RESID. STRESS - Experiments and Numerical Simulations of The Annealing Temperature Influence in S700MC

materials
Article
Experiments and Numerical Simulations of the
Annealing Temperature Influence on the Residual
Stresses Level in S700MC Steel Welded Elements
Tomasz Kik 1, * , Jaromír Moravec 2 and Martin Švec 2
1 Department of Welding Engineering, Silesian University of Technology, Konarskiego 18A,
44-100 Gliwice, Poland
2 Department of Engineering Technology, Technical University of Liberec, 461 17 Liberec, Czech Republic;
jaromir.moravec@tul.cz (J.M.); martin.svec@tul.cz (M.Š.)
* Correspondence: tomasz.kik@polsl.pl; Tel.: +48-32-237-1681

Received: 4 November 2020; Accepted: 20 November 2020; Published: 22 November 2020
Abstract: The article presents the results of research on the influence of temperature and time changes
of the annealing process on the values and distribution of stresses in the simulated heat-affected zone
of S700MC steel welded joints. For this purpose, tests were carried out on a thermal cycle simulator,
as well as heating the prepared samples in accordance with the recorded welding thermal cycles,
and then annealing at temperatures from 200 to 550 ◦ C. The stresses values in the tested samples before
and after the annealing process were measured by using X-ray diffraction (XRD). The performed
tests were verified with the results of numerical analyses using the finite element method (FEM)
performed in the VisualWeld (SYSWELD) environment as, on the one hand, the verification of the
obtained results, and, on the other hand, the source of data for the development of a methodology for
conducting analyses of heat treatment processes of S700MC steel welded structures. Also presented
are three examples of numerical analyses for Gas Metal Arc (GMAW), laser and hybrid welding and
then the annealing process of the obtained joints at selected temperatures. The main purpose of the
work was to broaden the knowledge on the influence of annealing parameters on the values and
distribution of stresses in welded joints, but also to signal the possibility of using modern software in
engineering practice.
Keywords: S700MC; welding thermal cycle; annealing; FEM; numerical analyses; stresses distribution
1. Introduction
Today it is impossible to imagine a world without welded structures. From household appliances
to huge vessels or oil rigs, construction elements connected using many known welding methods are
used everywhere. When designing these modern constructions, we strive for ever higher efficiency,
demanding strength, lower weight and high durability from the materials used during operation,
sometimes in very diverse environmental conditions [1–3]. Due to the intensive development of the
iron metallurgy and the use of modern, highly specialized production processes, it is now possible to
produce steel materials with properties compliant with the requirements of the engineering materials
market, and not yet available for this group of materials. A complicated process of thermomechanical
rolling based on applying the deformation of the material in a certain temperature ranges, leading to
the achievement of a state characterized by specific, higher properties that cannot be achieved by steel
forming process alone [4,5]. Thanks to this, it is possible to reduce the weight while maintaining the
strength of the materials, which brings enormous economic benefits, resulting not only from the costs
of the material itself, but also its preparation for welding, processing and also transport. The properties
Materials 2020, 13, 5289; doi:10.3390/ma13225289 www.mdpi.com/journal/materials

Materials 2020, 13, 5289 2 of 26
of thermomechanically rolled steels (TMCP—Thermo Mechanical Controlled Processing) have been

enthusiastically received by industries such as automotive, machinery, construction and shipbuilding.
They are used for statically and dynamically loaded components, especially in the automotive industry
in the production of axles and chassis of cars and trucks, but also for components of cranes and other
structural units [2,6–8].
The advantages of these group of steels are associated with much higher production requirements.
In addition to the aforementioned thermomechanical treatment, the content of micro-alloying
additives such as niobium, vanadium and titanium in a strictly defined content, in combination
with thermomechanical treatment, enables grain refinement and precipitation hardening to take
place [9–12]. It is also to be expected that the complicated manufacturing process of these steels will
be a problem if they are subjected to the welding thermal cycle. A different method of production
compared to conventional techniques, as is easy to guess, has a significant influence on the properties
of TMCP steel welded joints [9,12]. The research results available in the literature indicate as the main
problems that arise during the welding process:
- loss of the initial properties during and after the welding process, which may cause partial
dissolution of fine-dispersion strengthening precipitates (carbides, Nb, Ti, V carbonitrides) and
their uncontrolled re-precipitation
- excessive growth of strengthening precipitates and loss of their ability to inhibit grain growth,
- transfer of microalloying elements to the weld and the difference in the carbon equivalent between
the base material and the weld [5,13–16].
However, this is not the only problem faced by welding engineers using these TCMP steels.
Welded structures, usually consisting of dozens of welds, very often require heat treatment to reduce
the level of stresses remaining after the welding process and to homogenize their distribution. This is
of great importance if we consider the fact that the unfavorable changes caused by welding thermal
cycles may increase the tendency to brittle and fatigue cracking [16,17]. Also, in this case, TMCP steels
require the use of very precise settings of the process itself due to the risk of loss of strength and plastic
properties [10,14,15,18]. From the product data sheets, it can be determined that normalizing and
stress relieving annealing at temperatures above 580 ◦ C with long holding times results in a reduction
of the yield point and tensile strength compared to the “as delivered” condition [19]. The available
research results in this field also indicate that the annealing process conducted in a wide range of
temperatures (from 100 to 1300 ◦ C) showed that the stability of the bainitic structure is maintained up
to the temperature of 600 ◦ C. Above this temperature, a slight grain growth is noted and exceeding the
processing temperature of 1000 ◦ C causes an increase in the amount of ferrite in the grain structure
and size. The steel retains its strength properties and hardness after heat treatment for processing
temperatures not exceeding 600 ◦ C. After exceeding this value, the tensile strength and hardness
are much lower compared to the base material [20,21]. With this in mind, the use of TCMP steels
must go hand in hand with a broad knowledge of their behavior in the production process. In order
to control the properties and microstructure of welded joints, it is necessary to conduct extensive
research explaining the impact of both the welding process and the applied heat treatment on the
structural changes taking place [22–25]. Due to the interaction of thermal cycles and the resulting
stresses, identifying the causes of deterioration of properties is not very easy [23,26,27].
Modern software for simulating welding and heat treatment processes may come in handy in this
situation. However, it often happens that the welding technology tested on small samples does not
give the same results on large structures where we deal with different heat dissipation or stiffening of
elements. In such situations, one solution to realistically determine the distribution and level of stresses
in the structure are the mentioned calculation programs. Correctly prepared numerical simulations
allow for obtaining a lot of additional information about the welding results, but also about the
process [28–30]. This allows for the optimization of the process, the detection of potential faults as well
as the recent determination of the lifetime of the structure under specific operating conditions. It should
Materials 2020, 13, 5289 3 of 26
be mentioned that of all the numerical analyses carried out using the finite element method (FEM),
simulations of welding and heat treatment processes are some of the most complex and difficult types
to perform. They require the correct preparation of both the model itself and the material database
containing material data in function of the temperature and metallurgical changes taking place in the
material, as well as precise calibration and validation of calculation models [31,32].
The main problem of numerical simulations of welding and heat treatment processes is the
appropriate methodology for conducting these analyses and the correct definition of the adopted
assumptions and material models. A frequent error in calculations, i.e., not considering the structural
changes occurring with temperature changes, may result in significant differences in the obtained
results compared to reality. Therefore, to maintain the credibility of the results of the numerical
analyses, the tool used to carry it out must cover the phenomena that occur during the modeled process
as much as possible. The use of random calculation tools which are not specialized unfortunately may
lead to incorrect results [31,33–36].
In the numerical simulations presented in the article, the VisualWeld (SYSWELD) software
package (ESI Group, Paris, France) was used. The SYSWELD solver, like many tools for calculating
thermal problems in FEM, is based on the calculation of temperature fields based on the Fourier
differential formula. However, in SYSWELD it is modified by the shares of metallurgical phases heat
convection equation, which, thanks to the coupled thermo-metallurgical analysis, is represented by the
relationship [35,36]:
    
X  ∂T X   X
 Pi (ρC)i  − ∇
  Pi λi ∇T + Lij (T ) · Aij = Q (1)
∂t

i i i< j
where:
T—temperature [K],
t—time [s],
x, y, z—individual coordinates in the axes of the system [m],
a—thermal diffusivity coefficient [m2 ·s−1 ],
λ —heat conductivity coefficient [W·m−1 K−1 ],
c—specific heat [J·kg−1 K−1 ],
ρ —mass density [kg·m−3 ],
P—phase proportion,
i,j—phases index,
Q—heat source,
Lij (T)—latent heat of i → j transformation,
Aij —proportion of phase i transformed to j in time unit.
This allows for a much more precise representation of the material behavior during heating and
cooling operations at different speeds of the materials to be welded or processed.
To sum up, the use of modern research techniques in combination with the possibilities offered by
modern software for conducting numerical analyses, with a correctly adopted research methodology,
allows us to obtain many answers to many questions regarding the behavior of advanced materials
in manufacturing processes and to fully use their specific properties. The presented work presents
both laboratory results of research on the influence of the annealing temperature of S700MC steel,
later confirmed by the results of numerical analyses, as well as several examples of real joints welded
using various techniques.
Materials 2020, 13, 5289 4 of 26
2. Description of the Problem

The aim of the experimental part of the research was to determine the relationship between the
applied parameters of thermal cycles during welding and heat treatment (annealing) and the values and
distributions of residual stresses in the tested samples of S700MC steel. For this purpose, on a test stand
equipped with a Gleeble 3500 thermal cycle simulator (Dynamic System Inc., New York, NY, USA),
three heat cycles differing in the maximum temperature value were applied on specially prepared
samples. Then the residual stresses in the samples were determined by using X-ray diffraction analysis.
These samples were annealed in a vacuum furnace for a few selected maximum temperatures ranging
from 300 to 650 ◦ C. After the samples had cooled down, the stresses were measured again. As the
next step, an attempt was made to perform numerical analyses in order to illustrate the phenomenon
under study, and create the basis for the methodology of conducting numerical analyses of the impact
of changes in annealing temperature after welding thermal cycle on the values and distribution of
residual stresses. Several sample results for various welding methods were also presented.
2.1. Material Used for Annealing Tests

The material used for the experiments was S700MC steel, from the group of HSLA
(High Strength Low Alloy) steels. It is a structural, fine-grained high-strength steel with a minimum
guaranteed yield strength of 700 MPa, which is suitable for cold forming and can be welded. This steel
is thermomechanically rolled and microalloyed. The alloys are predominantly aluminum, titanium,
niobium and vanadium. Said elements form fine carbides, nitrides or carbonitrides, which contribute
to the refinement of the grain and the strengthening of the matrix. The secondary effect of grain
refinement is in increased values of yield and tensile strength, as well as in reduced values of transient
temperature and brittle fracture properties. The chemical composition of S700MC steel defined by
the standard and determined experimentally using a Bruker Q4 TASMAN spectrometer was shown
in Table 1.
Table 1. Chemical composition according to EN 10149-2 and measured on S700MC steel samples.
Chemical Composition, % wt.

C Si Mn P S Al Nb V Ti B Mo
max. max. max. max. max. min. max. * max. * max. * max. max.
0.12 0.60 2.10 0.008 0.015 0.015 0.09 0.20 0.22 0.005 0.50
Chemical composition, % wt. measured on spectrometer
0.051 0.197 1.916 0.006 0.006 0.038 0.063 0.072 0.056 0 0.113
* Total content of Nb, V and Ti should not exceed max. 0.22%.
The measurements on the spectrometer showed that the tested steel has a very specifically low
carbon content (0.056%), therefore the strength and plastic properties of welded joints may be reduced
as a result of limiting the influence of austenite changes during the welding process. Because of this,
it was also necessary to modify the SYSWELD material base used in the performed numerical analyses
to compensate for the differences in the chemical composition of the existing base and the base needed
for calculations. The mechanical properties were determined experimentally by means of a static
tensile test and they are presented in Table 2.
Materials 2020, 13, 5289 5 of 26
Table 2. Results of mechanical properties tests of used S700MC steel.
Yield Tensile
Elongation Elongation
Specimen Specimen Diameter Strength Strength
Ag A40
Designation [Mm] Re Rm
[%] [%]
[MPa] [MPa]
S1 6.52 759 851 10.97 23.91
S2 6.53 758 852 12.33 25.22
S3 6.50 741 849 10.72 23.62
Average
Materials 2020, 13, x FOR value:
PEER REVIEW 753 ± 8.0 851 ± 1.0 11.01 ± 0.71 24.25 ± 0.75 of 28
Remarks: Ag —percentage non-proportional elongation at maximum force.
2.2. Assumptions for Testing Stresses Distribution in S700MC Steel under the Conditions of Welding and
Annealing Process Thermal
2.2. Assumptions for TestingCycles
Stresses Distribution in S700MC Steel under the Conditions of Welding and
Annealing Process Thermal Cycles
In order to determine the influence of the annealing temperature on the magnitude and stresses
In order
distribution to determine
after the welding theprocess,
influencetests
of the annealing
were carriedtemperature
out on the on the magnitude
Gleeble 3500 thermaland stresses
cycle
distribution after the welding process, tests were carried out on the Gleeble
simulator. The experiment was planned in such a way that it was possible to determine both the 3500 thermal cycle simulator.
The experiment
influence was planned
of the simulated in such
thermal cyclea of
waythethat it wasprocess
welding possibleand to determine both theheat
the post-welding influence of the
treatment
simulated thermal cycle of the welding process and the post-welding
thermal cycle. For this purpose, samples with dimensions 10 × 10 × 102 mm were prepared. An M10 heat treatment thermal cycle.
purpose, samples with dimensions 10 × 10 ×
thread was cut at the ends of the specimens to secure the specimen in the jaws of the instrument. The at
For this 102 mm were prepared. An M10 thread was cut
the ends
square of the specimens
cross-section to secureso
was selected thethat
specimen
when in the jaws ofthe
measuring the residual
instrument. The square
stresses using cross-section
the X-ray
was selected so that when measuring the residual stresses using the
diffraction method (XRD), there was no need to compensate for the shape of the sample surface, X-ray diffraction method (XRD),
there was no need to compensate for the shape of the sample surface, which
which affects the accuracy of the measurement. For the finished samples, it was necessary to remove affects the accuracy of the
themeasurement.
undesirable For the finished
residual stressessamples, it was necessary
in the surface to remove
layer created afterthe undesirable
machining. residual
For stresses
first group of in
the surface layer created after machining. For first group of samples,
samples, this was achieved by heating in a vacuum furnace at a temperature of 650 °C for 2 h, which this was achieved by heating
◦ C for 2 h, which are annealing parameters commonly
areinannealing
a vacuum parameters
furnace at a commonly
temperature of 650
used in technical practice, in the heat treatment of welded
used in technical
assemblies. In the practice,
second groupin the heat treatment
of samples, of surface
the welded layer
assemblies. In the second
was removed group
by local of samples,
electrolytic
the surface layer was removed by local electrolytic etching. For this case,
etching. For this case, a special silicone nozzle with dimensions 7 × 15 mm was made with the help a special silicone nozzle with
ofdimensions
3D printing, 7 ×which
15 mm was made
copied with the
the surface ofhelp
the of 3D printing,
sample and thus which copied
resulted inthe surfaceetching.
uniform of the sample
The
and thus resulted in uniform etching. The thickness of the etched layer was
thickness of the etched layer was 0.2 mm. The sample with the etched surface layer is shown in Figure 0.2 mm. The sample with
1a.the etched surface layer is shown in Figure 1a.
(a) (b)
Figure
Figure AA
1. 1. view
view ofof
(a)(a)test
testsample
samplewith
withetched
etchedarea
area for
for measurements
measurements and
and (b)
(b)aasample
sampleattached
attachedtotothe
tooling with a welded thermocouple.
the tooling with a welded thermocouple.
Thesamples
The sampleswerewerethen
thenprepared
preparedfor fortests
testsononthe
theGleeble
Gleeble35003500simulator.
simulator. First,
First,a aK-type
K-type
thermocouple was welded to the center of each sample to control the heating and
thermocouple was welded to the center of each sample to control the heating and cooling temperature cooling temperature
cycles.
cycles. After
After welding
welding the
the thermocouple,two
thermocouple, twonuts
nutswere
werescrewed
screwedonto onto both
both ends
ends of of the
the sample,
sample, firstly
firstly to
ensure proper clamping in the temperature simulator, but mainly so that the
to ensure proper clamping in the temperature simulator, but mainly so that the sample could be sample could be exposed
to tensile
exposed and compressive
to tensile stresses.
and compressive Finally,
stresses. the sample
Finally, waswas
the sample inserted intointo
inserted a copper jaws
a copper jawswith
withfull
contact,
full contact,as as
shown
shown in Figure 1b, 1b,
in Figure so that a temperature
so that a temperaturegradient corresponding
gradient to thetoconditions
corresponding during
the conditions
real welding is created in the sample. For this reason, the free length of the sample
during real welding is created in the sample. For this reason, the free length of the sample between between the jaws
was limited to 10 mm. The whole assembly was inserted into the working
the jaws was limited to 10 mm. The whole assembly was inserted into the working chamber of the chamber of the Gleeble
3500 simulator.
Gleeble For correct
3500 simulator. and sufficient
For correct contactcontact
and sufficient between the cooper
between the jaws
cooperand the and
jaws clamping system of
the clamping
system of the device, the sample was fixed with spacer bolts. The primary reason for this step is to
ensure good current transfer heating the sample and to ensure intensive heat dissipation through the
copper jaws. A secondary reason for the above procedure is to secure the specimen against movement
so that tensile and compressive stresses can be applied to it. The clamped specimen located in the
working chamber of the Gleeble system is shown in Figure 2.
Materials 2020, 13, 5289 6 of 26
the device, the sample was fixed with spacer bolts. The primary reason for this step is to ensure good
current transfer heating the sample and to ensure intensive heat dissipation through the copper jaws.
A secondary reason for the above procedure is to secure the specimen against movement so that tensile
and compressive stresses can be applied to it. The clamped specimen located in the working chamber
Materials
of 2020, 13,system
the Gleeble x FOR PEER REVIEW
is shown in Figure 2. 6 of 28
Figure 2. AAview
Figure 2. view
of of a Gleeble
a Gleeble 35003500 simulator
simulator working
working chamber
chamber with sample
with sample attached
attached to the
to the tooling
tooling equipment.
equipment.
The Gleeble 3500 system allows heating at speeds of up to 10,000 ◦ C/s. Therefore, real welding
The Gleeble 3500 system allows heating at speeds of up to 10,000 °C/s. Therefore, real welding
cycles measured by arc welding thermocouples can be applied to the sample. Figure 3 shows the
cycles measured by arc welding thermocouples can be applied to the sample. Figure 3 shows the
temperature cycle used, which was recorded during GMA welding of a S700MC steel fillet weld.
temperature cycle used, which was recorded during GMA welding of a S700MC steel fillet weld.
Based on it, three temperature cycles with different maximum temperatures were programmed: 1080,
Based on it, three temperature cycles with different maximum temperatures were programmed: 1080,
1200 and 1365 ◦ C. Thus, the temperature cycles corresponding to the three different locations in the
1200 and 1365 °C. Thus, the temperature cycles corresponding to the three different locations in the
Heat Affected
Materials 2020, 13,Zone
x FOR (HAZ).
PEER REVIEW 7 of 28
Heat Affected Zone (HAZ).
The next step was to set the method of program control. The Gleeble system allows several ways
to control the process, with power control being the most commonly used method. In such a case, the
control system makes it possible, thanks to the feedback control and the load cell, to apply a defined
force (or stress) or a zero force to the samples during heating and cooling. In this case, the jaws are
moved according to the data from the load cell so that no residual stresses or plastic deformations
occur in the sample. In welding simulations, control by means of a strain gauge is usually used, when
a defined displacement of the specimen or zero elongation of the specimen can be realized. This
simulates the clamping stiffness of the specimen, as in the real welding case. As the thickness of the
welded materials increases, the intrinsic stiffness of the entire system also increases. Therefore, a
welding method using a strain gauge is used in welding simulations. Then it is possible to define
how much the sample can be extended during heating, or with zero possibility of elongation it is
possible to define a perfectly rigid clamping of the sample. This method was chosen when applying
the programmed temperature cycles to the test specimens. In addition, it was tested whether and, if
so, how
Figure the3. increase
Comparison
Comparison in of
cooling ratecooling
controlled
controlled to thethermal
value cycles
of residual stresses
(green curve) in the
which sample
simulate would
welding cycleoccur.
with
with thickness
Therefore, thickness ofof1010
both temperature mmmm andand
thermal cyclecycle
thermal
cycles withwith
corresponding highertocooling
higher rate
cooling
reality and(red curve)
rate (red which
temperaturecurve) simulate
whichwith
cycles welding
simulate
so-called
cycle
welding
free cooling with thickness
cycle
were with
used, higher
wherethan
thickness 10 reaching
higher
after mm.
than 10 mm.the maximum temperature the controlled reheating is
switched off and the sample is cooled only by heat dissipation through high temperature jaws with
The
Afternext step was to set the
programming method of program control. The the
Gleeble system allows to several ways
full contact. In this case,the thetemperature
sample wascycle andapproximately
cooled evacuating working
2.5 times chamber
faster than acontrolled
value of
to control
0.533 Pa the
(4·10 process,
−3 Torr), with
the power
test itself control
could bebeing the
started. most
For commonly
all samples, used
the method.
preparation In such
and testa case,
were
cooling. Both used thermal cycles with different cooling methods were shown in Figure 3. The free
the controlinsystem
repeated theshows makes
same it only
possible,
withthanks to the feedback control
the setand the load cycle
cell, totemperature
apply a defined
cooling curve away,
delay at about the
480difference
°C, caused between
by a transformationmaximum that releases latent heat.and
force
the above-mentioned cooling method. During the experiments, the temperature, forces the
(or stress) or a zero force to the samples during heating and cooling. In this case, andjaws are
stresses
moved according to the data from the load cell so that no residual stresses
in the sample were measured and the possible elongation of the sample was also checked. Figure 4 or plastic deformations
occur
showsinthe thedependence
sample. In of welding simulations,
the stresses generated control by sample
in the means of a strain
during gaugetoisthe
heating usually
maximum used,
temperature of the cycle.
Materials 2020, 13, x FOR PEER REVIEW 7 of 28
Materials 2020, 13, 5289 7 of 26
when a defined displacement of the specimen or zero elongation of the specimen can be realized.
This simulates the clamping stiffness of the specimen, as in the real welding case. As the thickness of
the welded materials increases, the intrinsic stiffness of the entire system also increases. Therefore,
a welding method using a strain gauge is used in welding simulations. Then it is possible to define
how much the sample can be extended during heating, or with zero possibility of elongation it is
possible to define a perfectly rigid clamping of the sample. This method was chosen when applying
the programmed temperature cycles to the test specimens. In addition, it was tested whether and, if so,
how the increase in cooling rate to the value of residual stresses in the sample would occur. Therefore,
both temperature cycles corresponding to reality and temperature cycles with so-called free cooling
were used, where after reaching the maximum temperature the controlled reheating is switched off
and the sample is cooled only by heat dissipation through high temperature jaws with full contact.
Figure
In this case,3.the
Comparison
sample wasof controlled cooling thermal 2.5
cooled approximately cycles (green
times curve)
faster which
than simulatecooling.
controlled welding Both
cycle used
withcycles
thermal thickness
withofdifferent
10 mm cooling
and thermal cycle were
methods with shown
higher cooling
in Figurerate
3. (red curve)
The free whichcurve
cooling simulate
shows a
welding cycle with
◦ thickness higher than 10 mm.
delay at about 480 C, caused by a transformation that releases latent heat.
After programming the temperature cycle and evacuating the working chamber to a value of
After programming
0.533 Pa (4·10 −3 Torr), thethetesttemperature
itself could cycle and evacuating
be started. the working
For all samples, chamber and
the preparation to a test
value of
were
0.533 Pa (4·10
repeated in the Torr),
−3 the test
same way, onlyitself
withcould be started.
the difference For all the
between samples, the preparation
set maximum and test were
cycle temperature and
repeated in the same way, only with the difference between the set maximum
the above-mentioned cooling method. During the experiments, the temperature, forces and stresses cycle temperature andin
the above-mentioned cooling method. During the experiments, the temperature,
the sample were measured and the possible elongation of the sample was also checked. Figure 4 shows forces and stresses
in
thethe sample were
dependence measured
of the and the possible
stresses generated elongation
in the sample duringof heating
the sample was
to the also checked.
maximum Figure of
temperature 4
shows the
the cycle. dependence of the stresses generated in the sample during heating to the maximum
temperature of the cycle.
Figure 4. Graph
Figure Graph of
ofthe
thedependence
dependenceof the sample
of the stresses
sample on the
stresses ontemperature duringduring
the temperature heatingheating
and cooling.
and
cooling.
However, it should be noted that the temperature in the sample was not uniform (this was a
temperature
However, gradient),
it shouldand therefore
be noted thateach
the part of the sample
temperature in the behaves
sample was differently, as in HAZ
not uniform (this welds.
was a
It is also cleargradient),
temperature from the andgraph at Figure
therefore 4 that
each after
part the sample
of the start of behaves
heating in the sample,
differently, asthe compressive
in HAZ welds.
stresses ◦ C (red line at Figure 4).
It is alsofirst
clearincrease,
from the and onlyat
graph reach the 4yield
Figure thatpoint
after at
thea temperature
start of heatingof about
in the300
sample, the compressive
Furthermore,
stresses the sample
first increase, andwasonlyplastically
reach thedeformed,
yield point soatthea stress value stabilized
temperature of about 300at around
°C (red 400 MPa
line at
and the4).temperature
Figure Furthermore, continued
the sampleto rise. After reaching
was plastically a temperature
deformed, of approx.
so the stress 750 ◦ C, there
value stabilized was a
at around
significant
400 MPa and decrease in compressive
the temperature stresses,
continued which
to rise. Afterwas caused
reaching by the transformational
a temperature of approx. 750 conversion
°C, there
of thea material
was significant to austenite.
decrease After reaching the
in compressive maximum
stresses, cycle
which temperature
was caused byand thethe start of cooling
transformational
(blue line atofFigure
conversion 4), the compressive
the material to austenite.stresses were converted
After reaching into tensile
the maximum cyclestresses, the value
temperature and ofthewhich
start
increased during cooling. When the transformation temperature of austenite was
of cooling (blue line at Figure 4), the compressive stresses were converted into tensile stresses, the reached, the stress
was further
value of whichsignificantly
increasedreduced.
during With further
cooling. Whencooling, the tensile stresses
the transformation continue to
temperature of increase,
austeniteaswascan
be seen in Figure 4.
reached, the stress was further significantly reduced. With further cooling, the tensile stresses
continue to increase, as can be seen in Figure 4.
Materials 2020, 13, 5289 8 of 26
3. Results
3. Results of
of the
the Tests on the
Tests on the Thermal
Thermal Cycles
Cycles Simulator
Simulator
The experiment
The experiment plan
plan consisted
consisted of
of the
the following
following points:
points:
1)
(1) use
use of
of welding
weldingthermal
thermalcycles
cyclesatatGleeble
Gleeble3500
3500simulator
simulatorwith a maximum
with a maximum cycle temperature
cycle of
temperature
1365, 1200 and 1080 °C and a cooling rate corresponding to the measured welding cycle,
of 1365, 1200 and 1080 ◦ C and a cooling rate corresponding to the measured welding cycle,
2) application of the above thermal cycles at Gleeble 3500 simulator with controlled cooling rate,
(2) application of the above thermal cycles at Gleeble 3500 simulator with controlled cooling rate,
3) determination of the conditions for lowering the value of residual stresses and quantifying the
(3) determination of the conditions for lowering the value of residual stresses and quantifying the
values for individual temperatures and strength at these temperatures. It was decided to
values for individual temperatures and strength at these temperatures. It was decided to perform
perform the experiments at the temperatures 200, 300, 400, 500◦ and 550 °C with the hold time at
the experiments at the temperatures 200, 300, 400, 500 and 550 C with the hold time at 2 and 4 h,
2 and 4 h,
(4) determination of mechanical properties of S700MC steel after annealing in selected
4) determination of mechanical properties of S700MC steel after annealing in selected process
process conditions.
conditions.
3.1. Residual Stresses Distribution after Welding Thermal Cycle Application
3.1. Residual Stresses Distribution after Welding Thermal Cycle Application
After applying the programmed temperature cycles to all tested samples, it was possible to
After applying the programmed temperature cycles to all tested samples, it was possible to
proceed to the measurement of internal residual stresses. The measurement was performed by X-ray
proceed to the measurement of internal residual stresses. The measurement was performed by X-ray
diffraction (XRD) analysis using a Proto iXRD combo instrument (Proto Manufacturing Inc., Taylor,
diffraction (XRD) analysis using a Proto iXRD combo instrument (Proto Manufacturing Inc., Taylor,
MI, USA) with a chromium X-ray tube, as can be seen in Figure 5.
MI, USA) with a chromium X-ray tube, as can be seen in Figure 5.
Figure 5. A view of a X-ray diffraction testing stand.

Figure 5. A view of a X-ray diffraction testing stand.
3.1.1. Samples with a Maximum Cycle Temperature at 1365 ◦ C
3.1.1. Samples with a Maximum Cycle Temperature at 1365 °C
In the case of testing samples subjected to the impact of the welding thermal cycle with a maximum
In the case
temperature of testing
of 1365 ◦ C, the samples
followingsubjected to the
samples were impact of the welding thermal cycle with a
made:
maximum temperature of 1365 °C, the following samples were made:
• samples no. 1 and 4—these were annealed to reduce residual stresses at temperature 650 ◦ C for 2 h.
• samples no. 1 and 4—these were annealed to reduce residual stresses at temperature 650 °C for
This was followed by a thermal cycle with a maximum temperature of 1365 ◦ C and free cooling.
2 h. This was followed by a thermal cycle with a maximum temperature of 1365 °C and free
• sample no. 3—this was annealed to reduce residual stresses at temperature 650 ◦ C for 2 h. This was
cooling.
followed by a thermal cycle with a maximum temperature of 1365 ◦ C with controlled cooling.
• sample no. 3—this was annealed to reduce residual stresses at temperature 650 °C for 2 h. This
• sample no. 5—a 0.2 mm surface layer was etched away. This was followed by a thermal cycle with
was followed by a thermal cycle with a maximum temperature of 1365 °C with controlled
a maximum temperature of 1365 ◦ C with controlled cooling without annealing at the beginning.
cooling.
• sample
The no. 5—a
achieved 0.2 mm
stresses surface
profiles for layer wasno.
samples etched
1 andaway. This was
4 are shown in followed
Figure 6. by a thermal cycle
The reason why two tests were performed under completely identicalwithout
with a maximum temperature of 1365 °C with controlled cooling annealing
conditions at the
was to assess
beginning.
the repeatability of the experiments, and thus the relevance of the obtained results. It can be seen
fromThe
the Figure 6 that
achieved the measured
stresses stresses
profiles for samplesdistributions
no. 1 and 4 and values in
are shown were almost
Figure 6. identical in these
two samples. Thus, the repeatability of the measurements for the same boundary conditions of the
experiment can be assumed. Therefore, and also due to the high time and cost of each experiment,
measurements were performed for only one sample at a time. The green line in Figure 6 indicates the
Materials 2020, 13, 5289 9 of 26
initial stresses state in the sample after annealing to reduce the residual stresses (650 ◦ C for 2 h) and
before testing thermal cycle application. The stresses on sample no. 3 with controlled cooling and
sample no. 1, which was freely cooled, were also compared, see Figure 6.
Stresses distribution on sample no. 5 where the thermal cycle with a maximum temperature
of 1365 ◦ C was applied with no initial annealing was performed in comparison with sample no. 3
were shown on Figure 7. In such a sample, high values of residual stresses reaching a depth of about
0.1 mm are introduced into the surface layer after machining. Therefore, the affected surface layer was
etched to a depth of 0.2 mm so that only the unaffected material was measured. In order to remove
the thin surface layer, closed circuit electrolytic etching with the Electrolyte type A supplied by Proto
Manufacturing Europe was used. Due to the constant flow of electrolyte, the etched area was not
thermally affected. The internal stresses after the welding thermal cycle simulation were measured
in only one half of the sample. The reason was to maintain the same boundary conditions of the
experiment, as the size of the etched area was limited by the dimensions of the etching nozzle to
7 × 15 mm. From previous measurements, the symmetry of the distribution of internal stresses in the
right and left halves from the center of the sample was assumed. In sample no. 5, in comparison to the
previous samples, the internal stresses were further measured into the base material so as to reach a
steadyMaterials
state. 2020,
In the
13, xcase of both
FOR PEER samples, it was a simulation with controlled cooling.
REVIEW 9 of 28
FigureFigure 6. Comparison
6. Comparison of the
of the residual
residual stressesdistribution
stresses distribution in
in samples
sampleswith
withfree (specimen
free 1 and
(specimen 4) and
1 and 4) and
controlled cooling (specimen 3)—welding thermal cycle with a maximum temperature
controlled cooling (specimen 3)—welding thermal cycle with a maximum temperature of 1365 C. of 1365 °C. ◦
The reason why two tests were performed under completely identical conditions was to assess
the repeatability of the experiments, and thus the relevance of the obtained results. It can be seen
from the Figure 6 that the measured stresses distributions and values were almost identical in these
two samples. Thus, the repeatability of the measurements for the same boundary conditions of the
experiment can be assumed. Therefore, and also due to the high time and cost of each experiment,
measurements were performed for only one sample at a time. The green line in Figure 6 indicates the
initial stresses state in the sample after annealing to reduce the residual stresses (650 °C for 2 h) and
before testing thermal cycle application. The stresses on sample no. 3 with controlled cooling and
sample no. 1, which was freely cooled, were also compared, see Figure 6.
Stresses distribution on sample no. 5 where the thermal cycle with a maximum temperature of
1365 °C was applied with no initial annealing was performed in comparison with sample no. 3 were
shown on Figure 7. In such a sample, high values of residual stresses reaching a depth of about 0.1
mm are introduced into the surface layer after machining. Therefore, the affected surface layer was
etched to a depth of 0.2 mm so that only the unaffected material was measured. In order to remove
the thin surface layer, closed circuit electrolytic etching with the Electrolyte type A supplied by Proto
FigureFigure
Manufacturing 7. Comparison
7. Comparison of
Europeofwas residual
residual stresses
Due todistribution
used.stresses distribution
the constantin etched
in andelectrolyte,
etched
flow of not
andannealed samples
not annealed
the with a non-
samples
etched area withnot
was a
etched
non-etched sample after a welding cycle with maximum temperature of 1365 °C and
◦ controlled cooling.
thermally sample after
affected. Thea internal
welding stresses
cycle with maximum
after temperature
the welding thermalofcycle
1365simulation
C and controlled cooling.
were measured
in only one half of the sample. The reason was to maintain the same boundary conditions of the
experiment, as the size of the etched area was limited by the dimensions of the etching nozzle to 7 ×
In the
15 mm. caseprevious
From of samples heated by the welding
measurements, thermal
the symmetry ofcycle with a maximum
the distribution temperature
of internal stressesofin1200
the
°C, theand
right following samples
left halves fromwere tested:of the sample was assumed. In sample no. 5, in comparison to
the center
the previous samples, the internal stresses were further measured into the base material so as to reach
• sample no. 6—heated with a welding thermal cycle to a maximum temperature of 1200 °C with
a steady state. In the case of both samples, it was a simulation with controlled cooling.
controlled cooling,
Materials 2020, 13, 5289 10 of 26
3.1.2. Samples with a Maximum Cycle Temperature at 1200 ◦ C

Figure 7. Comparison of residual stresses distribution in etched and not annealed samples with a non-
In the etched sample
case of after a welding
samples heatedcycle
by with
the maximum
weldingtemperature of 1365 with
thermal cycle °C andacontrolled
maximum cooling.
temperature of
◦
1200 C, the following samples were tested:
• sampleInno. 6—heated with a welding thermal cyclecycle
to awith
maximum temperature ofof1200 ◦
the case of samples heated by the welding thermal a maximum temperature 1200 C with
°C, the following
controlled cooling,samples were tested:
• sample no. 7—heated with a welding ◦
• sample no. 6—heated with a weldingthermal cycletoto
thermal cycle a maximum
a maximum temperature
temperature of 1200of
°C1200
with C with
free cooling.
controlled cooling,
Both samples had an etched surface layer to a depth of 0.2 mm. As with sample no. 5, the residual
free cooling.
stresses were measured
Both onan
samples had only onesurface
etched half of thetosample,
layer to0.2
a depth of maintain the sample
mm. As with same boundary conditions of
no. 5, the residual
the experiment, as measured
stresses were can be seen in Figure
on only one half8.of the sample, to maintain the same boundary conditions of
the experiment, as can be seen in Figure 8.
8. Comparison
Figure Figure of the
8. Comparison of residual stresses
the residual distribution
stresses distributionin
in etched samplesafter
etched samples aftera awelding
welding cycle with
cycle
with a temperature
a maximum maximum temperature
of 1200 of◦ C1200
and°Cfree
andand
free controlled
and controlled cooling.
cooling.
3.1.3. Samples
3.1.3. Samples withwith a Maximum
a Maximum CycleTemperature
Cycle Temperature atat1080
1080°C◦ C
The last experiment on Gleeble was the application of a welding thermal cycle with a maximum
The last experiment on Gleeble was the application of a welding thermal cycle with a maximum
temperature of 1080 °C. In this case, only one type of experiment was performed as sample no. 2. This
temperature of 1080 ◦ C. In this case, only one type of experiment was performed as sample no. 2.
sample was annealed to reduce residual stresses at 650 °C for 2 h. This was followed by a thermal
This sample was aannealed
cycle with maximumtotemperature
reduce residual
of 1080stresses
°C with at 650 ◦ C for
controlled 2 h. This
cooling. was followed
The course by a thermal
of the residual
cycle with a maximum temperature of 1080 C◦
stresses across the sample after the application of with controlled
this cycle cooling.different
was significantly The course ofcycles
from the the residual
with a maximum temperature of 1365 and 1200 °C. The curve in the center of
stresses across the sample after the application of this cycle was significantly different from the sample did not tend
the cycles
to decrease significantly, as can be seen in Figure
◦ 9.
with a maximum temperature of 1365 and 1200 C. The curve in the center of the sample did not tend
to decrease significantly,
Materials as can
2020, 13, x FOR PEER be seen in Figure 9.
REVIEW 11 of 28
Figure
Figure 9. 9. Course
Course of residual
of residual stressesininsample
stresses sample 2
2 after
afteraawelding
weldingthermal cyclecycle
thermal with awith
maximum
a maximum
temperature of ◦1080 °C and controlled cooling.
temperature of 1080 C and controlled cooling.
3.2. Influence of Annealing on the Change of the Magnitude of Residual Stresses
To assess the effect of annealing on the magnitude of residual stresses in the material, two
annealing diagrams were experimentally compiled. One lasts 2 h and the other lasts 4 h at the
annealing temperature. The samples used in this experiment were annealed in a vacuum furnace at
200, 300, 400, 500 and 550 °C. After annealing all samples, the residual stresses values were measured
Materials 2020, 13, 5289 11 of 26
3.2. Influence of Annealing on the Change of the Magnitude of Residual Stresses

To assess the effect of annealing on the magnitude of residual stresses in the material, two
annealing diagrams were experimentally compiled. One lasts 2 h and the other lasts 4 h at the annealing
temperature. The samples used in this experiment were annealed in a vacuum furnace at 200, 300,
400, 500 and 550 ◦ C. After annealing all samples, the residual stresses values were measured by X-ray
diffraction analysis. In order to measure the initial stress of the base material, a part of the material was
etched from the surface of the sample to a depth of 0.2 mm. The measured values of residual stresses
after annealing with endurance at temperature for 2 and 4 h were summarized in Figure 10 and Table 3.
Table 3, on the one hand, indicates the initial state of stress at a given point after thermomechanical
rolling of the S700MC steel, and then the subsequent value of the residual stress after the selected
annealing temperature cycle to reduce the residual stresses. To measure the initial stress of the base
material, a part of the material was etched from the surface of the sample to a depth of 0.2 mm.
Figure 10. Annealing diagram for S700MC steel for annealing times of 2 h and 4 h.
Figure 10. Annealing diagram for S700MC steel for annealing times of 2 h and 4 h.
Table 3. Maximum values of residual stresses after annealing by 2 and 4 h at selected temperatures.
3.3. Influence of Annealing on the Reduction Of Residual Stresses after Temperature Cycles
Annealing Temperature (2 h Heating)
Sample
To assess Designation
the influence of the BM/200
annealing temperature on the residual stresses BM/550
value results after
BM/300 BM/400 BM/500
welding thermal cycle sample no. 1 was tested as an example. First the specimen was annealed at 650
AN1_2 508/413 468/306 431/218 388/107 525/62
°C for 2 h before heating through the welding thermal cycle. Then the welding cycle with a maximum
temperature of 1365AN2_2
°C was applied 472/403 435/269After431/198
to the sample. applying the 330/117 456/59 cycle and
welding thermal
measuring the internal stresses, annealing was Annealing Temperature
performed to reduce(4 hthe
Heating)
internal stresses level. For
this case, annealing at 300 °C for BM/200
2 h was chosen.
BM/300 After annealing
BM/400 the internal stresses
BM/500 BM/550 was again
measured at the same
AN1_4 points by X-ray diffraction
499/416 545/323analysis, Figure
341/184 11. It should
538/108 be noted here that
500/9
after annealing, the compressive stresses almost disappeared and the tensile stresses decreased, of
AN2_4 382/307 540/315 341/196 489/108 482/4
course, considering the dependence that the higher the annealing temperature, the lower◦ the residual
Remarks: designation—BM/200—max. value of stress [MPa] in base material/after annealing at 200 C etc.
stresses values. This relationship was also shown later in the numerical simulation results.
To assess the influence of the annealing temperature on the residual stresses value results after
welding thermal cycle sample no. 1 was tested as an example. First the specimen was annealed
at 650 ◦ C for 2 h before heating through the welding thermal cycle. Then the welding cycle with a
maximum temperature of 1365 ◦ C was applied to the sample. After applying the welding thermal
cycle and measuring the internal stresses, annealing was performed to reduce the internal stresses level.
For this case, annealing at 300 ◦ C for 2 h was chosen. After annealing the internal stresses was again
measured at the same points by X-ray diffraction analysis, Figure 11. It should be noted here that after
To assess the influence of the annealing temperature on the residual stresses value results after
welding thermal cycle sample no. 1 was tested as an example. First the specimen was annealed at 650
°C for 2 h before heating through the welding thermal cycle. Then the welding cycle with a maximum
temperature
Materials 2020, 13, 5289of 1365 °C was applied to the sample. After applying the welding thermal cycle and 12 of 26
measuring the internal stresses, annealing was performed to reduce the internal stresses level. For
this case, annealing at 300 °C for 2 h was chosen. After annealing the internal stresses was again
measured
annealing, at the same points
the compressive by almost
stresses X-ray diffraction analysis,
disappeared and Figure 11. Itstresses
the tensile should be noted here
decreased, ofthat
course,
after annealing,
considering the compressive
the dependence that the stresses
higher thealmost disappeared
annealing and the tensile
temperature, stresses
the lower the decreased, of
residual stresses
course, considering the dependence that the higher the annealing temperature,
values. This relationship was also shown later in the numerical simulation results. the lower the residual
stresses values. This relationship was also shown later in the numerical simulation results.
11. Comparison
FigureFigure of theofresidual
11. Comparison stresses
stresses in samples
distribution subjected
in samples to the
subjected towelding thermal
the welding
cycle with a maximum
thermal cycle with temperature of 1365 ◦ C of
a maximum temperature with free
1365 °C cooling
with freethat was”that
cooling annealed at 300 ◦at
was” annealed C 300
for 2 h.
°C for 2 h.
3.4. Determination of Mechanical Properties after Annealing
3.4. completeness
For Determination ofof
Mechanical Properties aafter
measurements, Annealing
static tensile test was performed on selected annealed
samples. From this, the mechanical properties of tensile
For completeness of measurements, a static the heat-treated sampleson
test was performed were determined
selected annealed and
compared with the mechanical properties of the base unannealed material, which were listed inand
samples. From this, the mechanical properties of the heat-treated samples were determined Table 2.
First, two samples annealed at 450 ◦ C for 2 h were measured. Subsequently, a tensile test was performed
on two samples annealed at 550 ◦ C for 4 h, as can be seen in Table 4.
Table 4. Mechanical properties of the sample annealed at 450 ◦ C for 2 and at 550 ◦ C for 4 h.
Sample Upper Yield Tensile

Lower Yield Elongation Elongation
Sample Diameter Point Strength
Point Rel Ag A40
Designation Do Reh Rm
[MPa] [%] [%]
[mm] [MPa] [MPa]
annealed at 450 ◦ C for 2 h
T1_2 6.49 724.39 743.50 807.42 9.36 21.21
T2_2 6.52 744.01 773.54 825.36 9.38 20.88
annealed at 550 ◦C for 4 h
T1_4 6.49 722.66 745.38 784.61 9.51 20.93
T2_4 6.49 721.04 742.98 786.19 9.77 21.35
Remarks: Ag —Percentage non-proportional elongation at maximum force.
4. Numerical Simulations
While on small samples or test joints, the laboratory tests of the influence of the annealing
temperature on the mechanical properties and stresses distribution do not pose any major problems
except to the measurement accuracy and by causing possible disturbances in the measured values,
though this is problematic in the case of large structures and ready-made implementations. The fact
that it is difficult or impossible to carry out XRD tests and impossible to remove the surface layer under
these circumstances, means that modern techniques used in welding, such as numerical analyses,
Materials 2020, 13, 5289 13 of 26
can help. For this purpose, four examples related to the issue of annealing effect on stresses distributions
and values are presented later in this article. All analyses presented were calculated in VisualWeld
(code SYSWELD). The S700MC steel material database used in the simulations contains the full thermal,
metallurgical and mechanical material properties depended on the temperature and metallurgical
phases changes and was modified according to measured chemical composition, as can be seen
in Table 1.
4.1. An Example of the Numerical Model of Tests Carried Out on the Simulator of Thermal Cycles
The demonstration of the usefulness of the tool, which are numerical analyses of welding and heat
treatment processes, began with an attempt to simulate the tests carried out on a thermal cycle simulator.
For this purpose, a two-dimensional numerical model was created containing only 800 2D elements
and 861 nodes, as can be seen in Figure 12a. The model mesh was refined in the place of the defined
thermal cycle. A model of this type drastically reduces the computation time and, in some cases,
allows for almost immediate results to be obtained. This model was developed as a 2D-symetrical
model and the calculations were performed using the ‘transient’ (step-by-step) technique with the
Imposed Thermal Cycle load [31]. This method relies on loading selected mesh nodes with a defined
heat cycle instead
Materials ofx using
2020, 13, the
FOR PEER frequently used moving heat source model, as can be seen14inofFigure
REVIEW 28 12b.
(a) (b)
FigureFigure
12. A 12.
viewA view
of aof
2D a 2D numericalmodel
numerical model with
with (a)
(a)clamping conditions
clamping and (b)
conditions and“load” nodes area.
(b) “load” nodes area.
In the In the of
case case of using
using thisthis computational technique,
computational technique, it it is is
essential to correctly
essential calibrate
to correctly the model
calibrate the model
and reproduce the course of the heating curve, and then determine the boundary conditions so that
and reproduce the course of the heating curve, and then determine the boundary conditions so that
the temperature values and cooling speed obtained in the simulation correspond to those obtained
the temperature
in real testsvalues
duringand cooling speed
thermoelectrical obtained inAfter
measurements. the simulation correspond
the initial calibration to those
of the obtained in
numerical
real tests during
model, thermoelectrical
analyses were carried out measurements.
for three selectedAfter the initial
thermal calibrationthe
cycles simulating ofinfluence
the numerical
of the model,
analyses were process
welding carriedwithoutmaximum
for threetemperatures:
selected thermal1360, 1200cycles 1080 °C, as can
and simulating thebeinfluence of the
seen in Figure 13a.welding
According to the methodology of calculating with the use◦
process with maximum temperatures: 1360, 1200 and 1080 C, as can be seen in Figure 13a. the
of the predefined thermal cycle, According
selected mesh elements were loaded with the heating curve until the maximum temperature was
to the methodology of calculating with the use of the predefined thermal cycle, the selected mesh
reached, and then the cooling proceeded according to the assumed boundary conditions. The
elements were loaded with the heating curve until the maximum temperature was reached, and then
boundary conditions corresponding to the clamping were set to simulate heating and cooling of the
the cooling proceeded
elements according
corresponding to the assumed
to the conditions boundary
of mounting the realconditions.
sample in theThe jawsboundary conditions
of the Gleeble
corresponding
simulator.toAdditionally,
the clampingthewere set to simulate
symmetry boundaryheating
condition andwascooling
used dueof the
to elements corresponding
the fact that the
computational
to the conditions ofmodel was a symmetrical
mounting model, as
the real sample in can
thebejawsseen of
in Figure 12a. Thesimulator.
the Gleeble conditions forAdditionally,
heat
dissipation were determined by such heat dissipation at the external nodes
the symmetry boundary condition was used due to the fact that the computational model was a of the model mesh (except
for the nodes marked as the axis of symmetry) that in the welding cycle simulation phase, the
symmetrical model, as can be seen in Figure 12a. The conditions for heat dissipation were determined
calculated cycle was consistent with the one recorded on the simulator and then in the phase heat
by suchtreatment
heat dissipation
according to at the
the given
external nodes
thermal of of
cycle theannealing,
model mesh as can(except
be seenfor in the nodes
Figure marked as
13. The
the axiscalculated
of symmetry)
distribution of average stresses for the selected thermal cycles do not differ noticeably consistent
that in the welding cycle simulation phase, the calculated cycle was in
with the oneofrecorded
terms on the
the obtained simulator
values and then stresses.
of the calculated in the phase heatittreatment
However, should be according to the given
noted that their
thermaldistributions differ slightly,
cycle of annealing, as can
as can be be seen
seen ininFigure
Figure 13.
14. The calculated distribution of average stresses
for the selected thermal cycles do not differ noticeably in terms of the obtained values of the calculated
stresses. However, it should be noted that their distributions differ slightly, as can be seen in Figure 14.
for the nodes marked as the axis of symmetry) that in the welding cycle simulation phase, the
calculated cycle was consistent with the one recorded on the simulator and then in the phase heat
treatment according to the given thermal cycle of annealing, as can be seen in Figure 13. The
calculated distribution of average stresses for the selected thermal cycles do not differ noticeably in
terms 2020,
Materials of the obtained values of the calculated stresses. However, it should be noted that
13, 5289 their
14 of 26
distributions differ slightly, as can be seen in Figure 14.
(a) (b)
Figure
Figure13. Imposed thermal
13. Imposed thermalcycle
cycleofof(a)(a)heating
heating
forfor welding
welding conditions
conditions (dashed
(dashed lineline is applied
is applied cycle
cycle and dotted
and dotted is calculated)
is calculated) and
and (b) (b) thermal
thermal cycle
cycle of postofweld
postheat
weld heat treatment
treatment (annealing
(annealing with
with different
different temperature).
temperature).
[MPa] [MPa]
(a) (b)
Figure14.14.Comparison
Figure Comparison of the
of residual mean stresses
mean stresses after welding
distribution afterthermal cycle
welding application
thermal cycle
for (a) 1360 ◦ for
application C and
(a) (b)
13601080 ◦ C thermal cycles.
°C and (b) 1080 °C thermal cycles.
In
Inthe
thenext
nextstep,
step,numerical
numericalanalyses
analysesof ofthe
theannealing
annealingprocess
processwerewerecarried
carriedout
outatat400,
400,550
550and
and
650 ◦ C withstanding the element at these temperatures for 2 h in accordance with the cycle shown
650 °C withstanding the element at these temperatures for 2 h in accordance with the cycle shown in
in Figure13b.
Figure 13b.The
Thesamples
sampleswerewereheated
heatedon onall
allmesh
meshnodes
nodesatat the
the rate
rate ofof 55 ◦°C/min
C/min and
and cooled
cooled totothe
the
ambient temperature at the rate of 3 ◦ C/min. The stresses distribution and graphs of their values on
ambient temperature at the rate of 3 °C/min. The stresses distribution and graphs of their values on
the
thecross-section
cross-sectionin inthe
themiddle
middleofofthethesample
samplethickness
thicknessforforselected
selectedthermal
thermalcyclescyclesand
andtemperatures
temperatures
of
ofthe
theannealing
annealingprocess
processwere wereshown
shownininFigures
Figures15 15and
and16.
16.There
Thereisisaavisible
visibledecrease
decreasein inthe
thevalue
valueofof
stresses with the increase of the annealing temperature and at the temperature
stresses with the increase of the annealing temperature and at the temperature value of value of approximately
◦ C, the stresses value does not exceed the value of 50 MPa, as can be seen in Figures 16 and 17.
650
approximately 650 °C, the stresses value does not exceed the value of 50 MPa, as can be seen in
The
Figures 16 and 17. Theand
calculated values their course
calculated valuesareand
also similar
their in some
course areas
are also to the
similar in values recorded
some areas to theduring
values
the
recorded during the actual tests carried out on the thermal cycle simulator, Figures 6, 7, 10 real
actual tests carried out on the thermal cycle simulator, Figures 6, 7, 10 and 11. In the and tests,
11. In
for
thethe cycle
real with
tests, for athe
maximum
cycle withtemperature
a maximum oftemperature
1080 degrees,ofit1080
was degrees,
not measurements
it was not for the etched
measurements
samples. Both for
for the etched cycle 1360
samples. Both were provided,
for cycle i.e., for
1360 were the sample
provided, with
i.e., for theand without
sample withtheandsurface
withoutlayer
the
removed. The FEM simulation does not show the peak compressive stresses
surface layer removed. The FEM simulation does not show the peak compressive stresses because because the grinding of
the
thesample
grinding during
of theitssample
preparation
duringwas itsnot simulated.was
preparation Consequently,
not simulated. the simulation
Consequently, results
thecorrespond
simulation
to the results obtained for the etched samples.
results correspond to the results obtained for the etched samples.
[MPa] [MPa]
(a) (b)
Figure 15. Comparison of the residual mean stresses distribution after annealing in 400 °C by 2 h for
Figures 16 and 17. The calculated values and their course are also similar in some areas to the values
recorded during the actual tests carried out on the thermal cycle simulator, Figures 6, 7, 10 and 11. In
recorded during the actual tests carried out on the thermal cycle simulator, Figures 6, 7, 10 and 11. In
the real tests, for the cycle with a maximum temperature of 1080 degrees, it was not measurements
the real tests, for the cycle with a maximum temperature of 1080 degrees, it was not measurements
for the etched samples. Both for cycle 1360 were provided, i.e., for the sample with and without the
for the etched samples. Both for cycle 1360 were provided, i.e., for the sample with and without the
surface layer removed. The FEM simulation does not show the peak compressive stresses because
surface layer removed. The FEM simulation does not show the peak compressive stresses because
the grinding
Materials 2020, 13, of the sample during its preparation was not simulated. Consequently, the simulation
5289 15 of 26
the grinding of the sample during its preparation was not simulated. Consequently, the simulation
[MPa] [MPa]
[MPa] [MPa]
(a) (b)
(a) (b)
Figure 15. Comparison of the residual mean stresses distribution after annealing in 400◦ °C by 2 h for
Figure 15.Comparison
Figure15. Comparisonof ofthe
theresidual
residualmean
meanstresses
stressesdistribution
distributionafter
afterannealing
annealingin 400 °C
in400 C by
by22hhfor
for
(a) 1360◦ °C and (b) 1080◦ °C thermal cycles.
(a) 1360 °C and (b) 1080 °C thermal cycles.
(a) 1360 C and (b) 1080 C thermal cycles.
[MPa] [MPa]
[MPa] [MPa]

(a) (b)
(a) (b)
Figure 16. Comparison of the residual mean stresses distribution after annealing in 650 °C◦ by 2 h for
Figure 16. Comparison of the residual mean stresses distribution after annealing in 650 C by 2 h for
(a) 1360 °C◦ and (b) 1080 °C◦ thermal cycles.
(a) 1360 C and (b) 1080 C thermal cycles.
(a)
(b)
Figure
Figure 17.17. Mean
Mean stressesdistribution
stresses distributiongraphs
graphsafter
afterannealing in 400 ◦°C
annealing in 650 ◦°C
C and 650 C by
by22hhfor
for(a) 1360◦ C
(a)1360
◦
°Cand
and(b)
(b)1080
1080°CCmaximal
maximaltemperature
temperatureofofwelding
weldingthermal
thermalcycles.
cycles.
4.2. An Example of a Numerical Analysis of a Multipass Butt Joint Gas Metal Arc Welding and Annealing
Process
After numerical analyses of laboratory tests performed on a thermal cycle simulator, the analysis
of a model butt joint, welded with two beads, was carried out on a 2D cross-section model using the
(b)
Figure 17. Mean stresses distribution graphs after annealing in 400 °C and 650 °C by 2 h for (a) 1360
Materials 2020, 13, 5289 16 of 26
°C and (b) 1080° C maximal temperature of welding thermal cycles.
4.2. An
4.2. An Example
ExampleofofaaNumerical
NumericalAnalysis
Analysisof aofMultipass
a Multipass
ButtButt
JointJoint
GasGas Metal
Metal Arc Welding
Arc Welding and Annealing
and Annealing Process
Process
After numerical analyses of laboratory tests performed on a thermal cycle simulator, the analysis
After numerical
of a model butt joint, analyses of laboratory
welded with two beads,tests performed
was carried out on on
a thermal cycle simulator,
a 2D cross-section modeltheusing
analysis
the
of a model butt joint, welded with two beads, was carried out on a 2D cross-section
‘transient’ technique [31]. The numerical model prepared for the Gas Metal Arc Welding (GMAW) model using the
‘transient’
welding oftechnique
two 10 mm [31]. Theplates
thick numerical model552
contained prepared
elements for(440
the Gas Metal Arc
2D elements Welding
and (GMAW)
112 elements of
welding of two 10 mm thick plates contained
1D type) and 485 nodes, as can be seen in Figure 18. 552 elements (440 2D elements and 112 elements of 1D
type) and 485 nodes, as can be seen in Figure 18.
Figure 18. A view of 2D cross-section model of butt joint multipass GMAW welded and heat source
double-ellipsoidal heat
models used for (a) 1st and (b) 2nd bead welding with visible double-ellipsoidal heat source
source models.
models.
The boundary conditions describing the clamping of plates to be welded have been adopted in
such a way as to simulate the welding situation without additional fastening. The heat transfer from
the model to environment was defined at the outer edges of the model as convective heat dissipation
and radiation to the environment at 20 ◦°C. C.
Heat source models in VisualWeld (SYSWELD) are described by a volume density of energy
applied to elements
13, x FOR Q(x,y,z)
elements
Materials 2020, Q(x,y,z)
PEER [31,32,35].
[31,32,35].In
REVIEW Inthe
thedescribed
describedcase
caseas
asaamodel
model of
of aa moving
moving heat
heat source17inofthe
28
‘transient’ technique (step-by-step), a double ellipsoid volumetric model (Goldak’s model) was used,
as can be seen
‘transient’ in Figures
technique 18 and 19. a double ellipsoid volumetric model (Goldak’s model) was used,
(step-by-step),
as can be seen in Figures 18 and 19.
Figure
Figure 19.
19. AAview
viewof
ofthe
thedouble-ellipsoidal
double-ellipsoidal(Goldak’s)
(Goldak’s) heat
heat source
source model
model [31,33].
[31,33].
In
In SYSWELD,
SYSWELD, itit is
is possible
possible to
to introduce
introduce the
the power
power density
density function
function applied
appliedto
to the
the structure.
structure.
Accordingly,
Accordingly, the energy was divided into:
Q
Qff—as
—as the
the heat
heat energy
energy density
density introduced
introduced inin the
the front half of the model
6√3𝑓√𝑄 −3𝑥 2 ! −3𝑦 ! −3𝑧!
𝑄 (𝑥, 𝑦, 𝑧) = 6 3 f f𝑒𝑥𝑝 Q −3x 𝑒𝑥𝑝 −3y2 𝑒𝑥𝑝−3z2 (2)
= 𝜋√𝜋 √ exp
Q f (x, y, z) 𝑎𝑏𝑐 𝑎 exp 𝑏 exp 𝑐 (2)
abc f π π a 2 b 2 c 2
Qr—as the heat energy density introduced in the rear part:

Qr —as the heat energy density introduced in the rear part:
6√3𝑓 𝑄 −3𝑥 −3𝑦 −3𝑧
𝑄 (𝑥, 𝑦, 𝑧) = √ 𝑒𝑥𝑝 𝑒𝑥𝑝 ! 𝑒𝑥𝑝 2 ! (3)
𝑎𝑏𝑐6𝜋√3𝜋fr Q 𝑎 2 𝑏 2
−3z𝑐
!
−3x −3y
Qr (x, y, z) = √ exp exp exp (3)
where Q is the total power source; a, b, abccfr/cπr are
π respectively
a2 b2 c2
width/depth/length of the front/rear part
of the molten pool; ff, fr are constants that affect the flow of energy in a material. It must be also
mentioned that ff + fr = 2 [31,33].
After calibration procedure of the heat source models, the optimal parameters were determined,
as can be seen in Figure 20 and Table 5. Then the mechanical analysis for welding process was carried
out, as can be seen in Figure 21.
Table 5. Welding parameters used in numerical simulations of multipass butt joint GMAW
mentioned that ff + fr = 2 [31,33].
After calibration procedure of the heat source models, the optimal parameters were determined,
as can be seen in Figure 20 and Table 5. Then the mechanical analysis for welding process was carried
out, as can be seen in Figure 21.
Materials 2020, 13, 5289

Table 5. Welding parameters used in numerical simulations of multipass butt joint GMAW 17 of 26
welding.
where Q isBead
the total
Number power EPUL a, b, cf /cr v
source;(J/mm) are respectively
(mm/s) k width/depth/length
Heat Source Model ofDimensions
the front/rear* part of
the molten pool; f ,
1st bead
f rf are constants that
800 affect the flow
5.0 of energy
0.8 in a material.
10.0/10.0/5.0 be also mentioned
It must
that ff + fr = 2nd
2 [31,33].
bead 1000 4.0 0.8 15.0/12.0/5.0
After calibration
cooling conditions: free air in temperature 20 °C, preheatingparameters
procedure of the heat source models, the optimal temperature:were65 determined,
°C
as canRemarks:
be seen EPUL—energy
in Figure 20 and Table 5. Then the mechanical analysis for welding
per unit length, v—welding speed, k—energy efficiency factor; process
*—inwas
carriedsequence:
out, as can be seen
molten pool in Figure 21. pool width/penetration depth, as can be seen in Figure 19.
length/molten
[°C]
(a) (b)
Figure 20.AAview
Figure20. viewofoftemperatures
temperaturesdistribution
distributionduring
during(a)(a)1st
1stand
and(b)
(b)2nd
2ndbead
beadwelding.
welding.
Table 5. Welding parameters used in numerical simulations of multipass butt joint GMAW welding.
Bead Number EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

1st bead 800 5.0 0.8 10.0/10.0/5.0
2nd bead 1000 4.0 0.8 15.0/12.0/5.0
cooling conditions: free air in temperature 20 ◦ C, preheating temperature: 65 ◦ C
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—in sequence: molten
Materials
pool 2020, 13, x FORpool
length/molten PEER REVIEW
width/penetration depth, as can be seen in Figure 19. 18 of 28
[MPa]
[MPa]
(a) (b)
Figure 21.AAview
Figure21. ofof
view thethe
residual mean
residual stresses
mean distribution
stresses afterafter
distribution (a) 1st
(a)and
1st (b)
and2nd
(b)bead welding
2nd bead and
welding
cooling to the to
and cooling ambient temperature.
the ambient temperature.
After
Aftersimulation
simulation of welding,
of welding,as inas
theinprevious described
the previous case, numerical
described analyses ofanalyses
case, numerical the annealing
of the
◦ C were carried out in accordance with the thermal cycles
annealing process at temperatures of 400, 550 and 650 °C were carried out in accordance with the
process at temperatures of 400, 550 and 650
shown
thermalin Figure 13b. The
cycles shown inwelded
Figure joint
13b. was
The heated
weldedinjoint
all mesh nodes at
was heated inthe
all rate
mesh 5 ◦ C/min,
of nodes withstood
at the rate of 5
at°C/min,
a givenwithstood
temperature ◦
at afor 2 h and
given then cooled
temperature for at
2 the ratethen
h and of 3 cooled
C/minattothetherate
ambient temperature.
of 3 °C/min to the
The obtained stresses distributions and graphs of their values analyzed in three parallel
ambient temperature. The obtained stresses distributions and graphs of their values analyzed in three lines along the
cross-section
parallel linesofalong
the model were shown
the cross-section ofin
theFigures
model22 andshown
were 23. There is observed
in Figures 22 anda 23.
noticeable
There isdecrease
observed
ina the value of the calculated stresses, while maintaining their type
noticeable decrease in the value of the calculated stresses, while maintaining their (compressive/tensile) withtype
an
increase in the annealing temperature. In the case of annealing temperature of 650 ◦ C, a decrease in
(compressive/tensile) with an increase in the annealing temperature. In the case of annealing
the calculatedof
temperature stresses
650 °C,toa values
decrease notinexceeding approximately
the calculated 30 MPanot
stresses to values was visible, as
exceeding can be seen in30
approximately
Figures 22b and 23.
MPa was visible, as can be seen in Figure 22b and Figure 23.
[MPa] [MPa]
ambient temperature. The obtained stresses distributions and graphs of their values analyzed in three
parallel lines along the cross-section of the model were shown in Figures 22 and 23. There is observed
a noticeable decrease in the value of the calculated stresses, while maintaining their type
(compressive/tensile) with an increase in the annealing temperature. In the case of annealing
temperature of 5289
Materials 2020, 13, 650 °C, a decrease in the calculated stresses to values not exceeding approximately
18 of30
26
MPa was visible, as can be seen in Figure 22b and Figure 23.
[MPa] [MPa]
(a) (b)
Materials
Figure2020,
Figure 22.13,A
22. Ax view
FOR
viewPEER
of
of theREVIEW
the residual
residual mean
mean stresses
stresses distribution
distributionafter
afterannealing
annealingby
by22hhat 19 of 28
at temperatures:
temperatures:
(a)
(a) 400
400 and
and (b) 650 °C.
(b) 650 ◦ C.
Figure 23. Graphs of mean stresses values in the multipass butt-joint GMA welded, depending on the
Figure 23. Graphs of mean stresses values in the multipass butt-joint GMA welded, depending on the
distance from the weld axis for annealing temperatures equal to 400, 550 and 650 ◦ C.
distance from the weld axis for annealing temperatures equal to 400, 550 and 650 °C.
4.3. An Example of a Numerical Analysis of the Laser Butt Joint Welding and Annealing Process
4.3. An Example of a Numerical Analysis of the Laser Butt Joint Welding and Annealing Process
As a further example, a symmetrical three-dimensional numerical model of the of 10 mm thick
As a further
laser-welded butt example,
joint witha filler
symmetrical
materialthree-dimensional
was prepared. This numerical model of of
model consisted the32,091
of 10 mm thick
3D solid
laser-welded butt joint with filler material was prepared. This model consisted of 32,091
elements with 29,987 nodes. As in the previous case, the boundary conditions related to heat dissipation 3D solid
elements
to with 29,987
the environment werenodes.
definedAsoninallthe previous
external case,ofthe
surfaces theboundary conditionsheat
model as convective related to heat
dissipation
dissipation to the environment were defined
◦ on all external surfaces of the model as
and radiation to the environment at 25 C. The boundary conditions related to the fixing of the welded convective heat
dissipation
elements andset
were radiation
in suchto the environment
a way that the model at 25 °C. Thelike
behaves boundary
a sampleconditions
lying onrelated to the fixing
the welding table
of the welded elements were set in such a way that the model behaves like a sample
without any additional fixing. The additional boundary condition of symmetry was used on the model lying on the
welding
split planetable without
because any additionalmodel
the computational fixing.
wasThe additional boundary
a symmetrical model. The condition
analysesofwere
symmetry
carried was
out
used on the model split plane because the computational model was a symmetrical
using the ‘transient’ technique with a moving model of the heat source in the form of 3D conical which model. The
analyses were carried out using the ‘transient’ technique with a moving
can be determined by the defined power and its dimensions, as can be seen in Figure 24. model of the heat source in
the form of 3D conical which can be determined by the defined power and its dimensions, as can be
seen in Figure 24.
dissipation and radiation to the environment at 25 °C. The boundary conditions related to the fixing
of the welded elements were set in such a way that the model behaves like a sample lying on the
welding table without any additional fixing. The additional boundary condition of symmetry was
used on the model split plane because the computational model was a symmetrical model. The
analyses were carried out using the ‘transient’ technique with a moving model of the heat source in
Materials 2020, 13, 5289 19 of 26
the form of 3D conical which can be determined by the defined power and its dimensions, as can be
seen in Figure 24.
heat transfer to the material structure depending on the given coordinates

Figure 24. A
Figure 24. A view
view of
of the
the 3D
3D conical
conical heat
𝑥 source
heat + 𝑦 model
source model [30,32].
[30,32].
𝑄(𝑥, 𝑦, 𝑧) = 𝑄 𝑒𝑥𝑝 − (4)
The mathematical description of this model is contained 𝑟 in(𝑧)two equations which describes [31,33]:
The mathematical description of this model is contained in two equations which describes
heat transfer to the material structure depending on the given coordinates
[31,33]: the change in radius at depth
and
𝑟 −𝑟 2
𝑟 (𝑧) = 𝑟 +  x(𝑧 +− y2)
𝑧

(5)
Q(x, y, z) = Q0𝑧exp −𝑧−  (4)
r20 (z)
where: Q0 is the heat flux density, re, ri, ze, zi are the cone geometrical dimensions according to Figure
24. and the change in radius at depth
The calculations were performed using the “transient” calculation technique with a moving heat
ri − re
source model. After calibrating procedure r20 (z) =ofrthe
e + heat source
(z − ze model
) and obtaining full remelting(5)of
z − ze
the welded sheets, a mechanical analysis was also icarried out to determine the stresses distribution
in the analyzed
where: joint flux
Q0 is the heat afterdensity,
welding re ,process, as the
ri, ze , zi are cancone
be seen in Figuredimensions
geometrical 25, Table 6.according to Figure 24.
The calculations were performed using the “transient” calculation technique with a moving heat
source model.Table
After6.calibrating
Welding parameters
procedure used in aheat
of the lasersource
butt-joint welding
model and simulation, Fig. remelting
obtaining full 25. of the
welded sheets, a mechanical analysis was also carried out to determine
Designation EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions * the stresses distribution in the
analyzed joint after
laser welding process,
360 as can be
16.0 seen in
0.8Figure 25, Table 6.
2.0/1.5/10.0
After cooling
weldingconditions:
and coolingfree
theair
sample,
in temperature 25 °C, preheatingalltemperature:
a thermal cycle was set for external nodes of the model
65 °C
corresponding to annealing at 400 and 650 ◦ C, Figures 26 and 27. The calculated stresses distributions
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—for 3D
were conical
analyzed in the same way as before in three parallel lines drawn on the cross-section of the model,
model in sequence: top diameter/bottom diameter/penetration depth.
as shown in the Figure 28 with graphs of the stress distribution in individual lines.
[°C]
Figure 25.
Figure A view
25. A view of
of (a)
(a) temperatures
temperatures distribution
distribution during
during weld
weld and
and (b)
(b) 3D
3D conical
conical heat
heat source
sourcemodel.
model.
Table 6. Welding parameters used in a laser butt-joint welding simulation, Figure 25.
After welding and cooling the sample, a thermal cycle was set for all external nodes of the model
Designation
corresponding EPUL (J/mm)
to annealing at 400 and 650 °C, Figures 26
v (mm/s) k and 27.
Heat
TheSource Modelstresses
calculated Dimensions *
distributions
were analyzed
laser in the same 360way as before 16.0in three parallel
0.8 lines drawn 2.0/1.5/10.0
on the cross-section of the
model, as shown in the Figure 28 with graphs of the stress
◦ distribution in individual
cooling conditions: free air in temperature 25 C, preheating temperature: lines.
65 ◦ C
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—for 3D conical model in
[MPa]
sequence: top diameter/bottom diameter/penetration depth.[MPa]
Figure 25. A view of (a) temperatures distribution during weld and (b) 3D conical heat source model.
After welding and cooling the sample, a thermal cycle was set for all external nodes of the model
corresponding to annealing at 400 and 650 °C, Figures 26 and 27. The calculated stresses distributions
were analyzed
Materials in the same way as before in three parallel lines drawn on the cross-section20ofof the
2020, 13, 5289 26
model, as shown in the Figure 28 with graphs of the stress distribution in individual lines.
[MPa] [MPa]
(a) (b)
Figure
Materials 26.Ax A
2020,26.
Figure 13, view
FOR of distributions
view
PEERofREVIEW of (a)ofresidual
distributions mean stresses
(a) residual and (b)and
mean stresses VonMises stresses after
(b) VonMises welding
stresses after
21 of 28
on the specimen
welding on the cross-section in the middle
specimen cross-section of model
in the middlelength.
of model length.
[MPa] [MPa]
[MPa] [MPa]
(a) (b)
(a) (b)
Figure 27. A view of the distribution of (a) residual mean stresses and (b) VonMises stresses after
Figure27.27.AAview
Figure viewofofthe
thedistribution
distributionofof(a)
(a)residual
residualmean
meanstresses
stressesand
and(b)
(b)VonMises
VonMisesstresses
stressesafter
after
annealing 400◦°C and 650◦°C on the specimen cross-section in the middle of the model length.
annealing 400 °C and 650 °C on the specimen cross-section in the middle of the model length.
annealing 400 C and 650 C on the specimen cross-section in the middle of the model length.
Figure 28.Graphs
Figure28. Graphs of
ofresidual
residualmean
meanstresses
stressesvalues
valuesin
inthe
thelaser
laserwelded
weldedbutt-joint,
butt-joint,depending
dependingononthe
the
Figure from
distance 28. Graphs
the of residual
weld axis for mean stresses
annealing values in
temperature the laser
equal to: welded
400, 550 and 650 ◦ C. depending on the
butt-joint,
distance from the weld axis for annealing temperature equal to: 400, 550 and 650 °C.
distance from the weld axis for annealing temperature equal to: 400, 550 and 650 °C.
4.4. An Example of a Numerical Analysis of a Hybrid Butt Joint Welding and Annealing Process
The last of the presented examples was the complex process of hybrid welding. As a combination
of two welding processes: laser and arc welding, the method allows for high efficiency on the one
hand and, on the other hand, significantly reduces the risk of high stresses, as is often the case with
laser welding. For this purpose, a discrete full three-dimensional model of a hybrid-welded butt joint
was prepared, consisting of 45,748 3D solid elements with 41,321 nodes. The mesh was concentrated
was prepared, consisting of 45,748 3D solid elements with 41,321 nodes. The mesh was concentrated
in the weld area to increase calculation accuracy, as can be seen in Figure 29. Boundary conditions
Materials 2020, 13, 5289 21 of 26

was prepared, consisting of 45,748 3D solid elements with 41,321 nodes. The mesh was concentrated in
the weld area to increase calculation accuracy, as can be seen in Figure 29. Boundary conditions related
to clamping conditions during welding were set to simulate welding without any additional fastening.
For boundary conditions corresponding to heat dissipation to the environment, it was assumed that
on each external surface, convective heat dissipation occurred to the environment at 20 ◦ C and the
Materials 2020,
radiation 13, x FOR PEER REVIEW
point. 22 of 28
Figure 29. A view of 3D model of butt joint hybrid welding and heat source models used for (a) laser
and (b) arc welding (MAG).
In the case of numerical analyses of the hybrid welding process, a double ellipsoid-shaped heat
source model to model the arc welding process and a 3D conical model for laser welding modeling
were used, as can be seen in Figure 29. The calculations, as in the previous case, were performed
using the “transient” technique. In order to calibrate the heat source models, an in-built “Heat Input
Fitting” module was used to optimize the shape of the resulting molten metal pool, Figure 30.
Additionally, it should be noted that, in order to be fully consistent with reality, the Goldak model
followed the cone model at a distance of 4.0 mm. The results of the calibration were presented in
Figure 30, Table 7.
Figure 29. A view of 3D model of butt joint hybrid welding and heat source models used for
Figure 29. A view of 3D model of butt joint hybrid welding and heat source models used for (a) laser
(a) laser and (b)7.arc
Table weldingparameters
Welding (MAG). used in a hybrid butt-joint welding simulations, Fig. 30.
and (b) arc welding (MAG).
In theDesignation
case of numerical analyses of the hybrid
EPUL (J/mm) weldingkprocess,
v (mm/s) HeataSource
doubleModel
ellipsoid-shaped
Dimensions heat
*
In the
source model case of numerical
to model analyses
the arc welding of the hybrid welding process, a double ellipsoid-shaped heat
3D conical (laser) 360process and a 3D conical 0.8 model for laser welding modeling
2.0/2.0/10.0
source model to model the 16.0
were used,
double as can be seen
ellipsoidal in arc
(MAG)
welding
Figure 29.490 process
The and a 3D
calculations, conical
as in model for
the previous
0.6
laser
case, were welding modeling
performed
9.0/8.0/2.0 using
were used, as can be seen in Figure 29. The calculations, as in the previous case, were performed
the “transient” technique.
cooling medium: In order to calibrate
free air the heat20
in temperature °C, preheating
source models, antemperature:
in-built “Heat °C Fitting”
65Input
using thewas
module “transient”
used to technique.
optimize Inshape
the orderof tothe
calibrate the molten
resulting heat source
metalmodels,
pool, an in-built
Figure 30. “Heat Input
Additionally,
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—for laser
Fitting”
it should module
be noted was used
that, in to
order optimize the shape
to be fully consistent of the resulting
with reality, molten
the Goldak metal
model pool,
followedFigure 30.
theforcone
model in sequence: top diameter/bottom diameter/penetration depth, as can be seen in Figure 24;
Additionally,
modelMAG it
at a distance should be noted
of 4.0 pool that, in order
mm.length/width/penetrationto be fully
The results of the calibration consistent
were with reality, the Goldak model
model: molten depth, as can presented in Figure
be seen in Figure 19. 30, Table 7.
followed the cone model at a distance of 4.0 mm. The results of the calibration were presented in
Figure 30, Table
[°C] 7.
Table 7. Welding parameters used in a hybrid butt-joint welding simulations, Fig. 30.
Designation EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

3D conical (laser) 360 0.8 2.0/2.0/10.0
16.0
double ellipsoidal (MAG) 490 0.6 9.0/8.0/2.0
cooling medium: free air in temperature 20 °C, preheating temperature: 65 °C
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—for laser
(a)
model in sequence: top diameter/bottom (b)be seen in Figure 24; for
diameter/penetration depth, as can
MAG model: molten pool length/width/penetration depth, as can be seen in Figure 19.
Figure
Figure30.30. AAview
viewofoftemperature
temperature fields
fields on on
the the cross-section
cross-section of hybrid
of hybrid weldedwelded buttfor:
butt joint joint for:
(a) laser
(a)beam
laser beam
and
[°C]
and
(b) arc (b) arc
welding welding
heat heat
source. source.
The obtained stresses distributions after the welding process and the sample cooling down
correspond to the heat distribution during welding and the maximum stresses values are
concentrated in HAZ, Figure 31. As occurred previously, numerical analyses of the annealing process
were carried out at temperatures of 400, 550 and 650 °C in accordance with the cycles shown in Figure
13b.
(a) (b)
Materials 2020, 13, 5289 22 of 26
Table 7. Welding parameters used in a hybrid butt-joint welding simulations, Figure 30.
Designation EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

3D conical (laser) 360 0.8 2.0/2.0/10.0
16.0
double ellipsoidal (MAG) 490 0.6 9.0/8.0/2.0
cooling medium: free air in temperature 20 ◦ C, preheating temperature: 65 ◦ C
Remarks: EPUL—energy per unit length, v—welding speed, k—energy efficiency factor; *—for laser model in
sequence: top diameter/bottom diameter/penetration depth, as can be seen in Figure 24; for MAG model: molten
pool length/width/penetration depth, as can be seen in Figure 19.
The obtained stresses distributions after the welding process and the sample cooling down
correspond to the heat distribution during welding and the maximum stresses values are concentrated
in HAZ, Figure 31. As occurred previously, numerical analyses of the annealing process were carried
out at temperatures
Materials of PEER
2020, 13, x FOR 400, 550 and 650 ◦ C in accordance with the cycles shown in Figure 13b. 23 of 28
REVIEW
[MPa] [MPa]
[MPa] [MPa]
(a) (b)
Figure
Figure AA
31.31. view
viewofof
distribution
distributionofof
(a)(a)
residual mean
residual meanstresses and
stresses (b)(b)
and VonMises stresses
VonMises after
stresses welding
after welding
onon
thethe
specimen cross-section in the middle of model length.
specimen cross-section in the middle of model length.
(a) (b)
Also,
Also, in in
thisthis
case, the the
case, calculated
calculatedstresses distributions
stresses in theinanalyzed
distributions joints behave
the analyzed as in theascases
joints behave in the
described
casesFigure earlier.
described Graphs
31. A view
earlier. of the stresses
of distribution
Graphs ofofthe values
residualin
(a) stresses three
mean
values parallel
stresses
in threeandlines, drawn
lines,in
(b) VonMises
parallel a cross-section
stresses
drawn after
in awelding on the
cross-section
surface, insurface,
on the
on the the middle
specimen of middle
the jointin
in cross-section
the ofand
the on
themiddletheand
joint bottom,
of model
on thealsobottom,
length.showedalso thatshowed
an increase
that inanthe annealing
increase in the
temperature causes a decrease in the stress value while maintaining
annealing temperature causes a decrease in the stress value while maintaining their character. their character. In the caseInofthe
caseAlso,
annealing atinthe
of annealing this case,
at thethe
temperature of 400 ◦ C, of
calculated
temperature stresses
only400a °C,distributions
slight
onlydecrease
a slightin in thevalue
the
decreaseanalyzed
in is jointswhich
visible,
the value isbehave in as
visible, the incase
which thein
cases described the earlier. Graphs byofanother
the stresses ◦ values
of increasing
the case temperature
of increasing the temperature by150 C
another 150in°Cthree
is more parallel
is more lines, drawn
noticeable—especially in abottom
at the
noticeable—especially cross-section
at theof the
bottom
on
weld the surface,
joint,
of the weldas can in
joint, the
beasseenmiddle
canin of
beFigures the
seen in32 joint
and 33.
Figures and on
32 and 33.the bottom, also showed that an increase in the
annealing
Annealingtemperature causes a decrease
at the temperature of 650inC,◦ thesimilar
stress [MPa]
value
to the while maintaining
previous analyzedtheir cases, character.
resultedIninthe a
[MPa]
case of annealing
decrease in the stressesat thevalues
temperature
in the weldof 400 °C,and
area only a slight
HAZ to thedecrease
maximum in the value
value is visible,
of 53 MPa and which
in thein
the caseparts
further of increasing
of the joint thepractically
temperature up tobyaanother
few MPa, 150as°Ccanis more
be seennoticeable—especially
in Figures 32b and 33. at the bottom
of the weld joint, as can be seen in Figures 32 and 33.
[MPa] [MPa]
(a) (b)
Figure 32. A view of residual mean stresses distributions after welding and annealing in: (a) 400 °C
and (b) 650 °C by 2 h.
(a) (b)
Annealing at the temperature of 650 °C, similar to the previous analyzed cases, resulted in a
Figure
Figure
decrease in 32.
32.
theAA view
view
stresses of residual
of residual
values in themean
mean weldstresses
stresses distributions
distributions
area and HAZafter after
welding
to the welding andofannealing
and annealing
maximum value in: MPa
53 (a) 400 °C in the
in:
and
◦ C and (b) 650 ◦ C by 2 h.
(a)
and400
(b) 650 °C by 2 h.
further parts of the joint practically up to a few MPa, as can be seen in Figures 32b and Figure 33.
Annealing at the temperature of 650 °C, similar to the previous analyzed cases, resulted in a
decrease in the stresses values in the weld area and HAZ to the maximum value of 53 MPa and in the
further parts of the joint practically up to a few MPa, as can be seen in Figures 32b and Figure 33.
Materials 2020, 13, 5289 23 of 26
FigureFigure 33. Graphs

33. Graphs of residual
of residual meanmean stressesvalues
stresses values in
in the
thehybrid
hybridwelded
welded butt-joint, depending
butt-joint, on theon the
depending
distance from the weld axis for annealing temperature equal to 400 °C,
◦ 550 °C
distance from the weld axis for annealing temperature equal to 400 C, 550 C and 650 C. and
◦ 650 °C. ◦
5. Conclusions
5. Conclusions
The use Theofuse of modern
modern construction
construction materials, which
materials, which often
oftenobtain
obtain their high
their strength
high properties
strength as
properties as
a result of complex thermo-mechanical treatment, often creates problems of a technological nature.
a result of complex thermo-mechanical treatment, often creates problems of a technological nature.
On the one hand, the thermal cycle of the welding process, and on the other hand, post-welding heat
On thetreatment
one hand, the thermal cycle of the welding process, and on the other hand, post-welding heat
used to reduce and homogenize the stresses distribution, often cause changes in the
treatment
precisely obtained, and
used to reduce homogenize
fine-grained the stresses
structure as well distribution,
as in strength often
andcause changes
plastic in theThe
properties. precisely
obtained,
temperature and duration of the post-weld heat treatment are usually both the cost of this operation and
fine-grained structure as well as in strength and plastic properties. The temperature
duration
andof thethe post-weld
impact heat of
on the recall treatment are usually
material properties. Theboth thedetermination
precise cost of this operation andisthe
of these values onlyimpact
on thesimple
recall for homogeneous
of material laboratory
properties. Thesamples.
preciseIndetermination
the case of real of constructions,
these values it is is
only virtually
simple for
impossiblelaboratory
homogeneous to do precisely.
samples.In such situations,
In the case ofnumerical analyses of it
real constructions, welding processes
is virtually come in to do
impossible
handy.
precisely. In such situations, numerical analyses of welding processes come in handy.
The test results showed that the essentially similar in nature stresses distributions in the tested
The test results showed that the essentially similar in nature stresses distributions in the tested
samples differ depending on the maximum cycle temperatures in the sizes of max. achieved tensile
samples differ
stresses evendepending
by about 250 onMPa theinmaximum
the case of thecycle
cycletemperatures
with temperature in 1360
the and
sizes of °C,
1080 max.Figuresachieved
6
tensileand
stresses evenalso
9. It should by be
aboutnoted 250
thatMPa
in theincase
theofcase
max.ofvalues
the cycle with temperature
of compressive stresses in the1360 and for
sample 1080 ◦ C,
Figures 6 and
each 9. tested
of the It should
cycles,also
thisbevalue
noted that inand
is similar theoscillates
case of max. values
at around of compressive
500–550 MPa, as can be stresses
seen in in the
sample for each of the tested cycles, this value is similar and oscillates at around 500–550 MPa,ofas can
Figures 6, 7 and 9. Such high values were caused by the process of grinding samples at the stage
be seentheir preparation.
in Figures 6, 7 Removal
and 9. Such of a small
high (0.2 mm were
values thick) caused
top layerby of the
the sample
processcaused a decrease
of grinding of this at the
samples
value to about 100 MPa, Figures 7 and 9. Of course, it should be mentioned that etching does not have
stage of their preparation. Removal of a small (0.2 mm thick) top layer of the sample caused a decrease
the same effect as annealing. During annealing, residual stresses in the entire cross-section of the
of this value to about 100 MPa, Figures 7 and 9. Of course, it should be mentioned that etching does not
sample are minimized, not only in the surface layer affected by machining. Only the affected surface
have the
layersame effect asbyannealing.
is removed etching, but During annealing,
the residual stressesresidual stresses in the entire
after thermomechanical cross-section
processing remain in of the
samplethe are minimized,
entire notof
cross section only in the surface layer affected by machining. Only the affected surface
the sample.
layer is removed
For the by etching,
cycles but the
with slow andresidual
controlled stresses after thermomechanical
(accelerated) cooling, some differences processing
are alsoremain
visible in the
in the stresses diagrams.
entire cross section of the sample. For a cycle with a maximum temperature of 1365 °C, it is visible that the
stresses in the case of forced cooling obtain values much lower, even by 150
For the cycles with slow and controlled (accelerated) cooling, some differences are also visible in MPa (approximately 400
MPa in relation to almost 500–550 MPa), as well as in the center of the
the stresses diagrams. For a cycle with a maximum temperature of 1365 ◦ C,sample, their significant
it is visible that the stresses
decrease in relation to for the case of slow cooling, as can be seen in Figures 6 and 8. The most likely
in the case of forced cooling obtain values much lower, even by 150 MPa (approximately 400 MPa in
relation to almost 500–550 MPa), as well as in the center of the sample, their significant decrease in
relation to for the case of slow cooling, as can be seen in Figures 6 and 8. The most likely reason for is a
longer time for compensation of residual stresses by plastic deformation and a slightly longer time for
stresses relaxation. The annealing tests carried out in the temperature range from 300 to 550 ◦ C showed
Materials 2020, 13, 5289 24 of 26
that the value of stresses in the workpiece decreased with the temperature increase, as can be seen in
Figure 10 and Table 3. In addition, on the basis of the research, it was also found that the extension
of the holding time at the given temperature of the heat treatment from 2 to 4 h causes a decrease in
the final value of the measured stresses from approx. 60 to 4–9 MPa for the annealing temperature
equal to 550 ◦ C, as can be seen in Table 3 and Figure 10. Transferring this experiment to a sample
subjected to the effect of the welding thermal cycle confirmed the decrease in the stress level to the
value of 250 MPa, which seems to be close to the previously created annealing diagram, as can be seen
in Figs. 10 and 11. However, it should be also noted that both the increase in the temperature of the
treatment and the extension of its duration cause a decrease in the tensile strength limit by an average
of 20–40 MPa with almost constant plastic properties, as can be seen in Table 5. It may therefore be
caused by a partial, slight stress relaxation at these temperatures, which, however, does not remove the
strengthening mechanisms of the material resulting from its production process.
Numerical analyses of the samples processed in the thermal simulator showed a high correlation
of the results with the actual tests. The differences between the actual and calculated values should
be explained on the one hand by the simplifications of the numerical model (defining of boundary
conditions, accuracy of the prepared material base, etc.), and on the other hand also by possible
measurement errors and disturbances, which may also occur in the case of real test. Nevertheless,
a significant convergence of the simulation results with reality is observed, which in the case of purely
engineering applications is more than acceptable, as can be seen in Figures 6, 10, 11 and 14–17.
The distributions and values of the calculated stresses were consistent with the results obtained for
samples subjected to electrolytic etching. This is understandable due to the fact that this operation has
not been modeled. However, it requires adopting an assumption in the methodology of conducting this
type of research that the measurements of stresses in real samples should be performed after removing
the effects of the stage of sample preparation for testing or possibly take into account the presence
of a thin layer of high compressive stresses. Based on the proven methodology, several examples of
numerical analyses carried out for various welding processes as well as various calculation techniques
used in modern programs for the analysis of welding and heat treatment processes are presented.
In the presented models, both full three-dimensional models and techniques with a very reduced time
needed to perform the calculations were used to answer the key question: whether and if the levels of
residual stresses for the selected heat treatment variant will be lowered. In each of the analyzed cases,
as a result of the thermal cycle of the annealing process, the value of the calculated stresses decreases,
and after exceeding the temperature of 650 ◦ C, the stress value was practically completely reduced to
values not exceeding 30–50 MPa, as can be seen in Figures 16 and 17.
The result of the GMAW multipass calculations were the stresses distributions, which show that
with the increase of the annealing temperature, the stresses level in the joint decreases, as can be
seen in Figure 23. When processed at the temperature of 650 ◦ C, the residual stresses values drop
practically to a level not exceeding 30 MPa. In addition, the use of the 2D cross section technique
when building the model allows for quick calculation of many variants that the user considers when
developing the necessary technology. In case of numerical analyses of the laser welding process,
the calculated stresses after welding reach their maximum values at the level of 544 MPa in the
HAZ area, as can be seen in Figure 26. These values are higher than those obtained in the case of
arc welding (a maximum of about 300 MPa) and characteristic for laser welding, and cover a much
narrower zone. Also, in this case, the applied heat treatment shows a decrease in the value of residual
stresses with increasing temperature of the applied annealing process. However, in this case, visible
changes are only observed after exceeding the annealing temperature of 550 ◦ C, as can be seen in
Figures 27 and 28. The results of the calculations carried out for hybrid welding were a kind of
confirmation of the above observations. The use of an arc heat source in hybrid welding “softens” the
impact of the acute thermal cycle of the laser beam, as shown by the obtained stress distributions on
the cross-section of the joint after the welding process, laser welding, though this only happens in
practice after exceeding the annealing temperature of 550 ◦ C, as can be seen in Figure 33.
Materials 2020, 13, 5289 25 of 26
To sum up, the comparison of the results of the tests and numerical analyses shows that the
increase in the annealing temperature in each of the analyzed cases causes a decrease in the level of
residual stresses and their distribution. This information should be considered during preparation of
the welding procedure and when creating a Welding Procedure Specification (WPS). It is also visible
in this case how useful the results can be of a correctly performed numerical analysis of the welding
process at this stage, which greatly facilitates the selection of appropriate conditions for carrying out
these types of operations. As mentioned earlier, this will be especially important in the case of large,
complex and responsible structures.
Author Contributions: Conceptualization, J.M. and M.Š.; methodology, T.K., J.M. and M.Š.; software, T.K.;
validation, T.K.; formal analysis, T.K.; investigation, T.K, J.M. and M.Š.; data curation, M.Š.; writing—original
draft preparation, T.K.; writing—review and editing, T.K, J.M. and M.Š.; visualization, M.Š.; supervision, T.K.
and J.M.; project administration, J.M.; funding acquisition, T.K. All authors have read and agreed to the published
version of the manuscript.
Funding: This publication was co-funded by the statutory grant of the Faculty of Mechanical Engineering at the
Silesian University of Technology in 2020.
Conflicts of Interest: The authors declare no conflict of interest
References
1. Chmielewski, T.; Golański, D. The role of welding in the remanufacturing process. Weld. Int. 2015, 29, 861–864.
[CrossRef]
2. Shipitsyn, S.Y.; Babaskin, Y.Z.; Kirchu, I.F.; Smolyakova, L.G.; Zolotar, N.Y. Microalloyed steel for railroad
wheels. Steel Transl. 2008, 38, 782–785. [CrossRef]
3. Żuk, M.; Górka, J.; Jamrozik, W. Simulated Heat-Affected Zone of Steel 4330V. Mater. Perform. Charact. 2019,
8, 606–613. [CrossRef]
4. Zhao, M.-C.; Yang, K.; Shan, Y. The effects of thermo-mechanical control process on microstructures and
mechanical properties of a commercial pipeline steel. Mater. Sci. Eng. A 2002, 335, 14–20. [CrossRef]
5. Górka, J. Microstructure and properties of the high-temperature (HAZ) of thermo-mechanically treated
S700MC high-yield-strength steel. Mater. Tech. 2016, 50, 616–621. [CrossRef]
6. Adamczyk, J.; Grajcar, A. Structure and mechanical properties of DP-type and TRIP-type sheets obtained
after the thermomechanical processing. J. Mater. Process. Technol. 2005, 267–274. [CrossRef]
7. Willms, R. High strength steel for steel constructions. In Proceedings of the Nordic Steel 2009, Malmö, Sweden,
2–4 September 2009; pp. 597–604.
8. Miki, C.; Homma, K.; Tominaga, T. High strength and high performance steels and their use in bridge
structures. J. Constr. Steel Res. 2002, 58, 3–20. [CrossRef]
9. Rakshe, B.; Patel, J. Modern high strength Nb-bearing structural steels, Forming processes. Available
online: http://millennium-steel.com/wp-content/uploads/articles/pdf/2010%20India/pp69-72%20MSI10.pdf
(accessed on 4 November 2020).
10. Wang, S.H.; Chiang, C.C.; Chan, L.I. Effect of initial microstructure on the creep behaviour of TMCP EH36
and S490MC steels. Mater. Sci. Eng. A 2003, 344, 288–295. [CrossRef]
11. Wang, G.R.; Lau, T.W.; Weatherly, G.C.; North, T.H. Weld thermal cycles and precipitation effects in
Ti-V-containing HSLA steels. Met. Mater. Trans. A 1989, 20, 2093–2100. [CrossRef]
12. Górka, J. Welding thermal cycle-triggered precipitation processes in steel S700MC subjected to the
thermo-mechanical control processing. Arch. Met. Mater. 2017, 62, 321–326. [CrossRef]
13. Górka, J. Assessment of the Weldability of T-Welded Joints in 10 mm Thick TMCP Steel Using Laser Beam.
Mater. 2018, 11, 1192. [CrossRef] [PubMed]
14. Shin, Y.; Kang, S.; Lee, H. Fracture characteristics of TMCP and QT steel weldments with respect to crack
length. Mater. Sci. Eng. A 2006, 434, 365–371. [CrossRef]
15. Park, K.S.; Cho, Y.H. Comparison of Fatigue Properties of Welded TMCP Steels and Normalized Steel;
Pohang University of Science and Technology: Pohang, Korea, 2003.
16. Yurioka, M. TMCP steel and their welding. Weld. World 1995, 35, 375–390.
17. De Meester, B. The Weldability of Modern Structural TMCP Steels. ISIJ Int. 1997, 37, 537–551. [CrossRef]
Materials 2020, 13, 5289 26 of 26
18. Porter, D.; Laukkanen, A.; Nevasmaa, P.; Rahka, K.; Wallin, K. Performance of TMCP steel with respect to
mechanical properties after cold forming and post-forming heat treatment. Int. J. Press. Vessel. Pip. 2004,
81, 867–877. [CrossRef]
19. Available online: https://www.salzgitter-flachstahl.de/en/products/hot-rolled-products/steel-grades/high-
strength-steels-for-cold-forming-thermomechanically-rolled.html (accessed on 4 November 2020).
20. Górka, J.; Janicki, D.; Fidali, M.; Jamrozik, W. Thermographic Assessment of the HAZ Properties and
Structure of Thermomechanically Treated Steel. Int. J. Thermophys. 2017, 38, 183. [CrossRef]
21. Kik, T.; Górka, J.; Kotarska, A.; Poloczek, T. Numerical Verification of Tests on the Influence of the Imposed
Thermal Cycles on the Structure and Properties of the S700MC Heat-Affected Zone. Met. 2020, 10, 974.
[CrossRef]
22. Lisiecki, A. Welding of Thermomechanically Rolled Fine-Grain Steel by Different Types of Lasers/ Spawanie
Stali Drobnoziarnistej Walcowanej Termomechanicznie Laserami Różnego Typu. Arch. Met. Mater. 2014,
59, 1625–1631. [CrossRef]
23. Fydrych, D.; Łabanowski, J.; Rogalski, G.; Haras, J.; Tomków, J.; Świerczyńska, A.; Jakóbczak, P.; Kostro, Ł.
Weldability of S500MC Steel in Underwater Conditions. Adv. Mater. Sci. 2014, 14, 37–45. [CrossRef]
24. Mičian, M.; Harmaniak, D.; Novy, F.; Winczek, J.; Moravec, J.; Trško, L. Effect of the t8/5 Cooling Time on the
Properties of S960MC Steel in the HAZ of Welded Joints Evaluated by Thermal Physical Simulation. Metals
2020, 10, 229. [CrossRef]
25. Skowrońska, B.; Chmielewski, T.; Golański, D.; Szulc, J. Weldability of S700MC steel welded with the hybrid
plasma. Manuf. Rev. 2020, 7, 4. [CrossRef]
26. Fydrych, D.; Łabanowski, J.; Rogalski, G. Weldability of high strength steels in wet welding conditions.
Pol. Marit. Res. 2013, 20, 67–73. [CrossRef]
27. Skowrońska, B.; Szulc, J.; Chmielewski, T.; Sałaciński, T.; Swiercz, R. Properties and microstructure of
hybrid Plasma+MAG welded joints of thermomechanically treated S700MC steel. In Proceedings of the
27th Anniversary International Conference on Metallurgy and Materials (METAL), Brno, Czech Republic,
23–25 May 2018.
28. Chang, K.-H.; Lee, C.-H.; Park, K.-T.; Um, T.-H. Experimental and numerical investigations on residual
stresses in a multi-pass butt-welded high strength SM570-TMCP steel plate. Int. J. Steel Struct. 2011,
11, 315–324. [CrossRef]
29. Danielewski, H.; Skrzypczyk, A. Steel Sheets Laser Lap Joint Welding—Process Analysis. Materials 2020,
13, 2258. [CrossRef]
30. Perić, M.; Nižetić, S.; Tonković, Z.; Garašić, I.; Horvat, I.; Boras, I. Numerical Simulation and Experimental
Investigation of Temperature and Residual Stress Distributions in a Circular Patch Welded Structure. Energies
2020, 13, 5423. [CrossRef]
31. Kik, T. Computational Techniques in Numerical Simulations of Arc and Laser Welding Processes. Materials
2020, 13, 608. [CrossRef]
32. Sysweld manual ESI Group. Welding Simulation User Guide; Sysweld manual ESI Group: Paris, France, 2016.
33. Kik, T.; Górka, J. Numerical Simulations of Laser and Hybrid S700MC T-Joint Welding. Materials 2019, 12, 516.
[CrossRef]
34. Kik, T.; Górka, J. Numerical Simulations of S700MC Laser and Hybrid Welding. In Laser Technology 2018:
Progress and Applications of Lasers, Proceedings of SPIE; Article Number: UNSP 109740K; SPIE: Bellingham,
WA, USA, 2018; Volume 10974. [CrossRef]
35. Kik, T. Heat Source Models in Numerical Simulations of Laser Welding. Materials 2020, 13, 2653. [CrossRef]
36. Kik, T.; Moravec, J.; Novakova, I. New Method of Processing Heat Treatment Experiments with Numerical
Simulation Support. Modern Technologies in Industrial Engineering V. In Proceedings of the ModTech 2017
International Conference, Sibiu, Romania, 14–17 June 2017.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional
affiliations.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

RESID. STRESS - Experiments and Numerical Simulations of The Annealing Temperature Influence in S700MC

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RESID. STRESS - Experiments and Numerical Simulations of The Annealing Temperature Influence in S700MC

Uploaded by

Copyright:

Available Formats

materials

Materials 2020, 13, 5289; doi:10.3390/ma13225289 www.mdpi.com/journal/materials

of thermomechanically rolled steels (TMCP—Thermo Mechanical Controlled Processing) have been

2. Description of the Problem

2.1. Material Used for Annealing Tests

Chemical Composition, % wt.

Table 2. Results of mechanical properties tests of used S700MC steel.

Figure 5. A view of a X-ray diffraction testing stand.

3.1.2. Samples with a Maximum Cycle Temperature at 1200 ◦ C

3.2. Influence of Annealing on the Change of the Magnitude of Residual Stresses

Sample Upper Yield Tensile

Materials 2020, 13, x FOR PEER REVIEW 16 of 28

Qr—as the heat energy density introduced in the rear part:

Materials 2020, 13, 5289

Bead Number EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

Materials 2020, 13, x FOR PEER REVIEW 20 of 28

heat transfer to the material structure depending on the given coordinates

Materials 2020, 13, x FOR PEER REVIEW 22 of 28

Designation EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

Designation EPUL (J/mm) v (mm/s) k Heat Source Model Dimensions *

Materials 2020, 13, x FOR PEER REVIEW 23 of 28

FigureFigure 33. Graphs

You might also like