D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
775–782
NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE
EFFECT OF HYDROSTATIC PRESSURE ON THE RESIDUAL
STRESS IN BOILER-TUBE WELDS
NUMERI^NA IN EKSPERIMENTALNA PREISKAVA VPLIVA
HIDROSTATI^NEGA TLAKA NA ZAOSTALE NAPETOSTI V
ZVARU NA KOTLOVSKI CEVI
Daryoush Danyali, Eslam Ranjbarnodeh
Amir Kabir of University of Technology, Department of Mining and Metallurgical Engineering, Tehran, Iran
islam_ranjbar@yahoo.com, islam_ranjbar@aut.ac.ir
Prejem rokopisa – received: 2015-08-13; sprejem za objavo – accepted for publication: 2015-10-09
doi:10.17222/mit.2015.255
A tube to tube-sheet weld joint is a critical section in many boilers. Leakage and failure are two common problems that can
occur with this type of joint. The tensile residual stresses associated with welding can play a major role in these problems. In the
present study, the Finite-Element Method is used to predict the residual stresses in a tube to tube-sheet weld and the effect of
hydrostatic pressure on the residual stresses’ redistribution. Investigations were performed by numerical analyses and
experimental methods. The joint included a circumferential fillet weld in one pass using a gas tungsten arc welding process. The
thermomechanical behavior of the joint is simulated with a two-dimensional axisymmetric model and a subroutine developed in
ANSYS software. The thermography method is used for the thermal verification and the hole-drilling strain gauge under
post-weld heat treatment is used for the mechanical analysis verification. The numerical and experimental results showed that
applied hydrostatic pressure can reduce, by about 58 %, the axial residual stress.
Keywords: hydrostatic test, residual stress, boiler tubes, tube-sheet, finite element method (FEM)
Kriti~ni podro~ji v kotlih sta zvarjen spoj cevi in cevne predelne stene. V tej vrsti spoja sta glavna problema, ki se pojavljata,
pu{~anje in poru{itev. Zaostale natezne napetosti povezane z varjenjem so glavni dejavnik pri teh problemih. V {tudiji je bila
uporabljena metoda kon~nih elementov za napovedovanje zaostalih napetosti v zvarih cevi in cevne predelne stene in vpliv
hidrostati~nega tlaka na razporeditev zaostale napetosti. Raziskave so bile izvedene z numeri~no analizo in z eksperimentalnimi
metodami. Spoj je vklju~eval obodni zvar izdelan v obloku z volframovo elektrodo in za{~ito s plinom. Termomehansko
obna{anje spoja je bilo simulirano z uporabo dvodimenzijskega osnosimetri~nega modela in programa razvitega z ANSYS
programsko opremo. Termografska metoda je uporabljena za preverjanje toplote in pri toplotni obdelavi zvarov je bila
uporabljena metoda z merilnimi listi~i na izvrtini za preverjanje mehanske analize. Numeri~ni in eksperimentalni rezultati so
pokazali, da uporabljen hidrostati~ni tlak, lahko za 58 % zmanj{a osne zaostale napetosti.
Klju~ne besede: hidrostati~ni preizkus, zaostala napetost, kotlovske cevi, cevna predelna stena, metoda kon~nih elementov
(FEM)
1 INTRODUCTION
Boilers are a very important part of many industries,
such as power generation, food, hospitals, and hotels.
Boilers are the most troublesome components of elec-
trical power generation plants. It costs the US utility
industry over $5 billion per year in unscheduled shut-
downs, repairs and power replacements.1 In general, the
safety and structural integrity of a boiler are of para-
mount importance. Welding is common method in the
boiler-manufacturing process and the process of welding
has a direct influence on the integrity of the components
and their thermal and mechanical behaviors during
service. Micro-cracking in tube to tube-sheet welds, in
both the radial and circumferential directions, is
commonplace.2
The nature of failures in the vicinity of the tube to
tube-sheet interfaces is particularly difficult to determine
because of the large temperature gradients as well as the
relatively high pressures. Due to the high temperatures
introduced during welding and the subsequent cooling of
the welded metal, welding can produce undesirable
residual stresses and deformations. Radial cracking is
more likely to occur, but circumferential cracking also
occurs and results from the thermal cycling, which is
exacerbated by a high residual stress.3 The thermal
stress, mechanical stress and welding residual stress of
tube to tube-sheet has received a lot of attention in recent
years.
B. Yildrim and H. F. Nied4 investigated the residual
stress and distortion in boiler-tube panels with welded
overlay cladding. X. Qingren5 reported that residual
stress in the circumferential weld in pipes is reduced when
hydrostatic pressure is increased. To the best of the
knowledge of the author, a very limited number of FEM
models have been proposed for predicting the residual
stresses in a tube to tube-sheet weld and the effect of
hydrostatic pressure test on the residual stresses in this
type of joint. The present study uses a two dimensional
Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782 775
UDK 519.6:620.179:67.017 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(5)775(2016)
axisymmetric FEM model for the calculation and
distribution of the residual stresses in a tube to
tube-sheet joint. The Thermography Method is used for
the thermal verification and the hole-drilling strain gauge
under post-weld heat treatment is used for the
mechanical analysis verification and then the effect of
hydrostatic pressure on the redistribution residual
stresses is investigated by numerical analyses and experi-
mental methods.
2 FEM AND VERIFICATION PROCEDURES
2.1 Welding verification procedures
To verify the results of the FEM method, an
experimental test with the same geometry, material and
welding parameters was produced, as shown in Figure 1.
2.2 Material properties
The base metal of the tube was steel St35.8 and
tube-sheet was steel 17Mn4. The chemical compositions
are listed in Table 1.
Table 1: Chemical composition in mass fractions (w/%)
Tabela 1: Kemijska sestava v masnih odstotkih (w/%)
Composition C Si Mn P S Cr
St35.8 0.17 0.35 0.60 0.03 0.03 –
17Mn4 0.20 0.20 0.120 0.03 0.025 0.30
For the thermal and mechanical analyses, the rela-
tionship between the thermo-physical and the thermo-
mechanical properties of the materials with temperature
is incorporated, as shown in Figure 2.
The welding parameters used were exactly as same as
the boiler-repair parameters. The dimensions of the
welded parts are presented in Table 2.
Table 2: The dimensions of the pieces used
Tabela 2: Dimenzije uporabljenih kosov
Material Diameter(mm)
Width
(mm)
Width
(mm)
Thickness
(mm)
Tube 51.8 – 200 3.2
Tube-sheet – 80 80 12
2.3 Thermal verification procedure
To verify the thermal model, thermography was used.
In this study, the welding temperature was monitored
with a model T8 thermography camera. It helps to record
the temperature during the welding and after the weld
cooling.
2.4 Mechanical model verification procedure
To verify the weld profile obtained using the FEM
method, a section was prepared for macroscopic exami-
nation of the welded samples. The section was prepared,
polished and etched in a solution with 50 mL of 37 %
hydrochloric acid and 50 ml of distilled water for 30
minutes. After the etching period it was examined
macroscopically.
2.5 Residual stress verification procedure
The hole-drilling strain gauge under post-weld heat
treatment is based on the principle that if a hole in a
piece is created, stresses around the hole and the hole
will be released. When this amount increases, the mobi-
lity of the hole will increase. The strain and thus the
amount of residual stresses can be calculated by deter-
mining the displacement of the hole. This procedure is
based on Tait’s research.3
The amounts of axial residual stresses were produced
by the drilling under post-weld heat treatment technique.
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
776 Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782
Figure 1: Schematic of geometrical model for verification
Slika 1: Shematski prikaz geometrijskega modela za preverjanje
Figure 2: Thermo-physical and mechanical properties
Slika 2: Termofizikalne in mehanske lastnosti
The samples were drilled using a TOSKURIM drilling
machine.
The high-speed steel drill was 2 mm in diameter,
known as the drill chuck. The rotation speed was
2743 min–1 and the forward speed of the drill was 0.05
mm/min. After drilling, the distances between the holes
were measured and recorded with a MITUTOYO
micrometer type with an accuracy of up to 0.02 mm.
After measuring the distance between the holes, the
samples were heat treated according to the guidelines for
heat treatment.5
The applied heating rate was 220 °C/h. The welded
parts were held at 620 °C for an hour and then they were
allowed to cool with a cooling rate of about 275 °C/h,
cooled in furnace to 300 °C and after that in air. Figure 3
shows the applied heat-treatment cycle for the welded
parts in this study.6
After that the samples were cooled, the distances
between the holes were measured and recorded again in
the same manner using the same micrometer. The resi-
dual stresses in the specimens are released by drilling the
specimens, and the axial strain, ex, and transverse re-
leased strain, ey, are measured. Using the measured
strains, the axial residual stress, x, can be obtained from
Equation (1):7
x = [–E/(1 – v
2)](ex + vey) (1)
where E is Young’s modulus and v is Poisson’s ratio.
2.6 Hydrostatic test verification procedure
Before applying hydrostatic pressure, the distances
between holes were measured and recorded again. The
drilling apparatus and procedure were the same as used
in the calculation of the residual stresses. Figure 4
shows a hydrostatic test setup and Figure 5 shows the
sample drilled to investigate the behavior of residual
stresses after the hydrostatic pressure.
To create the real conditions, like hydrostatic test
conditions, the welded part was restrained between the
fixed jaws of a clamp. The fittings for the pressure-test-
ing machine were installed and water with a temperature
of about 15 °C was injected into the welded part. When
the air bubbles exit completely, a pressure equal to 1.5
times the boiler’s design pressure was applied to the
welded part.6 Due to the design the pressure of the
investigated boiler in this study was 0.4 MPa, while the
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782 777
Figure 3: Heat-treatment cycle
Slika 3: Potek toplotne obdelave
Figure 5: The sample drilled for axial residual stress measurement
and hydrostatic tests
Slika 5: Zvrtan vzorec za merjenje aksialnih zaostalih napetosti in
hidrostati~ni preizkus
Figure 4: Hydrostatic test setup
Slika 4: Sestav za hidrostati~ni preizkus
Figure 6: Applied hydrostatic pressure as a function of time
Slika 6: Uporabljen hidrostati~ni tlak v odvisnosti od ~asa
applied hydrostatic pressure to the welded part was about
0.6 MPa.
Figure 6 shows the applied hydrostatic pressure as a
function of time. After 1800 s, the hydrostatic pressure
was removed. After the completion of the hydrostatic
testing procedure, the distances between the drilling
locations were measured and recorded with the same
micrometer.
2.7 FEM meshed model
The FEM model of the welding process is performed
on the St35.8 steel tube with an outer diameter of 51 mm
and a wall thickness of 3 mm, and 17Mn4 steel tube-
sheet with a wall thickness of 12 mm using a single pass.
There was no gap between the tube and tube-sheet. The
welding process used was gas tungsten arc welding. For
a more convenient calculation, a part of the tube-sheet
was selected for the FEM analysis. Although the real
welding is a three-dimensional procedure, it is often
considered sufficient to represent a circumferential weld
with a two-dimensional axisymmetric FEM model.7 The
two-dimensional axisymmetric FEM model is much
faster and easier to perform.8,9 Therefore, the methodo-
logy described here is based on a two-dimensional
axisymmetric model.
The meshed model is shown in Figure 7. For the
meshing, plane55 and Surf151 elements were used for
thermal analysis and the plane82 element was used for
the mechanical analysis. In total, 346 nodes and 303
elements were generated. The welding of the tube to
tube-sheet was performed with a single pass weld. In the
weld adjacent, the meshing was very fine due to the high
thermal gradient of this region during the welding
process.
2.8 Thermal analysis
The heat equation is a parabolic partial differential
equation that describes the distribution of heat (or vari-
ations in temperature) in a given region over the time and
is generally describes by Fourier’s relation, Equation
(2):6
c
t x
K
T
x y
K
T
y z
Kx y zp
d
d
d
d
∂
∂
∂
∂
∂
∂
∂
∂
⎛
⎝
⎜ ⎞
⎠
⎟ = ⎡⎣⎢
⎤
⎦⎥+
⎡
⎣⎢
⎤
⎦⎥
+
d
d
T
z
Q
⎡
⎣⎢
⎤
⎦⎥+ (2)
where  is the density (kg/m3), Cp is the specific heat
capacity (J/kg K), K (W/m K) is the thermal conduc-
tivity, Q is the heat input (W), and T is the temperature
(K).
The Gaussian function is calculated by following
Equation (3):6,10
q r
Q
r
r
r
( ) exp=
⎛
⎝
⎜
⎞
⎠
⎟ −
⎛
⎝
⎜
⎞
⎠
⎟
⎡
⎣⎢
⎤
⎦⎥
3
3
0
2
0
2π
(3)
where q(r) is the heat flux (W/m2), r0 is the radius of the
welding heat source that is estimated to be about one
half of the Gaussian curve width, r is the distance from
the middle point of the heat source to the center, and Q
is the heat input (W) that is calculated using Equation
(4):6
Q = V·I· (4)
where V is the welding voltage, I is the welding current,
and ç is the welding efficiency. In this paper V, I,  and
r0 were 10 V, 120 A, 0.85 and 10 mm, respectively. The
filler metal, welding speed and gas flow rate were
ER70S-6, 5cm/min and 15 L/min, respectively.
The fluid flow in the welding pool increases the
heat-transfer rate, which has an important effect on the
temperature field in the welding process. The effect of
fluid flow on the weld pool temperature for the whole
temperature field is considered by the heat-transfer
coefficient, h. In this study, as in a previous study, h = 15
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
778 Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782
Figure 8: The applied thermal boundary conditions
Slika 8: Uporabljeni toplotni robni pogoji
Figure 7: FEM meshed model
Slika 7: Model mre`e za FEM
W/K m2.11 Due to the small thickness of the parts,
preheating before welding is not done and the initial
temperature of the welded parts was assumed to be
27 °C. Figure 8 shows the applied thermal boundary
conditions.
2.9 Mechanical analysis
The residual stress was calculated by using the tem-
perature distribution obtained from the thermal analysis
as input data. The material properties relevant to the re-
sidual stress are the Young’s modulus, the yield strength,
the Poisson’s ratio, and the coefficient of thermal
expansion. The total strain can be decomposed into three
components, as in Equation (5):11
thotal = e + p + th (5)
where e, p, and t are the elastic strain, plastic strain
and thermal strain, respectively. The elastic strain was
modeled using the isotropic Hooke’s law with a tempe-
rature-dependent Young’s modulus and Poisson’s ratio.
During the mechanical analysis, boundary conditions
were applied to prevent the rigid-body motion. Figure 9
shows the applied mechanical boundary conditions.
Because of the tube-sheet rigidity, BC-X is constrained
in the X-direction and BC-Y is constrained in the
Y-direction. The S.L lines are axisymmetric lines.
3 RESULTS AND DISCUSSION
3.1 Thermal results
The thermal contour recorded by the experimental
method is shown in Figure 10 and the thermal results
recorded by experiment and the FEM are shown in
Figure 11. The weld-pool peak temperatures of the FEM
model and the experiments are 1900 °C and 1716 °C,
respectively. This 10 % difference shows the good agree-
ment between the experimental results and the FEM
model. This difference could be due to changes in
thermo-physical properties of the materials as a function
of the temperature, the difference in the thermal con-
ductivity. In the same way, Majzobi and colleagues con-
ducted a study in which a thermography technique was
used to record the thermal history.12 The thermal history
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782 779
Figure 10: Contours of the thermal history
Slika 10: Plastnice termi~ne zgodovine
Figure 9: The mechanical boundary conditions
Slika 9: Mehanski robni pogoji
Figure 12: Temperature contours for the FEM model during welding
Slika 12: Razporeditev temperature pri FEM modelu med varjenjem
Figure 11: Curves of the thermal history
Slika 11: Krivulji toplotne zgodovine
of the contours recorded by the other researches also
shows a difference of about 11 %, which confirms the
results of the present study.
The weld profile that is created with the FEM is
shown in Figure 12. The peak of the weld’s penetration
is 1.06 mm into the tube. The weld profile that is created
using the experimental method is shown in Figure 13.
The peak of the weld penetration is 1.2 mm into the tube.
Comparing these two sets of results shows that great-
est difference in the amount of weld penetration into the
tube-sheet is about 12 %. The results are proved to be in
a logical agreement for the FE method and experimental
results.
3.2 Mechanical behavior under hydrostatic effect
The residual stresses contour in the axial direction is
shown in Figure 14. The axial residual stress calculated
by the FEM and experimental methods is shown in Fig-
ure 15. The peak axial residual stress of 105 MPa is
located in the weld root zone. The peak axial residual
stress calculated by the experimental method was 85.94
MPa. The axial residual stresses first decrease from the
peak value, and then start to increase. All the axial resi-
dual stresses decrease along the weld and the outer
surface of the tube. The axial residual stress values are
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
780 Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782
Figure 17: Residual stresses in axial direction using the FEM and
experimental methods, after the hydrostatic test
Slika 17: Zaostale napetosti v osni smeri po FEM in po eksperimen-
talni metodi, po hidrostati~nem preizkusu
Figure 14: Residual stresses results in axial direction using the FEM
method, before the hydrostatic test
Slika 14: Zaostale napetosti v osni smeri po FEM metodi, pred hidro-
stati~nim preizkusom
Figure 13: The weld profile using the experimental method
Slika 13: Profil zvara pri preizkusu
Figure 15: Residual stresses results in axial direction using the FEM
method, before the hydrostatic test
Slika 15: Zaostale napetosti v smeri osi, po FEM metodi, pred hidro-
stati~nim preizkusom
Figure 16: Residual stresses in axial direction using the FEM and
experimental methods, after the hydrostatic test
Slika 16: Zaostale napetosti v smeri osi po FEM in eksperimentalni
metodi, po hidrostati~nem preizkusu
reasonable, compared to Sattari-Far’s research re-
sults.13,14 The maximum difference between the two
methods is 18 % and the lowest is 2 %. As compared to
Tait’s results2, the findings of the present study are
acceptable. The peak axial tensile residual stress is 105
MPa in the root of the joint, which caused the initiation
and growth of fatigue cracks under cyclic loading in this
boiler joint.
In order to investigate the effect of the hydrostatic
test on the residual stress in the axial direction in a
boiler-tube weld, hydrostatic pressure is applied using
the FEM and experimental methods. Figure 16 shows
the curve of the residual stress results in the axial direc-
tion using the FEM and experimental methods, before
the hydrostatic test. It is clear that the hydrostatic pres-
sure affects the residual stress in the axial direction, as
shown in Figure 17. The maximum difference between
the two methods is about 9 %. The results of the FE
method show a peak in the residual stress in the axial
direction after the application of the hydrostatic pressure
is 52.4 MPa. The results of the experimental method
show that the peak residual stresses in the axial direction
after the application of hydrostatic pressure is 47.7 MPa.
The residual stresses in the axial direction calculated
by the FEM method in the weld root are shown in Fig-
ure 18. The residual stresses in the axial direction cal-
culated by the FEM method in weld root show that the
application of a hydrostatic test pressure after the
welding process can reduce the tensile residual stresses
in the axial direction from 105 MPa to of 52.4 MPa, i.e.,
by approximately 51 %. The results of the residual
stresses in the axial direction calculated using the FEM
method in the weld toe shows that hydrostatic pressure
can reduce the tensile residual stresses in the axial
direction from 40.8 MPa to 20.4 MPa, i.e., a reduction of
about 50 %. The residual stresses in the axial direction
calculated using the experimental method in the weld
shows that applying a hydrostatic test pressure after the
welding process can reduce the tensile residual stresses
in the axial direction from 20.74 MPa to 8.71 MPa. This
means a reduction of about 58 %. The reason for this
reduction can be sought in the accumulation of the
hydrostatic pressure and the previous residual stresses
that were caused by plastic deformation. In the plasti-
cally deformed region, the total stresses exceeded the
yield strength and partial relaxation of the residual
stresses can be occurred.
4 CONCLUSIONS
This study predicted the axial residual stresses and
their behavior under hydrostatic pressure using the FEM
and experimental methods. The results of this study can
be summarized as follows:
1. The root of tube to tube-sheet joint and the weld toe
were two critical points with the maximum axial
tensile residual stress.
2. Applying a hydrostatic pressure after the welding
process could reduce the axial tensile residual
stresses in the weld by about 58 %.
3. Applying a hydrostatic pressure after the welding
process could reduce the axial tensile residual
stresses in the root of tube to tube-sheet joint by
about 50 %.
4. Applying hydrostatic pressure after the welding
process could reduce the axial tensile residual
stresses in the weld toe by about 51 %.
5. It was easier and cheaper to calculate the residual
stresses in the design of a complex geometry in
connection with a corner joint with a very low
thickness using the technique of strain measurement
accuracy of the hole under the heat treatment.
6. The results of the present study could be generalized
for other critical boiler joints, such as the shell to
furnace joint and the furnaces to tube-sheet joint.
7. The results of the present study could be generalized
to other high-pressure equipment including heat
exchangers, hot-water boilers, hot-oil boilers and any
other equipment that has a tube to tube-sheet joint.
5 REFERENCES
1 X. Shugen, W. Weiqiang, Numerical investigation on weld residual
stresses in tube to tube sheet joint of a heat exchanger, Int. J. Pres.
Ves. Pip., 101 (2013), 37–44, doi:10.1016/j.ijpvp.2012.10.004
2 R. Tait, J. Press, Investigation An experimental study of the residual
stresses, and their alleviation, in tube to tube-sheet welds of
industrial boilers, Engineering Failure Analysis, 8 (2001), 15–27,
doi:10.1016/j.engfailanal.2013.04.01
3 K. Abouswa, F. Elshawesh, A. Abuargoub, Stress corrosion cracking
(caustic embrittlement) of super heater tubes, Desalination, 222
(2008), 682-688, doi:10.1016/j.desal.2007.02.073
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782 781
Figure 18: Residual stresses in the axial direction using the FEM and
experimental methods, before and after the hydrostatic test
Slika 18: Zaostale napetosti v osni smeri po FEM in po eksperimen-
talni metodi, pred in po hidrostati~nem preizkusu
4 B. Yildrim, H. F. Nied, Residual Stress and Distortion in Boiler Tube
Panels with Welded Overlay Cladding, Journal of Pressure Vessel
Technology – Transactions of ASME, 126 (2004), 426–431,
doi:10.2320/matertrans.126.426
5 X. Qingren, F. Yaorong, H. Chun-Yong, The measurement and
control of residual Stress in spiral submerged arc welded pipe,
Proceedings of 4th International pipeline conference, Calgary, 2002,
615–622
6 British Standards Institute, BS2790, Design and manufacturing of
shell boilers of welded construction, 1992, 80–100
7 C. Lee, K. Chang, Numerical investigation of residual stresses in
Strength mismatched dissimilar steel butt welds, J. Strain Analysis,
43 (2008), 55–66, doi:10.1243/03093247JSA313
8 T. Soanes, W. Bell, A. Vibert, Optimizing residual stresses at a repair
in a steam header to tube plate weld, Int. J. Pres. Ves. Pip., 82 (2005),
311–318, doi:10.1016/j.ijpvp.2004.08.009
9 D. Deng, H. Murakawa, Numerical simulation of temperature field
and residual stress in multi-pass welds in stainless steel pipe and
comparison with experimental measurements, Comp. Mater. Sci., 37
(2006), 269–277, doi:10.1016/j.commatsci.2005.07.007
10 X. Shugen, Z. Yanling, Using FEM to determine the thermo
mechanical stress in tube to tube–sheet joint for the SCC failure
analysis, Engineering Failure Analysis, 34 (2013), 24–34,
doi:10.1016/j.engfailanal.2013.07.01
11 T. Tso-Liang, F. Chin-Ping, C. Peng-Hsiang, Y. Wei-Chun, Analysis
of residual stresses and distortion in T-joint fillet welds, Int. J.
Pres.Ves. Pip., 78 (2001), 523–538, doi:10.1016/j.ijpvp.2000.07.008
12 J. Otegui, P. Fazzini, Failure analysis of tube–tubesheet welds in
cracked gas heat exchangers, Engineering Failure Analysis, 11
(2004), 903–913, doi:10.1016/j.engfailanal.2004.01.003
13 I. Sattari-Far, Y. Javadi, Influences of welding sequence on welding
distortion in pipes, Int. J. Pres. Ves. Pip., 85 (2008), 265–274,
doi:10.1016/j.engfailanal.2007.02.004
14 I. Sattari-Far, M. Farahani, Effect of the weld groove shape and pass
number on residual stresses in butt-welded pipes, Int. J. Pres. Ves.
Pip., 86 (2009), 769–777, doi:10.1016/j.ijpvp.2009.07.007
D. DANYALI, E. RANJBARNODEH: NUMERICAL AND EXPERIMENTAL INVESTIGATION OF THE EFFECT ...
782 Materiali in tehnologije / Materials and technology 50 (2016) 5, 775–782