Simulation of Residual Stresses in Hardfaced Component

Simulcia zvykovch napt tvrdench komponentov

Punitharani, K., Murugan, N., Sivagami, S.M
.

Abstract
Residual stress remain the single largest unknown in
industrial damage situations. With the continuing drive to
optimize material performance and minimize component
weight, there is an increasing emphasis on understanding
the role of residual stresses. This paper presents the results
of a study aimed at establishing temperature distribution,
distortion and residual stress field developed during plasma
arc hard facing of Stellite-6 over low alloy steel (AISI 4140)
gate valve by finite element analysis (FEA). In FEA, a three
dimensional solid model was generated and simulated using
ANSYS Parametric Design Language (APDL) code. After
simulation, residual stresses were compared and validated
by the stress measured by the experimental method viz., XRay Diffraction technique. Numerical results were compared with experimental test, which showed good agreement.
Keywords: Residual Stress, Plasma Transferred Arc
Welding, Material Property, Low alloy steel, Stellite-6

Abstrakt
Zvykov naptie je poslednou vekou neznmou v situcich priemyselnch pokoden. S pokraujcou snahou
optimalizcie materilovch vlastnost a minimalizovanm
vhy komponentov, sa stle v draz kladie na pochopenie lohy zvykovho naptia. lnok prezentuje vsledky
tdie zameranej na stanovenie pol teplotnej distribcie,
skrtenia a zvykovch napt vzniknutch procesom tvrdenia povrchu materilu Stellite-6 a nzkolegovanej ocele
(AISI 4140) dvernej zklopky plazovm oblkom, analzou
pomocou metdy konench prvkov (MKP). V MKP bol
vytvoren trojdimenzionlny model, na ktorom bola vykonan simulcia s pouitm kdu parametrickho kontruknho jazyka systmu ANSYS. Po simulcii boli porovnvan zvykov naptia a nsledne konfrontovan meranm
napt experimentlnymi metdami, menovite Rntgenovou
technikou difrakcie. Vsledky simulcie boli porovnan s
experimentlnym testom, o dokzalo vysok zhodu.
Kov slov: zvykov naptie, zvranie plazmou
prenesenm oblkom, vlastnosti materilu, nzkolegovan
oce, Stellite-6

1 Introduction
Development of residual stresses and distortion in a
welded component are strongly affected by many parameters and by their interactions. In particular, there are
structural, material and welding parameters. The structural
parameters include geometry of the plates, thickness, width
and the type of joint. Among the material parameters,
mechanical and physical properties at various temperatures
and the type of filler-metal are important parameters. Welding process parameters include type of process employed,
welding procedure: current, voltage, arc travel speed, feed
rate and nozzle to plate distance and also arc efficiency [1].
Hardfacing is one of the popular surfacing techniques employed to enhance the surface property of metal for specific
applications. It is a process of depositing a filler material on

the surface of carbon and low alloy steel base metal [2]. In
Petroleum Industrial applications, the gate valve made of
low alloy steel substrate (AISI 4140) is used extensively to
withstand high temperature and heavy load conditions. During opening and closing of the valve, metal to metal contact takes place on both the sides, leading to wear and tear
and oil leakage. This wear and tear reduces the service life
of the gate valve.
Among many welding processes, PTA hardfacing has
been increasingly employed due to the advantages such as a
higher deposition rate, minimum oxidation, relatively lower
dilution, improved metallurgical bond between the substrate
and finally the wide applicability of the material [3].
During hardfacing deposition by PTA process, the components are often subjected to high temperature and heavy
thermal load conditions. As a result of the non-uniform heating and cooling, a part of the material close to the weld is
subject to different rates of expansion and contraction causing development of three-dimensional complex residual
stresses [4,5]. To understand the formation of residual
stress, node temperature history during surfacing technique
and temperature dependent physical and mechanical
material properties must be known. In particular, during the
process of depositing Stellite-6 over low alloy steel substrate, mechanical material properties change drastically,
especially when the material approaches its melting point.
Therefore, due to the temperature dependent material properties and large deformation, material and geometrical
non-linearity have to be taken into account [9]. It has been
reported that the initial expansion of material due to temperature increase is constrained by material placed away
from the heat source, therefore generating compressive
stress at the centre of the weld pool [10]. Plate stiffness
affects strongly the magnitude and distribution of residual
stress, in particular, by the boundary conditions applied
during the hardfacing process. In some cases, the residual
stress may equal or exceed the yield stress of the base
material. The plastic strains resulting from the heating
causes stress, which in turn produces internal forces that
may cause buckling, bending and rotation. These deformations are called distortion [4].
In recent years, advanced numerical analysis has been
applied to resolve complex problems. The finite element
method is the conventional means of calculating welding
residual stresses. In 1971, the application of finite element
methods to analyze the welding residual stress problem has
been pioneered [6,7]. Bonifaz theoretically analyzed the
butt weld on the basis of the finite element method, while
considering the effects of changes with temperature in the
modulus of elasticity, yield stress and the coefficient of
linear thermal expansion [8].
The residual stress combined with distortion results in
premature failure during service. It is found from literature
survey that very few attempts have been made to analyse
residual stresses in gate valves. The present work aims to
predict the residual stresses in a hardfaced gate valve with

55

the help of finite element analysis, which is validated with

the measured values of residual stresses by X ray diffraction
technique.

2 Experimental Procedure
A 50.8 mm thick low alloy steel (AISI 4140) gate valve
having 3 mm groove on both sides, as shown in Fig.1. was
preheated to 450C for an hour before depositing Stellite-6
alloy in order to avoid moisture, cold crack, entrapped
hydrogen and to minimize the residual stress induced. In a
preheated gate valve, a 14 mm width and 5 mm height circular bead was deposited using hardfacing technique by
PTA process with 10 overlap simultaneously on both sides,
as shown in Fig. 2. The chemical composition and mechanical properties of valve body and hardfacing alloy are
given in Tab.1 and Tab.2 respectively. The process parameters used during hardfacing are given in Tab. 3.

Fig. 1 Gate Valve before hardfacing

Obr. 1 Dvern zklopka pred tvrdenm

Table 1 Chemical composition of gate valve and hardfacing

alloy
Tab. 1 Chemick zloenie dvernej zklopky a tvrdenej
zliatiny
Elements
C
Si
Mn
Cr
Mo
Co
Ni
P
S
V
W
Fe

% by weight
AISI 4140
Stellite-6
0.37
1.08
0.15
1.09
0.60
1.00
0.80
28.75
0.15
Bal
0.50
2.5
0.04
0.04
0.10
4.37
2.50

Table 2 Mechanical properties of gate valve and hardfacing

alloy
Tab. 2 Mechanick vlastnosti dvernej zklopky a tvrdenej
zliatiny
Properties
Tensile Strength (MPa)
Yield Stress (MPa)
Density (kg/m3)
Poissons Ratio

AISI 4140
1075
1172
7750
0.29

Materials
Stellite -6
758
655
8380
0.25

Table 3 PTA Hardfacing parameters values

Tab. 3 Hodnoty parametrov tvrdenia
Parameters
Welding current (Amps)
Voltage (Volts)
Welding speed (mm/min)
Powder feed rate (grams/min)
Plasma gas flow rate (lpm)
Oscillation frequency (cycles/min)
Pilot arc current (Amps)
Shielding gas flow rate (lpm)
Powder gas flow rate (lpm)
Nozzle diameter (mm)
Nozzle to plate distance (mm)
Circle overlap (degree)

Values
170
35.5
150
30
3.5
39
60
18
4
10
9
10

After hardfacing, the gate valve was kept in a electric

furnace, which was maintained at a temperature of 630C
for 2 hours in order to release the induced residual stresses
during hardfacing.

56

Fig. 2 Gate valve after circular bead deposition

Obr. 2 Dvern zklopka po rotanom zvarovom nnose

3 Analysis of Residual Stresses

Residual stresses can be introduced into engineering
components during manufacture as a result of eg., forming,
bending and welding processes [1]. They can also be caused
by the forces and thermal gradients imposed during surfacing operation. These stresses can affect load-carrying capacity and resistance to fracture the components. In order to
quantify their effect it is necessary to know their magnitude
and distribution [9].
Several techniques are available for measuring residual
stress. They include X-ray diffraction, neutron diffraction,
hole drilling, slicing and magnetic methods. However, a
popular method for predicting the residual stresses is X-ray
diffraction technique which is one of the non destructive
techniques capable of measuring 3D stress [10].
The residual stresses in the gate valve were estimated
using
(i) X Ray Diffraction technique and
(ii) Coupled field finite element analysis using ANSYS
10.0.

3.1 X Ray Diffraction Technique

After stress relieving the hardfaced gate valve, the oxide
coatings were removed using different grades of emery
sheet and electrochemical polishing was applied on the locations at which residual stress measurements were carried
out. Those locations are indicated in Fig. 3.
The residual stress was measured using X-ray stress
analyzer (Rigaku strainflex MSF-2M) having a back reflection type goniometer with a 2 scan range from 140C to
170C. The strain in the surface layer was estimated by

measuring peak shift due to deformation of the crystal

lattice using CuK radiation. Finally these strains were
converted into stresses by assuming linear elastic distortion
of the crystal lattice. The measured values are plotted and
shown in Fig. 5.

in Fig. 4. The hardfaced material was modeled above and

below the base metal with 5 mm thickness. The hardfacing
process simulated is a single pass plasma transferred arc
hardfacing. The finite element model in this study employs
the technique of element birth and death. All elements have
to be created, including those that are born in the later
stages of the analyses. In the method proposed in this study,
the elements were not removed to achieve the elements
deaths. Instead, the elements were deactivated by multiplying their stiffnesses by a large reduction factor. The cylindrical block absorbs a part of the heat generated.

3.2.2 Mesh creation

Fig. 3 Residual Stress Measurement Points

Obr. 3 Body merania zvykovch napt

Using 8-noded brick element the solid model was

meshed. Totally 55680 nodes and 53145 elements were generated in the present model. The accuracy of the finite
element method depends upon the density of the mesh used
for the analysis. The temperature around the arc is generally
higher than the melting point of the material, and it drops
sharply in regions away from the weld pool. Therefore in
order to obtain the correct temperature field in the region of
high temperature gradients it was necessary to have a more
refined mesh close to the weld line while in regions located
away from weld-line a coarse mesh was generated. Sensitivity analysis for mesh density has been performed and a
satisfactory mesh has been applied. The mesh used in the
stress analysis was identical to that in the thermal analysis.
Fig. 4 shows the final meshed model.

3.2.3 Load conditions

Fig. 4 Meshed Model of Finite Element

Obr. 4 Mriekov model konench prvkov

The thermal analysis has been performed by using temperature dependent thermal material properties for substrate
and hardfaced layer were considered, therefore non-linear
equations were solved, with all complexities related to their
solutions. The hardfacing parameters used for calculating
heat is displayed in Tab.3. Tab. 4 and Tab.5 show the
thermal and structural material properties used for both base
metal and hardfaced alloy. The temperature changes encountered in the heat-affected-zones are so large that the
change of thermal properties could not be neglected. To
determine the temperatures and other thermal quantities that
vary over time, there was a need to perform a transient
thermal analysis. Implicit method of time discretisation was
employed which allows for larger time steps. Here the time
step size is actually not a problem for calculation stability
but it determines the accuracy of the solution. Convection
losses are modeled at all free surfaces of the gate valve
component.
Table 4 Material Property - AISI 4140
Tab. 4 Vlastnosti materilu - AISI 4140

Fig. 5 Comparison of residual stress measured by XRD

technique and ANSYS
Obr. 5 Porovnanie zvykovch napt nameranch
rntgenovou technikou difrakcie a ANSYS-om

3.2 Finite Element Analysis

Finite element modelling and analysis using ANSYS
10.0 were performed to predict residual stresses and distortion within the hardfaced gate valve component.

3.2.1 Model Description

A cylindrical block of substrate with inner and outer
radius of 21 mm and 44.5 mm respectively and 50.8 mm
thick were considered to build the 3D solid model as shown

The FE analysis was carried out in two steps. A nonlinear transient thermal analysis was conducted first to
obtain the global temperature history generated during the
hardfacing process. The general purpose Finite Element
package ANSYS 10.0 was used for both thermal and stress
analysis performed sequentially.

57

Table 5 Material Property - Stellite 6

Tab. 5 Vlastnosti materilu - Stellite 6

In the thermal analysis, the heat was input in 370 load

steps, including 10 overlap on top side and the same procedure is repeated for bottom side also. As the number of
load steps is fairly very high, the programmic language
called ANSYS Parametric Design Language, coding was
employed to perform both thermal and structural analysis
easily. The nodal temperature solutions obtained from the
thermal analysis were read as loading into the stress analysis. In order to capture the residual stresses induced due to
the heating and cooling cycle, the temperature history needed to be read at a sufficiently large number of time points,
especially where the temperature gradient is large. However, the greater the number of the thermal solution steps
used, the greater is the computational time and the larger the
store space required.

5 Conclusion
Based on the results, the following conclusions were
made.
1. A generalized plane strain assumption yielded reasonnable stress distributions in both longitudinal and
transverse direction, as was verified by XRD experimenttal data.
2. The temperature near the circular bead and the HAZ
decreased rapidly with the distance from the center of the
heat source.
3. A very large tensile residual stress occurs on the surface
of circular bead, and a compressive stress appears below
the circular bead.
The maximum deviation between the measured and
predicted residual stresses is found to be 19.5%, due to the
assumptions made in the finite element analysis.
Punitharani. K1 , Murugan. N 2 and Sivagami. S.M3
1
Faculty of Mechanical Engineering,
PSG College of Technology, Coimbatore,
Tamil Nadu - 641 004, India,
Email address: punitharani@rediffmail.com
2
Faculty of Mechanical Engineering,
3
Research Scholar,
Coimbatore Institute of Technology, Coimbatore,
Tamil Nadu - 641014, India

3.2.4 Analysis of structural model

Solid 45 elements having 8 nodes and 3 degrees of
freedom per node were used for static structural model
analysis. With the available global temperature history from
the thermal analysis, and with proper constraints, structural
analysis has been carried out to predict the residual stresses
induced and thereby distortion in the hardfaced gate valve.

4 Results and Discussion

4.1 Residual stress measurement by XRD
technique
From the measured longitudinal residual stresses on the
surface of hardfaced gate valve using XRD technique
plotted in Fig.5, it is found that tensile stress reaches a
maximum value of 32 MPa at a distance of 2 mm from the
beginning of hardfaced metal and then reduces. It becomes
compressive residual stress as the distance increases. The
maximum compressive residual stress of 400 MPa is found
at a distance of 12 mm from the hardfaced metal.

4.2 Residual stresses analysis by Finite

Element Method
Fig.5 shows the longitudinal residual stress distribution
on hardfaced gate valve using coupled field finite element
analysis. It is clear from Fig.5 that the same trend is observed between the two methods with small variations due to
assumptions made in FE analysis. On the base metal side,
maximum tensile residual stress value of 28 MPa is found at
a distance of 2 mm and compressive stress of 322 MPa is
found at a distance of 12 mm and comparing this with allowable limits of yield stress of base metal, which is 1172
MPa, it is much low, which is due to the stress relieving of
gate valve after hardfacing. It is observed that, tensile stresses occur at the hardfaced surface and below that compressive stresses were noticed. Maximum stress could be noticed at the sharp edges and wherever the cross section
varies.

58

References
[1]

M M Mahapatra, G L Datta and B Pradhan: Three-dimensional finite

element analysis to predict the effects of shielded metal arc welding
process parameters on temperature distributions and weldment zones
in butt and one-sided fillet welds. IMechE 2006, 837-845.
[2] Q.X.Yang, M.Yao, J.K.Park: Numerical simulations and measurements of temperature and stress field in medium-high carbon steel
specimen after hard-face-welding. Computational Materials Science
29 (2004) 37-42.
[3] Garg AK.: Plasma Arc Welding and Surfacing Process. Welding
Research Institute, Department of Science and Technology, SERC
School on Advances in Welding Technology. 2002: 101 - 114.
[4] Michaleris P., DeBiccari. :Prediction of Welding Distortion. Welding
Journal, April 1997.
[5] Canas J, R Picon, F Paris, J Marin: Experimental and numerical
analysis of residual stress in welded AL-5083-O Aluminum plates.
Welding International 1994.
[6] Siva Prasad, Sankara Narayanan,.: Finite Element Analysis of
Temperature Distribution during Arc Welding using Adaptive grid
Technique. Welding Journal, April 1996.
[7] Murugan.S, Kumar.P.V, Gill.T.P.S, Raj.B, Bose. M.S.C.: Numerical
Modeling and Experimental Determination of Temperature Distribution During Manual Metal Arc Welding. Journal of Science and
Technology of Welding and Joining, Vol.4, No.6, March/1999, pp.
357-364.
[8] E.A.Bonifaz :Finite Element Analysis of Heat Flow in Single-Pass
Arc Welds. Welding Research Supplement, May 2000, 121-125.
[9] Peng-Hsiang Chang, Tso-Liang Teng: Numerical and experimental
investigations on the residual stresses of the butt-welded joints,
Computational Materials Science 29 (2004) 511522.
[10] Maeder G, Leburn JL and Sprauel JM,: Present Possibilities for the
X-ray Diffraction Method of Stress Measurements. NDT International 1981; 14 (5): 235 -248.