyama', ISS-AIME Ironmaking Conference Proceedings, 1985, pp 149-163.
35
43. Takata K., Koike .: 'Still in Good Operation, A Status Report on Kashima No. 3
Blast Furnace', ISS-AIME Ironmaking Conference Proceedings, 1988, pp 439-447.
44. Shourong Z., Zhongjie Y.: 'Maintenance of Blast Furnace Hearth and Bottom Lining
by using Titania-Bearing Material', ISS-AIME Ironmaking Conference Proceedings,
1988, pp 537-542.
45. Carmichael I.F., Lowing J. : 'Recent Operation of Redcar No. 1 Blast Furnace', Steel
Times, August 1981, pp 416-420.
46. Garcia ., Rodriguez R., Borrego J. , Aldama F., Garcia ., Banuelos Y.: 'Monitoring
and Maintenance of Blast Furnace Hearths, AHMSA Experience', ISS-AIME
Ironmaking Conference Proceedings, 1994,'pp 333-339.
47. Okada T., Kuwano K., Shimomura K., Hori R., Miyatani H., Ochiai Y., Uemura K.:
'Protection of Blast Furnace Hearth Refractories by Ti 0
2
Injection Through Tuyeres',
ISS-AIME Ironmaking Conference Proceedings, 1991, pp 307-312.
48. Calcagni M., Federico G., Guili M.: 'Hearth Conditions and BF Life Prolongation at
ILVA Taranto Works', European Blast Furnace Committee, Redcar, 28-29 October
1993.
49. Shimomura K.: 'Hearth Protection for Long Campaign Life', Proceedings of the 2nd
European Ironmaking Conference, Glasgow , 1991. pp 179-187.
50. Graffeuille F., Jager J.M., Capelani R., Libralesso J.M.: 'Hearth Design and Control of
the Sollac Dunkerque BF 4', ISS-AIME Ironmaking Conference Proceedings, 1990,
pp 289-297.
51 . La Carrubba S., Salvatore E., Barnaba P., Staffilani .: 'Second Campaign of Taranto
No. 5 BF and Developments for the Third Campaign', Proceedings of the 2nd
European Ironmaking Conference, Glasgow , 1991. pp 62-76.
52. Hebel R., Jeschar R, Steiger R., Streuber C: 'Temperatures in the Blast Furnace
Refractory Lining. Theoretical Calculations, Measurements and Determination of the
Wear Profile', ISS-AIME Ironmaking Conference Proceedings, 1995, pp 101-111.
53. Sakamoto Dr Y., Sakai T., Yamamoto T., Kurashige I.: 'Diagnosis Techniques for
In-Operation Blast Furnace Linings', Taikabutsu Overseas, April 1982, Vol.2, No. 1,
pp 65-70.
54. Harz , Dr.-lng. Dissertation (Dr.-lng. Thesis), TU Clausthal, 1982.
55. Schulte M., Krause E., Voss M.: 'Development and Operational Test of a New
Method of Wear Control at the Blast Furnace Hearth by Direct Heat-Flux
Measurement', ECSC Contract 7210-AA/133, Report No. 3, March 1995.
56. Ponghis N.: 'Investigations on Optimum Blast Furnace Working Based on the Use of
Objective Strategies Founded on the Association of Quantified Symptoms', ECSC
Research Contract 7210-AA/208, Report No. 7, March 1988.
36
57. Warren P.W.: 'Recent Developments in Blast Furnace Process Control Within British
Steel', ISS-AIME Ironmaking Conference Proceedings, 1995, pp 281-284.
58. James J.D., Nanavati K.S., Wakelin D.H.: 'LTV Steel's First Experience With Stack
Staves', ISS-AIME Ironmaking Conference Proceedings, 1992, pp 247-254.
59. Hughes F., Rees A. J. : 'Optimise Burden Descent, Gas Efficiency, and Heat Losses
by Controlling Gas Distribution in British Steel Blast Furnaces', European Blast
Furnace Committee, Taranto, 21-22 September 1995.
60. van Laar J. , Tijhuis G.J., Spreij M., Stokman R.J.M.: 'Blast Furnace Lining Life - A
Quantitative Analysis of Lining/Cooling Systems', Iron and Steelmaker, December
1994, pp 25-29.
61 . Strimple R.A., Snyder J.E., Shoemaker B.F.: 'Method for Measuring the Thickness of
a Refractory in a Metallurgical Apparatus', U. S. Patent 4,269,397 (May 26 1981),
Bethlehem Steel Corporation.
62. Kawate Y., Sonoi H., Yokoe K., Takano S., Shimomura K.: 'Measuring Methods for
Lining Erosion of the Blast Furnace', Transactions of the Iron and Steel Institute of
Japan, 1982, Vol.22, No.10, pp 799-806.
63. Gao ., Huang L, Wang Y., Kou ., Zhang X., Deng B.,Tao R., Wen Y, Gu J. :
'Measurement of Lining Thickness in a Blast Furnace by Electrical Resistance
Method and its Application in No. 1 BF of Shanghai Baoshan Iron and Steel
Complex', Iron and Steel (China), March 1992, Vol. 27, No. 3, pp 6-9.
64. Higuchi M., Wakuri S., Isoyama M.: 'Development of New Sensors and Improvement
of Control System for Blast Furnace', Nippon Steel Technical Report, No. 35, October
1987, pp 43-51
65. Notman C, Hill P.: 'Control of the Blast Furnace Process Within BSC' Metals and
Materials, December 1986, Vol.2, No. 12, pp 762-767.
66. Schneider W.: 'Probing the Blast Furnace for Better Gas Distribution', Steel Times,
April 1988, Vol. 216, No. 4, pp 199-200.
67. Federico G., Salvatore E., Guili M., Barnaba P.: 'ILVA Experience in Extending Blast
Furnace Campaign', ISS-AIME Ironmaking Conference Proceedings, 1 991 ,
pp 357-364.
68. Kurita K., Hatano M., Hariki M. Takashima H., Mori K.: 'A Mathematical Model
Simulating Wear Contour of Blast Furnace Hearth', Transactions of the Iron and Steel
Institute of Japan, 1980, Vol. 20, No. 1 , pp B3.
69. Singh B.N., Chatterjee L.M., Chatterjee .: 'Monitoring the Erosion of Blast Furnace
Hearth Pad by the Use of Sub-hearth Thermocouples', Transactions of the Indian
Institute of Metals, Vol. 35, No. 5, October 1982, pp 475-481.
37
70. Suh Y.K., Huh N.H., Choi J.S., Park H.K.: 'Development of Mathematical Model to
Predict Residual Thickness of Refractory and Solidified Layer in Hearth of
Kwangyang No. 3 Blast Furnace', Research Institute of Industrial Science and
Technology (Pohang City) Technical Research Report, June 1992, Vol. 6, No. 2,
pp 221-228.
71 . Preuer ., Winter J. , Hiebler H.: 'Computation of the Erosion in the Hearth of a Blast
Furnace', Steel Research, April 1992, Vol.63, No. 4, pp 147-151.
72. Yoshikawa F., Nigo S., Kiyohara S., Taguchi S.: 'Application of a Computer to
Estimation of Refractory Wear and Solidified Layer Distribution in the Blast Furnace
Hearth', Taikabutsu (Refractories), October 1990, Vol. 42, No. 10, pp 579-587.
73. Tanski J.A.: 'The Development of Advanced Simulation Models to Determine Hearth
Erosion at Inland's No. 7 Blast Furnace', ISS-AIME 10th Process Technology
Conference Proceedings, 1992, pp 45-53.
74. Gdula S.J., Bialecki R., Kurpisz K., Nowak ., Sucheta .: 'Mathematical Model of
Steady State Heat Transfer in Blast Furnace Hearth and Bottom', Transactions of the
Iron and Steel Institute of Japan, 1985, Vol. 25, No. 5, pp 380-395.
75. Sandu I.F., Munteanu V., Nedelcu P., lonescu M.P.: 'Progress and Prospective
Directions of Ironmaking Technology at the Iron and Steel', Complex S.C. "SIDEX"
S.A. Galatz Romania First International Congress of Science and Technology of
Ironmaking, 1994, Sendai. ISIJ, pp 462-468.
76. Rex ., Zulu P., Plat P., Tanzil W.,Skimmings T., Jelenich L.: 'Determination of the
State of the Hearth of BHP Steel's Blast Furnaces', ISS-AIME Ironmaking
Conference Proceedings, 1993, pp 603-609.
77. Leprince G., Steiler J.M., Sert D.: 'Blast Furnace Hearth Life: Models for Assessing
the Wear and Understanding the Transient Thermal States', ISS-AIME Ironmaking
Conference Proceedings, 1993, pp 123-132.
78. Ohno J. , Nakamura M., Hara Y., Tachimori M., Arino S.: 'Flow of Iron in Blast
Furnace Hearth', Proceedings of the International Hearth and Raceway Symposium,
1981, Wollongong.
79. Shibata K., Kimura Y., Shimizu M., Inaba S.: 'Dynamics of Dead-man Coke and Hot
Metal Flow in a Blast Furnace Hearth', ISIJ International, 1990, Vol.30, No. 3,
pp 208-215.
80. Steiler J-M. , Venturini M.J., Le Coq X., Leprince G., Dufor ., Billon ., Lao D.,
Provost G.: 'Blast Furnace Hearth Erosion', Steel Technology International, 1994/5,
pp 47-51 .
81 . Sert D., Steiler J-M. , Venturini M.J., Negro P.: 'The Internal State of the Blast
Furnace Hearth', Steel Technology International, 1993 pp 47-51.
82. IRM S.A.: 'Blast Furnace Monitoring - a Tool for Better Operation', Steel Times,
December, 1994, pp 496-498.
38
83. Matsuda K., Nakayama M., Otsuka Y., Konishi M.: 'Application of Al Technology for
Blast Furnace', First International Congress of Science and Technology of
Ironmaking, 1994, Sendai. ISIJ, pp 302-307.
84. Ferstl ., Brunnbauer G., Stohl ., Druckenthaner .: 'Blast Furnace Process Control
at VOEST-ALPINE Slahl Linz GmbH', Young European Blast Furnace Engineers'
Conference, Lulea, September 7-8 1995.
85. Otsuda Y., Matsuda K., Hanaoka K., Kadoguchi K., Maki T.: 'Application of Neural
Network Systems to Pattern Recognition of Blast Furnace Operation Data', Kobelco
Technology Review, No. 15, October 1992, pp 12-16.
86. Peters H., Plum -D.: 'Artificial Neuronal Networks and their Application in th Steel
Industry', Stahl und Eisen, 114,18 April 1994, pp 39-46.
87. Togino Y., Tateoka M., Sugata M., Yamaguchi K., Okubo ., Nakagome M.:
'Application of Thermoviewer at Blast Furnace Top', ISS-AIME Ironmaking
Conference Proceedings, 1979, pp 295-304
88. Franssen R., Pirlet R., Danloy G., Cantigniaux .: 'Spirotherm for the Measurement
of Temperature Distribution at the Blast Furnace Stockline', ISS-AIME Ironmaking
Conference Proceedings, 1979, pp 305-312
89. lizuka M., Yoshida H., Matsumura K., Sano ., Shinoda .: 'NKK Thermo Camera
System for Furnace Top and its Application to Blast Furnace Operation', ISS-AIME
Ironmaking Conference Proceedings, 1979, pp 313-326
90. Tonteling P.: 'Computer Analysing of Infrared Blast Furnace Burden Data', ISS-AIME
Ironmaking Conference Proceedings, 1979, pp 327-335
91. Kowalski W., Bachhofen HJ. , Beppler E., Krebich K., Mlheims ., Peters M.,
Weiters, CU. : Process Control at the Blast Furnaces of Thyssen Stahl AG, Young
European Blast Furnace Engineers' Conference, Lulea, September 7-8 1995.
92. Hallin M.: 'Recent Developments in Ironmaking', Steel World, August 1992, pp 54-57
93. Donaldson R.J., Harshaw A.S., McClure A.M.: 'Leak Detection at Dofasco's No. 4
Blast Furnace', ISS-AIME Ironmaking Proceedings, 1988, pp 467-474.
94. Peters K.H., Beppler E., Gerstenberg ., Lorch R.: 'Parameter Governing the
Nitrogen Content in Hot Metal', ISS-AIME Fifth Process Technology Proceedings,
1986, pp 1005-1009.
95. Nakajima R., Kisimoto S., Hotta H., Ishii K., Kobiyama H., Yamaoka Y.: 'Innovative
Technology of Blast Furnace Operation with Hot Blast Control Valves', NKK
Technical Review, No. 59, 1990, pp 1-7.
96. Shinohara K., Akizuki H.: 'Design and Performance of New, Long-life Blast Furnace
Tuyere at Mizushima Works', Iron and Steel Engineer, October 1988, pp 26-28.
39
97. Parratt J.E.: 'Control of Hearth Sidewall Temperatures at Redcar', Colloquium on
Blast Furnace Disturbances and their Prevention, MatTech Centre, Helsinki, 21-22
November 1994.
98. Gravemaker G., Broekstra .: 'Blast Furnace Process Control at Hoogovens,
Umuiden', Young European Blast Furnace Engineers' Conference, Lulea, September
7-8 1995.
99. Bell P.F., Duckworth R.F., Green J.W.: 'Stack Renewal by Remote Control,
Development of a Technique', ISS-AIME Ironmaking Proceedings, 1990, pp 389-393.
100. Keaton D.: 'Remote Gunning Technique Maximises Top Change Downtime at
Armco's Amanda Blast Furnace', Iron and Steelmaker, April 1993, pp 33-35.
101. Moore ., Bowler K.: 'Maintenance - An Integral Part of the 'Process", ISS-AIME
Ironmaking Proceedings, 1992, pp 625-635.
102. Schuett K.J., Muniz S.A., Workman J.A., Restivo V.D.: 'Reduction of Blast Furnace
Downtime Through Hot Repairs of Trough Refractories', ISS-AIME Ironmaking
Proceedings, 1991, pp 301-305.
103. Ricketts J. , Salinas R., Fodness K.: 'Blow-in Practice with a Pipe From the Stoves to
the Taphole', ISS-AIME Ironmaking Proceedings, 1992, pp 311-319.
104. Donaldson R.J., Fischer A. J. , Sharp R.M., Stothart D.W.: 'Dofasco's No. 4 Blast
Furnace Hearth Breakout, Repair and Rescue', ISS-AIME Ironmaking Proceedings,
1995, pp 183-193.
105. Oltolini G.P., van Stein Callenfels J. E. , van Laar J. , Tijhuis G.J., de Pagter J. : 'Open
Hearth Surgery and Subsequent Resuscitation of Comigliano BF 2', ISS-AIME
Ironmaking Proceedings, 1992, pp 329-336.
106. Dufor ., Billon B., Lao D., Provost G., Didelon F., Venturini M.J., Le Coq .:
'Mechanisms of Hearth Erosion Following Examination of the State of Sollac Blast
Furnace Hearths after Shut Down', European Blast Furnace Committee, Redcar
28-29 October 1993.
107. Lathlean D., Mellor D., Mitchell S., Tanzil F.: 'Experience with Hearth Drainage and
the Influence on Stable Operations at BHP Port Kembla Blast Furnaces', McMaster
Symposium No. 19, Raceway Control for Optimum Blast Furnace Performance,
May 1991.
108. Kamano H., Kawamura M., Ichifuji K., Matsumoto T., Marushima H., Obata H.: 'Long
Term Stable Operation of Chiba No. 6 BF', ISS-AIME Ironmaking Proceedings, 1989,
pp 515-521.
109. Uenaka T., Shimomura K., Kuwano K., Uemura K.: 'Refractory Technology for
Extending the Life of Blast Furnaces', ISS-AIME Ironmaking Proceedings, 1986,
pp185-191.
40
110. Burgess J.M., Jenkins D.R., McCarthy M.J., Nowak L.P., Pinczewski W.V., Tanzil
W.B.U.: 'Mathematical and Physical Simulation of Non-Uniform Liquid Drainage in
Packed Beds', Proceedings of the International Blast Furnace Hearth and Raceway
Symposium, 1981, Newcastle, Australia.
111. Operation Data of Sinter Plants and Blast Furnaces, EBFC European Data Exchange
1994.
112. Clixby G.: 'The Effect of Hearth Liquid Level Upon Blast Furnace Operation', ECSC
Contract 7210.AA/817, Technical Report No. 4, June 1995.
41
TABLE 3.1
COKE SPECI FI CATI ON FOR REDCAR BLAST FURNACE, 1986
Micum (BS) test
M40
M10
Extended Micum
Coke size*
AMS. mm.
>75mm. %
<30mm, %
Reactivity
cope
Postreaction strength
Chemical analysis
Moisture
Ash
Sulphur
wt%
Sample point
After coke oven screen
Minimum specification
84 min.
7 max.
068 max.
5152 min.
10 max.
6 max.
29 max.
58 min.
4 max.
10 max.
085 max.
Typical result
860
60
06507
51
8
4
24
67
2 5 3 0
10
055060
After blast furnace screen
Minimum specification
86 min.
65 max.
065 max.
50 min.
8 max.
3 max.
29 max.
58 min.
4 max.
10 max.
085 max.
Typical result
880
5 5
0607
50
3
3
24
67
2530
100
055060
Square aperture basis.
TABLE 3.2
REQUIREMENTS ON THE PHYSI CAL AND METALLURGI CAL
PROPERTIES OF SINTER AND PELLETS
I Sinter
Strength
Grain size

R0I ISO 4696


Reducbility
II Pellets
Grain size
Tumbler strength
LTB

Redutibility R
iQ
Swelling
ISO 3271 > 6.3 mm
< 5 mm or < 6 mm
< 10 mm
>50 mm
<3.15 mm
ISO 4695
> 16 mm
8 1 0 mm
< 6.3 mm
ISO 3271
> 6.3 mm
< 0:5 mm
ISO DIS 4697 (dynamic)
> 6.3 mm
< 0.5 mm
ISO 4695
ISO DP 4698
7 0 8 0 %
max. 5%
max. 30%
max. 10%
max. 30 33%
1.41.6%/min
max. 5%
min. 85%
max. 5%
min. 95%
max 5%
min. 80%
max. 15%
min. 0.8%/min
max. 15%
43
TABLE 3.3
REDCAR BLAST FURNACE LI FE EXTENSI ON
OPERATI ONAL PRACTI CES
Limit Productivity, eg 2.2t/m
3
/day
Apply Production rules
eg Coke Size, Output Changes/Burden
Parameters
Control Maintenance Shutdowns
Improve Instrumentation to identify
trigger points
Use Ilmenite
Reduce Tuyere Bore in trouble areas
Apply Shell Spray Cooling
Enhance Hearth Cooling
TABLE 3.4
HEARTH PROTECTI ON PLAN
TEMP
100
150
200
250
300
325
CAUTION
LEVEL
SAFE
BEWARE
ATTENTION
DANGEROUS
V.
DANGEROUS
ALARM
CARBON
THICK-
NESS (m)
2.14
1.40
1.14
0.80
0.64
0.55
ACTION
Mon i t or
Ch e c k Th e r moc oupl e , Ma x Cool i n g
Re duc e Tuye r e Si ze , Re duc e Out put
Re duc e Out put by 5%, Pl ug Tuye r e ,
Add Il me n i t e
Re duc e Out put by 10%
Off Bl a s t 24 Hour s , Pl ug Tuye r e ,
Re duc e Out put By 15%
44
Productivity
" 3 5 I " 3 5 I " 3 7 ' 3 8 3 9 . ' 9 0 "9 1 " 9 2 I 9
, I-
\/Cf^X^l^a^
OPERATI ONAL ACHI EVEMENT OF KURE NO. 1 BF FIG. 3.1
o
E
-*.
o
o
o
C
3
5

IM
LU
l U
9
t
7
6
5
4
at
2
1
s.
".
^^^
^ V
\
*
s , *
"! ?". Si .
_

a

1
r s



^ ^

1
^
1
10 20 30 40
Stillstandszeit je Ofenreise in %
50
PRODUCTI ON PERFORMANCE OF BLAST FURNACES AFTER FI G. 3.2
COMPLETED FURNACE CAMPAI GNS vs PERCENTAGE DOWNTI ME
45
3u80
^ : 6 0
9 LU 40
co -

a
CL

^1 40-
g 120
100-1
LU

(b)
" ' " ' 9 f ' ' -
5 15 25 4 14 24 4 14 24 3 13 23 3 13 23
March April May June
1984
a sidewall temperature; b centre temperature
July
TRENDS I N HEARTH TEMPERATURE
(HI GHLI GHTED BY ARROWS), MARCH - JULY 1984
FI G. 3. 3
a

E
c
C
r-\
re
c
CJ
re
re
CJ
5. 5- 4. 0-
- 5. 0- 3.5
4.5- 3.0-
(1BF) (2BF)
18 19 20 21 22
Mean size of coke in deadman, mm

CJ

c
CJ

c
re
"33
CJ
- D
re
CJ
E
CJ
EFFECT OF MEAN SIZE OF DEADMAN COKE ON
GAS AND LI QUI D FLOW
FIG. 3.4
46
JZ
>
es
c "5
i


S
Q.
ce U
"
150
100
50
0
1,000
800
600
90
70
50
30
I
i
k^\
X

\

I
|
X


V. ^ _ _
I

, ! , I , I I , [ t :_!__, L _,_ _
10 20 30 40
Time (day)
50 60 70
THE EFFECT OF CENTRAL COKE CHARGI NG ON
THE TEMPERATURE OF THE HEARTH BOTTOM AND
HEARTH WALL
FIG. 3.5
71 6.0
CT
c
S LO
OJ
>
1 2.0
OJ
M


>
30 LO
Z ei t ( h)
EFFECT OF SLAG ALKALI NI TY ON THE RESIDENCE TI ME
OF THE ALKALI S I N BLAST FURNACE NO. 2 AT
SOLN/ER I N FOS-SUR-MER
FIG. 3.6
47
0.9 1,0 1.1 1.2
Besengrad __LJ9L_ der Schlacke
S1O2
+
AI2O3
ALKALI CONTENT OF THE SLAG I N VARI OUS
BLAST FURNACES
FI G. 3.7
Je r
0.86
O.Sa OS: O.SO '"O.Tli 0.76 0.74 0.73 0. 70068". "
Thermal flow ratio
3.00-.
1.90.
LO/LC-0. 70 0.80 0.90
1.0
1.1
L
0
= Ore Layer Thickness at Wal l
L
c
= Coke Layer Thickness at Wal l
OPTI MUM LQ/ I - C DI AGRAM ON THE VARIOUS FURNACE
OPERATI NG CONDITIONS
FI G. 3.8
48

/* *'
=
c
e
n
t

b
o
t
t
o

^

<
/
60
50
40
30
20
"
.
> *. "
.

" *" *
: .:.::
; " * .
" " .
: _ _ : . .
: ; ; " : = : ; - . ' . _
. . ; ; . _ a ^
" . ; . . " : a . . ; ; . ; . "
a a a ; a a "" " .

1 I 1
0 10 20 30 40
QQ at center of upper shaft (%)
(CQ2%)

CO (CO

7c)+(C02

7c)
RELATIONSHIP BETWEEN nCO AT THE CENTRE OF
UPPER SHAFT AND AT CENTRE OF HEARTH BOTTOM
FIG. 3.9
Rotating chute
ZL 3
H^H
Ci
C:
OL
Os
Small Ore
1514 131211109 8 7 6 5 4 3 2 1
(13-5*) (52.0
e
)
Center Bell-less notch Wall
TYPICAL CHARGING PATTERN OF NO. 3 BLAST FURNACE FIG. 3.10
49
Distance from center (m)
1.0 2.0 3.0 4.0
1 . 0-
"5 2.0
g 3.0
4.0
5.0
Cml Coi J On i Cs i i l Oi ul Os o'
^^m^-**4
^ ^ * ^ s/%f/^2
/
^
P^"
Os o
C sc
-1
\
C i a
Ci x'
c
o
(.w)
i Cu( y) i 0i 2
C f y O i z '
Coa rc e c oke
Gn e s in te r
Movi bl e a rmor
;:c s ( oi i Oui wl i Os i o)
:cs(o;:o
u
:)x-2-
ttotdi
C: Nut coke
O,: Pel sinter
N' : Number of chirps
: Number of charges
*
inonecyde
without O
BURDEN DISTRIBUTION PROFILES (AS CALCULATED BY
MATHEMATI CAL MODEL FOR PREDI CTI NG
BURDEN DISTRIBUTION)
FIG. 3.11
50 100 150 200
Belly temperature (C)
250
COMPARI SON OF PERMEABI LI TY BEFORE AND AFTER
NUT COKE ORE MI X CHARGI NG
FIG.3.12
50
Mixing rate of small-size
coke (kg/t-p)
EFFECT OF SMALL-SIZE COKE ON REPOSE
ANGLE OF PELLETS
FIG. 3.13
Q.
O
Q .

OJ
+-"
OJ
E

'a
OJ
"

OL
OJ
>
OJ
cr
Charging
chute
Flow control.gate
Segregation
control plate
Bunker
parallel
3 bunker type
(control plate)
O parallel
3 bunker type
( without control
plate )
center feed type
O 25 50 75 100
Discharged weight ( '/. )
CHANGES IN ORE SIZE DISCHARGED FROM THE
TOP BUNKER
FIG. 3.14
51
Wall Center:
Notch
Coke
Large
Ore
Center
Coke
Small
Sinter
1 I 2 I 3 I 4
F
r
-|
J
, , , , > , , : ,
**:.:: .:::::
5
6 I 7 I 8 ! 9
-.-: xc. . .

Nute
t 10 I 11
Oke mixing
. . . . t E I
te # . .
ffif;
VX.,..'
:-. -,:: 1
, : , > . , : .
t
-,,-. ,
TYPI CAL CHARGI NG PATTERN AT KWANGYANG FIG. 3.15
Hopper
EQUIPMENT FOR CENTRE COKE CHARGI NG AT
KAKOGAWA' S NO. 2 BLAST FURNACE
FIG. 3.16
52
1.76
1.75 -
co
O
~

Q.
<3

I
.
1.74
1.73
1.72
1.71
1.70
10.0
9.5
9.0
8.5
s
3
Vi
E
Q.
C
. ^
^ * ^
^*w
^ \
^ ^
* ^
w
\


I \
\
\
\
^ \
N
*
* ^ .
s
*>
V*
k * >
\
\
\ \
\ \
\ \
\ \
\ \
\ \
\ *
\ . \
"^
\
\
CX
\ \
\
\
\ #
V.
^ \ _
\ . . \

I
0 50
^ 1K^
^ ^ ^
"^
^ V
'
"
\
I I
100 150
1 -
Wc (kg/eh)
THE EFFECT OF THE VOLUME OF CENTRE CHARGED
COKE ON , PI AND SLIP TIMES
FIG. 3.17
53
20
Strand I :ROI . 58 Ti 23
Strand : Ol . SI Ti 2 3
0.1
0.2
INFLUENCE OF Ti ON RDI
0. 3 0.4
% Ti in sinter
FIG. 3.18
Q
CC
SO
40
30
20
IO
>
100 %
00
80
( =

O
CC
CL
Ui
o >
70
OC

io

o
<
50
-L1 3.2 5 10.2 19.0
Tl 0
2
KG/THM
UI
cc
EFFECT OF T i a ADDITION ON RDI
AND SINTER STRAND PRODUCTIVITY
FIG. 3.19
54
C ; : Q: i ?0?i 303S 0 SSC SeO 65707 90
Sedar " " " E ^ '<**'='
TpnoeOture 2O ^ ^ U " * ' , No.3
of heonn 220J ; > ; { * T S 0 ^ \ * " ,
siaewoii
u u
l P. \ No.1
er. N - -
0
- -
160)- C/ A ^ ^ N i .
wol , . / ' ^ VJ
I20
0
[*C1
\
Bolonce
2 0
(kg/T) ,
5
Input
- Sl og
o- Iron
Resiouol
12
(kg/T) -1.0 i-
1.01
05
-0 5?-/
AW
Temperorure
) 0

d no metal
w 9 0
( .
c )
W80-
130
CoO/SO 125
120
/
THE RESULTS OF I NCREASED T O*
CONTENT OPERATI ON AT NO. 5 BF
FIG. 3.20
'85 '86
Oct. Nov. Dec. Jan. Feb. March April May June July
l : ;
C O
Temper at ur e 2 0 0
of side wal l
100
Titanium
additives
(kg/charge)
Grouting
ref ract ory
800
600
400
200
RESULTS OF I NCREASI NG TO2 OPERATI ON
AT KASHI MA NO. 3 BF
FIG. 3.21
55
5 11

s*
"
3 '
t
3
V


/ X I
/ '
/ ' X. '
W/m*
KjTTOt/ntH
-
I X
te
t f *
lf-26 2J-3 i7-il 1-4 f'* IS- 33 t-4*U~B -H
Mr*A 4fr. Hm/
TRANSI TI ON OF MAXI N/ UM HE AT LOAD OF HE ARTH
STAVE S AND THE DOSAGE OF TO2
BE ARING MATE RI AL I N THE BURDE N
FI G. 3 .2 2
(u se d)
I No . 4 T H| No . I T. H
WD^
Dam age d po sitio n i H 2 02 )
o f he art h wal l
Ti 02 in je c t e d t u ye re s
(N0. 4 an d No . 6 )
Grc u m f e re n rial wal l
t e m pe rat u re dist ribu t io n
@ : be f o re TiOz in je c t io n
() .af t e r TiOz in je c t io n
(u se d)
DAMAGE D POSITION OF HE ARTH WALL AND
Ti 0
2
I NJE CTE D TUYE RE S
FI G. 3 .2 3
56
250
5 IO
May, 9 0
TRANSI TI ON OF HEARTH WALL TEMPERATURE FIG. 3.24
Tuyere
TiOz injected tuyeres (No.26 ond No.28)

29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13
--
N0.3T.H
y
^ /No.2T.H
R8I7 R8I6 * ^
R7I8'
(used)
Thermocouple
position
Damaged position
of hearth wall
RELATI ONSHI P BETWEEN DAMAGED POSITION OF HEARTH
WALL AND Ti 0
2
I NJECTED TUYERES
FIG. 3.25
57
150
IO 15
Oct . . 85
Days
Nov.
TRANSI TI ON OF HEARTH WALL TEMPERATURE FI G. 3. 26
C aso
- kg/t-p
S" s
~ o
5 lOOO
=0
c
soo
t / h
M^
w
U"wi
< 1 0-30-
o
>
85 86 "87 88 ' 89 90
1.
5*C-
91 "92 '93
ACHEVEMENT OF HEARTH BRI CK TEMPERATURE,
I NPUT T 0
2
, VOLUN/E OF FLOWI NG WATER
FI G. 3.27
58
WALL TEMPERATURE HOLE 2
TEMPERATURE,^
U I
I D
350
300
250
200 -
150 -
100 - - - \
50
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 f 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 f f
1/N 5 15 25 1/D 5 15 25 10 20 1/F 5 15 25
10 20 10 20 1/J 5 15 25 10 20 1/M 5
Tc : E21 +50 mm G3 +50 mm F3 +50 mm
ma x 321 *C ma x 240 *C ma x 256 *C
CASTING
EFFECT OF ILN/ENITE INJECTION (SOREFLUX) ON THE WALL TEMPERATURE
NEAR THE TAPHOLE NO. 2
FIG. 3.28
T. H. 4^- ~J" . H. 1
TUJ. 23 INJ. TK32
T.H.3 T. H. 2 TAPPINO T.H.
HEART WALL
20

TEMPERATURE
*
c
ieo
180
170
160 3
03 F7 T
O ^ . . . . .
TI 02 tujere
lnjectlon(71 RUTILE)
150 U 1 ! 1 I I ' ' I I I L
12 13 14 15 18 17 I S 19 20 21 22 23
July 93
HEART WALL
TEMPERATURE
c
196
194
192
190
188
186
184
182
inn
/ .*..
/ a'' - " VX\
/ /
/ a'4
._,
/

_ . . . / . Injection RUTILE
- 4 ^^^ ^ ^ ^
1 3 5 7 9 11 13 15 17 19 21 23 1 3 5 7 9 11
17/07/93 18/07/93
HEARTH WALL TEMPERATURE CONTROL AT TAPHOLE
NO. 3 BY TO2 I NJECTI ON AT TUJ23
FIG. 3.29
60
THE TREND OF TEMPERATURE AT
HEARTH SI DEWALL
FI G. 3.30
SHELL
TC 2
BOTTOM
LEVEL I
TC I
WEST . -f, J _ IA5T WH T J . . , V u i UI T
/ \
TC 4 ^ C j TC < * TC J
SOUTH SOUTH
WALLS
NORTH
T C
. ' I .
Te
/
wtST3
TC * TC 3
HOLES
I . 1 . ' '.
BF A 1987
THERMOCOUPLES I NSTALLED I N THE HEARTH
FIG. 3.31
61
I L
ORIGINAL _
J_3
1 LEVEL 9150 mm. FOUR AROUND B.F. HEARTH AT 200 mm
DEPTH.
2 LEVEL 8250 mm. FOUR AROUND B.F. HEARTH AT 200 mm
DEPTH.
3 LEVEL 6165 mm. EIGHT ACROSS B.F. HEARTH
h
LJJ
PRESENT

LEVEL 10.000 mm. FOUR SETS OF TWO TC EACH.
AT 45 OF IRON NOTCHES.
2 LEVEL 9150 mm FOUR SETS OF TWO TC EACH.
BELOW ANO AT 90 FROM IRON NOTCH.
3 LEVEL 8700 mm. EIGHT SETS OF TWO TC EACH.
4 LEVEL 8250 mm EIGHT SETS OF TWO TC EACH.
5 LEVEL 8090 mm, EIGHT SETS OF TWO TC EACH.
6 LEVEL 6165 mm SAME AS THE ORIGINAL DESIGN
GAN/AHMSA NO. 5 BF HEARTH ORIGINAL
AND PRESENT TC POSITIONS ELEVATION VIEW
Slag notch
Te.2
Tc6
LEVEL 9150
Slag notch
Tc9
T c S i / ^
/\yfe.
No.1 Iron //
notch 1 / \
Tc.-i \-\y
\ TVOI
T t l X S .
Te
Slag notch
" ^NTe^a
19X/ C
JcV\03Vlo^ li
f l Jnotch
NTe.lfl lTc11
y VV
i rysf.-2
10
Tc27
LEVEL 8700
Tc 15 |
s l a
9
n o t c i l
0.2 Iron No.1 Iron
notch
No.2 Iron
notch
LEVEL 8250
ACTUAL Tc ARRANGEMENT PLAN VIEW FIG. 3.32
62
direction West North East South West
LeveT^ '
8 0
'
2 7
' '
w
'
l x 0
'
NL+16200
NL+15100
NL+14200
NL+13200
NL+11500
-O i

Y
(D >


- * -
: 0: 0 !

T T T
'D >


- -
U:
:
:
- 4 !
NL+ 9820
=1-2
C : double thermometers
' ? : single thermometers
C
:
Tap hole
ARRANGEMENT OF THE HEARTH TEMPERATURE
MEASURI NG POINTS
FI G. 3. 33
1977 *80 "85 '90
TRANSI TI ON OF STAVE HEAT LOAD FI G. 3. 34
GCals/Hr
30
20
10
-

r i
Campaign 1 Campaign 2
*
. v<>*
1 1 . . 1 1 1 1 1 1
0 1 2 3 4 5 6 7 8 9 10
Years into Campaign
THERMAL STRESSES ON THE FURNACE SHELL DURI NG
CAMPAI GN 2 ARE LESS THAN I N CAMPAI GN 1 DUE
TO A LOWER HEAT LOAD
FI G. 3. 35
63
800
.
I CO
COQ
s
X
200
DAILY AVERAGES

mmkM r
BTu/MiN* ,' Mpn^ l i Ff ' ;
,> M 'I II ' . " .
> v w
l
J I L
30
25
20
15
io
CM CO
i Q
U 2
?"
? Z>
O O
ss
8/5/87 4/1/88 11/27 7/25 3/22/9C 11/17___,
12/3 7/30 3/27/89 11/22 7/20 3/17/91
AUG 1987 JUN 1991
H-4 FURNACE HEAT LOSS FROM STACK STAVES FIG. 3.36
Profile meter (luer.
microwave or mechanical
type)
Burden surface profile, radial distribu
tion of burden descent velocity and that
of ore/coke layer thickness ratio
Top probe
Verticil probe
Layer thickness meter
Upper shaft probe
Cross shaft probe
Middle shaft probe
Radial fas temperature and composition
distributions
Gas temperature and composition,
solid temperature, observation and
solid sampling in vertical direction
Ore/coke layer thickness and bur
den descent velocity
Radial (diametrical) gas tempera
ture and composition distributions
in upper shaft
Radial gas temperature and com
position distributions in middle
shaft
Shaft pressure gauge
Stave thermometer
Cas pressure at several points in
vertical and circumferential direc
tions of shaft
Thermal load distributions in cir
cumferential and vertical
directions
Lower shaft short probe
Radial gas temperature and com
position distributions in lower
shaft
Belly probe
Gas temperature and composition,
solid temperature, direct observa
tion and solid sampling in radial
' direction
Tuyere coke sampler Coke sampling in raceway
Deadman probe
Hearth bottom and side
wall thermometers
Gas temperature and composition, solid
temperature, dina observation and
solid sampling in deadman and raceway
Hearth bottom and sidewall tem
perature distributions
LOCATIONS AND FUNCTIONS OF MAIN
BLAST FURNACE SENSORS
FIG. 3.37
64
Furnace
Probe
C
Indicating/
'Recording
Chute
(or Bell)
Material / /
S t r e a m
/ /Activated
Sensors^ / j f Sensors
r r t r / r n l r r t T / y
I
/ I
TRAJECTORY PROBE FIG. 3.38
Mi -
so
40
30
20
November 1980
X September 1980
Center
Will
CHANGE I N GAS FLOW DI STRI BUTI ON
(AS MEASURED WITH UPPER SHAFT PROBE)
FIG. 3.39
CO +co
2
60
SO
40
30
20
10
Wall f
Clean Walls Fuel Economy
T
Good Driving
Center
AN I DEAL GAS PROFI LE ESTABLISHED BY
BY UNDERBURDEN PROBE
FIG. 3.40
65
Optimum above stockline temperature profile
400Ci
200 C-
0"C
I 00C
Centre Wall
Optimum over burden gas profile
+ 60%
Max attainable
(For high efficiency)
%CO, xl 00
^
CO
* %CO + %CO,
44-55%
Centre Wall
TYPI CAL OVER-BURDEN TEMPERATURES AND GAS
COMPOSITION PROFILES FOR OPTI MUM OPERATI ON
a (ABOVE) OPTI MUM ABOVE-STOCKLI NE
TEMPERATURE PROFI LE, b (BELOW) OPTI MUM
OVER-BURDEN GAS PROFI LE
FIG. 3.41
CO
600
500
400
300
200
100
n
-
-
-
_l
_. 1 1 _
i
/v
M
// \\
/ / | \
l \ S
i
, . i
TUAITO If S
, ^- IF 1
I 1 1 1 1
3 2 1 0 1 2 3 4 5 (m)
TYPI CAL GAS TEMPERATURE PATTERNS AT THE TOP FIG. 3.42
66

i
g
c 5 c o o o o c o c o o o ~
Th* calculated mod temperaturas at th batom surtac of th carbon blocks aning
JZ
Th*rrnl flow through th bottom wall [ WW]
A Th* radioacBv sourcat xnnng at th dat of dannai flows and temperatur manurcnwnt
THE CALCULATED WEAR RATE OF nc6 BLAST FURNACE
HEARTH FROM SC " SI DEX" SA GALATZ - ROMANI A
FI G. 3.43
Sjo
Si
X
.
w
380
C Sat urat i on Li nei
HO 1. Schweigern
O HO 7 Hambom
HO 7. without Ilmenite
H08.
Diameter BF Hearth
I m]
13.6
7.6
7.6
9.0
Q60
V. Si ' Hot Metal
0.80
roo
CARBON AND SI LI CON PERCENTAGES
I N HOT METAL FROM DI FFERENT
BLAST FURNACES
FI G. 3.44
67
Unodjustemem of tuyere diameter (120* mm 39)
Adjustment of tuyere cfiomerer
/I20>mm 31, 1300 mm 8 \
VQMarking ,l20cemm130mm change/
DI STRI BUTI ON OF BLAST VOLUME EACH TUYERE
FI G. 3.45
Degrees C
350
23.26 hrs OB
3.15 hrs OB
0
* <, ^ <a<^
INCREASE I N TEMPERATURE UNDER
NOS. 1 & 4 TAPHOLES
FI G. 3.46
68
Refractory wall
Corrugated raking
surface
Tuyere o p e n i n g
PLAN VIEW SCHEMATI C OF THE AREA
COVERED BY THE CORRUGATED I RON
FIG. 3.47
Sl e w i n g motor
L
Tilting motor
Pickup
Ampere
meter
O O
isplacement
Current
c o nve ne r
Vibration
c o nve ne r
Sl e wi n g chute
Displacement
c onve ne r
Interface
sequencer
O
Monitor
MONI TORI NG SYSTEM FOR THE BELL-LESS DRI VE UNI T
FIG. 3.48
69
eter scr-v csol i nq
e- r t xn er emi t s rarniin mix
blank flange
& ffi
iTTXi
t remacel e
I i n \ cresstre t i ent screwing
MEASUREMENT OF STRESS RELATED DATA FIG. 3.49(a)
197 Mia. according FEM calculation (Model 21
I I I I I I I I I I I I I I I I I I I I I I I I
IMI II I MI I / S ] < S 1/5? 1 I S It II K/ VI 3 4 S t I t S 11 II M 1/51 ! J t 5 I 1 I 1 11 II l!/H
Time (monlh)
MEASUREMENT OF TENSION ON THE HEARTH WALL FOR A FI G. 3.49(b)
PERI OD OF THREE YEARS AFTER COMMISSIONING
70

tu
"O
ro
si
t
ro

.c
Q.
E

r
200
Mud Quantity
T a p n o l e c h
at stopping
3
_
N a 2
increased
i_L
J I I L_J_
Measuring point
: below No.3 Tap Hole
10 15 20
1987. January
25 30
(day)
TYPI CAL OPERATI ON ACTI ONS FOR EROSION
OF HEARTH CORNER
FI G. 3.50
Q
al


u. o
OS
* J o
P!
< O
ttX <
. or
2 4
UJ J X
r


PREVIOUS MUD NEW MUD o '.NO.ITH
:N0.4TH
3 8
3. 6
3.4
32
140
120
looH
80
60
8 4 / i 2 3 4 5
PREVIOUS MUD . NEWA
V
MUD N0.2TH
~ ^ ^ . : N 0 3 T H
f\
lrt&S&.
: NO. I TH
: N0. 2TH
: N0. 3TH
+ : N0. 4TH
&&&
MEASURING POINT
OF WALL TEMPERA
TURE BELOW THE
TAPHOLE
84/ 1
EFFECT OF TAPHOLE DEPTH EXTENSION ON
HEARTH WALL TEMPERATURE AT KAKOGAWA NO. 2 BF
FI G. 3.51
71
9
1
2 3 4 5 (m)
' '
/ R=0 . 0
* * / K= 0 . 3 3
* l / R= 0 . 4 5
POSITION AND DEPTH OF A
TAPHOLE FOR CALCULATI ON
FI G. 3.52(a)
*/ R=0. 0 /R= 0. 33 l/R= 0. 45
o
u

<
I
<
1480 1520
U80 1520 U80 1520
CHANGE OF CI RCUMFERENTI AL DI STRI BUTI ON OF
REFRACTORI ES TEMPERATURE AT THE HORI ZONTAL
CROSS SECTION UNDER A TAPHOLE
FI G. 3.52b)
72
8 5. 0-
CHANGE OF HEARTH WALL TEMPERATURE DURI NG TAPPI NG
(KOBE NO. 3 BF)
FIG. 3.53
/ t -
f 260 \
.. . .
CI RCUMFERENTI AL DI STRI BUTI ON OF HEARTH
WALL TEMPERATURE ( KAKOGAWA NO. 3 BF)
FIG. 3.54
73
_
o _
a - 300
2
W
3 O
S o
41 3
, 100
Ef
S
*~
Temperature measuring
A3. A
.

-
i
,

. !

1^~
<
1 2 3 4
Time (month)
T.H
T.H.
CHANGE OF REFRACTORI ES TEMPERATURE I N A HEARTH
WITH PROLONGATI ON OF THE DEPTH OF A TAPHOLE
( KAKOGAWA NO. 2 BF)
FI G. 3. 55
200
LOCATION OF THERMO-
COUPLES
X
t - 50
rr ^
<
LU
I
UJ
<
tu
LU
3
PERICO OF SHORT TAP HOLE
T
I.6
82.3
o
85. IO
83.12 PERIOOOF
LONG TAP HOLE
300 400 500 600
HEARTH FAD TEMPERATURE ,Tp ("C )
1981
1982

1983
* 1984
1985
HEARTH TEMPERATURE CHANGE OF
KAKOGAWA NO. 3 BF(1 )
FI G. 3. 56
74
Ca r bon bot t om t e mpe r a t ur e , C
1000
900
800
700
600
500
400
300
200
100
0
24 HOURS BF SHUT-DOWN
1. UNDER HEARTH COOLING PIPES CLEANING
2. WATER FLOW RATE INCREASE IN BOTTOM PIPES
3. WATER FLOW RATE INCREASE IN STAVES
4. T02 INCREASE IN BURDEN: 5 -12 kg/t HM
STANDARD CONDITIONS
CARBON
PASTE
INJECTION
T02LOAD
5 kgrt HM
TUYERE
DIAMETER
DECREASE
WIND RATE
DECREASE
[SI] INCREASE
TUYERE
BLINDING
NEAR CRITICAL
AREA
50 100 150 200 250 300
He a r t h wa l l t e mpe r a t ur e , C
350 400
TARANTO NO. 5 BF - SECOND CAMPAI GN
OPERATI NG SCHEME FOR HEARTH
TEMPERATURE CONTROL
FI G. 3.57
75
4. REMEDIAL ACTIONS
Examples of campaign extension possibilities are given in Table 4.1
(1)
.
4.1 Gun /Spr a y
The stack refractory brickwork of the blast furnace is mainly exposed to mechanical wear in the
upper part and also to thermal and chemical wear in the lower part. Stable operation of a blast
furnace throughout its campaign life is ensured by always keeping the stack lining profile balanced
in the circumferential direction'
3
'. Intermediate relinings in that area are possible in a blown down
blast furnace. A remote controlled gunning system'
3
"
8
', enabling repairs to be made to both the
stack and bosh areas using closed circuit television to monitor the work'
3
', is state of the art
technology. This system eliminates the need for men to enter a hot furnace. It was reported that
4651 of castables were applied to the bosh and stack of Anshan's No. 9 blast furnace in under
four days by remote controlled gunning'
3
'.
In order to decrease the stoppage time compared to "cold gunning" (cooling the furnace
completely and constructing scaffolding from which the gunners work), Armco'
4
' also chose a
remote controlled gunning device known as the SEQUAD process, for the stack and bosh of its
Amanda blast furnace (hearth diameter 10.2 m). The device used was suspended from a
cantilever beam by cables and inserted through a door in the side of the stack near the furnace
top. It was then positioned in the centre of the furnace using a gantry'
4
'. The operator used a
continuous feed refractory gun to apply a mixture of monolithic material and water to the wall of
the furnace at a uniform distance and from a constant 90 angle. The gunning mechanism rotated
360 degrees and could be raised and lowered to allow the operator to reach all parts of the stack.
As each revolution is completed, the gunning device is simply raised to the next level. The
operator can stop the rotation to fill in more seriously eroded areas. The unit is equipped with
lights to improve operator visibility'
4
'. It was reported that the advantage of this technique is that
the reprofiling of the Amanda furnace could be completed within the planned eight-day outage
(cold gunning took 30 days to complete) required for the top change'
4
'. Since the equipment is
operated by remote control from a platform outside of the furnace, there was no need to wait for
the furnace to cool before gunning could begin. This gunning device was also used at Dofasco
No. 4 blast furnace, Fig. 4.1
(8)
. Before the gunning operation, cleaning of the stack and the
removal of large scabs is an important prerequisite'
5

7
'. The removal of loose refractory materials
and/or accretions can be done, for example, by water jetting'
3,5
' and cleaning of the stack can take
between 12 and 14 hours'
5
'. When the gunning procedure is completed, the furnace can be dried
by controlled blowing of low temperature hot blast through the furnace'
5
'. The top gas temperature
is progressively raised to 400C by increasing the blast temperature, and the target is to raise the
temperature 150 mm back in the applied material to 120C to drive off moisture from the gunning
operation. If this is not done, the rapid heating once the furnace is operated could cause massive
spalling'
5
'.
The length of a shutdown for robot gunning is reported at AK Steel Ashland to be in the range 3 to
5 days'
10
'. Here gunning is carried out every 9 to 10 months (gunning to profile). Nippon Kokan
reported that the campaign life of the gunned refractories is six to twelve months, in the upper
section of the shaft, and about three to nine months in the medium section
(7)
. Hoogovens report
three to five months of life extension by gunning'
11
'. To give the furnace an additional three to four
years campaign life, carrying out a professional intermediate repair is proposed, if the only reason
for the reline is the bad condition of the stack or bosh refractory'
11
'.
A comparison of upper stack lining repairs is shown in Table 4.2'
2
'. The refractories for the repair
method are shown in Table 4.3. The life of gunned repairs can be extended by the addition of
77
attachments to the furnace shell. In the hex-frame gunning method, a steel frame with hexagonal
grates is fixed to the shell, where the brick lining is badly worn, and a castable refractory is
gunned onto the frame to a thickness of 150 to 200 mm to rebuild the lining. The advantages are
that the castable can be fully gunned up to the back side of the grates and that the gunned
castable has high strength. The hex-frame gunning method is expected to extend the life of the
upper stack lining by about six months, under low thermal loads'
2
'. The installation of cooling
pipes in the upper stack can also give additional support to gunned refractory'
9
'.
The precast panel method was developed to improve the brittleness of the gunned castable
layer'
2
'. In this method, a castable refractory is vibration-formed in a hex-frame and dried into a
panel at 350C. The precast panel thus obtained is fixed to the shell and castable is grouted into
the gap between the furnace wall and the panel. The panels are reported to have lasted more
than one year'
2
'. The development of gunning repair materials for the upper shaft, as well as for
the middle and lower shaft, is described for extending campaign life of Kashima No. 3 furnace'
9
'.
4.2 En h a n c e d Cool i n g
Enhanced cooling of the blast furnace wall in critical areas is one way to promote prolongation of
campaign life. One critical area is the taphole of a blast furnace. At the Redcar blast furnace
(14.0 m hearth diameter) of British Steel, a refrigeration unit, a chiller, was installed below No. 4
taphole, after the hearth sidewall temperature peaked under this taphole in 1992'
12

13
'. The unit
cools the water being pumped to the two staves directly below the taphole, i.e. the area with the
most carbon wear. Stave water temperature was reduced to approximately 5C compared with
the normal 30C. This cooler water aids a faster build-up and thicker skull in this area. After the
success of the chillers for the staves under No. 4 taphole, the scheme of using chillers was
extended to the other three tapholes of the Redcar furnace.
To further improve the durability of the upper stack lining, a method in which water-cooled iron
castings are installed was developed'
2
', Table 4.2. Water-cooled iron castings are installed in such
a manner that they are exposed in the furnace, and a castable refractory is grouted into the gap
between the worn brick lining and the water-cooled iron castings. The water-cooled iron castings
can more adequately withstand thermal load variations than refractories, such as precast panels'
2
'.
Water-cooled panels have also been installed in the upper shaft area at blast furnace No. 6 of
Chiba works, for campaign prolongation'
17
'.
At the bosh of Kashima No. 3 blast furnace, a hole was bored in the shell at the locations of stave
wear and stave pipe failure'
9
'. A small cooling panel (pin) made of copper was then attached
where low temperature setting mortar was injected, Fig. 4.2. This should promote formation of a
layer of adhesive inside the furnace to protect the shell of the furnace body. The small cooling
panel can ordinarily be attached in blowing-out, and its shape does not ensure the high retention
of adhesive. These measures become insufficient when stave wear progresses'
9
'. For example,
stave wear advanced in the Kashima No. 3 blast furnace between B3 and B2 staves (Fig. 4.2), so
that, in blow-out, with the stock level lowered to the bosh, a small water-cooled metal (plate) was
attached as a method of repairing the furnace wall from the furnace interior'
9
'. It has shape to
cover the wear part of the stave and is composed of a steel pipe wrapped with cast iron. In
addition, it provides a structure to allow attachment with feed and drain pipes to the attachment
hole for the small cooling panel'
9
'. The wear condition of these small coolers suggests that the
repair effect corresponds to 1 -1.5 years.
Cylindrical cooling elements (also known as cigar coolers) can be utilized as a water-cooled
refractory anchor or stave insert, when remedial repairs are required and shutdowns of the blast
78
furnace should be avoided'
14
"
16
'. Cylindrical coolers (Fig. 4.3) also provide cost effective
alternatives for extending the lifetime of new refractory linings, whether installed during relines,
repairs or in new furnaces'
14
'. Apart from the installation of such coolers in a newly relined blast
furnace and as refractory anchors, their mounting plate configurations can include built-in grout
injection ports to save installation costs and time, Fig. 4.4. In addition, special twin cylindrical
coolers can be mounted on a common flange or plate containing such a built-in grouting port,
Fig. 4.5'
14
'. This twin unit can then be installed in any location on the furnace stack that requires
grouting, without the need to drill holes in the stack and weld grout nozzles in place'
14
'. Cylindrical
coolers can also be utilized to re-establish cooling in failed staves, Fig. 4.6'
15)
. The coolers are
inserted through the shell into openings which are provided in the stave. These openings can be
lanced or high pressure water cutting employed. The location pattern and size of the coolers is
dependent upon the actual stave configuration and the location of the cast-in pipes, mounting
bolts and piping connections'
14,15
'.
A hearth cooling system using a refrigerator for bottom cooling water is shown in Fig. 4.7'
18
'. The
system was established to protect the hearth carbon bricks from heat load by hot metal. The
cooling water temperature was 30 to 35C and is now controlled at 5C. As a result, the thickness
of the solidification layer in the hearth increased and the carbon brick temperature decreased.
An underhearth cooling system was installed at the air-cooled bottom of BHP Port Kembla No. 4
blast furnace'
19
'. This was an emergency measure because of increasing temperatures (maximum
805C at 4 metres from hearth shell) in the hearth bottom. Underhearth cooling pipes were
installed at that furnace on programmed stops from August 1979 to April 1980, Fig. 4.8. A total of
18 pipes was installed at approximately the desired spacing of 500 mm, being cooled with
20 000 l/h of recirculating water, extracting about 70 kW of heat. As a result, localized
temperatures fell sharply when the pipes were installed'
19
'. Later, a chiller system was also
installed on the underhearth cooling system of No. 5 blast furnace, lowering the water coolant
temperature from a summer average of 30C to 15C'
20
'.
For improvement of cooling of the furnace shaft, cooling plates were used at Fukuyama No. 5
blast furnace, taking into consideration the brick supporting ability and exchangeability at the time
of burnout'
21
'. Six path cooling plates were employed for the lower portion of the shaft where
thermal load is high, and two path cooling plates for the upper and middle portion. It was reported
that, in ten years of operation, none of the six path cooling plates at the lower portion were burnt
out. From investigations, it was found out that, due to smooth and uniform flow speed of water, a
six path cooling plate is more durable than a two path type
121
'. Other successful measures against
burnout of cooling plates were increasing the water supply volume, decreasing the protruding
length of cooling plates by gunning repair on the shaft and reduction of the thermal load on the
cooling plate by attempts to stabilise the furnace condition.
At Thyssen Stahl AG a blast furnace repair method has been developed'
22
', discussed within
VDEh Committee in 1989/1990 and successfully used at several blast furnaces. It is called
intermediate hearth repair'
22
'. In the framework of a planned shut-down a thermovision camera is
used to check the hearth to establish whether it is necessary to carry out partial repair of the
hearth. Depending on the work necessary 10 to 20 days are required to carry out the hearth
repairs, described in reference (22). During this period other necessary refractory repair work can
be carried out, such as shaft spraying with an injection robot. The repair work includes the
removal of a piece of shell where there is insufficient carbon brick remaining, the exchange of
worn carbon brick by new ones, the welding of a new piece of shell and the filling of the free space
between the shell and the carbon brick with high thermal conductivity grouting material, with a
special binder.
79
4.3 Gr out i n g a n d We l di n g of Sh e l l
Grouting of blast furnaces is a well known technology to promote a long campaign life. Grouting
of blast furnace hearths, i.e. the injection of plastic refractory material through the shell from the
exterior, is carried out to backfill following repairs: to counteract gas leakage in the taphole area; in
"brittle zones" of the hearth wall and in case of temperature rises in the hearth wall'
23
'. This is
done at nearly all blast furnaces in the world. Figure 4.9 shows for example that, at Kashima No. 3
blast furnace, a temperature increase in the wall was counteracted by titanium additives, by water
spray and by grouting'
24
'.
When carrying out grouting, the following safety criteria are recommended'
24
':
During all grouting in the region of the hearth, tuyeres and elbow tops should be
closed.
Since the grouting material flows well, it is sufficient to set the pressure on injection to
5 bar. This prevents any brick residues due to build-up of grout pressure from being
pushed into the furnace.
The furnace shell should be cooled to prevent excessively rapid vaporisation of
plasticisers and consequent crack formation.
Water-free masses should be used.
To prevent gas blow out from the taphole, caused by insufficient burning of mud because of the
high thermal conductivity of SiC bricks used in that area and by deterioration of mortar used
between brick and shell, improvement countermeasures have been carried out at Wakayama
Steel Works'
25
'. These included:
Injection to around the taphole. To plug the gas passage, a total of 10,000 litres of
materials such as tar and resin were injected into the tapholes and surrounding areas
in 199 places through the shell.
Taphole inside ramming. In order to intercept gas passage through taphole brick
joints, ramming of a resin type material was performed.
Injection of resin into tapholes. For intercepting gas passage through taphole brick
joints, resin injection into tapholes using a mud gun was performed.
Improvement of taphole casting material and casting method. In order to accomplish
interception of gas passage between shell and stave and between stave and brick
from the taphole side, the conventional method of casting material into tapholes in
one operation was changed to "gunning + 2 phase casting", as shown in Fig. 4.10, to
prevent generation of voids.
The shell of a blast furnace must withstand high operating and refractory pressures, thermal
stresses, burden loads and have numerous cut-outs for internal water cooling systems
connections'
26
'. Cracks in the shell cause gas leakage. The following concepts to minimize cracks
are recommended:
80
The shape must be as simple as possible'
27
'.
A design that restricts shell movement, such as a ring girder, must be avoided'
27
'.
Shell thickness must be as thin as possible'
26,27
'.
To meet these conditions, the shell material must have excellent elasticity and weldability'
27)
. A
steel with the following chemical composition and mechanical properties is recommended'
27
':
Chemical composition: C 0.18%; Si 0.35%; Mn 1.4%; 0.04% (maximum);
S 0.04% (maximum); carbon equivalent 0. 31 % (maximum).
Mechanical properties: tensile strength (minimum) 41 to 52 kgf/sq mm (58,000 to
74,000 psi); yield point (minimum) 23 kgf/ sq mm (33,000 psi).
The use of finite element techniques allows significantly thinner shell plate, compared to
sophisticated design practice and optimised design, to give a greater ability to withstand cracking
in the latter part of the campaign'
26
'.
The welding of blast furnace jackets and the main requirements for achieving safe welded joints is
reported for the new construction or rebuilding of blast furnaces'
28
"
30
'.
Cutting out of cracked sections of the shell and replacement with a prelined piece of shell was
carried out at Fukuyama Works'
7
'. It was reported that nearly simultaneous with the start of the
damage of cooling plates, cracks occurred on the shell, commencing at the corner of the section
where the cooling plate is mounted. Part of the shell was therefore cut out during blowing out and
a shell prelined with castable refractories for about 150 mm was installed'
7
'. At the same time, the
method to weld fitting flanges for the cooling plates had been changed, and grouting holes to
mortar have been made on the inner side of the prelined shell. It is imperative with the shell to
detect at an early opportunity the sections with thin residual brick thickness and to grout or spray
castable refractories to prevent cracks'
71
.
4.4 Replacing of Staves and Coolers
Damaged cooling plates can be easily replaced, whereas it is practically impossible to change
staves without blowing down the furnace. Examples of how to proceed for replacing stave coolers
are described, for removing staves inside the furnace from the top'
9,31
' or for replacement from
outside, together with shell segments'
32,33
'.
For example, after a hearth wall break out near the taphole of USS/Kobe's Lorrain No. 4 blast
furnace, a hearth wall and stave repair was carried out'
32
'. The iron notch insert stave cooling
circuits were damaged, so that the iron notch stave would have to be replaced. It was not feasible
to replace the entire left iron notch stave, due to time constraints. The agreed repair consisted of
replacing as much of the left iron notch stave as possible with fabricated staves'
32
'.
These staves were designed with serpentine piping, embedded in a carbon ram refractory. Grout
nipples were included in the design of the repair staves. The nipples provided the possibility for
filling any voids between the stave and carbon brick hearth after installation. Hearth jacket repair
plates were designed with holes to accommodate the repair staves' grout and cooling piping.
Grout nipples were also used to fill any voids between the repair staves and hearth jacket'
32
'.
81
Stave replacement activities are reported for the prolongation of Kashima No. 3 blast furnace for
its second campaign'
9,31
'. Stave replacement in the shaft was conducted at the Kashima No. 3
blast furnace, but, in the replacement of staves at the bosh, it is important to secure a working
space in the furnace because of a limitation in minimizing stock level. The limit is the exposure of
a tuyere, and it is necessary to enter into blow-out after charging a sealing material (ballast), to
secure visibility in the furnace during repair work at blow-out period. Minimizing both the stock
level (to the tuyere level) and the amount of sealing material allows the required work space.
However, considering the problem of variations in the level of the charged matter in the
circumferential direction, along with a rise in its level by the removal of sticking matter, methods of
removing the charged matter after entering blowout period were also studied (raking the charged
matter out of the tuyere, etc.)'
9
'.
Based on the above results, replacement of three staves, having serious wear, was conducted in
March 1989 at the Kashima No. 3 blast furnace. Although there was a slight adhesion in front of
the staves, the level of the charged matter in the furnace almost agreed with that planned. In
addition, staves could be replaced without removing the charged matter in the furnace.
Through the establishment of a bosh stave replacement technique, the bosh was no longer the
part which determined the lifetime of the furnace'
9
'.
4.5 An c i l l a r y Equi pme n t
4.5.1 Bl owe r a n d Col d Bl a s t Ma i n
Although few references relate this topic to blast furnace life, it is important to avoid blower and
cold blast main failures that would take a furnace off blast suddenly, in an uncontrolled manner.
In such circumstances, slag may enter tuyeres and blowpipes, and, in exceptional cases, the
bustle and hot blast mains. Such problems necessitate long unplanned periods off blast, which
have a detrimental effect on campaign life. To reduce the possibility of such failures, a standby
blower, good maintenance, and routine inspection of the cold blast system are necessary.
Insulation of the cold blast system, often carried out to save energy, may protect the steelwork but
increases its working temperature and may make routine inspections more difficult.
4.5.2 Hot St ove s a n d Hot Bl a s t Ma i n
High temperature hot stoves with external combustion chambers, for hot blast temperatures of
over 1200C, are state of the art technology. The newly constructed Schwelgern No. 2 blast
furnace (14.9 m hearth diameter), commissioned in October 1993, was equipped with three hot
blast stoves with external combustion chambers'
34
'. They are equipped with ceramic burners and
each has a heating surface of 109,000 m
2
. The blast temperature, of a maximum of 1300C, is
regulated in a spherical mixing chamber of a type which had already been successfully used at
Schwelgern No. 1 blast furnace (13.6 m hearth diameter). A special protective coating is used to
prevent the occurrence of intercrystalline stress corrosion on all the process vessels and
pipework. If necessary, external insulation can also be fitted at a later date.
Not only for the hot stoves but also for the hot blast main system, attention has to be paid to
operating safely and economically. Concerning the fixed point of hot blast main shell to the blast
furnace, it is well known that there are only a few furnaces where equal expansion of the bustle
pipe to the furnace centre is guaranteed. It is recommended to use a large number of three point
hangers at the bustle pipe'
35
'.
82
The equal ring expansion will be achieved by small vertical movements. With this design, an
optimum of operational safety at the tuyeres is guaranteed, along with easy access to change the
tuyeres and the tuyere holder'
35
'.
Tuyere stocks with multi-layer compensators are characterized by low pressure losses and good
flow characteristics, thus ensuring low heat radiation with optimal lining'
36
'.
Concerning the fixed point of the hot blast main to the stove, it is known that the hot blast main
shell expansion is great, as it has to cover long distances between the blast furnace and the first
hot stove. With a three-link arrangement, expansion and refractory lining problems can be
overcome'
35
'. If this three-link arrangement is not possible horizontally, it may be possible
vertically, Fig. 4.11.
Normally the nozzles of the hot blast main and the hot blast nozzle of the stove are on the same
axis. Depending on the length of the hot blast valve expansion joint, movement in this area is
normally limited to 10 mm'
35
'. For this reason, fixed points must be built in front of or between
the stoves which are able to take the thrust of the expansion joints as well as the frictional force of
the refractory lining. Depending on the size of the hot blast main system, these forces could be
between 60 and 1201'
35
'. If no fixed points exist, there will still be a kind of restrictive measure due
to the build-up of a centre of mass at the centre of the hot blast main arrangement, whereby the
expansion moves both in the direction of the furnace and the last stove. This will generate many
problems at the bustle pipe and the hot blast valve expansion joints'
35
'.
In modern blast furnace stoves, there are three to six hot blast valves (depending on the plant
system and the number of stoves) installed as stove shut-off valves, blast furnace isolation valves
and vent valves for the bustle main. These hot blast valves usually operate at hot blast
temperatures of up to 1500C and at blast pressures of up to 5.5 bar absolute. Current designs of
hot blast valves are, for the most part, steel fabricated and water cooled'
37
'. A hot blast valve was
developed in which air instead of water could be used as the cooling medium'
37
'. For the cooling
air supply, provision is made to take a small quantity from the cold blast before the stove block
and feed it to the hot blast valves for cooling purposes. During its flow through the hot blast valve,
the air is warmed and it is subsequently re-fed into the hot blast. This design, unlike the
traditional water cooling system with considerable heat energy being continuously extracted from
the stove process, reduces energy losses considerably. A further advantage of air cooling
compared with water cooling is that, in the event of leakages, there is no water to flow into the hot
blast main. Therefore using air eliminates the possibility of damage to the refractory lining of the
hot blast valves, hot blast mains and combustion chamber'
37
'.
Concerning troublefree hot stove operation, a burner with good combustion performance is one
prerequisite'
38,39
'. Stove burner replacement can be done cold within three months, or hot within
two weeks'
38
'. The repair of damaged stove burners and the implementation of a computer model
to adjust stove combustion settings enables more efficient stove operation and allows the hot blast
system to provide higher hot blast temperature'
40,41
'. In general, pulsation and stove vibration has
to be controlled'
38
"
42
'.
4.5.3 Hot Bl a s t Con t r ol Va l ve s
NKK and Asahi Glass Co Ltd jointly developed a hot blast control valve (HBCV) which can control
the flow rate of the hot blast at each tuyere of a blast furnace'
43,44
'. The structure of the valve is
shown in Fig. 4.12 and the installation in Fig. 4.13. It was tested at Keihin No. 2 blast furnace.
With these valves, it is possible to control the circumferential balance of the stockline level at the
top, which led to improvements in furnace conditions'
44
'. Also innovative furnace operations, like
83
pulse blowing and new tapping control, had been developed'
44
'. Operation with HBCV enabled the
blast furnace operators to adjust the speed of burden descent, to remove scaffolds on the furnace
wall and to extend tapping time. As a result of the reduction of tapping frequencies, blast furnace
operation was more stable. No. 1 blast furnace of Keihin Works was equipped with hot blast
control valves at all tuyeres in 1989. Also No. 4 blast furnace of Newcastle Works of BHP Steel
had HBCV fitted for ten of its twenty tuyeres'
45
'.
4.5.4 Ca s t Hous e Equi pme n t
4.5.4.1 Run n e r s
With the increase of production, the flow rate of hot metal and slag in the runner system has risen.
The main troughs in Germany normally have the dimensions of 15 m length, 1.9 m width to back
lining, 0.95 m depth and 0.4 m wall thickness of wear lining'
46
'. The main trough width of 1.9 m is
sufficient to adjust good flow characteristics of the flow channel if the thickness of the wear lining
is kept low. The main trough depths are 0.9 m because the blast furnaces are equipped with a
pool system'
46
'. Using pooling design compared to conventional design of main trough resulted in
a decrease of hot metal velocity adjacent to the trough walls and increased the residence times
for slag/hot metal separation'
47)
.
The durability of the refractory materials is determined by the mechanical, chemical and thermal
load.
The requirements concerning the refractory material are determined by'
46
':
The resistance to chemical attack from slag and hot metal.
The erosion and oxidation resistance of the individual constituents of the trough
material.
The volume stability at high temperatures.
Resistance to thermal shocks.
Ease of working.
High-purity raw materials are used to make suitable construction materials. Care is taken to
ensure that they are chemically neutral in contact with slag and hot metal. For the materials used
today, fused corundum, bauxite, andalusite and chamotte are the basic raw materials. SiC and
Si
3
N
4
are added as anti-oxidation agents. Furthermore, metallic materials such as aluminium and
silicon powder are sometimes needed to bind the materials. Mainly water is added as a liquid
phase.
To mix the casting materials on the spot, it is absolutely necessary to check the mixing time and
the addition of water precisely.
A non-toxic ramming mix and non-hazardous, low moisture castable has also been developed for
use in the blast furnace trough, hot metal and slag runners'
48
'.
For blast furnaces equipped with only one taphole, the installation of a changeable trough system
is a suitable solution'
46,49
'.
84
The lining techniques for the main troughs have developed in parallel with the improvement in the
quality of the trough materials from normal, monolithic materials with ramming to the very modern
casting technique'
46
'.
4.5.4.2 Ta ph ol e Dr i l l i n g a n d Gun Ma c h i n e
It is recommended that taphole drilling and gun machines should be installed where they are safe
concerning fire, dust and dirt'
50
'. In addition to this, availability must be high and maintenance
easy, to avoid casting delays.
4.5.4.3 Ca s t Hous e De -dus t i n g
Cast house dedusting of blast furnaces is not required by the authorities in most countries. In
those where it is required, full availability of the dedusting system is necessary to ensure
continuity of blast furnace operation. Due to ever tightening environmental legislation, a reduction
in cast house fume emission will become necessary for some operators, during the current
furnace campaign.
The dust sources that need extraction to provide dedusting of a cast house are the taphole, the
main trough, the slag skimmer, the hot metal and slag troughs as well as the hot metal and slag
transfer point'
51
', Fig. 4.14. At full exhaust of the trough systems by means of hoods arranged
above the troughs, volume flows of up to 1 million m
3
(S.T.PVh are attained, and cleaned in
electrostatic precipitators or bag filters. Electrostatic precipitators require higher capital
expenditure than bag filters, but have lower operating costs, as bag filters are operated with
higher pressure losses. Bag filters cause problems in terms of temperature capability, because
waste air with a temperature of over 150C has to be cleaned upon opening the taphole. In the
operation range of the taphole guns, the trough is often covered with a lid.
Another way to keep the casthouse clean is to prevent hot metal from coming into contact with
atmospheric oxygen, by the injection of inert gas, e.g. nitrogen, via hoods arranged over the main
and hot metal troughs, and the transfer points'
52
'. This suppresses oxidation of the metal and thus
the formation of dust. If the taphole area, during opening and closing, and the slag troughs and
runners have also to be cleaned, additional extraction must be performed there'
51
'.
4.5.4.4 Ga s Cl e a n i n g Pl a n t
The top gas cleaning system of a modern blast furnace normally consists of a dustcatcher, the
cyclone group and the wet scrubber. Pressure losses occur in the dust catcher and cyclone
group, as well as in the venturi scrubber. Operating a blast furnace with higher chlorine input, for
example through injectants, can cause corrosion problems in the washer system. Therefore
special importance was given to the cyclone dust catcher and washer of the new Schwelgern No.
2 blast furnace of Thyssen Stahl AG, Fig. 4.15'
34
'.
In the past, gas cleaning systems were designed as doublestrand systems. Following the positive
results obtained during the past few years with a single strand system (1 cyclone/1 scrubber),
Thyssen decided to have a single strand system built with 1 cyclone and 1 scrubber, each for a
gas flow rate of up to 750,000 m
3
/h.
Particular importance was attached to the wear of the single strand cyclone/scrubber system. At
the entry-end elbow and the horizontal line, protection is provided by means of Kalcor or by way of
gunning, depending on the excessive erosion.
85
On the cyclone, the entry-end duct was coated with fusion-cast basalt. Kalcor-type protection was
used at the tangential entry end and half of the perimeter of the bottom-end exit end, on the baffle
plate and the lower-end cone. The roof and the rest of the cylinder are protected by means of
fusion-cast basalt. The connecting lines between the cyclone and the scrubber are coated with
fusion-cast basalt. In the scrubber, the entry-end cone is coated with acid-resistant material, and
the scrubber vessel is completely lined with a coat of Keraflake. Starting from an untreated gas
dust content of 20 g/m
3
(S.T.P.), the gas dust content after treatment downstream of the scrubber
is < 5 mg/m
3
(S.T.P.), with > 75% of dust precipitation achieved in the cyclone.
In the cyclone, the deposited dust is discharged by passing through 3 cone locks with one
intermediate bin each, through dust humidifiers and double mixing screws, before being taken
100% to the sintering plant.
The wet gas cleaning is done in two steps. The first step is the pressure zone scrubber, where
the gas is cooled and precleaned. Following this, the gas passes into the annular-gap scrubber
for fine cleaning. Downstream of this stage, there is a two-stage water trap. Separation of the
dust-laden, residual drops occurs in the mist collector.
The ring gap scrubber not only ensures the fine cleaning of the gas, but also maintains the top
pressure'
34
'.
It has to be mentioned here, that the Subcommittee Blast Furnace Operation of the VDEh Blast
Furnace Committee has discussed intensively, in its meeting on April 27, 1995, the corrosion
problem of wet gas cleaning systems caused by higher chlorine input into the blast furnace via
coal injection. The minutes of the meeting and the individual contributions have not been
published, but it can be summarized that, depending on the rate, the injection of coal introduces
an extra 0.2-0.3 kg/tHM of chlorine into the blast furnace and that this causes corrosion problems
at the wet top gas cleaning system. This makes coating of the washer necessary, as well as
additional action concerning the pH value of the water.
4.5.4.5 Ch a r gi n g/Di s t r i but i on Sys t e m
From blast furnaces operated with central gas flow, it is known that top gas temperatures in the
middle of up to 1000C may arise. Cooling of the chute of the bell-less top system may be a
solution.
On bell-less tops, the hopper wear tile life may determine the maximum time between furnace
stoppages. The frequency of stoppages can be extended by improving the materials used'
53
'.
The rotating chute, for example, at Redcar is of the "dirtbox" design. Utilising quality repair
techniques, a life cycle of 20 months has been achieved'
53
'. The Paul Wurth gearbox is critical to
good blast furnace operation. Problems have been experienced at the Redcar blast furnace'
53
'
due to loss of oil from the angle adjustment gearboxes and damage to the slew-ring tooth profile.
Modifications have been carried out to the spare angle adjustment gearboxes to allow
replacement of the segment shaft seals during shutdown conditions, thus avoiding the protracted
operation of changing the gearboxes.
4.5.4.6 Ge n e r a l El e c t r on i c s a n d Con t r ol Sys t e m
Few papers concerning furnace life refer to electronic systems. With increasing campaign life, the
problem of corrosion and deterioration of major electronic systems, that were previously expected
to last the duration of a campaign, is more important. In addition, over a long campaign, the
86
availability of electronic spares may become difficult, as a result of the rapid development of new
technology.
It is advisable to have a stand-by unit for critical control systems, which can have an effect on
furnace life by causing sudden off-blast periods or chilled conditions, e.g. tuyere hydrocarbon
injection or stove control. Backup services, e.g. electrical power or compressed air supply, or the
capability of operating a system in manual control will also reduce the risk.
4.5.4.7 Cool i n g Ci r c ui t s
Build up in cooling systems reduces the flow rate and the heat transfer by conduction. To prevent
internal corrosion and deposition in cooling circuits, chemically treated water is usually used. Acid
cleaning and pressurised water jetting are sometimes used to remove deposits from cooling
pipework and heat exchangers'
20
'.
To increase cooling efficiency, the water flow rate may be increased, reducing the water
temperature at the face of the coolers.
4.5.4.8 Coa l In je c t i on - Me a s ur e me n t t o e a c h Tuye r e
A measure of the blast flow rate at each tuyere is important when tuyere hydrocarbon injection is
used. Should a tuyere aperture become blocked, perhaps by a large accretion, the injectant may
combust in the tuyere stock or bustle main, causing extensive damage and requiring an
immediate, possibly prolonged, stoppage. The pressure drop across the tuyere stocks is often
used for this purpose, with an automatic cut-off of tuyere injectant when a restricted tuyere flow is
indicated.
Measurement of tuyere hydrocarbon injectant at each tuyere is necessary to enable good control
of raceway conditions, for even circumferential operation. To optimise circumferential balance, it
may prove beneficial to match the injectant flow to the blast flow.
Coal flow measurement systems for each line to the furnace or each lance, like Endress & Hauser
and Tribloc, have been tested on several furnaces'
54
'.
Two new developments of coal flow measurement were reported at the meeting of Subcommittee
Blast Furnace Operation of the VDEh Blast Furnace Committee'
55,56
'.
4.5.4.9 Th r oa t Ar mour Pl a t e s
For prolonging blast furnace service life, and to reduce maintenance costs, Nippon Steel has
developed an arrangement for changeable throat armour plates, Fig. 4.16'
57
'. This arrangement is
in practical use in Japan'
17,58,59
'. The plates may be fitted retrospectively, on top of worn throat
armour. Preparations for their installation, involving pressure drilling and welding, can be carried
out during production, with few preparatory shutdowns, followed by a longer (typically 2-3 days)
shutdown. A much longer shutdown would be required to repair traditional throat armour.
87
4.6 References
1. Anon: 'Repairing, rebuilding to extend BF life. A Staff Report', 33 Metalproducing
7/94, pp 26/7
2. Murai, Y.; Mochizuki, S.; Wakuri, S.; Inoue, .: 'Development of technology for
repair of damage to blast furnace stack refractories', 1985 Ironmaking Conference
Proceedings, pp 173/80
3. Anon: 'Gunning for blast furnaces', Steel Times, January 1994, p. 15
4. Keaton, D.: 'Remote gunning technique maximizes top change downtime at Armco's
Amanda blastfurnace', Iron and Steelmaker, April 1993, p. 33/5
5. Bell, P. F.; Duckworth, R. F.; Green, W.: 'Stack renewal by remote control
development of a technique ',1990 Ironmaking Conference Proceedings, pp 389/93
6. Inoue, H.; Niwa, Y.; Sumigama, T.; Maki, .; Ymaguchi, .; Sakai, A: 'Technologies
for prolongation of blast furnace campaign life', 1991 Ironmaking Conference
Proceedings, pp 293/9
7. Kajikawa, S.; Nakajima, R.; Ohzeki, S.; Yoshida, H.: 'Prolonging the life of a large
blastfurnace', Mc Master Symposium No. 10, pp 391/410
8. Anon: 'Success of gunning/rundown procedures of Dofasco blast furnace No. 4',
Responses to BHP Questionnaire, March 30,1993
9. Hiroki, N.; Yamanishi, I.; Koike, .; Tanizawa, Y.; Koba, M.: 'Extending the
campaign life of the Kashima No. 3 blast furnace', Taikabutsu Overseas, Vol. 12,
No. 2, June 1992, pp 23/9
10. Lungen, . .; Vos, D.:' Manahmen zur Verlngerung von Hochofenreisen in den
besuchten Werken in Nordamerika', Fachausschubericht No. 1.024 des
Hochofenausschusses des VDEh, February 1996
11. van Laar, J.; van Stein Callenfels, J. E.: 'Blast furnace campaign life and repairs',
Steel World, Sept. 1991, pp 31/7
12. Campbell, D.A.: 'Extra long furnace campaign life', Paper presented at the
Ironmaking 2000 Symposium, Myrtle Beach, SC, USA, October 3-4,1994 (Compare
also 13)
13. Lungen, . .: 'Ironmaking 2000', Cokemaking International, Vol. 7, 1/1995,
pp 10/20
14. Dzermejko, A. J.; Hoelpes, G.; Sliepen, J.; Solvi, M.: 'Design considerations for
utilizing cylindrical cooling elements in the blast furnace', 1988 Ironmaking
Conference Proceedings, pp 489/504
88
15. Hyle, F. W.; Jefferson, G. K.; Yearsich, T. G.; Helenbrook, R. G.; Cox, I. J.: 'Bosh
repairs to U. S. Steel's Gary No. 13 blast furnace ', Ironmaking Proceedings, Vol.
45, II, 1986
16. Storm, L. P.: 'A short mid-campaign bosh repair at Sparrows Point "L" blast furnace',
1994 Ironmaking Conference Proceedings, pp 323/32
17. Kobayashi, K.; Matsumoto, T.; Yanagisawa: 'Technology for prolonging campaign
life of blast furnace', Kawasaki Steel Technical Report, No. 32, March 1995, pp 9/18
18. Terayama, O.; Funakashi, T.; Hoshikuma, Y.; Nunomura, Hirota, N.: 'Technical
improvements to keep the high productivity operation for long period at Kure blast
furnace', Proceedings of the First International Congress on Science and
Technology of Ironmaking, 1994, Sendai, ISIJ, pp 456/61
19. Alleyn, A. G.; Munive, L.: 'Hearth behaviour of Port Kembla blast furnaces',
Proceedings of the International Hearth and Raceway Symposium, 1981,
Wollongong, pp 111/1122
20. Lathlean, D.; Mellor, D.; Mitchell, S.; Tanzil, F.: 'Experience with hearth drainage
and the influence on stable operations at BHP Port Kembla blast furnaces',
McMaster Symposium No. 19, May 1991, pp 204/36
21. Yamamoto, R.; Nakajima, R.; Koyama, Y.; Niiya, K.: 'Ten years life of No. 5 BF in
Fukuyama', AIME 1985, pp 149/63
22. Kowalski, W.; Bachhofen, HJ. ; Rther, P.; Ballewski, T.; 'Intermediate hearth repair
at Thyssen Stahl AG', 55th Ironmaking Conference Proceedings, Pittsburgh, 1996
23. Wilms, E.; Janz, J.; Lungen, . .; Peters, M.; Schmle, P.: 'Prolongation of
blast-furnace campaigns - factors which affect the life of the furnace hearth', Stahl
und Eisen 113 (1993), Nr. 6, p. 81/92
24. Koike, .; Takata, .; Kariya, J.; Kojima, M.; Sato, K.: 'Extending the campaign life
of Kashima No. 3 blast furnace', Proceeding of the 2nd European Ironmaking
Congress, Glasgow 1991, pp 155/67
25. Yamashita, M.; Kashiwada, M.; Shibuta, H.: 'Improvement of tapholes at Wakayama
No. 5 blast furnace', 1995 Ironmaking Conference Proceedings, pp 177/82
26. Macauley, D.: 'Current blast furnace design philosophies', Steel Times International,
March 1995, pp 20/3
27. Kanetsuna, I.; Akimoto, E.; Saito, S.: 'New Blast Furnace Revamping Technology',
Iron and Steel Engineer, June 1990, pp 32/7
28. Leidl, R.; Stock, G.: 'Schweiverfahren, angewandt beim Bau der druckfhrenden
Anlagenteile des neuen Groraumhochofens der Voest Alpine im Werk Linz',
Schweitechnik Nr. 9, Sept. 1976, 30. Jahrg., pp 169/73
29. Pass, H.: 'The electroslag welding of a blast furnace hearth jacket', Welding and
Metal Fabrication, January/February 1976, pp 21/7
89
30. Gifford, .; Baker, D.: 'Rebuilding the blast furnace shell at Redcar', Metal
Construction, May 1987, pp 268/73
31. Senga, Y.; Kosaka, T.; Nishizawa, S.; Yoshioka, H.: 'Replacing staves at Wakayama
No. 4 blastfurnace', 1991 Ironmaking Conference Proceedings, pp 281/86
32. Chemorov, M.; Gale, J. P.; VanderSluis, F. H.: 'Hearth wall and stave repair at
Lorrain No. 4 blastfurnace', 1992 Ironmaking Conference Proceedings, pp 255/9
33. Oltolini, G. P.; van Stein Callenfels, J. E.; van Laar, J.; Tijhuis, G. J.; de Pagter J.:
'Open hearth surgery and subsequent resuscitation of Comigliano BF2' , 1992
Ironmaking Conference Proceedings, pp 329/36
34. Schulz, E.; Kowalski, W.; Bachhofen, HJ. ; Peters, K.H.; Wilms, E.; Land, S.:
'Planning, construction, start-up and initial operating results of Schwelgern No. 2
blastfurnace', Stahl und Eisen 125 (1995) Nr. 11, 41/54
35. Cronert, W.: 'Modernization of hot blast main system', Fachberichte Httenpraxis
Metallweiterverarbeitung, Vol. 24, No. 8,1986, pp 569/75
36. Nickel, F.: 'Dsenstocke als flexible Verbindung zwischen Heiwindringleitung und
Blasformen am Hochofen', Fachberichte Httenpraxis Metallweiterverarbeitung, Vol.
23, No. 7, 1985, pp 488/91
37. Seeger, J.; Kuckertz, W.: 'Air cooled hot blast valve', Ironmaking and Steelmaking,
1989, Vol. 16, No. 3, pp. 138/43
38. Hodgson, W. R.; Robertson T.: 'Improvements to stove combustion at Redcar',
McMaster Symposium No. 15, May 1987, 137/51
39. Harp, G.; Klima, R.D.; Sucker, D.: 'Effect of operational measures on the formation
of corrosive condensates in hot-blast stove operation', Stahl und Eisen 110 (1990)
No. 6, pp 121/7
40. Greenwald, P.; Quinsberry, P.: 'Blast temperature control at No. 7 blast furnace',
McMaster Symposium No. 15, May 1987, pp 235/45
41. Mohnkern, .; Vo, M.; Janz, J.; Dhl, K. W.; Kuhn, P.; Sucker, D.: 'New optimising
Cowper process control at Klckner Stahl Bremen', Stahl und Eisen 114 (1994) No.
5, pp51/7
42. Snowden, P. F.; Ballard, P. D.; Devin, G. L.: 'Blast temperature considerations for
the Port Kembla blast furnaces', McMaster Symposium No. 15, May 1987,
pp 265/75
43. Anon: 'Ceramic valve for controlling hot blast', Steel Times, December 1985, p. 598
44. Nakajima, R.; Kishimoto, S.; Hotta, H.; Ishii, K.; Kobiyama, H.; Yamaoka, Y.:
'Innovative technology of blast furnace operation with hot blast control valves', NKK
Technical Review No. 59 (1990) pp 1/7
90
45. Anon: 'Hot blast control on Newcastle No. 4 blast furnace', Steel Times International,
March 1994, 27
46. Rther, .P.; Lungen, . .: 'Refractory technology and operational experience with
tapholes and troughs of blast furnaces in the Federal Republic of Germany',
Metallurgical Plant and Technology 3/1989, pp 12/29
47. Cameron, . .; Tudhope, J. M.: 'Improved trough design using water modelling',
1988 Ironmaking Conference Proceedings, pp 505/15
48. Banerjee, S.; Shah, S.: 'Nonhazardous high performance blast furnace casthouse
refractories', I & SM, March 1990, pp 89/93
49. Rusila, H.; Haimi, S.; Ojala, K.: 'Casthouse practice with exchangable runner
system', Iron and Steel Engineer, June 1990, pp 52/5
50. Stichlochstopfmaschinen an Hochfen; Fachausschubericht Nr. 5.021 des
Ausschusses fr Anlagentechnik des VDEh, Dsseldorf, August 1980
51. Lungen, H. B.; Theobald, W.: 'Environmental measures in European sinter plants
and blast furnaces', Proceedings of the 2nd European Ironmaking Congress 1991,
pp 275/88
52. Grtzmacher, K.; de Haas, H.; Mohnkern, H.; Ulrich, K.; Kahnwald, H.: 'Dust
suppression in blast-furnace casting bays', Stahl und Eisen 101 (1991), No.3,
51/63
53. Moore, .; Bowler, .: 'Maintenance an integral part of the "Process"', 1992
Ironmaking Conference Proceedings, pp 625/35
54. Cappel, J.; Geerdes, M.; Langner, K.; Lungen, . .: 'Present state of blast furnace
coal injection technology in VDEh member works', Metallurgical Plant and
Technology 5/1988, pp 20/32
55. Block, F. R.; Nowak, H. G.; Schmitz, R.: 'Elektrische Metechnik zur Bestimmung
der Massenstrme pneumatisch gefrderter feinkrniger Kohle', Paper presented at
the subcommittee Blast Furnace Operation of the VDEh Blast Furnace Committee
on October 25, 1995
56. Eckstein, HJ. ; Fabinder, H.G.; Kupfer, R.: 'CORIOLIS Meter fr genauer
Massenstrommmessungen', Paper presented at the Subcommittee Blast Furnace
Operation of the VDEh Blast Furnace Committee on October 25,1995
57. Kase, M.: 'Technical progress in long life of blast furnace campaign', Tetsu to
Hagane, 68,1982, No. 2, pp 200/3
58. Kamano, H.; Kawamura, M.; Ichifuji, K.; Matsumoto, T.; Marushima, H.; Obata, H.:
'Long term stable operation of Chiba No. 6 BF', 1989 Ironmaking Conference
Proceedings, pp 515/21
59. Ayuha, .; Mizuno, Y.; Yamada, K.: 'Technique for prolongation of blast furnace
campaign life', La Revue de Metallurgie, August/September 1988, pp 661/73
91
TABLE 4. 1
EXAMPLES OF CAMPAI GN EXTENSION POSSIBILITIES
Area
Upper Stack
Middle Stack
Middle &
Upper Stack
Lower Stack
Lower, Middle
& Upper Stack
Belly/Lintel
and Bosh
Tuyere
Surrounds
Tapholes
Hearth
Sidewalls
Practice
Remote gun
in 3 days
Rebrick in
21 days
Remote gun
in 4 days
Rebrick in
21 days
Remote gun
in 5 days
Rebrick in
21 days
Remote gun
5 days
Grout every 8
weeks, 'A day
each grouting
Rebrick in
21 days
Remote gun
in 6 days
Rebrick in
26 days
Remote gun
in 6 days
Rebrick in
28 days
Rebrick
Upper 'A
Rebrick
complete
Replace from
exterior in
14 days
Gun and
local repairs
Rebrick in
30 days
-*
*
~*
*
*
-"
-*
-*
-
-*
"*
-*
-


-*
+
Result
1 year
extension
5 year
oxtonslon
1 year
exrension
5 year
extension
1 year
extension
5 year
extension
1 year
extension
2 months
extension
5 year
extension
1 year
extension
5 year
extension
1 year
extension
5 year
extension
3 year
extension
5 year
extension
5 year
extension
3 - 5 year
extensions
+5 year
extension
~*
-*
~*


~~
-*
Expanded
12 days down to get to 5 years
20 days down to get to 5 years
20 days down to get to 5 years
20 days down to get to 5 years
2 - 3 years maximum duration
24 days down to get to 5 years
24 days down to get to 5 years
92
TABLE 4. 2
OUTLI NE OF UPPER STACK LI NI NG REPAI R TECHNIQUES
APPLI ED TO BLAST FURNACE OF NI PPON STEEL
Hex frame
gunning
Precast panel
Water-cooled
iron casting
Gunning
Machine
Hex Fram'i
( st eel )
Prccas t
Panel
Grou t
Water-
cooJed
Iren
Cast ing
Grout ing
Feature Hex frame is
fixed to
shell to
support cast-
able to be
gunned and
castable is
gunned onto
hex frame
while furnace
is shut down.
Castable is
vibration
formed in hex
frame and
dried.
Precast panel
is installed
where brick
lining is
damaged and
castable is
grouted into
gap between
panel and
brick lining.
Water-cooled
iron cast-
ings are
installed
where brick
lining is
damaged and
castable is
grouted into
gap between
water-cooled
iron casting
and brick
lining.
Profile
repair
accuracy
Satisfactory
Good Good
Durabil-
ity
6 months
1 year or
more
4 years
or more
93
TABLE 4.3
COMPARISON OF UPPER STACK LI NI NG REPAIR METHODS
Repair method
Material
Qual i t y
Water added
Repair method
Dr yi ng
Cunni ng
Chamotte
10 % 12%
Cunni ng
Not dr yi ng
Precast panel
Corundum
6 %
Precasting by
vibration
350C 45h
Repair method
Qual i t y
Apparent porosity
(%, at 500
e
C)
Hot modulus of r upt ur e
( k g/ c m
2
, at 1200C)
Wear volume
( cm
1
, at 800
e
C)
Resistance to spelling
( 1200
e
C wat er x10)
Resistance to alkali
(%, 1200
e
Cx5hx5)
Cunni ng
Chamotte
30 \, 32
30 ^ 40
not mesure
Cruck at
2 times
not mesure
Precast panel
Corundum
18.4
150
2.4
Good
(Hai r cruck)
+ 7.7
Bri ck ( r ef . )
AUO,
15.6
130
3.8
Good
( Hai r cruck)
14.8
94
IO
UI
DOF ASCO BLAST FURNACE NO. 4 GUNNING EQUIPMENT
FIG. 4.1
"!!. !
3 '
I P " Coding Pi o e ^
B3 i
o u t l e t
/
w i t t r e ~J[
O i n et
Injection i B2 j
i 11]
o u t l e t
If
^ Gunning \
Injection
Steve f i
liivsj
Wi t e r i n l e t
1 \ ^ Cooling Plate
FIG. 4.2
CYLI NDRI CAL COOLER, SORELOR TYPE, WITH HI GH
VELOCI TY HELI COI DAL WATER FLOW
FIG. 4.3a
MULTI PLE MOUNTED CYLI NDRI CAL COOLERS USED AS FI G. 4.3b
REPLACEMENTS FOR CONVENTI ONAL FLAT COOLING PLATES
96
Furn a c e Sh e
Gr out Nozzl e
Twi n Cyl i n dr i c a l
Code r s
Common Moun t i n g Ra t
Gr out An c h or Ba r s
CYLI NDRI CAL COOLER MOUNTI NG SCHEME WITH
BUI LT- I N, REFRACTORY GROUT I NJECTI ON NOZZLE
FI G. 4.4
Gr out Nozzl e
Skul or
Re s i dua l
Re fr a c t or y
Gr out Impr e gn a t e d
Bur de n
Gr out An c h or Ba r s
TWI N MOUNTED CYLI NDRI CAL COOLER WITH
BUI LT- I N GROUT I NJECTI ON NOZZLE, FOR " I N- KI ND"
REPLACEMENT OF FLAT COOLER
FI G. 4.5
97
COPPER TO STEEL
WELD
ORIGINAL STAVE
HOT FACE
SHELL
4" CYLI NDRI CAL COOLER I NSTALLATI ON FI G. 4.6
t h e r moc oupl e
r e fr i ge r a t or
TEMPERATURE CONTROLLI NG SYSTEM OF
COOLI NG WATER
FI G. 4.7
98
NO. 4 BF ARRANGEMENT EMERGENCY UNDERHEARTH
COOLI NG PIPES
FI G. 4.8
Year ' 7778 79 ' 80*81 ' 82' 83' 84' 85' 86' 87 ' 88*89
Bottom 40C
(C) 200
Wall 300
(Max) 200-
CO 100
Titanium 15
additives 10
(kg/pt) 5
Intensificador
of cooling
(Grouti
inq
Intensification (of water sproy
Life of BF 2 3 4 5 6 7 8 9 10 II 12 13
FURNACE WEAR AND COUNTERMEASURES I N THE
REPAI R AT KASHI MA NO. 3 BF
FI G. 4.9
99
Gunning refractory / 1 /
lining to upper holder t!l\ ^1^1
Casting refractory
lining to outer tap hole
Casting refractory
lining to inner tap hole
TSS
MET HOD OF T A P HOL E CASTI NG FIG. 4.10
Ho
Q
* *
Vf
< s >.
^
V .J .
Stove No. 4
Stove No. 2 ,
SS*
1
vi
^
%
V.
.*"
./Bat
4 ^ m / ' >5L_I **" I im Gerst ' **.-2SS'
M
_ o
P ; " *
i t ^ - H l
%
' ' ' \ - - . '
m Ge, a
v ' ' " " ^ Stove No. 3
^ ^ e y ^ \ W p r > ' ^ ' , ' ' ^ ' * Inclined Sliding Point
^ " * <~$s> "'
!
' **"' Support for Longitudinal
Stove No. 1
and Height Compensation
Three Point
Suspension
COMPL ET EL Y COMPENSATED HOT BLAST MA I N SYSTEM FIG. 4.11
100
Coupling Heat insulating caster
STRUCTURE OF HBCV
FIG. 4.12
Down leg
Hot blast
control valve r
Controller
Pressure of hot blast main
Recorder
-j "
Pressure difference
transmitter
Penstock
Blow pipe
SCHEMATI C DRAWI NG OF I NSTALLATI ON OF HBCV FIG. 4.13
101
CAST HOUSE DEDUSTING SYSTEM OF HKM BLAST FURNACE
(SCHEME: MANNESMANN DEMAG)
FIG. 4.14
Schmelzbasalt
Kalcor
Kaloear
CYCLONE DUST CATCHER AND WASHER OF BLAST FURNACE
SCHWELGERN 2 OF THYSSEN
FIG. 4.15
102
Il
I
Changeable throat
armour plates
GENERAL ARRANGEMENT OF THROAT ARMOUR FI G. 4.16
103
5. FUTURE DESIGNS
5.1 Four t h Ge n e r a t i on St a ve s
The development of stave coolers has led to today's so-called fourth generation of staves,
Fig. 5.1
(1)
, of Nippon Steel Corporation. The characteristic features are additional cooling pipes in
the stave corners, an additional meandering shaped (serpentine) pipe on the shell side and
extended holding ribs for the refractory lining. This ought to make the protective brickwork
superfluous. It is known from blast furnace inspections that 3rd generation staves showed that
cracking and excessive wear occurred at the corners and upper and lower ends of the staves'
2

3
'.
In the fourth generation, corner pipes were incorporated to eliminate these problems. The
addition of the meandering cooling pipe serves as a backup cooling system, in the event of failure
of any of the other cooling pipes.
A further development of 4th generation staves of NSC, where the furnace lining is attached to the
stave, secured by tapered ribs, in the foundry, is shown in Figs. 5.2 and 5.3. Taking into
consideration the fact that the wear speed of cooled stave metal is 1/10 or less than that of the
brick in front of the stave metal, the brick thickness is reduced, from 450 or 600 mm, to 200 mm,
and the stave thickness was increased from 340 mm to 400 mm, so that the life of the front brick
and stave would remain almost unchanged, Fig. 5.4'
4,5
'. Figure 5.5 shows that the improvements
in stave coolers, especially the 4th generation stave, have greatly increased the durability of
staves'
4
'.
The general advantages of the 4th generation staves are:
The refractory lining life is extended to an operation period of up to 3 years.
The cracking and most excessive wear at stave corner were reduced.
The loss of cooling capacity due to pipe failures is countered by addition of a
meandering shaped cooling pipe.
The stave life is increased to 17 years'
5
'.
5.2 Coppe r St a ve s
Staves have the major advantage of cooling the entire area of the blast furnace shell, whereas
cooling plates only cool localized areas'
6
'. . However, damaged cooling plates can be easily
exchanged, whereas it is practically impossible to change staves without blowing down the
furnace.
After detailed discussion, Heinrich et al'
6
' assumed that the advantage to cool the whole area of
the shell is more important, and the disadvantage of not being able to easily exchange the cooling
element has to be overcome by designing it for an extremely long and safe lifetime. This meant
the development of a stave with a target lifetime identical with the lifetime of a blast furnace, i.e.
30 to 50 years, versus 10 to 12 years with state of the art cast iron staves'
6
'.
Experience with cast iron staves has shown wear and loss of material, first at the corners, and
then at the edges'
3
"
6
', as already mentioned. That is because the cast-in pipes must be located
approximately 80 mm from the edges and 110 mm from the corners of the staves, to meet casting
requirements and provide permissible bending geometry'
6
'. These conditions, together with the
relatively low conductivity of cast iron, prevent the edges and corners from being adequately
105
cooled'
6
'. Cooling conditions are further exacerbated by the relatively wide uncooled gaps
between two adjacent cast iron staves'
6
'. Also, cooling problems result from the insulating pipe
coating used to prevent carbon pick-up during casting, which results in higher hot side
temperature, giving cracking and increased wear. To avoid cracking at the corners and edges of
cast iron staves, corner pipes have been installed on 4th generation staves, which meant
additional pipes, holes in the blast furnace shell and additional water and energy requirements.
These problems do not exist with rolled copper staves. Copper plate is rolled to narrow tolerances
and can be machined close to the sides'
6
'. The cooling channels can be drilled accurately, and
located as close to the edges and corners as dictated by the minimal wall thickness needed to
avoid leaks under the water pressure'
6
'. Finally, the gap between staves can be narrow, so that
this critical, uncooled area is virtually eliminated'
6
'. Concerning physical properties, see Table
5.1'
6
', the 10 times higher heat conductivity of copper'compared to cast iron and the fact that, due
to mechanical strength of copper, the temperature should be held below 120C, means that more
heat would be removed from the blast furnace when using copper staves compared to other stave
cooling systems'
6
'. To avoid this effect, MAN GHH chose a design solution where
low-conductivity, high alumina refractory bricks were installed in the hot side of the copper stave'
6
'.
According to operation experience, an extremely stable accretion layer additionally protects the
stave and avoids high thermal losses of the blast furnace.
Results with test copper staves, installed at Hamborn No. 4 and Ruhrort No. 6 blast furnaces of
Thyssen Stahl AG, showed very good results. After 9 years of operation and a production of 12.9
million t of hot metal, the cast iron staves of Hamborn No. 4 blast furnace showed heavy signs of
wear, cracks and exposed cooling pipes, whilst the copper staves had a virtually unused
appearance, with sharp edges and corners'
6
'. The remaining copper material had maintained its
original physical properties. Performance of the second set of copper staves, Fig. 5.6, installed at
Ruhrort No. 6 blast furnace, also showed good results, as the average daily temperatures of the
staves have been between 70 and 80C. The major technical results of copper staves can be
summarized as follows'
6
':
For a copper stave system with three rows of copper staves in the belly/lower shaft
area and cast iron staves in the remaining area, the heat removal and cooling water
circulation rates are comparable with a pure cast iron staves system. In some cases,
the heat removal of copper staves was even lower than that of cast iron staves, due
to the stable accretion layer.
With the installation of copper staves in an existing furnace, there is the possibility of
obtaining additional useful furnace volume because copper staves are thinner than
nodular cast iron staves and cooling plates.
Copper staves, in contrast with cast iron staves and copper cooling plates, do not
require lugs or noses. Copper staves present a smooth surface to the furnace
interior and do not interfere with the charge layers, which would adversely effect gas
permeability in the wall area.
Since stable accretion layers form in front of the copper staves, it is not necessary to
install expensive refractory material which will be replaced by crusts after a certain
time. In a copper stave system, the possible additional life gained by costly
refractory material is negligible when compared with the expected overall life. Staves
can, therefore, be installed with inexpensive refractories.
106
Up to now, good results with copper staves are reported from Thyssen Stahl AG with the new
Schwelgern No. 2 blast furnace (1 row) and from Preussag Stahl AG with the new blast furnace
Salzgitter (3 rows)'
7,8
'. Due to further orders, an additional 8 furnaces will be equipped with
copper staves in the lower stack, bosh and belly area'
9
' in the near future (Sept. 1995), in Belgium,
Germany, Spain, Sweden and the USA. In total, the copper stave cooled furnace is, in its cooling
element related parts, about 10 % higher in investment costs than a blast furnace with a classic
stave system, and 5.4 % more expensive than a blast furnace with densely packed copper cooling
plates. This will be compensated by cheaper refractory material in front of copper staves and by
the longer life time of copper staves.
5.3 Improved Refractories
5.3.1 Blast Furnace Hearth
In principle, the refractory lining of blast furnace hearths constructed in the last decade can be
subdivided into "Black" and "Black/White" refractory lining designs, as shown in Fig. 5.7'
10
'. For
"Black" hearths, either small format (predominantly in North American blast furnaces) or large
format carbon blocks of variable qualities are used. The "Black/White" designs have either
ceramic layers in the bottom under or above the carbon bottom layers, with relined hearth walls
made of carbon blocks or additive with varying heights of ceramic cap facing. Another definition
for a construction of hearth lining is the "thermal solution", using high conductivity refractories
combined with powerful cooling, and the "refractory solution", involving low conductivity refractory
materials'
11
"
14
'.
The lining variant "thermal solution" had to be improved concerning thermomechanical and
thermochemical properties. The properties of the bricks, which are affected by the choice of raw
materials and the brick manufacturing process, largely act in opposite directions and have to be
assessed for their effect in the operating application. Table 5.2 shows a summary of the general
characterisation of carbon bricks.
Improved modern grades of hearth lining bricks based on carbon aim to improve one property
which is considered to be important, if possible without worsening the others. This is achieved
through the choice of the raw material, manufacturing process and metallic and/or ceramic
additives in the raw mixture. Thus the various brick producers work in different ways, often in
close cooperation with blast furnace plants and research institutes'
10
'.
The following have led to new developments'
10
':
Introduction of very finely distributed Al
2
0
3
in the bonding matrix
Impregnation with colloidal Si 0
2
solutions
Addition of Si or Si-containing modifications to the bonding matrix, which form SiC
and/or Si-0-N bonds under suitable burning conditions
Tar impregnation with post-burning
Additives are not usually used in graphite bricks. Metal oxide additives are reduced at high
combustion temperatures. Part or semi-graphite bricks become microporous by such additives.
There are some part-graphite bricks with only Si and SiC additives and some in which Al
2
0
3
powder is mixed during compounding.
107
The microporous amorphous carbon brick is also produced from traditional amorphous carbon
brick based on anthracite, by the addition of silicon or modifications of silicon to the brick
compounding. Al
2
0
3
was also added during compounding by other producers to improve
resistance to dissolution by under-saturated hot metal. The disadvantage, that Al
2
0
3
can react
with infiltrated alkalis resulting in an increase in volume and brick flaking, was tolerated.
Resistance to dissolution in hot metal and to hot metal infiltration is considered more important by
manufacturers and users. With the trend towards even finer pores and towards types of
super-microporosity bricks, the raw material was optimised with regard to the preparation and
purity in compounding. If thought advisable, the compounding also incorporates a larger
proportion of graphite.
Stahlwerke Bremen GmbH, during the last relining of No. 2 blast furnace in 1992, installed, in
combination with a ceramic cup, a European-manufactured amorphous carbon brick which was
compressed by extruder and which, by mixing with Si, has more than 60% of pores of a size
< 1 , Fig. 5.8'
10
'.
Httenwerke Krupp Mannesmann GmbH chose a microporous carbon brick, with around 95 % of
its pores < 1 , for the endangered hearth wall region from the bottom to the taphole, for relining
blast furnace A in 1989, Fig. 5.9'
15
'. The carbon brick was also made resistant to dissolution in hot
metal by increased Al
2
0
3
powder addition. This European-manufactured brick is also installed in
the new blast furnace of Preussag Stahl AG'
16
'.
Thyssen Stahl AG decided on a Japanese-produced microporous carbon brick, based on
anthracite with higher graphite addition in compounding, Fig. 5.10'
17
'. This brick contains Si for
good microporosity, but somewhat less Al
2
0
3
than those mentioned previously. Schwelgern No. 2
blast furnace also got a ceramic cup.
Hoogovens, Umuiden, supports the thermal solution, using semi-graphite and copper plate
coolers, which promotes skull formation and decreases lining temperature'
13
'.
A total of 30 layers of hot-pressed carbon bricks was installed in the Bengang No. 5 blast furnace
(China) in June 1990, Fig. 5.11'
18
'. The potential advantages pointed out are the higher thermal
conductivity, the reduced length of the blocks in the through-wall direction, that the use of cement
releases thermal stress, the improved thermal conductivity of the hearth because bricks are in
close contact with staves, that the low porosity, developed during manufacture by concurrent
heating and pressure, minimizes water and iron penetration, and the high alkali resistance. It was
reported that, after one year of operation, the temperature of the hearth cooling water has
remained unchanged'
18
'.
As a refractory solution'
19

20
' for the hearth, the bricks incorporated into this are characterised by
high resistance to wear mechanisms at high temperatures. It is attempted to minimize the rate of
wear until application of the cooling system from outside. Such carbon bricks are based on the
raw material anthracite'
10
'.
For an anthracite brick the high strength and low solubility in hot metal are mentioned as
advantages from laboratory and blast furnace outbreak investigations. The low thermal
conductivity, the high -modulus and the higher coefficient of thermal expansion compared to
graphite, and lower resistance to alkalis are referred to as disadvantages'
10
'.
A "ceramic cup" has also been suggested, and installed at several blast furnaces, as a refractory
solution for the hearth'
21
"
25
'. Figure 5.12 compares, for new blast furnace hearths, the isotherms of
a conventional hearth and a hearth relined with a ceramic cup. It can be seen from the isotherms
108
calculation, that the 800C isotherm, which is responsible for the formation of the brittle layer in
carbon refractories, lies within the ceramic material. In combination with a sufficient sump depth
of the hearth, the ceramic cup is a further supporting measure to prevent wear of the carbon
blocks, especially the "elephant foot".
During a study journey of members of the VDEh Blast Furnace Committee to USA and Canada,
the problem of hearth durability was also discussed'
26
'. Except in one case, all furnaces of the
seven plants visited are relined in the hearth wall with small dense carbon bricks. It was reported
that no wear problems are registered in the hearth wall. In most cases, the hearth wall wear is
not monitored with thermocouples. Two of the blast furnaces visited had problems in the taphole
area. These were mainly related to blast furnace stoppages. A continuous and smooth blast
furnace operation, as well as good coke quality, are prerequisites for long hearth life. The critical
area was in some cases the bosh and belly.
5.3.2 Refractories above Tuyere Level
In the upper stack, where the wear is mainly mechanical, the refractory blast furnace lining has
evolved over the years from a normal fireclay brick, with 39% Al
2
0
3
, into a dry-pressed, dense
fireclay grade, with around 45% Al
2
0
3
. Its advantage for that area is good abrasion resistance
and high resistance to oxidation'
21
'. Poor thermal shock resistance is a problem, so that in critical
areas of the upper stack various composite linings are being tried, involving bricks with 60% Al
2
0
3
and silicon carbide to improve thermal shock resistance, Fig. 5.13'
27
'. Its adequate durability,
which can be improved only at significantly higher costs by using higher grade bricks, does not
give rise to any expectations of new developments for this zone'
21
'.
The zones below the upper stack, i.e. bosh, belly, lower stack down to the tuyere level, are
additionally exposed to thermal and chemical wear. That means wear by high temperatures, high
and fluctuating heat fluxes and chemical attack by alkalies, zinc, slag, etc. coupled with abrasion.
The wear solutions for these zones can also be divided into thermal solution and refractory
solution.
The thermal solution is graphite, semi-graphite and SiC bricks, with SiC bricks having found
acceptance for this zone because of the hardness of the raw material'
21
'. Figure 5.14 shows, as
an example for the thermal solution, the relining of No. 7 blast furnace of Hoogovens in 1991'
28
'.
There the bosh lining has been upgraded compared to the 1980 relining, from a combined
graphite/semi-graphite system to a full graphite design, because a full graphite refractory design
might better cope with high temperatures and temperature fluctuations'
28
'. From the bosh parallel
up to the middle stack, the refractory system is a combination of graphite and silicon carbide. The
main purpose of the graphite is to cool down the silicon carbide to relatively low temperatures,
while the main purpose of the silicon carbide is to protect the graphite from abrasion'
28
'. As a
refractory solution for bosh, belly and lower stack, the use of Sialon 'bonded corundum has been
developed'
14
', Table 5.3. The conductivity of this brick is good, although it does not vary much
with temperature and is lower than for silicon carbide bricks'
14
'.
An example of a refractory solution using corundum sialon brick is shown in Fig. 5.15 for Hamborn
No. 9 blast furnace of Thyssen Stahl AG'
21

29
'. The brick grades have a corundum grain and a
non-oxide bonding system similar to the SiC bricks. Satisfactory operating results have already
been achieved with these bricks'
21
'.
109
5.4 Mor e Compr e h e n s i ve a n d Re l i a bl e In s t r ume n t a t i on
As the blast furnace hearth is seen as the key factor for the length of a campaign, the monitoring
of the wear progress in the hearth is of great importance. Temperature measurements by
thermocouples are used to monitor the advance of wear in the hearth and the thermocouples are
installed in a grid of varying levels in the hearth wall. NiCr-Ni thermocouples are, of course,
"poisoned" in the temperature range above 600C and can then indicate too low a temperature
level'
30
', so that thermocouples should only be placed where temperatures above some 400C are
not expected. These problems are avoided by using changeable thermocouples with a suitable
protective tube'
10
'.
It should be noted that thermocouples indicate only local wear and comprehensive monitoring is
only possible with an extremely high density of thermocouples. Generally speaking, it is,
therefore, possible to detect only the average wear course. An alternative method of monitoring
wear progress is the use of a thermovisual camera or heat flux probes.
In a research programme'
31
', a series of 12 heat flux probes have been installed in the hearth
brickwork of No. 2 blast furnace of Stahlwerke Bremen, Fig. 5.16, for measuring the heat flux
density directly. The following advantages of heat flux probes compared the thermocouples are
pointed out'
31
':
Heat flux probes can be installed near the blast furnace jacket in colder areas of the
refractory lining. Therefore the durability is higher compared to thermocouples. The
carbon brick refractory lining will not be damaged. Instead of several thermocouples,
only one sensor at a measuring place is needed to determine the heat flux density.
Heat flux probes react more sensibly to damage in the refractory brickwork.
The calculations regarding to an optimum local distribution of the heat flux probes
have shown that a distance between two probes of 3 m seems to be sufficient for
reliable and complete wear control. In order to achieve an equal certainty with
thermocouples, these would have to be placed in the middle of the carbon brickwork
with distances from each other of about 1 m. The wear profile will be calculated on
the basis of the given data on heat flux density.
In another project, the measuring of hearth liquid level using EMF is in the development stage'
32
'.
5.5 Th r oa t Cool i n g
For the upper shaft (throat area), stave coolers were adopted to maintain the furnace profile and
to prevent disorder of the burden distribution coupled with brick wear under the throat armour
(wearing) plates'
33

34
'. The installation of a water-cooled support plate or cooling box for a fixed
throat armour system was performed to secure parts of the armour'
34
"
36
'.
5.6 Impr ove d He a r t h a n d Ta ph ol e De s i gn
The sump depth (distance from the taphole to the bottom surface) of the hearth appears to be the
decisive factor in the wear of the hearth wall by hot metal flow during tapping'
32

37
'. It is
recommended that the deadman should either rest completely on the hearth bottom or float
completely in the hot metal sump'
10
'. To prevent "elephants foot" wear, partial floating of the
deadman should be avoided. Due to bottom wear with time, an object initially resting completely
on the bottom may become partially floating. The sump depth should therefore be chosen from
110
the outset at such a level that the deadman can float, Fig. 5.17'
37
'. The required critical sump
depth depends on the size and the method of blast furnace operation. The molten metal vector
and constant temperature line is shown in Fig. 5.18, for two sump depths'
38
'. The results of the
calculation show that, when the dimensionless hearth depth H/R (H: distance from taphole to
hearth surface, R: hearth radius) is 0.46 instead of 0.29, the peripheral flow is decreased and the
flow speed of molten metal near the hearth decreases. On the other hand, the heat transfer
behaviour near the hearth surface corresponds to the change in flow described above. The
change occurs gradually from convection heat transfer to conduction heat transfer, so that the
temperature of the hearth brick is lower at an H/R of 0.46.
It has to be taken into consideration that changes in the burden composition can lead to a change
in the situation and in the hot metal flows in the hearth on tapping, if the average burden layer
thickness and thus the compressive stress of the burden on the deadman is changed. An
example is the difference between all coke operation, and operating with high oil or coal injection
rates'
10
'. A further possibility for preventing advancing wear of the hearth wall under the taphole is
to taper the hearth jacket and to increase the wall thickness of the refractory lining of this area'
10
',
Fig. 5.19.
Tapping technique and taphole design have a great influence on blast furnace campaign life.
Model calculation and tapping trials were carried out in a 1:10 water model (1.4 m diameter) of a
14 m hearth diameter blast furnace, having a packed bed simulating the deadman'
39
'. From this, it
was recommended that the blast furnace should not be emptied to the point where gas can
escape. The observation of this rate can prevent the formation of wear-promoting eddies in the
bottom-to-wall joint area. Interruptions of the tapping process should be kept as short as
possible. This prevents hot and unsaturated molten metal, from the melting zone, from coming
into contact with the refractory lining, where it may dissolve the carbon brick. Taphole diameter
should be kept small, because this helps to reduce the overall flow rate level.
The influence of the taphole length on the flow stress of the hearth wall was investigated by
calculations and by modelling'
40
'. Figure 5.20 shows the variation of temperature distribution in
the refractory lining at an increasing ratio of taphole length to hearth radius. With longer tapholes,
the point of attack of the flow drifts away to the left and the right of the taphole. The intensity of
the point of attack decreases with increasing distance from the taphole. The process control
objective for blast furnaces is therefore to keep the taphole as long as possible in order to shift the
tapping flow from the hearth wall to the inside of the hearth.
A taphole design was installed at No. 6 blast furnace at Umuiden, where a copper encapsulation
not only makes the taphole region gas tight but also protects it from the ingress of water'
41
'.
In general, taphole bricks should be superior in resistance to alkali, spalling, oxidation, melting in
hot metal, slag, and opening by oxygen'
42
'. In order to stabilise the taphole, AI
2
0
3
-C-SiC bricks,
with good durability, were developed and applied in a Japanese blast furnace, Fig. 5.21. The
developed material shows excellent performance, with the exception of resistance to oxidation. In
addition, carbon blocks have lower resistance to melting in hot metal and lower resistance to
oxidation than the AI
2
0
3
-C-SiC material. It was reported that improvements of these properties
are necessary'
42
'.
111
5.7 Fur n a c e De s i gn s a n d Di me n s i on s
The furnace design has to cope with the reactions and the metallurgy. The height of the shaft has
to be a compromise between opposite requirements. For fuel efficiency, it is desirable to have
extended gas/solid contact, so that use of the sensible heat and reducing power of the gas may
be closer to the theoretical limits'
43
'. It is also very important to have the height of the shaft above
a certain value to ensure that iron ores enter the cohesive zone with a maximum degree of
metallization, so that the hearth will not be chilled to create problems in smelting'
43
'.
A concept of construction of Chiba No. 6 BF is referenced'
34
'. With freestanding blast furnaces,
shaft brackets are provided for use during emergencies. As blast furnaces were scaled up, throat
bracket supports, lintel supports, and shaft bracket supports were adopted in large numbers at
Kawasaki Steel's Mizushima Works and other companies'
34
'. However, with the shaft bracket
support and lintel support, the shell is under constraint at high thermal load parts of the shaft and
belly, and the use of expansion joints in the structure may cause cracks in the shell and other
problems. As a countermeasure, the freestanding type was adopted, but, with the shaft bracket
support used, the load of the shell is locally supported by lower furnace columns when the shell
buckles.
The furnace profile was decided on the basis of results obtained with Mizushima No. 4 BF, from
the viewpoint of the profile and operating results, furnace body refractory wear, stable operation,
and other characteristics of large-scale blast furnaces in Japan. As one noteworthy feature, the
height of the lower part of the furnace below the belly is great. The bosh angle of No. 6 BF was
set at 80, based on the finding that, in dismantling research with previously blown out furnaces,
the erosion angle of the bosh was 68-80, and the finding reported in the Babarykin study that a
funnel-shaped loosely packed bed exists between the deadman and the bosh and has a stabilized
angle of 79-80. Also based on the Babarykin report, the distance from the belly to the iron notch
was increased from that used in the past, and the belly height, bosh height, and hearth height
were set at 2.7 m, 4.1 m, and 5.3 m, respectively.
For preventing advancing wear in the hearth, the construction of the hearth is discussed in
Section 5.6, showing also, by the example of the Schwelgern No. 2 blast furnace of Thyssen Stahl
AG, the taper of the hearth jacket to increase the wall thickness of the refractory lining in the area
under the taphole'
10

17
*.
112
5.8 References
1. Ishikawa, Y.; Yamamoto, T.; Abe, Y.; Okuda, K.: Operation methods and improved
blast furnace cooling techniques for prolongation of campaign life', Proceedings of
the 1st European Ironmaking Congress 1986, Vol. 3, Paper V2.
2. Heinrich, P.; Hille, H.; Richert.K.: 'Blast furnace casing cooling - design and cost of
cooling systems', Stahl und Eisen 106 (1986) Nr. 5, p. 205/10
3. Nash, A. D.: 'Stave utilization in North American blast furnaces', Iron and Steel
Engineer, December 1992
4. Murai, Y.; Yamamoto, T.: 'Progress in blast Furnace plant technology', Nippon Steel
Technical Report No. 35 October 1987, p. 12/18
5. Ayuha, .; Mizuno, Y.; Yamada, K.: 'Technique for prolongation of blast furnace
campaign life', La Revue de Mtallurgie, August/September 1988, p. 661/72
6. Heinrich, P.; Hille, H.; Bachhofen, HJ. ; Kowalski, W.: 'Copper blast furnace staves
developed for multiple campaigns', Iron and Steel Engineer, February 1992, p. 49/55
7. Kowalski, W.; Bachhofen, H. J.; Kannappel, ,M.; Korthas, B.: 'Erfahrungen mit
Kupferstaves bei der Thyssen Stahl AG', Paper presented at the meeting of the
Subcommittee Blast Furnace Operation of the VDEh Blast Furnace Committee on
October 25, 1995
8. Gropietsch, K.H.: 'Erfahrungen und CuStaves am Hochofen in Salzgitter', Paper
presented at the meeting of the Subcommittee Blast Furnace Operation of the VDEh
Blast Furnace Committee on October 25,1995
9. Reference list of MAN GHH
10. Wilms, E.; Janz, J.; Lungen, . .; Peters, M.; Schmle, P.: 'Prolongation of
blast-furnace campaigns - factors which affect the life of the furnace hearth', Stahl
und Eisen 113 (1993) Nr. 6, 81/92
11. Fujihara, S.; Ikeda, M.; Nakai, M.; Tamura, S.: 'High duty carbon blocks', Nippon
Steel Technical Report, No. 41, April 1989, p. 1/6
12. Van Laar, J.; Noort, N.; Tijhuis, G. J.; Trouw, J.; Stolwijk, C. S. M.: 'Stress related
wear phenomena in blast furnace hearths', Proceedings 2nd European Ironmaking
Congress 1991, Glasgow, pp 110-114
13. Van Laar, J.; Tijhuis, G. J.; Spreij, M.; Stokman, R. J. M.: 'Blast furnace lining life a
quantitative analysis of lining/cooling systems', 1994 Ironmaking Conference
Proceedings, pp 417/26
14. Anon: Thermal insulation of a blast furnace', World Ceramic Refractory Journal,
1990, Nr. March/April, pp 24/6
113
15. Altland, R.; Beseoglu, M. K.; Engel, K.; Hning, M.; Winzer, G.: 'Second furnace
campaign and relining of blast furnace A in the Huckingen steelworks', Stahl und
Eisen 112(1992) Nr. 1, pp 23/30
16. Sthler, .; Jacob, .; Zischkaie, W.; Gropietsch, K.H.: 'Neubau des Hochofens
der Preussag Stahl AG in Salzgitter', Stahl u. Eisen forthcoming
17. Schulz, E.; Kowalski, W.; Bachhofen, H. J.; Peters, K. H.; Wilms, E.; Land, S:
'Planning, construction, start-up and initial operating results of Schwelgern No. 2
blastfurnace', Stahl u. Eisen 115 (1995) Nr. 11, pp 41/54
18. Lin, .; Lin. F.; Zhang, L.: 'Blast furnace carbon block wear and application of hot
pressed carbon brick at Bengang', Iron and Steel Engineer, December 1992,
pp. 36/9
19. Knig, G.; Weidenmller, C; Pietzko, G.: 'Wear of refractory blast furnace walls.
New conceptions and their application', Stahl und Eisen 91 (1971), No. 2, pp 63/9
20. Kuster, D.; Schoennahl, J.: 'Thermochemical attack in blast furnace refractory
linings', Steel Times No. 1 (1988), pp 42/6
21. Fusenig, R.; Still, G.: 'Refractory materials for the lining of blast furnaces', 4th
German -Japanese Seminaron Refractory Materials, November8, 1989, Dsseldorf
22. Anon: 'Ceramic cups to extend BF hearth life', Steel Times, January 1989, pp 17/18
23. Delavos, G.; Morti, G.: 'Experience with a ceramic Picrochromite cup in the hearth of
blast furnace A, Voest Alpine, Linz', Proceedings of XXXIst International Colloquium
on Refractories, Aachen, 1988, pp 131/41
24. Graffeuille, F.; Jager, J. M.; Capelani, R.; Libralesso, J. M.: 'Hearth design and
control of the Sollac Dunkerque BF 4', 1990 Ironmaking Conference Proceedings, pp
289, 97
25. Bachhofen, H. J.; Bauer, J. M.; Schoennahl, J.; Spindeler, H.: 'Development and
experience with ceramic linings in blast furnace hearths of Thyssen Stahl AG',
Proceedings of the XXXIst International Colloquium on Refractories, Aachen, 1988,
pp 142/67
26. Lungen, H. B.; Vos, D.: 'Manahmen zur Verlngerung von Hochofenreisen in den
besuchten Werken in Nordamerika', Fachausschubericht Nr. 1.024 des
Hochofenausschusses des VDEh, February 1996
27. Whiteley, P. G.: 'Developments in critical areas of blast furnace linings', Steel Times
International, November 1990, pp 30/2
28. Koen, W.; van Laar, J.; Noort, N.; Schmidt, K.; Stolwijk, C; van Strhn, J. 'The
modernization and reline of Hoogovens No. 7 blast furnace', 1992 Ironmaking
Conference Proceedings, pp 275/81
114
29. Wilms, E.; Vogelpoth, . .; Knig, G.: 'Development and application of
chrome-doped refractory materials in the blast furnaces of Thyssen Stahl AG',
Proceedings of the XXXIst International Colloquium on Refractories, Aachen 1988,
pp 202/19
30. Harz, .: Dr.lng. Dissertation (Dr.lng. Thesis), TU Clausthal, 1982
31. ECSC Contract 7210-AA/133, 'Development and operational test of a new method of
wear control at the blast furnace hearth by direct heatflux measurement' (BFI,
Stahlwerke Bremen GmbH)
32. ECSC Contract 7210-AA/817, The effect of levels and flow of hearth liquid upon
blast furnace operation and wear' (British Steel pic.)
33. Koike, .; Takata, J.; Kariya, J.; Kojima, M.; Sato, K.: 'Extending the campaign life of
Kashima No. 3 blast furnace', Proceedings of the 2nd European Ironmaking
Congress 1991, pp 155/67
34. Kobayashi, K.; Matsumoto, T.; Yanagisawa, K.: Technology for prolonging campaign
life of blastfurnace', Kawasaki Steel Technical Report No. 32, March 1995, pp 9/18
35. Inoue, H.; Niwa, Y.; Takashi, S.; Maki, .; Yamaguchi, .; Sakai, .: 'Technologies
for prolongation of blast furnace campaign life', 1991 Ironmaking Conference
Proceedings, pp 293/9
36. Motomachi, T.; Kanatani, T.; Matsuda, K.; Shoji, S.: 'Techniques to extend campaign
life of blast furnace at Mizushima Works', Taikabutsu Overseas Vol. 12, No. 2, pp
11/17
37. Still, G.: The behaviour of coke in the lower part of the blast furnace', Dr.lng.
Dissertation (Dr.lng. Thesis), RWTH Aachen 1984, INF/TRANS/62/96
38. Okada, T.; Strimomura, K.: 'Measures to extend the campaign lives of blast furnaces
at Kakogawa Works', Taikabutsu Overseas, Vol. 12, No. 2, June 1992, pp 4/10
39. Schulz, V.; Rther, H. P.: 'Investigation and improvement of iron flow in the hearth of
blast furnaces and of related methods for reducing wear of linings', ECSC Contract
7210-AB/112, Final Report (BFI, Thyssen Stahl AG)
40. Shibata, K.; Kimura, Y.; Shimizu, M.; Inaba, S.: 'Influence of taphole depth and
hearth configuration on hot metal flow in blast fumee hearth', Curr. Advanc. Mater.
Proc. 2(1989), No. 1, 92
41. Burteaux, M. et. al.: 'Latest development in cooling and refractories of European blast
furnaces', Proceedings 1st European Ironmaking Congress, Aachen 1986, Vol.3,
Paper V 3
42. Aoyama, K.; Ohtuki, Y.; Asano, .: 'Development of AI
2
0
3
-C-SiC brick for blast
furnace taphole', Taikabutsu Overseas, Vol. 12, No. 2, June 1992, pp 30/33
115
43. Lu, W.K.; He, O.S.: 'Reactions in the shaft of a blast furnace', Proceedings of the
First International Congress of Science and Technology of Ironmaking, 1994, Sendai,
pp 175/85
116
TABLE 5.1
PHYSI CAL AND MECHANI CAL PROPERTIES OF
TECHNI CAL COPPER AND NODULAR CAST I RON
Pr ope r t y
He a t c on duc t i vi t y, w/me t r e /'K*
De n s i t y, kg/c u dm
He a t c a pa c i t y, kJ/kg/K
Yi e l d s t r e n gt h , kg/s q mm
Me l t i n g poi n t , C
Te c h n i c a l c oppe r
360
8.9
0. 383
25-40
1083
Nodul a r c a s t i r on
30-35
7.2
0.544
35
1180-1380
a t 400C.
TABLE 5.2
CARBON- CONTAI NI NG HEARTH BRI CK LI NI NG MATERI AL
Pr oduc t -
Cla s s ific a tion
Amorph ous
c a rbon
bric ks
Th e r ma lly
c ompre s s e d
c a rbon bric ks
Gra ph ite
Pa r t -or
Se mi -gr a ph i t e
Th e r ma lly c ompre s s e d
pa r tgr a ph ite
Se mi -gr a ph i t i c
c a rbon
Burn in g
te mpe r a tur e
800-1 400
1000
2400-3000
800-1 400
1000
1 600-2200
Coarse gr a in
Anthracite plus
graphite if
thought advisable
An th ra c ite
Gra ph ite
Gra ph ite
Gra ph ite
Gra ph ite
Bon din g Ma trix
Carbon plus
additives if
th ough t advisable
Ca rbon
Gra ph ite
Carbon plus
additives if
th ough t advisable
Ca rbon
Se mi -gr a ph i t i c
c a rbon

117
TABLE 5. 3
PROPERTIES OF BRI CKS OF CORUNDUM AND WITH
SI ALON BONDI NG
Properties
Analysis: Al
2
0
3
Si 0
2
SiC
N
2
Sialon '
Apparent density
Porosity
Average pore size
Modules of rupture
at 1500C
Thermal conductivity
at 1000C Wm
Corrosion resistance to
Alkalis
Hot metal
Slag
/o
/o
%
%
%
%

M pa
-1 K-1
Corundum
Mullite
bonded
86
11



3.05
14
20
6
3.2
Average
Excellent
Poor
Silicon
carbide
Siaion
bonded
13

73
7
25
2.70
14
1
50
16
Excellent
Good
Excellent
Corundum
Sialon
bonded
82


7
25
3.15
15
1
20
3.5
Good
Excellent
Good
118
3r d ge n e r a t i on
4t h ge n e r a t i on
Chamotte
Graphite / =
SI C
Ca s t a bl e
530-460
OO- U- J- L
400
COMPARI SON BETWEEN THI RD AND FOURTH
GENERATI ON STAVES
FIG. 5.1
GENERATI ON 4 PLUS STAVE FIG. 5.2
4t h s t a ge
[ M od ification]
1 N o b rick lay ing
on site
B rick work is
held b y stav es
FI G. 5.3
119
(mm)
3 4 5 6 7 8
Years after blowin
e
o
ESTIMATED DURABI LI TY OF FOURTH- GENERATI ON STAVES FIG. 5.4
100
80-
e
o
60
o
O

40
20

A
Istjeneraiion u v e : BF blown om
2ndgtner2ix>n oaves: BF blown out
2ndfeoenuion s um: BF in operation
3rjayrrnion sura: BF m operation
4th{eDention sura: BF m operation
j _ A
K5ICCO
1968 1970 1972 1974 1976 1978 1980 1982
Blowin year
OOO
1984 1986
NUMBER OF STAVE FRONT PIPES BROKEN DURI NG
ONE CAMPAI GN
FIG. 5.5
120
-535 -j
1 45-^
DESIGN DRAWING OF SECOND TYPE OF EXPERIMENTAL
COPPER STAVE COOLERS FOR THYSSEN/RUHRORT NO. 6
BLAST FURNACE
FIG. 5.6
Re fr a c tor y lin in g c on s truc tion s of BF h e a r th
Bla c k
Sma ll-for ma t
Bla c k/wh it e
La r ge -for ma t
Bottom
bla c k/wh it e
Ce r a mic
s a uc e r
High t h e r ma l
c on duc tivity
Low t h e r ma l
c on duc tivity
Dope d bric ks
Ce r a mic
c up
- La r ge /s ma ll-for ma t bric ks
- C-bric ks c e r a mic dope d
- Ce r a mic bric ks C-dope d
REFRACTORY DESIGNS OF BLAST FURNCE HEARTHS FIG. 5.7
121
Tuye r e
a r e a
Ta ph ole
a r e a -
Ce r a mic Cup
V///////////////////////////////7777Z?/,
AetoitiM tiiiAM Jti^^
mic r opor e
a mor ph ous
C-b ric ks
Ra mmin g
ma s s
Gr a ph i t e
LI NI NG OF THE HEARTH OF BLAST FURNACE NO. 2
OF THE STAHLWERKE BREMEN (1992)
FI G. 5.8
Tuye r e
a r e a
Ta ph ole
a r e a
^ ^ ^
mikropore
a morph ous
C-b rick
Ra mmin g
ma s s
Gr a ph it e
LI NI NG OF THE HEARTH OF BLAST FURNACE A OF THE
HTTENWERKE KRUPP- MANNESMANN (1989)
FIG. 5.9
122
Tuye r e a r e a
Un de r h e a r t h
c oolin g
Sia lon boun de d
c or un dum bric ks
Ra mmin g
join ts
Ch a mot t e
Mikropore
a morph ous
C-br ic ks
Gr a ph it
ES a mor ph ous C-br i c ks S2222Z
Ra mmin g
join ts
LI NI NG OF THE HEARTH OF BLAST FURNACE
SCHWELGERN NO. 2 OF THE THYSSEN STAHL AG
FIG. 5.10
Carb on M ock
Course
! - a
U ot Pressed
Semi-Grapliilc
I M-Pr c a i i c 4
Ca r bon
HEARTH CONSTRUCTI ON
AFTER REBUI LD
TAPHOLE CONSTRUCTI ON
FIG. 5.11
123
Temperatur at taphole level: ,1500'C Taphole level
4500
3000
1500
E
E
0
E?
' OJ
a) Carbon.
Hot metal
5ocro
800C
Chromcor Mullite 1 1 00
e
C- =- Z/

t
Chromcor
-Carbon-
iGraphite
Temperatur at taphole level: ^
4500
3000
1500
b)
Chromcor-1 /
Hot metal
Chromcor Mullite
l J
500"C-
800'C
1100
e
C-
3
Carbon
Carbon
J_
iGraphitei
1500 3000 4500 6000 7500
Hearth radius, mm
CALCULATED ISOTHERMS FOR A CONVENTI ONAL HEARTH FI G. 5.12
CONSTRUCTI ON (A) AND FOR A CERAMI C CUP (B)
124
BS REDCAR - 2n d CAMPAIGN
1600 Gr a de
Ca s t a bl e
60% Al umi n a
l a c e d wi t h
Silic on Ca r bi de
45% Alumina
Silic on Ca r bi de
l a c e d with
1600 Gr a de low
Ir on Ca s t a bl e
Silic on Ca r bi de
Mi c r opor e
Ca r bon
St a n da r d
Ca r bon
POSCO KWANG YANG 8F1
1600 Gr a de
Ca s t a bl e
S/Dut y Fir e br ic k MuHite
40% Al umi n a
ba c ke d with
95% Al umi n a
60% Al umi n a
Hi gh K-Va h i e
Ca r bon
Se mi gr a ph it e
POSCO KWANG YANG BF3
1600 Gr a de
Ca s t a bl e
40% Alumin a
Supe r mi c r opor
Ca r bon
Hi gh K-Va k*
Ca r bon
Se mi gr a ph it e
40% Alumin a
65% Al umn a
LI NI NG PROFILES OF REDCAR, KWANG YANG NO. 1 AND
KWANG YANG NO. 3
FI G. 5.13
125
1980 1991
45 V. Al, O


SB -GRAPHI TE
CARBON
GRAPHI TE
rCARBON
GRAPHITE*
SILICONCARBIDE
ii Qj
SEMIGRAPHITE
1980 1991
REFRACTORY LINING AND COOLING SYSTEM
OF BF7, 1980 AND 1991
FIG. 5.14
Kinds of bricks

inside: Corundum -7500
Fireclay 43
Andalusite 63
hot side: Andalusite 63
shell side: Mullite chrome (type 5)
hot side: Andalusite 63
shell side: Nitride bonded corundum
hot side: Andalusite 63
shell side: Corundum chrome (type 3)
Corundum c o n c re te _w i th _Cr
2
0
3
(type 2) K ' ^ J J O T C H ^ T ^ " "
o

co
o
tuyere level
Graphi t p
RELINING OF BLAST FURNACE 9, HAMBORN IN 1987 FIG. 5.15
126
tap hole
ceramic
heat flux probes:
5 to 12
1 to U
carbon blocks
chrome carborundum blocks

* y ^ | graphite blocks
chrome corundum concrete
tamping clay
LOCATI ON OF HEAT FLUX PROBES I N
STAHLWERKE BREMEN NO. 2 BLAST FURNACE
FIG. 5.16
127
3.0
2.5 -\
h sump
>
h crit.
2.0-

i e .
'
"D
Q.
I 1.0
co
0.5 i
0
h sump <
n
crit.
calculated critical sump depth
geometrical measured critical sump depth
Sump depth of blast furnaces without
mushroom formation
10
11 12 13
Hearth diameter, m
14 15
COMPARI SON OF CALCULATED AND MEASURED
CRI TI CAL LI QUI D SUMP DEPTH
FI G. 5. 17
*** / / / / / ////;
M/R - 0 . 2 9
CSEA: Health depth H/R=0.29
CASE : Hearth depth H/R=0/.46
MOLTEN METAL VECTOR AND CONSTANT
TEMPERATURE LI NE WITH INCREASED
HEARTH DEPTH (H/R)
FIG. 5.18
128
ro
io
3000
2500-
E
2000

co

_ 1500
o
=5 1000
500
0 1 3 4 5 6 7 8 9
Steel jacket angle , degrees
10 11 12
HEARTH WALL THICKNESS vs STEEL JACKET ANGLE FI G. 5. 19
O 1 2 3 4 5 (m)
' '
/ R=0 . 0
* / R= 0 . 3 3
* / R=0 . 4 5
Posi t i on and depth of a t aphol e. f or cal cul at i on
Jt/R=0.0 /R= 0. 33 /R= 0. 45
c
o


>

V)

<

<
1 4 8 0 1 5 2 0
fC)
1480 1520 1480 1520
CHANGE OF CI RCUMFERENTI AL DISTRIBUTION OF
REFRACTORI ES TEMPERATURE AT THE
HORI Z ONTAL CROSS SECTION UNER A TAPHOLE
FI G. 5.20
130
Resistance to melting in
hot metal
Resistance to slag
Resistance to spalling
Resistance to alkali
Resistance to oxidation
Resistance to opening
by oxygen
Comprehensive
evaluation
Newly
developed
material
AI2O3-C-SC


O
O
O

O
Material
conventionally
used
Sillimanite




O


Ordinary
carbon
Carbon block

@



O
0
superior, O somewhat superior, somewhat inferior, inferior
COMPREHENSIVE EVALUATI ON OF TAPHOLE BRI CKS FI G. 5.21a
Boring position
?
Tap-hole
Cooler
TAPHOLE STRUCTURE OF KAMAI SHI NO. 1 BF FI G. 5.21b
131
6. CONCLUSIONS
In recent years, major improvements have been achieved in campaign life extension. Campaign
lives achieved in the late 1970's and early 1980's were generally less than 7 years, whereas today
lives of more than double this figure are being achieved. There has been a corresponding
doubling in campaign output, from less than 5000 t/m
3
inner volume to over 11,000 t/m
3
. These
improvements have been achieved by a combination of new technology and materials, more
consistent operation and the development of repair techniques.
The study has highlighted that, not only are there a vast number of techniques in use to extend
blast furnace campaign life but that opinion sometimes varies as to their relative effectiveness and
the best course of action to be taken for a given situation. It is, however, important that an
operator is aware of the experience gained by others, to allow an informed choice of action for
particular circumstances. Despite the varying opinions on specific campaign life extension
techniques, there is some agreement on the basic requirements to achieve a long campaign life.
For a long campaign life, a blast furnace should be operated at a productivity that enables stable,
smooth operation. Good, consistent quality burden materials and coke are paramount in
achieving this, together with accurate control of burden distribution; the latter controlling central
and peripheral gas flow to encourage furnace stability and protect the furnace walls. It is also
essential to have a means of quantifying a given distribution, usually a radial probe, to ensure that
the desired distribution is maintained.
Comprehensive instrumentation and routine monitoring techniques are essential to support stable
operation and to enable early detection of likely problem areas, so remedial actions can be taken
to minimise the effect on furnace life. Maximum use should be made of computers, to process and
analyse primary data, giving the operations staff both rapid information and advice on potential
problem areas. Computerisation of procedures and action plans, in the form of operator guidance
or 'expert' systems, can make a significant contribution to ensuring a rapid, consistent and correct
response to such problems.
The hearth, being the most difficult region to repair during a campaign, is critical for long campaign
life. Successful hearth wall repairs have been completed, but their longevity is uncertain and they
require a long outage. It is therefore very important that the hearth is built to a good design, with
the best refractories affordable. The hearth must be comprehensively equipped with
thermocouples or heat flux probes, efficiently cooled and regularly monitored. A regime which
retains a stable, frozen layer on the hearth refractories is of great benefit in prolonging campaign
life.
The number of stop-start operations must be minimised, by practices which avoid unplanned
off-blasts, efficient maintenance and an integrated steelworks strategy which avoids an imbalance
in iron and steel capacity. Many unplanned stoppages, particularly in the later stages of a
campaign, relate to water leakage from cooling members, which should be minimised by cooling
member design and efficient leak detection systems.
The potential for a long campaign life will be affected by the level of investment when the furnace
is rebuilt. The level of capital expenditure for maximum campaign life may be difficult to justify at
the time of planning a rebuild, as economic factors change over the several years involved. Given
adequate instrumentation and operational control, additional expenditure should be concentrated
in reducing potential problem areas and in regions in which it is difficult to effect a mid-campaign
repair. Awareness of likely mid-campaign repair requirements at the design stage of a rebuild,
133
could allow minor additions at minimal cost, that would simplify and reduce the cost of future
repairs.
Mid-campaign repairs to the furnace body play a major role in maximising campaign life,
especially on existing furnaces which may not incorporate the most modern design features.
These repairs include spraying of shaft refractories (with a variety of methods available to prolong
refractory retention), the repair/replacement/enhancement of cooling members, refractory
grouting, repairs/replacement of the furnace shell and repairs to throat armour. Mid-campaign
repairs to ancillary parts of plant are becoming more important, but by good planning may often be
executed with minimal disruption to production. Design improvements or upgrading may be
incorporated with such repairs to reduce future problem areas.
With more advanced repair techniques available, and the large capital and production costs
involved in a full furnace rebuild, more operators are adopting the philosophy of mini rebuilds,
particularly when there is no spare ironmaking capacity to cover for furnaces under repair. These
mini-rebuilds may involve periods of up to 30 days, where large sections of the furnace are to be
replaced, but some operators are suggesting the possibility of a full rebuild not being necessary.
For further improvements in blast furnace campaign life, continual development of materials and
techniques is essential, particularly in critical areas. The effect of a stack spray is often limited to
a few months and an extension of this by improvements in anchoring, materials or techniques
would be of great benefit. The reduction in cooling member failure and subsequent water leakage
is an important factor in extending furnace life; by improved design, adoption of copper staves, the
use of surface coatings, and improved leak detection. As furnace life increases, it is important to
ensure the reliability of ancillary parts of plant and to design plants which are readily upgradeable
to meet the ever tightening environmental constraints.
134
o
CORDIS
The Community Research and Development Information Service
Your European R&D Information Source
CORDIS represents a central source of information crucial for any organisation - be it industry, small and
medium-sized enterprises, research organisations or universities - wishing to participate in the
exploitation of research results, participate in EU funded science and technology programmes and/or seek
partnerships.
CORDIS makes information available to the public through a collection of databases. The databases cover
research programmes and projects from their preparatory stages through to their execution and final
publication of results. A daily news service provides up-to-date information on EU research activities
including calls for proposals, events, publications and tenders as well as progress and results of research
and development programmes. A partner search facility allows users to register their own details on the
database as well as search for potential partners. Other databases cover Commission documents, contact
information and relevant publications as well as acronyms and abbreviations.
By becoming a user of CORDIS you have the possibility to:
Identify opportunities to manufacture and market new products
Identify partnerships for research and development
Identify major players in research projects
Review research completed and in progress in areas of your interest
The databases - nine in total - are accessible on-line free of charge. As a user-friendly aid for on-line
searching, Watch-CORDIS, a Windows-based interface, is available on request. The databases are also
available on a CD-ROM. The current databases are:
News (English, German and French version) - Results -
Partners - Projects - Programmes - Publications -
Acronyms - Comdocuments - Contacts
CORDIS on World Wide Web
The CORDIS service was extended in September 1994 to include the CORDIS World Wide Web (WWW)
server on Internet. This service provides information on CORDIS and the CORDIS databases, various
software products, which can be downloaded (including the above mentioned Watch-CORDIS) and the
possibility of downloading full text documents including the work programmes and information packages
for all the research programmes in the Fourth Framework and calls for proposals.
The CORDIS WWW service can be accessed on the Internet using browser software (e.g. Netscape) and
the address is: http://www.cordis.lu/
The CORDIS News database can be accessed through the WWW.
Contact details f or further Information
If you would like further information on the CORDIS services, publications and products, please contact
the CORDIS Help Desk :
CORDIS Customer Service Telephone: +352-401162-240
B.P. 2373 Fax: +352-401162-248
L-1023 Luxembourg E-mail: helpdesk@cordis.lu
WWW: http://www.cordis.lu/
European Commission
EUR 17247 Reduction of iron ores
Technical study into the means
of prolonging blast furnace campaign life
D. Jameson, H. Lungen, D. Lao
Luxembourg: Office for Official Publications of the European Communities
1997 134 pp. 21.0 29.7 cm
Technical steel research series
ISBN 92-827-9912-3
Price (excluding VAT) in Luxembourg: ECU 23
The objective of this study was to carry out a comprehensive literature
review of techniques used to extend the campaign life of blast furnaces
and identify important aspects of operational practices, remedial actions
and future plant design, to establish cost-effective methods of extending
European blast furnace campaign lives.
The study has highlighted that, not only are there a vast number of
techniques in use to extend blast furnace campaign life, but that opinion
sometimes varies as to their relative effectiveness and the best course of
action to be taken for a given situation. It is, however, important that an
operator is aware of the experience gained by others, to allow an informed
choice of action for particular circumstances. Despite the varying opinions
on specific campaign life extension techniques, there is some agreement
on the basic requirements to achieve a long campaign life.
For further improvements in blast furnace campaign life, continual
development of materials and techniques is essential, particularly in critical
areas. The effect of a stack spray is often limited to a few months and an
extension of this by improvements in anchoring, materials or techniques
would be of great benefit. The reduction in cooling member failure and
subsequent water leakage is an important factor in extending furnace life,
by improved design, adoption of copper staves, the use of surface
coatings, and improved leak detection. As furnace life increases, it is
important to ensure the reliability of ancillary parts of plant and to design
plants which are readily upgradeable to meet the ever tightening
environmental constraints.
Venta Salg Verkauf Sales Vente Vendita Verkoop Venda Myynti Frsljning
BELGIQUE/BELGI
Moniteur beige/Belgisch Staatsblad
Rue de Louvain 40-42/
Leuvenseweg 40-42
B-1000 Bruxelles. Brussel
Tl. (32-2) 552 22 11
Fax (32-2) 511 01 84
Jean De Lannoy
Avenue du Roi 202/
Koningslaan 202
B-1060 Bruxelles Brussel
Tl. (32-2) 538 51 69
Fax 32-2) 538 08 41
E-mail: jean.de.lannoy@infoboard.be
Librairie europenne Europese Boekhandel
Rue de la Loi 244/
Wetstraat 244
B-1040 Bruxelles/Brussel
Tl. (32-2) 295 26 39
Fax (32-2) 735 08 60
DANMARK
J. H. Schultz Information A/S
Herstedvang 10-12
DK-2620 Albertslund
Tlf. (45) 43 63 23 00
Fax (45) 43 63 19 69
E-mail: schultz@schulu.dk
URL: www.schultz.dk
DEUTSCHLAND
Bundesanzeiger Verlag
Breite Strae 78-80
Postfach 10 05 34
D-50667 Kln
Tel. (49-221)20 29-0
Fax (49-221 i 20 29 278
GREECE/
G.C. Eleftheroudakis SA
International Bookstore
Panepistimiou 17
GR-105 64 Athens
Tel. (30-1)331 41 80/1/2/3
Fax (30-1) 323 98 21
E-mail: elebooks@netor.gr
ESPAA
Mundi Prensa Libros, SA
Castell, 37
E-28001 Madrid
Tel. (34-1 ) 431 33 99/431 32 22
Fax (34-1) 575 39 98
E-mail: mundiprensa@tsai.es
URL: www.tsai.es/mprensa
Boletn Oficial del Estado
Trafalgar. 27-29
E-28071 Madrid
Tel. (34-1) 538 22 95 (Libros)/
384 1715 Suscripciones)
Fax (34-1) 538 23 49 (Libros)/
384 17 14 (Suscripciones)
URL: www.boe.es
Mundi Prensa Barcelona
Conseil de Cent, 391
E-08009 Barcelona
Tel. (34-3) 488 34 92
Fax (34-3) 487 76 59
Journal officiel
Service des publications des CE
26, rue Desaix
F-75727 Paris Cedex 15
Tl. (33-1)40 58 77 01/31
Fax (33-1) 40 58 77 00
IRELAND
Government Supplies Agency
Publications Section
4-5 Harcourt Road
Dublin 2
Tel. (353-1)66131 11
Fax (353-1) 475 27 60
ITALIA
Licosa SpA
Via Duca di Calabria. 1/1
Casella postale 552
1-50125 Firenze
Tel. (3955)64 54 15
Fax (39-55) 64 12 57
E-mail: licosa@ftbcc.it
URL: id382.cilea.itA/irtual_Library/bibliofA'etnna/
licosa/ttl.htm
GRAND-DUCH DE LUXEMBOURG
Messageries du livre Sari
5, rue Raiffeisen
L-2411 Luxembourg
Tl. (352)40 10 20
Fax (352) 490 661
E-mail: mdl@pt.lu
Abonnements:
Messageries Paul Kraus
11, rue Christophe Plantin
L-2339 Luxembourg
Tl. (352) 499 88 88
Fax (352) 499 888 444
E-mail: mpk@pt.lu
URL: www.mpk.lu
NEDERLAND
SDU Servicecentrum Uitgevers
Christoffel Plantijnsfraat 2
Postbus 20014
2500 EA 's-Gravenhage
Tel. (31-70)378 98 80
Fax (31-70) 378 97 83
E-mail: sdu@sdu.nl
URL: www.sdu.nl.
OSTERREICH
Manz'sche Verlags und Universitts
buchhandlung Gmbh
Siebenbrunnengasse 21
Postfach 1
A-1050 Wien
Tel. (43-1)53 161334/340
Fax (43-1) 53 161 339
E-mail: auslieferung@manz.co.at
URL: www.austria.EU. net:81/manz
PORTUGAL
Imprensa NacionalCasa da Moeda, EP
Rua Marques de S da Bandeira, 16 A
P-1050 Lisboa Codex
Tel. (351-1)353 03 99
Fax (351-1) 353 02 94/384 01 32
Distribuidora de Livros Bertrand Ld.'
Rua das Terras dos Vales, 4 A
Apartado 60037
P-2701 Amadora Codex
Tel. (351 -1 ) 495 90 50/495 87 87
Fax (351-1 496 02 55
SUOMI/FINLAND
Akateeminen Kirjakauppa /
Akademiska Bokhandeln
Pohjoisesplanadi 39/
Norra esplanaden 39
PL/PB 128
FIN-00101 Helsinki/Helsingfors
Tel. (358-9) 121 41
Fax (358-9) 121 44 35
E-mail: akatilaus@stockmann. mailnet fi
URL: booknet.cultnet.fi/aka/index.htm'
SVERIGE
BTJAB
Traktorvgen 11
PO Box 200
S-22100Lund
Tel. (46-46)18 00 00
Fax (46-46) 18 01 25
E-mail: bti_tc@mail.btj.se
URL: www.btj.se/media/eu
UNITED KINGDOM
The Stationery Office Ltd
(Agency Section)
51, Nine Elms Lane
London SW8 5DR
Tel. (44-171)873 9090
Fax (44-171) 873 8463
URL: www.the-stalionery-office.co.uk
ICELAND
Bokabud Larusar Blondel
Sklavrdustig, 2
IS 101 Reykjavik
Tel.(354)5515 650
Fax (354) 55 25 560
NORGE
NIC Info AS
stenjoveien 18
Boks 6512 Etterstad
N-0606 Oslo
Tel. (47-22) 97 45 00
Fax (47-22) 97 45 45
SCHWEIZ/SUISSE/SVIZZERA
OSEC
Stampfenbachstrae 85
CH-8035 Zrich
Tel. (41-1)365 53 15
Fax (41 -1)365 54 11
E-mail: urs.leimbacher@ecs.osec.inet.ch
URL: www.osec.ch
ISRAEL
CESK REPUBLIKA
NIS CR - prodejna
Konviktsk 5
CZ-113 57 Prahat
Tel. (42-2) 24 22 94 33
Fax (42-2) 24 22 94 33
E-mail: nkposp@dec.nis.cz
URL: www.nis.cz
CYPRUS
Cyprus Chamber Of Commerce & Industry
38, Grivas Digenis Ave
Mail orders:
PO Box 1455
CY-1509 Nicosia
Tel. (357-2) 44 95 00/46 23 12
Fax (357-2) 361 044
E-mail: cy1691_eic_cyprus@vans.infonet.com
MAGYARORSZAG
Euro Info Service
Europa Haz
Margitsziget
PO Box 475
H-1396 Budapest 62
Tel. (36-1)11 16 061/11 16 216
Fax (36-1 (302 50 35
E-mail: euroinfo@mail. matav.hu
URL: www.euroinfo.hu/index.htm
MALTA
Miller Distributors Ltd
Malta International Airport
PO Box 25
LOA 05 Malta
Tel. (356) 66 44 88
Fax (356) 67 67 99
POLSKA
Ars Polona
Krakowskie Przedmiescie 7
Skr. pocztowa 1001
PL-00-950 Warszawa
Tel. (48-2)26 12 01
Fax 48-2) 26 62 40
TURKIYE
Dnya Infotel A.S.
Istikll Cad. No: 469
TR-80050 Tnel-Istanbul
Tel. (90-212)251 91 96
(90-312)427 02 10
Fax (90-212) 251 91 97
BLGARUA
EuropressEuromedia Ltd
59. Bid Vitosha
BG-1000 Sofia
Tel. (359-2) 80 46 41
Fax (359-2) 80 45 41
HRVATSKA
Mediatrade Ltd
Pavia Hatza 1
HR-10000 Zagreb
Tel. (3B5-1)43 03 92
Fax (385-1) 44 40 59
ROMANIA
Euromedia
Str. G-ral Berthelot Nr 41
RO-70749 Bucuresti
Tel. (40-1) 210 44 01/614 06 64
Fax (40-1 ) 210 44 01 /312 96 46
SLOVAKIA
Slovenske Technicka Kniznlca
Nmestie slobody 19
SLO-81223 Bratislava 1
Tel. (42-7)53 18 364
Fax (42-7) 53 18 364
E-mail: europ@tbb1 .sltk.stuba.sk
SLOVENIA
Gospodarskl Vestnlk
Zalozniska skupina d.d.
Dunajska cesta 5
SI-1000 Ljubljana .
Tel. (386)61 133 03 54
Fax (386) 61 133 91 28
E-mail: belicd@gvestnik.sl
URL: www.gvestnik.si
R.O.Y. International
17, Shimon Hatarssi Street
PO Box 13056
61130 Tel Aviv
Tel. (972-3)546 14 23
Fax (972-3) 546 14 42
E-mail: royil@netvision.net.il
Sub-agent for the Palestinian Authority:
Index Information Services
PO Box 19502
Jerusalem
Tel. (972-2)2716 34
Fax 972-2) 27 12 19
RUSSIA
CCEC
60-letiya Oktyabrya Av.
117312 Moscow
Tel.(095)135 52 27
Fax (095) 135 52 27
AUSTRALIA
Hunter Publications
PO Box 404
3167 Abbotsford, Victoria,
Tel. (61-3)9417 53 61
Fax (61-3) 9419 71 54
CANADA
Uniquement abonnements/
Subscriptions only:
Renouf Publishing Co. Ltd
1294 Algoma Road
K1B 3W8 Ottawa, Ontario
Tel. (1-613)741 73 33
Fax (1-613) 741 54 39
E-mail: renouf@fox.nstn.ca
URL: fox.NSTN.Ca/-renouf
EGYPT
The Middle East Observer
41, Sheril Street
Cairo
Tel. (20-2) 39 39 732
Fax (20-2) 39 39 732
JAPAN
PSIJapan
Asahi Sanbancho Plaza #206
7-1 Sanbancho, Chiyoda-ku
Tokyo 102
Tel. (81-3)3234 69 21
Fax (81-3) 3234 69 15
E-mail: psijapan@gol.com
URL: www.psi-japan.com
SOUTH AFRICA
Sarto
5th Floor Export House,
CNR Maude & West Streets
PO Box 782 706
2146Sandton
Tel. (27-11)883 37 37
Fax (27-11)883 65 69
UNITED STATES OF AMERICA
Beman Associates
4611 -F Assembly Drive
MD20706 Lanham
Tel. (301 459 2255 (toll free telephone)
Fax 800) 865 3450 toll free fax)
E-mail: query@bernan.com
URL: www.beman.com
MEXICO
MundiPrensa Mexico, SA de CV
Ro Panuco, 141
Delegacin Cuauhtemoc
ME-06500 Mxico DF
Tel. (52-5) 553 56 58/60
Fax (52-5) 514 67 99
E-mail: 104164.23compuserve.com
REPUBLIQUE DE CORE
Kyowa Book Company
1 F1. Phyung Hwa Bldg
411-2 Hap Jeong Dong, Mapo Ku
121-220 Seoul
Tl. (82-2) 322 6780/1
Fax 82-2) 322 6782
E-mail: kyowa2@ktnet.co.kr.
ANDERE LANDER/OTHER COUNTRIES/
AUTRES PAYS
Bitte wenden Sie sich an ein Bro Ihrer
Wahl / Please contact the sales office of
your choice / Veuillez vous adresser au
bureau de vente de votre choix
12796 #
NOTICE TO THE READER
All scientific and technical reports published by the European Commission are announced in the
periodical 'e ur o a bs t r a c t s ', published every two months, and in the R&TD publications database of
CORDIS, the Community Research and Development Information Service.
For subscription (1 year: ECU 65) please write to the address below.
o
o
-t-
->l
o
Price (excluding VAT) in Luxembourg: ECU 23
I SBN ^ - . 3 5 7 - ^ 1 2 - 3
OFFICE FOR OFFICIAL PUBLICATIONS
OF THE EUROPEAN COMMUNITIES
* * *
k EUR -k
op *
* L-2985 Luxembourg
9 789282V99123 >

You might also like