Optics & Laser Technology 56 (2014) 158166

Contents lists available at ScienceDirect

Optics & Laser Technology

journal homepage: www.elsevier.com/locate/optlastec

Study of metal transfer in CO2 laser GMAW-P hybrid welding

using argonhelium mixtures
Wang Zhang a,b, Xueming Hua a,b,n, Wei Liao a,b, Fang Li b, Min Wang a,b
a
b

Shanghai Key Laboratory of Material Laser Processing and Modication, Shanghai Jiaotong University, Shanghai 200240, PR China
State Key Laboratory of Metal Matrix Composites, Shanghai Jiaotong University, Shanghai 200240, PR China

art ic l e i nf o

a b s t r a c t

Article history:
Received 15 May 2013
Received in revised form
26 July 2013
Accepted 10 August 2013
Available online 7 September 2013

The metal transfer in CO2 Laser GMAW-P hybrid welding by using argonhelium mixtures was
investigated and the effect of the laser on the mental transfer is discussed. A 650 nm laser, in conjunction
with the shadow graph technique, is used to observe the metal transfer process. In order to analyze the
heat input to the droplet and the droplet internal current line distribution. An optical emission
spectroscopy system was employed to estimate default parameter and optimized plasma temperature,
electron number densities distribution. The results indicate that the CO2 plasma plume have a signicant
impact to the electrode melting, droplet formation, detachment, impingement onto the workpiece and
weld morphology. Since the current distribution direction ow changes to the keyhole, to obtain a metal
transfer mode of one droplet per pulse, the welding parameters should be adjusted to a higher pulse time
(TP) and a lower voltage.
& 2013 Elsevier Ltd. All rights reserved.

Keywords:
Metal transfer
Hybrid welding
Shielding gas

1. Introduction
Laser GMAW-P hybrid welding is an increasingly accepted
joining technology for a variety of industrial sectors [1,2]. The
main advantages are the wider gap tolerance, higher welding
speed, and controlling the bead humping defect, improving the
weld metal microstructure by using ller materials [3,4].
In order to achieve a stable and repeatable welding process, it is
essential to select optimum combinations of parameters for a
controlled metal transfer process over wide ranges of heat and
mass input levels. GMAW-P which is a modied spray transfer
process provides the best of both short-circuiting and spray
transfer, by using a low base current to maintain the arc and a
high peak current to melt the electrode wire and detach the
droplet [57]. Although pulsed GMAW is capable of reducing
spatters and improving arcing stability through obtaining spray
transfer, such capability is conditional [8]. So far, no one welder
specically designed for hybrid welding. Thus, it is not easy to
attain the repeatability and controllability in Laser GMAW-P
hybrid welding.
The plasma characteristics and metal transfer process must be
affected by the additional laser beam. A number of experiments
have been conducted to investigate the behavior of droplet
n
Corresponding author at: Shanghai Jiaotong University, Shanghai Key Laboratory of Material Laser Processing and Modication, Shanghai 200240, PR China.
Tel.: 86 21 54748940; fax: 86 21 54748940.
E-mail addresses: xmhua@sjtu.edu.cn, mrzhangwang@163.com (X. Hua).

0030-3992/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.optlastec.2013.08.006

transfer in laser or hybrid welding [2,9,1012]. Liu et al. investigated a direct-current GMAW CO2 laser hybrid welding, it was
found that the droplet transfer mode is changed from globular
transfer to spray transfer with the increasing distance between
laser and arc [11]. Campana et al. found that spray transfer is more
appropriate than short circuiting transfer for hybrid welding [13].
Roepke et al. found that low laser powers produced the large
globular/short circuiting transfer in the arc, while high laser
powers produced the mid frequency small globular free ight
transfer [12]. Huang and Zhang found the laser affect the metal
transfer process as an additional detaching force [14,15]. However,
GMAW-P often referred to as synergic power supplies, the pulse
welding process parameters only aimed at specic parameters and
shielding gas [16]. In case of hybrid welding, the changing of
shielding gas and energy input to the plasma would given the
mental transfer a destabilizing effect [2,17]. This is due to the
droplet transfer mode related to both arc power and laser power
affects droplet transfer in laser-GMA hybrid welding process.
Therefore, the pre-selected conventional GMAW-P pulse parameters need to be optimized before use.
Appropriate selection and matching of the shielding gas and
the welding parameters will contribute to renements in
Laser GMAW-P hybrid welding. However, no attention has been
paid to study how to obtain a metal transfer mode of one droplet
per pulse up to now. Therefore, the primary objective of this
investigation is to compare the effect of laser-induced plasma on
the metal transfer in Laser GMAW-P hybrid welding, and nd the
appropriate sets of pulse parameters to achieve the transfer mode

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

of one-droplet-per-pulse. Finally, we given a guide on how to

design specically welder for hybrid welding. The arc plasma and
the metal transfer was observed by a high speed video camera. The
plasma temperature, electron number densities distribution was
estimated by an optical emission spectroscopy system, respectively. On the basis of the measured results, the heat content and
force, which exerts a great inuence on the droplet transfer, was
discussed.

2. Theoretical background
How much the heat content of droplet and how the forces act
on the droplet were the major factors affecting the type of droplet
transfer. In the conventional GMAW-P progress, Heat content of
droplet is composed of the heat input by the Joule heating, H J , and
arc heating H A [18]. In the case of hybrid welding, an additional
laser plume radiation heat input H L must be considered.
H J f I rms =S2 l=v

HJ is a function of (Irms/S)2 l/v, where S is the cross section of

electrode, l the electrode extension, v the electrode feeding rate,
and Irms is the effective current.
H A EA UI

where U is the equivalent melting voltage, and I the average

current.
H L f LP ; DLA; V

H L is a function of (Lp, DLA, V),where Lp is the laser power, DLA the

distance between laser and arc, V the welding speed. This can be
reected by plasma prole and temperature.
In the conventional GMAW-P progress, the metal transfer was
considered in a state of force balance and thermal equilibrium to
achieve one droplet per every pulse. However, the application of
the laser destroyed the balance of welding progress. In the case of
hybrid welding progress, a large amount of laser-induced plasma
energy is continuously put into the electrode. The additional laser
power changed the arc prole, electron density, current density
and other parameters of conventional GMAW. Which nally
changed either the direction or size of the force acted on the
droplet. The major forces acting on the droplet include the
gravitational force, electromagnetic force (Lorentz force), aerodynamic drag force, surface tension, and vapor jet force. These forces
are shown in Fig. 1.
According to the static-force balance theory, the balance of
these forces determines the metal transfer process, i.e., droplet
formation, size, and frequency. Detaching force and retention force
occur in hybrid welding process. Detaching forces include gravitational force, plasma drag force, and electromagnetic force, while
retention forces include surface tension and vaporization force.

159

The force due to gravity can be expressed as

4
F g R 3 d g
3

where R is the droplet radius, d is the droplet density, and g is the

acceleration of the gravity. This force is higher with higher droplet
radius.
The surface tension is given as
F 2

where a is the electrode radius, while is the surface tension

coefcient. The surface tension mainly determined by the diameter of the wire.
The aerodynamic drag force can be expressed as;
!
f v2f
F d C D Ap
6
2
where C D is the aerodynamic drag coefcient, f and vf are the
density and uid velocity of the plasma, Ap is plasma acting area.
This force is higher with higher droplet radius and plasma velocity.
The electromagnetic force, Fem, is given by [19]


I2 1
r
ln i
7
F em 0
4 2
ru
where 0 is the magnetic permittivity, I is the welding current, r i is
the exit radius of the current path, and r u is the entry radius of the
current path. However, in case of hybrid welding, the category of
electromagnetic force is changed with changing current line
distribution in the droplet.
The vapor jet force is given by [20]
8

3=2 
< 1 C D A2 V 2 Na K B T s expU=T s expD 2 =2R2 D r R
2
LA
LA
h
m o
h
M a Bo
F RL 4 Rh
:
0
D Z R
LA

8
where Rh is the metal vapor distribution parameters, C D is the
aerodynamic drag coefcient, A is plasma acting area, m and is
the density of the plasma, Vo is constant (3.4  102 m s  1), Na is
Avogadros constant, KB is Boltzmann constant, TS is the surface
temperature of melted zone. DLA is the distance between laser and
arc. As can be seen, this force is mainly related with the DLA .
As can be seen, in case of hybrid welding, both the heat content
and the forces of droplet are different from conventional GMAW-P
progress.

3. Experimental set-up and procedure

The experimental setup is shown in Fig. 2. It composed of ve
main sections: (1) the welding torch with power supply and the
system of gas ow control, (2) the welding parameter data
acquisition system, (3) the high speed camera, (4) the high power
laser and (5) the spectrometer system. These sections are
described in later paragraph.
3.1. The welding process

Fig. 1. Schematic of forces affecting droplet in hybrid welding.

A welding arc is generated at atmospheric pressure by a plasma

torch, with 45 degree angles to a workpiece. Electrical power, gas
ow and wire feed control are performed by a Fronius machine
TPS5000. The power source is operated with positive polarity on
the wire. The wire is steel CHW-50C6, and is 1.2 mm in diameter. It
is fed at a velocity of 4 m min  1. A 10 mm thick steel (E36)
workpiece was employed. The chemical compositions (percent
by weight) of the workpiece and the wire are given in Table 1.
Shielding gas are argonCO2 (80% Ar20% CO2) and argonhelium
mixtures (50% argon50% helium), owing at 40 L min  1.

160

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

Fig. 2. Experimental set-up for CO2 laser hybrid welding.

Table 1
Chemical composition of the workpiece and the wire.
C

Mn

Si

Cu

CHW-50C6

0.060.15

1.401.85

0.801.15

r0.035

r 0.025

r 0.5

E36

r0.18

0.901.60

r 0.50

r0.035

r 0.035

r 0.35

The torch was set up with a distance of 15 mm between

workpiece and contact tip. During the welding process, the
distances between contact tip, nozzle and workpiece were kept
constant in order to create plasma with reproducible plasma
properties. The workpiece was moved in the positive Y-direction
at 75 cm min  1 by means of a translation table. The wire feeding
speed is 4 m min  1, the laser power is 6 kW, and the distance
between the laser and arc is 4 mm.
In laser welding, the role of the shielding gas is not only to
prevent the molten metal from oxidizing, but also to suppress the
laser-induced plasma. Because of the ionization of argon, the laserinduced plasma disturbs the laser energy transfer into the workpiece. In contrast with argon, helium has high ionization potential,
high heat conductivity, and low inverse Bremsstrahlung, It is
thought that Helium is the most effective way in plasma control.
However, helium is much more expensive than argon. argon is
often mixed with various percentages of helium to take the
advantages of each gas [21]. When the helium content was above
50%, the amount of laser-induced plasma signicantly reduced
[22]. Therefore, a 50% helium and 50% argon was chosen for the
optimized experiments.
In the laser-GMA hybrid welding process, shielding gas can be
supplied in three different ways: (1) One through the copper
orice coaxially and another through the MAG torch, respectively
[9,11,23]; (2) An additional gas nozzle and a GMA torch used
simultaneously [24]; (3) Only through a GMAW welding torch
[2,22]. In the case where shielding gas is supplied by two nozzle,
a kind of turbulence ow can be generated, which leads to the
formation of porosity in the welds [22]. Finally, in this experiment,
the shielding gas was supplied only through a GMAW
welding torch.
The rst phase of the experiment was aimed at investigating
the role of the high power laser on the plasma characteristic and
droplet transfer behavior in situation of conventional GMAW-P
welding mode. The test was carried out exploiting factory default
parameters and using the 80% Ar 20% CO2 shielding gas. The
welding parameters were set as follows: pulse frequency, 110 Hz;
peak arc current, 515 A; base arc current, 25 A; peak time, 1.35 ms;
voltage command value 31.2 V.

Cr

Ni

Other
r 0.5

r0.20

0.01

The second phase of the experiment was aimed at understanding how the plasma characteristic and droplet transfer
affected by the 50% argon 50% helium shielding gas and a Laser
beam. The test was carried out exploiting factory default parameters and using the 50% argon 50% helium shielding gas.
The last experimental phase was presented to optimizing the
appropriate sets of pulse parameters to achieve the transfer mode
of one-droplet-per-pulse. The welding parameters were adjusted
to as follows: peak time, 1.6 ms; voltage command value 26 V.
3.2. Data acquisition system
The current was measured with a Hall-effect transducer
(CS600E2/4V, Chieful). The voltage was measured with a Halleffect voltage sensor (VSM025A, Chieful) at the contact tip. The
electrical measurements were performed using high-performance
multifunction analog, digital, and timing I/O board(6023E, NI)to PC
and recorded by a LabVIEW software.
3.3. High speed camera system
A high speed camera (Fastcam-Super10KC, Photron) system
based on principle of optical polarization was used to obtain
images of the plasma and droplet transfer [25]. To suppress the
arc light and obtain the shadow graph of droplet, an optical band
pass lter (6507 1.5 nm) and two polarizers were mounted in
front of the camera lens. As a light source, a 100 mW 650 nm
semiconductor laser with the beam expander was mounted on the
other side of the arc. The sample frequency for droplet was 3000
frames per second and 1000 frames per second for plasma.
Synergized trigger mode was used in the experiment.
3.4. High power laser
TLF15,000 was used as the laser source with a maximum
output power of 15 kW and a focal length of 357 mm. The K factor
was 0.26 and the focus diameter was 0.75 mm.

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

3.5. The spectrometer system

Welding plasma generates a two times image on the image
screen by a convex lens, An aperture diaphragm is mounted at the
focus to allow only parallel light passing. The radiation emitted
perpendicularly to the arc axis and propagating in the X-direction
is recorded as a function of the lateral position Y, the datum was
seted at the focus. The HR4000 Fiber Optic Spectrometer is used in
this experiment.
In thermal plasmas under local thermo dynamical equilibrium,
electron excitation temperature Te is equal to heavy particles
temperature Th. The plasma can then be described as a uid with
Table 2
Characteristic of the iron lines selected for the Boltzmann-diagnostic method.
ij (nm)

Aji (106 s  1)

Ej

gj

530.23
532.4
532.8
537.1
538.3
539.3
539.7

6.30E6
1.500E7
1.05E6
1.15E6
5.6E7
3.10E6
2.59E6

5.6206837
5.5392422
3.2409687
3.2657053
6.6149260
5.5392422
3.2111889

5
9
7
5
13
9
9

161

a single temperature T Te Th [26]. The excitation temperature of

the excited atomic levels is then equal to the electron temperature
and their densities are related to the next ions ground state
population density by the SahaBoltzmann relations [27]. Following the Boltzmann plot method, we analyzed iron lines to deduce
the plasma temperature. The electron number density was determined using the Stark broadening of the 532.4 nm FeI line.
Unlike the one based on Stark broadening, the Boltzmann plot
method needs the LTE hypothesis validity. In fact, Valensi et al.
[26] already veried the LTE in the core region of the GMAW
plasma. Also, we given some assumptions for the plasma: the
existence of local thermodynamic equilibrium (LTE) conditions;
and the optically thin plasma. Then, the excitation temperature
can be easily calculated using the Boltzmann plot method [28,29].
!


ij ij
Ej
ln
C
9

KBT
g j Aji
The left-hand term is called a Boltzmann function. The quantities ij , ij and Aji are the emissivity (Wm3 sr1 ), the wavelength
(m), the transition probability (s1 ) of the line corresponding to
the transition from upper level j to lower level i, respectively.
Ej and gj are the energy (eV) and the statistical weight of level j,
respectively. K B is the Boltzmann constant. T is the temperature of
the plasma (K) and C is a constant for a xed temperature. A linear
t of lnij ij=g j Aji on Ej can deduce the plasma temperature.
Seven iron lines are selected because of the large differences in
their upper energy levels Ej (Table 2).
The electron densities of pulse time (tp) were directly obtained
by measuring the Stark widths of the FeI 532.4 nm. Experimentally
measured spectra are convolutions of the different broadening
effects. The convolution of a Lorentzian and Gaussian is given by a
Voigt prole [30,31]
V

1
1

GLd

10

For non-hydrogenic ions the Stark effect gives rise to a

Lorentzian spectral prole whose full width at half maximum
(FWHM) is directly related to the electron density of the plasma.
Spectral lines, whether hydrogenic or non-hydrogenic, can be
tted to determine plasma density is: [30,32].
Fig. 3. Characteristics of current and voltage during GMAW and hybrid welding
(shielding gas 80% Ar 20% CO2).

1=2 2  106 wN e 1 1:75  104 Ne1=4 10:068Ne1=6 Te1=2 

11

Fig. 4. The high-speed photographs of droplet transfer in pulsed GMAW and hybrid welding (3000 frames per second, Ip 540 A, Tp 1.3 ms, shielding gas:
80% Ar 20% CO2).

162

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

Fig. 5. The high-speed photographs of plasma generated in hybrid welding (1000 frames per second, Ip 540 A, Tp 1.3 ms, shielding gas: 80% Ar 20% CO2).

Fig. 6. Probability density distribution of voltage for GMAW and hyrid welding
(Ip 540 A, Tp 1.3 ms, shielding gas: 80% Ar 20% CO2).

Fig. 7. Characteristics of current and voltage during GMAW and hybrid welding for
50% Ar 50% He shielding gas.

The values are reported by Valensi et al. [33], we got the

relation

1:6700
Ne
Te
Fe

12
S 0:2648
1023 13; 000
To get the stark broadening density prole, we used Fourier
expansion deconvolution method [34,35].

4. Results
4.1. Optimization of shielding gas
In GMAW welding, the consumable metal electrode is both
ller material and arc carrier. The shielding gas that ows out
prevents chemical reactions between the hot workpiece surface
and the surrounding air. This maintains the strength and durability
of the weld metal. Inert and active gases can be used as
shielding gases.
Conventional GMAW-P usually employs pure argon or argon
with the addition CO2 as shielding gas, Fig. 3 demonstrate the
welding current and voltage waveform for conventional GMAW-P
and CO2 laser GMAW-P process where the shielding gas is 80%
argon and 20% CO2. As can be seen, the period cycle of current
waveform is divided into three phases (pulse current time, droplet
detachment time and, pulse off time), during which the electrode
is molten, a molten drop propagates to the workpiece, and the
plasma is sustained by a small holding current, respectively.
Fig. 4 shows the droplet transfer high-speed photographs. In
the case of conventional GMAW-P, the droplet is detached with a
stable mode of one-droplet-per-pulse. On the other hand, in the

Fig. 8. Probability density distribution of voltage for GMAW and hybrid for
50% Ar 50% He shielding gas.

case of the CO2 laser GMAW-P hybrid process, a large amount of

plasma energy induced by laser is continuously put into the
electrode. Thus, the molten electrode tip tends to be prolonged.
The arc length of hybrid was relatively small compared to
GMAW welding due to the presence of the laser induced plasma
causes the arc resistance and radius to decrease. In Fig. 3,, the
arc voltage of pulse current time and pulse off time in GMAW
progress are a little higher than that in hybrid. However, because
of the unstable plasma plume, the measured voltage of dropletdetachment time for hybrid progress is a uctuating waveform, as
Fig. 3, shows. As indicated by the high-speed photographs and
the probability density distribution, as shown in Fig. 6, the mode
of metal transfer is short circuiting. It also found that the hybrid

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

progress have a slightly lower pulse frequency compared to

the GMAW.
Fig. 5 shows the plasma generated in hybrid welding. Because
argon gas has a lower ionization potential, lower heat conductivity,
and higher inverse Bremsstrahlung than helium. Thus, it absorbs
most of the lasers irradiation energy. The laser-induced plasma is
very strong and unstable.
Fig. 6 shows the density probability distributions of voltage for
GMAW and hybrid progress. The distributions present three
humps to the corresponding welding waveform. This feature is
more evident for GMAW, while the distributions corresponding to
the hybrid show a slight gentle transition. There is an extension
trend toward the 0 V values indicating that the mode of metal
transfer have a tendency of short circuiting.

163

4.2. Optimization of current waveforms

From what mentioned above, it can be seen clearly that the
pure argon or argon with the addition CO2 shielding gas which
conventional GMAW-P usually employed can not be singly applied
to the CO2 laser GMAW hybrid welding progress. In the next two
phases of the experiment, the 50% argon 50% helium shielding
gas was used.
Fig. 7 shows the measured voltage/current waveforms when
the shielding gas is 50% helium and 50% argon. As can be seen, the
hybrid progress have a remarkable higher pulse frequency compared to the GMAW.
As Figs. 810 shown, with decreasing arc length, short circuit
and arc broken usually happens in the case of GMAW progress,

Fig. 9. The high-speed photographs of droplet transfer in pulsed GMAW for 50% Ar 50% He shielding gas.

Fig. 10. The high-speed photographs of plasma generated in GMAW welding for 50% Ar 50% He shielding gas.

Fig. 11. The high-speed photographs of droplet transfer in hybrid welding for 50% Ar 50% He shielding gas with default parameters.

164

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

Fig. 12. The high-speed photographs of plasma generated in hybrid welding for 50% Ar 50% He shielding gas with default parameters.

Fig. 13. The high-speed photographs of droplet transfer in hybrid welding for 50% Ar 50% He shielding gas with optimized parameters.

Fig. 14. The high-speed photographs of plasma generated in hybrid welding for 50% Ar 50% He shielding gas with optimized parameters.

Fig. 15. Temperature distribution of default and optimized hybrid plasma on

2.5 mm above the surface of workpiece.

Fig. 16. Electron number density distribution of default and optimised hybrid
plasma on 2.5 mm above the surface of workpiece.

4.3. Electron excitation temperature and electron number densities

due to the hard ionization and unstable cathode spot, the measured waveforms and the probability density distribution of
voltage validate the phenomenon.
Fig. 11 is the droplet transfer images of hybrid welding in
default welding parameter where the shielding gas is 50% helium
and 50% argon. Fig. 12 shows the plasma images. Although the
pendant droplet is constricted, that a neck is formed during the
pulse time (as shown in Fig. 11, 56), the current did not keep at
the peak value to exercise its electromagnetic force over the
droplet. The current falls down just after 1.35 ms, the electromagnetic force acting on the droplet is suddenly lowered, due to
the surface tension of the droplet, neck is disappeared (as shown
in Fig. 11, 79), the pendant droplet never transfers, Observation
shows that the pendant droplet swells into a large globe during
several pulses. In this case, the droplet and the plasma turned
toward keyhole side instead of along the axis of electrode.
Fig. 13 is the droplet transfer images captured when the
welding parameters was optimized. Fig. 14 shows the plasma
images. After the pulsing time reached 1.6 ms and the arc voltage
reduced to 26 V, as shown in Fig. 7,, the droplet transfer
become one droplet with a diameter close to that of the electrode
transfers regularly and reliably per droplet every pulse.

In order to analyze the heat input to the droplet and the droplet
internal current line distribution. Optical emission spectroscopy
was employed to estimate default parameter and optimized
plasma temperature, electron number densities distribution.
Figs. 15 and 16 shows the temperature distribution and electron
number density distribution of default and optimized hybrid
plasma on 2.5 mm above the surface of workpiece, respectively.
As can be seen, in the transition region from arc to laser plasma,
spatial variations of the electron temperature and electron density
strongly depend on the arc voltage and exhibit much larger
gradients in the optimized mode than in the conventional mode.
In the arc core, resulting in a decrease in the temperature and ne,
the ne decreases by about 50% with the optimized mode. Moreover, the results show that the electron density and electron
temperature of the laser region independently of the
transfer mode.
4.4. Weld sample
Fig. 17 shows the bead appearance and cross section at different
hybrid welding condition. For 80% Ar 20% CO2 shielding gas, the

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

penetration depths were dramatically decreased owing to the

disturbance of laser irradiation to the base metal by laserinduced plasma. Compared with the default parameter, in the
case of optimized parameter, weld width was presented to reduce
the uctuations, and shown a nail-head weld bead prole.

5. Discussion
As has been observed above, the application of the laser would
given the mental transfer a destabilizing effect. For 80% Ar 20%
CO2, penetration depths were dramatically decreased owing to the
disturbance of laser irradiation to the base metal by laser-induced
plasma. This because argon and helium differ in physical properties, the major difference is that argon has a lower ionization
potential, 15.755 V as compared to 24.58 V for helium. With a
wavelength of 10.6 m, CO2 lasers often experience greater plasma
absorption. When a plasma phase is present, partially-ionized
argon gases absorb infrared wavelength by the inverse Bremsstralung mechanism and refract the laser beam so as to defocus it.
Finally, the power of the laser beam are modied as it traverses the
plume and the laser power density impinging on the material
surface is decreased.
Due to the wire feeding at a constant rate, the arc length is
lengthened to balance the arc by reducing the Joule heat, as show
in Figs. 9, 11 and 13.
The laser-induced plasma changed the size and direction of the
force on the droplet. As can be observed, the arc tends to bend to
the keyhole. The current diverged downward from the electrode
tip, then converged at the surface of the workpiece and the laserinduced plasma. It was seen the outer shape of the plasma prole
between the electrode tip and the workpiece for the case of
optimized looks like a broom. In the default cases, a typical
cocktail shape plasma arc is found, as Figs. 12 and 14 show.
According to the calculated electron number density distribution, in the transition region between arc and laser. Compared to
the default parameter, the optimized result trend to have a higher
gradient descent, which mean that exist a much more arc current
contraction underneath the droplet in case of default case. Since
the arc follows the path of least electrical resistance, the arc tends
to bend, and the arc root forms within close proximity to the
keyhole.
Due to the change in the direction of the current ow, there
exists an upward and inward electromagnetic force near the
bottom of the droplet which becomes an attaching force that

165

sustains the droplet. An obvious arc contraction appears underneath the droplet due to the attraction of laser to the electric arc
plasma, as shown in Fig. 18(a). The arc contraction causes current
ows to leave the droplet from a smaller area at the bottom
surface of the droplet and thus induces the inward and downward
current ows in the lower part of the droplet, which, in turn,
produces an upward and inward electromagnetic force for repelling the droplet, as shown in Fig. 18(b).
The downward electromagnetic force near the top of the
droplet and the upward electromagnetic force near the bottom
of the droplet tend to squeeze the droplet to become an oblate
shape, and droplets center-of-mass deviated from the wire axis.
Thus, produces a weaker pinch effect, which also increases the
droplet formation time.
As Fig. 19 shown, after optimization, since the voltage is lower,
and therefore, the cathode spot is not completely dependent on
the laser-induced plasma, the current line distribution becomes
more divergent, the direction of electromagnetic force is inward
the droplet, which would tend to increase to accelerate the
separation of droplet from the wire.
Based on the above research, in order to obtain the droplet
transfer mod of one droplet per pulse in hybrid welding progress,
On the one hand, the heat input to the droplet should be
controlled. The heat input of conventional GMAW is mainly
composed of the Joule heating, H J and arc heating, H A . In hybrid
welding, the additional laser heating causes an increase in the
plasma temperature, thus, to balance the heat input to the droplet,
the arc voltage need to be reduced for decreasing the arc heating,
as Fig. 15 shows.

Fig. 18. The distributions of current and corresponding electromagnetic force

distribution in the metal for the cases of default parameters.

Fig. 17. The bead appearance and cross section at different welding condition.

166

W. Zhang et al. / Optics & Laser Technology 56 (2014) 158166

Fig. 19. The distributions of current and corresponding electromagnetic force

distribution in the metal for the cases of optimized parameters.

The changed current distribution in hybrid welding leads to an

upward and inward electromagnetic force near the bottom of the
droplet and the increase of droplet formation time, so the pulse
duration should be increased.
6. Conclusion
The metal transfer in CO2 Laser GMAW-P Hybrid Welding by
using argonhelium mixtures was investigated, Both the inuences of shielding gas compositions and current waveform on the
metal transfer were studied.
The phenomena of arc prole, droplet transfer mod, the
distribution of temperature and electron number density and
probability density distribution of voltage in the default welding
parameter case and optimized case were compared.
In order to obtain the droplet transfer mod of one droplet per
pulse in hybrid welding progress, On the one hand, the heat input
to the droplet should be controlled, the arc voltage need to be
reduced for decreasing the arc heating.
The changed current distribution in hybrid welding leads to an
upward and inward electromagnetic force near the bottom of the
droplet and the increase of droplet formation time, so the pulse
duration should be increased.
Acknowledgments
This research effort was supported by the National Science
Foundation of china [Grant Number: 51275299 and 51035004].
References
[1] Zhang T, Wu CS, Qin GL, Wang XY, Lin SY. Thermomechanical analysis for
Laser GMAW-P hybrid welding process. Computational Materials Science
2010;47:84856.
[2] Tani G, Campana G, Fortunato A, Ascari A. The inuence of shielding gas in
hybrid laser-mig welding. Applied Surface Science 2007;253:80503.
[3] Bagger C, Olsen FO. Review of laser hybrid welding. Journal of Laser Applications 2005;17:2.
[4] Choi H, Farson D, Cho M. Using a hybrid laser plus GMAW process for
controlling the bead humping defect. Welding Journal 2006;85:1749.
[5] Choi S, Yoo C, Kim Y. The dynamic analysis of metal transfer in pulsed current
gas metal arc welding. Journal of Physics D: Applied Physics 1999;31:207.
[6] Palani P, Murugan N. Selection of parameters of pulsed current gas metal arc
welding. Journal of Materials Processing Technology 2006;172:110.
[7] Wu C, Chen M, Lu Y. Effect of current waveforms on metal transfer in pulsed
gas metal arc welding. Measurement Science and Technology 2005;16:2459.
[8] Zhang Y, Li P. Modied active control of metal transfer and pulsed GMAW of
titanium. Welding Journal (USA) 2001;80:54.

[9] Kutsuna M, Chen L. Interaction of both plasmas in CO2 Laser-MAG hybrid

welding of carbon steel, LAMP 2002: international congress on laser advanced
materials processing. International Society for Optics and Photonics
2003:3416.
[10] Campana G, Fortunato A, Ascari A, Tani G, Tomesani L. The inuence of arc
transfer mode in hybrid laser-mig welding. Journal of Materials Processing
Technology 2007;191:1113.
[11] Liu S, Liu F, Zhang H, Shi Y. Analysis of droplet transfer mode and forming
process of weld bead in CO2 laser-MAG hybrid welding process. Optics & Laser
Technology 2011.
[12] Roepke C, Liu S, Kelly S, Martukanitz R. Hybrid laser arc welding process
evaluation on DH36 and EH36 steel. Welding Journal 2010;89:1409.
[13] Campana G, Fortunato A, Ascari A, Tani G, Tomesani L. The inuence of arc
transfer mode in hybrid laser-mig welding. Journal of Materials Processing
Technology 2007;191:1113.
[14] Huang Y, Zhang Y. Laser enhanced metal transferPart 1: System and
observations. Welding Journal 2011;90.
[15] Xu GX, Wu CS, Qin GL, Wang XY, Lin SY. Adaptive volumetric heat source
models for laser beam and laser pulsed GMAW hybrid welding processes.
The International Journal of Advanced Manufacturing Technology
2011;57:24555.
[16] Ghosh P, Dorn L, Hbner M, Goyal V. Arc characteristics and behaviour of
metal transfer in pulsed current GMA welding of aluminium alloy. Journal of
Materials Processing Technology 2007;194:16375.
[17] Ribic B, Palmer T, DebRoy T. Problems and issues in laser-arc hybrid welding.
International Materials Reviews 2009;54:22344.
[18] Ueguri S, Hara K, Komura H. Study of metal transfer in pulsed GMA welding.
Welding Journal 1985;64:242s50s.
[19] Huang Y, Zhang Y. Laser enhanced GMAW. Welding Journal 2010;89.
[20] Liu D. Analysis of Droplet Behavior in CO2 Laser-Mig Hybrid Welding Process.
Harbin Institute of Technology; 2006.
[21] Rao Z, Hu J, Liao S, Tsai H. Modeling of the transport phenomena in GMAW
using argonhelium mixtures Part IThe arc. International Journal of Heat and
Mass Transfer 2010;53:570721.
[22] Rhee S, Kim JH, Kim CH, Chae HB. The effect of shielding gas composition in
CO2 lasergas metal arc hybrid welding. Proceedings of the Institution of
Mechanical Engineers, Part B: Journal of Engineering Manufacture
2008;222:131524.
[23] Z Liu, M Kutsuna, L Sun, CO2 laser-mag hybrid welding of 590 MPa high strength
steel-study on laser-arc hybrid welding of high strength steel (Report 1),
, 24; 2006. 1725.
[24] Hu L, Huang J, Li Z, Wu Y. Effects of preheating temperature on cold cracks,
microstructures and properties of high power laser hybrid welded 10Ni3
CrMoV steel. Materials & Design 2011;32:19319.
[25] Tan Xiu-wen HX-M, Meng-Xing LI, Xin YE, Fei Wang. High Speed Video
Photography system based on principle of optical polarization. Journal of
Shanghai Jiaotong University 2011;45.
[26] Valensi F, Pellerin S, Boutaghane A, Dzierzega K, Pellerin N, Briand F. LTE
experimental validation in a gas metal arc welding plasma column. Contributions to Plasma Physics 2011;51:2936.
[27] Zieliska S, Musio K, Dzierga K, Pellerin S, Valensi F, De Izarra C. Investigations of GMAW plasma by optical emission spectroscopy. Plasma Sources
Science and Technology 2007;16:832.
[28] Rouffet M, Wendt M, Goett G, Kozakov R, Schoepp H, Weltmann K, et al.
Spectroscopic investigation of the high-current phase of a pulsed GMAW
process. Journal of Physics D: Applied Physics 2010;43:434003.
[29] Rouffet ME, Wendt M, Goett G, Kozakov R, Schoepp H, Weltmann KD, et al.
Spectroscopic investigation of the high-current phase of a pulsed GMAW
process. Journal of Physics D: Applied Physics 2010;43:434003.
[30] Vogman G, Shumlak U. Deconvolution of Stark broadened spectra for multipoint density measurements in a ow Z-pinch. Review of scientic instruments 2011;82:103504.
[31] G Vogman, U Shumlak, R Golingo, B Nelson, Stark broadening density prole
measurements using Fourier expansion deconvolution, (2009).
[32] Dimitrijevi MS, Christova M, Sahal-Brchot S. Stark broadening of visible Ar I
spectral lines. Physica Scripta 2007;75:80919.
[33] Valensi F, Pellerin S, Boutaghane A, Dzierzega K, Zielinska S, Pellerin N, et al.
Plasma diagnostics in gas metal arc welding by optical emission spectroscopy.
Journal of Physics D: Applied Physics 2010;43:434002.
[34] He J, Zhang C. The accurate calculation of the Fourier transform of the pure
Voigt function. Journal of Optics A: Pure and Applied Optics 2005;7:613.
[35] Letchworth KL, Benner DC. Rapid and accurate calculation of the Voigt
function. Journal of Quantitative Spectroscopy and Radiative Transfer
2007;107:17392.