Académique Documents
Professionnel Documents
Culture Documents
a r t i c l e i n f o
abstract
Article history:
Received 13 August 2012
Received in revised form
10 November 2012
Accepted 17 December 2012
Available online 24 January 2013
This paper investigates laser beam welding of dissimilar AISI430F and AISI440C stainless steels.
A combined welding and pre-and-postweld treatment technique was developed and used successfully
to avoid micro-crack formation. This paper also examined the effects of laser welding parameters and
line energy on weld bead geometry and tried to obtain an optimized laser-welded joint using a full
factorial design of experiment technique. The models developed were used to nd optimal parameters
for the desired geometric criteria. All the bead characteristics varied positively as laser power increased
or welding speed decreased. Penetration size factor decreased rapidly due to keyhole formation for line
energy input in the range of 1520 kJ/m. Laser power of 790810 W and welding speed of 3.64.0
m/min were the optimal parameters providing an excellent welded component. Whatever the
optimization criteria, beam incident angle was around its limiting value of 151 to achieve optimal
geometrical features of the weld.
& 2012 Elsevier Ltd. All rights reserved.
Keywords:
Laser-welding
Steels
Butt joint
1. Introduction
Laser welding with high power density, high degree of automation and high production rate is extremely advantageous in
industrial applications as stated in [1]. Joints between dissimilar
metals are particularly common in components used in the
automotive, power generation, chemical, petrochemical, nuclear
and electronics industries as described in [2]. The ability to use
different metals and alloys in a product provides the designer and
production engineer with greater exibility, and often results in
technical and economic advantages over components manufactured from a single material.
Welding of metals and alloys is an experienced subject,
dissimilar welding represents a major scientic and technical
challenge. Emerging new technologies increasingly require dissimilar metals and alloys to be joined. Most metals and alloys are
very weldable either in conduction or keyhole modes. Two factors
most important to weldability are hardenability and susceptibility of the hardened structure to cracking. Hardenability is related
to the cooling rate of metals. The faster cooling rate tends to
produce higher hardness and hence, the hardened structure
becomes more susceptible to cracking. According to Kou [3], for
a constant mass of the metal and the particular welding process
126
Fig. 1.1. butt welded surface (a) before and (b) after applying the combined preheat-and-postheat welding technique for the line energy, LE 15 kJ/m.
Fig. 1.2. Schematic diagram to show the variation in incidence angle, A in a plane
tangential to the sample trajectory.
no result has been reported yet in the literature on the laser butt
welding of these particular dissimilar ferritic AISI430F and martensitic AISI440C stainless steels or for similar industrial applications. Moreover, in this study, a novel combined laser welding and
pre-and-postweld treatment technique is developed for overcoming the problems associated with laser welding of dissimilar
stainless steels and applied successfully to the fabrication of fuel
injector. This paper, therefore, examines and optimizes the laser
welding of dissimilar martensitic/ferritic stainless steels in a
constrained butt joint conguration. This study is focused on:
Combined welding and pre-and-postweld treatment concept
and its effects,
Effects of laser welding parameters such as laser power,
welding speed and incidence angle of the beam, impinged
tangentially in the direction of rotation as shown in Fig. 1.2, on
the geometrical features of the weld i.e., on weld width,
resistance length, and penetration depth,
Effects of line energy or energy per unit length on the same
weld geometrical features to understand the energy dependent welding phenomena, and
127
9.560.03
AISI
440C
440C
AISI
430F
430F
Rm
Inner Shell
Inner
Shell
Outer
Shell
Dp
Inner
Shell
Outer Shell
Outer
Shell
7.4580.015
Fig. 2.1. Characterization of welding cross-section (W: Weld width, Dp: Weld penetration depth, Rm: Minimum crack-path).
Table 2.1
Chemical compositions of base metals of the weld.
Base metal
AISI430F
AISI440C
Cr
Ni
Mn
Si
Mo
Se
Fe
0.12
0.951.2
16.018.0
17.2
0.75
1.0
1.00
0.04
0.03
0.015
1.00
1.00
0.75
0.20
Remainder
Remainder
128
LE 0:06
P
S
kJ=m
Table 2.2
Experimental conditions and response factors.
Process factors
Laser power (W)
Welding speed (m/min)
Angle of incidence (1)
Constant factors
Base material
Laser source
Shielding gas
Response factors
Weld bead characteristics
Symbols
P
S
A
Actual values
600
2.0
15
800
3.0
30
Outer shell
Inner shell
Continuous wave Nd:YAG laser
Type
Flow rate
1000
4.0
45
AISI 430F
AISI 440C
Argon
29 l/min
Weld width (W), weld penetration depth (Dp), and minimum crack-path (Rm)
Table 2.3
Design matrix with actual factors and measured mean responses.
Standard order
Laser
Head
Specimen
Shielding
Gas Nozzle
Specimen
Holder
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Process factors
Response factors
P (W)
S (m/min)
A (deg)
W (mm)
Rm (mm)
Dp (mm)
600
800
1000
600
800
1000
600
800
1000
600
800
1000
600
800
1000
600
800
1000
600
800
1000
600
800
1000
600
800
1000
2.0
2-0
2.0
3.0
3.0
3.0
4.0
4.0
4.0
2.0
2.0
2.0
3.0
3.0
3.0
4.0
4.0
4.0
2.0
2.0
2.0
3.0
3.0
3.0
4.0
4.0
4.0
15
15
15
15
15
15
15
15
15
30
30
30
30
30
30
30
30
30
45
45
45
45
45
45
45
45
45
930.7
1050.3
1021.9
639.6
712.5
851.5
584.6
712.6
881.5
1002.1
1146.8
1217.7
742.3
897.6
899.5
733.6
820.8
848
912.8
1024.6
1073.5
715.8
843.4
933.5
643.3
707.9
901.4
532.5
832.8
1180.6
397.9
822.1
1129.1
292.3
563.4
618.2
375.8
524.9
799.4
195.7
601
669.2
249.3
456.1
510.7
166.8
448.1
539
174.5
376.6
486.1
146.2
298.2
413.6
599.9
956.6
1234.5
438.5
833.8
1148
301.7
594
1043.9
406.2
593.5
959.7
282.8
603.9
1074.4
287.4
464.2
819.2
256.3
631.3
801.4
261.9
458
643.6
205
306.4
505.3
and after the cut using the optical microscope. Hermetic weld is
ensured by performing leak test in vacuum for each of welded
specimens. During leak test, nitrogen is inated into the
assembled part at a pulsed pressure in the range 10150 bar for
the expected life cycles. This method also guarantees that the
weld will not fail during its service life. In case of failure, the
internal cracks generated during the welding process propagate
up to the free surface and N2 leakage is detected by a loss of
vacuum into the chamber.
3. Results and discussion
Various weld prole characteristics are measured with axially
cut specimens using an optical microscope and are recorded for
further analyses described in the succeeding sections.
Perturbation plots are used to illustrate the effects of individual process parameter such as laser power (P), welding speed (S),
and beam incident angle (A) on geometrical features of the weld
e.g., weld width, weld penetration depth, and minimum crackpath. Contour plots are used to show the two-factor interaction
effects on the same weld bead geometry.
The line energy is plotted against weld width, weld penetration depth, and minimum crack-path with a view to demonstrate
GL Ta
Tm
GF
Ta
Tv
GB Ta
Melt pool
Spot
GL
Ta
GL
GF
Tv
Melt pool
GB
Spot
GL
io
n
Fig. 3.1. Top views of melt pools and basic draft of thermal gradients (G) in
different directions with: (a) laser beam perpendicular to the surface (b) laser
beam inclined on the tangential plane.
ct
La
di
re
se
rb
ea
nt
id
e
ri
nc
se
La
Melt pool
129
Fo
pl cal
an
e
+ defocus
sample rot. speed
-defocus
Solidified seam
AISI 430F
AISI 440C
Fig. 3.2. (a) Top and cross sectional views of an instant during the welding process, (b) draft of the basic concept to obtain a less severe heating rate on the fresh material.
130
Perturbation
Perturbation
1220
Weld Width, m
1060
P
900
740
S
A
P
A
1300
1025
P
750
A
S
S
A
475
P
200
580
-1.00
-0.50
0.00
0.50
- 1.00
1.00
-0.50
0.00
0.50
1.00
Perturbation
1200
925
A
650
S
A
375
P
100
-1.00
-0.50
0.00
0.50
1.00
Weld Width, m
737
3.5
791
871
3.0
950
2.5
4.0
4.0
131
3.5
355
487
3.0
618
750
882
2.5
1030
1110
2.0
2.0
600
700
800
900
1000
600
700
Laser Power, W
900
1000
4.0
800
Laser Power, W
3.5
307
3.0
406
504
603
2.5
701
2.0
600
700
800
900
1000
Laser Power, W
Fig. 3.4. Contour graphs to show the interaction effects of P and S on (a) weld width, (b) weld penetration depth, and (c) minimum crack path at A 301.
1400
1400
1200
1000
800
600
400
15
15
30
30
45
45
200
132
1200
1000
800
600
400
15
30
45
200
0
10
15
20
25
30
35
10
1300
1200
1100
1000
900
800
700
15
30
45
600
500
5
10
15
20
25
15
20
25
30
35
30
15
30
3.0
2.5
2.0
1.5
1.0
0.5
0.0
5
35
10
15
20
25
30
35
Fig. 3.5. Effect of line energy on (a) weld penetration depth, (b) minimum crack-path, (c) weld width, and (d) penetration size factor for different beam incident angle.
Table 3.1
Sequential model sum of squares for weld width model.
900
800
700
600
500
400
Source
Sum of
squares
df
Mean
Linear
2FI
Quadratic
Cubic
Residual
4.372E 007
1.325E 006
3042.62
2.340E 005
53388.46
41493.34
Total
1
3
3
3
7
9
Mean
square
F
value
p-value
prob4 F
4.372E 007
4.417E 005 29.28 o0.0001
1014.21
0.059
0.9808
77994.39
13.15
0.0001
7626.92
1.65
0.2363
4610.37
Suggested
Aliased
300
200
30
100
0
200
400
600
800
1000
Table 3.2
Model summary statistics for weld width model.
Source
Std. dev.
R2
Adj R2
Pred R2
PRESS
Linear
2FI
Quadratic
Cubic
122.83
131.56
77.01
67.90
0.7997
0.8015
0.9427
0.9750
0.7724
0.7389
0.9105
0.9304
0.7298
0.6687
0.8369
0.7707
4.477E 005
5.490E 005
2.702E 005
3.799E 005
1200
Suggested
Aliased
Again, the ANOVA table for the weld width model shows that
there is a quadratic relationship between weld width and welding
parameters. The linear terms of laser power (P) and welding speed
(S); and the quadratic terms of welding speed and incident angle are
the signicant model terms associated with the weld width. However, linear term of beam incident angle is added to support hierarchy
of weld width model. For the weld penetration depth model, ANOVA
table demonstrates that all three linear terms i.e., laser power (P),
welding speed (S) and beam incident angle; and two-factor
interactions (2FI) of laser power-welding speed (PS), are the signicant model terms.
From the results shown in Tables 3.13.6, it is, therefore,
apparent that the developed statistical models for predicting weld
width and penetration depth are fairly accurate and can be of
following forms:
(i) Weld width
W1:06 2163:33 0:76P1154:6S 36:97A
154:87S2 0:58006A2
(ii) Weld penetration depth
Dp 446:02 2:16 P106:24 S 7:674 A0:024 P x A
Table 3.3
Sequential model sum of squares for weld penetration depth model.
Source
Sum of
squares
df Mean
square
Mean
Linear
2FI
Quadratic
Cubic
Residual
9.442E 006
2.046E 006
61301.75
8834.15
50208.79
37119.61
Total
1.165E 007
1
3
3
3
7
9
F
value
9.442E 006
6.821E 005 95.30
20433.92
4.04
2944.72
0.54
7172.68
1.74
4124.40
p-value
prob4 F
o0.0001
0.0223
0.6620
0.2158
Suggested
Aliased
26 4.479E 005
Table 3.4
Model summary statistics for weld penetration depth model.
Source
Std. dev.
R2
Adj R2
Pred R2
PRESS
Linear
2FI
Quadratic
Cubic
84.60
71.14
73.88
64.22
0.9285
0.9564
0.9604
0.9832
0.9188
0.9426
0.9381
0.9532
0.8994
0.9240
0.9006
0.8310
2.216E 005
1.674E 005
2.191E 005
3.725E 005
Suggested
Aliased
133
Normality of residual data and amount of residuals in prediction are then checked to ensure statistical validation of the
developed models. The normality of data is veried by plotting
the normal probability plot (NPP) of residuals. The residual is the
difference between observed and predicted values (or tted
value) obtained from the regression model. The data set is
normally distributed if the points on the plot fall fairly close to
the straight line. The normal probability plots of residual values
for weld width, and penetration depth are illustrated in Fig
3.7(a)(b), respectively. The experimental points are reasonably
aligned with predicted or tted points suggesting the normality of
data. This is an implication that empirical distribution of residual
data is well-compared with a normal distribution having the
same mean and variance
Fig 3.8(a)(b) are showing the relationships between the
actual and predicted values of weld width and penetration depth.
Since the points plotted are close to and around the diagonal line,
the difference between the predicted and actual value for each
point can be considered to be minimal. It is also an indication that
the statistical models for prediction are adequate and predicted
results are in good agreement with the measured data.
3.4.2. Numerical optimization
Two criteria are introduced in this numerical optimization.
The rst set of criteria is to maximize weld penetration depth
with no limitation on either process parameters or weld width. In
this case, all the process parameters and weld width (rst
response) are set within a specied range. Furthermore, lowering
the laser power and increasing the welding speed are the most
common techniques used in automotive industries to produce
relatively low-cost and excellent weld joints. Taking these cost
and quality aspects into account, second set of criteria for process
parameter optimization is xed to maximize weld penetration
depth and welding speed, and minimize the laser power and weld
width. Table 3.7 summarizes these two criteria, lower and upper
limits as well as importance for each input and response factor.
Tables 3.8 and 3.9 show the optimal solution based on the two
optimization criteria as determined by design-expert software.
The optimization results clearly demonstrate that, whatever the
Table 3.5
ANOVA table for weld width quadratic model.
Source
Sum of squares
df
Mean square
F value
p-value prob 4F
Model
P
S
A
S2
A2
Residual
1.549E 006
3.876E 005
8.518E 005
19093.30
1.405E 005
93119.36
1.083E 005
5
1
1
1
1
1
20
3.097E 005
3.876E 005
8.518E 005
19093.30
1.405E 005
93119.36
5417.49
57.17
71.54
157.24
3.52
25.93
17.19
o 0.0001
o 0.0001
o 0.0001
0.0751
o 0.0001
0.0005
Cor total
R2 0.9346
1.657E 006
Adj R2 0.9183
25
Pred R2 0.8889
Signicant
Table 3.6
ANOVA table for weld penetration depth 2FI model.
Source
Sum of squares
df
Mean square
F value
p-value prob 4F
Model
P
S
A
PA
Residual
2.101E 006
1.388E 006
1.885E 005
4.895E 005
55065.56
1.024E 005
4
1
1
1
1
21
5.254E 005
1.388E 006
1.885E 005
4.895E 005
55065.56
4876.13
107.74
284.74
38.65
100.38
11.29
o 0.0001
o 0.0001
o 0.0001
o 0.0001
0.0030
Cor total
R2 0.9535
2.204E 006
Adj R2 0.9447
25
PredR2 0.9324
Signicant
134
99
99
95
90
80
70
95
90
80
70
Normal % Probability
Normal % Probability
50
30
20
10
5
50
30
20
10
5
1
-1.70
-0.82
0.07
0.95
1.84
-1.98
Studentized Residuals
-0.85
0.27
1.40
2.53
Studentized Residuals
Fig. 3.7. Normal probability plot for weld (a) width, and (b) penetration depth.
1900
1300
1625
1000
Predicted
Predicted
1350
700
1075
400
800
100
856
1108
1360
1613
1865
124.7
407.6
Actual
690.5
973.4
1256.3
Actual
Fig. 3.8. Scatter diagrams of weld (a) width, and (b) penetration depth.
Table 3.7
Optimization criteria used in this study.
Parameters or
responses
Limits
Table 3.8
Optimal solutions as obtained based on rst criterion.
Lower Upper
P (W)
S (m/min)
A (1)
W (mm)
DP (mm)
600
2.0
15
584
206
Soln no.
Importance Criterion
1000
4.0
45
1218
1235
3
3
3
5
5
First
Second
Is in range
Is in range
Is in range
Is in range
Maximize
Minimize
Maximize
Is in range
Minimize
Maximize
1
2
3
4
5
6
7
8
9
10
P (W)
S (m/min)
F (lm)
DP (lm)
W (lm)
998.6
999.3
998.9
997.9
999
999.5
998.9
998.5
999.6
999.7
2.02
2.16
2.03
2.00
2.02
2.04
2.03
2.14
2.18
2.01
15.2
15.1
15
15
15.5
15.6
15.2
15.2
15
15.8
1248.1
1235.9
1250
1251.7
1243.8
1240.4
1247.4
1235.3
1236
1240.8
1084.6
1039.7
1078.5
1088.6
1088
1082.3
1081.1
1048.7
1033.9
1095
In this case, as shown in Table 3.9, the weld width and penetration
depth would be of 665 mm and 706 mm, respectively, which are
much higher than the prerequisite values for the weld. These results
also indicate the fact that laser welding ought to be conductionlimited for this particular joint type and laser-material combinations
in order to obtain the optimal geometrical features of the weld.
Since optimal range of laser power and welding speed selected
based on second criterion is, respectively, much lower and higher
than that obtained for rst set of criterion, any combination of process
parameters of the second optimal set would cause less energy input
to constrained butt joints to be made. This reduced energy input to
weld materials would, ultimately, result in less distortion, and
135
Table 3.9
Optimal solutions as obtained based on second criterion.
Soln no.
1
2
3
4
5
6
7
8
9
10
S (m/min)
F (lm)
DP (lm)
W (lm)
792.7
794.9
790.1
795.9
789.2
797.2
826.3
816.9
789.5
810.9
3.95
3.94
3.98
3.99
4.00
3.73
3.99
3.78
3.66
3.63
15.00
15.00
15.00
15.00
15.00
15.00
15.00
15.00
15.00
15.00
676.3
681.3
668.5
678.2
664.6
707.6
731.9
737.6
700.9
742.8
705.4
706.1
705.5
708.7
706.2
702.8
723.9
712.6
699.5
710.3
Table 3.10
Visual inspection of weld quality.
Process parameters
P (W)
600
800
1000
600
800
1000
600
800
1000
Visual check
S (m/min)
A (1)
Cracks
Blow holes
Spatter
2.0
2.0
2.0
3.0
3.0
3.0
4.0
4.0
4.0
15
15
15
15
15
15
15
15
15
0
1
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
2
2
1
1
2
0
0
1
This study is carried out to investigate the various microstructures typically formed at the weld zone and various locations
close to both base metals for a line input of 12.0 kJ/m.
The particular sample selected for the metallurgical study is one
of those samples that provide the weld with designed geometry.
Fig 4.1(a) shows the micrograph of the weld made of AISI440C
and AISI430F stainless steels.
As shown in Fig 4.1(b) and (c), the microstructures at the inner
shell are cellular dendritic, whereas columnar dendritic structures
form at the outer shell. These microstructures are a result of
solidication behavior and subsequent solid-phase transformation controlled by melting ration of two materials to be joined
and weld cooling rate [17]. Since base metals consist of martensitic AISI440C and ferritic AISI430F stainless steels, the microstructures that form in the fusion zone must contain a variety of
complex martensite-ferrite structures. Moreover, as shown in
Fig (c), the intra-granular Cr23C6 carbide formation is evident in
the microstructures forming near the fusion zone boundary. As a
result, at the outer shell, average microhardness of the fusion
zone is found to be much higher than that of both the heat
affected zone and the base metal as can be seen in Fig. 4.2.
Fig. 4.1(b) shows the microstructures of base metal and HAZ of
the inner shell, which is pre-hardened and-tempered. Both
microstructures contain partly spherodized primary carbide particles in a tempered martensitic matrix. However, microstructures
Fig. 4.1. (a) typical micrograph of laser welding of AISI440C and AISI430F stainless steels, and base metal (BM), heat affected zone (HAZ) and Fusion zone (FZ) of (b) inner
shell and (c) outer shell.
136
600
Local Microhardness
Intersection
Line
500
400
300
AISI
440C
AISI
430F
HAZ
200
BM
HAZ
Welding Pool
FZ
FZ
Boundary
Boundary
BM
Fig. 4.2. Vickers local microhardness prole along the line shown in Fig. 4.1(a).
5. Conclusions
Using the laser machine and within the limits of the laser
parameters considered in this study the following points can be
concluded:
The developed combined welding and pre-and-postweld
treatment technique is able to overcome the crack formation
problem associated with laser welding of dissimilar stainless
steels.
Laser power and welding speed are the most signicant laser
welding input factors and have opposite effects.
For welding dissimilar stainless steels in a butt joint conguration, weld penetration depth determines the minimum
crack-path of the weld.
Formation of keyhole results in rapid change in the weld
geometrical features within a certain range of energy input. After
the upper limiting value, formation of upper keyhole plasma
plume only contributes to the change in shape of the weld bead.
Various, complex martensiteferrite microstructures develop
in the fusion zone. As a consequence, average microhardness
of the fusion zone becomes much higher than that of base
metal of the outer shell and lower than that of base metal of
the inner shell.
By means of design of experiment inspired by full factorial
design, it is possible to achieve the best operating parameter
window.