Science and Technology of Welding and Joining
ISSN: 1362-1718 (Print) 1743-2936 (Online) Journal homepage: http://www.tandfonline.com/loi/ystw20
Visualization of gas metal arc welding on globular
to spray transition current
T. Methong, M. Shigeta, M. Tanaka, R. Ikeda, M. Matsushita & B. Poopat
To cite this article: T. Methong, M. Shigeta, M. Tanaka, R. Ikeda, M. Matsushita & B. Poopat
(2018) Visualization of gas metal arc welding on globular to spray transition current, Science and
Technology of Welding and Joining, 23:1, 87-94, DOI: 10.1080/13621718.2017.1344454
To link to this article: https://doi.org/10.1080/13621718.2017.1344454
Published online: 30 Jun 2017.
Submit your article to this journal
Article views: 248
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ystw20
SCIENCE AND TECHNOLOGY OF WELDING AND JOINING, 2018
VOL. 23, NO. 1, 87–94
https://doi.org/10.1080/13621718.2017.1344454
Visualization of gas metal arc welding on globular to spray transition current
T. Methong
a , M. Shigeta
a , M. Tanakaa , R. Ikedab , M. Matsushitab
and B. Poopatc
a Joining and Welding Research Institute, Osaka University, Ibaraki, Osaka, Japan; b JFE Steel Corporation, Chuo-ku, Chiba, Japan;
c Department of Production Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi, Thung Khru, Bangkok,
Thailand
ABSTRACT
ARTICLE HISTORY
The transfer modes in gas metal arc welding have important effects on welding quality. However,
present study of metal transfer modes is not yet fully understood. In this study, welding arcs was
visualised using the optical emission spectroscopy technique. The carbon steel wire electrode
was used for welding with 80% Ar + 20% CO2 shielding gas. The results showed that the globular
to spray transition current was 330–350 A. During globular to spray transition, argon,CO2 and
Fe plasma tended to gradually change from localising near the arc axis to a two-layer structure
having 11,000 K in high-temperature region away from the arc axis and around 7000 K in lowtemperature region near the arc axis.
Received 21 March 2017
Accepted 6 June 2017
Introduction
In gas metal arc welding (GMAW) is an arc welding
process that uses a plasma arc between a filler-metal
electrode and the weld pool. The high-temperature
plasma arc melts the electrode and forms a droplet at the
electrode tip. The droplet is detached and transferred
in the arc to the workpiece. The transfer of metal from
the electrode to the workpiece influences penetration,
bead morphology, fume generation, process stability
and spatter. Shielding gas had significantly effects on
arc characteristic, heat distribution and metal transfer
behaviour [1].
The addition of CO2 in the argon and CO2 gas
mixture shielding gas increases the transition current
and decreases the maximum droplet detachment frequency [2]. Poopat and Warinsiriruk [3] studied the
effect of CO2 addition in argon and CO2 gas mixture by using high-speed camera. The results showed
clear behaviour in three regions; short circuit, globular and spray transfer. Haidar and Lowke [4] reported
that the transition current from globular to spray transfer was around 350 A when using 75% Ar + 25% CO2 .
Soonrach and Poopat [5] studied the effect of oxygen
addition in argon and CO2 gas mixture on metal transfer behaviour in GMAW process. The results shows
that the mixed mode of globular and spray transfer
occurs when welding current above 250 A. However,
the plasma characteristic of globular to spray transition
has not yet been studied accurately.
According to the literature, it was found that the
arc plasma of GMAW with argon-rich shielding gas in
CONTACT T. Methong
567-0047, Japan
t.methong@jwri.osaka-u.ac.jp
KEYWORDS
Transition current; GMAW;
optical emission
spectroscopy; metal transfer;
plasma temperature; Fe
vapour
the spray transfer mode has a two-layer structure. The
high-temperature region exists away from the arc axis
and low temperature exists near the arc axis. This can
be explained by the higher metal vapour concentrations
near the arc axis will increase the radiative emission
coefficient for a given temperature [6–9].
There are several methods to diagnose thermal
plasma characteristics. The Langmuir probe is a simple and flexible tool. However, the interpretation of
obtained data is difficult because no simple, comprehensive underlying theory exists for high-temperature
thermal plasmas [10]. Consequently, optical emission
spectroscopy techniques have come to be one of the
most extensively applied diagnostic methods for study
of thermal plasmas. From measured spectroscopic
data, the plasma temperature can be determined using
atomic line intensities, intensity ratios of two or more
spectral lines or the ratios of line-to-continuum intensities [11]. Among these methods, the Fowler–Milne
method [12–15] and two-line relative intensity method
have been used widely due to their precisions and stabilities. However, only a few studies of GMAW processes
by spectroscopic methods were reported [15,16]. The
investigations up to date have not covered in the transition current range in terms of the arc plasma behaviours
and metal droplet transfer.
In this work, experimental measurements of the
plasma main parameters, plasma temperature distribution and metal vapour concentration, were presented. The optical emission spectroscopy using the
Fowler–Milne method and two-line relative intensity
Joining and Welding Research Institute, Osaka University, 11-1, Mihogaoka, Ibaraki, Osaka
© 2017 Institute of Materials, Minerals and Mining. Published by Taylor & Francis on behalf of the Institute.
88
T. METHONG ET AL.
method were carried out for the GMAW with 80%
Ar + 20% CO2 . The images of argon and Fe line spectra
and arc appearances were obtained by high-speed camera system which can provide information of dynamic
change of the plasma and electrodes. The spectroscopic
data were converted to the arc plasma temperature
and metal vapour concentration by the data processing
under the assumption of local thermodynamic equilibrium (LTE).
Temperature measurement methods
Fowler–Milne method for argon plasma
temperature measurement
Fowler–Milne method is based on the relationship
between the line spectrum intensity Inm and the temperature of arc plasma. The line spectrum intensity Inm
is given by
Inm =
Anm hc gn Nn exp(−En /kT)
·
4π λnm
Z(T)
(1)
where Anm is the transition probability from upper level
n to lower level m, h is the Planck’s constant (J s),
c is the speed of light in vacuum (m s−1 ),λnm is the
wavelength of the radiation emitted, Nn is the number
density (m−3 ) of upper level n [17], gn is the statistical weight of upper level n, En is the energy of upper
level n (eV), k is the Boltzmann’s constant (J K−1 ), T is
the temperature (K) and Z(T) is the partition function.
Equation (1) shows that the line spectrum intensity
depends on the arc plasma temperature. Therefore, the
plasma temperature can be inversely calculated by measuring the line spectrum intensity and convert using
Equation (1). Fowler–Milne method relies on the facts
that, for a plasma in LTE and at a constant pressure. The
line spectrum intensity Inm has a maximum at the welldefined temperature Tc known as the normal temperature. The number density of argon used in this study
can be obtained from a calculated values of computational modelling [17]. Figure 1 shows the calculated
normalised intensity as a function of temperature.
In this study, the arc plasma of GMAW by using 80%
Ar + 20% CO2 was observed. The Ar I 696.5 nm line
Figure 1. Calculated normalised intensity as a function of temperature.
was selected by the monochromator for the temperature measurement of argon plasma. The line spectrum
was normalised to its maximum value. The normal
temperature Tc of Ar I 696.5 nm line was 15,000 K.
Two-line relative intensity method for Fe plasma
temperature measurement
The two-line relative intensity method obtained the
plasma temperature from two-line spectra. The twoline spectra must belong to the same atomic or ionic
species. The relative line intensity is described using the
following equation.
I(1)
Anm(1) gn(1) λ(2)
En(1) − En(2)
=
exp −
I(2)
Anm(2) gn(2) λ(1)
kT
(2)
The temperature can be calculated from the relative
intensity using Equation (3) which is obtained by the
deformation of Equation (2)
T=−
En(1) − En(2)
1
· Anm(2) gn(2) λ(1) I(1)
k
ln Anm(1) gn(1) λ(2) I(2)
(3)
In this study, iron plasma temperature was determined
from two iron lines intensity of Fe I 537.1 and Fe I
538.3 nm. Figure 2 showed the calculation result of the
relationship between the two iron line intensity ratios
as a function of temperature according to Equation (3).
Metal vapour concentration measurement
The Fe vapour concentration was obtained by the relationship of mole fraction, Fe atoms number density and
temperature [17]. For calculation of the Fe vapour concentration, liner interpolation was used. The number
density of Fe atoms was given by the following equation
which was obtained by the deformation of Equation (1).
Nn =
Z(T)
· Inm
Anm gn hvn exp(−En /kT)
(4)
where Inm is intensity distribution of Fe line spectrum
obtained from the experiment, Nn is the number density (m−3 ), Z(T) is the partition function. The Fe vapour
Figure 2. Relationship between two iron line intensity ratio of
Fe I 537.1 and Fe I 538.3 nm as a function of temperature.
SCIENCE AND TECHNOLOGY OF WELDING AND JOINING
concentration can be obtained using the relationship
between Fe plasma temperature and the number density of Fe atoms of each Fe mole fraction in Figure 3
[17]. However, the intensity distribution obtained from
the experimental measurement was corrected by the
apparatus constant. The apparatus constant can be
obtained by comparing the intensity distribution by
89
Stark broadening technique [11] with that obtained
from the experimental measurement.
Experimental method
The optical measurement system has a capable of capturing four intensity images at the same time during
welding. Figures 4 and 5 showed a schematic illustration of the experimental set-up. It consists of three
main sections, including the arc plasma source, the
optical system and high-speed camera system. These
components are described in later paragraphs.
Arc plasma source and welding condition
Figure 3. Relationship between temperature and Fe atoms
number density for each mole fraction.
GMAW welding was carried out on a direct current
power source and common GMAW torch. The shielding gas was 80% Ar + 20% CO2 and the gas flow
rate was 20 L min−1 . The distance between the contact tip and the workpiece was 20 mm. The wire electrode JIS Z3312 YGW11 (AWS A5.18 ER70S-G) with
Figure 4. Optical system and monochromator set-up (a) and schematic diagram of the experimental set-up (b).
90
T. METHONG ET AL.
Figure 5. Optical system for high-speed colour images (a) and schematic diagram of the experimental set-up (b).
1.2 mm of diameter was used in this experiments. The
welding was performed in direct current electrode positive mode with welding current of 270–370 A on carbon
Figure 6. Arc appearance of GMAW of different welding currents.
steel grade SS400. The travel speed was 60 cm min−1 .
These welding conditions can produce both globular
and spray transfer mode in the welding arcs.
SCIENCE AND TECHNOLOGY OF WELDING AND JOINING
The spectral system
The spectral system consists of four monochromators
and high-speed cameras that are available for recording
Figure 7. Influence of welding current on droplet diameter and
droplet frequency using 1.2 mm YGW 11 wire electrode in 80%
Ar + 20% CO2 shielding gas.
91
intensity images of Ar I and Fe I lines at the same time.
The Ar I 696.5 nm line being the maximum emission
coefficient and the 694.0 nm line being the background
spectrum, respectively, were used to measure the
plasma temperature by using Fowler–Milne method.
The background spectrum was selected from arc light
spectra of GMAW with argon shielding gas and used for
subtracting the continuum spectra from a discrete spectral line. This technique can improve the accuracy and
precision of spectral measurements. The Fe I 537.1 and
538.3 nm line which were the maximum emission coefficient of Fe atoms were used to measure the Fe plasma
temperature by two-line relative intensity method.
The arc plasma radiation was focused on to a 0.5 mm
wide entrance slit of a monochromator (Princeton
Instruments, Acton Series 2300) by objective lens
and mirror system. The monochromator was the
Figure 8. Instantaneous distribution of the arc plasma temperature in GMAW with different welding current.
92
T. METHONG ET AL.
Czerny–Turner type and it had diffraction grating with
wavelength resolution of 0.4 nm.
High-speed camera system
The high-speed cameras (NAC, MEMRECAM GX-1)
were used to capture the arc plasma radiation. The
cameras were placed at the back of each monochromator. The 16-bit images were captured with the frame
rate of 2000 frames per second (fps) and the exposure
time of 1000 μs. These camera settings allowed recording plasma radiation at high-temporal resolution. The
optical emission spectroscopy system and monochromators were illustrated in Figure 4. Moreover, colour
images of arc plasma were captured by using highspeed digital colour camera (Vision Research, Phantom, New Jersey, USA, Miro eX). These camera settings allowed counting droplet frequency and droplet
diameter. The optical system for high-speed colour
images and schematic diagram were presented in
Figure 5.
Results and discussion
Arc appearance of GMAW with different welding
current
Figure 6 showed the colour images of the arc plasma
and metal droplet transfer at the time after detachment of droplet. A white dashed line indicated the
YGW 11 wire electrode with 1.2 mm in diameter. The
arc plasma images were captured from the direction
which was perpendicular to travel direction. The metal
droplets were formed at the tip of wire electrode and
then detached to the weld pool. It was seen that the
metal droplet size decreased with increase of welding
current. When the welding current was 270 A, the metal
droplet was obviously larger than the diameter of wire
Figure 9. Instantaneous distribution of the Fe plasma temperature in GMAW with different welding current.
SCIENCE AND TECHNOLOGY OF WELDING AND JOINING
electrode. It meant that the metal droplet transfer mode
was globular transfer. In contrast, the metal droplet
was smaller than the diameter of wire electrode and
the transfer mode became spray transfer mode at the
welding current of 350 A.
Figure 7 shows the influence of welding current on
droplet diameter and droplet frequency. The results
showed that the droplet diameter decreased as the welding current increased and it was smaller than the diameter of wire electrode when the welding current was
around 330 A. As shown in Figure 6, the globular to
spray transition was observed at the welding current
range of 330–350 A.
Arc plasma temperature distribution
Figure 8 shows the temperature distribution profile
with different welding current of the argon and CO2
plasma. These temperatures were calculated from the
93
Ar I 696.5 nm line using the Fowler–Milne method.
The welding current seems to have a little influence
on plasma temperature. The maximum temperature is
about 11,000 K. It can be clearly identified that the profile shapes of temperature change depending on metal
transfer mode. At the welding current of 310 A, the
transfer mode was globular and the high-temperature
region still included the arc axis. However, the hightemperature region moved away from the arc axis when
the transfer mode changes from globular to spray transfer. It should be noted that the plasma temperature
obtained from the Fowler–Milne method is valid only
in high-intensity distribution of Ar I line spectrum at
the region apart from the arc axis.
Figure 9 showed the temperature distribution profile
with different welding current of Fe plasma. These temperatures were calculated from the iron lines intensity
ratio of Fe I 537.1 and Fe I 538.3 nm using two-line relative intensity method. It can be seen that the profile
Figure 10. Instantaneous distribution of the Fe vapour concentration in GMAW with different welding current.
94
T. METHONG ET AL.
shapes of temperature changed little. In all welding
currents ranging from 310 to 370 A, the uniform distribution of Fe plasma was found in the arc axis with
the average temperature around 7500 K. These results
indicated that the temperature of Fe plasma was lower
than that of argon and CO2 plasma.
Behaviour of metal vapour in plasma
Figure 10 showed the metal vapour concentration with
different welding current and it was increased gradually with increasing welding current. It can be explained
that the increase of the welding current leads to the
increase of ohmic heating and promotes the vaporisation of Fe at the surface of wire electrode.
It is well known that the presence of metal vapour
decreases the plasma temperature by increasing the
plasma radiation [6–9,18]. That is the main factor for
the temperature fall of central region observed in spray
transfer mode. This means that the centre of arc column
in spray transfer mode had a high-radiative emission
coefficient and the temperature of arc become lower due
to the radiative cooling.
Conclusions
The emission spectroscopic study was performed to
study plasma characteristics in globular to spray transition for GMAW with 80% Ar + 20% CO2 shielding
gas. Plasma characteristics have been determined utilising the Fowler–Milne analysis and two-line relative
intensity analysis. The diagnostics allowed sufficiently
high-accurate determination of the plasma temperature
as well as estimation of the behaviour of metal vapour
concentration. The centre of the arc in globular transfer
mode was composed mainly of argon and CO2 as well as
Fe, where the plasma temperature was around 7500 K.
In this study, the globular to spray transition current
was determined by using high-speed camera and found
to be in the range of 330–350 A. During transition current, argon, CO2 and Fe plasma tended to gradually
change from localising near the arc axis to a two-layer
structure consisting of high-temperature region away
from the arc axis and low-temperature region near the
arc axis. Above welding current of 350 A, the metal
transfer was spray transfer mode which the plasma temperature reached over 11,000 K away from the arc axis
and around 7000 K of Fe-rich region near the arc axis.
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
T. Methong
M. Shigeta
http://orcid.org/0000-0001-6871-3478
http://orcid.org/0000-0001-9320-1351
References
[1] Annette OB. Welding handbook. Welding processes,
part 1. 9th ed. Miami (FL): American Welding Society;
2004. p. 148–149.
[2] Ogino Y, Hirata Y, Murphy AB. Numerical simulation of
GMAW process using Ar and an Ar–CO2 gas mixture.
Weld World. 2016;60:345–353.
[3] Poopat B, Warinsiriruk E. Investigation of parameters affecting behavior of metal transfer in MAG-M
welding process: the effect of shielding gas composition. Bangkok, Thailand: The Second South-east
Asia International Welding Congress; 2010. p. 128–
137.
[4] Haidar J, Lowke J. Effect of CO2 shielding gas on metal
droplet formation in arc welding. IEEE Trans Plasma
Sci. 1997;25(5):931–936.
[5] Soonrach R, Poopat B. Effect of oxygen addition in
argon/carbon dioxide gas mixture on metal transfer
behavior in gas metal arc welding. IJMPE. 2016;4(9):
50–54.
[6] Rouffet ME, Wendt W, Goett G, et al. Corrigendum:
spectroscopic investigation of the high-current phase
of a pulsed GMAW process. J Phys D Appl Phys.
2012;45:189501.
[7] Valensi F, Pellerin S, Boutaghane A, et al. Plasma
diagnostics in gas metal arc welding by optical emission spectroscopy. J Phys D Appl Phys. 2010;43:
434002.
[8] Schnick M, Fuessel U, Hertel M, et al. Modelling
of gas–metal arc welding taking into account metal
vapour. J Phys D Appl Phys. 2010;43:434008.
[9] Zielinska S, Musiol K, Dzierzega K, et al. Investigations
of GMAW plasma by optical emission spectroscopy.
Plasma Sourc Sci Technol. 2007;16:832–838.
[10] Fanara C, Vilarinho L. Electrical characterization of
atmospheric pressure arc plasmas. Eur Phys J D.
2004;28:241–251.
[11] Griem HR. Principles of plasma spectroscopy. Cambridge: Cambridge University Press; 1997. p. 291–293.
[12] Olsen HN. The electric arc as a light source for
quantitative spectroscopy. J Quant Spec Radiat Transf.
1963;3:305–333.
[13] Vilarinho LO, Scotti A. Proposal for a modified FowlerMilne method to determine the temperature profile in
TIG welding at low currents. J Braz Soc Mech Sci Eng.
2004;26(1):34–39.
[14] Wang L, Hua X, Xiao X, et al. Analysis of arc physical
property of pulsed tungsten inert gas welding based on
Fowler-Milne method. J Shanghai Jiaotong Univ (Sci).
2013;18:343–347.
[15] Ma S, Gao H, Wu L. Modified Fowler–Milne
method for the spectroscopic determination of
thermal plasma temperature without the measurement
of continuum radiation. Rev Sci Instrum. 2011;82:
013104.
[16] Ton H. Physical properties of the plasma-MIG welding
arc. J Phys D Appl Phys. 1975;8:992.
[17] Murphy AB, Arundelli CJ. Transport coefficients of
argon, nitrogen, oxygen, argon-nitrogen, and argonoxygen plasmas. Plasma Chem Plasma Process. 1994;
14:451–490.
[18] Schnick M, Füssel U, Hertel M. Metal vapour causes a
central minimum in arc temperature in gas–metal arc
welding through increased radiative emission. J Phys D
Appl Phys. 2010;43:022001.