Lecture Notes in Mechanical Engineering

Banh Tien Long · Hyung Sun Kim ·

Kozo Ishizaki · Nguyen Duc Toan ·
Ivan A. Parinov · Yun-Hea Kim Editors

Proceedings
of the International
Conference on Advanced
Mechanical Engineering,
Automation, and Sustainable
Development 2021
(AMAS2021)
 Thermal Analysis in Wire Arc Additively
Manufactured SS308L Walls via Numerical
Simulations

Manh Cuong Bui, Van Thao Le(B) , Dinh Xuan Ta, Dinh Si Mai, Duc Manh Dinh,
Tat Khoa Doan, and Van Chau Tran

Le Quy Don Technical University, Hanoi, Vietnam

Abstract. Wire-arc additive manufacturing (WAAM) is recently adopted for the

fabrication of components with wide dimensions thanks to its high rate of depo-
sition and low devices’ costs. In this process, an electrical arc is utilized as the
heat source for melting the metallic wire, and the melted wire is directly added
into the parts in a layer-by-layer manner. One of the major challenges related
to this method is the heat accumulation in fabricated parts, which causes high
thermal distortions and residual stresses. The thermal variation also results in the
anisotropy of microstructures and mechanical properties of the built component.
Therefore, the thermal evolution analysis in WAAM is crucial. In this paper, the
thermal evolution in WAAMed thin-walled parts is analyzed through numerical
simulations. The effects of different depositing strategies on the thermal evolution
are also discussed. The results achieved in this work will be served important
evidences to interpret the microstructure formation and mechanical properties of
as-built components.

Keywords: Thermal analysis · Wire-arc additive manufacturing · Stainless

steel · Numerical simulation · Element finite method

1 Introduction
Additive manufacturing (AM) has emerged and developed giving a new prospect for pro-
ducing complex components without the requirement of cutting tools, molds and cooling
fluid as in conventional manufacturing processes. Particularly, metallic AM has recently
been applied to produce end-use components instead of the rapid prototyping purpose.
Among metallic AM techniques, wire and arc-based additive manufacturing (WAAM) is
considered as a promising method for the manufacture of large dimensional components
with medium shape complexity [1]. The main advantages of such a technique consist of
a high rate of deposition and low devices’ costs. In addition, using metallic wires as the
input materials is a cheaper and safer approach than metallic powder [2, 3]. However,
WAAM also introduces several inherent limitations such as high heat accumulation, high
thermal distortion, and residual stresses, as well as low dimension accuracy and poor

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022

B. T. Long et al. (Eds.): AMAS 2021, LNME, pp. 10–15, 2022.
https://doi.org/10.1007/978-3-030-99666-6_2
 Thermal Analysis in Wire Arc Additively Manufactured SS308L Walls 11

surface quality. It is also accepted that complex thermal histories in the WAAM pro-
cess significantly influence the microstructure evolution of components [1]. Therefore,
a comprehensive understanding on the thermal evolution during the WAAM process is
essential.
In this paper, the thermal evolution in WAAMed thin-walled parts is analyzed through
numerical simulations. For this aim, the finite element model (FEM) of thin-walled parts
is firstly built (Sect. 2). Subsequently, the effects of different depositing strategies on the
thermal evolution are also discussed (Sect. 3). Section 4 is intended for some conclusions
of this research work.

2 Finite Element Method-Based Simulation

2.1 Materials and the Geometries of WAAM Parts
In this article, the manufacture of thin-walled parts was studied. The GM-308L stainless
steel wire with a diameter of 1 mm and two SS308L substrates with dimensions of
200 × 100 × 10 mm (length × width × thickness) were used to build the walls. Two
thin-walled parts of twenty layers (Wall A and Wall B) were fabricated by a welding
robot (TA14000 Panasonic), using the mono-direction deposition method [2]. An idle
time of 40 s was applied between two successive deposits for the cooling, and a shielding
gas (99.99% argon) with a flow rate of 15 L/min was utilized during the deposition. The
process parameters used to build the walls and their actual dimensions are shown in
Table 1.

Table 1. Processing parameters used to build the walls and the walls’ geometries.

Wall Processing parameter Actual geometry

Wall A Welding current: I = 120 A Average wall width = 7.80 mm
Arc voltage: U = 22 V Average layer thickness = 3.34 mm
Traveling speed: v = 0.2 m/min
Wall B Welding current: I = 120 A Average wall width = 6.02 mm
Arc voltage: U = 22 V Average layer thickness = 1.84 mm
Traveling speed: v = 0.5 m/min

In this study, the thermal physical properties of SS308L (i.e., thermal conductiv-
ity, density and specific heat) used for finite element-based simulations were calculated
based on the SS308L chemical composition [2, 3] by using JMatPro software. Moreover,
in WAAM, the interchange of heat takes place between built walls, the substrates and the
environmental surrounding. As a result, the convection and the radiation can be mod-
eled by using the coefficients of convection and emissivity, and the Stefan-Boltzmann
constant. The emissivity and convection coefficients were fixed to 0.9 and 30 W/m2 .K,
respectively, for the top surface of the substrate and deposited layers, while the convection
coefficient was set at 300 W/m2 .K for the other substrate surfaces. The Stefan-Boltzmann
constant of 5.67 × 10–8 W/m2 .K was adopted [4].
12 M. C. Bui et al.

2.2 Modeling of the Heat Input

To model the heat input in the WAAM process, the double ellipsoidal power density
distribution model of Goldak [5] was adopted. Accordingly, the heat flux for the points
in the first semi-ellipsoid located in front of the weld arc is expressed by Eq. (1). On
the other hand, the heat flux for the points in the second semi-ellipsoid covered the rear
section of the arc is depicted by Eq. (2).
√
6 3ff Q 3x2 3y2 3z 2
qf (x, y, z) = √ exp(− 2 − 2 − 2 ) (1)
af bcπ π af b c
√
6 3fr Q 3x2 3y2 3z 2
qr (x, y, z) = √ exp(− 2 − 2 − 2 ) (2)
ar bcπ π ar b c

Herein af , ar , b and c are geometric parameters; f f and f r are the heat input proportions
in the front and rear parts, f f + f r = 2. The values of these parameters reported in [4]
were adopted. The heat input was computed as Q = ηUI, where U is the arc voltage, I
is the welding current (Table 1), and η is the arc efficiency, η = 0.8 [6].
The simulation of the WAAM process was performed by using LS-DYNA software
and element deactivation/activation technique. Before the simulation, the all elements
of the deposited layers are deactivated because they do not exist before the depositing
process. The deactivated elements are activated sequentially following the movement
of the welding torch with the traveling speed. The keyword in LS-DYNA was used for
heat source modeling of WAAM process is “*BOUNDARY THERMAL WELD TRA-
JECTORY”. The thermal analysis of WAAM process was carried out using the material
model “*MAT_THERMAL_CWM T07”. The 3D models of the walls composed of 20
layers were constructed based on the geometries of the substrates and the built walls, as
mentioned in Table 1.

Fig. 1. The mesh model and the temperature simulation result.

For the WAAM process modeling, linear brick elements with eight nodes are used
for the thermal simulation. To capture the thermal performance around current position
of welding torch, a refined mesh were used for the deposited layers and the surrounding
 Thermal Analysis in Wire Arc Additively Manufactured SS308L Walls 13

areas of the substrate that is close to the wall. The meshes became coarser in the direc-
tions far away from the wall. The entire FE model (the substrate and deposited layers)
comprised 24042 elements and 29981 nodes. Figure 1 shows the mesh model and the
thermal simulation results.

3 Results and Discussion

To analyze the thermal evolution characterization at a point in the built walls, three
thermal cycle curves at the middle point of the layers L2, L10 and L18 were extracted
(Fig. 2). Generally, a point in the layer j undergoes (20 − j + 1) times of heat treatments
and the thermal curve having (20 − j + 1) thermal peaks. For example, the thermal cycle
curve at the middle point of the layer L2 exhibits nineteen thermal peaks. Similarly,
there are eleven and three temperature peaks appearing in the thermal cycle curves of
the middle points of L10 and L18, respectively.
As the arc source goes to a point, the temperature increases up to the peak value,
and subsequently it gradually decreases with the displacement away of the arc source
(Fig. 2). If the value of a temperature peak exceeds the melting point at a point, it will be
melted under the effect of the arc source. As a result, the first peak in the thermal cycle
curve “Wall B-L2” depicts the formation of the melting pool. On the other hand, the
temperature value of other peaks is not superior to the melting point, leading to a series of
heat treatments to the layer L2, which do not result in melting. The thermal cycle process
at other locations occurs in a similar manner (as shown in the thermal curve “Wall B-L10”
and “Wall B-L18”, Fig. 2). Moreover, it is found that the values of thermal peaks increase
with an increase in the wall height, indicating the heat accumulation phenomenon during
the WAAM process. In this process, the previous layer has an effect of preheating on
the next layer, while the current layer has an effect of reheating the previous layers.

Fig. 2. Thermal evolution at the middle point in different layers of the walls.
14 M. C. Bui et al.

Based on the thermal cycle curves, the cooling rate at one point of the built wall
can also be estimated. For example, the average cooling rates in the first, second and
third peaks of the “Wall B-L2” curve are 95 °C/s, 51 °C/s, and 27 °C/s, respectively,
whereas those values for the first, second and third peaks in the “Wall B-L10” and “Wall
B-L18” curves are {66, 30, 16 °C/s} and {61, 27, 14 °C/s}, respectively. This is due to
the heat accumulation in the WAAM process, resulting in the increase in the substrate
temperature and the decrease in the average cooling rates. Therefore, it can be concluded
that the cooling rate decreases with the increase in the wall height.
Additionally, it is observed that the curve “Wall B-Li” is under the curve “Wall A-
Li” with i = 2, 10 and 18. In particular, the temperature of the first peak of the curve
“Wall A-Li” is significantly higher than that of the curve “Wall B-Li”. For example,
the temperature of the first peak in the “Wall A-L2” is 2274 °C, whereas that value in
the “Wall B-L2” is 1800 °C. As a result, the heat accumulation in the wall A is more
significant than that in the wall B. The average cooling rate in the first peak in the curves
“Wall A-Li” are 62, 37 and 32 °C/s for i = 2, 10 and 18, respectively, which are lower
than those in the first peaks in the curve “Wall B-Li” (95, 66 and 61 °C/s for L2, L10
and L18, respectively). Hence, it is deduced that the wall B built with a higher traveling
speed features a higher cooling rate in all regions as compared to that in the wall A. This
finding allows us to predict that the wall B will exhibit finer microstructures.

4 Conclusions
In this paper, the thermal evolution in the SS308L wall during the WAAM process was
investigated based on the FE simulations. The main conclusions of this research are
summarized as follows:

• During the WAAM process, the wall passes through a complex thermal cycle. A point
in the layer j features a thermal cycle curve with (N − j + 1) of temperature peaks,
where N is the number of deposited layers. The heat accumulation in the wall appears
when the depositing height increases, and the cooling rate decreases along the wall
height.
• The wall built with a higher traveling speed exhibits a higher cooling rate than that in
the wall built with a lower traveling speed.
• The prediction of the thermal evolution in this work is very important to interpret the
microstructure formation and mechanical properties of as-built components.

Acknowledgements. This research is funded by Vietnam National Foundation for Science and
Technology Development (NAFOSTED) under grant number 107.99-2019.18.

References
1. Jafari, D., Vaneker, T.H.J., Gibson, I.: Wire and arc additive manufacturing: opportunities and
challenges to control the quality and accuracy of manufactured parts. Mater. Des. 269, 109471
(2021). https://doi.org/10.1016/j.matdes.2021.109471
 Thermal Analysis in Wire Arc Additively Manufactured SS308L Walls 15

2. Le, V.T., Mai, D.S., Paris, H.: Influences of the compressed dry air-based active cooling on exter-
nal and internal qualities of wire-arc additive manufactured thin-walled SS308L components.
J. Manuf. Process. 62, 18–27 (2021). https://doi.org/10.1016/j.jmapro.2020.11.046
3. Le, V.T., Mai, D.S., Doan, T.K., Paris, H.: Wire and arc additive manufacturing of 308L
stainless steel components: optimization of processing parameters and material properties.
Eng. Sci. Technol. Int. J. 24, 1015–1026 (2021). https://doi.org/10.1016/j.jestch.2021.01.009
4. Lee, S.H.: CMT-based wire arc additive manufacturing using 316L stainless steel: effect of
heat accumulation on the multi-layer deposits. Metals (Basel) 10, 278 (2020). https://doi.org/
10.3390/met10020278
5. Goldak, J., Chakravarti, A., Bibby, M.: A new finite element model for welding heat sources.
Metall. Trans. B 15, 299–305 (1984). https://doi.org/10.1007/BF02667333
6. Quintino, L., Liskevich, O., Vilarinho, L., Scotti, A.: Heat input in full penetration welds in
gas metal arc welding (GMAW). Int. J. Adv. Manuf. Technol. 68(9–12), 2833–2840 (2013).
https://doi.org/10.1007/s00170-013-4862-8