Effect of external magnetic field on weld pool flow conditions in high-speed gas metal arc welding

Abstract

In high-speed gas metal arc welding process, weld defects like humping bead and undercutting appear and the weld bead quality worsens when welding speed exceeds a critical value. The previous investigations have proved that the backward flowing molten jet with high momentum in the weld pool is responsible for the formation of humping bead in high-speed gas metal arc welding. In this study, an external electromagnetic field generator is developed to change and control the flow condition in the weld pool and suppress the occurrence tendency of the humping bead. Bead-on-plate welding experiments are performed on mild steel Q235 test-pieces, and the influences of the exerted magnetic strength on the arc behavior, the fluid flow inside the weld pool, the weld bead shape and the weld penetration and width are investigated. The reason why the external electromagnetic field can suppress the humping bead is explained. The results show that external magnetic field can remarkably adjust the weld pool fluid flow field and decrease the momentum of the backward flowing molten jet. Therefore, weld defects like humping bead and undercutting can be effectively suppressed, the quality of weld bead is remarkably improved, and the critical welding speed is notably increased.

Keywords

External magnetic field fluid flow in weld pool high-speed gas metal arc welding humping bead

Introduction

Arc welding processes are still the most widely used materials joining technologies in manufacturing industry.^1–4 To improve the competitiveness in today’s industry environment, manufacturing enterprises must consistently seek for higher welding speed to achieve higher productivity.^1,5 However, the weld bead quality is worsened, and severe geometric defects such as humping bead begin to occur if the welding speed is increased to some extent.^6,7 That is to say, there exists a critical welding speed beyond which humping bead will deteriorate the weld quality in gas metal arc welding (GMAW) and gas tungsten arc welding (GTAW). The welds with humps cannot satisfy the requirements of manufacturing industry, and effective measures must be taken to avoid or suppress the occurrence of humping bead. In recent years, special welding equipment and processes have been developed to obtain high-quality welds at high welding speed, such as tandem or triple wire GMAW or GTAW and hybrid welding.^8,9 Although these modified welding processes are able to increase the welding productivity without sacrificing the weld quality, they have some drawbacks. For example, some rely on special shielding gas mixtures and expensive equipment, and others add difficulties in operation and process optimization. Since GMAW is one of the most economical, efficient and versatile fusion welding processes,¹⁰ it is of great significance to develop low-cost and easy-operation high-speed GMAW.

Experimental observations have demonstrated that there is a strong backward flow of molten metal inside the weld pool during high-speed GMAW.^6,7 The high momentum of the backward flow of molten metal results in the formation of humping bead. High-speed GMAW usually uses high level of welding current to maintain the required heat input. Because the arc pressure is proportional to the square of the welding current, a severely depressed region is created in the front of the weld pool by the combined actions of the arc force and the momentum of the droplets from the electrode. The molten metal layer beneath the arc is very thin, and the molten metal is pushed to the back of the weld pool where it accumulates and grows into a swelling. After solidification, humped bead forms.^5–7,10 To suppress humping bead, it is crucial to reduce the momentum of the backward molten metal flow in the weld pool. Some modified GMAW used two wires to expand the action area of the arc force so that the backward molten metal flow in the weld pool is weakened.^11–13 But two welding torches make the equipment and operation more complex. There is current flowing through the molten metal in the weld pool. If an external magnetic field is exerted into the molten metal inside the weld pool, the generated electromagnetic force can change the flow velocity of molten metal, and the momentum of the backward molten metal flow in the weld pool may be decreased by such a simple and practical way.

Magnetic control methods were proposed around 1960s¹⁴ and were implemented in arc welding to control the arc and the weld pool behaviors.^9,14,15 Li’s group used external magnetic field to improve the weld quality in resistance spot welding of advanced high-strength steels, and promising results were obtained.^16–19 For arc welding, though external magnetic field has been employed in adjusting the arc behavior,^9,14 stirring the weld pool to refine grain size²⁰ and supporting the weld pool,^21,22 it has not been tried to control and change the fluid flow in weld pool to realize high-speed GMAW. Kern et al.²³ used magneto-fluid dynamic mechanism in laser welding to change flow conditions in weld pool so as to stabilize them and achieve greater welding speed. However, the weld pool in laser welding is very small in volume, and there is no current passing through the laser weld pool. For high-speed GMAW, the weld pool is much larger in volume, and there is current flowing inside the weld pool. The interaction between the current in the weld pool and the external magnetic field is more complicated, and the adjustment effect of the electromagnetic force on the flow conditions in the weld pool may be more powerful.

In this study, a self-developed magnetic field generator is applied to control the backward flow jet of molten metal inside the weld pool in high-speed GMAW. Both the arc and weld pool are observed by a vision sensor under the conditions with and without actions of external magnetic field. It is found that exerted electromagnetic force is able to change and control the relative position between the arc axis and the weld pool, the direction of the droplet transfer and the flow status of the molten metal in weld pool, so that the momentum of the backward flow of molten metal is reduced in the weld pool. Consequently, the humping bead is dramatically suppressed.

Experimental setup

As shown in Figure 1, the experimental system consists of a control computer, a GMAW power source, a work-table, a magnetizer (magnetic field generator), a high-speed vision sensor and a data acquisition subsystem. The magnetizer includes the two coils (excitation winding) and an excitation power source. Each coil has 270 turns. The wire of diameter 2.3 mm is winded around the electrical iron bar with diameter 20 mm and length 145 mm. The distance between the centers of two coils is 38 mm. The excitation current is 0–8 A, and the corresponding magnetic flux density is 0–80 mT. The two coils are assembled under the work-table and located in the vertical plane passing the welding torch. There is a 5-mm distance from the bottom surface of the backing plate to the top end of the coils. During the welding process, both the torch and the coils are stationary, while the workpiece travels along the welding direction. For generating the electromagnetic force toward the front edge of weld pool, the external magnetic field should be parallel to the welded plate but perpendicular to the welding line.

Figure 1.

Schematic of experimental system.

During the welding process, the behaviors of the arc and the weld pool are observed by a mono high-speed camera equipped with filter in the visible light region. The sampling frequency is about 836 frames per second (fps). The camera aims at the target from one side of the work-table, that is, the axis of the camera is perpendicular to the welding direction, and the viewing angle is about 45° with respect to the workpiece surface.

The workpieces are mild steel Q235 plates of 3 mm thickness, and the filler wire is steel wire H08Mn2SiA. The experimental conditions are listed in Table 1. The bead-on-plate welds are made under different conditions.

Table 1.

Experimental conditions.

Welding current I_A	300 A	Length of welds	170 mm
Arc voltage E	30 V	Filler wire	H08Mn2SiA
Excitation current I_E	0–8 A		Φ1.2 mm
Magnetic flux density B	0–80 mT	Shielding gas	Ar + 8%CO₂
Welding speed V	2.01 m/min		20 L/min
Contact tube-to-work distance	17 mm

Electromagnetic control of flow condition in weld pool

Figure 2 shows the action mechanism of external electromagnetic force applied to high-speed GMAW process. The electromagnetic force, current flux density and magnetic flux density are represented by F, I and B, respectively. Because of the N pole and S pole fixed on both sides of the welding line, the external magnetic field is coplanar with the workpiece surface and perpendicular to the welding line. The current in the weld pool consists of the downward component I_Z- , the backward component I_X- and the forward component I_X . Under the interaction of these current components with the external magnetic field, the forward electromagnetic force F_X , downward electromagnetic force F_Z- and upward electromagnetic force F_Z are generated, respectively. Since the forward electromagnetic force F_X is toward the front edge of the weld pool, it will reduce the momentum of the backward molten flow and suppress its backward flow velocity and accumulation at the pool rear. If the magnetic strength is large enough, the force F_X will push the molten metal to the front part of weld pool so that molten metal may accumulate there. Since F_Z is upward electromagnetic force, it will promote the layer of molten metal to get thicker in the forepart of weld pool beneath the arc. On the contrary, the downward force F_Z- contributes to a thinner layer of molten metal at the tail of the weld pool.

Figure 2.

Action principle of external electromagnetic field (⊗: the magnetic field toward y-axis).

Before the arc ignites, the magnetic flux density is localized on the workpiece and is just about 1 mT in the air between the wire end and the workpiece, because mild steel has the magnitude of magnetic permittivity more than 1000 times higher than that of the air at ambient temperature. However, once the arc ignites, much heat reaches the workpiece, and the temperature of the base metal surrounding the weld pool exceeds the Curie point, which causes the magnetic permittivity of mild steel to be almost the same as that of the air. In such a case, the magnetic flux density in the arc region can be much larger than 1 mT. So, the effect of the external magnetic field on the arc and droplet cannot be neglected. The interaction of downward welding current I_A in the arc column and the magnetic field generates the forward electromagnetic force F_A which will push the arc and droplets slightly forward in the welding direction. The combined action of the forward tilted arc and flight droplets and the forces F_Z- and F_X will result in a thicker layer of molten metal in the forepart of weld pool, which is advantageous to absorb and dissipate the arc pressure and the impact of the filler metal droplets. Under normal welding conditions, the weld pool is not large in volume (less than 40 × 8 × 3 mm³),²⁴ so that the downward current component I_Z- is much larger than the other two components I_X- and I_X . Thereby, the forward electromagnetic force F_X related to I_Z- plays a more remarkable role in affecting the fluid flow in the weld pool. In this study, the forward electromagnetic force F_X is employed to change and control the backward flow jet of molten metal inside the weld pool so that the humping bead is suppressed.

Results and discussion

The high-speed camera is used to observe the arc and the weld pool during the welding process. Figure 3 shows the images of the arc and weld pool under different values of magnetic strength. When there is no external magnetic field (B = 0) as shown in Figure 3(a), the arc keeps almost straight and the droplets are transferred into the weld pool vertically. Under the action of the arc pressure and droplet impact, the weld pool surface is depressed and a crater forms beneath the arc. The molten metal is pushed into the boundary of the crater. The molten metal in the front and two sides of the crater flows backward along the side channels surrounding the arc and merges with the molten metal behind the crater and forms the backward flow jet with higher velocity and momentum. Such a backward flow jet of molten metal inside the weld pool with high momentum makes a gradually accumulated molten metal at the rear part of the weld pool, and a hump bead occurs. It can be seen from Figure 3(a) that the flowing channel of backward flow jet is wider, which appears as the uniform bright zone.

Figure 3.

Effect of magnetic flux density on the arc and molten pool: (a) B = 0 mT, (b) B = 40 mT, (c) B = 60 mT and (d) B = 80 mT.

When the external magnetic field is applied, as shown in Figure 3(b)–(d), the arc is inclined forward due to the forward electromagnetic force (F_A ) in the arc region. The droplet at the wire end is also inclined forward and impinges upon the fore edge of the crater in the weld pool. The molten metal in the front part of the weld pool is much thicker. The molten metal flows backward through two channels, one on each side of the crater. Due to the action of the forward electromagnetic force Fx, the molten metal flow behind the crater is suppressed, so that flowing along both sides of the crater are two main channels. It is clear that the bright zone in Figure 3(a) disappears when the external magnetic field is applied, as shown in Figure 3(b)–(d). If the magnetic strength is high enough, the molten metal at the center of the weld pool is also pushed to the forepart of the weld pool by the forward electromagnetic force Fx. Thus, only bright zones are at the two sides of the weld pool in Figure 3(b)–(d). When the magnetic strength is 80 mT, the force Fx becomes so strong that the molten metal accumulated at the forefront of the weld pool becomes bulky, as shown in Figure 3(d). When the droplet impinges the bulky molten metal there, great spatter will take place with loud noise. If the magnetic strength is over 80 mT, the arc becomes unstable and the spatter is too severe. In this study, 80 mT is the upper limit of the external magnetic strength.

Figure 4 shows the weld bead morphology at different levels of external magnetic strength. Without the external electromagnetic field, humping bead appears, as shown in Figure 4(a), and undercutting occurs together. The spacing between two humps is about 35 mm. After the external electromagnetic field is applied, the quality of weld formation is much improved, and no humping bead occurs. Good and smooth weld bead is obtained when the magnetic strength B = 60 mT, as shown in Figure 4(c). If B < 60 mT or B > 60 mT, undercutting still appears at some positions (see Figure 3(a) and (b)). If B = 80 mT, spatter is severe, and the weld bead is not uniform (see Figure 4(d)). Although the humping bead is suppressed by the external magnetic field, it causes some spatter, as shown in Figure 4(b)–(d). This side effect may have resulted from the interference of the external field on metal transfer. The investigation is still on-going to avoid the spatter.

Figure 4.

Weld bead morphology at different levels of magnetic strength: (a) B = 0 mT, (b) B = 40 mT, (c) B = 60 mT and (d) B = 80 mT.

The test samples are made from the weld bead in Figure 4. Figure 5 demonstrates the macrographs of the welds. Figure 5(a) is the transverse cross section at the valley when B = 0. When the external magnetic field is applied, there is smooth transition between the deposited metal and the base metal, and symmetric welds are obtained, as shown in Figure 5(b) and (c). If the magnetic strength is too high (B = 80 mT), the transition between the deposited metal and the base metal gets worsened, and there is a step on the left-hand side and a undercutting at the right-hand side, as shown in Figure 5(d). Clearly, there exists a corresponding optimum scope of the magnetic strength for a set of welding process parameters. Under the action of the matched external magnetic field, the fluid flow condition inside the weld pool is adjusted, and good weld bead quality is obtained.

Figure 5.

Cross-sectional shape of weld beads at different magnetic strengths: (a) B = 0 mT, (b) B = 40 mT, (c) B = 60 mT and (d) B = 80 mT.

Figure 6 shows the effect of the external magnetic strength on the weld width and penetration. With the increase in the external magnetic strength from 40 to 80 mT, the weld penetration decreases from 1.2 to 0.9 mm, while the weld width increases from 6.6 to 7.4 mm. This is because a higher magnetic strength causes a thicker molten layer in the front part of the weld pool which weakens the “digging” action from the arc and the impacting droplets beneath the arc. The weakened digging action makes a shallower weld pool, and the heat inside the pool has to transfer in transverse direction. Thus, the weld width expands a little, and the penetration gets less.

Figure 6.

Effect of the magnetic strength on weld width and penetration.

Conclusion

A magnetizer is developed to exert external magnetic field in weld pool during high-speed GMAW. It is found that the added electromagnetic force is able to change the flow conditions of molten metal and lower the momentum of backward flow jet in the weld pool, so that humping bead is suppressed.

Both the arc and the weld pool are observed by a vision sensor under the conditions with and without the actions of external magnetic field. With the action of external magnetic field, the forward tilted arc axis and droplet transfer as well as the forward electromagnetic force in weld pool combine together to form a thicker layer of molten metal in the front part of the weld pool, while the accumulation of the molten metal at the pool rear does not occur and no hump is found to form.

There exists an optimum range of the external magnetic strength for obtaining good quality of weld beads. For the study case, satisfactory weld bead without humping and undercutting is obtained when the magnetic strength is 60 mT at a welding speed of 2.01 m/min. However, the external magnetic field causes some spatter, and the investigation is still ongoing to avoid this side effect.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research received the financial support from the National Natural Science Foundation of China (Grant No. 51275276) and the Research Fund for the Doctoral Program of Higher Education of China (No. 20120131130009).

References

Dilthey

. Keynote Lecture—increased productivity by improved arc- and beam-welding. In: 8th international symposium on innovations in welding and joining for a new era in manufacturing (ed Hirata

Tanaka

), Kyoto, Japan 16–18 November 2008, pp.1–2. Tokyo, Japan: Japan Welding Society.

Tanaka

Ohata

Suenaga

. Now and then of welding science and technology. J Jpn Weld Soc 2011; 80(1): 1–11.

Cho

. A study on V-groove GMAW for various welding positions. J Mater Process Tech 2013; 213: 1640–1652.

Zhang

. Reconstruction of three-dimensional gas metal arc weld pool surface from reflected laser pattern. J Manuf Sci E: T ASME 2013; 135(2): 021002.

Nguyen

Weckman

Johnson

. High speed fusion weld bead defects. Sci Technol Weld Joi 2006; 11: 618–633.

Nguyen

Weckman

Johnson

. The humping phenomenon during high speed gas metal arc welding. Sci Technol Weld Joi 2005; 10: 447–459.

Zhang

. Suppression of weld-bead defects and increase in the critical welding speed during high-speed arc welding. Proc IMechE, Part B: J Engineering Manufacture 2009; 223: 751–757.

Woloszyn

Melton

Sende

. Comparison of advanced MIG/MAG and TIG welding processes. International Institute of Welding IIW Document XII-2076-12, July 2012. Paris, France: International Institute of Welding.

Wang

Chen

. Research progress in high efficiency modification of arc welding processes. Weld Join 2013; (8): 1–10.

10.

Nguyen

Weckman

Johnson

. Predicting onset of high speed gas metal arc weld bead defects using dimensional analysis techniques. Sci Technol Weld Joi 2007; 12: 634–648.

11.

Michie

Blackman

Ogunbiyi

TEB

. Twin-wire GMAW: process characteristics and applications. Weld J 1999; 78(5): 31–34.

12.

Chen

Zhang

. Double-electrode GMAW process and control. Weld J 2007; 86: 231s–237s.

13.

Zhong

. Prevention of humping bead associated with high welding speed by double-electrode gas metal arc welding. Int J Adv Manuf Tech 2012; 63: 573–581.

14.

Hicken

Jackson

. Effects of applied magnetic fields on welding arcs. Weld J 1966; 45(8): 513s–518s.

15.

Wang

Lin

. Study on moving GTA weld pool in an externally applied longitudinal magnetic field with experimental and finite element methods. Model Simulat Mater Sci Eng 2002; 10(6): 781–798.

16.

Lin

Shen

. Numerical analysis of transport phenomena in resistance spot welding process. J Manuf Sci E: T ASME 2011; 133(3): 031019-1–031019-8.

17.

Shen

Lin

. Quality improvement in resistance spot welding of advanced high strength steel using external magnetic field. Sci Technol Weld Joi 2011; 16(5): 465–469.

18.

Shen

Lin

. Impact of external magnetic field on weld quality of resistance spot welding. J Manuf Sci E: T ASME 2011; 133(5): 051001-1–051001-7.

19.

Shen

Lin

. Effect of external constant magnetic field on weld nugget of resistance spot welded dual-phase steel DP590. IEEE T Magn 2011; 47(10): 4116–4119.

20.

Luo

. Mechanism of the gas tungsten arc welding in longitudinal magnetic field controlling. Acta Metall Sin 2001; 37: 217–220.

21.

Wang

Chen

. Study on the electromagnetically controlled horizontal position MAG welding process. J Beijing Univ Technol 2003; 29: 147–150.

22.

Shoichi

Yukio

Koki

. Study on the application for electromagnetic controlled molten pool welding process in overhead and flat position welding. Sci Technol Weld Joi 2013; 18: 38–44.

23.

Kern

Berger

Huegel

. Magneto-fluid dynamic control of seam quality in CO₂ laser beam welding. Weld J 2000; 79: 72s–78s.

24.

Chen

. Numerical simulation of forming process of humping bead in high speed GMAW. Acta Metall Sin 2009; 45: 1070–1076.