Welding in the World

Modelling of Temperature Field During Metal Inert Gas Welding Process Using Finite
Difference Method
--Manuscript Draft--

Manuscript Number: WITW-D-22-00148

Full Title: Modelling of Temperature Field During Metal Inert Gas Welding Process Using Finite
Difference Method

Article Type: Research Paper (unsolicited)

Corresponding Author: BENSADA MOUAD

Université Moulay Ismail de Meknes
Meknes, MOROCCO

Corresponding Author Secondary

Information:

Corresponding Author's Institution: Université Moulay Ismail de Meknes

Corresponding Author's Secondary

Institution:

First Author: BENSADA MOUAD

First Author Secondary Information:

Order of Authors: BENSADA MOUAD

A LAAZIZI

K FRI

Order of Authors Secondary Information:

Funding Information:

Abstract: Predicting temperature field in an arc welding process is essential and precondition for
any study of thermal or mechanical behavior of the weld. In fact, thermal, material
proprieties and physical phenomena have to be taken into account in order to predict
welding morphology, residual stresses and the final product quality of the weld. In the
case of analytical studies, simulations are done with many simplifications that affect
result precision. In this study, a 2D numerical model is developed by taking into
account heat transfer and distribution of heat source. Gaussian distribution for heat
source is presented and simulated by using Finite Difference Method (FDM) with
explicit scheme. This method permits following the evolution of temperature field
throughout the studied material during welding process. Also, operating parameters
effects such as current and voltage have been discussed.

Suggested Reviewers:

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript Click here to access/download;Manuscript;Manuscript
bensada mouad.docx
Click here to view linked References

1
2
3
4
5
6 Modelling of Temperature Field During Metal Inert Gas Welding Pro-
7
8
cess Using Finite Difference Method
9
10 M BENSADA1,*, A LAAZIZI1 and K FRI1
11 1
SECNDCM-L2MC, ENSAM- Moulay Ismail University, 50000 Meknes, Morocco
12
*Author for correspondence (mouad.bensada@edu.umi.ac.ma)
13
14
15
16 Abstract
17
18 Predicting temperature field in an arc welding process is essential and precondition for any study of thermal
19 or mechanical behavior of the weld. In fact, thermal, material proprieties and physical phenomena have to
be taken into account in order to predict welding morphology, residual stresses and the final product quality
20
of the weld. In the case of analytical studies, simulations are done with many simplifications that affect
21
result precision. In this study, a 2-dimensional numerical model is developed by taking into account heat
22
transfer and distribution of heat source. Gaussian distribution for heat source is presented and simulated
23 by using Finite Difference Method (FDM) with explicit scheme using the mathematical software,
24 MATLAB. This method permits following the evolution of temperature field throughout the studied mate-
25 rial during welding process. Also, operating parameters effects such as current and voltage have been dis-
26 cussed. Comparison with experimental work has been caried out, good correlation between numerical and
27 experimental results in terms of fusion zone penetration has been proved.
28
29 Keywords Numerical simulation; welding; heat transfer; explicit scheme.
30
31
32 1. Introduction
33
34 The use of welding as joining process replacing riveting permits to minimize the weight of structure either in the auto-
35 motive or the aeronautic industries [1]; it contributes also in reducing the production cost. In fact, arc welding process is
36 one of the most used processes in this kind of applications, which consist of heating of metals by using an electric arc to
37 melt and join separate pieces. Often a power supply creates an electric arc between a consumab le or non-consumable
38 electrode, the base material which may induce in some cases thermal effects due to the overheating and generate defects
39 that can lead to metal failure [2].
40
41 Therefore, predicting temperature field and controlling defect creation are important for any preventive study of the
42 welding process parameters [3]. In addition, temperature gradients are considered one of the main parameters that affects
43 the microstructure of the welded materials then their final mechanical proprieties. As a resul t, thermal proprieties and
44 physical phenomena have to be taken into consideration in order to predict welding morphology, residual stresses and
45
the final product quality of the welds [4]. Last decade, it has been challenging to simulate the welding behavi ors in
46
whole piece because of software requirements and long computational time [5]. Although many improvements of com-
47
48 putational background, it is now possible to examine welded joints and simulate temperature and strain in an adequately
49 accurate way.
50
51 The aim of the current research is to develop a 2D model of welding process by including heat source distribution and
52 the heat transfer. Finite difference method is used to solve numerical model. The effect of welding operating parameters
53 is discussed.
54
55 2. Metal Inert Gas Welding Process (MIG)
56
57 The study of the relationship between welding process parameters and final welding shape began in the mid-1900s,
58 regression analysis was applied to welding geometry research by Lee and Raveendra [6].
59
60 Many efforts have been made in the development of various algorithms in modeling the arc welding process [7,8].
61 1
62
63
64
65
 1
2 McGlone and Chadwick [9] reported a mathematical analysis correlating process variables and submerged arc welding
3 bead geometry.
4
5 Chandel [10] first applied this technique to the arc welding process and investigated the relationship between process
6 variables and the melted zone. These results showed that the arc current has the greatest influence on the melt ed zone
7 geometry, and that mathematical models can be used to accurately predict the weld bead geometry.
8
9
Nomenclature
10
11
12 Cp Specific heat (J/Kg.K) T(t,x,z) Temperature (°C)
13
14 e Thickness of workpiece(mm) To Initial temperature, 23 (°C)
15
hconv Heat transfer coefficient t Time (s)
16
17
I welding current (A) v Welding source velocity (m/s)
18
19 k Thermal conductivity (W/m.K) V Welding voltage (V)
20
21 L Length of workpiece x, z Cartesian coordinates (m)
22
23 m Gaussian mean σ Standard deviation
24
25 Qm Welding input source (W/m2) η Welding process efficiency
26
27 r radial distance of the Gaussian function  Density (Kg/m3 )
28
29
30
31 3. MIG Welding process modeling
32
33 The MIG welding process is based on the fusion of metals using intense heat generated by an electric arc between the
34 metals to be joined and a filler wire. The wire gradually enters into, and contributes in part to the formation of the m elted
35 pool. The arc and the melted pool are protected from contaminating agents present in the atmosphere by an inert gas.
36
37
38 Rosenthal [11] has been the first to develop an analytical model to calculate temperature distribution during welding by
39 using simplifying assumptions. Heat source was considered as a punctual source with a constant velocity. The calculated
40 temperature field was:
41 -vX
42 T=T0 + exp ( 2 K )ψ(x,y,z) (1)
43
44
45 With ‘T 0’ is the initial temperature, (x, y, z) is the coordinate of the punctual heat source,’v’ is the velocity of the heat
46 source and ψ (x, y, z) is a function to be defined.
47
48 In the case of the thin plate, one of the most known solutions is [12]:
49 q
-r 2
50 T=T0 + e(4πλρvct)1/2 exp ( 4Kt ) (2)
51
52
53 Where ‘e’ is the thickness of the plate, ‘q’ is the welding power and ‘t’ is time. The heat source and melted zone were
54 considered moving with a velocity ‘v’ and a length ‘l’ (Fig.1) [13].
55
56 With “r” is the distance between the considered point and the heat source:
57
58 𝑟 = √𝑥 2 + 𝑦 2 + 𝑧 2 (3)
59
60 Nunes [9] has proposed an extended Rosenthal weld model. The weld obtained by a heat sour ce of effective power ‘q’,
61 2
62
63
64
65
 1
2 the power supplied by the welding unit multiplied by a process efficiency factor, moving with constant velocity ‘v’ in
3 the x-direction on the surface (x-y plane) of a wide, long plate of thickness ‘e’, the temperature field ‘T’ r epresented
4 approximately as following:
5
6
7
T=T0 +q[F(n=0)+ ∑∞n=1(F(rn )+F(rn ' ))] (4)
8 Where ‘F’ is a function to be defined.
9
10 rn =√x 2 +y 2 +(z-2e)2 (5)
11
12 rn ' =√x 2 +y 2 +(z+2e)2 (6)
13
14
15 Therefore, modeling the heat source as a punctual source does not allow determining the maximal temperature near to
16 the heat source center, which limits the comparison between the experimental and the analytical study.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36 Fig. 1 Coordinates of the heat source moving with velocity v
37
38
3.1 Heat source modeling
39
40
41 In this study, the spatial distribution of the power welding input has been taken into account. The heat source is presented
42 as a Gaussian function distribution as centered normal law with a radial distance ’R’, standard deviation ‘σ=0.4’ and a
43 Gaussian mean =0. Fig.2 shows the used heat source distribution [14].
44
45 An axisymmetric 2D model is considered due to the piece geometry. The heat input is depending to the voltage ‘V’, the
46 current ‘I’ and the efficiency of the welding process ‘η’.
47 By assuming the energy distribution is Gaussian, the absorbed energy flux on the top surface (z=0) is given as:
48
VIη -R2
49 Q m = 2πσ2 exp (2σ2 ) (7)
50
51 In the case of MIG welding, the process efficiency "η" takes into account the heat absorption coefficient, since this
52 process is characterized by the presence of an inert gas which protects the melted zone.
53 In general, during the MIG process with an inert gas composed of 82% Ar and 18% CO2, the efficiency of the process
54 is equal to 80% [15,16].
55
56
57
58
59
60
61 3
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 Fig. 2 Gaussian heat source distribution
25
26
27 4. Heat transfer modeling
28
29 Nowadays, Steel is considered as one of the most preferable materials in automotive application due to low cost com-
30 paring to others metallic materials. A new generation of high strength steel is now used in the automotive field offering
31 a high capacity of fuel efficiency, enhanced safety a good formability characteristic.
32 Mild 140/270 steel is employed principally on the fuel tank of the car, the structure is assembled by arc welding process
33 including the arc welding technology, the main characteristic of the used materiel is presented in Tables 1 and 2.
34 In this numerical study, a 2D dimensional model is introduced as a plat with a thickness ‘e’ and width ‘L’.
35
36
37 Table 1. Material properties of mild steel
38 Steel grade Yield Ultimate ten- Total n-Value Density
39
strength sile
40 Elongation (5-15 %) (Kg/m3)
41 (MPa) strength(MPa)
42
43 Mild 140 270 38-44 0.23 7200
44 140/270
45
46
47 Table 2. Thermal characteristics of mild steel
48
49 Steel grade k(W/m.K) Cp (J/K.kg) hconv(W/m2.K)
50
51 Mild 25 500 40
52 140/270
53
54
55 The governing equations are presented as follow:
56
∂T ∂2 T ∂2 T
57 ρCp =k ( 2 + 2) (8)
58 ∂t ∂x ∂z

59 Cartesian geometry was used; the x–z plane was taken for the surface of the work piece that absorbs the heat generated
60
61 4
62
63
64
65
 1
2 by electric arc. The length and thickness of the studied piece were 5×5 mm2 respectively.
3 In order to enhance the accuracy of calculation, the finite difference method was used to solve numerically the mathe-
4 matical model (Eq. 9) and (Eq.8). The explicit scheme is applied. It builds up an iterative calculation in time. As a result,
5 the temperature is written as :
6
7 T(t,j-1,k)-2T(t,j,k)+T(t,j,k)
8 Δt
T(t+1,j,k)=T(t,j,k)+ [ Δx 2 ] (10)
9 𝜌𝐶𝑝 T(t,j,k-1)-2T(t,j,k)+T(t,j,k+1)
10 +
Δz 2
11
12 However, the explicit scheme has to respect the following relation (Eq. 11) in order to be stable. For this reason, the
13 choice of Δt is limited:
14
( + 2 ) ≤ 1⁄2
15 Δt 1 1
(11)
16 ρCp Δx 2 Δz
17
18 Although, the timescale required the use of fine grids and very small-time steps to obtain accurate results, that will lead
19 to longer calculation time. The space grid was taking as Δx=Δz=0.5 mm.
20
21
4.1 Boundary condition
22
23
24 The heat source of welding Q m has been introduced as a boundary condition (Table 3) in addition to the conviction
25 exchange with the air (Eq. 9).
26
27 In fact, the upper surface is in contact with air, it submitted to convective heat transfer as shown below:
28
∂T
29 -
∂z
=hconv (T-T0 )+Q m . (9)
30
31
32 For both lateral surfaces were considered at 23 °C. The model has been developed and simulated in MATALB environ-
33 ment.
34
35 5. Results and discussions
36
37 According to the flowchart illustrated in Fig.3, the temperature field in all points of the work piece was calculated at
38 every step of time by using finite difference method.
39
Using Matlab code, it was possible to calculate temperature field for different welding parameters, such as current and
40
voltage (Fig. 4, 5 and 6) during 10 s of arc welding process.
41
42 As a result, varying welding parameters, especially the welding power, affect the dimension of the melted zone. In fact,
43 Increasing the welding power, voltage and current intensity leads to the growth of the melted zone (Fig.7).
44
45
46 Table 3. Boundary conditions of the work piece
47
48 Position Boundary conditions
49
50
x=0 T=T0
51 x=L T=T0
52
53 z=0 𝜕𝑇
− = ℎ𝑐𝑜𝑛𝑣 (𝑇 − 𝑇0 ) + 𝑄𝑚
54 𝜕𝑧
55
56
57
58
59
60
61 5
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55 Fig. 3 Problem solving flowchart (Explicit Scheme).
56
57
58
59
60
61 6
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 Fig. 4 Temperature field after 10 s of weld- Fig. 5 Temperature field after 10 s of welding
25
ing process I=70A, U=20V. process I=50A, U=20V
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Fig. 6 Temperature field after 10 seconds of
51 welding process I=100A U=25V.
52
53
54
55
56
57
58
59
60
61 7
62
63
64
65
 1
2
3
4 8
7.2
5
6 7
7
5.8
8 6
9
10 Width
5
D, W (mm)

11 4
12 Depth
13 4
14
15 3
16
17 2
18 1.00
0.80
19 1
20 0.20
21 0
22
23 875 1875 3500
24 Welding Power (W)
25
26
27
28
29
30
31 Fig.7 Evolution of the melted zone dimensions versus welding power .
32
33
34 5.1 Comparison with experimental studies
35
36 In order to ensure of the accuracy of the developed model, it was necessary to compare with experimental results that
37 describe the effect of the welding parameters (Table 4), especially welding power, on melted zone morphology. Ka-
38 radeniz [17] has studied the effects of various welding parameters on welding penetration in Low Carbon steel welded
39 by robotic gas metal arc welding. The depths of penetration were measured for each specimen after the welding opera-
40 tions. They showed that increasing welding current and voltage increased the predicted melted zone morphology which
41 was in good agreement with the corresponding numerical results.
42 Fig.8 shows that the comparison with the experimental study reveals a slight difference in terms of the depth of penetra-
43 tion, this can be explained by the estimation of the process efficiency taken in the study. This coefficient depends on
44 several parameters such as the type of shielding gas and the material to be welded.
45
46
47 Table 4. Comparison between experimental and numerical welding penetration
48 Welding Arc Voltage (V) Experimental Numerical Penetration
49
current(A)
50 Penetration (mm) [11] (mm)
51
52 22 2.46 2.4
53
54 95 24 2.61 2.5
55 26 2.64 2.5
56
57
58
59
60
61 8
62
63
64
65
 1
2
3
4
5 Numerical penetration experimental penetration
6
7 2.8
8
9
10
11 2.7
12 2.64
13 2.61
14 2.6
Penetration(mm)

15
16
2.5 2.5
17
18 2.5 2.46
19
20 2.4
21 2.4
22
23
24
2.3
25
26
27
28 2.2
29 2090 2280 2470
30 Welding Power (W)
31
32
33
34
Fig.8 Comparison experimental and numerical value of penetration
35
36
37
6. Conclusion
38
39 A numerical model of arc welding process has been developed to analyze the effect of the welding parameters on the
40 temperature distributions and the shape of weld pool. Finite Difference Method with explicit scheme was used, including
41 a description of a gaussian heat source distribution.
42 Computational results showed that increasing voltage and intensity leads to the growth of the melted zone, which may
43 induce high residual stress and risks of metal failure. Comparison with experimental results was made in terms of welding
44 penetration. The behavior of measured widths of the melted zones was in good agreement with the model predictions.
45
46
47
48 7. Declarations
49
50
51 Conflict of interest The authors declare no competing interests.
52
53
54 References
55
56 [1] S. El mouhri, S. Ettaqi, A. Laazizi and S. Benayoun, FSW of aluminum alloys : Effect of operating parameters on the weld
57 quality, J. Key Engineering Materials, 640 (2015) 43-50. https://doi.org/10.4028/www.scientific.net/KEM.640.43
58 [2] Laazizi, B. Courant, F. Jacquemin and H. Andrzejewski, “Applied Multi-Pulsed Laser in Surface Treatment and Numerical–
59 Experimental Analysis,” J. Optics & Laser Technology 43, pp. 1257–1263, 2011. http://dx.doi.org/10.1016/j.optlas-
60 tec.2011.03.019
61 9
62
63
64
65
 1
2
3 [3] X. K. Zhu, Y. J. Chao. Effects of temperature-dependent material properties on welding simulation. J. Computers and Struc-
4 tures. 80 (2002) 967-976. https://doi.org/10.1016/S0045-7949(02)00040-8
5 [4] Y. Rong, J. Xu, Y. Huang, G. Zhang Review on finite element analysis of welding deformation and residual stress, J. Science
6 and Technology of Welding and Joining, 23:3, 198-208, (2018). DOI: 10.1080/13621718.2017.1361673
7 [5] J. Tao, T. C . Zhang, P.T. Liu, J. Li. Numerical Computation of a Linear Friction Welding Process. J. Materials Science Forum,
8 575–578, 811–815. 2008. DOI: 10.4028/www.scientific.net/msf.575-578.811
9 [6] J. Raveendra, R.S. Parmar, Mathematical Models To Predict Weld Bead Geometry For Flux Cored Arc Welding, Metal Con-
10 struction 19/2 (1987) 31-35
11 [7] Kim IS, Son JS, Park CE, Kim IJ, Kim HH (2005) “An investigation into an intelligent system for predicting bead geometry in
12 GMA welding process”. J Mater Process Technol 159:113–118. https://doi.org/10.1016/j.jmatprotec.2004.04.415.
13 [8] Modenesi P.J., Avelar R.C. de, The influence of small variations of wire characteristics on gas metal arc welding process
14 stability, Journal of Materials Processing Technology, 1999, Vol. 86, pp. 226-232.
15 [9] Yong-Rae Kim, Hyun-byung Chae and Jae-Woong K im : Welding Heat Source Modeling for Heat Flow Analysis of GTA
16 Overlay Welding, Journal of KWJS, 31-4 (2013), 62-66 (in Korean) https://doi.org/10.5781/KWJS.2013.31.4.62
17 [10] R. S. Chandel, “Mathematical Modeling of Gas Metal Arc Weld Features,” Proceedings of the 4th International Conference on
18 Modeling of Casting and Welding Processes,” Palm Coast, 17-22 April 1988, pp. 109-120.
19 [11] D. Rosenthal. Mathematical Theory of Heat Distribution during Welding and Cutting. J. Welding. 20, 220 -234. (1941)
20 [12] Poorhaydari K, Patchett BM, Ivey DG (2005) Estimation of cooling rate in the welding of plates with intermediate thickness.
Weld J 84(10):149–155
21
[13] Song S, Dong P, Pei X, A Full-Field Residual Stress Estimation Scheme For Fitness-For-Service Assessment Of Pipe Girth
22
Welds: Part I – Identification Of Key Parameters, International Journal of Pressure Vessels and Piping (2015), doi:
23
10.1016/j.ijpvp.2015.01.002
24
[14] A.C. Nunes, Jr. An extended Rosenthal weld model. J. Welding Journal. 1983.
25
[15] Pépe, N., Egerland, S., Colegrove, P. A., Yapp, D., Leonhartsberger, A., & Scotti, A. (2011). Measuring the process efficiency
26 of controlled gas metal arc welding processes. Science and Technology of Welding and Joining, 16(5), 412 –
27 417. doi:10.1179/1362171810y.000000002
28 [16] Pittner A., Weiss D., Schwenk C. and M. Rethmeier: Methodology to improve applicability of welding simulation, Science and
29 Technology of Welding and Joining, 2008, vol. 13, no. 6, pp. 496–508. https://doi.org/10.1179/136217108X329322
30 [17] E. Karadeniz, U. Ozsarac and C. Yildiz. The effect of process parameters on penetration in gas metal arc welding processes. J.
31 Materials and Design (28), 649–656 (2007). https://doi.org/10.1016/j.matdes.2005.07.014
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61 10
62
63
64
65