See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319368819

Recovery mechanism of the sticking-type breakout during continuous casting

of steel

Article in Ironmaking & Steelmaking · August 2017

DOI: 10.1080/03019233.2017.1368956

CITATIONS READS

4 630

7 authors, including:

Fei He Zhenghai Zhu

Anhui University of Technology University of Leicester
15 PUBLICATIONS 228 CITATIONS 2 PUBLICATIONS 11 CITATIONS

SEE PROFILE SEE PROFILE

Jiashun Shi
Southeast University (China)
1 PUBLICATION 4 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

National Natural Science Foundation of China (Grant No. 51504002). View project

MoldExpert View project

All content following this page was uploaded by Fei He on 31 August 2018.

The user has requested enhancement of the downloaded file.

 Ironmaking & Steelmaking
Processes, Products and Applications

ISSN: 0301-9233 (Print) 1743-2812 (Online) Journal homepage: http://www.tandfonline.com/loi/yirs20

Recovery mechanism of the sticking-type breakout

during continuous casting of steel

Haitao Qian, Fei He, Xi’an Xie, Zhenghai Zhu, Lingying Zhang, Linlin Zhang &
Jiashun Shi

To cite this article: Haitao Qian, Fei He, Xi’an Xie, Zhenghai Zhu, Lingying Zhang, Linlin Zhang &
Jiashun Shi (2017): Recovery mechanism of the sticking-type breakout during continuous casting of
steel, Ironmaking & Steelmaking, DOI: 10.1080/03019233.2017.1368956

To link to this article: http://dx.doi.org/10.1080/03019233.2017.1368956

Published online: 29 Aug 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=yirs20

Download by: [Anhui University of Technology] Date: 29 August 2017, At: 16:48
 IRONMAKING & STEELMAKING, 2017
https://doi.org/10.1080/03019233.2017.1368956

Recovery mechanism of the sticking-type breakout during continuous casting of

steel
Haitao Qian, Fei He, Xi’an Xie, Zhenghai Zhu, Lingying Zhang, Linlin Zhang and Jiashun Shi
School of Metallurgical Engineering, Anhui University of Technology, Ma’anshan, Anhui, China

ABSTRACT ARTICLE HISTORY

Sticking-type breakout is one of key factors which restrict the development of high-speed or wide- Received 2 August 2017
heavy plate continuous casting technology. In order to better heal and control the sticking-type Accepted 11 August 2017
breakout, this paper presents a comprehensive study for its recovery mechanism. First, according to
KEYWORDS
force characteristics of solidified shell at mold exit, calculation method of safe shell thickness is Sticking-type breakout;
proposed. Then, considering sticker propagation behaviour and safe shell thickness, a more recovery mechanism; safe
practical recovery model of sticking-type breakout is established and verified by actual examples. shell thickness; casting speed;
The result shows the proposed recovery strategies avoid casting speed to be decreased too low or mould oscillation parameters;
cast stop, thus save unnecessary operating time, and are better than original strategies.
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

continuous casting
Furthermore, influencing factors of recovery behaviour are analyzed. The result shows hot spot
location, operating casting speed, and mold oscillation frequency have a significant effect on
minimum speed for the recovery and oscillation stoke has little influence on it.

Introduction
To solve the sticking-type breakout problem is of great sig-
During the continuous casting process of steel, hot liquid steel nificance for ensuring smooth production and high quality
in the ladle is poured into the tundish through a shroud, and slabs. At present, in addition to improving the casting
then into the water-cooled copper mould from the tundish factors of inducing the sticker, the main method is mould
through a submerged entry nozzle. In the mould called the breakout prediction technology based on measuring temp-
‘heart’ of a caster, the liquid steel is solidified into the shell erature with thermocouples [3–6]. Its basic principle is that a
with enough thickness. And then, under mould oscillation, number of thermocouples are embedded in mould copper
the solidifying shell could be continuously and safely with- plates, and used to measure and monitor mould temperature
drawn from the mould as a strand, and upon exiting the at different positions, and the breakout symptom is identified
mould. But sometimes, due to large mould level fluctuations, as soon as possible by logical algorithm or intelligent algor-
rapid speed changes, too high casting speed, mismatching of ithm, and then reducing the casting speed could increase
steel grades and mould fluxes etc., the poor lubrication results negative strip time, eliminate the sticking between the shell
in the sticking between the shell and mould. Especially in and mould, and re-solidify the tear of the sticker and form a
high-speed continuous casting and wide and heavy plate con- shell with enough thickness. When the new shell reaches
tinuous casting process, the sticker of the shell is more likely the exit of the mould, it has sufficient strength and could be
to occur, because the growth uniformity of the initial solidified tolerable for static pressure of molten steel, thus the breakout
shell and the stability of inflow of the flux slag decrease signifi- could be prevented.
cantly [1,2]. After a sticker has occurred, if no action is taken, At present, the research of mould breakout prediction tech-
under mould oscillation and shell withdrawing, the shell is nology focuses on prediction method of the sticker. Although it
torn and re-solidified again and again, and the tear line is important to identify the sticking phenomenon of the shell as
moves downward continuously, depending on the casting soon as possible, but often in the actual production process,
speed. Meanwhile, the ruptured portion also expands large temperature fluctuation, or failure of thermocouple
towards both sides and forms a V-type tear line. When the caused by terrible and complex field condition, may make
tear reaches the exit of the mould, the breakout will happen breakout prediction algorithm send late alarm. The late break-
due to be not tolerable for static pressure of molten steel. out alarm may result in this phenomenon that even if the
This kind of breakout caused by the sticker is called the stick- measure of decreasing casting speed after alarm is taken, the
ing-type breakout, which has very high frequency of occur- breakout accident still happens [7]. This phenomenon has a
rence in actual continuous casting process. The breakout lot to do with improper casting speed control measures after
not only disturbs smooth production, but also damages the the breakout alarm, in addition to late alarm time. For most of
devices of the caster, and results in huge economic losses. the casters, actual control measure after the breakout alarm is
Therefore, the sticking-type breakout has become one of a kind of fixed decreasing casting speed mode or direct stop
the key factors which restrict the development of high- the caster. All of these problems occur due to the lack of
speed continuous casting and wide and heavy plate continu- clarity on the recovery mechanism of the sticking-type breakout.
ous casting technology. And it is an important problem that Therefore, it is necessary to further study the recovery
urgently needs to be solved for the continuous casting. behaviour of the sticking-type breakout, and analyse its

CONTACT Fei He [email protected], Zhenghai Zhu [email protected] School of Metallurgical Engineering, Anhui University of Technology,
Ma’anshan, Anhui 243002, China
© 2017 Institute of Materials, Minerals and Mining
 2 H. QIAN ET AL.

recovery mechanism, and deduce its recovery conditions, and study with Moon for slab continuous casting process, and
then put forward the effective speed control measures for in the derivation of the recovery conditions, the propagation
repairing the sticker’s tear, thus reducing the risk of breakout, behaviour’s law of the sticker was considered quantitatively.
saving unnecessary operating time, improving production The above studies deduced the recovery conditions of the
efficiency, and reducing the loss of equipment. In this sticking-type breakout, such as the minimum casting speed
paper, based on recovery theory of the sticking-type break- and holding time. However, the influence of mould oscil-
out, mathematical model of recovery behaviour of the stick- lation parameters on the recovery process is not taken into
ing-type breakout has been established, and its recovery account. At the same time, it is unreasonable that the safe
conditions are deduced in detail. At the same time, the influ- shell thickness simply replaces with the thickness of the
ence of the casting speed and mould oscillation parameters sound shell, which is related to the force status of the shell
on recovery behaviour of the sticking-type breakout is at the mould exit.
discussed. Therefore, in order to accurately and scientifically under-
stand recovery mechanism of the sticking-type breakout, it is
necessary to further quantitatively study the recovery behav-
Theory and mathematical model of sticking-type
iour. This paper is based on those proposed theories of the
breakout recovery
sticking-type breakout. In a healing cycle for a sticker event,
The research about recovery behaviour of the sticking-type as shown in Figure 1(a,b), by reasonable decreasing casting
breakout has attracted some attention since the 1980s. Tsu- speed and controlling mould oscillation with large amplitude
neoka et al. [8] found that for slab continuous casting and low frequency to increase the time of negative strip
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

process, when a sticker was identified, the sticking-induced period, the three conditions of the recovery are satisfied: (1)
rupture could be repaired by decreasing the casting speed elimination condition: after decreasing speed, the stuck shell
immediately. According to the force status of the stuck is liberated from copper plate during the negative strip
shell, the mechanism of rupture recovery was put forward, period; (2) mechanical condition: the liberated shell is
and the control variables were obtained, namely, the moving downward with the withdrawn shell and is not tore
minimum casting speed and the time of negative strip. again; (3) thickness condition: the thickness of repaired shell
However, only the recovery condition at the shell’s rupture at the mould exit is larger than the safe shell thickness.
is taken into account, the thickness of the newly formed Based on the force status and solidification law of the stuck
shell at the rupture and the shell whether can withstand shell, mathematical model of sticking-type breakout recovery
the static pressure of molten steel at the mould exit are is established according to the recovery conditions, as
not taken into account. shown in Figure 1(c), and then by the model repairing par-
Itoyama et al. [9] explicitly proposed three conditions of ameters of a sticker could be obtained, such as the
recovery of the sticking-type breakout: ① The stuck shell was minimum healing speed and holding time of minimum
liberated from copper plate of mould. ② The liberated shell healing speed, decreasing speed gradient, increasing speed
was moving downward. ③ The healed shell had a sufficient gradient and mould oscillation parameters.
thickness at the exit of mould, which could withstand the
static pressure of molten steel. But the conditions ① and ②
are only analysed qualitatively, Itoyama’s emphasis is on the Recovery conditions of sticking-type breakout
condition ③, and according to it, the minimum casting speed
According to Figure 1, three recovery conditions of sticking-
to prevent breakout could be deduced. However, in the deri-
type breakout need to be analysed. The elimination condition
vation process, the calculation method of the thickness of
is easily to be met by decreasing casting speed in conjunction
the shell at the mould exit which can withstand the static
with mould oscillation with high amplitude and low fre-
pressure of molten steel (namely safe shell thickness) is not
quency, so the emphasis is put on the derivation of mechan-
given. Moreover, it is unreasonable to put the casting speed
ical condition and thickness condition.
as the descending speed of sticker, because the actual des-
cending speed of the sticker is less than the casting speed.
Blazek et al. [10] observed the phenomenon of sticker recovery Mechanical condition of sticking-type breakout
by experiments and found that the sticking-induced rupture recovery
not only could be repaired by means of decreasing the
casting speed, but sometimes the rise of liquid surface can As shown in Figure 1(c), the forces acting on the stuck shell
also make it liberate from the mould copper plate to achieve include friction force, inertial force to meet the mechanical
recovery. However, there is no quantitative analysis of the condition, the resultant force acting on the stuck shell could
recovery mechanism of the sticking-type breakout. not exceed its shear strength. The related forces could be cal-
According to the method of Tsuneoka, Moon et al. [11] culated as follows:
analysed the mechanism of tearing and repairing of shell Friction force:
for thin slab continuous casting process, and obtained the
aVC t
bVC t

1
recovery conditions of the sticking-induced rupture. And to Ff = mrl gydydx = mrl ga2 bVC3 t3 (1)
0 0 3
further make up for the lack of Tsuneoka, by combining
recovery conditions of the rupture and thickness require- where, µ is the coefficient of friction between the mould and
ment of the newly formed shell at the rupture, the recovery shell (=0.3); ρl is the density of the molten steel
parameters of the sticking-type breakout were deduced (=7200 kg m−3); α is the ratio of the propagation speed
quantitatively, such as the minimum healing speed and its of sticker in vertical (y) direction to the casting speed
holding time, decreasing speed gradient and increasing (α = 0.7613); β is the ratio of the propagation speed of
speed gradient. He et al. [12] also carried out a similar sticker in horizontal (x) direction to the casting speed (β =
 IRONMAKING & STEELMAKING 3
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

Figure 1. Recovery model of sticking-type breakout recovery. (a) Recovery behaviour of sticking-type breakout; (b) Mould oscillation curve before and after decreas-
ing casting speed; (c) Recovery conditions of sticking-type breakout.

1.267); VC is the casting speed (m min−1); t is the time elapsed Shear strength:
from the meniscus (s); g means gravitational acceleration
l
mtNn
(=9.8 N kg−1). Fs = 2 tN cos udzdl = 2lmtNn tN cos u (6)
Inertial force: 0 0

aVC t bVC t mtn where, θ is the acute angle between the tear line and ver-
rs gdzdydx (2p · f )2 s
Fa = 0 0 0
· tical (y) direction, u= tan−1 (b/a) depends on the propagation
g 2 speed of sticker in vertical and horizontal directions.
2mp2 f 2 srs abVC2 t n+2 To make the newly formed shell be not torn, the function F
= (2)
(n + 1)(n + 2) (VC, t) must meet the following formula:
Where, ρs is the density of the shell (=7800 kg m−3); m, n are F(Vc , t) = Fs − (Ff − Fa ) ≥ 0 (7)
constant (shell thickness d (mm) = mtn; m = 2.58, n = 0.5); f is
oscillation frequency of mould (=150 min−1); s is the oscil- Through the above analysis, F(VC, t) depends on the time t
lation stroke (=12 mm). elapsed from the meniscus and the casting speed VC, the
Length of tear line: minimum casting speed to ensure the repaired shell is not
torn again could be obtained when the time t elapsed from

the meniscus is given.
l= a2 + b2 VC t (3)
Figure 2 provides the minimum speed values that should
be controlled at each position of mould when the operating
Shear stress [11]:
casting speed VCo is 1.2 m min−1. Figure 3 provides the
sN 1 minimum speed values for different frequencies and different
tN = √ = √ × (0.118m2 nt 2n−1 − 0.0367)g (4) strokes when the operating casting speed VCo is 1.2 m min−1
3 3
and the time t is 40 s (namely, the distance from the meniscus
Where, subscript ‘N’’ means the negative strip period; σN is L is 610 mm).
the average yield stress of the shell (MPa). From Figure 2, the function value F(VC, t) becomes smaller as
Negative strip time [13]: the casting speed VC increases. When F(VC, t) = 0, VC is the criti-
  cal value of the required minimum casting speed. For example,
60 1000VC
tN = cos−1 (5) when L = 100 mm, as long as the speed is less than
p·f p · sf 5.26 m min−1 the shell will not be torn here. It could also be
 4 H. QIAN ET AL.

the function value F(Vc, t) increases, which means that the

shell will not be easily torn. Curves’ spacing decreases
with s increases, anti-tear effect is getting worse.

Calculation of safe shell thickness at the mould exit

The thickness of repaired shell at the mould exit should
withstand the static pressure of molten steel to avoid break-
out. In this paper, as shown in Figure 4, by the force analysis
of actual shell at the mould exit, the equivalent simply sup-
ported beam model is established, and the strength check-
ing is carried out, and then the calculation formula of safe
shell thickness could be obtained. The specific steps are as
follows.
Figure 2. Recovery curves when operating casting speed is 1.2 m min−1. (a)
s = 12 mm, t = 40 s. (b) f = 150 min−1, t = 40 s.
(1) Calculating support reaction:

lB −lA
F = 0 ⇔ R A + RB = rgb(lA + x)dx
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

0
1
= rgb(lB2 − lA2 ) (8)
2


A
lB −l

M = 0 ⇔RB (lB − lA ) = rgb(lA + x)xdx (9)

0

1
RA = rgb(lB2 − 2lA2 + lA lB )
6
(10)
1
RB = rgb(2lB2 − lA2 − lA lB )
6
where, lA is the distance from molten steel surface to
mould exit (mm); lB is the distance from molten steel
surface to the first foot roller centre (mm); RA is the
support reaction at point A (N ); RB is the support reaction
at point B (N ); ρ is the density of molten steel (kg m−3); b is
the shell width (mm).
(2) Solving shear force and bending moment:
1
Q(x) + rgb(2lA + x)x = RA (11)
2

x
Q(x) x + rgb(lA + x)xdx = M(x) (12)
0

1 1
Q(x) = rgb(lB2 − 2lA2 + lA lB ) − rgblA x − rgbx 2 (13)
6 2

Figure 3. Recovery curves at different mould oscillation parameters. 1 1 1

M(x) = rgb(lB2 − 2lA2 + lA lB )x − rgblA x 2 − rgbx 3 (14)
6 2 6
From Equations (13) and (14), Equation (13) is the deriva-
seen that as the distance from the meniscus L increases, tive function of Equation (14), and there are two solutions
the required minimum casting speed is getting smaller. to the equation Q(x) = 0, as follows.
When L = 300 mm, the minimum speed is 2.32 m min−1. 
When L = 500 mm, the minimum speed is 1.45 m min−1. When lA2 + lB2 + lA lB
L = 800 mm, the minimum speed is 0.92 m min−1. x1 = − − lA
3
From Figure 3(a), when VCo = 1.2 m min−1 and f =  (15)
lA2 + lB2 + lA lB
150 min−1, the shell will be torn at t = 40 s (namely, L = x2 = − lA
610 mm), because the function value F(Vc, t) < 0. As the fre- 3
quency f decreases, the function value F(Vc, t) increases, (3) Drawing shear force diagram and bending moment
which means that the shell will not be easily torn. Curves’ diagram. From Figure 5, only when x ≥ 0 it has practi-
spacing increases with f decreases, anti-tear effect is cal significance, the dangerous cross section should
getting better. From Figure 3(b), when VCo = 1.2 m min−1 be located at x = 0 (shear force has the maximum
and s = 12 mm, the shell will be torn at t = 40 s, because value) or x = x 2 (bending moment has the maximum
the function value F(Vc, t) < 0. As the stroke s increases, value).
 IRONMAKING & STEELMAKING 5

Figure 4. Schematic diagram and the force analysis of the shell at the mould exit.


Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

6M(x)
d≥ (22)
b[s]

where, σmax is the maximum value of bending moment,

[σ] is the allowable tensile stress of shell.

Because lB − lA .. d, and the shear force is relatively

small, the beam can be taken as a slender beam, and just
the action of bending moment needs to be considered [14].
Therefore, the safe shell thickness at the mould exit could
be calculated as follows.
Figure 5. Shear force diagram and bending moment diagram.

6M(x=x2 )
dn ≥ w
b[s]
When x = 0: 
 
6 1 1 1
1 =w rg(lB − 2lA + lA lB )x2 − rglA x2 − rgx2
2 2 2 3
Q(x=0) = rgb(lB2 − 2lA2 + lA lB ) (16) [ s] 6 2 6
6
(23)
when x = lB − lA :
Where, dn is the safe shell thickness; φ is the influence coeffi-
1 cient of steel grade, superheat degree and slab-withdrawn
Q(x=lB −lA ) = rgb(2lB2 − lA2 − lA lB ) (17)
6 force (φ = 1–1.3).
 Take the steel called SS400 as an example, the yield
when x = x2 = lA2 + lB2 + lA lB /3 − lA , M(x) takes the strength σs of the shell at different temperatures could be
maximum: obtained by high temperature mechanical properties’
1 1 experiment, as shown in Table 1. And here, lA = 800 mm,
M(x) = rgb(lB2 − 2lA2 + lA lB )x2 − rgblA x22 lB = 910 mm, the allowable stress of shell [s] = ss /n, where
6 2
1 the safe coefficient n is 2.5 [15].
− rgbx2 3
(18)
6
(4) Whenx = 0, the beam is subjected to pure shear force Thickness condition of sticking-type breakout
[14]: recovery
3Q(x) Figure 1(a) shows the recovery behaviour of sticking-type
tmax = ≤ [ t] (19)
2db breakout. There are some assumptions before analysing thick-
ness condition of sticking-type breakout recovery, as follows.
3Q(x)
d≥ (20)
2b[t] (1) The thickness of repaired shell at the mould exit is larger
where, τmax is the maximum value of shearing force, [τ] is than the safe shell thickness at the mould exit (namely
permissible shear stress of the shell. dBO ≥ dn).
 (2) The sticker is initiated near the meniscus. The initial
lA2 + lB2 + lA lB location of hot spot (namely sticking point or quasi-
When x = x2 = − lA , the beam is sub-
3 meniscus) is 0 mm from the meniscus.
jected to pure bending moment [14]:
(3) The casting speed is maintained for certain time at the
6M(x) minimum speed, and increased before the repaired
smax = ≤ [s ] (21)
d2 b shell leaves the mould.
 6 H. QIAN ET AL.

Table 1. Yield strength and minimum safe shell thickness for SS400 steel at (4) Increasing speed gradient kinc :
different temperatures. As show in Figure 1(a), assuming that at point B,
Shell temperatures/°C 950 1000 1050 1100 1150 1200 the stuck shell is just liberated from the mould
Yield strength/MPa 29 25 22 20 18 15 copper plate, and the recovery time of the sticker:
Allowable stress/MPa 11.6 10.0 8.8 8.0 7.2 6.0
Safe shell thickness at mould 9.33 10.05 10.71 11.18 11.78 12.91 tr = t2 + t3 .
exit/mm
1
Lm − Ld − DLdec − VBO tr = kinc t32 (34)
2
(4) Ld is The distance from the meniscus to the hot spot 2(Lm − Ld − DLdec − VBO tr )
kinc =
location when alarm signal is on, the average hot spot t32
speed Vst can be estimated from the statistical formula: 2(Lm − Ld − VBO tr )
Vst = 0.7613V c − 0.022. The value of response time for ≤ (35)
(tr − t2 )2
alarm ta could be calculated by the formula: ta = Ld /Vst .
where, t3 is the increasing time of casting
In view of recovery behaviour of sticking-type breakout, speed;△Linc is the distance during increasing the
according to the above assumptions, the thickness condition speed.
of sticking-type breakout recovery is analysed as follows.
Case analysis of sticking-type breakout recovery
(1) The recovery time of the sticker tr and holding time of
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

To further explain and verify the above recovery con-

minimum casting speed t2:
ditions, two different recovery cases of actual stickers
From dBO ≥ dn and the solidification square root law:
are calculated by above equations, and recovery par-
√
KS tr ≥ dn (24) ameters of actual healing cycle for each case are com-
pared with the healing cycle for which uses the
tr1 ≤ tr ≤ tr2 with tr1 = (dn /KS )2 proposed recovery parameters in this study, as shown
Lm − Ld − DLdec Lm − Ld (25) in Table 2 and Figure 6.
tr2 = , ,
VBO VBO From Figure 6(a), when the distance from hot spot to the
meniscus Ld is 0.505 mm, the actual repairing case uses cast
0 ≤ t2 ≤ tr (26) stop measure, the recovery time of the sticker tr is 51 s. But,
Where, dBO is the thickness of repaired shell for pre- if the casting speed decreases to 0.6 m min−1 when decreas-
venting the breakout at the mould exit; Ks is the solidifica- ing speed gradient kdec is −0.1642 m min−1 s−1, the recovery
tion constant. time is only 25 s. And the thickness of repaired shell at the
(2) The minimum healing speed VBO : mould exit is larger than the safe shell thickness. So, in most
From Equation (25): cases of recovery, there is no need for cast stop. From
Figure 6(b), when the distance from hot spot to the meniscus
Lm − Ld − DLdec Lm − Ld Ld is 0.386 mm, in the actual repairing case, the minimum
VBO ≤ , (27)
(dn /KS )2
(dn /KS )2 speed is 0.3 m min−1, the recovery time of the sticker tr is
66 s. But, if the casting speed decrease to 0.8 m min−1 when
From F(Vc, t) = Fs − (Ff − Fa ) ≥ 0:
decreasing speed gradient kdec is −0.0694 m min−1 s−1, the
VBO ≤ VBO
F
(F|L=Ld = 0) (28) recovery time is only 25 s. And also the thickness of repaired
shell at the mould exit is larger than the safe shell thickness.
From Equations (27 and 28): So, in most cases of recovery, casting speed does not need
  to be reduced too low.
Lm − Ld F
VBO = min ,V BO (29) Through the above analysis shows that by using the repair-
(dn /KS )2
ing strategy proposed in this study, not only the success rate
Where, Lm is the distance from the meniscus to mould of repairing the sticking-type breakout could be ensured, but
exit;△Ldec is the distance during decreasing the speed. more recovery time will be saved and equipment loss will be
(3) Decreasing speed gradient kdec : reduced, and so on.
As show in Figure 1(a), for the hot spot, the travelling
length during decreasing the speed △Ldec can be calcu- Table 2. Recovery parameters of sticking-type breakout.
lated as Equation (30). This movement can be seen as Parameters Unit Case 1 Case 2
uniform deceleration movement. Distance from the meniscus to mould m 0.80 0.80
exit Lm
Vst + VBO Operating casting speed VCo m min−1 1.20 1.20
DLdec = t1 (30) Distance from hot spot to the m 0.505 0.386
2
meniscus Ld
VBO = Vc + kdec t1 (31) Average hot spot speed Vst m min−1 0.8914 0.8914
Response time for alarm ta s 34.00 26.00
Range of healing period tr s ≥25.00 ≥25.00
From Equation (27): Range of minimum speed VBO m min−1 0–0.7091 0–0.9945
Holding time taken a fixed value t2 s 12 14
DLdec ≤ Lm − Ld − (dn /KS )2 VBO (32) Minimum speed taken a fixed value m min−1 0.6 0.8
VBO
(Vst + VBO )(VBO − Vc ) Range of decreasing speed gradient m min−1 s−1 ≤−0.1642 ≤−0.0694
kdec ≤ (33) kdec
2 × [Lm − Ld − (dn /KS )2 VBO ] −1 −1
Range of increasing speed gradient m min s ≤0.0322 ≤0.0576
kinc
Where, t1 is the decreasing time of casting speed.
 IRONMAKING & STEELMAKING 7
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

Figure 6. Comparison of healing cycles for actual and proposed recovery parameters. (a) Case 1. (b) Case 2.

Influencing factors of sticking-type breakout speeds, there are different minimum speeds, as shown
recovery in Figure 7. Here, VBO2 is the minimum casting speed
required by the mechanical condition, VBO3 is the
From the above derivation of recovery conditions of stick-
minimum casting speed required by the thickness
ing-type breakout, the location of the hot spot, the
condition.
casting speed and mould oscillation parameters have an
From Figure 7, when operating casting speed is less
important influence on the recovery behaviour of sticking-
than or equal to 0.8 m min−1, minimum speed for the
type breakout.
recovery of sticking-type breakout is influenced by mech-
anical condition and thickness condition. With the increase
of operating casting speed, the effect of mechanical con-
Effect of operating casting speed on sticking-type
dition will weaken and the recovery mainly depends on
breakout recovery
the thickness condition when operating casting speed
In this case which the mould has the same oscillation par- exceeds to 0.8 m min−1. For example, if the operating
ameters (namely f = 150 min−1, s = 12 mm), according to casting speed is 0.6 m min−1, when the distance from hot
the mechanical condition, at different operating casting spot to the meniscus is in the range of 0–175 mm and
 8 H. QIAN ET AL.

by the mechanical condition, and thus only the thickness

condition need to be considered.

Effect of mould oscillation frequency on sticking-type

breakout recovery
In this case which the operating casting speed are 0.6, 0.7, 0.8,
1.2 m min−1, and mould oscillation stroke is 12 mm, according
to the mechanical condition, at different oscillation frequen-
cies, there are different minimum speeds, as shown in
Figure 8. From Figure 8, with the increase of operating
casting speed, the oscillation frequency has an increasing
impact on minimum speed. But when operating casting
speed exceeds to a critical value (such as 1.2 m min−1), the
minimum speed depends on the thickness condition and
Figure 7. Comparison of minimum speed at different operating casting speeds. the effect of the oscillation frequency on the minimum
speed will be ignored.
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

610–800 mm, the minimum speed required by the thick-

ness condition is less than that required by the mechanical Effect of mould oscillation stroke on sticking-type
condition, so the thickness condition plays a leading role breakout recovery
and must be met. On the contrary, when L is in the In this case which the operating casting speed are 0.6, 0.7, 0.9,
range of 175–610 mm, the mechanical condition is the 1.2 m min−1, and mould oscillation frequency is 240 min−1,
leading factor and must be met. If the operating casting according to the mechanical condition, different minimum
speed is 1.2 m min−1, the minimum speed required by speeds of different oscillation strokes are shown in Figure 9.
the thickness condition is always less than that required As illustrated in Figure 9, the effect of mould oscillation

Figure 8. Comparison of minimum speed at different oscillation frequencies. (a) When operating casting speed is 0.6 m min−1. (b) When operating casting speed is
0.7 m min−1. (c) When operating casting speed is 0.8 m min−1. (d) When operating casting speed is 1.2 m min−1.
 IRONMAKING & STEELMAKING 9
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

Figure 9. Comparison of minimum speed at different oscillation strokes. (a) When operating casting speed is 0.6 m min−1. (b) When operating casting speed is
0.7 m min−1. (c) When operating casting speed is 0.9 m min−1. (d) When operating casting speed is 1.2 m min−1.

stroke on minimum speed for sticking-type breakout recovery shell thickness of different steel grades at different temp-
is small, compared with operating casting speed and mould eratures could be determined quickly and effectively.
oscillation frequency. (3) By analysis of the mechanical condition and thickness
condition, the calculation formulas of repairing par-
ameters could be obtained. According to the formulas, a
Conclusions repairing scheme could be given for avoiding the
In order to better repair and precisely control sticking-type casting speed to be decreased too low or cast stop, and
breakout after alarm, this paper deeply studies the recovery thereby reducing the loss. For example, when operating
mechanism of sticking-type breakout. Through the analysis casting speed is 1.2 m min−1 and the distance from hot
of recovery behaviour of the sticking-type breakout, its math- spot to the meniscus is 0.505 mm, if the casting speed
ematical model has been established, and the relevant recov- is decreased to 0.6 m min−1 with the acceleration
ery conditions are deduced. At the same time, the influencing −0.1642 m min−1 s−1, the recovery time is only 25 s and
factors of the recovery behaviour are analysed. The following there is no need for cast stop.
are main conclusions drawn from this study. (4) The locations of the hot spot, operating casting speed and
mould oscillation frequency have a significant effect on
(1) According to the mechanical condition of sticking-type the minimum speed for the recovery, and mould oscil-
breakout recovery, the minimum casting speed for ensur- lation stoke has little influence on it.
ing that the shell is not torn, could be obtained at differ-
ent operating casting speeds and hot spot’s locations. At
the same time, it is found that reasonable control of
Disclosure statement
mould oscillation parameters could improve the shell
strength, and the improvement effect of oscillation fre- No potential conflict of interest was reported by the authors.
quency is better than the oscillation stroke.
(2) The calculation method of safe shell thickness at the
mould exit has been proposed. By constructing the Funding
force analysis model of equivalent simply supported The authors are grateful to the financial support for this project from the
beam of the shell at the mould exit, the minimum safe National Natural Science Foundation of China (Grant No. 51504002).
 10 H. QIAN ET AL.

References [8] Tsuneoka A, Ohashi W, Ishitobi S, et al. Measurement and control

system of solidification in continuous casting mold. Proceedings of
[1] Liu Y, Wang X, Shi G, et al. Process factors of sticking breakout for ISS-AIME Steelmaking Conference. Detroit: Iron and Steel Society,
wide and thick slab continuous casting. J Univ Sci Technol Beijing. 1985:3–10.
2014;36(6):757–762. [9] Itoyama S, Yamanaka H, Tanaka S, et al. Prediction and prevention
[2] He J, Wang C, Wang Z, et al. Reason analysis and control system for sticking type breakout in continuous casting.
measures of container steel by CSP process. Steelmaking. Proceedings of the 71st Steelmaking Conference. Warrendale: Iron
2014;30(4):20–24. and Steel Society, 1988:97–102.
[3] He F, He DF, Deng ZH, et al. Development and application of [10] Blazek KE, Saucedo IG. Characterization of the formation, propa-
mould breakout prediction system with online thermal map gation, and recovery of sticker/hanger type breakouts. ISIJ Int.
for steel continuous casting. Ironmak Steelmak. 2015;42 1990;30(6):435–443.
(3):194–208. [11] Moon CH, Lee DM, Moon SC, et al. Re-start technology for redu-
[4] Liu Y, Wang X, Du F, et al. Visual detection based on computer vision cing sticking-type breakout in thin slab caster. ISIJ Int. 2008;48
for sticker breakout in slab continuous casting. Ironmak Steelmak. (1):48–57.
2014;31(5):357–363. [12] He F, Zhou L, Deng Z. Novel mold breakout prediction and control
[5] Blazevic D, Ikonic M, Mikac T. Algorithm for prevention of molten technology in slab continuous casting. J Process Control.
steel sticking onto mold in continous casting process. Metalurgija. 2015;29:1–10.
2008;47(1):47–50. [13] Gan Y, Ni M, Yu Z, et al. Practical handbook for modern continuous
[6] Langer M, Arzberger M. New approaches to breakout prediction. casting. Beijing: Metallurgical Industry Press; 2010
Steel Times Int. 2002;26(9):23–26. [14] Fan Q, Tang J. Engineering mechanics: static and material mech-
[7] Qian J. Analysis and application of sticking breakout prediction anics. Beijing: Higher Education Press; 2007.
system in CSP caster. J Anhui Vocational College Metall Technol. [15] Wang B, Huang X, Du W. Engineering mechanics. Hefei: China
Downloaded by [Anhui University of Technology] at 16:48 29 August 2017

2013;23(1):26–29. Science and Technology University Press; 2014.

View publication stats