Professional Documents
Culture Documents
S Arungalai Vendan Liang Gao Akhil Garg P Kavitha G Dhivy 1
S Arungalai Vendan Liang Gao Akhil Garg P Kavitha G Dhivy 1
Arungalai Vendan · Liang Gao
Akhil Garg · P. Kavitha · G. Dhivyasri
Rahul SG
Interdisciplinary
Treatment to Arc
Welding Power
Sources
Interdisciplinary Treatment to Arc Welding Power
Sources
S. Arungalai Vendan Liang Gao
•
Rahul SG
Interdisciplinary Treatment
to Arc Welding Power
Sources
123
S. Arungalai Vendan P. Kavitha
School of Electrical Engineering School of Electrical Engineering
VIT University VIT University
Vellore, Tamil Nadu Vellore, Tamil Nadu
India India
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Contents
1 Welding an Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Basics of Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Electric Charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Electric Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.3 Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.4 Electrical Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.5 Heat Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.6 Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Equivalent Electrical Analogous Representation
of Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Arc Welding as a Confluence of Several Disciplines . . . . . . . . . . 6
1.4 Applications of Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2 Insight into Arc Welding Power Source Terminologies . . . . . . . . . . . 15
2.1 Critical Arc Power Source Terminologies for Welding . . . . . . . . 15
2.1.1 Arc Plasma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.2 Arc Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.3 Arc Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.4 Arc Blow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.5 Pinch Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.6 Arc Shielding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Influence of Power Source Parameters on Weldment . . . . . . . . . 18
2.2.1 Open-Circuit Voltage (OCV) . . . . . . . . . . . . . . . . . . . . 18
2.2.2 Arc Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.3 Welding Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.4 Electrode Polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.5 Power Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.6 Duty Cycle and Current Rating . . . . . . . . . . . . . . . . . . . 22
2.2.7 Class of Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
v
vi Contents
This book categorically presents the mechanical, metallurgical and electrical termi-
nologies and the associated science for welding power sources. Key focus is laid
on arc welding processes owing to its growing demand in versatile applications.
Figure 1.1 illustrates the types of welding processes.
Arc welding is a fusion process wherein joining of materials is accomplished by
the application of sufficient electrical potential to an electrode and creating an arc
between the electrode and the work piece. The electric arc serves as a conductive
path for the current flow via the electrode. Consumable electrode conducts the current
while melting to provide filler material for the joint [1]. If the electrode is of a non-
consumable type, it simply conducts current between electrode and workpiece and
a separate filler material is used. The arc welding process is classified under the
following categories based on the work setup configuration (Fig 1.1):
(i) Shielded metal arc welding (SMAC) or manual metal arc welding (MMAW)
or stick Welding,
(ii) Gas Metal Arc Welding (GMAW) or Metal Inert Gas Welding (MIG),
(iii) Flux-Cored Arc Welding (FCAW),
(iv) Gas Tungsten Arc Welding (GTAW) or tungsten Inert Gas Welding (TIG),
(v) Submerged Arc Welding (SAW),
(vi) Cold Metal Transfer (CMT).
Power source is establishing the electric arc. It is a vital component in a welding
process that is responsible for delivering specific voltage–current characteristics with
a well-defined control to achieve preciseness while regulating the static and dynamic
characteristics. In general, the power source is designed to operate with 230 V single
phase or 440 V three phase while delivering a current of 0–300 A and a voltage
between 0 and 90 V. The operating modes vary based on AC/DC current, which
is identified and finalized depending on the material to be welded. The welding
process requires low voltage and high current to minimize the heat generated, which
is a challenge for the researchers. The dynamic behaviour of power sources and their
modernization through automation adds to the existing complications [2].
© Springer Nature Singapore Pte Ltd. 2019 1
S. A. Vendan et al., Interdisciplinary Treatment to Arc Welding Power Sources,
https://doi.org/10.1007/978-981-13-0806-2_1
2 1 Welding an Overview
Welding Process
• Friction
• Carbon arc • Ultrasonic
• Metal arc • Diffusion
• Butt
• Oxy-acetylene • Plasma arc • Explosive
• Spot • Electron beam
• Air,-Acetylene • Gas Metal Arc • Magnetically
• Seam • Laser
• Oxy-Hydrogen (MIG) impelled Arc Butt
• Projection
• Gas Tungsten Arc Welding (MIAB)
(TIG) • Induction Pressure
• Cold Metal Welding (IPW)
Transfer (CMT) • Flash Butt welding
(FBW)
• Magnetic Pulse
Welding (MPW) –
cold welding process
Despite its key role, the literature based on the behaviour of welding
power source and its implication on weld efficiency that accounts for mechani-
cal/metallurgical/electrical phenomena or behaviour are limited. Moreover, most of
the power source studies are presented by the authors in their native language which
poses more complexity and raises ambiguous pointers.
This book attempts to explore the inter- and intra-dependent parameters of the
joining process as a whole and the physics underlying. The physics of welding met-
allurgy involves the material being subjected to heat and pressure. This necessitates
the inclusion of welding power source, control module, electrode setup and the spec-
imens to be welded. Eventually, the weld specimens are subjected to non-destructive
analysis, mechanical and metallurgical characterization as illustrated in Fig. 1.2 to
establish the feasibility of that particular process for the chosen material.
Governing to the available literature of arc welding power sources, the following
terminologies are to be understood from the electrical perspective of welding. All
matter is made up of atoms and each atom contains fundamental particles; electrons,
protons and neutrons. Electrons, the negatively charged particles have the capability
to move from one place to another; and protons, the positively charged particles
remains stationary at normal conditions.
1.1 Basics of Arc Welding 3
Mechanical Behaviour
Input Parameters
(Current, Voltage, Metallurgical Characteristics
Properties
Power factor...) Tensile strength
WELDING SYSTEM Micro and Macro Images
Bead Width SEM analysis
Impact TEM analysis
Hardness
Power
Source
Convertors Non Destructive Analysis
Radiography
Intermediate
Ultrasonic
System
Liquid penetrant test(LPT)
Dye penetrant Test(DPT)
Controllers Visual Inspection
Optical Inspection
Eddy Current
Magnetic particle test
Thermal Analysis
Electrical current is a measure of the amount of electrical charge transferred per unit
of time. It represents the flow of electrons through a conductive material, such as a
metal wire. It is measured in amperes (A).
1.1.3 Voltage
external energy forces the electrons to move from one atom to another atom or
away from that atom. This energy is termed as the electromotive force or voltage.
Without voltage, there is no current flow. The transfer of electrons from a negative
to a positive charge throughout the length of a conductor in presence of a voltage
potential constitutes the electrical current. Materials vary in their ability to transfer
electrons. Insulators have a strong electron bonding and their atoms resist the free
movement of electrons. Hence, they are poor conductors of electricity. However,
conductors, on the other side have more number of free electrons due to their weak
electron bonding, thereby exhibiting superior electron flow leading to conduction.
The unit of voltage is the volts (V).
Electrical resistance can be said as the resistance to the flow of current. Resistance is
best understood through Ohm’s Law which emphasizes that even though the voltage
and current changes, the ratio of their relationship remains as per Eq. (1.4)
V I∗R (1.1)
where
V Voltage (V),
I Current (A),
R Resistance ().
The theory of electrical resistance is of immense importance in the arc welding
process. And it is this resistance in the air gap between the electrode and the base
material that contributes to the conversion of electrical energy into heat energy. In arc
welding process, this energy conversion causes the workpiece and the electrode/filler
material to melt resulting in the formation of weld pool (crater).
where
H Heat density (KJ/cm2 ),
ƒ Heat transfer efficiency,
V Voltage (V),
I Current (A),
v Travel velocity of heat source (mm/s).
In arc welding, as arc voltage forces the electrons to move faster, the energy
generated by them is used partially to overcome the resistance created by the arc
gap. Thus, this energy is converted into heat energy and the temperature increases to
the point where it brings the materials to a molten state.
1.1.6 Power
Watt is the term defined as the amount of power required to maintain a current of 1 A
at a pressure of 1 V. The circuit voltage that comes is typically a constant factor, but
the current drawn from the utility company depends on the number of watts required
to run the electrical appliance. Power is denoted by Eq. (1.4) as follows:
W V∗I (1.3)
where
W Power (W),
V Voltage (V),
I Current (A).
The current required by an electrical equipment can be calculated by dividing the
watts rating of the device by the primary voltage for which it is designed is shown
in Eq. (1.4),
I W/V (1.4)
The arc generated causes a heating effect which makes the workpiece and consumable
electrode or the filler material to melt and fuse. To understand the principle of energy
conversion, a fundamental knowledge on electricity and the electrical aspects of
welding are important. A typical arc welding process is depicted along with the
equivalent electrical analogous circuit which is shown in Fig. 1.3.
The equivalent electrical analogous notations are presented in Table 1.1.
6 1 Welding an Overview
Mains Electrode
Electric Welding Holder
Supply Cable
Arc Welding Welding current Welding Arc
Power Electrode
Source
Return Cable
Return Workpeice
Clamp
I2R Power or
diode in reverse
bias
Current I
AC/DC
Voltage Load
Source Resistor
(Or)
Ground
The flow diagram shown presents a detailed overview on the arc welding power
source and its interdependent terminologies. All the subsections in the flow diagram
are explained briefly in the subsequent chapters (Fig. 1.4).
Although most of the welding processes at the time of their developments were used
for repairing in production, gradually they started gaining relevance to actual compo-
nent manufacturing. Presently welding is widely being used in fabrication of pressure
vessels, bridges, building structures, aircraft and space crafts, railway coaches and
1.4 Applications of Arc Welding
Table 1.2 Application of arc welding. (Arc Welding in Ships, Aerospace and Automobile Applications - Google Search [Online]. Avail-
able: https://www.google.com/search?biw=1536&bih=710&tbm=isch&sa=1&ei=1eotW-z7I8nrvATroI4o&q=arc+welding+in+ships%2C+aerospace+and+aut
omobile+applications&oq=arc+welding+in+ships%2C+aerospace+and+automobile+applications&gs_l=img.3...15329.22442.0.22806.33.19.0.)
Application industry Type of arc welding Components (name with images)
Aerospace GTAW/TIG, CMT, MIG/GTAW Fuel pipes
(continued)
Table 1.2 (continued)
Application industry Type of arc welding Components (name with images)
Automotive GMAW/MIG Automobile components
1.4 Applications of Arc Welding
(continued)
9
Table 1.2 (continued)
10
(continued)
1 Welding an Overview
Table 1.2 (continued)
Application industry Type of arc welding Components (name with images)
Railways FCAW, GMAW, SAW, SMAW Railway tracks
1.4 Applications of Arc Welding
(continued)
11
Table 1.2 (continued)
12
(continued)
1 Welding an Overview
Table 1.2 (continued)
Application industry Type of arc welding Components (name with images)
Construction industry MMA, TIG Bridges
1.4 Applications of Arc Welding
Metal structures
13
14 1 Welding an Overview
References
Welding arc is an electric current flowing between two electrodes through an ionized
gas column. Arc stability is a critical problem which influences the metal transfer
during welding. When the arc is stable, metal transfer is uniform with minimum spat-
ter [1]. The terminologies of major concerns during the joining process are presented
below.
Arc plasma as shown in Fig. 2.1 is the ionized state of welding gas, which is a mixture
of equal amounts of ions and electrons [2]. To study the effects of the arc plasma
over the weld pool, four main factors to be considered are heat flux, current density,
shear stress and arc pressure. Heat flux and current density have a direct relationship
towards the weld pool. While, increase in shear stress (shear stress is the amount of
force per unit area perpendicular to the axle of the member) results in outward flow
at the top of weld pool and increase in arc pressure (force exerted by arc per unit
area) results in a more concave surface of weld pool.
Where,
1. Gas plasma,
2. Protection nozzle,
3. Shielding gas,
4. Electrode,
5. Nozzle constriction,
6. Electric arc.
Arc efficiency is defined as the heat input to the metal divided by the total heat
energy of the arc [2]. Or simply, it is defined as the amount of arc energy delivered
to the workpiece [3]. It is essential to know the arc efficiency to measure the melting
efficiency using heat-flow models and experimentation. The arc efficiency of non-
consumable electrodes is lower than that of the consumable electrodes [4]. Heat input
can be calculated by the Eq. (2.1) [3]
Arc voltage*Arc current
Heat input KJ/cm2 ∗ Arc efficiency (2.1)
Welding speed*1000
Arc stability is influenced by arc power and metal transfer modes. Spattering is the
emission of metals during welding process, which is a negative effect of poor arc
stability. The properties of ideal and stable welding arc are
(i) Constant shape of all transferred materials,
(ii) Constant length of arc and
(iii) Low or no spattering.
Another factor affecting arc stability is the shielding gas mixture of CO2. The
stability is low when the shielding gas mixture has high CO2 .
2.1 Critical Arc Power Source Terminologies for Welding 17
Arc blow is a phenomenon where the arc tends to deviate from the welding point [5].
This is because of the imbalance in the magnetic field surrounding the workpiece.
This phenomenon occurs in the following three situations:
(i) Presence of magnetic materials around welding arc,
(ii) Changes in the direction of current,
(iii) Presence of magnetic materials at the end of the plate.
Arc blow mainly occurs in DC welding currents and can be reduced using AC
current. Arc with low voltage is stiff and resists deflection than the arc with a high
voltage. The melting rate is influenced by the heating effects of current which is
given by the Eq. (2.2)
β
Melting Rate α I + ∗ ls ∗ I 2 (2.2)
αω
where
α and β are constants
l s Electrode resistivity (m),
α Cross-sectional area of wire (m2 ),
l s Welding current (A).
The arc carrying the current is always surrounded by a magnetic field [5]. In arc
welding process the direction of electromagnetic force varies, and it depends on the
direction of flow of welding current. The magnetic field has a force directed towards
the centre of arc, called as Lorentz force. With increase in current, the arc radial
constriction increases due to increased magnetic force. This process is termed as
pinch effect (as shown in Fig. 2.2) [6]. The pinch force is directly related with the
welding current and electrode wire diameter and influences drop detachment to the
weld pool [7].
At elevated temperatures, metals tend to react with oxygen and nitrogen in the atmo-
sphere. When the molten metal pool is exposed to air it results in formation of oxides
and nitrides on metals which deteriorates the toughness and strength of the weld
joint. Arc welding processes require a protective shield of gas, vapour or slag which
is termed as arc shielding [2].
18 2 Insight into Arc Welding Power Source Terminologies
Critical electrical power source parameters influencing the final weldment are
explained briefly in the following context.
The open-circuit voltage refers to the potential difference between the two terminals
of a welding power source at no load condition. The selection of an optimal value
of OCV depends on type of workpiece, type of welding current, polarity, type of
arc welding, type of electrode and its composition, etc. The base metal with a lower
ionization potential such as potassium (K), sodium (Na), calcium (Ca) requires a
low OCV than the base metal with a higher ionization potential such as oxygen (O),
fluorine (F) and nitrogen (N). Among all arc welding processes, GTAW requires
less OCV since it uses tungsten electrode which has higher free electron emitting
capability [8].
• When an AC source is used then the setting up of correct OCV is very important
for arc stability.
• AC power source requires higher OCV than DC because AC is prone to arc insta-
bility, since the welding current continuously changes its magnitude and direction
for every half cycle. But, in DC-type power sources the current remains constant.
• Very high value of OCV also results in electric shock.
• OCV for transformers ranges from 70 to 90 V; and for rectifiers it is 20–60 V.
The OCV of a power source is always 5–10 V greater than the arc voltage.
2.2 Influence of Power Source Parameters on Weldment 19
Electrode Electrode
Tip Tip
Cathode (-) Anode (+)
Welding Welding
Arc Arc
Arc Voltage Arc Voltage
Cathode drop
Arc length (cm)
Arc Column
Anode drop
Arc Voltage
(Volts)
Arc voltage shown in Fig. 2.3 is the potential difference between the electrode tip
and the workpiece when there is current flow [9].
Any fluctuation in the arc length affects the resistance to the flow of current, which
in turn affects the arc voltage. Increase in arc length increases arc voltage. Increase
of electrode extension increases the arc voltage and increases the electrode heating
due to increase in resistance to current flow. The arc voltage can be categorized into
three zones as shown in Fig. 2.4;
The cathode drop is near the electrode tip where the potential drop occurs swiftly
in a narrow zone. Near the anode drop zone, the voltage drop is fast on the anode
surface and is uniform. Sharp drop in arc potential near cathode and near anode is
attributed to the cooling effect of electrode and workpieces.
There are two types of electric current; Alternating Current (AC) and Direct Current
(DC). DC is a stream of negatively charged electrons flowing through the cable,
moving from the negative pole to the positive pole, whose flow is always unidirec-
tional. While, AC is achieved when the power supply switches between positive and
negative polarity at a frequency of 50 or 60 Hz [10].
As welding current increases, the weld penetration also increases and vice versa,
which is illustrated in Fig. 2.5. Among the several types of arc welding, only TIG
welding of aluminium utilizes AC because the continuous change in polarity of
AC helps in tearing the layer of oxide on the surface of weld pool. Most welding
applications are implemented with negative DC polarity to get the required depth of
penetration.
2.2 Influence of Power Source Parameters on Weldment 21
When DC power source is used polarity of the electrode has a significant role. Polarity
in arc welding designates the direction of current flow. The types of connection are
Direct Current Electrode Positive and Direct Current Electrode Negative [10].
i) DCEP (Direct Current Electrode Positive)—The continuous flow of avalanche
of electrons in small passage produces the arc. After the electrons are emitted
from the base material (negative polarity), they are accelerated due by the poten-
tial difference and strike the electrode (positive polarity) with high velocity. The
kinetic energy of the electrons is partially converted partially to thermal energy
upon striking, which results in heat dissipation at the electrode [11]. Two-thirds
of arc heat is generated at electrode; while, only one-third heat is generated
at base material. Eventually, the electrode melts swiftly and weld penetration
increases.
(ii) DCEN (Direct Current Electrode Negative)—The electrons flow from electrode
to the base plates. Subsequently, more heat is generated at the base material as
compared to electrode. So, depth of penetration reduces.
Power factor is defined as the ratio of actual power (KW) used to produce the rated
load to the apparent power drawn from the supply (KVA) as shown in Eq. (2.3)
Real Power (KW)
Power factor (2.3)
Apparant Power (KVA)
Real power (W) is the one that does a work. For example, torque produced in
rotating a machine. It is a real value. Apparent power (VAR) is the power that bounces
back and forth between source and load. It is an imaginary value. Power factor is an
important parameter because it gives a measure of effectiveness of the real power
utilization in the system [10]. It also denotes the measure of distortion of the line
voltage and line current and the presence phase shift between them. The power source
should have high power factor for utilizing the energy effectively. When power source
has higher power factor it reduces reactive power in system and there by reduces
power consumption and cost. The power factor is high in welding transformers than
the other power sources. The advantages of having high power factor are improved
voltage quality, occasional voltages drop, less transmission losses, more economic
operation, reduced reactive power thereby reduced power consumption, reduced
cable cross-sectional area, etc. [12].
22 2 Insight into Arc Welding Power Source Terminologies
Duty cycle refers to the percentage of welding time to the total welding cycle (welding
type + rest time). Total welding time is taken as approximately 5 min. Duty cycle and
associated current are important for the safety of power sources. The welding current
drawn at a duty cycle can be expressed by the Eq. (2.4)
DR ∗ IR2 I100
2
∗ D100
2
(2.4)
where
I Current (A) at 100% duty cycle,
D100 100% duty cycle,
IR Current (A) at required duty cycle,
DR Required duty cycle.
At 100% duty cycle, minimum current will be drawn to avoid excessive heating of
winding and power source. Reduction of duty cycle results in withdrawal of higher
magnitude of welding current [12]. When high current is drawn from power source,
the flow of current causes heating of coil and cables due to electrical resistance of
heating. During such time welding, must be stopped to protect the welding power
source (resting time). Maximum current that can be drawn from a power source
at a duty cycle depends on size of wire and type of insulation. Effective cooling
systems reduce the withdrawal of higher current from the power source. Duty cycle
is inversely proportional to the current drawn.
Static characteristic of a welding power source tells the variations in voltage with
respect to current when a pure resistive load is connected to the power source. They
are also called as constant characteristics.
The CC power sources are designed to maintain the welding current approximately
constant, during changes in the arc length and arc voltage. The V–I curve shown in
Fig. 2.6 is called as droop characteristics because there exists a negative slope [8].
When consumable electrodes are used then the electrode melting rate remains con-
stant without making them to get stuck with the workpiece. The CC power sources are
suitable for thicker consumable electrodes and for non-consumable electrodes such
as tungsten electrode since the welding current is limited and maintained constant.
They are widely used in MMA, TIG welding, etc. [12].
In the CV power source, the output voltage is set on the source which then estab-
lishes the required arc voltage. The power source supplies the welding current that
is necessary to melt the electrode at a rate to maintain the arc voltage limit that is
pre-set. These types of power sources are used only for MIG welding and GMAW
[13]. The speed of electrode wire feeder drive becomes an important parameter to
control the amount of welding current, since the wire feed rate is directly proportional
to the welding current. When the power source is used in combination with a con-
stant electrode wire feed system, it then results in a system with a self-regulating arc
Constant Current
Characteristics
Voltage (Volts)
Current (Amperes)
24 2 Insight into Arc Welding Power Source Terminologies
Constant Voltage
Characteristics
Voltage (Volts)
Current (Amperes)
Constant
Arc length
Voltage
Long
35
Medium
25 Small
length. This power source maintains a constant arc voltage by increasing or decreas-
ing the welding current since the arc length can sometimes vary due to external or
internal fluctuations such as operator errors, puddle movement, plate irregularities,
etc., thereby the V-I characteristic curve shown in Fig. 2.7, which is called as flat
characteristics. But practically, it has a slight negative slope due to internal resistance
and inductance of the power source.
Dynamic characteristics is essential for success of welding because during arc weld-
ing, the arc is subjected to severe transients and fluctuations in arc voltage due to
minor changes in arc length. Thereby, the arc current rapidly fluctuates for a very
short time. Dynamic characteristics depicts the instantaneous variations in welding
voltage and current relationships caused by the metal transfer across arc, welding
parameters, electrical parameters of power source such as resistance, inductance
and capacitance, and determines the process stability. The load is subjected to rapid
changes from arc condition like open circuit and with load state to short circuit and
vice versa. During short circuit, there is rapid rise in welding current and electrode
wire tip melts and falls on weld pool resulting in a spatter. With advancement in tech-
nology, automated welding is available that necessitates stabilizing of the welding
parameters and controlling the process remotely using a computer [13].
A variety of welding power sources are available for electric arc welding and some of
the major types are discussed briefly. Arc welding power sources are majorly divided
into two main categories: static type and rotating type as shown in Fig. 2.9.
Transformer-type power source as shown in Figs. 2.10 and 2.11 produce only Alter-
nating Current (AC). They are termed as ‘Welding Transformers’. All the AC-type
sources are utilized with single-phase primary power and are constant current type
[14].
26 2 Insight into Arc Welding Power Source Terminologies
ARC WELDING
POWER
SOURCES
STATIC TYPE
ROTATING TYPE
Uses commercially
Uses electromechanical
generated power to energize
components during
a transformer that steps
absence of commercial
down the line voltage to
power
usable welding voltage
Welding Rectifiers
Only Alternating Produce DC
Current (welding Current Or DC & Motor coupled to
Gasoline Or
Transformer) AC Welding generator or
Diesel Engine
Currentt alternator
All AC types uses
single-phase Constant current or
Constant current Constant voltage or
type combination of
both Both have AC or DC
welding power sources or
combinations of both.
They also use Constant
voltage and Constant
current source
Ac Power
50/60Hz
Rectifier-type power sources also termed as welding rectifiers are introduced in mid-
1970s. They produce DC or, both AC and DC welding current. They are being utilized
with either single-phase or three-phase input supply. They consist of a transformer
and perform rectification of AC to DC using power semiconductor devices such
as selenium rectifiers, silicon diodes or Silicon-Controlled Rectifiers (SCR). These
2.4 Classification of Arc Welding Power Sources 27
Ac Power
50/60Hz
Voltage Welding
Control
transformation AC AC Rectifier DC Filter output
circuit
and isolation Power
Inverter-type power sources are introduced in the 1940s and became successful in
the 1970s. They boost the frequency as much as 1000 times of the input frequency.
They are beneficial in reducing the coil turns of transformers and core area resulting
in smaller and lighter machine than the conventional type transformers. They sup-
port both single- and 3-phase supplies. Their applications include GTAW, GMAW,
SMAW, FCAW processes. Figures 2.14 and 2.15 depict the inverter-type power
source. First, the high-voltage and low-current AC is rectified into DC which is then
filtered and fed to the switching components. Switching occurs at very high rates
28 2 Insight into Arc Welding Power Source Terminologies
Rectifier type
DC
Control Sensing
Circuit circuit
type of power sources. They are also used in combination with microcontrollers to
access and modify the electrical characteristics of the power source in real time
through software [14].
Motor generator power source as shown in Figs. 2.16 and 2.17 consist of an electric
motor coupled to a generator or alternator that produces the desired welding power.
These machines produce superior weld qualities. However, due to the moving parts
they require frequent maintenance. Welding generators can be of both AC and DC
type [15].
30 2 Insight into Arc Welding Power Source Terminologies
Shaft
coupling
Earlier, arc welding power sources used motor generator types to produce welding
current, which followed the principle of current generation by the movement of
conductor through a magnetic field. These engine-driven electromechanical devices
find its applications, where there is no availability of commercial power. They are
generally called as rotating-type power sources and are classified further as alternators
that produce AC and generators that produce DC [14]. A gasoline or diesel engine
is coupled to a generator or alternator that produces the desired welding power as
shown in Figs. 2.18 and 2.19. They are used extensively on applications beyond
commercial power lines and as mobile units. Both rotating types can deliver either
AC or DC welding power, or a combination of both. Both types are available as
constant current or constant voltage types.
Shaft
coupling
Gasoline or
Generator /
Diesel Power Out
Alternator
Engine
Some of the fundamental terminologies addressing the literature are briefly explained
below.
2.5.1 Diode
Depending on the voltage applied across a diode, it operates in one of the three
regions as shown in Fig. 2.21;
1. Forward bias: The diode is ‘ON’ and current conduction takes place when the
voltage across the diode is greater than the forward voltage.
2. Reverse bias: This is the ‘OFF’ state of a diode, where the voltage is less than
forward voltage but greater than reverse breakdown voltage. A very small amount
of current in the order of nA flows which is called as reverse saturation current.
3. Breakdown: When the voltage applied across the diode becomes very large and
negative, lots of current flows in the reverse direction from cathode to anode.
2.5.2 BJT
2.5.3 MOSFET
There are two ways in which a MOSFET works. The first is depletion mode;
when there is no gate pulse, the channel exhibits its maximum conductance. As
the gate voltage increases either towards positive or negative direction, the channel
conductivity decreases. The second is called enhancement mode; when there is no
voltage on the gate, there is no channel, and the device does not conduct (Fig. 2.26).
A channel is only produced by the application of a voltage to the gate. The greater
the gate voltage, better the conduction. Since the oxide layer is so thin MOSFET is
susceptible to permanent damage due to electrostatic charges.
MOSFET operating regions as shown in Fig. 2.27 are as follows:
1. Cut-off region: The MOSFET is off and there is no current flow. In this region,
MOSFET behaves like an open switch.
2. Ohmic or linear region: The drain-to-source current increases with an increase
in the value of drain-to-source voltage. They act as amplifiers in this region.
Pinch-off voltage
Cut-off region
The insulated gate bipolar transistor is a cross between a conventional BJT and
a MOSFET. It combines the insulated gate technology of the MOSFET with the
output performance characteristics of a conventional BJT [15]. The three terminals
of IGBT are collector, emitter and gate as shown in Fig. 2.28.
Collector and emitter are associated with the conductance path of current, while
its gate terminal controls the device.
Since IGBT is a voltage-controlled device, it requires only a small voltage on
the gate to maintain conduction unlike BJTs which require that the base current
is continuously supplied in a sufficient quantity to maintain the saturation. Also,
IGBT is a unidirectional device it switches the current in forward direction only
from collector to emitter. An IGBT is simply turned on or off by activating and
deactivating its gate terminal. Applying a positive voltage signal in the gate and the
emitter keeps device in its ‘ON’ state, while making the input gate signal zero results
in ‘OFF’ state. The IGBT characteristics are shown in Fig. 2.29.
The IGBT is presently one of the most popular devices due to its wide ratings,
switching speed of about 100 kHz, easy voltage drive and a square safe operating
area devoid of a second breakdown region.
Fully “on”
Active region
Fully “off”
SCRs are used in power-switching applications. In the, OFF state, SCR restricts
current flow. The SCR characteristics are shown in Fig. 2.31. When the gate-to-
cathode current exceeds a threshold value, the device turns ON and conducts. The
SCR remains the ON state even after gate current is removed so long as the current
through the device exceeds the holding current. If the gate is pulsed and the current
through the device is below latching current, the device will remain in the OFF state.
Holding current is the minimum current which must pass through from anode
to cathode for it to remain in the ‘ON’ state [15]. The minimum anode current is
required to maintain a thyristor in ‘ON’ state immediately after a thyristor has be
triggered ON. SCR can be turned ON by the application of a gate pulse when a
forward bias voltage is present at the main terminals. However, being regenerative
or ‘latching’, it cannot be turned off via the gate terminals, at the extremely high
amplification factor of the gate. The two main types of SCRs are: Converter grade
or Phase Control thyristor and Inverter grade thyristor.
38 2 Insight into Arc Welding Power Source Terminologies
Pulse width modulators are powerful tools for controlling analog circuits with a
microprocessor’s digital outputs. They widen the pulse width or narrow the pulse
width without changing the time as shown in Fig. 2.32. PWM is employed in a wide
variety of applications, ranging from measurement and communications to power
control and conversion [15]. The main advantage of using PWM is the power loss
becomes low in the switching devices.
2.5 Power Sources Components Briefing 39
2.5.7 Microprocessor
2.5.8 Microcontroller
FPGAs are semiconductor devices that are based around a matrix of Config-
urable Logic Blocks (CLBs) connected via programmable interconnects as shown
in Fig. 2.35. FPGAs can be reprogrammed to desired application or functionality
requirements after manufacturing [15].
40 2 Insight into Arc Welding Power Source Terminologies
and switching losses are decreased [29]. The power source is enhanced with single
phase welding current based on new energy conversion methods by replacing tradi-
tional transformer-rectifier systems [30]. Then fuzzy-based power sources are sug-
gested. The designed power source is implemented with the phase-shift full-bridge
Zero-Voltage Zero-Current Source (ZVZCS) [31]. Phase-shift full-bridge topology
is used in the high-power arc welding/cutting inverter power source that consisted of
three control methods; Sliding mode control, PID control and fuzzy control [32]. The
control technology widely used in inverter welding consists of full-bridge-isolated
pulse width modulation providing high current, low voltage and high power. The
Zero-Voltage Switching (ZVS) is realized easily by leading leg however the lag-
ging is difficult, that increases switching loss and Electromagnetic Inference (EMI)
[29]. Lebedev, V.A., et al., analysed the single-phase welding current source using
transformers and condenser [33].
To improve the blocking voltage capability of semiconductor devices, two meth-
ods are defined. The first method is to combine MOS and bipolar physics and the
second one is Wide-Bandgap (WBG) semiconductor material that replaces the Si
material. IGBT came out from the first method which is widely employed in power
electronics [34]. Second method with WBG material, where the silicon is replaced
by the silicon carbide in the power devices to reduce the resistance in drift regions.
The benefits of silicon carbide devices are demonstrated in [35] and have realized
the desired features such as high-power density and lightweight in medium voltage
application [36]. The optimal selection of welding current and arc voltage deter-
mines the quality of weld joints [37], which contributes to study of the power source
characteristics of power source under dynamic and static operating conditions. By
arriving at optimal parameter ranges for welding, the parameter settings are done in
the power sources. A reliable method based on Artificial Neural Networks (ANN) is
developed for detection of magnetic core saturation in welding transformers, with no
additional sensor. The proposed ANN is trained by the resilient backpropagation rule
using samples, then the dynamic model of the welding system and its measurements
are obtained [38]. The invention of high-speed solid-state devices leads to greater
saving of energy in the welding processes.
Later an arc welding Switched Mode Power Supply (SMPS)-type power source
is developed using simulation software PSIM, the proposed converter is built and
verified by experimental prototype measurements [1]. In the IGBT-type power
source used in electron beam welding, transient behaviour of the system is enhanced
by altering the characterization circuit and stability of the system [39]. Inverter
power supply for welding achieves the Zero-Voltage Source (ZVS). The controlled
resonant inductor technology is employed which reduced the loss of duty cycle
of the phase-shift full-bridge converter. The efficiency of the power supply is
increased, and the switching losses are decreased [29]. The current trends in various
welding methods are reported by JFE [40]. Recently, the performance of welding
parameters: Probability Density Distributions (PDDs), artificial neural network
comprising of Self-Organizing Maps (SOM), voltage and current is analysed using
inverter and generator power source [41]. The list of papers on evolution of welding
2.6 Evolution of Arc Welding Power Sources 43
1400
Min Max
1200
1000
800
Amps
600
400
200
0
1980 1985 1990 1995 2000 2004 2010 2014 2016
Year
Fig. 2.36 Year-wise variation of average welding current in the welding processes
power source is presented in the following table. The year wise variation of average
welding current is plotted in Fig. 2.36. Table 2.1 presents various papers pertaining
to the evolution of arc welding power sources.
In 1998 Du Dong et al., considered the inverter power source for the robotic arc
welding. The inverter power source shall meet the following conditions:
(i) The power source should have a control property, constant output current reg-
ulation at a wide range and improvise the welding quality.
(ii) The power source should have high duty cycle and rated output current to
ensure a long life time.
(iii) The power source should have electromagnetic-compatible interfaces.
With the implementation of modern electronic power techniques, the IGBT (Isolated-
Gate Bipolar Transistor) inverter has acceptable control properties and high switch-
ing frequency (more than 20 kHz). The IGBT power sources are electromagnetically
compatible, therefore, each unit must have the same phase, frequency (25 kHz) and
synchronized switching. The inverter-type power source system consists of substan-
tial number of independent, parallel IGBT inverter units combined with a controlled
single-chip microcomputer unit. This system has good static and dynamic indices,
100% duty cycle and high-power capacity. They not only meet the welding require-
ment, but in addition its lower cost leads to batch production of inverter power sources
[42]. Instantaneous output current control method is used in these power sources.
The major characteristics of this method are as follows;
44 2 Insight into Arc Welding Power Source Terminologies
(i) When compared to traditional inverter arc welding machine it reduces spatter
formation.
(ii) Short-circuit metal transfer process is stabilized and enhanced the bead state.
The authors proposed a power supply based on the voltage drop caused by leakage
inductance in the transformer. This leakage inductance is proportional to the excita-
tion frequency. The study also reveals that, the inverter drive for a transformer causes
effects in feedback mode. The equivalent load impedance of the inverter drive varies
approximately ±20% more than the variation in sine wave operation. Further testing
is required to improve the controlling methods [43]. Few authors analysed the weld-
ing power source equipped with digital filtering process. Welding power sources
46 2 Insight into Arc Welding Power Source Terminologies
with accurate control grasps instantaneous weld phenomenon, processes the data
ultrafast and performs control in real time. Research and developments are ongoing
in this domain and there are several reports available on weld instability where mul-
tiple abnormal voltage events’ occurrences are introduced as a side effect of digital
filtering [44].
In welding, low-voltage and high-current sources are used with either AC or
DC welding currents as an effective approach. To implement them in low-power
applications, the use of zero-voltage and zero-current switching of the inverter is
introduced [45]. The switching frequency can be increased up to 40 kHz due to the
applied zero-current, zero-voltage switching mode, however the switching losses
remain low. The iron core size of the transformer can be considerably reduced when
the transformer is supplied with increased frequency. Some authors modified the
topologies of the basic DC–DC converter structure and it is observed from the studies
the sensitivity of the system reduces [46]. With the evolution of fast converter and
inverter grade switches, waveform-controlled power sources are developed. The list
of waveform-controlled welding power sources and weldability for different arc
welding processes are illustrated in Tables 2.2 and 2.3 respectively.
In 2009, an inverter power source is designed in which, the increase of volt-
age in the individual elements is restricted. Thus, process of Weld with Modulated
Current (WMC) by double modulation of welding current provides an efficient
control [7]. Later, an inverter-type power source equipped with a microcontroller
(MCF5213CAF8) and FPGA is experimented and presented to attain high reliabil-
ity, fast dynamic response, stable arc length control and good welding quality [47].
Also, the compatibility of the arc welding power source is analysed.
The equipment of the power source are gradually replaced from transformer
devices to rectifier and inverter devices which had many advantages such as pre-
defined sets of parameters, smooth adjustment of welding parameters, smaller and
lighter size. They have also introduced other welding methods such as Cold Metal
Transfer (CMT) and Surface Tension Transfer (STT). The Electromagnetic Com-
patibility (EMC) in power sources show some disadvantages and relatively resistant
to originate interferences from the mains. Using semiconductors devices, the inter-
ference problems are eliminated. Also, the power supply operation is a nonlinear
character due to core saturation, which is due to the harmonic occurrences.
The FPGA digital control strategy is adopted to analyse the voltage, current and
switching frequency. Thus, switching losses occurring are reduced using the phase of
the current in synchronized gate signal by incorporating predictive digital algorithm
[48]. In the following Table 2.4, a review of few switch-based techniques in welding
power source-related papers are presented.
2.7 Switch-Based Techniques Adopted for Welding Power Sources 47
Table 2.2 Waveforms and its applications for different processes of welding
Welding Waveforms Materials used
AC-TIG Sine wave Aluminium, Al alloys,
Soft square wave Stainless steel, Copper
Rectangular wave
Sawtooth wave
Asymmetric (wave shape is
not balanced either positive or
negative)
DC-TIG Square wave Stainless, Mild, Copper Alloy,
Nickel alloy, Titanium alloy,
Low alloy
AC/DC MMA Square wave Steel, Iron, Aluminium,
Nickel, Copper alloy
AC SAW Sine wave Carbon steels, Low-alloy
Square Wave steels, Stainless steels,
Nickel-based alloys
DC SAW Square wave Steel, Nonferrous metals,
Aluminium
AC-GTAW Square wave Aluminium, Copper alloys,
Magnesium alloys
DC-MIG Triangular wave Carbon steel, Stainless steel,
Aluminium, Thin sheet metals
DC-RSW Square wave Aluminium, Magnesium,
Sine wave Inconel, Nickel, Nickel silver,
Monel, Low-carbon steel,
Low-alloy steel, Stainless steel
AC/DC CMT Sine wave Cold Rolled Steel, Galvanneal,
Square wave Stainless steel, Aluminium,
and Copper
Single-chip microcomputer is regulated in individual power unit consisting of parallel connected IGBTs to make the power source system to work over
a wide range of continuous output current with good dynamic properties
Low cost, good dynamic and static indices, high-power capacity and 100% duty cycle is produced
Chae et al. (1999) [58] Output current control using
single-chip microcontroller
Inverter circuit topology is implemented to improve welding performance, reduced spatter and improved bead state
2 Insight into Arc Welding Power Source Terminologies
(continued)
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
Inverter frequency is varied to easily control load current. The use of transformer leakage inductance for current control for obtaining low-voltage and
high-current power sources
Uezono & Hongjun (2008) Digital filtering process with digital
[44] calculation technology
2.7 Switch-Based Techniques Adopted for Welding Power Sources
For abnormal voltage events, the optimal value is different depending on the material. Digital filtering removes the noise in real time and proved its
advantages in manufacturing efforts
(continued)
49
Table 2.4 (continued)
50
AHC adjusts itself automatically to any change in iron core size and in the value of applied DC bus voltage. Proposed AHC assures a very short rise
time of the welding current and best possible utilization of transformer iron core
(continued)
2 Insight into Arc Welding Power Source Terminologies
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
Control algorithms of the power section in the IPS enables efficient control
(continued)
51
Table 2.4 (continued)
52
The software design of FPGA and software design of operating system µC/OS-II on MCF5213CAF80 are proposed which overcomes the
disadvantages of analogous control technology and single-chip control. Microcontroller and FPGA-based control system for MIG welding has faster
response, better reliability and more stable arc length
(continued)
2 Insight into Arc Welding Power Source Terminologies
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
The detailed analysis on impact of sub-harmonics and inter-harmonics on a power system is done. Efficient detection of system parameters is performed
by monitoring the distortion introduced into the system
(continued)
2.7 Switch-Based Techniques Adopted for Welding Power Sources
53
Table 2.4 (continued)
54
The algorithm generates two complementary signals to drive the 4QSWs, required to synchronize the switches and the output current of the transformer.
Switching losses on 4QSW are minimized by the proper synchronization of the gate signals with the phase of the input current using a predictive digital
algorithm. Phase lags are measured by zero-block crossing
(continued)
2 Insight into Arc Welding Power Source Terminologies
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
Delivered output voltage of 25 V and output current of 200A. Their operation at high frequencies has minimized the size, weight and reduced current
ripples. The power source is controlled using the control logic developed using Xilinx logic cell array
Janson and JWik [62] Series-to-parallel switching of
inductor and capacitor in
oscillatory circuits
2.7 Switch-Based Techniques Adopted for Welding Power Sources
AC-DC resonant is developed. The reactive power of the inductor and capacitor compensates each other in both parallel and series connection so the
power factor is maintained unity. There are no requirement for filters and reactive power compensative circuits
55
(continued)
Table 2.4 (continued)
56
Low EMI. Most overlooked features of this design are minimal engineering time and cost-effectiveness. As a future work, the author recommended
building and testing of a prototype with further investigations on size and economic design
Shklovski and Janson [63] Switch-mode load-resonant
converter using PSIM
The authors proposed that inherent short-circuit current limitation and a fast-parametrical response to the load would result in good weldment. It has
wide operating range from no-load to short-circuit without overloading, high power factor, THD of 0.1 PF between 0.99 and 1, and low transient
overshoots
2 Insight into Arc Welding Power Source Terminologies
(continued)
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
The proposed circuit consists of a bridgeless (BL) zeta converter in the front and a full bridge (FB) buck converter for high-frequency isolation. This
converter eliminated conduction losses and provided power factor correction
Narula et al. [65] Modified power factor
(PF)-corrected single stage and two
switches-isolated zeta converter
2.7 Switch-Based Techniques Adopted for Welding Power Sources
The proposed converter operates in a discontinuous inductor current mode (DICM) for achieving good PF correction and reduced control complexity
(continued)
57
Table 2.4 (continued)
58
A closed-loop adaptive control strategy to minimize switching losses for a lightweight ZVZCS-PS-FB-Convertor built with IGBTs and MOSFETs
Snubber capacitors are used to reduce switch turn-off loss and saturable inductors are used to prevent discharge of the blocking capacitor and to limit
the current rise
Chae et al. [67] Arc welding inverter
Proposed a new control algorithm for instantaneous current control using a single-chip microprocessor for CO2 arc welding inverter to achieve proper
metal transfer, reduced spattering and good digital control
(continued)
2 Insight into Arc Welding Power Source Terminologies
Table 2.4 (continued)
Author(s) Switch-based methodologies Block diagrams and inference
A three-phase welding inverter power supply made of IGBTs with power factor correction and validated for high quality and unity power factor. The
proposed design aims to minimize inactive power flow problems when the power source works under open-load condition. The author suggested further
research to improve the performance characteristics when the power source is used for arc GTAW and GMAW
Zhu et al. [69] Constant current IGBT-based
inverter-type power source
2.7 Switch-Based Techniques Adopted for Welding Power Sources
A constant current IGBT-based inverter-type power source for CO2 arc welding using LPC213-embedded ARM unit to realize output waveform control
digitally. The developed power source utilizes PWM control to adjust the waveform parameters, which maintained a stable welding condition and
improves the welding performance. The author suggested to further investigate and add devices in the main circuit and develop high-frequency power
source to improve the welding property
(continued)
59
Table 2.4 (continued)
60
A fuzzy adaptive PID control and PWM control algorithm for IGBT-based switching-inverter power source for obtaining smooth, constant DC output.
The proposal aimed to overcome nonlinearities and time variations and improve the steady-state precision and dynamic performance of the system
2 Insight into Arc Welding Power Source Terminologies
2.8 Literature Addressing Power Source Parameters 61
Fig. 2.37 Probability distribution of short-circuit period with CO2 , Ar/CO2 , and TIME shielding
gases. I 90 A, U 21 V (as illustrated in [57])
Based on the review of parameters associated with power sources, the following
points are discussed.
Suban and Tušek [6] have emphasized that poor arc stability causes spattering
which results in material losses, production time wastage, anaesthetic appearance
on workpiece and quality deterioration. The experimental study revealed a stable,
spatter-free weldment when carbon dioxide shielding gas (82% Ar/18% CO2 gas
mixture) is used for short-circuit metal transfer. TIME mixture (gas mixture of 65%
Ar/26.5% He/8% CO2 /0.5% O2 ) for spray transfer. (Variation of the curves during
the short-circuit periods are smaller than in the two other cases as in Fig. 2.37).
Palani and Murugan [49] have performed a study on pulsed current GMAW. The
study aimed at selection of optimal parameters such as average current, peak current,
peak time, base current, feed rate, frequency and shielding gas based on trial and error
experimentation. The review reveals that only a few researchers have used Design of
Experiments (DOE) to perform experiments for selecting the pulse parameters and
to study their effect on weldment.
A. K. Paul [50] has analysed the power electronic techniques and technologies for
the optimal performance of arc welding process. The analysis concluded that power
electronic technology is beneficiary in minimizing the energy usage, maximized
productivity, enhanced quality and minimal environmental impact.
Skrzyniecki et al. [51] performed an experimental study on power source stability
for MIG/MAG welding processes using LabVIEW software and a computer. The
computer-controlled process resulted in good weldment. The author emphasizes on
the choice of suggesting advanced methods for recording and online monitoring of
data.
Kolasa et al. [52] conducted an experimental study on dynamic characteristics
of TIG arc welding that is supplied with square and triangular current waveforms
as shown in Figs. 2.38 and 2.39. Two anodes; steel molten anode and water-cooled
copper anode are utilized. The outcomes revealed that for very low frequencies arc
voltage coincides with static characteristics, and at high frequencies, the arc dynamic
62 2 Insight into Arc Welding Power Source Terminologies
characteristics becomes like that of a constant resistance. It is observed that the square
waveform of current is optimal.
Lucas and Murch [21] conducted an experimental study on reduction of OCV by
developing a two new square wave power sources; squared AC and switched DC.
The experiments indicated that welding arc is sustained at 62 V on squared sine wave
and 40 V on square wave as shown in Figs. 2.40 and 2.41 respectively. The developed
power source had a greater current change rate during polarity reversal with reduced
OCV requirements.
Jinhong et al. [53] developed an optimal nonlinear model of control system for
arc welding inverter using MATLAB and proposed a Computer-aided analysis of
dynamic process. The dynamic process of open circuit to short circuit is simulated
and it is proven as an effective approach.
Yamane et al. [54] proposed a new power source characteristic to improve the arc
stability to avoid spattering. Experiments are carried out to compare the nonlinear
power source characteristic with the conventional characteristic under 100% CO2
shielding gas. The conventional power source characteristic revealed the occurrence
of unregulated metal transfer and spattering as shown in Fig. 2.42. While using
of nonlinear power source characteristic, the spatter is reduced, and arc stability is
improved as shown in Fig. 2.43.
2.8 Literature Addressing Power Source Parameters 63
Fig. 2.44 Typical photographs of surface appearance of weld bead with respect to spatter generation
under different welding processes of a MS-GMAW, b GMAW, c P-GMAW, and d DP-GMAW [55]
Fig. 2.46 Dynamic characteristic recorded for power source during welding with controlled load
[56]
References
1. J. Shklovski, K. Janson, Development of constant-power source for arc welding, in 13th Bien-
nial Baltic Electronics Conference, pp. 255–258 (2012)
2. H. Latifi, R. Suoranta, J. Martikainen, M. Pirinen, P. Kah, Usability of arc types in industrial
welding. Int. J. Mat. Eng. 9(1), 1–12 (2014)
3. T.W. Eagar, The Physics of Arc Welding Processes (1990), pp. 17–19
4. J.N. DuPont, A.R. Marder, Thermal efficiency of arc welding processes. Weld. Res. Suppl.
406–416(1995)
5. M. Schiedermayer, Improving reliability of inverter based welding machines. Weld. J. 76(2)
(1997)
6. M. Suban, J. Tušek, Methods for determination of arc stability. J. Mat. Process. Technol.
143–144, 430–437 (2003)
7. A.F. Knyaz’kov, S.A. Knyaz’kov, K.I. Dementsev, An inverter power source for welding with
modulated current. Weld. Int. 23(12), 957–962 (2009)
8. V. Agelidis, O. Anaya-Lara, T. Miller, E. Acha, Power Electronic Control in Electrical Systems
(Elsevier, India, 2002)
9. Arc Welding Power Sources:NPTEL online course (2017). Available: http://nptel.ac.in/course
s/112107090/module3/lecture1/lecture1.pdf
10. T.E. GROUP, Basics of Arc Welding. Available: http://www.esabna.com/euweb/awtclesson1_
21.htm. Accessed 2017
11. Pintu, Arc Welding Polarity—Types, Effects, Selection and Pros & Cons. Avail-
able: http://www.minaprem.com/physics-of-welding/arc-welding-polarity-types-effects-selec
tion-pros-and-cons/. Accessed 2017
12. T.L.E. Company, Weld Penetration Variables: Lincon Electric (1999). Accessed 2017
13. W.S. Service, Maritime Welding Handbook (2010)
14. M. Electric, Guidelines to GTAW process (2003). Available: http://manuals.chudov.com/Mill
er/Miller-TIG-Handbook.pdf. Accessed 2017
15. M.H. Rashid, Power Electronics Handbook (Elsevier)
68 2 Insight into Arc Welding Power Source Terminologies
16. A.F. Manz, Inductance vs. Slope Control for Gas Metal Arc Power, vol. 9 (1969), pp. 707–7012
17. E.H. Dagget, A power supply for pulsed power welding. IEEE Spectr. 1, 67–70 (1968)
18. J.D. Van Wyk, J.A. Ferreira, Transistor invertor design optimization in frequency range above
5KHz up to 50 KVA. IEEE Trans. Ind. Appl. 2, 296–302 (1983)
19. A. Kolasa. A. Matsunawa, Y. Arata, Experimental study on dynamic properties of power sources
for MIG/MAG Welding. Trans. JWRI, 2, 255–265 (1985)
20. J.F. Lancaster, The physics of fusion welding Part-1: the electric arc welding. IEEE
Proc.—Electric Power Appl. 134(5), 233–254 (1987)
21. W. Lucas, M.G. Murch, Arc reignition characteristics when welding with sine-wave and square-
wave power supplies. IEEE Proc.—Electr. Power Appl. 134(86), 348–354 (1987)
22. T. Mita. A. Sakabe, T. Yokoo, Quantitative estimates of arc stability for CO2 gas shielded arc
welding. Weld. Int. 2(2), 152–159 (1988)
23. I.V. Pentegov, S.V. Rymar, V.P. Latanskij, Optimisation of welding parameters in power sources
with condenser voltage multipliers. Weld. Int. 12(2), 89–91 (1988)
24. M.J.M. Hermans, G. Den Ouden, Modelling of heat transfer in short circuiting gas metal arc
welding. Sci. Technol. Weld. Joining 3(3), 135–138 (1998)
25. M.J. Kang, S. Rhee, The statistical models for estimating the amount of spatter in the short
circuit transfer mode of GMAW. Weld. J. (USA). 80(1) (2001)
26. Z. Jinhong, L. Wenlin, S. Yaowu, Study on the dynamic process of arc welding inverter. in
Proceedings of the Power Electronics and Motion Control Conference, 2000. IPEMC 2000,
vol. 1, pp. 308–322 (2000)
27. A.V. Agunov, M.V. Agunov, G.M. Korotkova, V.I. Stolbov, A.A. Shevtsov, Energy character-
istics of the power source-welding arc system. Weld. Int. 16(12), 966–969 (2002)
28. Y.N. Konovalov, Comparison of the properties of the MAGMA-350 universal inverter power
source for the welding arc with traditional rectifiers for mechanized welding. Weld. Int. 28(4),
317–319 (2014)
29. J. Ji, X. Hu, Z. Hua, G. Zeng, L. Guo, Research of soft switching arc welding inverter power
supply with high-frequency and high-power. Int. Power Electron. Appl. Conf. Exposition, 924
(2014)
30. V.A. Lebedev, V.S. Romanyuk, Single-phase welding current power sources for mechanised
carbon dioxide welding. Weld. Int. 18(6), 489–493 (2004)
31. G.R. Zhu, Z. Liu, X. Li, B.Y. Liu, S.X. Duan, Y. Kang, Research on digital soft-switch weld-
ing/cutting inverter power source, in IEEE International Conference on Power Electronics and
Drive Systems, vol. 7 (2007), pp. 325–329
32. Z. Guo-rong, L. Zhao, Z. Ai-yun, Y. Mi, S.X. Duan, Y. Kang, Sliding mode control and PI control
for arc welding/cutting inverter, in IEEE International Conference In Industrial Technology
(2008), pp. 1–4
33. V.A. Lebedev, A.V. Motrii, A.D. Glushchenko, N.I. Postolatii, Experimental examination of
welding current sources for semiautomatic welding equipment poared from 220 V single-phase
mains. Weld. Int. 21(6), 454–457 (2007)
34. B.J. Baliga, Power Mosfets—In Fundamentals of Power Semiconductor Devices (2008)
35. J.S. Glaser, J.J. Nasadoski, P.A. Losee, A.S. Kashyap, K.S. Matocha, J.L. Garrett, L.D. Ste-
vanovic, Direct comparison of silicon and silicon carbide power transistors in high-frequency
hard-switched applications, in Applied Power Electronics Conference and Exposition (APEC),
pp. 1049–1056 (2011)
36. R.C.S. Machado, J.C. Braz Filho, Next generation arc welding machines based on Silicon
Carbide MOSFETS and high frequency planar magnetics, in IEEE 13th Brazilian Power Elec-
tronics Conference and 1st Southern Power Electronics Conference (2015)
37. K. Devakumaran, P.K. Ghosh, Thermal characteristics of weld and HAZ during pulse current
gas metal arc weld bead deposition on HSLA steel plate. Mat. Manuf. Process 25(7), 616–630
(2010)
38. K. Dezelak, J. Pihler, G. Stumberger, B. Klopcic, D. Dolinar, Artificial neural network applied
for detection of magnetization level in the magnetic core of a welding transformer. IEEE Trans
Mag. 2, 46 (2010)
References 69
63. J. Shklovski, K. Janson, Development of constant-power source for arc welding. Proc. Bienn.
Balt. Electron. Conf. BEC, 225–258 (2012)
64. B. Singh, G. Bhuvaneswari, S. Narula, PFC bridgeless converter for welding power supply
with improved power quality, in IEEE International Conference on Power Electronic Drives
Energy System PEDES (2014)
65. B. Singh, G. Bhuvaneswari, S. Narula, Power factor corrected welding power supply using
modified zeta converter. in IEEE Conference (2016) pp. 617–625
66. W. Fischer, F. Werther, H. Mecke, Soft switching inverter power source for arc welding, in
EPE’79, Trondheim, vol. 1 (1998) pp. 333–337
67. Y.M. Chae, J.S. Gho, H.S. Mok, G.H. Choe, W.S. Shin, A New Instantaneous Output Current
Control Method for Inverter Arc Welding Machine (1999), pp. 0–5
68. C. Shu-Jun, Y. Shu-Yan, W. Dong-Ping, Z. Hua, H. Ji-qiang, Novel three-phase welding inverter
power supply with high power factor, pp. 1113–1118 (2003)
69. J. Zhu, H. Shi, K. Lu, K.Y. Lee, Study on IGBT inverter power supply for CO2 arc welding
and embedded arm based waveform control, in International Conference on Mechatronics and
Automation, 2007. ICMA 2007, pp. 2634–2638
70. S.Z. Wei, W. Qiong, X. Peng, J. De Li, A Switching-inverter power controller based on fuzzy
adaptive PID, in Proceedings of the 6th International Forum on Strategic Technology, IFOST
2011 (2011), vol. 2, pp. 695–699
Chapter 3
Control Terminologies and Schemes
for Arc Welding Processes
Some of the fundamental concepts involved in control system are presented in this
chapter.
3.1.1 Process
Electrode holder
Electrode cable
Electrode
Filler metal
Welding machine
AC or DC Power
Arc source
Work
3.1.2 System
3.1.4 Parameters/Variables
Parameters/variables are entities that are measured and controlled. Example: In arc
welding, welding current, arc length, arc voltage, depth of penetration and bead width
are few key parameters.
(i) Controlled Variable: It is the quantity or variable that is measured and controlled.
Example: Maintaining constant arc voltage in a Constant Voltage (CV) type
power source. Here, arc voltage is the controller variable.
3.1 Control System Terminologies 73
(ii) Manipulated Variable: It is the quantity or variable that is varied by the controller
to alter the value of the controlled variable. Example: To maintain a constant
voltage (controlled variable) in CV type power source, the gating signal applied
to a thyristor for modulating the pulse width is the manipulated variable.
3.1.5 Control
Control refers to measurement of controlled variable of the system and applying the
manipulated variable to the system to correct or limit deviation of the measured value
from a desired value.
3.1.6 Disturbances
A disturbance is an external signal that tends to adversely affect the value of the output
of a system. Example: Sudden change in arc length due to environmental effects
(operator mishandling, mechanical vibrations, etc.) are considered as disturbance.
3.1.7 Setpoint
3.1.8 Feedback
Feedback is an integral part of control system that permits the output dynamic
response to be altered without modifying the system components. The process param-
eter is measured using measuring devices called sensors or transducers. Example: In
arc welding process, the welding current is controlled after measuring it with a hall
effect sensor. The current is measured using a hall effect sensor and its feedback to
the input and the error is calculated.
74 3 Control Terminologies and Schemes for Arc Welding Processes
3.1.9 Error
The deviation of the measured value from setpoint is called as error. Example: If
the operator sets a current value of 70 A in a CC type power source and if the hall
effect sensor measurement indicates 68 A of current, then the error is the difference
between both which is 2 A. This error signal is fed to the controller.
Any physical system can be represented in the form of a differential equation denoted
by
After taking Laplace transform, the above time domain equation, we can obtain the
transfer function. The transfer function of a linear, time-invariant system is defined
as the ratio of the Laplace transform of the output (response function) to the Laplace
transform of the input (driving function) with zero initial conditions. The general
representation of transfer function is shown
C(s) bm s m + bm−1 s m−1 + · · · + b0
G(s) n (3.2)
R(s) an s + an−1 s n−1 + · · · + a0
Roots of the numerator and denominator polynomials are called as zeroes and
poles, respectively.
A process in which the output parameter has no effect on the input parameter is called
as an open loop system. It is a system without feedback and does not automatically
correct the variation in output as shown in Fig. 3.2.
Example: A current limit sensor is an open loop system. The sensor gets activated
when the current exceeds a prefixed limit and it starts activating buzzer. The system
needs human interaction to intervene and stop the buzzer (as shown in Fig. 3.3).
3.1 Control System Terminologies 75
error
Setpoint +
Controller Process Output
-
Feedback
Setpoint
error
current +
Arc welding
Controller Output
process
-
A process in which the output parameter alters the input parameter to achieve the
set point is called as closed-loop system as shown in Fig. 3.4. The output signal is
measured and feedback to the input. Hence, the closed-loop control system is often
referred as feedback control system.
A feedback control system tends to maintain a relationship between one parameter
and another parameter by comparing both the values. Their difference is fed as error
signal to the controller. Example: In a CC type power source for a GMAW process,
after a hall effect sensor measures the welding current, the difference between mea-
sured current and setpoint current is calculated and the corresponding error signal is
fed back to the controller (as shown in Fig. 3.5).
76 3 Control Terminologies and Schemes for Arc Welding Processes
There are two methods to analyse the functioning of a control system; time domain
analysis and frequency domain analysis.
(i) In time domain analysis: The system response is a function of time. The two
components of time domain response are transient response and steady-state
response. The time domain analysis can be applied only when the mathematical
model and nature of input is known. Time domain analysis is the preferred one
for welding applications.
Example: Transient analysis can be performed on welding current measured
from any arc welding process.
(ii) In frequency domain analysis: The system response is a function of frequency.
The two components of frequency domain response are magnitude response and
phase response. The frequency response of a system is defined as the steady-
state response to a sinusoidal signal. One major advantage of frequency response
is that we can use the data obtained from measurements on physical systems
without deriving its mathematical mode.
Both the analysis can be applied only when the mathematical model and the nature
of the input are known.
Order of the system refers to the highest derivative of the controlled quantity in the
equation describing the dynamics of the system, i.e. the highest power of ‘s’ in the
denominator polynomial of the transfer function without cancelling the common
terms in the numerator and denominator polynomial.
where
T Time constant.
Example: Thermocouple is an example of first-order system. Measurement of tem-
perature distribution in the Heat Affected Zones (HAZ) of the weldment is performed
using a thermocouple. As the temperature of the thermocouple rises, the differential
temperature between surroundings and HAZ decreases, the rate of heat flow and the
rate of change of temperature of thermocouple decreases proportionately.
C(s) ωn2
2 (3.4)
R(s) s + 2ξωn s + ωn2
where
ωn Natural frequency (rad/s)
ξ Damping ratio.
Example: An RLC circuit an example of a second order system. The two energy
storing elements are inductor and capacitor. Inductor stores energy in terms of mag-
netic field. Capacitor stores energy in terms of charge.
78 3 Control Terminologies and Schemes for Arc Welding Processes
3.2.5 Linearity
Linearity is the behaviour of a system in which the output signal varies in direct
proportion to the input signal. It is attributed to homogeneity and superposition.
Example: An amplifier amplifies the voltage signal proportional to the gain.
3.2.6 Sensitivity
Sensitivity is the relative change in the total transfer function of a system for a given
relative change in some parameter in the system.
The fundamental controllers are based on a fixed structure controller family, the
so-called PID controller family. These controllers have proven to be robust and
extremely beneficial in the control of important applications. PID stands for Pro-
portional, Integral and Derivative controller. The basic control actions are shown in
Table 3.1.
In general, PID controller is used in most closed-loop industrial processes despite
the abundance of sophisticated and advanced controllers. The general equation of a
PID control is represented by
t de
u(t) k p e(t) + ki ∫ e(t)dt + kd (3.5)
0 dt
3.4 Stability Analysis 79
Stability is defined as the state of a system that remains in a constant state unless
affected by an external action and which returns to a constant state when the external
action is removed. If a system is stable for all variations of its parameters, then the
system is called stable system. If the system is stable for limited range of variations
of its parameters, then the system is called conditionally stable system. If the system
does not adapt to the variations in parameter, then it becomes unstable. The concept
of stability is explained using an example shown in Fig. 3.6.
Force is represented as
F m∗a (3.6)
where
F Force (N)
m mass (Kg)
a acceleration (m/s2 )
G centre of gravity
From the figure, it is seen that the cone in position (a) experiences unstable equi-
librium as it topples over left and right directions. In the position (b) the cone falls
back. The cone remains in neutral equilibrium at position (c).
The stability of a feedback system is directly related to the location of the roots of
the characteristic equation (denominator polynomial) of the system transfer function.
When the poles of the closed loop transfer function of a given system are in the right-
half of the S-plane, the system becomes unstable. When the poles of the system are in
the left-half plane the system is stable. When the poles are located on the imaginary
axis ( jω-axis) then the system is said to be marginally or conditionally stable.
Some of the basic stability test criteria are given as follows:
The function of a feedback control system is to ensure that the closed-loop system
has desirable dynamic and steady-state response characteristics. Ideally, we would
like the closed loop system to satisfy the following performance criteria:
(i) Stability and robustness;
(ii) Minimal disturbance providing good disturbance rejection;
(iii) Good setpoint tracking (i.e. rapid smooth responses to set point changes);
(iv) Steady-state error (offset) is eliminated;
(v) Zero steady-state error;
(vi) Avoiding excessive control action.
Welding controller improves the efficient welding and the essential features are given
as follows:
Mains Supply
230V or 440V
Output
Input to Sensors/ Parameters (Welding
(Weld
Power Source Transducers current, arc voltage)
Efficiency)
Parameter Correction
Control Unit
(gating signals)
The variables used in control are sensed, scaled and fed to the processor where the
control schemes are applied. The signals include AC input voltage, welding cur-
rent, OCV, arc voltage, arc length and sometimes additional voltage like capacitor
or inductor voltage are used. The AC voltage is sensed using potential transformers.
For sensing DC voltages, hall effect sensors, optocouplers, isolation amplifiers are
used. The current signals are sensed using current transformers, shunt resistors, hall
effect current sensors, etc. The parameters are scaled and fed to the controller via
ADC channels which can be then used as a feedback signal in the control schemes
of the voltage and current controllers. These sensed parameters can be used to anal-
yse performance indices such as power factor, Total Harmonic Distortion (THD),
distortion factor, ripple factor, etc. [2].
The desired gating signal involves deriving the gating signals for the solid-state
switches used in power sources. Some of the controllers used in deriving gating sig-
nals are PID controller, pulse width modulator (PWM) controller, Neural network-
based controllers, SMC, etc. They are implemented through hardware using Digital
Signal Processor (DSP) or microcontrollers or digitally build using computer soft-
ware. The derived gating signals are measured and obtained from dedicated hardware
units and fed to optocoupler for isolation, and then given to the power switches. The
full integration of the control unit, interfacing unit and power source unit are in devel-
opment such that they are cost-effective, compact, highly efficient, reduced weight
and reliable.
The control methods and algorithms adopted for welding power source are repre-
sented as the backbone for improving the efficiency in welding. Different control
techniques have been proposed in research reports to achieve better weld quality.
Some published literature reviewed consist of various implementation to overcome
the shortcomings in welding. To obtain a uniform weld quality a microprocessor-
based online feedback control is suggested in 1985. In reference to the experimental
results, they obtained almost closer to the desired weld strength with the help of pro-
portional control algorithm. The amount of energy utilized for performing welding
action uses thyristor controls. Welding current, voltage and energy are the parameters
which depend upon different control methods. The desired welding current are con-
trolled using SCR controller. Based on this control system, input voltage to the SCR
controller is determined. To produce the controlling input voltage the authors partic-
ularly used P algorithm [3]. In 1995, robotic arc welding thyristor type power source
is investigated with the control of microprocessor measuring registering facility. The
investigations found the direct, functional relationship between the dynamic and
technological properties of power sources of the electric arc welding. The dynamic
properties affect only the magnitude of the spatter during welding whereas the weld
quality is based on welding parameters [4]. During 2005, the experimental investiga-
tions reveal that consistent weld penetration and fusion area is achieved by adopting
constant voltage control technique. Although welding setup, base material and torch
angle inconsistency are unavoidable to a certain extent. Using constant current tech-
nique coated surface damage and weld spatter are considerably reduced [5]. In 2006,
pulsed current GMAW power source is investigated based on selection parameters.
Peak duration, peak current and pulse parameters, play a major role in finding the
properties of weld bead [6]. In 2008, the following evolutions happened. The digital
control technology became reliable, exact and flexible. The phase-shift full-bridge
3.7 Controller Schemes Adopted for Welding Power Sources 83
topology is used in the high-power arc welding/cutting inverter power source that
consisted of three control methods; Sliding mode control, PID control and fuzzy
control [7].
Later in 2011, TIG welding power sources incorporated fuzzy controller to per-
form rapid current control [8]. The power source with conventional voltage con-
trolled short circuit mechanisms resulted in accomplishment of stable and efficient
weld quality [9]. Silva et al. determined the following factors of MIG/MAG weld-
ing process: globular transfer mode, metal transfer dynamics, drop formation, forces
acting on it, shielding gases, which developed the controlled short-circuiting [9]. In
2012, authors realized that Fuzzy PI control reduced current error than traditional
PI control in burning arc stage. Using STT, no splash transition is observed, and
the appearance of welding is improved [10] and also investigated the occurrence of
breakdown in electron beam gun welding. The power sources reactive parameters
on the time characteristics of voltage and current in different parts of the circuit
are presented [11]. In 2013, Dual-input/Single-output fuzzy control welding power
source is analysed. Using 89S51 microcontroller, change of current is checked that
showed that welding current remains stable, control effect is appreciable. This con-
trol system is low cost, flexible control and high reliability [12]. Recently in 2014,
authors presented a single-phase ac RSW power source which utilized two SCRs
connected in parallel. Current passing through the coil is dependent on the trigger
time and the system properties [13, 14]. In 2014, 1985 some papers stated that in
RSW system the most important input variable is SCR trigger time, which controls
the energy used for welding action. For each controlling phase, the SCR trigger time
is related to the real-time key factor to produce the accurate control of welding. Weld-
ing voltage, energy and current are the selected variables that depend upon various
control strategies [3, 14]. To maintain the weld quality, different controlling methods
in the welding current have been reported by many authors. Different control meth-
ods have been proposed, such as learning control in the frequency domain lookup
table method and fuzzy logical control to deal with the non-linear and time-varying
properties of RSW. The regulation is obtained using PD controller. The experimental
result showed the effectiveness of the controlled output which is compared with that
of the PID controller [13]. During 2014, the analysis of arc welding inverter type
power source and control of droplet transfer is studied at closed-circuit welding. The
control of the current in short circuit phase was used to optimize the welding process
and parameter control in welded joint. This has reduced the electrode metal losses.
Using digital controlling methods in each phase of the droplet formation and droplet
transfer of the electrode in short circuit arc welding produced different possibility of
joining with minimum energy consumption and electrode material losses to achieve
high weld quality [15].
In 2015, authors designed a new hysteresis current controller technique for steady-
state and dynamic performance. A non-linear wide-bandwidth digital current con-
troller for DC–DC and DC–AC converters has been presented. Controller imple-
mented in FPGA kit examined the oversampling of current error by using ADC. This
method is only used for current error signal acquisition. Then it is operated at a con-
stant switching frequency, wide small signal bandwidth and minimum large signal
84 3 Control Terminologies and Schemes for Arc Welding Processes
response delay of nonlinear controller [16]. In 2016, power supply system is devel-
oped with adaptive control method which can adjust to momentous change of system
parameters by using input–output variables for design instead of state variable [17].
Swati Narula also been reported in the literature of a full-bridge buck converter and
a half-bridge buck converter for the AWPS. In conventional isolated zeta AC–DC
converter, many limitations are identified across the switch. The proposed welding
converter is introduced with single stage power conversion, ease of control and high-
frequency isolation. The zeta converter-based AWPS has been designed, modelled
and implemented in hardware using closed-loop PI controller. Constant output volt-
age is maintained with respect to load and supply voltage variations. Due to the
reduced voltage stress in the devices, high reliability is obtained. The improved per-
formance during overload conditions is proportional to weld quality. Thus, the system
attained major advantages ,i.e. one gating signal is used to drive both the switches and
AWPS is simple in control [18]. The following Table 3.2 presents scholarly papers
based on the control strategy in welding as the main application areas.
• Constant voltage control technique and constant current technique is used to achieve uniform weld penetration, fusion area and reduced weld spatter
(continued)
85
Table 3.2 (continued)
86
• Sliding Mode control (Voltage loop) and PI control (Current loop) were implemented to solve voltage overshoots and faster current response is realized
• Proposed sliding mode control has good dynamic response for the control system and easy to apply the algorithms
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• Information of fuzzy controller in power source realized the control of molten pool width
(continued)
87
Table 3.2 (continued)
88
Shcherbakov, A. V.
(2012) [11]
• Electrical circuit of the power source of accelerating voltage is used to investigate the breakdowns
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
(continued)
89
Table 3.2 (continued)
90
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• New digital non-linear wideband current controller using ADC and FPGA is implemented
• Current error processing algorithm causes power converters to attain constant switching frequency
• This exploited the oversampling and signal conversion delays
• Adaptive control method is adjusted to the system parameters using input–output variables
(continued)
91
Table 3.2 (continued)
92
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
J Schupp, W Fischer
and H Mecke [19]
• Various possibilities of arc control for GMAW process using power electronics circuits and external magnetic field is emphasized
• To provide optimal control of welding power source, improves in weld quality and reduce spattering, modern power sources consists of inverter on primary side and rectifier with smoothening
3.8 Process Parametric Influences on Weld Quality
(continued)
93
Table 3.2 (continued)
94
• A mixed current and voltage control scheme for an inverter-controlled arc welding machine is proposed
• This scheme incorporates a closed loop current controller and an open-loop voltage controller to optimize the output voltage and current waveform
• An experimental comparison is made between the conventional and newly proposed control scheme and it revealed that the control schemes reduced the spatter generation by 30–50% by
stabilizing the metal transfer and improved the welding performance
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• Developed control strategies improves the control of highly unstable metal transfer in the GMAW process
• The research reviewed the issues involved in arc stability with CO2 shielding
• The authors proposed a solution by developing a new power source control technique to reduce the spatter levels when the wire feed rate increases
(continued)
95
Table 3.2 (continued)
96
• Authors designed a modular programmable GTAW welding controller with graphical user interface (GUI)
• The control algorithms are developed and implemented using Digital Signal Processor DSP
• The weld performance is evaluated readily by comprehensive monitoring capabilities
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• Developed a 20KW arc welding inverter power source with Sliding mode control (SMC) and PI control, based on average state space model
• PI controller is applied for current loop to improve the stability. SMC is applied for voltage loop to decrease the overshoot of voltage loop without affecting the current loop
• And phase-shift PWM generation is applied additionally for making a simple control and attain superior performance
(continued)
3.8 Process Parametric Influences on Weld Quality
97
Table 3.2 (continued)
98
• The proposed fuzzy and PI control system for MIG welding inverter power source overcomes the demerits of the analogous control schemes with improved reliability, faster response and stable
arc length
• The fuzzy control and PI control are implemented to control arc voltage and welding current, respectively
• The proportional controller is used to obtain good dynamic performance and integral control is used to reduce the static error, increase the precision
• And the fuzzy control system is an adjustor of the closed-loop control system to determine the parameters of PI controller for different welding conditions
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• The proposed sliding mode control (SMC) and PWM control based on average state space model for a 20 KW Arc Welding
• Inverter power source overcomes voltage overshoot and obtain fast response time of PI control
• The proposed Microcontroller(MCF5213CAF80)and FPGA based control system for pulsed MIG welding power source improves the flexibility, control, precision and reliability
• The overall hardware circuit, sampling, ADC circuits and PWM drive circuits and software design of FPGA and real-time embedded operating system μC/OS-II on MCF5213CAF80 are
designed. The experimental results showed that the microcontroller and FPGA based control system provided quicker response, better reliability and more stable arc length
(continued)
99
Table 3.2 (continued)
100
• Digitizing the arc welding power source using microprocessors and computer controls the spatter
• Digitisation of feed wire servo motor control is also implemented using feedback control system
• Recent improvements in digital computation capabilities paved a way to research on complex phenomena
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.2 (continued)
Author(s) Control circuits/inferences
• Authors developed a cheap, flexible arc welding power source using fuzzy control and PWM control schemes
• The traditional welding machine is applied to a simple systems and found to have demerits such as high energy consumption, poor welding performance, etc.
• When it comes to complex systems with too many variables the dynamics of the system is difficult to describe which emphasizes the significance of fuzzy controller which queries the fuzzy
control table and performs reasoning by saving time
(continued)
101
Table 3.2 (continued)
102
• Proposed Model reference adaptive control (MRAC) for Submerged Arc welding (SAW) power source is used to control arc voltage, feed wire speed, arc stability and to achieve, constant current
characteristics
• The MRAC is compared with PID control algorithm.
• The limitation of PID is that it can be used for simple systems and single variable linear systems. Welding process being more complex with multiple input, multiple output and non-linearities;
the MRAC is opted to monitor the welding process and eliminate error by modifying the controller parameters
3 Control Terminologies and Schemes for Arc Welding Processes
3.8 Process Parametric Influences on Weld Quality 103
Excessively high voltage lead to cracking, increases undercut and creates difficulty
in removing slag. Lowering the voltage produces stiffer arc, which improves penetra-
tion in a deep weld groove and resists arc blow. An excessively low voltage produces
a narrow bead and causes difficult slag removal along the bead edges.
Welding speed is the linear rate at which the electrode feed is moved along the
weld joint. With any combination of welding voltage and welding current, the effect
of changing the welding speed confirms to a general pattern. If the welding speed is
increased, power or heat input per unit length of weld is decreased and less filler metal
is applied per unit length of the weld, resulting in less weld reinforcement. Thus, the
weld bead becomes smaller. Weld penetration is affected more by welding speed than
any variable other than current. For excessively slow speeds the molten weld pool
is beneath the welding electrode. Excessive speed may cause undercutting, porosity,
arc blow, uneven bead shape, cracking and higher slag inclusion in the weld metal.
Higher welding speed results in less heat affected zone. Within limits, welding speed
can be adjusted to control weld size and penetration. Relatively slow welding speed
provides time for gases to escape from the molten metal, thus reducing porosity. If
the welding speed is too low, it may also result in a large molten pool that flows
around the arc, resulting in rough bead, slag inclusions and burn through of the weld
plate [26].
In 2002, authors studied the relationship between the process variable and bead
penetration, the effect of welding current and arc voltage on bead penetration on
increasing the voltage. The welding current and depth of penetration of bead also
increases correspondingly [27]. In welding process, deformation is unavoidable.
Inherent deformation of weld bead is analysed using elastic finite element modelling
(FEM). Major influence of the weld length to transverse shrinkage is observed [28].
In 2008, digital computation began to gain attention. The information obtained from
the device needs to be controlled. For the digital data used by the computer, numerical
calculations are performed on the data. Thus, controlling is done for the MIG welding.
The voltage and current control enables the overall stability of the molten pool after
solidification [29]. Dynamic response is the feature of transistor characteristics in
a transistorized technology, which determines the variation in the characteristics of
the electric circuit [30].
In traditional method, the process stability is determined based on the arc stabil-
ity by considering both dynamic and static characteristics assessment [31], which
resulted in better power source properties. In 2010, a thermal model for single seam
per layer multipass narrow gap P-GMA is published. Here, appropriate pulse param-
eter is decided by considering heat and current. Pulse parameters influence at a given
heat input. Then, the higher mean current assured the preparation of sound weld. This
also avoided the lack of groove wall fusion. Single seam per layer multipass narrow
gap P-GMA welding is operated by changing the pulse parameters, with the constant
heat input to produce a sound weld [32]. To reduce the bead widening AC frequency
control technique is done, such that by increasing the input frequency parameter,
widening of the bead during welding is reduced accordingly [25]. Deva Kumaran,
K and his team [33] has investigated the static and dynamic characteristics of an
inverter type synergic GMAW power source. The operational mode study concluded
104 3 Control Terminologies and Schemes for Arc Welding Processes
that there exists a direct correlation between welding wire feed rate, arc voltage
and mean current. The dynamic setting voltage of operating conditions is raised to
higher value than that of the static operating condition. The authors concluded that
using modified short circuit MS-GMAW mode of power source produced less spatter
when compared with that of the GMAW, Pulsed P-GMAW and double pulsation DP-
GMAW. The quality of the weld joint is largely determined by the optimal selection
of arc voltage and welding current. The relationship between welding voltage and
welding current in diverse types of welding processes are shown in Table 3.3.
It is required to study the power source characteristics under static and dynamic
operating conditions to decide about the optimal range of welding parameters [33]. In
2013, the stability of the power source is experimentally investigated. The arc stability
which determines the process stability is was assessed by implementing conventional
P, PID controller. The static (constant voltage) and dynamic (duty cycle) characteris-
tics assessment showed good power source properties. Using LabVIEW (Laboratory
Virtual Instrument Engineering Workbench), stability of the power source is anal-
ysed. The disturbances which affect the arc stability, welding parameters and metal
transfer irregularities are identified. Based on the voltage and current values, weld-
ing process stability and output current are related to each other [34]. Yamane et al.
investigated the effects of welding power source by analysing the current and volt-
age behaviour and identified the cause of spatter generation. Then, characteristics of
the power source are proposed to enhance the stability and self-regulation of the arc.
The developed power source with non-linear characteristics verified the performance
[35]. The analysis of arc welding inverter type power source and control of droplet
transfer during closed circuit welding is analysed. The control of the current in short
circuit phase was used to optimize the welding process and parameter control in
welded joint. This has reduced the electrode metal losses [15]. Overall study reveals
the influence of input parameters such as current, voltage, frequency and weld speed
that directly contributes to the mechanical behaviour and output characteristics of
welding. In the following Table 3.4, some of the comparative literature are illustrated.
Table 3.4 Process parametric effects on welding
Author(s) Welding process/input parameters Outputs and inferences
When
3.8 Process Parametric Influences on Weld Quality
(continued)
105
Table 3.4 (continued)
106
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
Bead appearance is uniform on stainless steel material using digital filtering process
107
(continued)
Table 3.4 (continued)
108
Mild steel subjected to Inconel deposition welding. Large grain sputter on the surface is observed
Dutra, Jair Carlos 2009 [30] MIG/MAG Welding Fillet weld carried out manually with CO2 at low current and voltage observed more splashes
20 A
21 V
Dutra, Jair Carlos 2009 [30] MIG/MAG Welding Fillet weld carried out manually with CO2 at high current and voltage observed minimum splashes
44 A
21 V
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
(continued)
Table 3.4 (continued)
110
AC frequency control (synchronized control method) technique in high-speed fillet lap welding by tandem pulse GMA welding method
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
Appearance of interrupted intimate contact of single seam per layer weld deposit with the groove wall in narrow gap welding
(continued)
111
Table 3.4 (continued)
112
Appearance of interrupted intimate contact of single seam per layer weld deposit with the groove wall in narrow gap welding
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
Lack of fusion at groove wall is typically observed in transverse section of weld joint
(continued)
113
Table 3.4 (continued)
114
Lack of fusion at groove wall typically observed in transverse section of weld joint
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
(continued)
115
Table 3.4 (continued)
116
(continued)
3 Control Terminologies and Schemes for Arc Welding Processes
Table 3.4 (continued)
Author(s) Welding process/input parameters Outputs and inferences
(continued)
Table 3.4 (continued)
118
(1) Resistance of the welding circuit; (2) controlling effect; (3) welding voltage; (4) welding current
Observed that splashing increases with the reduction of the rate of increase of current
3 Control Terminologies and Schemes for Arc Welding Processes
3.9 Real-Time Sample Reports on Formulating Adaptive Control … 119
3.9.1 Objective
In this chapter, an adaptive control scheme is employed for joining Aluminium 6061
alloy sheets by Cold Metal Transfer process. The transfer function model of the
CMT welding system is derived using empirical equations. The CMT plant transfer
function is estimated using system identification technique. For the estimated plant
model, a conventional PID controller is initially designed by tuning the controller
parameters k p , ki , and kd . The designed control system is tested for its ability to
control the welding current when short circuit phase and arcing phase are detected.
Following the conventional PID controller, a Model Reference Adaptive Controller
(MRAC) is implemented to maintain the welding current at desired range, during
melting and electrode wire short-circuiting. The performance analysis for the pro-
posed adaptive control scheme and the conventional PID controller is compared.
3.9.2 Implementation
The experimental trails for the CMT process is performed on 6061 Aluminium alloy
sheets (2 mm thickness) using the Fronius advanced CMT 7000 VR machine as
shown in Fig. 3.8. Argon shielding gas is used with a constant flow rate of 18 lit/min.
The CMT welding is performed in pulsed mode using a linear six-dimensional robotic
displacement machine. The welding trails were designed by varying the wire feed
speed and welding current. The remaining welding parameters are maintained con-
stant. The parameters evaluated in the study are weld bead width, Depth of Penetration
(DOP) and reinforcement height.
PID controller is the most preferred conventional controller. It is a feedback mech-
anism widely used in industrial control systems and a variety of other applications
requiring continuously modulated control. A PID controller continuously calculates
an error value and applies corrective action depending on the proportional, integral
or derivative terms. Its ideal transfer function is represented by
ki
G c (s) k p 1 + + kd s (3.7)
s
After obtaining the plant model from the system identification toolbox, the con-
ventional PID controller is implemented and tested as shown in Fig. 3.9. The plant
model obtained is a third-order system which consists two-zeroes and three-poles as
shown. The optimal controller gains were obtained using the auto tuning feature in
MATLAB. The optimal k p , ki and kd values provide the desired performance and
achieve appropriate time domain specifications. The obtained controller gains are
3.9 Real-Time Sample Reports on Formulating Adaptive Control … 121
Reference
Model
RM
Adaptive
Mechanism
AM e
ym
The PID controller is implemented to control the welding current when short circuit
of the electrode wire feed is detected. Initially, for the setpoint of 50A, the perfor-
mance of the auto-tuned PID controller is evaluated. The short circuit is fed as a
disturbance to the control system. The PID controller response is shown in Fig. 3.11.
The experimental weld images are presented.
From the observed results, the PID controller is unable to adapt to the disturbance
while detecting a short circuit. Initially, the welding current increases to a value of
122 3 Control Terminologies and Schemes for Arc Welding Processes
50A. After being subjected to a disturbance (short circuit phase), the welding current
reduces which thereby decreases the bead width and DOP. The experimental images
are shown in Table 3.5.
The response of the implemented MRAC is shown in Fig. 3.12. It is observed that
the current decreases when a short circuit is detected, and the electrode feed retracts.
The MRAC ensures that the current is maintained at 50 A with uniform bead width
and DOP.
The experimental weld images are shown in Table 3.6.
3.9 Real-Time Sample Reports on Formulating Adaptive Control … 123
References
1. D. Cuiuri, J. Norrish, Verstatile Welding Power Source Controller for Research and Product
Development (2006)
2. I. Singh, Bhim Senior Member, B. N. Singh, I. Chandra, Ambrish Senior Member, I. Al-
haddad, Kamal Senior Member, A. Pandey, I. Kothari, Dwarka P Senior Member, A review of
three-phase improved Power quality AC–DC converters. IEEE Trans. Indus. Electron. 51(3),
641–660 (2004)
3. H.S. Cho, D.W. Chun, A microprocessor-based electrode movement controller for spot weld
quality assurance, IEEE Trans. Indus. Electron. IE-32(3), 234–238 (1985)
4. J. Dziubinski, J. Slania, As assessment of the welding properties which are characteristic of
the power source used in robotic arc welding. Weld. Int. 9(8), 601–606 (1995)
5. G. Dean, J. Norrish, C. Cook, Evaluation of control techniques for dip transfer gas metal arc
welding. Australas. Weld. J. 50, 34–41 (2005)
6. P.K. Palani, N. Murugan, Selection of parameters of pulsed current gas metal arc welding. J.
Mater. Process. Technol. 172(1), 1–10 (2006)
7. Z. Guo-rong, L. Zhao, Z. Ai-yun, Y. Mi, D. Shan-xu, K. Yong, Sliding Mode Control and PI
Control for Arc Welding/cutting Inverter (2008), pp. 4–7
8. X. Dl et al., The Fuzzy Control Algorithm in Copper-Coated Aluminium Wire TIG System
Application (2011), pp. 2214–2218
9. R.H.G. Silva, J.C. Dutra, R. Gohr, Scientific and technological fundamentals for the develop-
ment of the controlled short-circuiting MIG/MAG welding process: a review of the literature.
Part 2 of 3. Metal droplet formation, shield gases, penetration mechanisms, heat input and
economical asp. Weld. Int. 23(2), 141–149 (2009)
10. S. Liu, Y. Wang, Research of CO2 welding inverter power source under current waveform
control. IEEE International Conference on Automation and Logistics, August, pp. 116–121
(2012)
11. A.V. Shcherbakov, Switching processes in power sources for electron beam welding gas with
formation of high-voltage breakdown. Weld. Int. 26(3), 221–226 (2012)
12. P. Tang, X. Jing, A Novel Fuzzy Control of Welder Power Source Using PWM Chip. 2013
Sixth International Symposium on Computing Intelligence and Design, pp. 224–227 (2013)
13. L. Gong, C.L. Liu, X.F. Zha, Model-based real-time dynamic power factor measurement in AC
resistance spot welding with an embedded ANN. IEEE Trans. Ind. Electron. 54(3), 1442–1448
(2007)
14. K. Zhou, L. Cai, A Nonlinear Current Control Method for Resistance Spot Welding.
IEEE/ASME Trans. Mechatron. 19(2), 559–569 (2014)
15. V.A. Lebedev, M.S. Sorokin, A.A. Belov, Algorithms for controlling inverter sources of welding
current to optimize the electrode metal transfer parameters. Weld. Int. 28(12), 957–961 (2014)
16. S. Buso, T. Caldognetto, A nonlinear wide-bandwidth digital current controller for DC–DC
and DC–AC converters. IEEE Trans. Ind. Electron. 62(12), 7687–7695 (2015)
17. Y. Liu, X. Miao, C. Zhang, The Model Reference Adaptive Control Method of Submerged Arc
Welding Power Supply System (2016), pp. 3646–3649
18. S. Narula, B. Singh, G. Bhuvaneswari, Power factor corrected welding power supply using
modified zeta converter. IEEE J. Emerg. Sel. Top. Power Electron. 4(2), 617–625 (2016)
19. J. Schupp, W. Fischer, H. Mecke, Control with power electronics (475), 18–19 (2000)
20. Y.M. Chae, Y. Jang, M.M. Jovanovic, A Novel Mixed Current and Voltage Control Scheme for
Inverter Arc Welding Machines. APEC 2001. Sixth Annual IEEE Applied Power Electronics
Conference and Exposition (Cat. No.01CH37181), vol. 0, no. C, pp. 308–313 (2001)
21. D. Cuiuri, J. Norrish, C. Cook, New approaches to controlling unstable Gas Metal Arc Welding.
Australas. Weld. J. vol. 47, no. 3 (2002)
22. P. Qingle, Design of fuzzy control and expert system based MIG arc welding invert power
source. ICEMI 2009—Proceedings of the 9th International Conference on Electron. Measure-
ments and Instruments (2009), pp. 489–492
126 3 Control Terminologies and Schemes for Arc Welding Processes
23. L. Zhao, X. Bai-lu, W. Shu-hui, Z. Guo-rong, The Sliding Mode Control for Arc Welding
Inverter Power Source. 2008 3rd IEEE (2008), pp. 1100–1104
24. Q. Pang, M. Zhang, Design of digital control system for pulsed MIG welding power source.
Proc. World Congr. Intell. Control Autom. 20090461204, 2492–2495 (2010)
25. T. Ueyama, Welding power sources_TF. Weld. Int. 24(9), 699–705 (2010)
26. I. The, Effect of welding parameters on bead. Flux 1 (1989)
27. I.S. Kim, J.S. Son, I.G. Kim, J.Y. Kim, O.S. Kim, A study on relationship between process vari-
ables and bead penetration for robotic CO2 arc welding. J. Mater. Process. Technol. 136(1–3),
139–145 (2003)
28. L. Wei, Measurement of inherent deformations in typical weld joints using inverse analysis
(part 2) prediction of welding distortion of large structures†. Trans. JWRI is (2005)
29. T. Uezono, T. Hongjun, Application to MIG welding using welding power source equipped
with digital filtering process. Weld. Int. 22(5), 299–303 (2008)
30. J.C. Dutra, MIG/MAG—Short circuit metal transfer—Welding power sources versus arc gases.
Weld. Int. 23(4), 231–236 (2009)
31. M. Suban, J. Tušek, Methods for the determination of arc stability (METAL TRANSFER). J.
Mater. Process. Technol. 143–144(1), 430–437 (2003)
32. B.P. Agrawal, P.K. Ghosh, Thermal modeling of multipass narrow gap pulse current GMA weld-
ing by single seam per layer deposition technique. Mater. Manuf. Process. 25(11), 1251–1268
(2010)
33. K. Devakumaran, N. Rajasekaran, P.K. Ghosh, Process characteristics of inverter type GMAW
power source under static and dynamic operating conditions. Mater. Manuf. Process. 27(12),
1450–1456 (2012)
34. K. Skrzyniecki, P. Kolodziejczak, P. Cegielski, A. Kolasa, Experimental Studies on Stability of
Power Source—Arc (2013), pp. 359–362
35. S. Yamane, S. Xiang, Y. Kaneko, K. Oshima, Effect of power source characteristic on CO 2
short circuiting arc welding_TF. Sci. Technol. Weld. Join. 10(3), 281–286 (2005)
Chapter 4
Power Sources and Challenges
for Different Arc Welding Processes
The choice of power source is based on the working principle of joining technique
and the material type to be welded. Power sources for various arc welding process
are discussed in the following sections.
In MMA welding, both AC and DC power sources are used. Mostly inverter type
power sources are preferred considering its adaptability for minimal maintenance.
The parameter range is presented in Table 4.1.
Welding Power Sources in SMAC process utilizes either AC or DC, but in either case,
the power source must be of the constant current (CC) type. Whether to use an AC,
DC, or AC/DC power source depends on the following factors presented in Table 4.2.
When DC power is used, polarity of electrode is a key factor. If electrode is connected
to negative polarity, the weld exhibits shallow penetration that contradicts to the deep
penetration exhibited when electrode is connected to positive polarity [1, 2].
In Gas Tungsten Arc Welding, also called as Tungsten Inert Gas Arc Welding, the
arc is established between a non-consumable tungsten electrode and the workpiece.
The electrode, the arc and the area around the weld puddle are protected from the
atmospheric reactions by inert gas shield of Helium or Argon. It is performed using
both DC and AC (30–300 Hz) for aluminium with pulsed current of 5–500A with
a frequency range of 2 Hz–20 kHz [3]. The choice between DC or AC depends on
the metal thickness and type of the metal to be welded. If DC power is chosen, then
the polarity is the key governing factor. If the electrode is connected to the negative
terminal of the power source, the electrons flow from the electrode to the workpiece,
and approximately 30% heat of the arc is concentrated at the electrode end and 70%
heat of the arc is concentrated at the workpiece [4]. This demands the use of small
tungsten electrodes to produce a narrow-concentrated arc. The resulting weld shape
indicates narrow and deep penetration. This polarity is suitable for welding most
metals.
Aluminium and magnesium have an oxide coating on their surface which must be
physically removed prior to welding if this polarity is to be used. If the electrode is
connected to the positive terminal of the welding power source, the flow of electron is
from the work to the electrode tip, concentrating approximately 30% at the workpiece
and 70% of the heat of the arc at the electrode. Such a higher heat at the electrode
side necessitates the use of larger diameter tungsten electrode to prevent its melting
and contamination of the workpiece. Since the electrode diameter is larger and the
concentration of heat is less at the work, the resulting weld penetration is wide and
4.3 Power Sources in Gas Tungsten Arc Welding (GTAW) … 129
shallow. For this polarity, the heavy oxide coating of Aluminium and magnesium must
be removed since it acts as an insulator for the electron flow [4] When DC power
sources are used, the arc should be initiated by touching the tungsten electrode on
the workpiece and quickly withdrawing that results in electrode contamination. So,
DC power sources for GTAW are designed with high-frequency stabilizer wherein,
the electrode when brought near 1/8th in. of workpiece, ionization of gas in the
occurs place thereby creating a conductive path for the arc to be established. When
AC power sources are used for GTAW it always incorporates high frequency which
maintains an ionized path for establishing, maintaining and reigniting the electric
arc each time, while the current changes the direction. The electrode size and the
resulting weld penetration is intermediate [1].
GMAW process utilizes a consumable metal electrode. The electrode is a metal wire
which is transferred across the arc and into the molten weld puddle. All GMAW
is done using DCEP. This polarity provides deep penetration, arc stability with low
spatter. AC power is not used for gas metal arc welding [5]. The knowledge about
three types of metal transfer is essential in GMAW which are depicted in Table 4.3.
A DC constant voltage power source is more suited for GMAW because of
lower open circuit voltage (OCV) and self-correcting arc length feature. It can be a
transformer-rectifier or a rotary type. Constant voltage power sources are used for
spray transfer welding and globular transfer welding. If it is used for short-circuiting
arc welding, slope control must be included which aims to limit the high short-circuit
current [6]. The power sources of GMAW have an outlet for providing power to the
wire feeder. The function of the wire feeder is to deliver the welding wire to the arc at
130 4 Power Sources and Challenges for Different Arc Welding Processes
Table 4.4 Welding process, current type and characteristics of power sources
Welding process Type of current Characteristics of power source
MMA DC/AC Constant current
TIG/GTAW DC Constant current
Plasma arc welding DC Constant current
SMAW DC/AC Constant current (if electrode thickness = 2.4 mm)
DC Constant voltage (if electrode thickness = 2.4 mm)
MIG/GMAW DC Constant voltage
constant speed [5]. Since the rate of wire feed determines the current, and ultimately
the amount of heat at arc region, wire feed speed influences the weld bead and pene-
tration. Inverter type power sources and microprocessor controls are popular features
in GMAW.
In Metal Inert Gas (MIG)/Metal Active Gas Welding (MAG) welding, the elec-
trode is of a continuous wire that is fed through a torch which would be attached to
a robotic arm. Welding continues for several minutes depending on electrode size.
MIG/MAG welding prefer DC sources predominantly while few AC sources are also
developed. The welding current is a pulsed type at 10–150 Hz frequency between
low level of 50A and high peak value of 400–600 A [3].
SAW process is mechanized and is like MIG/MAG welding process except that the
arc is submerged beneath the flux and along the joint instead of being protected by
shielding gas. AC or DC power sources are used with welding current Up to 1500 A
[3].
The following Table 4.4 presents the different welding processes, type of welding
current and the respective power source characteristics.
4.6.1 Harmonics
In AC power systems, voltage and current are purely sinusoidal. When their wave-
forms are distorted using Fourier series they can be analysed into components at
fundamental frequency and their multiples as shown in Fig. 4.1 [7]. The frequency
components other than the fundamental ones are termed as harmonics. The main
source of origins of harmonics are given as follows:
4.6 Major Challenges in Power Sources 131
Magnetic arc blow (shown in Fig. 4.2) is one of the challenges that occur when the
magnetic field around the arc is not symmetrical completely. It can often result in a
poor and defective weld.
Arc blow results in the following problems:
(i) The weld pool, and thus the weld bead, is non-uniform as the arc gets deflected
towards one side, producing a defective weld.
(ii) If the arc is deflected along the joint, the width of the bead and the penetration
gets affected.
(iii) The protection provided by molten slag or gas can be affected, resulting in the
formation of pores.
The problem worsens when the welding current increases and becomes more notice-
able. This will further increase all the electromagnetic forces rapidly in and around
the electric arc.
The major causes of arc blow are given as follows:
(i) Asymmetric return current connection
(ii) Asymmetric position of the workpiece while performing welding
(iii) Close Electrode position during multiwire welding
(iv) Permanent magnetic fields from magnetic clamps, bead plates, workpiece, etc.
4.6 Major Challenges in Power Sources 133
The welding arc behaves as a fluid conductor and generates a magnetic field at
a certain radius from the axis [11]. Over the last two decades, developments are
made in inverter technology for providing lighter and smaller power sources with
improvement in performance and efficiency to reduce the magnetic arc blow. In the
inverter power source, a single or three-phase input is rectified, and then switched
at high frequency using a power transformer. The secondary is then rectified to
obtain DC output, The DC output consists of high-frequency ripples which produce
a magnetic field at that switching frequency. Today, the switching frequencies are
increased from few KHz to 100 kHz.
The arc welding equipment’s tested are given:
(i) Conventional arc welding transformer
(ii) Thyristor arc welding power source
(iii) Inverter arc welding power source
(iv) Square wave inverter arc welding power source
(v) Pulsed MIG inverter arc welding power source.
For inspecting the magnetic field, QinetiQ’s 3 m linear plotter, magnetometer, hall
sensor, coil sensor and a data acquisition card are used. The magnetic field is made
up of many harmonic components which must be considered when analysing it
with standard ICNIRP reference levels. The results in the investigation carried by
GB Melton et al., of TWI Limited [2] reveals that for conventional AC transformer
power source, the magnetic field is predominantly 50 Hz, with harmonics at 100 and
150 Hz. For MMA and TIG welding, square wave, inverter power source, the results
indicate that as output frequency increases, the magnetic field decreases. For square
wave outputs, the harmonic components are higher than that of a sine wave output.
134 4 Power Sources and Challenges for Different Arc Welding Processes
The arc welding power sources are made of semiconductor materials and it is impor-
tant to determine semiconductor losses based on which adequate cooling means is
provided to keep the device temperature within prescribed limits. The semiconductor
losses are grouped as follows:
1. Conduction loss (on-state and dynamic saturation)
2. Switching (turn-on and turn-off)
3. Off state.
The conduction and switching losses depend on converter type, operating frequen-
cies, type of load and switching characteristics [7]. Off state losses are very negli-
gible. Snubber circuits reduce switching losses and associated stress due to voltage
fluctuations. The two types of snubber circuits are dissipative and non-dissipative.
1. Dissipative: The energy stored in reactive elements is dissipated in the resistors
and are converted into heat energy. This is not suited for higher operating powers
and frequencies.
2. Non-dissipative: The low losses are caused by nonideal device properties such
as transient switching loss and conduction losses of the switching devices in the
snubber circuits.
The power sources generate heat and they are not 100% efficient. There must be
a way to dissipate the excess heat generated to prevent power source damage and
maximize the reliability. This can be achieved by incorporating a cooling system.
A proper cooling system is necessary to increase the lifetime and reliability of the
power electronic devices in the power source unit [7]. The maximum allowable limit
of temperature for each component is prescribed in the datasheet and those limits
should be strictly maintained. The three mechanisms of heat transfer are shown in
Table 4.5.
4.6 Major Challenges in Power Sources 135
There are few popular methods to minimize losses in the power sources using the
following cooling methods shown in Table 4.6.
136 4 Power Sources and Challenges for Different Arc Welding Processes
(continued)
References 137
References
1. ESAB Welding &Cutting Products, Basic Welding Filler Metal Technology (2000). Available:
http://www.esabna.com/euweb/awtc/lesson1_1.htm
2. T.W. Eagar, The physics of arc welding processes. Adv. Join. Technol. 61–68
3. P. Kah, H. Latifi, R. Suoranta, J. Martikainen, M. Pirinen, Usability of arc types in industrial
welding. Int. J. Mech. Mater. Eng. 9(1), 1–12 (2014)
4. G.E. Cook, The Effect of High-Frequency Pulsing of Welding Arc,” vol. 5, no. 2 (1985), pp. 2–7
5. YAS HAM STD, Engineering, Costs, Quality, and Safety, vol. 5 (1997)
6. D. Cuiuri, J. Norrish, Verstatile Welding Power Source Controller for Research and Product
Development (2006)
7. E. Acha, V.G. Agelidis, O. Anaya-Lara, T.J.E. Miller, Power Electronic Control in Electrical
Systems—CH 2 (Elsevier, India, 2002)
8. C. Wang, Z. Wang, Q. Xu, Study on dynamic characteristics of inverter arc welding power
supply based on double-loop control, in 2009 IEEE 6th International Power Electronic Motion
Control Conference IPEMC ’09, vol. 3 (2009) pp. 1609–1612
138 4 Power Sources and Challenges for Different Arc Welding Processes
Sensors in arc welding process can obtain information about the geometry and posi-
tion of the workpiece, electrical parameters like voltage, current, wire feed rate, in
addition to providing suitable form of control over the welding process parameters.
The weld quality mainly depends on weld parameters like current, voltage, wire
speed and wire feed rate. The welding torch also influences the flow of the material.
The input heat for melting the component and constant heat flow is directly related
with the welding torch control and it further influences the weld quality. This chapter
presents the fundamental knowledge on data acquisition in arc welding process with
real-time case study illustrations.
Data Acquisition (DAQ) (the system shown in Fig. 5.1) is the process of sampling
signals which involves the measurements of any physical or electrical quantities
such as current, voltage, pressure and temperature with a computer. A DAQ system
comprises of devices such as sensors, measurement hardware and a computer with
a programmable software. In comparison to traditional measurement systems, PC-
based DAQ systems consist of fast processing power, productivity and display, cost-
effective and powerful measurement solution [1].
Data acquisition begins with the physical parameter measurement. This physical
parameter could be the temperature of a room, the intensity of a light source, the
pressure inside a chamber, the force applied to an object or many other things. An
effective data acquisition system can measure all these different parameters.
Analog to
Signal Digital Driver Application
Conditioning Converter Software Software
Sensor
Electric
Stimulus
e2 e3 Signal
e1
Transducer 1 Transducer 2 Transducer 3
5.1.2 Signals
Different signals demand different measuring systems. For this reason, it is important
to understand the different types of signals and their corresponding attributes. Signals
can be categorized into two groups:
• Analog
• Digital.
An analog signal can exist at any value with respect to time. Few examples of analog
signals include voltage, temperature, pressure, sound and load. The three primary
characteristics of an analog signal are level, shape and frequency (Fig. 5.3).
Level
Because analog signals can take on any value, the level gives vital information about
the measured analog signal. The intensity of a light source, the temperature in a room
and the pressure inside a chamber are all examples that demonstrate the importance
142 5 Sensors for Welding Data Acquisition
of the level of a signal. When we measure the level of a signal, the signal generally
does not change quickly with respect to time. The accuracy of the measurement,
however, is very important. We should choose a data acquisition system that yields
maximum accuracy to help with analog level measurements.
Shape
Some signals are named after their specific shapes—sine, square, sawtooth and tri-
angle. The shape of an analog signal can be as important as the level because by
measuring the shape of an analog signal, we can further analyse the signal, includ-
ing peak values, DC values and slope. Signals where shape is of interest, generally
change rapidly with respect to time, but system accuracy is still important. The anal-
ysis of heartbeats, video signals, sounds, vibrations and circuit responses are some
applications involving shape measurements.
Frequency
All analog signals can be categorized by their frequencies. Unlike the level or shape
of the signal, frequency cannot be directly measured. The signal is analysed using
software to determine the frequency information. This analysis is usually done using
an algorithm known as the Fourier transform. When frequency is the most important
piece of information, which considers both accuracy and acquisition speed. Although
the acquisition speed for acquiring the frequency of a signal is less than the speed
required for obtaining the shape of a signal, we still must acquire the signal at
the requisite speed to avoid the loos of the pertinent information while acquiring
5.1 Data Acquisition System 143
the analog signal. The condition that stipulates this speed is known as the Nyquist
Sampling Theorem. Speech, telecommunication and earthquake analysis are some
examples of common applications where the frequency of the signal must be known.
A digital signal is discrete in nature. Instead, a digital signal has two possible levels:
high and low. Digital signals generally conform to certain specifications that define
the characteristics of the signal. They are commonly referred to as transistor-to-
transistor logic (TTL). TTL specifications indicate a digital signal to be low when
the level falls within 0–0.8 V, and the signal is high between 2 and 5 V. The useful
information that can be measured from a digital signal includes the state and the rate
Fig. 5.4.
State
The state of a digital signal is essentially the level of the signal—on or off, high
or low. Monitoring the state of a switch—open or closed—is a common application
showing the importance of knowing the state of a digital signal.
Rate
The rate of a digital signal defines how the digital signal changes state with respect to
time. An example of measuring the rate of a digital signal includes determining how
fast a motor shaft spins. Unlike frequency, the rate of a digital signal measures how
144 5 Sensors for Welding Data Acquisition
DAQ hardware performs as the interface between signals from the real world and
computer. It mainly functions as a device that digitizes incoming analog signals,
so that a computer can interpret them. The three key components of a DAQ device
used for measuring a signal are: the signal conditioning circuitry, Analog-to-Digital
Converter (ADC) and computer bus. Many DAQ devices are also utilized for other
performances like automating measurement systems and processes. It primarily func-
tions as a device that digitizes incoming analog signals so that the computer interprets
them. Other data acquisition functionality includes the following:
• Analog input/output
• Digital input/output
• Counter/timers
• Multifunction—a combination of analog, digital, and counter operations on a sin-
gle device.
The most readily available platform is the desktop computer. PCI DAQ boards
that plug into any desktop computer are available now. In addition, DAQ modules for
PXI/Compact PCI, a more rugged modular computer platform specifically for mea-
surement and automation applications are available. For distributed measurements,
compact field point platform delivers modular I/O, embedded operation and Ethernet
communication are available. For portable or handheld measurements, DAQ devices
for USB and PCMCIA work with laptops or Windows Mobile PDAs are used. In
addition, DAQ devices for PCI Express, the next-generation PC I/O bus and for PXI
Express, the high-performance PXI bus are introduced.
• Radioactivity.
Some of the commonly measured quantities are tabulated in Table 5.2,
Arc Welding
Sensors
Sensor Input
Analog to
Buffer Low pass Active Communication
Digital Microcontroller
Amplifier Filter Port
Converter
Signal Conditioning
supply. The robot controls offsets in the programmed trajectory to bring the
weld current back to a specified level.
(iii) Vision-guided line scan systems—a laser camera is mounted at a particular
distance ahead of the weld torch. An accurate position of the weld, down to
0.1 mm, and process variables such as gap and joint angles are measured.
(iv) Vision-guided circular systems—using a circular scan rather than a line scan,
three-dimensional data can be obtained from a single measurement. Usually,
this method is slower than the line scan system mentioned above.
Signals from sensors are generally noisy and direct measurements of signal are not
conductive. Signal conditioning circuitry manipulates a signal into a suitable form
of input into an ADC. This circuitry includes attenuation, amplification, isolation
and filtering. Some DAQ devices include built-in signal conditioning system which
is intended to measure the specific type of sensors. Figure 5.6 represents the general
block diagram of a signal conditioning process.
148 5 Sensors for Welding Data Acquisition
Analog signals from sensors are converted into digital before they are manipulated
using a computer. An ADC is a chip, providing a digital representation of an analog
signal at an instant. ADC takes periodic ‘samples’ of the signal at a predefined rate
and these samples are transferred to a computer bus, where the original signal is
reconstructed from the samples in software.
A computer is connected to DAQ device using a slot. The computer bus serves as
the communication interface between the DAQ device and computer to measure
data’s and for sending information. DAQ devices are connected to computer buses
using USB (Universal Serial Bus), PCI, PCI Express, RS232, RS485, RS422, GPIB
(General Purpose Interface Bus) and Ethernet (as shown in Table 5.5). More recently,
DAQ devices have become available for 802.11 Wi-Fi for wireless communication.
There are some types of buses which offers different advantages for diverse types of
applications.
A programmable software in the computer controls the functions of the DAQ device
and is used for visualizing, processing and storing measured data. Many computers
are used for different types of applications. A desktop may be used in a lab for its
processing power, a laptop may be used in the field for its portability, or an industrial
computer may be used in a manufacturing plant for its ruggedness. LabVIEW is one
such graphical programming software which enables real-time data acquisition and
online monitoring of a process. The types of software’s used in computerized DAQ
system are given as follows:
Application software enables the interaction between the user and computer for
analysing, acquiring and presenting measured data. It is also a prebuilt application
with predefined functionality and programming environment for building applica-
tions with custom functionality. Example: National Instruments LabVIEW is an
application software for real-time data acquiring in welding process.
5.4 Role of Computer in a DAQ System 149
Ethernet
RS232
RS485
(continued)
150 5 Sensors for Welding Data Acquisition
PCI
PCI Express
5.4 Role of Computer in a DAQ System 151
Sensors
Driver software provides application software the ability to interact with a DAQ
device. It communicates with the DAQ device easily by extracting low-level hard-
ware commands. In general, DAQ driver software comprises of an Application Pro-
gramming Interface (API) that is used within a programming environment to build
application software [1]. NI-LabVIEW provides different driver software for various
DAQ modules and chassis.
The information about the weld quality is identified using different sensors. Sensors
in welding applications are mainly used for checking the components position from
start to end, process parameter changes at joint/grooves and joint tracking. They are
used during the time of welding or at pre-welding and post-welding conditions [4]
(Fig. 5.7).
Welding voltage is measured at the terminals of power source as shown in Figs. 5.8
and 5.9. However, this introduces small error due to the voltage drop across the
152 5 Sensors for Welding Data Acquisition
Current
Power source V
Wire feeder
A
Power source V
Fig. 5.9 Measuring the welding voltage between the wire feeder and the workpiece
which is at right angles to it. The effect of generating a voltage using a magnetic field
is called the Hall Effect.
The wire feed speed is a major parameter which has to be controlled in order to
achieve a stable welding process. Mostly, the welding power source is controlled to
produce a constant voltage and the preset parameters (voltage and current). However,
a current will represent a certain wire feed speed and the common method is to apply
a constant voltage and wire feed speed while adjusting the current. By keeping the
wire feed speed at a nominal preset value, a stable welding process can be achieved.
154 5 Sensors for Welding Data Acquisition
In normal robot systems, the wire feed unit is mounted on the robot arm. However,
in some cases, longer conduits must be used. This results in wire feed speed variations
when the conduit is bent and twisted. In practice, a push-pull wire feed system should
be used to counteract this problem. Measuring the wire feed speed is a major issue
and for laboratory purposes, custom-made solutions can be built that measure the
speed at the contact tube. A more realistic approach in a production system is to
measure the controlled speed of the drive wheel of the feeder unit. However, this
must be complemented with securing the functionality and reliability of the feeder
system as its robustness is important for the resulting quality of the welds. Special
sensors with idler rollers that generate a signal proportional to the wire feed speed
are available [4]. For occasional check measurements, it is sufficient to use a tape to
measure the length of wire delivered by the feeder in a certain length of time, from
which the speed can be calculated by the formula
Wire length
Wire feed speed ((1))
Measured time
Digital wire feed rate meters are also available as shown in Fig. 5.13.
The common method to measure the gas flow is to use a marked tapered glass
tube with a float inside. When using this to check performance, it should be held
vertically against the welding gun nozzle. Sensors that provide an electrical output
signal are also available. Digital shielding gas flow meters are also available as shown
in Fig. 5.14.
5.5.4 Temperature
Sensors for geometrical parameters must be able to obtain information about the
weld that relates to the geometry of the weld joint. This information is of immense
importance for the seam tracking to be done and use this information for quality
control of the weld. The challenge, however, is to use sensors during welding. Due to
the harsh environment with elevated temperature, liquid metal spattering, intensive
light and high currents, purpose-built sensors must be applied. The most commonly
used sensors are optical and through-arc sensors.
Optical sensors use the basic principle for detecting the weld joint during arc
welding. A laser beam that is projected in a scanning motion across the seam and
a CCD-array is used to measure the features of weld joint in combination with a
laser stripe. Variations of this method are in use, for example, the laser stripe may
not be a linear line on the weld joint but circular instead. In such case, the sensor is
more flexible to detect weld joints in corners. To measure the distance, the method
of triangulation is used as shown in Fig. 5.18
158 5 Sensors for Welding Data Acquisition
Objective lens
Reference
temprature
Temprature source
lamp
Observer
Eye-piece
Outer tube
Multimeter
Rheostat
Battery
Laser
Detector array
Laser beam
Lens
Scanning axis
A laser beam is focused on an object, and then the reflection from the object as
seen from a lens in the laser sensor is determined by the distance between the sensor
and the object. If the object is close to the sensor, then the angle between the outgoing
5.5 Data Acquisition in Arc Welding Processes 159
beam and the reflection through the focusing lens of the detector is large, while it is
small if the object is farther away.
Depending on the weld joint preparation and geometrical shape, the laser beam can
produce reflections like mirrors. For example, consider a V-groove weld joint where
the laser light will produce several reflecting positions but with different intensities
depending on the surfaces of the weld joint. Therefore, these sensors must have real-
time image processing capabilities to filter out reflections that do not belong to the
point of interest. It should be noted in this context that highly reflective materials
may cause problems during welding and a real test may be needed to verify the
functionality.
The widespread use of triangulation in welding is for seam tracking and this
requires measuring the weld joint geometry. This is achieved through a scanning
technique of the beam across the weld joint (Fig. 5.19). During the scanning, the
sensor acquires a two-dimensional picture of the joint profile as an array of 2D
coordinates. When the robot is moving, a weld joint geometrical model can be made
that contains a full 3D description of the joint.
These optical sensors have a capability for more than seam tracking. Information
that they acquire includes joint volume, gap size, misalignment, track welds, etc. This
information is useful for adaptive feedback control of both the welding power source
and the robot. From a control point of view, seam tracking is usually done with full
compensation of the position error. The nominal path is the assumed trajectory of
the weld joint and during tracking, the robot controller receives new target positions
from the sensor. The robot controller then overrides the nominal path by changing
the position of the Tool Centre Point (TCP) while keeping a constant orientation.
The benefits of this method lies in the fact that for a given nominal path, it is rather
straightforward to verify the ability of the robot to follow the path with some minor
Scanning
motion Viewing
mirror
Object
160 5 Sensors for Welding Data Acquisition
changes while keeping the orientation constant. The drawback is that the user has to
define and program a nominal path.
Instead, if the robot is only instructed where to start and where to end, it must be
able to follow the path measured by the sensor on the fly. This puts some additional
requirements on the robot system as it must be able to calculate the trajectory includ-
ing both target positions and orientations of the weld torch. Typical operating data of
a laser scanner is a scan sweep frequency of 10–50 Hz. If welding speed of 1.2 m/min
is assumed, it indicates one sweep per mm during welding. This is considered sat-
isfactory in most cases. However, new welding processes such as laser welding will
increase the welding speed considerably and for high requirements careful analysis
and trials are essential.
Acoustic signals can also be used to determine the metal transfer mode during
welding process. The main characteristic being used for determination of the short
arc metal transfer mode is the sound of reignition of the arc.
Ultrasonic testing of welds is a proven and efficient method for detection of
defects, irregularities, cracks, inclusions and other welding defects. The major advan-
tages of this method is its adaptability to versatile materials and show possibility of
detecting minute irregularities.
The first basic thing to achieve with a fully automatic robotic welding system is the
capability to follow precisely the joint to be welded. This is because the welding
quality depends very much on the welding pool position apart from its geometry. A
preprogrammed path cannot be obtained with the desired precision since deviations
from the programmed path are likely to occur due to deficient path definition. This
is also due to material plate deficiencies and the effect of heating the plates. Con-
sequently, an online joint recognition and seam tracking system must be available.
Several techniques have been used for joint detection and seam tracking, namely,
for welding robotic systems. Using the arc characteristics and exploiting the propor-
tional relationship between the welding current and the distance from the electrode
to workpiece was the pivotal approach. Seam tracking using a weaving motion and
the arc itself as the sensor, sometimes referred to as through-arc sensing, was intro-
duced in the 1980s. The principle behind the method is to make use of the change
in current when the distance between the contact tube and the workpiece varies. The
underlying principle is relatively easy and cost-effective. Therefore, it is the com-
monly used sensor for tracking methods in robotic welding based on gas metal arc
welding and related processes, like flux-cored arc welding, submerged arc welding,
etc. The approximate relationship between arc voltage (U), arc current (I) and the
contact tube to workpiece distance (l) is expressed by,
β3
U β1 I + β2 + + β4 l ((2))
I
5.5 Data Acquisition in Arc Welding Processes 161
where β 1, β 2, β 3 and β 4 are dependent on factors like wire, gas and the characteristics
of the welding power source. As can be seen from the equation above, when the value
‘l’ varies, the arc current will also change proportionally with opposite sign.
For implementing arc sensor technology, no additional sensors are needed which
makes them a very simple and cost-efficient technology. Furthermore, it is possible
to implement them into closed loop systems, and in comparison, to vision systems,
they are fast [4]. Also, arc sensor technology is a non-contact method and, therefore,
it does not intervene with the welding process.
DAQ hardware acts as an interface between the computer and the outside world.
It primarily functions as a device that digitizes incoming analog signals so that the
computer can interpret them as A DAQ device (Data Acquisition Hardware) which
usually has these following functions:
• Analog input,
• Analog output,
• Digital I/O,
• Counter/timers.
Driver software is the layer of software for easily communicating with the hard-
ware. It forms the middle layer between the application software and the hardware.
164 5 Sensors for Welding Data Acquisition
• NI-DAQmx
• NI-DAQmx Base.
Application software adds analysis and presentation capabilities to the driver soft-
ware. Our software application normally does such tasks as follows:
• Real-time monitoring,
• Data analysis,
• Data logging,
• Control algorithms,
• Human machine interface (HMI).
If you want a ‘continuous’ acquisition, put a While loop around the DAQ Assistant
like this (Fig. 5.28).
The temperature plots for three different parameter settings are shown in Figs. 5.30,
5.31, 5.32, 5.33, 5.34 and 5.35.
5.9 Case Study 2: Characterization of Gas Metal Arc Welding … 169
The data acquisition system employed in this study is an Arc Tracker, which is self-
contained data acquisition (DAQ) system manufactured by Lincoln Electric Com-
pany. The purpose of the DAQ unit is to obtain the welding data using a computer
(Fig. 5.36).
5.9.1 Description
The Arc Tracker system shown in Fig. 5.35, acquires voltage and current data at a
rate of 120 kHz. The rate of the controller in Lincoln Electric Power Waves 455R
and the main power supply model studied is about 10 kHz. This ensures that the data
observed at any change of state in the power supply is observed in the DAQ system.
This study utilized Lincoln Electric’s Power Wave Manager Software which is used
to connect the power supply directly and observes the operation of power supply.
This software can send the current pulse to the system and changes the calibration
set point of the power supply and represents the output in graphs. This results in
providing false offset voltage of the parameter setting and troubleshoots the entire
system.
170 5 Sensors for Welding Data Acquisition
The study is undertaken to monitor the system functioning. The control architectures
and effects of system circuitry were relatively unexplored. The system model of the
welding circuit is shown in Fig. 5.37 and 5.38.
172 5 Sensors for Welding Data Acquisition
CDAQ with
LabVIEW
NI-9219
Software Module
SS316L Base
Material
The main goal of the study is to determine the differences in welding parameters
across various power supplies. The weld system is established and bead on plate welds
are performed on the sample for different power sources. The weld parameters are
maintained at a constant, and testing is done to find the intrinsic difference of the
system. All data is collected using Arc tracker and at 120 kHz of sampling rate in
software. Voltage and current are recorded, then the acquisition is triggered at 10 V,
and data is stored. The stored signals are investigated for irregularities such as high
or low voltage power drops and spikes. Voltage and current are recorded along with
the arc stability and process or plate changes.
5.10 Results
parameters. Testing results include simple voltage and current measurements based
on different power sources from the same variations. Figure 5.39 shows the graph of
voltages recorded from six different power supplies in three different systems.
The error is calculated as V error V set − V weld such that a negative error caused
by the weld voltage is above the set voltage and a positive error is caused by the weld
voltage being below the set voltage.
Figure 5.40 shows the same power supplies and their current error related to the
voltage setting. The power supplies which has higher voltage output than the others
174 5 Sensors for Welding Data Acquisition
Power
Controller
Supply
Weld
+ Vposcable - + Vtorch -
Welding Wire Feeder
Power Supply +
Positive sense lead
Vweld
AC Work -
V
Work sense lead +
A Vfixturing
-
Fixturing Fixturing
- VGndcable +
have a higher current output in this graph, and vice versa for the power supplies with
lower voltages (Fig. 5.41).
References 179
References
The suitable model for Arc welding process can be obtained using a limited number
of experimental measurements of the input and outputs parameters, with prior system
knowledge.
The three basic steps are given as follows:
1. Collecting useful data,
2. Choosing a convenient model set and
3. Computing the best model within the model, followed by a certain identification
criterion, i.e. parameter estimation.
System identification toolbox in MATLAB enables us to estimate the system model
by using the experimental data. This toolbox provides MATLAB functions, Simulink
blocks and an app for constructing mathematical models of dynamic systems from the
measured input–output data. The dynamic systems can be easily modelled using this
toolbox compared to the first principles modelling. The time-domain and frequency-
domain input–output data are used to identify continuous-time and discrete-time
transfer functions, process models and state-space models. The toolbox also provides
algorithms for embedded online parameter estimation.
A frequent problem being faced by the manufacturer is the control of the process
input parameters to obtain a good joint. Traditionally, it has been necessary to deter-
mine the weld input parameters for every new welded product to obtain good joint
with the required specifications. To do so, it requires a time-consuming trial and
error approach, with weld input parameters chosen by the skill of the engineer or
machine operator [3]. Welds are then examined to determine whether they meet the
desired specification or not. Finally, the weld parameters can be chosen to produce a
welded joint that closely meets the joint requirements. Also, there are often a more
ideal welding parameters combination, which can be used only if the parameters are
determined. To overcome this problem, various optimization methods are applied to
define the desired output variables through the development of mathematical models
to specify the relationship between the input and output parameters. Presently, appli-
cations of Design of Experiment (DoE), evolutionary algorithms and soft computing
techniques are useful to develop an empirical relationship between the welding pro-
cess input and output parameters to determine optimal parametric window which
leads to the desired weld quality.
6.3 ANN-Based Optimization Techniques to Arc Welding Processes 183
Artificial Neural networks are advanced soft computing tools which find its promi-
nence in optimization of arc welding parameters. Due to the high-speed learning
capability of ANN, experimental data can be used to train the neural networks using
appropriate algorithms and they predict the characteristic parameters of the weld-
ments in showing good agreement with experimental values. This would help in
creating optimal parametric window.
Artificial neural networks are simplified model of biological network structure. The
building block of ANN is the artificial neuron. These neurons interact with the real
world to receive and transmit data. Some of the neurons remain hidden in their hidden
layers. Neurons are connected to one another by synapse (shown in Fig. 6.1), and each
synapse is associated with a weight factor [4]. Artificial neural networks have their
profound role in real-time, complex and interactive applications like speech recog-
nition, pattern recognition, medicine, finance, weather forecasting and control of
manufacturing processes. The meritorious feature is that they can be modelled using
the experimental data without making any assumption. Several types of ANN algo-
rithms, like Back Propagation (BPN), Multilayer Perception (MLP), Self-Organizing
Map (SOM), Radial Basis Function (RBF), etc. are widely used models [5].
Input Layer
X1
Activation
W1 Summation Function
Yin
W2 Output
X2 ∑ f
Y
Wn
Xn
X1
Y1
X2 Artificial
Neural
Network
Yn
(Hidden Layer
X3 with bias)
Xn
Yin x1 w1 + x2 w2 + · · · + xn wn (6.1)
n
Yin w0 + xi wi (6.2)
i0
Y f (Yin ) (6.3)
The ANN can be created and trained using neural network toolbox in MATLAB
software. This toolbox provides easy access and implementation of ANN. The input
data set is divided into time steps based on the number of data samples. The time of
steps are classified into three categories, such as training, validation and testing;
(1) Training: The data set is fed into the neural network during training stage, and
the network adjusts itself according to the error.
6.3 ANN-Based Optimization Techniques to Arc Welding Processes 185
Unsupervised
Input Output
Learning
A network topology is the arrangement of a network along its nodes and connecting
lines (Based on this arrangement the ANN is classified into feedforward ANN and
backpropagation ANN).
Learning is the method of modifying the weights of connections between the neurons
of a specified network. Learning in ANN can be classified into three categories,
namely, supervised learning, unsupervised learning, and reinforcement learning.
Unsupervised Learning
In this algorithm, there is no specific target or outcome variable to predict/estimate.
It is used for clustering population in diverse groups, which is widely used for seg-
menting customers in diverse groups for specific intervention (shown in Fig. 6.3).
Examples: Apriori algorithm, K-means.
Supervised Learning
This algorithm consists of a target/outcome variable (or dependent variable) which
is to be predicted from a given set of predictors (independent variables). Using these
set of variables, a function is generated that map inputs to desired outputs (a shown
in Fig. 6.4). The training process continues until the model achieves a desired level of
accuracy on the training data. Examples: Regression, Decision Tree, Random Forest,
KNN, Logistic Regression, etc.
Reinforcement Learning
Using this algorithm, the machine is trained to make specific decisions. When the
machine is exposed to an environment it trains itself continually using trial and error
186 6 Optimization in Arc Welding Process
Target
Error
Comparator
Supervised
Input Output
Learning
Reinforcement
Input Output
Learning
and also it machine learns from past experience and tries to capture the best possible
knowledge to make accurate business decisions (as shown in Fig. 6.5). Examples:
Markov Decision Process.
The additional force or effort is applied at the input to obtain an exact output. (Exam-
ple: Linear, sigmoid). An activation function is a decision-making function that deter-
mines the presence of particular neural feature. Non-linearity is needed in activation
functions because its aim in a neural network is to produce a non-linear decision
boundary via non-linear combinations of the weight and inputs.
Inputs Weight
Adjustments
Output Desired
(Wij)
Responses Outputs
(Wjk)
Network
Errors
Back Propagating
Errors Output Layers (k)
between neurons in each layer and the neuron thresholds (as shown in Fig. 6.6). Due
to its robustness, BP-NN is used in many fields [6]. Backpropagation or ‘backward
propagation of errors’, is an algorithm for supervised learning of artificial neural
networks using gradient descent method.
1 2 1
E k(ek ) e2 (6.4)
2 2
where
ek error,
e a vector with element ek.
For a given artificial neural network and an error function, this method calculates
the gradient of the error function with respect to the weights of neural network. The
‘backwards’ part of the name means that the calculation of the gradient proceeds
backwards through the network, with the gradient of the final layer of weights being
calculated first and the gradient of the first layer of weights being calculated last.
Partial computations of the gradient from one layer are reused in the computation of
the gradient for the previous layer.
This backward flow of the error information allows for efficient computation of
the gradient at each layer versus the naive approach of calculating the gradient of
each layer separately. Backpropagation’s popularity has experienced a recent resur-
gence given the widespread adoption of deep neural networks for image recognition
and speech recognition. It is considered as an efficient algorithm, and modern imple-
mentations take advantage of specialized GPUs for performance.
188 6 Optimization in Arc Welding Process
Some of the best-suited algorithms to train ANN are Particle Swarm Optimiza-
tion (PSO) algorithm, Levenberg–Marquardt (LM) and Genetic Algorithm (GA) for
optimizing and predicting the weldment characteristics.
The above-mentioned algorithms and their applications are discussed in the following
context. The main advantage of PSO is its ease of implementation as there are only
few parameters to adjust [7].
where
v[] the particle velocity
present[] current particle (solution)
pbest[] and gbest[] stated already
rand () random number between (0,1)
c1, c2 learning factors (usually c1 c2 2).
The PSO algorithm is used to optimize the BP Neural Network. The algorithm
is combined with NN using initial weights between BP-NN layers and the initial
thresholds between the nodes and to optimize the distribution; execute searches
within solution space and to find optimal initial weights and thresholds for the BP-
NN at a faster convergence rate. Subsequently, the weights and thresholds obtained
by the BP-NN are used to train and test the sample data set. Figure 6.8 shows a
flowchart of this optimization model.
The steps involved in the algorithm are given as follows:
1. Initialize the parameters of the PSO; population size, position, velocity of parti-
cles and number of iterations.
2. Decide the structure of BP-NN and generate the population of particles.
Particles: X i (x i1 , x i2 … x iD )T , i 1, 2, … n
D RD1 + D1 D2 + D1 + D1 (6.7)
where
R number of nodes in input layer
190 6 Optimization in Arc Welding Process
Initialize the
speed and Determine the ANN
position of network topology
the particle
Calculate
Initialize the weights and
fitness value
threshold of BP-NN
of the particle
Search the
Obtain optimal values of
pbest and
weight and threshold
gbest
Update the
position and Calculate error
velocity
No
Update pbest
and gbest Terminal
conditions
satisfied?
Yes
Terminal
conditions
satisfied? Output
3. Define the fitness function. The BP-NN is trained to set weights and thresholds
using the samples as per the Particle population, X i . The fitness function of the
X i , is denoted as
M−1
2
fiti yj − yj where, i 1, 2, . . . n (6.8)
j1
6.4 Development of PSO-Based Backpropagation Neural Network 191
4. Calculate the fitness value for each particle and evaluate the Population X. Cal-
culate the fitness value of each particle X i .
5. Update the optimal fitness value of the individual (Pbest) and the position of the
particles. For each of the particle X i , if the current fitness value is less than its
optimal fitness value, then use the current fitness value to update the optimal
fitness value and the position of particle X i .
6. Update the population fitness value (Gbest). For each of the particle X i , if the
fitness value is less than the optimal fitness value for current population, then use
the fitness value of the current particle to update the optimal fitness value.
7. Update the speed and position of particles iteratively to generate new populations
as per the below equations.
k
Vidk+1 ωVidk + c1r1 Pidk − X id
k
+ c2 r2 Pgd − X id
k
(6.9)
k+1
X id k
X id + Vidk+1 , d ∈ [1, D], i ∈ [1, n] (6.10)
where
ω inertia weight
k number of iterations
Vid velocity of particle Xi
C 1 and C 2 acceleration factors (not smaller than zero)
r 1 and r 2 are arbitrary numbers between 0 and 1.
8. The iteration is stopped based on the optimal value of initial weights and thresh-
olds of the BP-NN. The overall fitness variance of the particle population is used
to judge whether the algorithm converges or not. If the overall fitness variance
is less than given threshold, then the algorithm converges. The optimal solution
is the outputs; initial weights and thresholds. Else, the iterations continue. The
overall fitness variance is denoted by
n
f i − f avg 2
σ −
2
(6.11)
i1
f
where
n number of particles in the current population,
f fitness value,
fi fitness value of particle X i ,
f avg average fitness of the current population particle.
192 6 Optimization in Arc Welding Process
where
J Jacobian matrix for the system
λ Levenberg’s damping factor
δ weight update vector
E error vector.
The LM algorithm is designed to minimize the sum of squared error function of
the form
1 2 1
E k(ek ) e2 (6.13)
2 2
where
ek is the error,
e is a vector with element ek .
Based on weight update vector, the network weights are changed to achieve appropri-
ate solution. The cross product of the Jacobian matrix is known as the approximated
Hessian. The Levenberg damping factor is adjusted in each iteration.
The Jacobian matrix consists of all first-order partial derivatives of a vector function.
In the neural network, it is a N × W matrix, where N is the number of elements in
training set and W is the total number of parameters (weights + bias). It is created by
taking the partial derivatives of each output with respect to each weight denoted by
the form
6.5 Development of Levenberg–Marquardt (LM) Algorithm … 193
⎡ ⎤
∂ F(x1 ,w) ∂ F(x1 ,w)
∂w1
··· ∂wW
⎢ ⎥
⎢ ⎥
J ⎢
⎢
..
.
..
.
..
.
⎥
⎥ (6.14)
⎣ ⎦
∂ F(x N ,w) ∂ F(x N ,w)
∂w1
··· ∂wW
where F(x1 , w) is the network function evaluated at i-th input vector of training set
using the weight vector w and wW is the w-th element of the weight vector w.
g JTE (6.15)
H JTJ (6.16)
(H + λI )δ g (6.17)
find δ.
6. Updating of the network weights w using δ.
7. Recalculation of the sum of squared errors using updated weights.
8. If the sum of squared errors not decreased, discard the new weights, increase λ
using some adjustment factor and go to step 4.
9. Else decrease λ and stop.
The flow chart of the implemented ANN model is depicted in Fig. 6.9.
194 6 Optimization in Arc Welding Process
YES
Calculate the Network
Output
Evaluate result to the
requirement?
Chromosome
0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0
0 Gene
Population
(Set of Chromosomes)
0 Allele
evolved, until a stopping criterion is reached. Some of the key terminologies are
discussed below.
• Population—It is a subset of all the workable solutions for a given problem. The
population for a GA is analogous to the human population but instead of human
beings, GA has Candidate Solutions representing human beings.
• Chromosomes—A chromosome is one solution to the given problem (shown in
Fig. 6.10).
• Gene—A gene is one element position in a chromosome.
• Allele—It is a value a gene takes in a particular chromosome.
• Genotype—Genotype is the population in a computation space. In the computation
space, the solutions are represented such that it is manipulated by a computing
system.
• Phenotype—Phenotype is the population in the actual real-world solution space
in which solutions are represented in real-world situations.
• Decoding and Encoding—For simple problems, the phenotype and genotype
spaces are the same. In some cases, the phenotype and genotype spaces are differ-
ent. Decoding is the process of transforming a solution from the genotype to the
phenotype space and encoding is a process of transforming from the phenotype
to genotype space. Decoding is done fast as it is carried out repeatedly in a GA
during the fitness value calculation (shown in Fig. 6.11).
• Fitness Function—A fitness function is one which takes the solution as input
and produces the suitability of the solution as the output. Sometimes, the fitness
function and the objective function may be same.
• Genetic Operators—They alter the genetic composition of the offspring. These
include crossover, mutation, selection, etc.
GA is used to optimize the arc welding parameters like bead width, depth of
penetration, reinforcement height, front height to front width ratio, back height to
back width ratio, etc. The basic flow structure of GA is given in Fig. 6.12.
196 6 Optimization in Arc Welding Process
Encoding
Genotype Space
(Computation Space)
Phenotype Space
(Actual Solution Space)
0 1 0 1 0 1 0 1
Decoding
6.7.1 Objective
B. Senthil Kumar and T. Kannan and R. Madesh investigated the effect of FCAW pro-
cess parameters on the quality of the super duplex stainless steel (SDSS) claddings
using Taguchi L9 design of experiments. In this investigation, deposits are made with
30% bead overlap. Establishing the optimum combination of process parameters is
essential to ensure better bead geometry and desired properties. The above objectives
are achieved by identifying the significant input process parameters as input to the
mathematical models like welding voltage (X 1 ), wire feed rate (X 2 ), welding speed
(X 3 ) and nozzle to plate distance (X 4 ). The identified responses governing the bead
geometry are bead width (W ) and height of the reinforcement (H). The mathemati-
cal models are constructed using the data collected from the experiments based on
Taguchi L9 orthogonal array. Then, the responses are optimized using non-traditional
nature-inspired technique like genetic algorithm (GA) [11].
6.7.2 Experimentation
Start
Generate Number of
Individuals
Crossover
Assign fitness to off-springs
Finished?
YES
Mutation
NO Assign fitness to off-spring
Finished?
YES
Apply replacement operation to
incorporate new induvidual
Terminate?
YES
Finish
6.7.3 Optimization
The electrode melting and heat input rate regulates the bead dimensions. The deposi-
tion rate has greater impact on the bead dimensions and dilution but does not affect the
welding current. The main objective of weld cladding is to produce the desired bead
dimension with low percentage dilution without affecting the bead integrity. To meet
the above requirements, the objective function is composed of both reinforcement
height and bead width.
6.7 Case Study 1: Optimization of Flux Cored Arc Welding … 199
The table presents the parameter combinations for non-dominated Pareto optimal
points (as shown in Table 6.2). The details of the optimization environment are given
as follows:
GA settings,
Population 60,
Selection function tournament,
Crossover fraction 0.8,
Crossover ratio 1.0,
Distance measure function distance crowding,
Number of variables 4,
Upper limit [30, 300, 200, 23],
Lower limit [22, 200, 120, 15].
200 6 Optimization in Arc Welding Process
6.8.1 Objective
In this study, Friction stir welding is applied to Aluminium alloy 6061 sheets of 3 mm
thickness. The influence of the process parameters; spindle speed, plunge speed and
plunge depth on the Shear Strength and Brinell’s Hardness of the joints are inves-
tigated by performing a single lap joint. Finally, a parallel design method based on
Particle Swarm Optimized—Backpropagation Neural Network is developed to pre-
dict the shear strength and hardness of the aluminium alloy joints. The PSO algorithm
improves the accuracy and runtime efficiency of the BP Neural Network and thereby
optimizes the initial weights and thresholds of the same. Based on the experimental
outcomes, validations are carried out. The results reveal good coherence between
experimental data and predicted data. Considering the vitality, this soft computing
algorithm can be incorporated to replicate the experimental results thereby saving
resources, energy consumption, manpower and material wastage during investiga-
tions in manufacturing processes.
6.8.2 Experimentation
Brinell Hardness
Welding Speed
Hidden Layer
(mm/sec)
with bias
Shear Strength
(Mpa)
Plunge Depth
(mm)
equally spaced values, usually coded as −1.682, −1, 0, +1, +1.682. The range of
the input parameters spindle speed, welding speed and plunge depth are selected as
563–1236 rpm, 0.46–2.14 mm/s, 0.02–0.18 mm, respectively as shown in Table 6.3.
6.8.3 Implementation
The development of BP-NN and its training are performed using neural network
toolbox in MATLAB. The hidden network structure contains five layers with a bias
node. The neural network deployed in this study is a feedforward backpropagation
network with ‘PSO’ as learning algorithm. The input layer has three neurons (Spin-
dle speed, Welding speed and Plunge depth) and the output layer has two neurons
(Sher strength and Brinell Hardness) as shown in Fig. 6.16. Initially, various BP-NN
structures with one to five hidden layers (each with varying neuron numbers) in each
layer are evaluated. Out of all the evaluated NN, a NN with five hidden layers with
five hidden neurons in each layer resulted in superior performance.
202 6 Optimization in Arc Welding Process
The results for the Brinell hardness and Shear strength obtained by developed BP-
NN model is in good agreement with the experimental results as shown in Figs. 6.17
and 6.18.
Three test samples are taken for validating the obtained results. The test results
reveal that the experimental and predicted values are in close coherence with each
other as shown in Table 6.4. The percentage difference of error while measuring
the spread of prediction by the BP-NN model (in predicting the hardness and shear
strength) is estimated. It is observed that the developed BP-NN model predicts param-
eters under study within the acceptable error limits.
6.9 Case Study 3: LM Algorithm-Based ANN Model to Predict Strength … 203
6.9.1 Objective
Fig. 6.19 Ultrasonic welding machine used for joining Al–Cu wires
predict and validate the strength and joint resistance of the welded samples. It is
observed that there is a good agreement between the simulation and experimental
results.
6.9.2 Experimentation
The ultrasonic welding is performed for Al6061 and Cu alloys by varying parameters
based on L9 Taguchi Method. Based on the recent demands in industrial sector,
dissimilar materials aluminium and copper are chosen in this study. The power rating
of the ultrasonic machine is 10 KW and the process model used is UWMM 1000 as
shown in Fig. 6.19. This machine provides the option to vary the process parameters;
pressure, amplitude and weld time.
The parameters that have insignificant influence on the weldment as observed
from trials are maintained constant. For both Al and Cu, the thickness of the weld
samples is 0.5 mm and welding is performed for 30 trails to analyse the weldments
and to frame an optimum parameter window. The parameters levels in this study are
provided in Table 6.5.
6.9.3 Implementation
The ANN model is developed with four set of input parameters: Pressure (bar), weld
time (sec), amplitude (microns) and holding time (s). The hidden layers are chosen in
the implemented LM algorithm. The output layer consists of two parameters: Strength
(N/m2 ) and Joint resistance () (as shown in Fig. 6.20). Thirty data samples are used
for training the ANN model and three samples are used for prediction and validation.
Based on the experimental data, the ANN is trained using LM algorithm in the
MATLAB tool platform to predict the strength and joint resistance of ultrasonically
6.9 Case Study 3: LM Algorithm-Based ANN Model to Predict Strength … 205
Strength
(N/m2)
Time
(sec) Artificial
Neural
Network
Joint Resistance
(Hidden Layer (Ω)
Amplitude
(microns) with bias)
Holding time
(sec)
joined Al-Cu alloy wires. Figure 6.21 shows that the strength and joint resistance
predicted by the developed LM based ANN model is in good coherence with the
experimental validations.
Three test samples are taken for validating the obtained results. The test results
reveal that the experimental and predicted values are in close agreement with each
other as shown in Table 6.6.
The percentage difference of error while measuring the spread of prediction by
the ANN model (in predicting the strength and joint resistance) is estimated. It is
observed that the developed LM-ANN predicts within the acceptable error limits.
The error plots are shown in Fig. 6.22.
The optimization methods are appropriate for modelling, control and optimiz-
ing the different welding process. Most works indicate the adaptation of ANNs to
predict responses and optimize the welding process. Nevertheless, there is a lack of
comparative study regarding the performance of the optimization methods based on
206 6 Optimization in Arc Welding Process
Fig. 6.22 Error plot in prediction of strength (left) and Joint resistance (right)
its suitability. Future work should focus on the application of these modelling and
optimization techniques to identify the optimal welding combinations for a certain
welding process for which the process is considered safe, environment friendly and
economical.
References 207
References
Arc welding becomes a safe procedure when adequate safety precautions are con-
sidered and followed to protect the workers from harmful hazards. When such safety
measures are neglected, the workers come across potential hazards such as exposure
to gases and fumes, arc radiation, electric shock and fire accidents, which could be
serious and fatal.
A person in charge of the work (welding) has the main responsibility to ensure
that welders are not exposed to safety and health hazards. A person directing a
business on welding activities must eradicate risks due to welding or minimize the
risks that are reasonably practicable [1].
The WHS (Workplace Health and Safety) Regulations possess specific neces-
sities for managing the hazards due to airborne contaminants, chemicals and other
associated risks. This necessitates the analysis about the substance and providing
adequate information.
Officers, such as directors of an industry, have the responsibility to ensure that
the undertaking business fulfils the regulations provided by WHS.
Workers hold the responsibility of taking care of their own safety and must
not affect the safety of other workers. Workers must cooperate with procedures and
policies relating to safety and health aspects. Workers must use the personal protective
equipment (PPE) as per the training instructions given to them.
The preliminary step to manage the risks associated with arc welding is to find all
potential risks and hazards. Some welding results in the production of infrared (IR)
and ultraviolet (UV) radiation which cause blindness, burns and cancer.
The risk assessment process involves considering the consequences when a worker
is exposed to a hazard. As per the regulations of WHS, a risk assessment is not
compulsory for welding, but it is essential while working in a confined space.
In certain circumstances, a risk assessment will assist to:
Different welding processes influence risk in a different manner. For example, the risk
of electric shock is less in Gas Metal-Arc Welding (GMAW) process than Manual
Metal-Arc (MMA) welding process since the open-circuit voltage is lower and only
direct current is used.
The aim of eliminating a potential hazard and its associated risk is significant. If it
is not possible, the risks can be minimized by the following methods:
Substitution—Replacement of hazardous material with a less hazardous material,
for example, preferring Submerged Arc Welding (SAW) rather than Flux-Cored Wire
Welding (FCAW) reduces the risk of exposure to fumes and radiations.
Isolation—Removal of the workers from hazardous location or isolating the haz-
ards from the workers. For example, in the processes such as plasma cutting, grinding
7.1 Risk Management Process 211
can be performed in away from the usual manufacturing location to reduce the expo-
sure to noise.
Use of engineering controls—Risks can be eliminated using engineering con-
trols, for example, ventilation to remove welding fumes.
If risk then persists, it must be minimized by implementing administrative con-
trols. For example, if a welding is performed in a hot environment, allowing the welder
to weld for a limited time by taking break causes reduction in the heat exhaustion.
Any remaining risks can have minimized using personal protective equipment
(PPE). For example, if the worker stands on metal surfaces which belongs to a part
of the electric circuit, rubber-soled or leather-based boots can be used to prevent the
risk of electric shock.
A person performing the welding business must accomplish enough control measures
to manage the associated risks in a workplace [1]. This comprises ensuring of correct
labels for the hazardous chemicals that the workers access.
Welding produces fumes, dust, smoke, gases and vapours (as shown in Fig. 7.2). The
type and quantity emitted varies depending on the material and the joining process.
The health effects of exposure to these fumes can include the irritation in the nasal
cavity, respiratory tract, throat, asthma and tightness in chest area (Fig. 7.1).
A few basic terms to be known before approaching the control measures are given
as folows:
Natural Ventilation: It is the air movement through the workplace, which is caused
due to natural forces. Outside the workplace, it is caused by the wind. Inside, it occurs
through open doors and windows.
7.2.2 Radiation
Radiation is the energy that travels in the form of waves such as electromagnetic
radiation or subatomic particles (as shown in Fig. 7.3). Arc welding emits ultraviolet
radiation, infrared radiations and visible radiation. The effect of radiation on the
body depends on type, intensity and the distance of the person being exposed. It may
7.2 Specific Hazards and Control Measures 213
result in eye disorders and skin burns when exposed to intense UV or IR radiation
during welding. Exposure of radiation to eyes results in inflammation and pain in the
cornea called as ‘welders flash’ or ‘arc eye’. The cornea can repair itself in one or
two days, and sometimes may lead to loss of vision if the exposure is severe. Workers
involved in the welding environment are at greatest risk; however, other workers in
the workplace could be exposed to radiation.
Some of the control measures are given as follows:
Using welding equipment may result in the risk of electric shock. Exposure to elec-
tromagnetic fields is a potential hazard for workers with certain medical conditions
[1] (Fig. 7.4).
Electric shock results in burns or sometimes death by electrocution. Electrocution
occurs by direct contact with the electrode, live part, workpiece or contact with any
unearthed cable or equipment. Electric shock risk worsens with moisture and high
humidity.
Figure 7.5 illustrates melting of wires due to short circuit and excessive current.
214 7 Codes and Safety Standards During Welding
• Usage of electrode holders with insulations. The holder should not be wet for
cooling.
• Prevention of contacting welding wire or and electrodes with bare hands.
• Ensure that holders are not held under the armpits.
• Prevention of holders or electrodes to meet any other person.
• Checking the working area for the presence of any live components or wetness.
• Inspection of equipment like power switches, connections, cables and insulation
whether they work in good condition.
7.2 Specific Hazards and Control Measures 215
Arc welding results in the production of strong electromagnetic fields close to the
current-carrying cables and power source. They disturb the functioning of permanent
defibrillators, pacemakers or other medical devices which can slow down the heart
or even stop it. A worker will not be aware of a magnetic field hazard unless a heart
pacemaker or other device behaves irregularly. Some of the control measures are
given as follows:
Burns are most common injuries during welding. The IR and UV radiation result in
harmful effects to the workers. Burns occur on hands and skin surface eyes due to
sparks and weld fragments (Fig. 7.6).
Some of the control measures are given below:
• Isolation of workers from contacting the workpieces. For example, carrying post-
weld heat treatment where workpieces cannot be touched.
• Mark or label as ‘hot’ equipment.
• Contact with heated surfaces can be evaded with thermal insulators.
• Ventilation of work areas.
These gases are used as shielding gases in arc welding, which are stored in cylinders
as large volumes under high pressure. Precautions are to be taken when storing and
handling the cylinders.
The potential hazards associated with these gases are fire, toxicity, explosion
asphyxiation, oxidization and uncontrolled pressure release. Gas leakage is also
dangerous which can be recognized by its odour, however, leakage of oxygen is
harder to sense and therefore it is more hazardous. Some of the control measures are
given below:
In most cases, PPE (as shown in Fig. 7.7) must be worn by workers when welding
to supplement higher levels of controls such as ventilation systems or administrative
controls.
Some of the critical Personal Protective Equipment (PPE) are presented in
Table 7.1.
• Turn off the engines while troubleshooting and maintenance work are carried out
and except when the maintenance work requires it to be running.
• Operating engines in open and in well-ventilated areas helps the engine to exhaust
fumes outdoors.
• Avoid adding the fuel when the engine is running or near an open flame welding
arc.
• Before refuelling, stop the engine and allow it to cool in order to prevent spilled
fuel from vaporizing on contact with hot engine parts and igniting.
• Do not spill fuel while filling the tank. If fuel is spilled, wipe it up and do not start
the engine until fumes have been eliminated.
• Keep all equipment covers, safety guards and devices in position and in good
repair. Keep hands, clothing, hair and tools away from gears, fans, V-belts and all
other moving parts when starting, operating or repairing equipment.
7.3 Standard Operating Procedures During Arc Welding 221
• In few cases, it can be necessary to remove safety guards to do the required main-
tenance. Remove guards only when necessary and replace them when the main-
tenance requiring their removal is complete. Always use the greatest care when
working near moving parts.
• Do not put hands near the engine fan.
• Do not attempt to override the governor or idler by pushing on the throttle control
rods while the engine is running.
• To prevent accidentally starting gasoline engines while turning the engine or weld-
ing generator during maintenance work, disconnect the spark plug wires, distrib-
utor cap or magneto wire as appropriate.
• To avoid scalding, do not remove the radiator pressure cap when the engine is hot.
• Route the electrode and work cables together—secure them with tape when pos-
sible.
• Never coil the electrode lead around the body.
• Do not place the body between the electrode and work cables. If the electrode
cable is on our right side, the work cable should also be on our right side.
• Connect the work cable to the workpiece as close as possible to the area being
welded.
• Do not work next to welding power source.
• The electrode and work (or ground) circuits are electrically ‘hot’ when the welder
is on. Do not touch these ‘hot’ parts with our bare skin or wet clothing. Wear dry,
hole-free gloves to insulate hands.
• Insulate our self from work and ground using dry insulation. Make certain the
insulation is large enough to cover our full area of physical contact with work and
ground.
• In semiautomatic or automatic wire welding, the electrode, electrode reel, welding
head, nozzle or semiautomatic welding gun are also electrically ‘hot’.
• Always be sure the work cable makes a good electrical connection with the metal
being welded. The connection should be as close as possible to the area being
welded.
• Ground the work or metal to be welded to a good electrical (earth) ground.
• Maintain the electrode holder, work clamp, welding cable and welding machine
in good, safe operating condition. Replace damaged insulation.
• Never dip the electrode in water for cooling.
• Never simultaneously touch electrically ‘hot’ parts of electrode holders connected
to two welders because voltage between the two can be the total of the open-circuit
voltage of both welders.
• When working above floor level, use a safety belt to protect our self from a fall
should we get a shock.
222 7 Codes and Safety Standards During Welding
• Use only compressed gas cylinders containing the correct shielding gas for the
process used and properly operating regulators designed for the gas and pressure
used. All hoses, fittings, etc. should be suitable for the application and maintained
in good condition.
• Always keep cylinders in an upright position securely chained to an undercarriage
or fixed support.
• Cylinders should be located; away from areas where they may be struck or sub-
jected to physical damage; a safe distance from arc welding or cutting operations
and any other source of heat, sparks or flame.
• Never allow the electrode, electrode holder or any other electrically ‘hot’ parts to
touch a cylinder.
• Keep our head and face away from the cylinder valve outlet when opening the
cylinder valve.
• Valve protection caps should always be in place and hand-tight except when the
cylinder is in use or connected for use.
Welding, fabrication, testing, repair and inspection requirements are covered under
three main governing organizations: American Welding Society (AWS), the Amer-
ican Petroleum Institute (API) and the American Society of Mechanical Engineers
(ASME). All these organizations have specific codes for several types of processes,
construction and/or materials [5].
7.4 Welding Codes: American Welding Society (AWS) 223
The American Welding Society was founded in with a global mission to advance
the science and technology in the application of welding, cutting, allied joining. It
is headquartered in Florida. The organization is popular for its code and standard
procedures, for the welding and joining of materials. Some of the critical codes
pertaining to arc welding are presented as follows.
AWS D1.1: It contains the requirements for fabrication and erection of welded steel
structures. It applies to steel materials with a thickness of 3.2 mm or more. Most of
the provisions are mandatory.
AWS D1.2: It consists of the Structural Welding Code for Aluminium. The require-
ments are applicable to all types of welded aluminium alloy. This code is applied for
fabrication of supporting structures. It is not intended to the codes developed for use
in applications specifically to ASME Boiler, aerospace, pressure vessels and military
codes.
AWS D1.3: It consists of the Structural Welding Code for Sheet-based Steel. It covers
the arc welding of steel sheet including the cold-formed members which are less than
or equal to 4.8 mm thickness.
AWS D1.4: It consists of the Structural Welding Code for Reinforcing Steel. It applies
to the welding of reinforcing, carbon and low-carbon structural steel.
AWS D1.5: It consists of the Bridge Welding Code. It covers welding fabrication
requirements applicable to welded highway bridges. This code is not intended to
be used for steels with a minimum specified yield strength greater than 690 MPa,
pressure vessels, base metals other than carbon or low-alloy steels.
AWS D1.6: It consists of the Structural Welding Code for Stainless Steel. It covers
welding requirements applicable to stainless steel weldments. It shall be used in
conjunction with any complementary code for the design of stainless steel weldments.
AWS D3.5-93R: It consists of the guide for Steel Hull Welding. It is referenced in
many contract specifications for building vessels from barges to tugboats.
AWS D3.6M: It consists of the specification for Underwater Welding.
AWS D3.7: It consists of the guide for Aluminium Hull Welding. Like the Steel Hull
Welding Guide, but with a special emphasis on the unique properties of aluminium.
AWS D8.8-97: It consists of the specification for Automotive and Light Truck Weld
Quality: Arc Welding.
AWS D14.1: It consists of specification for Welding Earth Moving and Construction
Equipment.
AWS D14.5: It consists of specification for Welding Presses and Press Components.
It establishes minimum acceptable requirements for weld joint design and the fabri-
cation by welding of presses and press components.
224 7 Codes and Safety Standards During Welding
It sets out the conditions for welding procedure tests on standardized test pieces for
the arc welding methods for welding steel-based alloys as shown in Table 7.4.
The standard specifies the shape and minimum dimensions of standardized test pieces
to be used in connection with the welding procedure. The test pieces must be suf-
ficiently large to ensure that there is sufficient material to conduct away the heat.
When impact testing of the heat-affected zone is required, the test pieces must be
marked with the rolling direction. All welding of test pieces must be carried out
in accordance with the preliminary WPS, and under the same conditions as can be
expected in production.
Welding positions, angles of slope and rotation must be as specified in EN ISO
6947. Tack welding must be included in the test welds if it is to be used in production.
Welding and testing must be supervised by an examiner or examining body.
• Visual inspection,
• Radiographic or ultrasonic testing,
• Crack detection,
• Transverse tensile test,
• Transverse bend test,
• Impact testing,
• Hardness test and
• Macro- and micro-examination.
7.5.2.3 Retesting
If the welding procedure test pieces do not meet all the test requirements, the results
cannot be qualified. It is permissible to perform a further procedure test. If any single
test piece fails to meet the requirements due to geometrical defects, two new test
pieces may be selected for retesting. If either of them fails, then the entire WPS also
fails.
A WPS that has been qualified by a manufacturer is valid for welding in workshops
and at sites under the same technical management. Welding procedure tests form
the basis for qualification of a WPS, of which the important variables lie within the
qualification range of the procedure test. Essential variables are as follows:
• Parent material,
• Material thicknesses,
• Welding method,
• Welding position,
226 7 Codes and Safety Standards During Welding
• Type of joint,
• Consumables,
• Type of welding current,
• Heat input,
• Preheat temperature,
• Intermediate pass temperature and
• Post-heat treatment.
Records from welding and testing shall include all the information needed for qual-
ification. Welding Procedure Qualification Records (WPQR) must be signed by the
examiner. Model forms of WPQR are included in the standard.
Liquid penetrant testing (as shown in Fig. 7.11) is also a surface testing method
and is used to reveal discontinuities in the surface of non-porous materials.
Eddy current testing, or inductive testing (as shown in Fig. 7.12) is used to reveal
discontinuities on or immediately below the surface of electrically conducting mate-
rials. A common feature of all these test methods is that the test personnel must hold
certificates in accordance with EN ISO 9712. Testing must also be carried out in
accordance with qualified test procedures.
References 229
References