materials

Article
Investigation of the Residual Stress in a Multi-Pass T-Welded
Joint Using Low Transformation Temperature Welding Wire
Zhongyuan Feng 1,2, * , Ninshu Ma 1, * , Seiichiro Tsutsumi 1 and Fenggui Lu 3

1 Joining and Welding Research Institute, Osaka University, Osaka 567-0047, Japan; [email protected]
2 Graduate School of Engineering, Osaka University, Osaka 565-0871, Japan
3 School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;
[email protected]
* Correspondence: [email protected] (Z.F.); [email protected] (N.M.)

Abstract: We investigated whether low transformation temperature (LTT) welding materials are
beneficial to the generation of compressive residual stress around a weld zone, thus enhancing the
fatigue performance of the welded joint. An experimental and numerical study were conducted in
order to analyze the residual stress in multi-pass T-welded joints using LTT welding wire. It was
found that, compared to the conventional welded joint, greater tensile residual stress was induced
in the flange plate of the LTT welded joints. This was attributed to the reheat temperature of the
LTT weld pass during the multi-pass welding. The formerly-formed LTT weld pass with a reheat
temperature lower than the austenite finish temperature converted the compressive residual stress
into tensile stress. The compressive residual stress was generated in the regions with a reheat
temperature higher than the austenite finish temperature, indicating that LTT welding materials are
more suitable for single-pass welding.

Keywords: low transformation temperature weld metal; residual stress; martensitic transformation;



reheat temperature; multi-pass welding



Citation: Feng, Z.; Ma, N.; Tsutsumi,

S.; Lu, F. Investigation of the Residual
Stress in a Multi-Pass T-Welded Joint
1. Introduction
Using Low Transformation
Temperature Welding Wire. Materials
It is well known that fusion welding has been applied to a large number of engi-
2021, 14, 325. https://doi.org/
neering components. However, the non-uniform temperature distribution and constraint
10.3390/ma14020325 during the welding process engender residual stress in the welded joints [1]. Generally,
tensile residual stress is detrimental to the fatigue strength, corrosion resistance and crack
Received: 1 December 2020 resistance of welded joints [2–4]. In order to minimize these adverse effects, it is necessary
Accepted: 7 January 2021 to take appropriate measures, such as ultrasonic peening [5], TIG dressing [6], local heat
Published: 10 January 2021 treatment [7] and so on. Nevertheless, these post-weld treatments are costly and time
consuming.
Publisher’s Note: MDPI stays neu- During the 1970s, Jones and Alberry found that the stress accumulation in steels under
tral with regard to jurisdictional clai- constraint conditions can be reduced by bainitic or martensitic transformation [8]. With
ms in published maps and institutio- the development of materials science, Murata et al., in the 1990s, successfully developed
nal affiliations. an iron-based alloy using chromium and nickel as the main alloying elements, producing
a martensite start (Ms) temperature of approximately 180 ◦ C [9]. Owing to the lower
Ms temperature, this type of iron-based alloy is also referred to as low transformation
Copyright: © 2021 by the authors. Li-
temperature (LTT) alloy. The LTT alloys exploit the plasticity associated with displacive
censee MDPI, Basel, Switzerland.
transformation in order to offset thermal contraction strain, thus mitigating the tensile
This article is an open access article
residual stress around the weld zone [10,11]. Due to tensile residual stress mitigation,
distributed under the terms and con- numerous studies have shown that LTT alloys can significantly enhance the fatigue per-
ditions of the Creative Commons At- formance of welded joints [12–14]. Additionally, the LTT alloys were combined with TIG
tribution (CC BY) license (https:// dressing for the further improvement of the fatigue strength [15]. In addition to fatigue
creativecommons.org/licenses/by/ life improvement, the researchers also used the LTT alloys to reduce distortion [16] and to
4.0/). resist cold cracking [17]. In order to obtain an optimized residual stress around the weld

Materials 2021, 14, 325. https://doi.org/10.3390/ma14020325 https://www.mdpi.com/journal/materials

Materials 2021, 14, 325 2 of 14

zone, much attention was focused on the effects of the martensite start (Ms) temperature,
which involved both experimental measurements and numerical simulation [18–20]. It was
found that if the martensitic transformation occurred at an elevated temperature of more
than 400 ◦ C, the martensite finish (Mf) temperature would be located above the ambient
temperature, resulting in a build-up tensile stress due to thermal contraction during the
cooling process [21,22]. Regarding Ms temperatures lower than 80 ◦ C, the martensitic
transformation was not fully completed, which led to a smaller volume expansion strain
and rendered the tensile residual stress mitigation insufficient [23]. Another relevant factor
that needs to be considered is whether the interpass temperature or the reheat temperature
is above or below the Ms temperature. Since a single-pass weld goes through only one
thermal excursion, it is very easy to control the welding’s residual stress. In practice,
most engineering structures are fabricated with multi-pass welds. Unlike a single-pass
weld, multi-pass welds undergo several thermal excursions as subsequent weld passes
are deposited. Whether tensile residual stress can be induced in a multi-pass welded joint
is greatly dependent on the reheat temperature. Experimental and numerical studies on
residual stress in multi-pass butt-welded joints have shown that, if the reheat temper-
ature is higher than the Ms temperature, much of weld zone is left under compressive
stress; otherwise, the thermally-induced residual stress increases again [24]. Of course,
the specimen dimensions, welding parameters and number of weld passes also affect the
reheat temperature. According to previous studies, the final weld bead usually has a lower
compressive residual stress, and there are smaller residual stresses induced at the weld toe
of butt-welded joints due to the strong reheat effect of the final weld pass [25]. Generally,
fatigue cracks are initiated at these critical positions, such as the weld toe and the weld
root. Therefore, LTT butt-welded joints with smaller residual stresses at the weld toe may
still have a better fatigue life in comparison with a conventional welded joint. Apart from a
butt-welded joint, the fillet welded joint, like the T-welded joint, is also one of the most
common connections in welded fabrication. However, all of the previous studies have
focused on the effect of the reheat temperature on butt-welded joints, rather than fillet
welded joints. Hence, it is of great significance to conduct a study of the reheat effect on
the residual stress in fillet-welded joints.
The aim of this paper is to investigate the effect of the reheat temperature on the
residual stress in multi-pass T-welded joints using LTT welding wire. Accordingly, the
residual stress distribution of the conventional T-welded joint was used for reference.
In order to clarify the residual stress development in the course of the welding process,
a finite-element analysis was performed using the in-house software JWRIAN [26,27].
Finally, the computed residual stresses were compared with those obtained from the
experimental measurement.

2. Materials and Methods

A schematic of a T-welded joint is displayed in Figure 1. Two T-welded joints were
fabricated using a gas metal arc welding process using two types of commercial welding
wires, LB-3AD and LB-47, respectively. Accordingly, the former welded joint belongs
to the LTT welded joint, while the latter is the conventional welded joint. The chemical
compositions of the SM490A welded plates, the filler wire and their weld metals are listed
in Table 1. During the welding process, the welding current, voltage and speed were 130 A,
22 V and 120 mm/min, respectively. Meanwhile, a shielding gas, 80% Ar + 20% CO2 , was
employed throughout. Each weld pass was cooled down to an ambient temperature of
15 ◦ C before the start of the subsequent weld pass. After the fabrication of the welded
joints, residual stress measurement was conducted by means of µ-X360 FULL 2D Portable
X-ray Residual Stress Analyzer (PULSTEC, Tokyo, Japan) using the cosα method [28]. In
the cosα method, the signal of a diffraction ring (Debye ring) is detected by an area detector.
Materials 2021, 14, 325 3 of 14

Then, the strain εα where α = 0–360◦ at the Debye ring is applied in order to compute the
residual stress in the welded specimen, and can be expressed as follows:

σx h 2  i σ h  i 2(1 + ν)
y
εα = n1 − ν n22 + n23 + n22 − ν n21 + n23 + τxy n1 n2 (1)
E E E
where n1 , n2 and n3 represent the orientation cosine, respectively.

n1 = cos η sin ψ0 cos φ0 − sin η cos ψ0 cos φ0 cos α − sin η sin φ0 sin α (2)

n2 = cos η sin ψ0 sin φ0 − sin η cos ψ0 sin φ0 cos α + sin η cos φ0 sin α (3)
n3 = cos η cos ψ0 + sin η sin ψ0 cos α (4)
where η, φ0 and ψ0 represent the diffraction angle between the reflection line and the input
X-ray, the orientation angle between the projection of the input X-ray on the sample’s
surface and the axis x, and the orientation angle between the normal line of the sample and
the input X-ray, respectively.
A set of four strains, ε α , ε π +α , ε −α and ε π −α , were measured for each α in order to
minimize the possible experimental error, which can be expressed as follows:

1
a1 ≡ [(ε α − ε π +α ) + (ε −α − ε π −α )] (5)
2
Then,
1+v
a1 = − σx sin 2ψ0 sin 2η cos α (6)
E
Hence, σx can be determined from the slope, M1 , of the linear relationship between a1
and cos α, as:
 
E 1 1 ∂a1 E 1 1
σx = − =− M (7)
1 + ν sin 2ψ0 sin 2η ∂ cos α 1 + v sin 2ψ0 sin 2η 1

The measuring parameters of the XRD device are shown in Table 2. Owing to the
geometry of the T-welded joint, it is very difficult to measure the longitudinal residual stress
(parallel to the welding direction) adjacent to the weld toe. Therefore, the transverse stress
(vertical with regard to the welding direction) in the flange plate was measured. In order
to identify the transformation temperatures of the LTT weld metal, a dilatometric test was
performed using Formaster equipment (FUJI, Tokyo, Japan). Generally, a higher heating
rate leads to higher austenite start and finish temperatures, Ac1 and Ac3 temperatures [29].
Considering the repeated thermal cycles and the different reheating rates in multi-pass
welds, a small heating rate of 10 ◦ C/s was adopted for the Formaster test. In this way, the
regions away from the newly-formed LTT weld pass can be taken into account due to the
smaller reheating rate and lower Ac1 and Ac3 temperatures. According to the empirical
equation proposed by Yurioka et al. [30], the critical cooling time TM from 800 to 500 ◦ C to
obtain a fully martensitic microstructure is given as:

ln TM = 10.6CE − 4.8 (8)

Si Mn Cu Ni Cr Mo
CE = C + + + + + + (9)
24 6 15 12 8 4
The value of TM for a LTT weld bead is about 37 s; in other words, a cooling rate
higher than 8.2 ◦ C/s is acceptable. Hence, a solid cylindrical sample extracted from a LTT
weld bead with a dimension of φ3.0 × 10 mm was heated from ambient temperature up to
1350 ◦ C at a heating rate of 10 ◦ C/s and held for 2 s, followed by continuous cooling to
ambient temperature at a cooling rate of 20 ◦ C/s.
 Materials 2021, 14, x FOR PEER REVIEW 4 of 16

Materials 2021, 14, 325 4 of 14

to 1350 °C at a heating rate of 10 °C/s and held for 2 s, followed by continuous cooling to
ambient temperature at a cooling rate of 20 °C/s.

.
Figure Schematic
1. 1.
Figure ofof
Schematic a multi-pass T welded
a multi-pass joint.
T welded joint.

Table Chemical
1. 1.
Table compositions
Chemical (%)(%)
compositions of of
thethe
welded plate,
welded thethe
plate, filler wires
filler andand
wires their weld
their metals.
weld metals.

Name
Name Condition
Condition CC SiSi Mn Mn NiNi
SM490A 0.200 0.55 1.60 /
SM490A 0.200 0.55 1.60 /
Wire 0.034 0.41 3.30 3.30
LB-3AD Wire 0.034 0.41 3.30 3.30
As welded 0.067 0.34 2.95 2.65
LB-3AD
As welded
Wire 0.067
0.080 0.34
0.55 2.95
0.79 2.65/
LB-47
As welded
Wire 0.093
0.080 0.49
0.55 0.64
0.79 / /
LB-47
As welded 0.093 0.49 0.64 /
Table 2. Parameters of the XRD device used for the residual stress measurement.

TableMeasuring device
2. Parameters μ-X360
of the XRD device usedFULL 2Dresidual
for the Portablestress
X-raymeasurement.
Residual Stress Analyzer
X-ray tube target Cr-Kα Diffractive surface Fe(α) (211)
Measuring device
X-ray tube voltage µ-X360 FULL
20 kV 2D Portable X-ray Residual
X-ray tube electricity Stress Analyzer
1.0 mA
X-ray tube target SingleCr-Kα
incident angle Diffractive surface Fe(α) (211)
Measuring method X-ray incident angle 45°
method
X-ray tube voltage 20 kV X-ray tube electricity 1.0 mA
Exposure time 30s Collimeter spot size Standard: Ф 1mm
Single incident angle
Measuring method X-ray incident angle 45◦
method
3. Results and Discussion
Exposure time 30 s Collimeter spot size Standard: Φ 1 mm
In order to compare the residual stress between the conventional and LTT welded
joints, as shown in Figure 2, the residual Sy stress was measured using the X-ray diffrac-
3. tion
Results and Discussion
technique in three positions of the flange plate. The measured residual Sy stress is
In order
plotted toright
at the compare theFigure
side of residual stress also
2, which between
shows thetheconventional
five moving and LTT welded
averages intended
joints, as shown
to minimize inexperimental
the Figure 2, the residual Sy stress was
error. Evidently, measured
the residual using
stress the X-ray diffraction
distribution is very sim-
technique
ilar amongin three
thesepositions of the flange
three positions, plate. The measured
demonstrating the highresidual
accuracy Syof
stress
the is plotted at
experimental
the right side of Figure
measurement. 2, whichthe
Furthermore, alsoconventional
shows the five movingjoint
welded averages
has aintended to minimize
lower tensile residual
the experimental error. Evidently, the residual stress distribution
stress of about 100 MPa at the weld toe compared to that of approximately 250 MPa in theis very similar among
these
LTTthree
weldedpositions, demonstrating
joint. With the distancethe highfrom
away accuracy
weld of the
toe, theexperimental
tensile residualmeasurement.
stress in the
Furthermore,
LTT welded the jointconventional welded joint
gradually decreases. has a lower
Accordingly, tensiletemperature
the reheat residual stress of about
is responsible
100forMPa at the weld
the higher toeresidual
tensile compared to that
stress of approximately
induced in the LTT welded 250 MPa in the
joint. LTT welded
In order to clarify
joint. With the distance
the difference away from
in the residual weld
stress toe, the tensile
distribution between residual
thesestress in the LTT
two welded welded
joints, a ther-
joint gradually decreases. Accordingly, the reheat temperature is
mal–elastic–plastic analysis was implemented using the in-house software, JWRIAN. responsible for the higher
tensile Figure
residual stress induced
3 shows in the LTT
the simulation welded
model joint.
for the In order toof
computation clarify the difference
the residual stress. in
The
the residual stress distribution between these two welded joints, a thermal–elastic–plastic
total numbers of nodes and elements are 41,648 and 36,646, respectively. A finer mesh of
analysis was implemented using the in-house software, JWRIAN.
Figure 3 shows the simulation model for the computation of the residual stress. The
total numbers of nodes and elements are 41,648 and 36,646, respectively. A finer mesh of
s 2021, 14, x FOR PEER REVIEW 5 of 16
Materials 2021, 14, 325 5 of 14

about 2 mm was adopted around the weld pass, while coarser meshes were utilized in
regions away from
aboutweld
2 mmzone,
waswith a maximum
adopted around size of 8 mm.
the weld pass,The boundary
while coarser condition
meshes were utilized in
was defined byregions
red arrows, which denoted the fixed displacement
away from weld zone, with a maximum size of 8 mm. in the corresponding
The boundary condition
direction. Six weld passes were
was defined by reddeposited at one denoted
arrows, which side of thetheT-welded joints, andineach
fixed displacement the corresponding
weld pass was cooled to an
direction. Sixambient temperature
weld passes of 15 °Catbefore
were deposited the of
one side start
theof the next joints,
T-welded weld and each weld
pass. The welding parameters
pass was cooled ofto
theanwelding
ambientsimulation
temperature were the◦same
of 15 as the
C before theexperiment.
start of the next weld pass.
In addition, theThe
conventional von Mises’ elastoplastic flow theory was
welding parameters of the welding simulation were the same taken into account:
as the experiment. In
addition, the conventional e vonp Mises’ th elastoplastic flow theory was taken into account:
dε =dε + dε + dε + dε tr
(10)
dε = dεe + dεp + dεth + dεtr (10)
where d ε , d ε , d ε and d ε represent the increments of elastic strain, plastic strain,
e p th tr

thermal strain and

wheretransformation plastic
dεe , dεp , dεth and dεtr strain,
representrespectively. Furthermore,
the increments the isotropic
of elastic strain, plastic strain, thermal
hardening model wasand
strain used for the computation
transformation of the residual
plastic strain, stress.
respectively. Furthermore, the isotropic hardening
Figure 4 shows
modelthe wasthermal
used forand mechanical
the computation properties of the LTT
of the residual weld metals. The
stress.
thermal properties were
Figure 4 shows the thermal and mechanical properties of[31,32].
obtained from other published papers as references the LTTEx-
weld metals. The
cept for the mechanical
thermal properties were at room temperature,
obtained the other
from other temperature-dependent
published papers as references [31,32]. Except
mechanical properties of LB-47 and
for the mechanical LB-3AD at
properties were
room computed usingthe
temperature, thermodynamic soft-
other temperature-dependent me-
ware, JMatPro.chanical
The mechanical
propertiesproperties
of LB-47 and of SM490A
LB-3AD werewerecomputed
experimentally measured
using thermodynamic software,
based on the JISJMatPro.
G3106 Standard. In addition,
The mechanical Figureof5SM490A
properties displayswere
the transient coefficients
experimentally measured based on
of thermal expansion
the JIS (CTE)
G3106of the welded
Standard. components,
In addition, Figureas5measured
displays theby transient
the dilatometric
coefficients of thermal
test. The CTE isexpansion (CTE) of the welded components, as measured by the dilatometric test. The CTE
given by [33]:
is given by [33]:
CTE = d ε / ΔT (11)
CTE = dε/∆T (11)
dε =ΔL / L0 dε = ∆L/L0 (12) (12)
where L0 is thewhere
sample
L0 length at roomlength
is the sample temperature
at roomand ΔL is theand
temperature ∆L is the temperature-dependent
temperature-depend-
length
ent length change. change.

Figure 2. MeasuredFigure 2. Measured

residual residual
Sy stress of the twoSywelded
stress of the two welded joints.
joints.
 Materials 2021, 14, 325 6 of 14
als 2021, 14, xMaterials
FOR PEER REVIEW
2021, 14, x FOR PEER REVIEW 6 of 16 6 of 16
als 2021, 14, x FOR PEER REVIEW 6 of 16

Figure 3. Simulation model

Figure for the computation of the residual stress.
Figure
Figure 3. Simulation 3.3.Simulation
model Simulation modelfor
model forthe
for the computation the computation
of computation
the ofofthe
theresidual
residual stress. residualstress.
stress.

Figure 4. Material properties of the properties

LTT weld metal: (a) thermal properties; (b) mechanical proper-
Figure 4. MaterialFigure 4. Material
properties of the LTT weld of the LTT
metal: (a) weld metal:
thermal (a) thermal
properties; (b) properties;proper-
mechanical (b) mechanical proper-
ties. 4. Material properties
Figure ties. of the LTT weld metal: (a) thermal properties; (b) mechanical properties.
ties.

Figure
Figure 5. CTE 5.welded
of the5. CTE of Figure
the welded
5. CTEcomponents
of thethe during
welded the heating
components and cooling
during processes:
the heating (a) heating
andprocess;
cooling processes:
cooling (a) heating
Figure CTE ofcomponents
the welded during
components heating
duringand
thecooling
heatingprocesses: (a) processes:
and cooling heating (b)
(a) heating process.
process; (b) cooling process.
process; (b) cooling process.
process; (b) cooling process.
3.1. Reheat Temperature of LTT Weld Passes
3.1. Reheat Temperature
3.1.In of LTT
Reheat Weld Passes
Temperature of LTT
3.1. Reheat Temperature multi-pass
of LTT welding,
Weld Passes theWeld Passes
formerly-formed weld beads undergo different thermal
In multi-pass
cycles.welding,
InAs the
multi-pass formerly-formed
mentioned before,
welding, the
the weld
reheat beads undergo
temperature
formerly-formed different
plays
weld thermal
an indispensable
beads undergo role thermal
different in the
In multi-pass welding, the formerly-formed weld beads undergo different thermal
cycles. As mentioned
residual
cycles. before,
stress
As the
mentionedreheat
development temperature
before,inthe
a LTT plays
welded
reheat an indispensable
joint. Hence,
temperature plays role in the
it isindispensable
an critical to evaluate
role inthe
the
cycles. As mentioned before, the reheat temperature plays an indispensable role in the
reheat temperature distribution induced by these thermal cycles in the formerly-formed
 Materials 2021, 14, x FOR PEER REVIEW 7 of 16

Materials 2021, 14, 325 7 of 14

residual stress development in a LTT welded joint. Hence, it is critical to evaluate the re-
heat temperature distribution induced by these thermal cycles in the formerly-formed
weldbeads.
weld beads.Additionally,
Additionally,the thereheat
reheattemperature
temperaturehas hasaastrong
strongrelationship
relationshipwith withthe
theAc3Ac3
temperature.According
temperature. According to to
thethe dilatometric
dilatometric results
results of of
thetheLTT LTT weld
weld metal
metal shown
shown in Figure
in Figure 6,
6, the
the Ac3Ac3 temperature
temperature during during the heating
the heating process process
is about 780 ◦ C.780
is about For°C. For regions
regions with a
with a reheat
reheat temperature
temperature less thanless ◦ C, martensite
780than 780 °C, martensite cannot
cannot fully fully transform
transform into austenite,
into austenite, resulting re-
insulting in an increase
an increase in the thermally-induced
in the thermally-induced tensile residual
tensile residual stress.
stress. In other Inwords,
other words,
the CTE the
ofCTE
thisofregion
this region
duringduring the cooling
the cooling processprocess
is the is the same
same as that asof
that
theofheating
the heating process.
process. If theIf
reheat temperature
the reheat temperature exceeds 780 ◦780
exceeds C, martensite
°C, martensitewill will
fullyfully
transform
transformintointo
austenite during
austenite dur-
the
ingheating process,
the heating and then
process, the austenite
and then transforms
the austenite transformsinto martensite
into martensiteagainagain
during the
during
cooling process,
the cooling thus producing
process, thus producing the beneficial compressive
the beneficial residualresidual
compressive stress. Figure
stress.7Figure
shows 7
the temperature
shows distribution
the temperature of the weld
distribution of thepasses after each
weld passes afterwelding.
each welding. According to theto
According
temperature distribution and Ac3 temperature, the regions with
the temperature distribution and Ac3 temperature, the regions with a temperature higher a temperature higher
than ◦ C can be defined as the ‘real’ LTT weld zone, which can produce beneficial
than780
780 °C be defined as the ‘real’ LTT weld zone, which can produce beneficial com-
compressive residual
pressive residual stress
stress afterafter
each each welding.
welding. Conversely,
Conversely, thethe regions
regions with with temperatures
temperatures less
less than ◦ C are defined as the ‘non-real’ LTT weld zone. Correspondingly, the CTEs
than 780780
°C are defined as the ‘non-real’ LTT weld zone. Correspondingly, the CTEs of
ofthe
the ‘non-real’
‘non-real’ and
and ‘real’
‘real’ LTT LTTweldweld metals
metals areare redefined
redefined againagain for residual
for the the residual
stressstress
com-
computation, as shown
putation, as shown in Figure 7. in Figure 7.

Figure6.6.Dilatometric
Figure Dilatometricresults
resultsofofthe
theLTT
LTTweld
weldmetal.
metal.

3.2. Comparison of the Residual Stresses between the Welded Joints

Figure 8 shows the longitudinal residual Sx stress distribution in the mid-section of the
conventional and LTT welded joints. It can be seen in the conventional welded joint that
the whole weld zone is accompanied by a relatively higher tensile residual stress. For the
LTT welded joint, compressive residual stress is mainly produced in the newly-formed LTT
weld bead and the weld zone with a reheat temperature higher than the Ac3 temperature,
balanced by the tensile stress adjacent to this region. Once the reheat temperature is up
to the Ac3 temperature, the microstructure of the formerly-formed weld zone transforms
into austenite again in the heating process, and then the martensitic transformation occurs
during the cooling process, thus producing compressive residual stress. Owing to welding
sequence, the compressive residual stress of the weld passes formed first converts into
tensile stress if the reheat temperature is less than the Ac3 temperature. Therefore, greater
tensile residual stress is generated at the weld toe of weld pass 4, while lower stress is
found at the weld toe of weld pass 6. Furthermore, the residual stress near the tack weld in
both welded joints increases with the number of weld passes, and the conventional welded
joint has a larger tensile stress.
 Materials 2021, 14, 325 8 of 14
terials 2021, 14, x FOR PEER REVIEW 8 of 16

Figure 7. Temperature
Temperature
Figure 7.Materials distribution
distribution during during the
the welding welding
process forprocess for the redefinition
the redefinition of the CTEof of
theLTT
CTEbead:
of (a) first pass;
2021, 14, x FOR PEER REVIEW 9 of 16
LTT bead: (a) first pass; (b) second pass; (c) third pass;
(b) second pass; (c) third pass; (d) fourth pass; (e) fifth pass; (f) sixth pass.(d) fourth pass; (e) fifth pass; (f) sixth pass.

3.2. Comparison of the Residual Stresses between the Welded Joints

Figure 8 shows the longitudinal residual Sx stress distribution in the mid-section of
the conventional and LTT welded joints. It can be seen in the conventional welded joint
that the whole weld zone is accompanied by a relatively higher tensile residual stress. For
the LTT welded joint, compressive residual stress is mainly produced in the newly-
formed LTT weld bead and the weld zone with a reheat temperature higher than the Ac3
temperature, balanced by the tensile stress adjacent to this region. Once the reheat tem-
perature is up to the Ac3 temperature, the microstructure of the formerly-formed weld
zone transforms into austenite again in the heating process, and then the martensitic trans-
formation occurs during the cooling process, thus producing compressive residual stress.
Owing to welding sequence, the compressive residual stress of the weld passes formed
first converts into tensile stress if the reheat temperature is less than the Ac3 temperature.
Therefore, greater tensile residual stress is generated at the weld toe of weld pass 4, while
lower stress is found at the weld toe of weld pass 6. Furthermore, the residual stress near
the tack weld in both welded joints increases with the number of weld passes, and the
conventional welded joint has a larger tensile stress.

Figure 8. Residual Sx stress distribution in the mid-section of the conventional and LTT -welded joints: (a,g) first pass;
Figure 8. Residual Sx stress
(b,h) second distribution
pass; (c,i) infourth
third pass; (d,j) the mid-section
pass; (e,k) fifthof the(f,l)
pass; conventional
sixth pass. and LTT -welded joints: (a,g) first pass;
(b,h) second pass; (c,i) third pass; (d,j) fourth pass; (e,k) fifth pass; (f,l) sixth pass.
Figure 9 shows the residual Sy stress distribution in the mid-section of the conven-
tional and LTT welded joints. Compared to the longitudinal residual Sx stress, both
welded joints have lower tensile residual Sy stresses. In the conventional welded joint,
most of the weld zone is left with a greater tensile stress. For the LTT welded joint, similar
to the longitudinal residual Sx stress distribution, the compressive residual Sy stress is
mainly produced in the newly-formed LTT weld zone with a reheat temperature higher
than the Ac3 temperature, balanced by the tensile stress near this region. Due to the weld-
ing sequence, the residual stress state of those weld passes formed first converts from
Materials 2021, 14, 325 9 of 14

Figure 9 shows the residual Sy stress distribution in the mid-section of the conventional
and LTT welded joints. Compared to the longitudinal residual Sx stress, both welded joints
have lower tensile residual Sy stresses. In the conventional welded joint, most of the weld
zone is left with a greater tensile stress. For the LTT welded joint, similar to the longitudinal
residual Sx stress distribution, the compressive residual Sy stress is mainly produced in the
newly-formed LTT weld zone with a reheat temperature higher than the Ac3 temperature,
balanced by the tensile stress near this region. Due to the welding sequence, the residual
stress state of those weld passes formed first converts from compressive to tensile because
the reheat temperature is lower than the Ac3 temperature. Hence, greater tensile residual
stress is also induced at the weld toe of weld pass 4, while compressive stress is produced at
the weld toe of weld pass 6. Moreover, the residual stress near the tack weld in both welded
Materials 2021, 14, x FOR PEER REVIEW
joints builds up gradually during the welding process, and the conventional welded 10 of 16 joint

also has a greater tensile stress.

Figure 9. Residual Sy stress distribution in the mid-section of the conventional and LTT welded joints: (a) & (g) first
9. Residual
Figurepass; Sy stress
(b) & (h) second distribution
pass; inpass;
(c) & (i) third the mid-section
(d) & (j) fourthofpass;
the (e)
conventional and(f)LTT
& (k) fifth pass; & (l)welded joints: (a,g) first pass;
sixth pass.
(b,h) second pass; (c,i) third pass; (d,j) fourth pass; (e,k) fifth pass; (f,l) sixth pass.
Figure 10 shows the residual Sz stress distribution in the mid-section of the conven-
Figure
tional and10 LTTshows
welded thejoints.
residual Sz stress
Compared distribution
to the residual Sxin theSymid-section
and stresses, bothofwelded
the conven-
tional and
joints haveLTT welded
lower joints.
tensile residualCompared
Sz stresses.toMeanwhile,
the residual Sx and
residual SzSy stresses,
stress both welded
in the conven-
joints have
tional lower
welded tensile
joint residualmore
is distributed Sz stresses.
uniformly,Meanwhile,
as it was againstresidual
the LTTSzwelded
stress in the con-
joint.
Similarly,
ventional the compressive
welded residual stress
joint is distributed more is mainly produced
uniformly, as it in theagainst
was newly-formed
the LTT LTT
welded
weld
joint. zone with
Similarly, thea reheat temperature
compressive higher
residual thanisthe
stress Ac3 temperature,
mainly produced in balanced by the
the newly-formed
LTT tensile
weldstress
zoneclose
withtoathis region.
reheat The weldinghigher
temperature sequence
thanhasthe
theAc3sametemperature,
effect on the residual
balanced by
the tensile stress close to this region. The welding sequence has theatsame
stress state of the weld passes, leading to a greater tensile residual stress the weld toeon the
effect
of weld pass 4 and compressive stress at the weld toe of weld pass 6. Additionally, the
residual stress state of the weld passes, leading to a greater tensile residual stress at the
conventional welded joint has a greater tensile residual stress near the tack weld in com-
parison with the LTT welded joint. This can be attributed to the greater tensile residual
stress induced in the conventional weld bead, which results in a greater distortion of the
web plate.
Materials 2021, 14, 325 10 of 14

weld toe of weld pass 4 and compressive stress at the weld toe of weld pass 6. Additionally,
the conventional welded joint has a greater tensile residual stress near the tack weld in
comparison with the LTT welded joint. This can be attributed to the greater tensile residual
Materials 2021, 14, x FOR PEER REVIEW 11 of 16 of the
stress induced in the conventional weld bead, which results in a greater distortion
web plate.

Figure 10. Residual Sz stress distribution in the mid-section of the conventional and LTT welded joints: (a,g) first pass;
Figure(b,h) Residual
10. second Sz stress distribution in the mid-section of the conventional and LTT welded joints: (a,g) first pass;
pass; (c,i) third pass; (d,j) fourth pass; (e,k) fifth pass; (f,l) sixth pass.
(b,h) second pass; (c,i) third pass; (d,j) fourth pass; (e,k) fifth pass; (f,l) sixth pass.
Figure 11 compares the computed residual Sy stress with the measured stress in or-
derFigure 11 compares
to verify the computed
the computation results. Itresidual
can be seenSy stress
that the with the measured
computed residualstress
stressesin order
to are
verify
in good agreement with the experimental measurement adjacent to the weld toe. With are
the computation results. It can be seen that the computed residual stresses
in good agreement
a distance from weld with the experimental
toe longer than 8 mm, theremeasurement
exists an error adjacent
betweentothe the weld toe. With a
measurement
distance
and thefrom weld toe This
computation. longercanthan 8 mm, there
be attributed to theexists anstress
initial errorofbetween
the weldedthe plate
measurement
in-
and duced before the welding,
the computation. which
This can be is highly dependent
attributed on the
to the initial manufacturing
stress of the weldedprocesses of
plate induced
the welded plate, such as bending or rolling [34,35]. In order
before the welding, which is highly dependent on the manufacturing processes of the to quantitatively compare
the residual
welded plate,stress
such between theseor
as bending two welded
rolling joints, Figure
[34,35]. In order 12 shows the simulatedcompare
to quantitatively resid- the
ual stress distribution starting from the weld root along the two red lines on the flange
residual stress between these two welded joints, Figure 12 shows the simulated residual
and web plates. Evidently, except for the residual Sz stress, the residual Sx and Sy stresses
stress distribution starting from the weld root along the two red lines on the flange and
at the weld root of LTT welded joint are greater than those of the conventional welded
web plates. Evidently, except for the residual Sz stress, the residual Sx and Sy stresses at
joint. Due to the effect of the reheat temperature, as depicted in Figure 12a, the weld toe
theofweld
weldroot
passof LTT
4 in thewelded
LTT weldedjointjoint
are has
greater than
greater those
tensile of theSx
residual conventional
and Sy stresseswelded
than joint.
Due to the
those effect
of the of the reheat
conventional temperature,
welded as depicted
joint. Conversely, as shownin Figure 12a,12b,
in Figure thethe
weld
weldtoetoeof weld
pass 4 in the
of weld passLTT
6 inwelded
the LTT joint
welded has greater
joint tensile residual
has compressive residualSx SxandandSySzstresses
stressesthan
of −186those of
theMPaconventional
and −20 MPa, welded joint. Conversely,
respectively. On the other as shown
hand, in Figure
the residual Sx 12b,
and Szthestresses
weld toe of weld
at the
passweld6 intoe
the
of LTT
weldwelded
pass 6 injoint has compressive
the conventional weldedresidual
joint are Sx433andMPaSz stresses
and 6 MPa,of −186 MPa
respec-
tively.
Materials 2021, 14, 325 11 of 14

Materials 2021, 14, x FOR PEER REVIEW 12 of 16

and
Materials 2021, 14, x FOR PEER REVIEW −20 MPa, respectively. On the other hand, the residual Sx and Sz stresses at the12weld
of 16
toe of weld pass 6 in the conventional welded joint are 433 MPa and 6 MPa, respectively.

Figure 11.Computed
Computed residualSySy stressversus
versus theexperimental
experimental result.
Figure 11. Computed residual
Figure 11. residual Sy stress
stress versus the
the experimental result.
result.

Figure 12. Comparison of the simulated residual stresses along the red line: (a) on the flange plate; (b) on the web plate.
Figure 12.
Figure 12. Comparison
Comparison of
of the
the simulated
simulated residual
residual stresses
stresses along
along the
the red
redline:
line: (a)
(a) on
on the
the flange
flangeplate;
plate;(b)
(b)on
onthe
theweb
webplate.
plate.

Apart from the weldtoe, toe, theweld weld rootisisalso also a criticalpositionposition thatisissusceptible
susceptibleto to
Apart from the weld weld toe, the the weldroot root is alsoa acritical
critical positionthat that is susceptible
fracture
fracture failure.
failure. Hence,
Hence, it is necessary
it isitnecessary to evaluate
to evaluate the
thethe residual
residual stress
stress development
development at the
at the
to fracture failure. Hence, is necessary to evaluate residual stress development at
weld root.
weld root. Figure
Figure 13 13 displays
displays the the simulated
simulated residualresidual stresses at at the weldweld roots of both both
the weld root. Figure 13 displays the simulated residualstresses stresses atthe the weld roots of both
weldedjoints
welded jointsafter
after each welding.Obviously, Obviously,the theresidual
residualstresses
stressesvaryvary after each welding,
welded joints after each welding. Obviously, the residual stresses vary after each welding,
and the residual
residual Sx
and the residual Sx stress
Sx stress
stress is is greater
is greater
greater than than
than the the other
the other
other two two stress
two stress components
stress components
components in inboth
in bothof
both ofthe
of the
the
weldedjoints.
welded
welded joints.For For
Forthe the conventional
theconventional
conventionalwelded welded
welded joint,
joint,
joint, the
the the residual
residual
residual SxSx
Sx stress
stress
stress decreases
decreases
decreases from
from from500
500
MPa
500
MPaMPa after
afterafterthe
the thefirst
firstfirstwelding
welding
welding to
to 100100
to 100 MPa
MPa MPa after
after the
after
thethe sixth
sixthsixth welding.
welding.
welding. Instead,
Instead,
Instead, the
thethe residual
residual
residual Sx
Sx
stress
Sx stress
stress in
in in the
thethe LTT
LTT LTT welded
weldedjoint
welded joint
jointhashas the
hasthe lowest
thelowest residual
lowestresidual
residualstress stress
stressof of 190
of 190 MPa after
190 MPa after the firstthe first
first
welding,which
welding,
welding, whichthen
which thenincreases
then increasesup
increases upto
up to260
to 260MPa
260 MPaafter
MPa afterthe
after thesixth
the sixthwelding.
sixth welding.On
welding. Onthetheother
otherhand,
hand,
bothof
both
both ofthe
of theresidual
the residualSy
residual Syand
Sy andSz
and Szstresses
Sz stresses
stresses ininin
twotwo
two welded
welded
welded joints
joints
joints decrease
decrease
decrease afterafter the
thethe
after sixthsixth weld-
welding
sixth weld-
ing compared
ing compared
compared to that
toto after
that after
that afterfirst
the thewelding.
the first welding.
first welding. Additionally,
Additionally,
Additionally, the LTT the
the LTT welded
welded
LTT welded
joint has joint has
thehas
joint lowerthe
the
lower
residual residual
Sy andSy
lower residual Sy and
Szand Sz
stresses stresses
after the
Sz stresses after
afterfirstthe first
thewelding, welding,
but thebut
first welding, but the
residual residual
Sy stress
the residual Sy stress
after the
Sy stress after the
sixth
after the
sixth welding
welding
sixth welding isis relatively
is relatively relatively higher
higher higher
than that than
thanof thethatconventional
that of the
of the conventional
conventional
welded joint,welded
welded whilejoint,
joint, while
thewhile
residual the
the
residual
Sz stress Sz
residual Sz stress
is similar is similar
stress istosimilar
that ofto to
the that of the
conventional
that conventional
welded joint.
of the conventional welded
welded joint.joint.
Based on the above studies, it can
Based on the above studies, it can be concluded that the be concluded that thereheat
reheattemperature
temperaturedefinitely
definitely
playsaacritical
plays criticalrolerolein inthe
theresidual
residualstress stressdevelopment
developmentin inaaLTT
LTTwelded
weldedjoint.joint.In Inaddition,
addition,
LTT welding materials are more suitable for single-pass
LTT welding materials are more suitable for single-pass welding for tensile residual stress welding for tensile residual stress
reduction around the weld zone. According to the results
reduction around the weld zone. According to the results of Shiga et al. [36,37], the single- of Shiga et al. [36,37], the single-
pass LTT
pass LTT elongated-bead
elongated-bead method method can can greatly
greatly extend
extend the the fatigue
fatigue lifelife of
of aa corner
corner boxing
boxing
 Materials 2021, 14, x FOR PEER REVIEW 13 of 15

Materials 2021, 14, 325 12 of 14

fillet-welded joint by more than four times without deteriorating the fracture toughness.
Hence, there is still a great potential for the engineering application of LTT welding ma-
terials.

Figure 13. Simulated residual stress at the weld roots of the conventional and LTT welded joints after each welding: (a)
Figure
residual Sx stress13. Simulated
in conventional residual
welded stress
joint; (b) at the
residual weld
Sx stress roots
in LTT of the
welded joint;conventional andinLTT
(c) residual Sy stress con- welded joints after
ventional welded
each joint; (d) (a)
welding: residual Sy stress
residual Sxinstress
LTT welded joint; (e) residualwelded
in conventional Sz stress in conventional
joint; welded joint;
(b) residual (f)
Sx stress in LTT welded
residual Sz stress in LTT welded joint.
joint; (c) residual Sy stress in conventional welded joint; (d) residual Sy stress in LTT welded joint;
4. Conclusions
(e) residual Sz stress in conventional welded joint; (f) residual Sz stress in LTT welded joint.
Experimental and numerical analyses were performed in order to investigate the re-
Based onsidual stress distribution
the above studies,initmulti-pass conventional and
can be concluded lowthe
that transformation temperature definitely
reheat temperature
(LTT) T-welded joints. It was found that the reheat temperature plays a significant role in
plays a critical role in the residual stress development in a LTT welded joint. In addition,
LTT welding materials are more suitable for single-pass welding for tensile residual stress
reduction around the weld zone. According to the results of Shiga et al. [36,37], the
single-pass LTT elongated-bead method can greatly extend the fatigue life of a corner
boxing fillet-welded joint by more than four times without deteriorating the fracture
toughness. Hence, there is still a great potential for the engineering application of LTT
welding materials.

4. Conclusions
Experimental and numerical analyses were performed in order to investigate the
residual stress distribution in multi-pass conventional and low transformation temperature
(LTT) T-welded joints. It was found that the reheat temperature plays a significant role in
the residual stress development of a LTT welded joint. Based on this study, the following
conclusions were drawn:
(1) LTT welding materials are more suitable for single-pass welding for the generation of
compressive residual stress.
(2) The formerly-formed LTT weld passes with reheat temperatures of up to the austenite
finish temperature can produce beneficial compressive residual stress due to marten-
sitic transformation during the cooling process.
Materials 2021, 14, 325 13 of 14

(3) The compressive residual stress of the formerly-formed LTT weld bead is converted
into tensile stress if the reheat temperature is less than the austenite finish temperature.

Author Contributions: Conceptualization, Z.F., N.M., S.T., and F.L.; methodology, Z.F., N.M., S.T.,
and F.L.; software, Z.F. and N.M.; validation, N.M., S.T., and F.L.; formal analysis, Z.F., N.M., S.T., and
F.L.; investigation, Z.F.; resources, N.M., S.T., and F.L.; data curation, N.M., S.T., and F.L.; writing—
original draft preparation, Z.F.; writing—review and editing, Z.F., N.M., S.T., and F.L.; visualization,
Z.F.; supervision, N.M., S.T., and F.L.; project administration, N.M., S.T., and F.L.; funding acquisition,
N.M. and F.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by “Creation of Life Innovation Materials for Interdisciplinary
and International Researcher Development” and “Project to Create Research and Educational Hubs
for Innovative Manufacturing in Asia”.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Xiong, Y.; Li, C.; Chen, Z.; He, J.; Xin, H. The Evolution of Residual Stress in Rib-Diaphragm Joints of Orthotropic Steel Decks
Subjected to Thermal Cutting and Welding. Materials 2020, 13, 3804. [CrossRef]
2. Shahani, A.R.; Shakeri, I.; Rans, C.D. Effect of residual stress redistribution and weld reinforcement geometry on fatigue crack
growth of butt welded joints. Int. J. Fatigue 2020, 139, 105780. [CrossRef]
3. Javadi, Y.; Sweeney, N.E.; Mohseni, E.; MacLeod, C.N.; Lines, D.; Vasilev, M.; Qiu, Z.; Mineo, C.; Pierce, S.G.; Gachagan, A.
Investigating the effect of residual stress on hydrogen cracking in multi-pass robotic welding through process compatible
non-destructive testing. J. Manuf. Process. 2020. [CrossRef]
4. Kessal, B.; Fares, C.; Meliani, M.H.; Alhussein, A.; Bouledroua, O.; François, M. Effect of gas tungsten arc welding parameters on
the corrosion resistance and the residual stress of heat affected zone. Eng. Fail. Anal. 2020, 107, 104200. [CrossRef]
5. Zha, X.-Q.; Xiong, Y.; Zhou, T.; Ren, Y.-F.; Hei, P.-H.; Zhai, Z.-L.; Kömi, J.; Huttula, M.; Cao, W. Impacts of Stress Relief Treatments
on Microstructure, Mechanical and Corrosion Properties of Metal Active-Gas Welding Joint of 2205 Duplex Stainless Steel.
Materials 2020, 13, 4272. [CrossRef] [PubMed]
6. Ferro, P.; Berto, F.; Bonollo, F.; Montanari, R. Numerical modelling of residual stress redistribution induced by TIG-dressing.
Frattura ed Integrità Strutturale 2018, 13, 221–230. [CrossRef]
7. Ren, S.; Li, S.; Wang, Y.; Deng, D.; Ma, N. Finite element analysis of residual stress in 2.25Cr-1Mo steel pipe during welding and
heat treatment process. J. Manuf. Process. 2019, 47, 110–118. [CrossRef]
8. Jones, W.K.C.; Alberry, P.J. A model for stress accumulation in steels during welding. Met. Technol. 1977, 11, 557–566.
9. Murata, H.; Kato, N.; Tamura, H. Effect of Transformation on Residual Stress in Welding. Stress Releasement by Transformation
Superplasticity(Part 5). Q. J. Jpn. Weld. Soc. 1993, 11, 545–550. [CrossRef]
10. Ooi, S.W.; Garnham, J.; Ramjaun, T. Review: Low transformation temperature weld filler for tensile residual stress reduction.
Mater. Des. 2014, 56, 773–781. [CrossRef]
11. Kromm, A.; Dixneit, J.; Kannengiesser, T. Residual stress engineering by low transformation temperature alloys—state of the art
and recent developments. Weld. World 2014, 58, 729–741. [CrossRef]
12. Hanji, T.; Tateishi, K.; Kano, S.; Shimizu, M.; Tsuyama, T.; Takebuchi, T. Fatigue strength of transverse attachment steel joints with
single-sided arc weld using low transformation temperature welding consumable. Weld. World 2020, 64, 1293–1301. [CrossRef]
13. Harati, E.; Karlsson, L.; Svensson, L.-E.; Dalaei, K. Applicability of low transformation temperature welding consumables to
increase fatigue strength of welded high strength steels. Int. J. Fatigue 2017, 97, 39–47. [CrossRef]
14. Wu, S.; Wang, D.-P.; Zhang, Z.; Li, C.; Liu, X.; Zhou, Z.; Di, X. Effect of dilution on fatigue behaviour of welded joints produced by
low-transformation-temperature fillers. Sci. Technol. Weld. Join. 2019, 24, 601–608. [CrossRef]
15. Xu, L.Y.; Wang, Y.F.; Jing, H.Y.; Han, Y.D. Fatigue strength improvement of stainless steel using weld toes dressing with low
transformation temperature welding wire. Sci. Technol. Weld. Join. 2014, 19, 664–672. [CrossRef]
16. Mikami, Y.; Morikage, Y.; Mochizuki, M.; Toyoda, M. Angular distortion of fillet welded T joint using low transformation
temperature welding wire. Sci. Technol. Weld. Join. 2009, 14, 97–105. [CrossRef]
17. Zenitani, S.; Hayakawa, N.; Yamamoto, J.; Hiraoka, K.; Morikage, Y.; Kubo, T.; Yasuda, K.; Amano, K.; Iordachescu, M.;
Iordachescu, D.; et al. Development of new low transformation temperature welding consumable to prevent cold cracking in
high strength steel welds. Sci. Technol. Weld. Join. 2007, 12, 516–522. [CrossRef]
18. Feng, Z.; Aung, T.L.; Shao, C.; Lu, F.; Tsutsumi, S.; Ma, N. A design method of tensile triangles and low transformation temperature
weld metal for reduction of stress concentration and residual stress of welded joints. Mar. Struct. 2020, 72, 102759. [CrossRef]
Materials 2021, 14, 325 14 of 14

19. Feng, Z.-Y.; Di, X.; Wu, S.; Wang, D.-P.; Liu, X.-Q. Comparison of Microstructure and Residual Stress Between TIG and MAG
Welding Using Low Transformation Temperature Welding Filler. Acta Met. Sin. (English Lett.) 2017, 31, 263–272. [CrossRef]
20. Feng, Z.-Y.; Di, X.; Wu, S.-P.; Zhang, Z.; Liu, X.-Q.; Wang, D.-P. Comparison of two types of low-transformation-temperature weld
metals based on solidification mode. Sci. Technol. Weld. Join. 2017, 23, 241–248. [CrossRef]
21. Chen, X.; Fang, Y.; Li, P.; Yu, Z.; Wu, X.; Li, N. Microstructure, residual stress and mechanical properties of a high strength steel
weld using low transformation temperature welding wires. Mater. Des. 2015, 65, 1214–1221. [CrossRef]
22. Feng, Z.; Di, X.; Wu, S.; Ma, N. Transformation temperatures, mechanical properties and residual stress of two low-transformation-
temperature weld metals. Sci. Technol. Weld. Join. 2020, 1–9. [CrossRef]
23. Wang, W.; Huo, L.; Zhang, Y.; Wang, D.; Jing, H. New Developed Welding Electrode for Improving the Fatigue Strength of
Welded Joints. J. Mater. Sci. Technol. 2002, 18, 527–531.
24. Ramjaun, T.I.; Stone, H.; Karlsson, L.; Kelleher, J.; Moat, R.J.; Kornmeier, J.R.; Dalaei, K.; Bhadeshia, H.K.D.H. Effect of interpass
temperature on residual stresses in multipass welds produced using low transformation temperature filler alloy. Sci. Technol.
Weld. Join. 2014, 19, 44–51. [CrossRef]
25. Moat, R.J.; Ooi, S.; Laik, A.; Dai, H.; Mark, A.; Bhadeshia, H.K.D.H.; Withers, P.J. Residual stress control of multipass welds using
low transformation temperature fillers. Mater. Sci. Technol. 2017, 34, 519–528. [CrossRef]
26. Feng, Z.; Ma, N.; Li, W.; Narasaki, K.; Lu, F. Efficient analysis of welding thermal conduction using the Newton–Raphson method,
implicit method, and their combination. Int. J. Adv. Manuf. Technol. 2020, 111, 1929–1940. [CrossRef]
27. Ma, N. An accelerated explicit method with GPU parallel computing for thermal stress and welding deformation of large
structure models. Int. J. Adv. Manuf. Technol. 2016, 87, 2195–2211. [CrossRef]
28. Lin, J.; Ma, N.; Lei, Y.; Murakawa, H. Measurement of residual stress in arc welded lap joints by cosα X-ray diffraction method. J.
Mater. Process. Technol. 2017, 243, 387–394. [CrossRef]
29. Cai, J.; Lin, J.; Wilsius, J. Modelling phase transformations in hot stamping and cold die quenching of steels. In Microstructure
Evolution in Metal Forming Processes; Elsevier: Amsterdam, The Netherlands, 2012; pp. 210–236.
30. Yurioka, N.; Okumura, M.; Kasuya, T.; Cotton, H.U. Prediction of HAZ hardness of transformable steels. Met. Constr. 1987, 19,
217R–223R.
31. Jahromi, S.J.; Khajeh, A.; Mahmoudi, B. Effect of different pre-heat treatment processes on the hardness of AISI 410 martensitic
stainless steels surface-treated using pulsed neodymium-doped yttrium aluminum garnet laser. Mater. Des. 2012, 34, 857–862.
[CrossRef]
32. Deng, D. FEM prediction of welding residual stress and distortion in carbon steel considering phase transformation effects. Mater.
Des. 2009, 30, 359–366. [CrossRef]
33. Liu, Y.; Sommer, F.; Mittemeijer, E. Calibration of the differential dilatometric measurement signal upon heating and cooling;
thermal expansion of pure iron. Thermochim. Acta 2004, 413, 215–225. [CrossRef]
34. Park, J.-U.; An, G.B.; Woo, W.; Choi, J.-H.; Ma, N. Comparison of Measured Residual Stress Distributions in Extra-Thick Butt
Welds Joined by One-Pass EGW and Multipass FCAW. Adv. Mech. Eng. 2014, 6, 861247. [CrossRef]
35. Park, J.-U.; An, G.; Woo, W.C.; Choi, J.-H.; Ma, N. Residual stress measurement in an extra thick multi-pass weld using initial
stress integrated inherent strain method. Mar. Struct. 2014, 39, 424–437. [CrossRef]
36. Shiga, C.; Murakawa, H.; Hiraoka, K.; Osawa, N.; Yajima, H.; Tanino, T.; Tsutsumi, S.; Fukui, T.; Sawato, H.; Kamita, K.;
et al. Elongated bead weld method for improvement of fatigue properties in welded joints of ship hull structures using low
transformation temperature welding materials. Weld. World 2017, 61, 769–788. [CrossRef]
37. Shiga, C.; Murakawa, E.; Matsuo, Y.; Ohsuga, U.; Hiraoka, K.; Morikage, Y.; Yasuda, K. Fatigue improvement in high-strength
steel welded joints with compressive residual stress. Weld. World 2014, 58, 55–64. [CrossRef]