Upload
phungdat
View
217
Download
3
Embed Size (px)
Citation preview
CCC Annual ReportUIUC, August 20, 2014
Department of Mechanical Science & Engineering
University of Illinois at Urbana-Champaign
Thermal Stress Cracking
of Ladle Sliding Gate PlatesHyoung-Jun Lee, Seon-Hyo Kim
UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
I L L I N O I S
Department of Materials Science & Engineering
Pohang University of Science & Technology
Brian G. Thomas
Doo-Hoa Cheong, Sang-Woo Han,Yong-Hwan Kim Steelmaking Research Group
POSCO Technical Research Laboratories
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 2
Research Background(Business Case)
A A
section A-A
Japanese Patent [1] Crack location of reusable ladle plate[2]
Longer lifetime of Sliding-Gate is needed JFE and POSCO are developing reusable outer
plate technology Research is needed to understand and present
cracks in ladle plates
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 3
Research Background(Operation Problems)
ShroudNozzle
Lower Plate
Upper Plate
Ladle Bottom
Why are cracksin sliding-gatea concern?
Schematic of ladle-nozzle system [3]
-Safety Problem (Steel leakage)-Clogging (Air penetration)-Require replacing plate every heat
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 4
Schematic of Ladle-Nozzle Sliding-Gate System (Animation)
Lower plate moves horizontally to control the molten steel flow rate through the nozzle
Springs generate cassette pressure on the ladle plate
Ladle bottom
Uppernozzle
Upper frame
Upper plate
Copperrods
Lowerframe
Connected to hydraulic cylinder
Grease
Lower plate
Sliding frame
Shroudnozzle
Tundishtop
Front view (opening change) Side view (cst. pressure)
Links4 Springs in
each side
6-Bolts
Ceramic sheet
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 5
Type of Ladle Plate Cracks [4]
Common thru-cracks (every plate breaks in two)
Rare radial crack (~50% of plates)
Top View Bottom View
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 6
Method of Elastic Modulus Calibration by 3-Point Bending Test
Data vs time from bending test at 1200°C, #1
Final load = 6505 NFinal displacement = 1.414 mm
# sec kgf mm
SymmetrySliding platerefractory specimen
Reference pointcoupled to load roll
Support roll
Elastic modulus is adjusted until reaction force on reference point in FEM matches to final load of measurement
Eroll = 206 GPaνroll = 0.3Eref = ??? GPaνref = 0.2
Load roll
Eref = 1.53 GPa
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 7
2-D Finite Element Model of3-Point Bending Test
Specimen size : 20*20*80 mmSpan distance : 50 mmRoll diameter : 10 mm
2-D transient modelMesh : 2D plane stressShell thickness : 20 mm
tension
compression1
1
22
Specimen : Hex-solid 13,093Load roll : Hex-solid 614Support roll : Hex-solid 3,100
surf. to surf. contactµref-steel = 0.45
UX, UY = 0
Prescribe displacement (t)Strain rate = 0.5mm/min
Symmetry
X
Y
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 8
3-Point Bending Test Resultsat Different Temperatures
Conditions Replicated Tests1 2 3 4
25 ˚CLoad(N) 2805 2266 1521 1793
Stroke(mm) 0.160 0.117 0.116 0.155
400 ˚CLoad(N) 3474 3630 3205 -
Stroke(mm) 0.909 0.958 0.873 -
800 ˚CLoad(N) 3366 3916 4147 -
Stroke(mm) 0.755 1.085 1.842 -
1200 ˚CLoad(N) 6505 7234 - -
Stroke(mm) 1.414 1.502 - -
1400 ˚CLoad(N) 7005 8165 - -
Stroke(mm) 1.598 1.925 - -
The lowest final load at different temperature test is input to 2-D simulation
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 9
Load vs. DisplacementMeasurement/Prediction
0.0 0.4 0.8 1.2 1.60
2000
4000
6000
8000
Load
(N
)
Displacement (mm)
25 Degree C - predicted 400 Degree C - predicted 800 Degree C - predicted 1200 Degree C - predicted 1400 Degree C - predicted 25 Degree C - measured 400 Degree C - measured 800 Degree C - measured 1200 Degree C - measured 1400 Degree C - measured
1521 N
3205 N3366 N
6505 N
7005 N
Failure load
Measured and predicted final load/displacement are matched
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 10
Physical Property of Ladle Plate [5](Chosun Refractories Co. Ltd.)
Model CSN-601N, Alumina-Carbon BrickPhysical Properties Value
Apparent Porosity 5~8 %
Bulk Density 3,100~3,200 kg/m3
Cold Crushing Strength 147.15 ≤ MPa
Modulus of Rupture at 1,400˚C 12.74 ≤ MPa
Thermal Expansion at 1,400˚C 0.8~1.0 %
Chemical Composition
Al2O3 72 ~ 76 %
ZrO2 4 ~ 6 %
C 10 ~ 13 %
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 11
Analytical Solutions of 3-Point Bending Test
Elastic modulusLoad, P
b
h
L
3
3max4static
PLE
bh δ
Max. bending stress
max
My
I
M : Bending momentZ : Modulus of sectionI : Inertia of momenty : Distance between centerline
and top or bottom
max 3 2
342 2
12
PLh PL
bh bh
max
M
Z
IZ
y
3
12
bhI
y
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 12
Elastic Modulus and Critical Tensile Strength Results
Apparent decrease in E with increasing temperature is likely due to creep during test (when temperature exceeds glass transition temperature of ceramic)
Predicted critical tensile strength is taken from FEM simulation at center-bottom of test piece (tension area)
0 400 800 12000
1
2
3
4
5
6
7
Ela
stic
Mod
ulus
(G
Pa)
Temperature (degree C)
2D Finite-element model for E Analytical equation to estimate E
0 400 800 12000
10
20
30
40
50
60
70
Crit
ical
Ten
sile
Str
engt
h (M
Pa)
Temperature (degree C)
Predicted max. tensile stress Analytical max. bending stress
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 13
0 400 800 1200 1600600
800
1000
1200
1400
Spe
cific
hea
t, C
p (J
/kg-
K)
Temperature (degree C)
Thermal Property Measurements (CSN-601N, Chosun Refractories)
y=-603*exp(-x/560)+1396
Conductivity, k (W/m·K) = density, ρ (kg/m3) * diffusivity, κ (m2/sec) * specific heat, CP (J/kg·K)
0 400 800 1200 1600
1
2
3
4
Diff
usiv
ity (
10-6 m
2 /sec
)
Temperature (degree C)
y=3*exp(-x/263)+1.21
measured
Diffusivity Measurement – KICET (Laser Flash, NETZSCH, LFA457)
Specific Heat Measurement – Ajou Univ. (Differential Scanning Calorimeter, NETZSCH, DSC200F3)
[6]
Density, measured by displacement method – POSCO = 3147.5 kg/m3
Alumina-carbon brick (C=30wt%)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 14
Thermal Conductivity & Thermal Expansion Coefficient Evaluation
[6] [6]
Alumina-carbon brick (C=10wt%)
Alumina-carbon brickof POSCO ladle plate
Magnesia brick
High-alumina brick (corundum type)
Alumina-carbon brickof POSCO ladle plate
Alumina-carbon brick(C=10wt%)
0 400 800 12000.0
0.4
0.8
1.2
1.6
Temperature (degree C)
The
rmal
Exp
ansi
on (
%)
Measured thermal properties are well matched to reference data[6]
Thermal Expansion Coefficient Measurement – POSCO (Dilatometer, NETZSCH, DIL402C)
0 400 800 1200 16000
2
4
6
8
10
The
rmal
Con
duct
ivity
(W
/m-K
)
Temperature (degree C)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 15
Ladle-Nozzle Sliding-Gate Components& Cassette Pressure Calculation
Upperframe
Sliding frame
Upper plate
Lower plateSteel band
Ceramic sheet
Settingblock
Cu sliding rods
*Lower frame is omitted
YX
Z
Stopblock
Bolted under ladle bottom
Spring force = 10,000 kgfCu rod area = 0.035 x 0.890 m
210,000 [kgf] 9.81 [m/s ] 11.57 [MPa]
0.035 [m] 0.890 [m] 2
Sym. model
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 16
Ladle-Nozzle Sliding-GateDomain / Finite Element Mesh
Top view
Front view
Right view
Calculation time : 1.5 hrs. (heat transfer), 30 hrs. (stress)
XY
Z
PartsElements
Hexahedral Wedge
Upper plate - 47,764
Upper band 288 -
Upper ceramic sheet - 901
Upper stop block 3,226 -
Upper setting block 2,520 -
Upper frame 147,983 -
Lower plate - 47,764
Lower band 288 -
Lower ceramic sheet - 901
Lower stop block - 2,495
Lower setting block 1,793 -
Sliding frame 51,328 -
Sliding frame Cu rod 632
Lower frame Cu rod 720
Total elements 308,603
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 17
Properties forLadle-Nozzle Sliding-Gate Model
Property Value
Refractory(Plate)
Density ρref 3147.5 kg/m3
Elastic modulus Eref Prescribed PaPoisson’s ratio νref 0.2 -Thermal conductivity kref Prescribed W/m·KSpecific heat Cp,ref Prescribed J/kg·°CExpansion coefficient αref Prescribed °C-1
Emissivity [7] εref 0.92 -
Steel(Band,
Cassette)[7]
Density ρsteel 7860 kg/m3
Elastic modulus Esteel 206 х 109 PaPoisson’s ratio νsteel 0.3 -Thermal conductivity ksteel 48.6 W/m·KSpecific heat Cp,steel 418.6 J/kg·°CExpansion coefficient αsteel 1.78 х 10-5 °C-1
Emissivity εsteel 0.75 -
Friction coefficient
Steel-Steel [8] μsteel-steel 0.3 -Steel-Refractory [7] μsteel-ref 0.45 -Refractory-Refractory [7] μref-ref 0.1 -
Stefan-Boltzmann constant σ 5.669 х 10-8 W/m2·K4
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 18
Variables and Boundary Conditions for Ladle Sliding-Gate Model
Preheating Refining Casting
Opening Ratio [4] - 100 0 70 %
Duration Time [4] t 210 160 33 min.
Initial Temperature Tinitial 25 - - °C
Internal Sink Temperature Ti750 [9]
(Gas)340 [10](Ladle filler)
1600(Molten Steel)
°C
Internal Convection Heat Transfer Coefficient (Forced) [9] hi 65.24 8.82 28.7 x 103 W/m2·K
External Ambient
Temperature[9]
Inside of Cassette area To,in 200 200 270 °C
Outside of Cassette area To,out 100 100 120 °C
External Convection Heat Transfer Coefficient (Free) [9] ho 8.82 8.82 8.82 W/m2·K
Lower plate movementwith 25 mm/min.
Surface pressure 10,000 kgfat Cu sliding rod
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 19
Thermal Behavior (Movie)
Preheating(750˚C, 210min)
Refining(340˚C, 160min)
Casting(1600˚C, 33min)
1000% 070% End of heat
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 20
Temperature & Hoop Stress Histories at Locations where Cracks are Observed
12 3
4 Inner bore
XY Z
0 100 200 300 400
-2
0
2
4
6
8 1, x-stress 2. x-stress 3, y-stress 4, y-stress
Hoo
p S
tres
s (M
Pa)
Time (min)
0
400
800
1200
1600 inner bore 1 2 3 4
Tem
p. (
degr
ee C
)
Preheating Refining Casting
Zoom-innext slide
Compression preload(spring loading against the link)
Tension
Compression
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 21
Temperature & Hoop Stress Histories during Casting (Log Scale on x-axis)
12 3
4 Inner bore
XY Z
Max.differenceat #1
Max.differenceat #2, 3, 4
Casting start Casting end
Tension
Compression1E-3 0.01 0.1 1 10
0
2
4
6
8 1, x-stress 2, x-stress 3, y-stress 4, y-stress
Hoo
p S
tres
s (M
Pa)
Time from Casting (min)
0
400
800
1200
1600 inner bore 1 2 3 4
Tem
p. (
degr
ee C
)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 22
Common Through-thickness Crack Formation Mechanism
X
Y
Growthdirection
Temp.[°C] Stress [MPa]
Crack initiates on outer surface max. (P1 at 5min) Crack grows towards inner bore
260˚C Max at P1:Crack starts
Section view of 10 mm below lower plate top surface, after 5 min of casting
1600˚C
Temp. StressX
StressY
StressZ
Slide frame
Steel band
Stopblock
Settingblock
Lower plate
Critical tensile strength ;~16 MPa at 260˚C
8.6 MPa
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 23
Rare Radial Crack Formation Mechanism
230˚C
Not connected to inner bore
Temp. StressX
StressY
StressX
Temp.[°C]
Stress [MPa]X
YView of lower plate, after 0.5 min of casting
StressY
Growth direction
Max at P2,3,4:Crack starts
1600˚C
1600˚C
Temp.
Bottom view Top view
Max. load is generated at 0.5 min of casting Crack starts on convex (colder) lump part of upper or lower plate
Slide frame
Steel band
Stopblock
Settingblock Temp.
X Y
ZCritical tensile strength ;~15.5 MPa at 230˚C
6.2 MPa
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 24
Mechanical Loading Effects
Z
X
Stress [MPa]
Stress X
Highest temperature causes compression against block
Y
X
Stress Y
Stop blockSetting block
Contact area
Misalignment during assembly leads to concentrated point loads at contact points
Temp.Temp.[°C]
Temp. and stress distribution in plates at the end of casting450˚C
240˚C 230˚C
200˚C 215˚C
Z
X
Stress Y
Temp.
Top view of upper plate at the end of casting
Note: steel band expands away from ceramic, leaving ~5mm gap during casting
5 mm
x10 distortion
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 25
Casting Result without Preheating (Worst Case)
Direct casting and cooling is simulated without preheating for worst case
12 3
4 Inner bore
XY Z
CoolingCasting
Compression preload(spring loading against the link)
0 50 100 150 200 250 300-4
-2
0
2
4
6 1, x-stress 2, x-stress 3, y-stress 4, y-stress
Hoo
p S
tres
s (M
Pa)
Time (min)
0
400
800
1200
1600 inner bore 1 2 3 4
Tem
p. (
degr
ee C
)
0.1 1 10
-2
0
2
4
6
Hoo
p S
tres
s (M
Pa)
Time from casting (min)
Zoom-in during casting
Tension
Compression
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 26
Thermal/Mechanical Distortion Affected by Cassette Pressure
Cassette is plastically deformed after several times using for casting according to plant engineer – Deformation shape effects of plate in the plant will be discussed in future work
Casting (1600˚C,33min) Cooling(25˚C, 270min)
End of cooling
Stress histories, x 10 distortion
End of heat
Compression preload
Front view - symmetric plane, XZ(Stress X)
Right mirror view (Stress Y)
Stress X [MPa] Stress Y [MPa]
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 27
Future Works
Ladle bottom
DistortionBolt distortion?
Cracks?
Plastic deformation of used cassette in the plant is needed to investigate
Thermal and mechanical distortion of ladle-nozzle system pushes against upper nozzle, creating forces between upper nozzle and ladle bottom refractory
Creep effect in ceramic materials is important for cracking
Uppernozzle
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 28
Reusable Ladle-Nozzle Sliding-GateDomain / Finite Element Mesh
Lower plate innerLower plate outer
XY
Z
PartElements
Hexahedral Wedge
Upper plate - 49,452
Upper band 288 -
Upper Ceramic sheet - 897
Upper stop block 1,729 -
Upper setting block 2,749 -
Upper frame 146,671 -
Lower Plate
Inner - 5,673
Outer - 31,150
Lower band 288 -
Lower ceramic sheet - 897
Lower stop block - 1,729
Lower setting block 2,749 -
Sliding frame 59,164 -
Sliding frame Cu rod 632
Lower frame Cu rod 2,664
Total elements 306,732
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 29
Reusable Lower PlateDomain / Finite Element Mesh
Lower plate bottom surface
XY
Z
Lower plate inner
Lower plate outer
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 30
0
500
1000
1500
Tem
p. (
degr
ee C
) inner bore conventional reusable inner reusable outer
Stress [MPa]
Hoop Stress Results Comparison(Conventional / Reusable)
Larger temperature difference between inside and outside surfaces generates larger tensile stress
Tensile stress can be reduced by using outer “reusable” plate
Casting start
ConventionalStress
XStress
X
8.5 MPa
6.45.3
X
Y
Inner crack starts
Outercrack starts
Crack starts
Section view of 10 mm below lower plate top surface, after 5 min of casting
Casting end
Max.difference
TensionCompression
StressY
Reusable
1E-3 0.01 0.1 1 10-2
0
2
4
6
8
Hoo
p S
tres
s (M
Pa)
Time from casting (min)
conventional reusable inner, x-stress reusable outer, x-stress
Growth direction
Growth direction
Growth direction
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 31
Stress [MPa]
Reusable Lower Plate Outer Crack Formation Mechanism
Crack locations after once use [2]
X
Y
StressX
StressY
Crack locations are well matched to used plate In addition to same common crack mechanism:
Expansion of inner plate causes tension in reusable outer plate
x20distorted
Top view of reusable lower plate top surface, after 5 min of casting
Growthdirection
Crack starts
Inner expansion
1
2
1E-3 0.01 0.1 1 10-2
0
2
4
6
8
10
Hoo
p st
ress
(M
Pa)
Time from casting (min)
1, x-stress 2, x-stress
TensionCompression
9.4 MPa
5.3 MPa
Casting start Casting end
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 32
Conclusions
Replicated 3-point bending tests measured refractory strengths at different temps.
Thermal expansion of hot inside of plate causes exterior tensile hoop stress and crack growth towards interior: leading to both common through thickness cracks (starting at cold outside of plate) and rare radial cracks (starting at cold outside of lump).
No cracks are predicted (so quantitative models and fracture criteria need more work.
Reusable plate are predicted to reduce tensile hoop stress, but through-thickness crack formation may be unavoidable.
Two different crack mechanisms can form through-thickness cracks in reusable outer plate middle.
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 33
Acknowledgments
• Continuous Casting Consortium Members(ABB, ArcelorMittal, Baosteel, MagnesitaRefractories, Nippon Steel and Sumitomo Metal Corp., Nucor Steel, Postech/ Posco, Severstal, SSAB, Tata Steel, ANSYS/ Fluent)
• Others - Chosun Refractories Co. Ltd.- Daejoo Machinery Co. Ltd.- POSCO Grant # 4.0009576.01
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Hyoung-Jun Lee • 34
References
[1] Japanese Patent, JP2012-121049A, 2012.6.28
[2] T. Hisanaga, et al., Japanese Journal of Refractories, No.1, pp.26, 2012
[3] J. Choi, POSCO Gwangyang works, Gwangyang, Jeonnam, Korea
[4] D.H. Jeong, et al.. POSCO Technical Research Lab., Pohang, Gyeungbuk, Korea, 2013
[5] Technical Data, Chosun Refractories Co. Ltd., http://www.chosunref.co.kr/home/kor/product/ndata/
[6] T.G. Kyokai, Refractories Handbook, The Technical Association of Refractories, Japan, June, 1998
[7] J.M. Sun, et al., Chosun Refractories Co. Ltd. Research Center, Pohang, Gyeongbuk, Korea, 2010
[8] B.G. Thomas, et al., Iron and steel maker, Vol.25, pp.125, 1998
[9] H.J. Lee, et al., “Thermal Stress Cracking of Sliding Gate Plates”, AISTech proceeding, 2012
[10] S.S. Kim, et al., Journal of Ceramic Society, Vol.38, No.9, pp.829, 2001