hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_presentations/04_gogi...
TRANSCRIPT
![Page 1: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/1.jpg)
ANNUAL REPORT 2006Meeting date: June 15, 2006
Go-Gi Lee(Visiting Scholar from POSTECH, South Korea)
Brian G. Thomas
Department of Mechanical & Industrial EngineeringUniversity of Illinois at Urbana-Champaign
Hook formation near slab corners and prediction of argon gas bubble size
in the slab casting mold
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 2
Acknowledgements
• Continuous Casting Consortium at UIUC(Accumold, Algoma steel, Bosteel, Corus, Mittal Riverdale, LaBein, LWB Refractories, Nucor, Postech, Steel Dynamics)
• POSCO
• Korea Research Foundation
• Professor Brian G. Thomas
• Professor Seon-Hyo Kim
• Dr. Ho-Jung Shin, POSTECH
![Page 2: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/2.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 3
1. Investigation of 3-D Frozen Meniscus Shape
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 4
3-D Frozen Meniscus Region- corners of slab
< Narrow face results >
H.-J. Shin, CCC Annual Report, 2005
• Micrographs of the hook come from 2-D vertical section view of certain location • Near the corners of the slab: deepest hook depth at corners; complex 3-D hook shape
3-D shape of frozen meniscus at corners is not clear
Curved hook at 20mm distance from the corner
-110 -90 -70 -50 -30 -10 10 30 50 70 90 1100.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
OutsideCenter of narrow face
Inside
Mea
n ho
ok d
epth
(m
m)
Perimeter of narrow face (mm)
2002 POSCO, measured 2003 POSCO, measured 2004 POSCO, measured
0 2mm
Scale
Bubble
J. Sengupta et al., Acta Materilia, 2006
![Page 3: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/3.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 5
Metallurgical Method
20mm
Corner
CL
115mm
Sample for microscopy analysis
13mm Wide face
Casting direction
Sample size: 20mm(width) * 13mm(depth) * 30mm(height)
OM
Narrow face
• Ten micrographs (2-D vertical section view) were obtained from YZ planeaccording to X distance(~6mm) from corners
< Location of sample > < Definition of axes >
Narrow face
Wide faceX
Y
ZCORNER
OM
Observing plane (YZ plane)
Casting Direction
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 6
Casting Conditions
0.2410.100121745.0
Positive strip time (sec)
Negative strip time (sec)
Modification ratio for non-sinusoidal mode (%)
Frequency(cpm)
Stroke(mm)
- Slab thickness: 230 mm; Slab width: 1300 mm; Casting speed: 1.45 m/min; Pour temperature: 1571˚C
~ 0.040.050.01
~ 0.020.010.01~ 0.01~ 0.015≤ 0.0050.08< 0.005
AlTiCuNiCrSPSiMnC
3.21118011701149
Viscosity at 1300 oC(Poise)
Melting Temperature(oC)
Softening Temperature(oC)
Solidif. Temperature (oC)Properties
0.3506.720.113.430.030.030.340.185.970.8439.836.31.10
Li2OB2O3FK2ONa2OP2O5MnO2Fe2O3TiO2Al2O3MgOCaOSiO2BasicityChemicalcomposition
(wt. %)
Steel composition of ultra-low carbon steel (wt. %)
Mold powder composition and properties
Mold oscillation conditions
![Page 4: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/4.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 7
Actual Measurement of Micrograph
Micrograph of YZ plane at X = 1.7mm
Blue color: Upper line of frozen meniscusRed color: Lower line of frozen meniscus Gray color: Profile of oscillation markGreen color: Tip of frozen meniscus
1.8
1.7
1.6 02
46
810
12
-25
-20
-15
-10
-5
0
3rd OM
2nd OM
Ca
stin
g di
r ect
i on
Wide face (mm)
Narrow face (mm)
1st OMScale
0 2mm
1st
OM
2nd
OM
3rd
OM
Z=0
Slab surfaceY=0
DT
DT : Distance from slab surface to tip of frozen meniscus
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 8
Analysis of Micrographs
6 5 4 3 2 1 0 02
46
810
12-25
-20
-15
-10
-5
0
5
3rd OM
2nd OM
Ca
stin
g d
ire
ctio
n (
mm
)
Wide fa
ce (m
m)
Narrow face (mm)
Corner
1st OM
2.5mm3.5mm
5.0mm
< Micrographs according to the X direction >
Blue color: Upper line of frozen meniscusRed color: Lower line of frozen meniscus Gray color: Profile of oscillation markGreen color: Tip of frozen meniscus
X = 2.5mm 3.5mm 5.0mm
Scale
0 2mm
1st
OM
2nd
OM
3rd
OM
![Page 5: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/5.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 9
Investigate of Frozen Meniscus Shape
Hook characteristics:Continuous defect along the OM; 3-D structure
0 2 4 6 8 10 12 14 16 18 20 220
2
4
6
8
10
12
14
16
18
20
22
Wid
e F
ace
(mm
)
Narrow Face (mm)
DT at 1st OM
DT at 2nd OM
DT atf 3rd OM
DT: Distance from slab surface to
the tip of the hook structure
Corner
Max. hook depth point
65
43
21
0 0 2 4 6 8 10 12-10
-8
-6
-4
-2
0
2
4
Ca
stin
g D
ire
ctio
n (
mm
)
Wide Face (mm)
Narrow Face (mm)
Upper lines of forzen meniscus Lower lines of forzen meniscus Tip of frozen meniscus
Corner
Front view of frozen meniscus at 1st OM Tops view of frozen meniscus
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 10
Conclusions
• Frozen meniscus shape near the corner of slabs is investigated with metallurgical analysis:- Micrographs show the 3-D frozen meniscus
shape
- Hook is continuous defect at the oscillation mark running completely around the strand perimeter- Deepest hook depth appears near corners
![Page 6: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/6.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 11
2. Investigation of Argon Bubble Sizein Water Model using Porous Nozzle Refractory
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 12
Water Model Apparatus (1/3 scale)
Water flowmeter
Surface
Ф 25*11 exit
Tundish
500mm
Slide gate
Inner diameter of SEN: 25mm
1200mm
Submergence depth
Gas flowmeter
Water flowmeter
Water bath
Refractory
Pump77mm
Gas injection
High speed camera
Size of MgO refractory brick: 14mm(W)*44mm(L)*17mm(D)
![Page 7: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/7.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 13
Schematic of Upper Tundish Nozzlein Water Model
UTN
Slidegate
SEN
Bottom of the tundish bath
140mm
45mm
340mm
10mm
10mm
Ar injection
Acrylic cap
Ar injection
High speed
camera
Refractory
Outlet port
80mm
22mm
Nozzle wall
25mm
Top view
Side view
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 14
Conditions of the Experiment
< Properties of MgO refractory bricks >
•Fixed test conditions - Submergence depth: 60mm - SEN outport angle: 35°downward - Height of tundish bath: 330mm
• Factors - Total gas flow rate: 0.1, 0.2, 0.3 l /min - Mean water velocity: 0.96, 1.10, 1.25 m/s
Brick 1 Brick 2Fired B.D. (g/cc) 2.9 2.9Porosity (%) 16.2 17.6Permeability (nPm) 7.52 16.32Modulus of rupture (psi) 1085 1119
< Test conditions in water model >
Note: MgO uses to simulate the dolomite refractories
![Page 8: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/8.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 15
Similarity Analysis between Steel Caster & Water Model
2
2 2
3
( )
( )
lliq
gas
l
l
l g
g g
l g
l
li
lq
UInertial forcedueto liquid momentum
Buoyancy force Dg
d VInertial forcedueto gas momentum
Buoyancy force D g
DUInertial forcedueto liquid momentum
Viscous forceof liq
Fr
F
uid
r
Re
Re
ρρ ρ
ρρ ρρμ
= =−
= =−
= =
2 2
2 2
g g
g
l l
g g
gas
liq
gas
dVInertial forcedueto gas momentum
Viscous forceof gas
D UInertial forcedueto liquid momentum
Surfacetension force D
d VInertial forcedueto gas momentum
Surfacetension force
We
eD
W
ρμ
ργ
ργ
= =
= =
= =
Ref.) H. Bai: Ph. D. Thesis. UIUC, Urbana, IL, 2000
,
: , : ,
: , : , : ,
: , : , : ,
: /
l g l
g l g
where
D averagebubblediameter at a pore d gas injection porediameter
liquid density gas density liquid viscosity
gas viscosity U liquid velocity V gas velocity
liquid gas surfacetensioncoefficient
ρ ρ μμγ , :g gravitational acceleration
**) Note: Gas density was calculated for the temperature in liquid and the pressure due to the height of liquid in tundish (see appendix 2)
Dimensionless groups
Physical properties used in calculation
ParametersWater-air
systemSteel-hotAr system
Liquid density (ρl , kg/m3) 1000 7020
Gas density (ρg , kg/m3) 1.331* 0.446*
Liquid viscosity (μl , kg/m·s) 0.001 0.0056
Gas viscosity (μg , kg/m·s) 1.70E-05 7.42E-05
Surface tension coefficient (γl/g , N/m) 0.073 1.192
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 16
Liquid Flow Rate between Two Systems
( )( )
( )( )
( )
3
3
,
,
2,
,
,
,
,
,
,
W
S
W nozzle
S nozzle
W nozzle
Here
Q is the water flowrate m s
Q is the steel flowrate m s
U is the water velocity in water mo del nozzle m s
U is the steel velocity in actual nozzle m s
A is the cross sectional area of water model nozzle m
A
−
( )( )
2, ,S nozzle
C
is thecross se ctional area of actual nozzle m
V is thecasting speed m min
−
Table. Relationship between the liquid flow rate and casting speed
We choose the same liquid velocity between two systems, because the refractory geometry scale is 1:1
, ,
, ,
W nozzle S nozzle
S S nozzle S nozzle
U U
Q U A
=
∴ = ⋅
Note) Bore diameter of SEN is 25mm in water model and 75mm in steel caster, respectively
Water flow rate (l /min) Qw 28.2 32.5 36.8Liquid velocity in nozzle (m/sec) Uw,nozzle(=Us,nozzle) 0.96 1.10 1.25
Steel thoughput (m3/min) Qs 0.25 0.29 0.33
Casting speed (m/min, 230mm*1500mm slab) Vc 0.74 0.85 0.96
SC
QV
Cross setional area of the strand
⎛ ⎞∴ = ⎜ ⎟−⎝ ⎠
, , ,,
WW W nozzle W nozzle W nozzle
W nozzle
QQ U A U
A= ⋅ ∴ =
From mass balance
![Page 9: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/9.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 17
Gas Flow Rate Ratiocaused Surface Area of Refractory
, ,
air in air inair out
T W U W W
Q QQ
N N A− −
− = =⋅
QAr-hot
QAr-cold
, ,
Ar cold Ar coldAr hot
T S U S S
Q QQ
N N A
β β− −−
⋅ ⋅= =⋅
Qair-out
Qair-in
d
WaterFlow
Liquid Steel Flow
UU
< Schematic of water model experiment > < Schematic of steel caster >
Ar hot W Ar cold
air out S air in
Q A Q
Q A Qβ− −
− −
=:
: *
*) 1
W
S
Ar gas expansincoefficient
A Surfacearea ratioof two refractoryA
seetheappendix
β
:
:
:
T
U
N Total active sites oninner wall surfaceof refractory
N Active sites per unit areaof refractory
A Inner wall areaof refractory
d
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 18
Summary of Dimensionless Number Analysis
0 1 2 3 4 5 6 7 8 9 100.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
High liquid velocity (U>1.4m/s) Fr
liq Re
liq We
liq
Frgas
Regas
Wegas
Qa
ir-in (
l/min
)
QAr-cold
(l/min)
At high liquid velocity (U>1.4m/sec):Dair≈DAr at the same gas flow rate
Qair-in Qair-in
QAr-cold QAr-cold
(This work) (Bai's calculation)Frliq 0.07 0.285Reliq 0.087 0.356Weliq 0.03 0.123Frgas 0.0152 0.067Regas 0.005 0.026Wegas 0.01 0.044
Dimensionlessnumber
Similarity requirement for gas flow rate
Matching all of the dimensionless groups at the same time is impossible
![Page 10: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/10.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 19
Active Sites in Stagnant Flow
LWB RefractoriesLow permeability(7.52nPm) Qg = 0.3 l/min
22 active sites
Chosun RefractoriesPorosity = 20% Qg = 0.018 l/min
40 active sites0 1mm
Scale
0 1mm
Scale
0.0 0.1 0.2 0.3 0.4 0.50
5
10
15
20
25
30
35
40
45
50
55
60
Permeability (LWB Ref.) 7.52 nPm 16.32nPm
Porosity (Chosun Ref.) 20%
Act
ive
site
s in
sta
gnan
t flo
w (
#)
Total gas flow rate (l/min)
• Number of active site and bubble size through the porous refractory in stagnant flow are mainly dependent of the texture of porous refractory
• Generally speaking, active site increases as gas flow rate increases
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 20
Comparison of Mean Active Sites between Flowing and Stagnant Flow
• Active sites in downward flow are about 2 times those in stagnant flow
• Gas flow rate per site of high permeability is slightly higher than that of low permeability in downward-flowing flow at the same total gas flow rate
0.0 0.1 0.2 0.3 0.4 0.50
5
10
15
20
25
30
35
40
45
50
55
60
Downward-flowing flow Low permeability High permeability
Stagnant flow Low permeability High permeability
Mea
n ac
tive
site
s (#
)
Total gas flow rate (l/min)
Test conditions- Low permeability (7.52nPm)- Mean liquid velocity: 1.25 m/s- Total gas flow rate = 0.1 l/min- Camera speed: 4000 frames/s
t=0 t=0.25ms t=0.5ms
< Example of experiment photograph series >
![Page 11: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/11.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 21
Calculation of Total Active Sites in the Steel Caster Nozzle Wall
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.400
1
2
3
4
5
6
7
8
9
10
Downward-flowing flow Stagnant flow Low permeability Low permeability High permeability High permeability
NU (
#/cm
2 )
QS (ml/s)
*) see appendix 1
0 1 2 3 4 5 6 7 8 9 10 11 120
5
10
15
20
25
NU (
#/c
m2)
QT (l/min, cold)
Low permeability High permeability measurement measurement estimated value estimated value
Downward-flowing flow
Nu (#/cm2) = 2.48 lnQT (l/min) + 9.8
Nu (#/cm2) = 2.15 lnQT (l/min) + 8.7
( ) ( )( )
TS
T
Q Total gas flowrateQ Mean gas flowrate per site
N Total active sites=
• Low permeability is much more active sites
Water Model Steel CasterQT (Total gas flow rate-cold, l/min) 0.2 7
NU (Mean active sites per unit area, #/cm2) 5.5 13.732
A (Inner wall area of refractory*, cm2) 6.2 353.4NT (Total active sites, #) 34 4853
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 22
Comparison of Estimated Mean Gas Flow Rate per Site between Two Systems
0.00 0.05 0.10 0.15 0.20 0.250.00
0.05
0.10
0.15
0.20
0.25
Me
asu
red
mea
n g
as
flow
ra
tep
er
site
in w
ate
r m
od
el (
ml/s
, ho
t)
Estimated mean gas flow rate per site in steel caster (ml/s, hot)
Similarity of mean gas flow rate per site between two systemsIt makes the result of water model is meaningful
( )* 2 2
, /
( ) (#/ )T
S U
Q cold l min
A cm N cm
β ×=
×
Estimated mean gas flow rate per sitein steel caster (ml/s, hot)
*) AS = 353.4cm2, see appendix 1
NU (Mean active sites per unit area on inner wall of refractory)= 2.3154 lnQT(cold, l/min) + 9.2263
QT (Total gas flow rate-cold, l/min) 0.1 0.2 0.3 0.4 5 7 9 11β (Gas volume expansion coefficient**)NT (Total active sites on inner wall area, #) 24 34 40 44 4578 4853 5058 5223Estimated mean gas flow rate per site (hot, ml/s) 0.069 0.098 0.125 0.151 0.073 0.096 0.119 0.140
1 4
Steel CasterWater Model
![Page 12: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/12.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 23
Bubble Size Distributions:measured in water model nozzle
1.0 1.5 2.0 2.5 3.0 3.5 4.00
10
20
30
40
50
60
70
80
90
100
Bub
ble
volu
me
per
cen
tag
e (
%)
The range of bubble diameter (mm)
Low permeability High permeability 0.96 m/s 0.96 m/s 1.1 m/s 1.1 m/s 1.25 m/s 1.25 m/s
Mean liquid velocity
Total gas flow rate = 0.1 l/min
1.0 1.5 2.0 2.5 3.0 3.5 4.00
10
20
30
40
50
60
70
80
90
100
Low permeability High permeability 0.96 m/s 0.96 m/s 1.1 m/s 1.1 m/s 1.25 m/s 1.25 m/s
Bu
bbl
e vo
lum
e p
erc
en
tag
e (
%)
The range of bubble diameter (mm)
Mean liquid velocity
Total gas flow rate = 0.3 l/min
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.20.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.20.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Low permeability (7.52nPm)
Sta
nda
rd d
evia
tion
of b
ubb
le d
iam
eter
(m
m)
Mean bubble diameter (mm)
Correlation coefficient = 0.83047Correlation coefficient = 0.72677
High permeability (16.32nPm)
• Bubble size variations increase with increasing gas flow rate and increasing bubble diameter
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 24
Effect of Liquid Velocity and Gas Flow Rate on Bubble Diameter (water model)
• The mean bubble size increases with increasing gas flow rate and decreasing liquid velocity
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.40.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
Low permeability High permeability
Mea
n bu
bble
dia
met
er (
mm
)
Gas injection volume fraction (%)
*100( )
Total gas flowrateGas injection volume fraction
Total gas flowrate Liquid flowrate=
+
![Page 13: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/13.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 25
Comparison of Measured and Predicted Gas Bubble Size
0.0 0.1 0.2 0.3 0.4 0.50.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mea
n bu
bble
dia
met
er (
mm
)
Mean gas flow rate per site (ml/s)
d=0.1mm, predicted in water model d=0.4mm, predicted in water model Low permeability, measured High permeability, measured Liquid steel argon prediction
Mean liquid velocity = 0.96 m/s
d: Gas injection hole diameter
0.0 0.1 0.2 0.3 0.4 0.50.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mea
n b
ubb
le d
iam
eter
(m
m)
Mean gas flow rate per site (ml/s)
d=0.4mm, predicted in water model d=0.1mm, predicted in water model Low permeability, measured High permeability, measured Liquid steel argon prediction
Mean liquid velocity = 1.25 m/s
d: Gas injection hole diameter
Ref.) Bai & Thomas, Met. Trans., 2001
Bubble size prediction model:
Total gas flowrateMean gas flowrate per site
Mean active sites=
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 26
Comparison of Estimated Ar Bubble Size between Steel Caster and Water Model
using the Bai’s PredictionRef.) Bai & Thomas, Met. Trans., 2001
0.0 0.5 1.0 1.5 2.0 2.50
1
2
3
4
5
6
7
8
Mea
n bu
bble
dia
met
er (
mm
)
Mean liquid velocity (m/s)
Air flow into water Ar flow into steel 0.069 ml/s 0.073~0.14 ml/s 0.151 ml/s
Gas injection flow rate per site (hot)
Gas injection hole diameter = 0.1mm
![Page 14: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/14.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 27
Conclusions
• Investigation of Ar bubble size in water model nozzle using porous refractory- Micro texture of the porous refractory greatly affects injected gas behavior in stagnant flow- Active sites in downward-flowing flow are about two times compare with stagnant measurement- Although matching different dimensionless groups at the same time is impossible, the similarity of mean gas flow rate per site between water model and steel caster is good- Bubble size variations increase with increasing gas flow rate and increasing bubble diameter- Mean bubble size increases with increasing gas flow rate and decreasing liquid velocity- The match of mean bubble size between the previous model prediction by Bai and the experimental data is reasonably good
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 28
Appendix 1. Surface Area Ratio
Effective length of UTN considered distribution of Ar gas velocity
0.00
0.05
0.10
0.15
0.20
0.25
0.000.010.020.030.04
Velocity (m/s)
Dis
tan
ce fr
om
bo
ttom
to to
p o
f th
e n
ozz
le (
m). Slab P
LeffLUTN
75mm
150mm
Distribution of Ar gas velocity
Cylinder-like refractory in actual plant
Surface area (AS)= 353.4cm2
Rectangular refractory fixed in water model
Surface area (AW)= 6.2cm2
vs 44mm
14mm
17mm
B. G. Thomas & Z. Hashisho, CCC Annual Report, 2005
![Page 15: Hook formation near slab corners and prediction of …ccc.illinois.edu/s/2006_Presentations/04_GoGi Lee...Sample for microscopy analysis 13mm Wide face Casting direction Sample size:](https://reader035.vdocuments.net/reader035/viewer/2022071105/5fdf41e193ed390a302716c8/html5/thumbnails/15.jpg)
University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Go-Gi Lee 29
Appendix 2. Calculation of Gas Density
Table. Properties used in calculation
Standard pressure Pstd 101325 Pa
The Density of the liquid (kg/m3)
Density of water ρW 1000 kg/m3
Density of steel ρS 7020 kg/m3
The Density of the gas (kg/m3, STP)
Density of Ar gas ρAr 1.783 kg/m3
Density of air gas ρAir 1.29 kg/m3
The height of steel in tundish Hs 1.0 m
The height of water in tundish Hw 0.33 m
( )(1) '
( )
(2) '
.(1) '& (2) '
g
g
g
is density of gas
weight m
volume V
Ideal gas eqution is PV nRT
Combining Eqs gives
mP
nRT
ρ
ρ
ρ
=
=
=
( )( )
( ) ( )
273
273
3
273
273
273
1.032
1.032* 1.331
Air K std std W W
Air std K stdstd
Air AirK std
T P g HP
P T P
kg m
ρ ρρ
ρ ρ
⎛ ⎞⎛ ⎞ ⎛ ⎞+ ⋅ ⋅ ⎛ ⎞= =⎜ ⎟⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎝ ⎠⎝ ⎠⎝ ⎠ ⎝ ⎠
=∴ = =
( )( )
( ) ( )
1833
1833
3
1833
273
1833
1 11.679* 0.25
6.714 4
0.25* 0.446
Ar K std std S S
Ar std K stdstd
Ar ArK std
T P g HP
P T P
kg m
ρ ρρ
ρ ρ
⎛ ⎞⎛ ⎞ ⎛ ⎞+ ⋅ ⋅ ⎛ ⎞∴ = =⎜ ⎟⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎝ ⎠⎝ ⎠⎝ ⎠ ⎝ ⎠
⎛ ⎞= ≈ =⎜ ⎟⎝ ⎠
∴ = =
( )( )
3
3
0.446
1.331
Ar hot
air out
kg m
kg m
ρ
ρ−
−
∴ =
∴ =
∴Ar volume expansion coefficient relative to STP = 4
1) Steel-Ar system
2) Water-air system