Welding for

Design
Engineers

Welding for
Design
Engineers

Copyright 8 2006 by The CWB Group

All rights reserved.
Although due care has been taken in the preparation of this book neither the Canadian Welding Bureau, the Gooderham
Centre nor any contributing author can accept any liability arising from the use or misuse of any information contained herein
or for any errors that may be contained in the module. Information is presented for educational purposes and should not be
used for design, material selection, procedure selection or similar purposes without independent verification. Where
reference to other documents, such as codes and standards, is made readers are encouraged to consult the original sources
in detail.

Canadian Welding Bureau

Gooderham Centre for Industrial Learning
7250 West Credit Avenue
Mississauga, ON L5N 5N1
Tel: 905-542-2176
Fax: 905-542-1837
www.gcil.org

ISBN 0-9739175-0-4

Welding for Design Engineers

Table of Contents

Chapter 1 - Introduction .........................................................................................................................1

1.0
1.1
1.2
1.3
1.4
1.5

Introduction ...................................................................................................................................3
Historical Background ...................................................................................................................4
Grouping of Welding Processes ...................................................................................................6
The Welding Arc............................................................................................................................7
Health and Safety .......................................................................................................................16
Welding Terms and Definitions ...................................................................................................17

Chapter 2 - Welding Codes and Standards ........................................................................................31

2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18
2.19
2.20
2.21
2.22

Introduction .................................................................................................................................33
Purpose of Standards .................................................................................................................35
Development of Standards..........................................................................................................36
Administration of Standards ........................................................................................................38
CSA Standard W47.1 - Certification of Companies for Fusion Welding of Steel........................39
CSA Standard W47.2 - Certification of Companies for Fusion Welding of Aluminum ................43
CSA Standard W48.01 - Filler Metals and Allied Materials for Metal Arc Welding .....................45
CSA Standard W59 - Welded Steel Construction (Metal Arc Welding) ......................................45
CSA Standard W59.2 - Welded Aluminum Construction ............................................................49
CSA Standard S6 - Design of Highway Bridges .........................................................................50
CSA Standard S16-01 - Limit States Design of Steel Structures ................................................50
CSA Standard W186 - Welding of Reinforcing Bars in Reinforced
Concrete Construction ................................................................................................................51
CSA Standard W178.1 - Qualification Code for Welding Inspection Organizations ...................53
CSA Standard W178.2 - Qualification Code for Welding Inspectors ..........................................55
National Building Code of Canada (NBC)...................................................................................57
CSA Standard Z662 - Oil and Gas Pipeline Systems.................................................................57
American Society of Mechanical Engineers (ASME)..................................................................58
American Welding Society (AWS)...............................................................................................60
AWS Codes of D-Series .............................................................................................................61
AWS A5 Specifications................................................................................................................61
ANSI/AWS D1.1 - Structural Welding Code - Steel ...................................................................62
ISO Standards (International Standards Organization)...............................................................63

Chapter 3 - Weld Joints and Welding Symbols..................................................................................65

3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.13
3.14
3.15
3.16
3.17
3.18
3.19

Introduction .................................................................................................................................67
Definition of Joint ........................................................................................................................68
Definition of Weld ........................................................................................................................71
Groove Welds .............................................................................................................................73
Prequalified Joints.......................................................................................................................76
Positions of Welding ...................................................................................................................79
Joint Edge Preparation ...............................................................................................................83
Fundamental Concepts of Welding Symbols ..............................................................................87
Basic Weld Symbols ...................................................................................................................88
Supplementary Weld Symbols ....................................................................................................95
Break in Arrow.............................................................................................................................98
Combined Weld Symbols..........................................................................................................100
Information in Tail of Welding Symbol.......................................................................................102
Extent of Welding Denoted by Symbols ...................................................................................103
Multiple Reference Lines ..........................................................................................................103
Complete Penetration ...............................................................................................................105
Groove Welds ...........................................................................................................................107
Fillet Welds ................................................................................................................................117
Plug Welds ................................................................................................................................129

Chapter 4 - Metal Arc Welding Processes ........................................................................................133

4.1
4.2
4.3
4.4
4.5

Introduction ...............................................................................................................................135
Shielded Metal Arc Welding (SMAW) .......................................................................................136
Gas Metal Arc Welding (GMAW) ..............................................................................................148
Flux Cored Arc Welding (FCAW) ..............................................................................................170
Submerged Arc Welding (SAW)................................................................................................182

Chapter 5 - Welding Metallurgy .........................................................................................................195

5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
5.12
5.13
5.14
5.15
5.16

Introduction ...............................................................................................................................197
Basic Concepts of Iron and Steel..............................................................................................198
Iron, Cast Iron and Steel ...........................................................................................................199
Phase Transformation During Heating and Cooling .................................................................200
Effect of Heating and Cooling on Steel.....................................................................................203
Alloy Elements in Steels............................................................................................................213
How Does Hardness Affect Welding .........................................................................................215
Heat Affected Zone (HAZ).........................................................................................................216
Weldability of Metals .................................................................................................................218
Solidification Cracking...............................................................................................................226
Strength and Toughness in the Weld Zone...............................................................................227
Hydrogen Cracking ...................................................................................................................229
Heat Treatment of Steels ..........................................................................................................234
Influence of Welding on Mechanical Properties........................................................................240
Designation of Steels ................................................................................................................240
Classification of Steels (Numbering System)............................................................................241

Chapter 6 - Residual Stress and Distortion......................................................................................249

6.1
6.2
6.3
6.4
6.5
6.6
6.7

Introduction ...............................................................................................................................251
Expansion and Contraction of Metals .......................................................................................252
Coefficient of Thermal Expansion and Thermal Stress.............................................................254
Residual Stresses .....................................................................................................................256
Distortion ...................................................................................................................................269
Welding Procedure and Distortion ............................................................................................278
Control and Correction of Distortions........................................................................................289

Chapter 7 - Fracture and Fatigue of Welded Structures .................................................................299

7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12

Introduction ...............................................................................................................................301
Stress-Strain Relationship .........................................................................................................302
Fracture of Steel Components ..................................................................................................303
Fracture Surface .......................................................................................................................304
Cleavage ...................................................................................................................................305
Grain Size Effect .......................................................................................................................306
Transition Temperature and Brittle Fracture .............................................................................306
Effect of Strain Rate ..................................................................................................................319
Fracture Mechanics...................................................................................................................321
Stress State of Crack Tips.........................................................................................................322
Stress Intensity Factor ..............................................................................................................324
Fatigue and Fatigue Cracks......................................................................................................326

Chapter 8 - Welding Design ...............................................................................................................351

8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
8.10

Introduction ...............................................................................................................................353
Scope and Objectives ...............................................................................................................354
Design Principles ......................................................................................................................357
Shear Resistance......................................................................................................................365
Fillet Weld Strength ...................................................................................................................370
Fillet Weld Groups.....................................................................................................................375
Restrained Members and Moment Connections ......................................................................382
Welding of Hollow Structural Sections (HSS) ...........................................................................397
Design Procedures....................................................................................................................405
Sizing Welds .............................................................................................................................406

Chapter 9 - Welds Faults and Inspection..........................................................................................413

9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.9
9.10
9.11
9.12

Introduction ...............................................................................................................................415
Weld Fault Characteristics ........................................................................................................416
Distortion or Warpage ...............................................................................................................420
Dimensional Faults....................................................................................................................422
Structural Faults in the Weld Zone............................................................................................434
Fusion Faults.............................................................................................................................441
Cracking ....................................................................................................................................445
Surface Defects.........................................................................................................................450
Defective Properties..................................................................................................................452
Summary of Weld Faults...........................................................................................................452
Welding Inspection....................................................................................................................453
Methods of Testing ....................................................................................................................455

Chapter 10 - Weld Cost Estimating ...................................................................................................481

10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
10.9
10.10
10.11
10.12
10.13

Introduction ...............................................................................................................................483
Consistent Application of Welding Methods..............................................................................483
Cross-Sectional Area of Weld (At) ............................................................................................484
Excess Weld (X) .......................................................................................................................484
Unit Weight of Weld (M)............................................................................................................486
Weight of Weld Metal ................................................................................................................486
Weld Metal Deposition Rate (D) ...............................................................................................487
Shielding Gas (G) .....................................................................................................................488
Flux for SAW Process (F) .........................................................................................................488
Process Deposition Factor (Dp)................................................................................................488
Welder/Operator Work Efficiency Factor (Dw)..........................................................................489
Weld Cost Estimating Procedure ..............................................................................................491
Computer Estimating.................................................................................................................499

Chapter 1
Introduction

Table of Contents
1.0

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

1.1

Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

1.2

Grouping of Welding Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

1.3

The Welding Arc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

1.3.1 Arc Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
1.3.2 Voltage Distribution Along the Arc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
1.3.3 Magnetic Field Associated with a Welding Arc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
1.3.4 Effect of Polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14
1.3.5 Effect of Electrode Extension (Stickout) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
1.3.6 Hydrogen in Weld Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

1.4

Health and Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

1.5

Welding Terms and Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

1.0

Introduction

This book has been mainly developed for civil engineers. For more than a century, civil engineering
students have been taught the design of reinforced concrete structures and riveted steel structures.
Welded construction was relatively novel in the 1920s and 1930s, but really took off during the Second
World War. To provide the basic design knowledge of welding during those years of rapid
development, the Canadian Welding Bureau, in the late 1940s, undertook the task of disseminating the
knowledge of welding construction. The Bureau compiled and administered a series of
correspondence home study courses, known all over the world, which form the foundation of the
CWB/Gooderham Centre for Industrial Learning home study modules today.
This volume encompasses the educational materials developed during the past five decades and
specifically directs it toward civil engineering applications. Efforts have been made to condense vast
amounts of technological information into this volume. Additional reading materials have been
referenced at the end of each chapter for the reader to pursue further study.
The following news item appeared in the Engineering New Record in 1985 and 1987. It is a reminder
to our fellow engineers of what could happen with a seemingly correct decision, but one made without
thorough understanding of the implications of structural application.
In 1985 in Uster near Zurich, Switzerland, the collapse of a suspended concrete
ceiling over an indoor swimming pool resulted in 12 people killed and 2 injured.
Two years later the investigators reported that the acidic vapour (containing chlorine
ions) coated the stainless steel hangers supporting the ceiling and led to pitting,
stress corrosion and cracking. This problem has been written up in books which
are well known among welding and corrosion engineers, but the typical structural
engineers would not have these references. The design engineer should have called
an expert when dealing with materials outside their experience, so said the expert.
The lesson of this story is the importance of having a knowledge of welding engineering. The design
engineers should be familiar with it as they are with concrete. Hopefully, after studying this volume, the
reader should be able to solve welding problems, or otherwise know when its necessary to consult a
welding expert.
The main purpose of this book is to be used both as a primer for civil engineers who are searching for
welding knowledge, and as an important sourcebook for welding information.

1.1

Historical Background

Until very recently, the only method available to join metals was by forge welding, which requires two
pieces of metal to be heated and then pressed or hammered together to develop a metallurgical bond
between the two. Modern welding technology can trace its origins to the first half of the nineteenth
century, when advances in electrical technology such as the production of an arc between two carbon
electrodes and the invention of the electric generator took place. By the end of the nineteenth century,
these advances had led to the development of three welding processes:
g
g
g

arc welding
resistance welding
oxy-acetylene welding

The arc welding process in its numerous variations is now the most important and widely used welding
process.
The first major patent for arc welding was awarded in the United Kingdom to Russians, Benardos and
Olszewski in 1885, who employed a carbon electrode as the positive pole to obtain an arc with the
workpiece (negative pole). The arc heated the workpiece (comprised of two adjoining pieces of lead or
iron) so that they locally melted and fused with each other. Soon thereafter, in 1889, Slavinoff, from
Russia, and Coffin, from the United States, were able to substitute a metal electrode for the carbon
electrode.
A significant advancement in welding came with the use of consumable metal electrodes. Carbon
electrodes previously in use could not provide filler metal. Further advances and applications of the
metal arc welding process depended on the development of improved metal electrodes for greater arc
stability, and a means of shielding the molten pool from contamination from the air surrounding the arc,
which embrittled the weld metal.
The earliest effort in this regard was the application of coating or covering to the metal electrode.
Kjellberg of Sweden applied the coating by dipping iron wires in a thick mixture of carbonates and
silicates, and then letting them dry. The British were the first to attempt application of the arc welding
technology on significant scales as a substitute for riveting in the fabrication of ships. In the United
States, around the time of the start of World War I, German ships interned in New York harbour and
scuttled by their crews were rapidly brought back into service by effecting repairs using arc welding.
The first all-welded ship, the Fulagar, was launched by the British in 1920.
During the 1920s, arc welding was applied for fabrication of heavy wall-pressure vessels and buildings.
In Canada, a 500 foot long, three span bridge having an all-welded construction was erected in Toronto
in 1923. However, widespread use of arc welding had to wait until 1927 when an extrusion process to
economically apply covering to the electrode was developed. For welding stainless steel, electrode
coverings that reduced the amount of hydrogen in the weld metal or that contained more easily ionized
ingredients for arc stabilization were developed soon after.

In 1930, Robinoff was awarded a patent for submerged arc welding (welding under powder or flux,
continuous wire without any covering) of longitudinal seams in pipes. Being a highly productive,
mechanized process, it is still a very popular welding process today.
The use of externally applied gases, instead of slag and gases formed from the electrode covering, to
shield the weld pool in arc welding had also been investigated. During the 1920s, Hobart and Devers
in the United States experimented with argon and helium as shielding gases and this was a precursor
to the development of the gas tungsten arc welding process used for welding of magnesium, aluminum
and stainless steel, during World War II. Their work also demonstrated the use of continuous wire
being fed through a nozzle for arc welding with external inert shielding gases, and this was later
developed into the gas metal arc welding process in 1948 at Battelle Memorial Institute. Availability of
smaller diameter wires and constant voltage power sources made this process more popular for joining
non-ferrous metals and alloys. Application of gas metal arc welding to steels had to await the
introduction of carbon dioxide as a shielding gas in 1953. Since then, there have been numerous
developments of gas mixtures containing argon, helium, oxygen and carbon dioxide for gas metal arc
welding of steels.
Another significant innovation in the 1950s was the development of tubular wires that contained fluxing
agents on the inside. The gases generated by the decomposition of the fluxing agents as well as an
externally applied gas are used for shielding the pool from atmospheric contamination. Initially known
as the Dualshield process, it is known as flux cored arc welding today. Other variations of tubular
wires that have a significant usage today include self shielded flux cored arc welding wires, i.e., without
using the external gas shield; and metal cored arc welding wires with metal powder and some arc
stabilizing materials inside and used with external shielding gas.
Welding processes other than those using an arc heat for welding have also been developed over the
years since the last part of the nineteenth century. Briefly, these can be summarized as follows:
g

resistance welding and its variations (spot welding, seam welding, projection
welding, flash butt welding) over the period 1885 to 1900

thermit welding for joining rails in 1903

electroslag welding during the 1950s

electrogas welding in 1961

plasma arc welding in 1957

electron beam welding in the late 1950s

laser welding in the early 1970s

1.2

Grouping of Welding Processes

There are many ways in which welding processes may be grouped and different countries have
adopted various classification schemes based on the application of heat, whether external pressure is
applied, the type of energy involved (mechanical, electrothermic, thermochemical), etc. The American
Welding Society (AWS) groups welding processes based primarily on the mode of energy transfer, and
secondarily on the influence of capillary action in effecting distribution of filler metal in the joint (as in
brazing and soldering).
AWS defines welding as a joining process that produces coalescence (i.e. growing together) of
materials by heating them to the welding temperature, with or without the application of pressure or by
the application of pressure alone, and with or without the use of filler material. In the AWS approach,
welding processes are grouped into the following major categories:
g
g
g
g

arc welding (AW)

solid state welding (SSW)
resistance welding (RW)
oxyfuel gas welding (OFW)

g
g
g
g

soldering (S)
brazing (B)
other welding
allied processes (such as cutting,
thermal spraying)

Arc welding processes are, by far, the most commonly used in the welding industry and are, therefore,
the main focus in this book. However, arc welding involves melting and most metals, when melted in
air, become contaminated with oxides and nitrides through contact with the oxygen and nitrogen in the
air. This contamination may result in a poor quality weld. Most arc welding processes have some
means of shielding (protecting) the molten metal from the air or some other means of removing the
harmful effects of oxygen and nitrogen. The two main methods of arc shielding are:
g
g

flux shielding
gas shielding

Most of the arc welding processes are distinguished principally by the method of shielding or the way in
which it is applied.
The exact selection of an arc welding process for a particular application involves several
considerations including:
g

Is the process suitable for welding the metal or alloy involved? In the required
thickness and position?

Would the welded joint have the required quality and physical (mechanical,
corrosion) properties?

Is it the most economical of the available choices?

Are the equipment and skilled welders available for the chosen process?

1.3

The Welding Arc

Most metals and alloys conduct electricity at room temperature due to the presence of free electrons.
A considerable amount of heat can be produced from the flow of the current in a circuit. Typical
examples of this heating effect, also called resistance heating, are tungsten filament bulbs and heating
coils in ovens. In comparison, gases like oxygen, nitrogen, carbon dioxide, etc. do not conduct any
electricity at room temperature. However, if sufficient energy is applied to a gas it also can become
conductive. When sufficient voltage is applied to a gas it can be ionized changed into positively
charged ions and negatively charged electrons. The electrons move in response to the applied voltage
to produce a current flow and this movement of electrons allows the initiation of an arc. The current
flow causes resistance heating in the gas which promotes further ionization and increased current flow.
As long as the voltage source is able to supply the necessary voltage and the current needed by the
arc, it can be sustained in a stable manner and used for welding applications.
Based on the above principle, a conventional arc is formed between two non-consumable electrodes in
a gas or vapour medium when an appropriate voltage, depending on the electrode material and gas
phase, is applied to the electrodes. As seen in Figure 1.1, one of the two electrodes forms a positive
terminal of the electrical circuit and is called the anode; the negative terminal of the circuit is called the
cathode. When an arc is created, electrons are evaporated from the cathode and transferred to the
anode through the ionized gas in between. Flow of electrons is the same thing as flow of current or
electricity.

Figure 1.1: An arc between two electrodes.

A welding arc is formed when a

fairly high current (10 to 2000 A)
is forced to flow across a gap
between two electrodes at
relatively low voltage (10 to 50
V). A welding arc is intensely hot
with temperatures exceeding
3000C (see Figure 1.2) and
forms a concentrated heat source
suitable for melting most metals
rapidly. The intense heat of the
welding arc causes the filler
metal to melt and when added to
the locally hot melted workpiece,
it forms the weld fusion zone. Its
subsequent freezing
(solidification) produces the bond
(weld) between the workpieces.
Arc welding processes do not
require application of pressure to
cause fusion.

Figure 1.2: Temperature distribution in a 200 A arc in argon

(from AWS Welding Handbook, Vol. 1).

In welding, the arc may be established between an electrode and the workpiece, or between two
electrodes.
When the workpiece is one of the electrodes of the electrical circuit, the other electrode may be
consumable or nonconsumable. A consumable electrode is designed to melt and add filler material to
the welding joint.
The electrical current for welding is provided by a power source that draws high-voltage electric
power from the main transformer and converts it into higher current and lower voltage suitable for
welding (Figure 1.3). Power sources are broadly classified as constant current or constant voltage
type, and the static volt/ampere output characteristics for these two types of power sources are shown
in Figure 1.4.

Figure 1.3: Transforming electrical power.

Figure 1.4: Characteristic volt/ampere curve for welding power sources.

1.3.1 Arc Efficiency

The welding arc provides the intense heat needed to locally melt the workpiece and the filler metal. In
fact, all the electrical energy supplied by the power source is converted into heat (current x voltage).
Some energy is lost in the electrical leads, and therefore the energy available for welding is the product
of the current (I) and voltage drop between the electrode where the current enters it and the weld pool
(V). For example, with 400 A current and 25 V drop from the contact tip to the weld pool, the arc
energy is 10,000 Joules/second. This arc energy is partly used up in heating the electrode, melting the
consumable electrode or the separately added filler metal in a nonconsumable electrode process, and
heating and locally melting the workpiece. The rest of the heat is lost by conduction, convection,
radiation, spatter, etc. The proportion of the energy that is available to melt the electrode/filler metal and
the workpiece is termed the arc efficiency.
The arc efficiency for some of the commonly used arc welding processes varies between 20% and 90%.
For a given process, factors like welding in a deep groove, arc length, etc. also influence the arc
efficiency. Higher arc efficiency usually means that for a given arc energy, a greater amount of weld
metal is deposited and the workpiece cools at a comparatively slower rate.

1.3.2 Voltage Distribution Along the Arc

In any welding set up, there is a continuous drop in voltage from the lower-most point of contact
between the contact tip and the wire, to the molten weld pool or the workpiece. Figure 1.5
schematically shows that this voltage drop occurs in four steps.

Figure 1.5: Voltage drop in the region of the welding arc.

10

First, there is a drop in voltage over the electrode extension, that is the length of electrode between the
point of electrical contact with the contact tip, and its melting tip, also called cathode spot for the
current flow direction shown in the sketch. The magnitude of this voltage drop depends on the
electrode extension and the wire diameter as well as the current; a longer electrode extension, a
smaller wire diameter or higher current all increase the voltage drop over the electrode extension
length.
The voltage drop over the arc length, that is the distance between the cathode spot and the anode spot
(the molten weld pool surface in Figure 1.5) takes place in three steps. Right next to the anode and
cathode spots are small, thin, gaseous regions called the anode drop zone and cathode drop zone,
respectively, and over these zones there can be a significant drop in voltage, in the range of 1 to 12 V
depending on the electrode material.
In between the two drop zones, there is the arc column with a relatively small drop in voltage, of the
order of 1 to 2 V per centimetre length of the arc column. There is a jet-like flow of ionized gases in
the arc column that gives it some stiffness and force (resistance to deflection). This enables the welder
to manipulate the gun and direct the molten metal to be deposited at the desired location in the weld
joint. Shorter arcs have greater stiffness than longer arcs.
Arc length is a critical and controllable parameter, which is directly related to the arc voltage. Arc
voltage depends on the space between electrodes; electrode composition, diameter and extension;
shielding gas composition; metal thickness; joint design; welding position, etc. The voltage measured
at the power supply is greater than the arc voltage. Output voltage represents the sum of arc voltage
and the voltage drop in the remaining part of the electrical circuit. The longer the electrical cables the
greater will be the difference between the voltage read at the power supply gauge and the actual arc
voltage.

11

1.3.3 Magnetic Field Associated with a Welding Arc

When an electric current passes
through a conductor, a magnetic field is
created that surrounds the conductor
(Figure 1.6). Unless this magnetic
field is balanced in all directions, the
welding arc will tend to be deflected
from its normal axial orientation in line
with the electrode. This phenomenon
is called arc blow. It is more likely to
be present during welding of magnetic
materials (steels) and can cause
incomplete fusion types of flaws in
welds.
Some degree of imbalance in the
magnetic field is always present. The
path of the magnetic flux in the
workpiece is continuous behind the arc
and discontinuous ahead, due to the
change in the direction of the current
as it goes from workpiece to electrode
(Figure 1.7). Since a shorter arc is
stiffer, it is also less susceptible to arc
blow.

Figure 1.6: Magnetic field surrounding a

current carrying conductor.

Figure 1.7: Imbalance in the magnetic field due to change in the direction
of current and part unwelded joint.

12

The magnetic field introduced by the current flowing in the electrode also plays a role in metal transfer.
When the tip of the electrode melts, there are several forces that act at the molten tip. These include
surface tension, gravity, plasma jet and electromagnetic pinch force. Surface tensions tends to prevent
the detachment of the liquid drop at the electrode tip, irrespective of the welding position. Gravity
supports droplet detachment when welding in the flat (downhand) position and attempts to prevent it in
the overhead position. The plasma jet in most situations tries to detach and propel the molten drop
across the arc column to the workpiece.
The electromagnetic pinch force helps in the process of detaching the molten metal drops from the
electrode tip. Generally, when there is some necking between the molten tip and the unmelted
electrode, the magnetic field introduces a pinch force acting in both directions away from the neck
(Figure 1.8). This helps to separate the drop from the electrode. Since this pinch force increases as
the square of the current, smaller and smaller drops are detached as the current increases.

Magnetic
Field

Anode (+)

Pinch Force Aiding

Drop Detachment
Pinch
Effect

Cathode (-)

Figure 1.8: Detachment of molten metal drop due to pinch force.

13

1.3.4 Effect of Polarity

The electric current used in a welding arc may be either direct current (DC) or alternating current (AC).
Direct current flows constantly in one direction. Alternating current is continually changing direction.
When direct current is used for welding, the welding electrode (consumable or nonconsumable) can be
the positive pole or negative pole in the electrical circuit. The workpiece will have the opposite polarity.
These two arrangements for current flow are called DC electrode positive (DCEP) and DC electrode
negative (DCEN), respectively (Figure 1.9). The type of current selected and its polarity can have a
significant influence on the shape and penetration of the weld bead .

Figure 1.9: DCEP and DCEN arrangements for electrical leads.

For example, in gas tungsten arc (GTA) welding, a nonconsumable electrode welding process, direct
current electrode negative (DCEN) is the polarity used most often. Electrons are easily emitted from
the tungsten electrode (cathode). When the electrons travel through the arc they accelerate to very
high speed. About 70% of the arc heat is released at the workpiece (anode or positive pole) due to
electrons striking the surface at high speed. This produces a weld bead with greater penetration.
When the polarity is reversed (DCEP) the workpiece becomes the cathode. The weld pool cannot
easily emit electrons because the molten pool is at a much lower temperature than the tungsten and
will resist the release of electrons. While DCEP is helpful in cleaning the weld pool by removing the
oxides, about 70% of the arc heat is now generated at the electrode (anode). This reduces the life of
the tungsten electrode and the weld bead has reduced penetration. The use of alternating current
provides arc characteristics that are average of those for DCEN and DCEP (Figure 1.10).

14

Current Type
Electrode Polarity

Electron and
Ion Flow

Ions

DCEN

DCEP

Negative

Positive

+ - Electrons
+
+
-

Ions

+ - Electrons
+
+
-

AC (balanced)

Ions

+ +
+
-

Electrons

Penetration
Characteristics
Heat Balance in
the Arc (approx.)
Penetration

Work End: 70%

Electrode End: 30%

Work End: 30%

Electrode End: 70%

Work End: 50%

Electrode End: 50%

Deep, Narrow

Shallow, Wide

Medium

Figure 1.10: Effect of current type and polarity in GTA welding (from AWS Handbook, Vol. 2).

The heat balance in consumable electrode processes differs from that in tungsten arcs. Thus, a
greater amount of heat is generated at the cathode rather than the anode.
When using the gas metal arc process, direct current electrode positive is the polarity of choice as it
leads to greater heat generation at the workpiece (cathode) and therefore greater penetration.
Conversely, DCEN polarity produces more heat at the electrode (cathode), and therefore increases the
electrode melt-off rate and reduces penetration.

1.3.5 Effect of Electrode Extension (Stickout)

When an electric current flows through a conductor, a certain amount of heat is generated due to the
current having to overcome the electrical resistance of the conductor. This is called resistance heating
and it is proportional to I2 x R where I is the current and R is conductor resistance. The resistance, R,
increases with the length of the conductor and decreases as the diameter increases.

15

In continuous wire consumable electrode welding, the electrode extension (Figure 1.5) represents an
electrical conductor through which a fairly high welding current passes. When the electrode extension
is increased, its resistance increases and therefore the magnitude of resistance heating also increases.
As a result, for the same welding current, the consumable wire melts at a faster rate and thus
increases the deposition rate for the same arc energy. However, this heating effect means that less
heat is available to heat and melt the workpiece. Consequently, penetration is reduced and the risk of
incomplete fusion type of flaws is increased. Also, due to an increase in voltage drop over a longer
electrode extension, a higher voltage setting is usually needed to maintain a constant arc length as
with the shorter electrode extension.
The effect of electrode extension for individual arc welding processes is addressed later.

1.3.6 Hydrogen in Weld Metals

Invariably, there is some amount of hydrogen present in the solidified and cooled weld zone. This
hydrogen is introduced by the arc heat breaking down the moisture present in and around the welding
arc. Possible sources of this moisture include the electrode covering, flux, shielding gas, atmospheric
humidity and condensation on the work pieces.
When welding steels, absorbed hydrogen can cause cracking (this will be discussed further in Chapter
9).Therefore, in welding carbon and low alloy steels, martensitic stainless steels, etc., an important
consideration in selecting the welding process and filler metals is the amount of hydrogen that might be
introduced into the weld zone. Non-ferrous materials react differently to hydrogen.
1.4

Health and Safety

Like most manufacturing and fabrication processes, the welding operation presents various hazards to
the health and safety of the welder and personnel working near a welding operation. These hazards
include:
g
g
g

Fire hazards
Electrical shock
Arc radiation

g
g
g

Smoke and fumes

Compressed gases
Other hazards related to specific processes, locations, etc.

These hazards are well recognized and when proper precautions are taken, welding is a safe
operation. It is therefore extremely important that before performing any welding operation the
operator be fully aware of these precautions as well as be knowledgeable about the equipment to be
used and its operation. The reader is referred to Module 1 for detailed guidance on the health and
safety aspects of welding, and is strongly urged to have the knowledge therein before performing any
welding.
As engineering personnel, you are always in an oversight position on the construction site. Safety is
everybodys concern, especially for engineers.

16

1.5

Welding Terms and Definitions

Some of the terms frequently used in this book are briefly described below to assist the reader in better
understanding the contents of this book.
Acceptance Criteria

A defined set of parameters against which the features of a

product or component may be judged.

Acceptance Weld

A weld that meets all the requirements and acceptance criteria

prescribed by the applicable welding code, standard, and/or
specification.

Active Flux

A flux for submerged arc welding that causes changes in weld

metal composition that depends on the welding parameters used,
especially voltage.

Alloy

A metallic material made up of two or more elements, where at

least one is a metal.

Alternating Current

Current flow in an electrical circuit where its direction (and

therefore, direction of electron flow) continually reverses itself,
usually at a pre-determined frequency.

Angle of Bevel

See preferred term Bevel Angle.

Arc Blow

The deflection of an arc from its normal path because of magnetic

forces.

Arc Force

The axial force developed by an arc plasma.

Arc Length

The distance from the tip of the welding electrode to the weld
pool.

Arc Plasma

A gas that has been heated by an arc to at least a partially ionized

condition, enabling it to conduct an electric current.

Arc Voltage

The voltage across the welding arc.

Arc Welding Gun

A device used to transfer current to a continuously fed

consumable electrode, guide the electrode and direct the shielding
gas and the arc.
A device used to transfer current to a fixed electrode, position the
electrode and direct the shielding gas and the arc.

Arc Welding Torch

Autogenous Weld

A fusion weld made without addition of filler metal.

17

Back Gouge

To remove base metal and/or unfused weld metal from the

unwelded root side of a joint to create a suitable groove for weld
metal to be deposited. See also GTSM.

Backing Ring

Joint backing in the form of a ring, generally used in welding pipe.

Backing Strip

Joint backing in the form of a strip.

Backing Weld

Joint backing in the form of a weld.

Bare Electrode

A filler metal electrode that is manufactured as a wire, strip or bar

with no coating or covering other than that which is incidental to its
manufacture or preservation.

Barium Titanate

A polarized ceramic material used to create a piezoelectric signal

in transducers (probes) for ultrasonic inspection. See also
Piezoelectric Crystal, Probe and Quartz.

Base Metal

The metal (material) to be welded, brazed, soldered or cut.

Bead

See preferred term Weld Bead.

Beam Angle

The angle at which a sound beam enters a material during

ultrasonic inspection.

Bevel

An angular type of edge preparation.

Bevel Angle

The angle formed between the prepared edge of a member and a

plane perpendicular to the surface of a member.

BHN

Brinell Hardness Number: In the Brinell Hardness Test, a number

which denotes a materials hardness, correlating directly to the
diameter of the indentation obtained by the test.

Buildup

A surfacing operation where material is deposited on the surface

to restore dimensions.

Butt Joint

A joint between two members aligned in approximately the same

plane.

Chill Ring

See preferred term Backing Ring.

Cobalt 60 (Co 60)

A radioactive isotope that emits gamma rays for use in GammaRadiography. See also Radioisotope.

18

Code

A document often considered synonymous with standard or

specification, however, more often it will be found to further
incorporate rules of good practice by which the results required by
a standard or specification may be obtained. In the USA, code
is used as an equivalent to standard in Canada.

Coefficient of Absorption

A nuclear property of a material characterizing its ability to absorb

radiation.

Cold Crack

A crack caused by the presence of hydrogen in the weld zone and

that occurs at relatively low temperatures as the weld cools,
usually below 100C.

Complete Joint Penetration

Joint penetration in which the weld metal completely fills the

groove and is fused to the base metal throughout its total
thickness.

Compression

The type of force which tends to press an object, or a surface of

an object, together.

Concave Weld Surface

A weld surface that has a cross-sectional profile curved like the

inner surface of a circle.

Consumable Electrode

An electrode that melts and provides metal to fill the joint.

Covered Electrode

A composite filler metal electrode consisting of a core of a steel

rod to which a covering sufficient to provide a slag layer on the
weld bead has been applied. The covering may contain materials
providing such functions as shielding from the ambient air,
cleansing the weld metal, arc stabilization, or source of metallic
addition to the weld.

Convex Weld Surface

A weld surface that has a cross-sectional profile curved like the

outer surface of a circle.

Crater

A depression in the weld face at the termination of a weld bead.

Defect

A discontinuity that has been evaluated and determined to exceed

the application acceptance criteria of the relevant code, standard
and/or specification, i.e., rejectable discontinuity. See also
discontinuity and flaw.

Deflection

The movement of a structure or object, usually referring to a beam

or column, as a result of being subjected to a load.

Deposition Rate

The weight of material deposited in a unit of time.

19

Deposition Efficiency

The ratio of the weight of filler metal deposited in the weld metal to
the weight of the filler metal melted, expressed as a percentage.

Depth of Fusion

The distance that weld fusion extends into the base metal or
previous pass from the surface melted during welding.

Destructive Testing

Also known as Mechanical Testing, it is the process of testing a

sample by loading until failure occurs. See Module 12,
Mechanical Testing of Welds.

Developer

In Liquid Penetrant Inspection, the developing agent used, after

the removal of excess penetrant, to draw out and form a
contrasting background for the penetrant.

Direct Current Electrode

Negative (DCEN)

An arrangement of direct current arc welding leads in which the

electrode is the negative pole and workpiece is the positive pole of
the welding arc.

Direct Current Electrode

Positive (DCEP)

An arrangement of direct current arc welding leads where the

electrode is the positive pole and workpiece is the negative pole of
the welding arc.

Discontinuity

Any disruption in the normal physical or compositional features of

a part. A discontinuity is not necessarily a defect.

Drag Angle

When the electrode points towards the start of the weld, the angle
between the electrode center line and the seam center line in the
direction of travel.

Ductility

A term referring to a materials ability to be plastically deformed

without fracturing.

Duty Cycle

The percentage of time during a specified test period that a power

source can be operated at the rated output without overheating.

Dwell Time

In Liquid Penetrant Inspection, the time that the penetrant is in

contact with the material being inspected.

Effective Throat

The minimum distance from the root of the weld to its face, less
any reinforcement. See also Size of Weld.

Elastic Limit

The maximum limit of stress a material is able to be subjected to

without being permanently deformed. See also Yield Point.

Elastic Deformation

The non-permanent change in an objects dimensions while being

subjected to stress that is below the elastic limit.
20

Electrode

A component of the electrical circuit that terminates at the arc,

molten conductive slag, or base metal.

Electrode Extension

The length of unmelted electrode extending beyond the end of the

contact tube.

Electrode Setback

The distance the electrode is recessed behind the constricting

orifice of the plasma arc torch, measured from the outer face of
the nozzle.

Elongation

In tensile testing, a term used to describe the increase in distance

between gauge marks on the test specimen after testing. It is
usually expressed as a percentage of the original gauge length.

Essential Variables

A variable that if changed would affect the mechanical properties

of the deposited weld metal and/or weldment. A change in an
essential variable of a prescribed welding procedure would require
re-qualification.

Fatigue

A phenomena usually resulting in fracture caused by repeated or

fluctuating stresses which, at a maximum, are below the ultimate
tensile strength of the material. These failures are progressive,
begin as minute cracks and propagate due to the action of the
cyclical stresses.

Fatigue Failure

Failure of an object or weldment as the result of fatigue.

Fatigue Strength

The maximum stress per specified number of cycles that can be

sustained without occurrence of failure.

Feather Edge

See preferred term Root Edge.

Ferrous Alloy

A metal composition consisting primarily of iron and one or more

other elements.

Filler Metal

The metal or alloy to be added in making a welded joint.

Flat Position

The welding position when welding is performed on the upper side

of the joint and the weld face is approximately horizontal.

Flaw

Synonymous with defect, a flaw is an unacceptable discontinuity.

See also Discontinuity.

Fluorescent Method

For either Liquid Penetrant Inspection or Magnetic Particle

Inspection, the use of a detecting media that is fluorescent under
ultra-violet (black) light.
21

Flux

A material used to provide a slag cover on the molten weld pool to

prevent its contamination from the atmosphere and control the
amount of impurities in the weld metal.

Fusion (Fusion Welding)

The melting together of filler metal and base metal, or base metal
only, to produce a weld.

Gamma Radiography

A radiographic technique which utilizes the gamma radiation by

the decay of a radioisotope to produce an image on a recording
media. See also Gamma Rays and Radioisotope.

Gamma Rays

The electromagnetic radiation emitted by the decay of

radioisotopes, such as Cobalt 60 and Iridium 192, used in Gamma
Radiography.

Groove Angle

The total included angle of the groove between parts to be joined

by a groove weld.

Groove Radius

The radius used to form the shape of a J or U-groove weld joint.

Groove Type

The geometric configuration of a groove.

Gouge to Sound Metal

(GTSM)

The process of back gouging to a depth to where sound weld

metal, previously deposited from the other side, is achieved so
that a weld with complete fusion through the root is obtained.

Hardness

The relative resistance of a metal to plastic deformation. May also

refer to resistance to abrasion, scratching or indentation.

Heat Affected Zone

The portion of the base metal adjacent to the weld metal whose
mechanical properties or microstructure have been changed due
to the heat of welding.

Heat Treatment

A procedure or combination of procedures involving the heating of

a metal or alloy to a predetermined temperature and then cooling
it at some specified rate so as to obtain desire properties.

Horizontal Position
(fillet weld)

The welding position when a fillet weld is deposited on the upper

side of an approximately horizontal surface and against an
approximately vertical surface.

Horizontal Position
(groove weld)

The welding position when the axis of the weld is approximately

on a horizontal plane, and the weld face lies in an approximately
vertical plane.

22

Incomplete Fusion

A weld discontinuity formed when the weld metal does not

completely fuse with the substrate (base metal or previous weld
beads).

Included Angle

See preferred term Groove Angle.

Inert Gas

A gas that does not participate in any chemical reaction at all.

Inspection Cycle

The complete cycle involved in inspection beginning with the

examination of drawings, specifications, weld procedures,
consumables, equipment, operator qualifications, etc., through to
fit-up and pre-weld operations. Inspection during welding should
ensure that deviation from the weld procedure does not occur.
The inspection cycle is not complete until all aspects of
fabrication, including repair work, final dimension checks, and heat
treatment, have been finished.

Ionizing Radiation

Electromagnetic radiation of sufficient energy to cause electrons to

be stripped from the atoms they strike. Typically capable of
damaging cellular tissue.

Iridium 192 (Ir 192)

A radioactive isotope which emits gamma rays for use in Gamma

Radiography. See also Cobalt 60.

Joint Build-Up Sequence

The order in which the weld beads of a multi-pass weld are

deposited with respect to the cross-section of the joint.

Joint Design

The joint geometry together with the required shape, dimensions

and strength of the welded joint.

Joint Geometry

The shape of the joint to be welded and its dimensions.

Joint Penetration

The distance that the weld metal extends from its top surface
(excluding reinforcement) into the joint.

Joint Welding Sequence

See preferred term Joint Build-Up Sequence.

Lack of Fusion

See preferred term Incomplete Fusion.

Land

See preferred term Root Face.

Layer

A stratum of weld metal or surfacing material. The layer may

consist of one or more weld beads laid side by side.

23

Longitudinal Waves

In ultrasonic inspection, sound waves in which the particle motion

or vibration within the test materials is in the same direction as the
propagated wave.

Manual Welding

Welding performed by a welder who holds and manipulates the

torch, gun or the electrode holder, and moves the arc/weld pool
along the weld joint.

Mechanized Welding

Welding performed with the torch or gun held and moved along by
a mechanical device, with the operator making occasional
adjustments based on visual observation of the weld.

Melting Rate

The mass or length of electrode melted in a unit time.

Modulus of Elasticity

Also known as Youngs Modulus, it is the ratio of stress, below the

elastic limit, to strain. In essence, it is the measure of the stiffness
or rigidity of a material.

Necking

The reduction of the cross-sectional area of a material, in a

localized area, when in tension. Necking begins to occur when
the ultimate tensile strength of the material has been exceeded.

Nonconsumable Electrode

An electrode that does not melt but sustains the welding arc.

Non-Destructive Testing
(NDT)

Any of several examination methods where a component or

assembly is evaluated without damaging or otherwise lessening its
intended service life.

Open Circuit Voltage

The voltage between the output terminals of a power source when

no current is being drawn from it.

Overhead Position

The welding position where welding is performed from the

underside of the joint.

Partial Penetration Joint

A joint where the design does not require the weld throat to equal
the workpiece thickness.

Pass

A single progression of a welding or surfacing operation along a

joint, weld deposit or substrata. The result of a pass is a weld
bead, layer or spray deposit.

Penetrameter

In Radiography, a device used for the validation of the techniques

image quality. Penetrameters are made from similar material as
the test specimen and its thickness is relative to the thickness of
the test piece. Also known as an Image Quality Indicator.

24

Penetrant

In Liquid Penetrant Inspection, a liquid that has the ability to enter

extremely small surface openings by capillary action.

Penetrating Ability

In Radiography, the ability of a particular technique to penetrate a

certain object. This depends primarily on wavelength, with shorter
wavelengths having greater penetration.

Permanent Set

The amount of plastic deformation remaining in a material after

the stress causing the deformation has been removed.

Piezoelectric Crystal

A material used in Ultrasonic probes (transducers) capable of

producing a Piezoelectric Effect. See also Barium Titanate,
Piezoelectric Effect, Probe and Quartz.

Piezoelectric Effect

In Ultrasonic Inspection, the property of certain materials to

generate mechanical vibrations when subjected to electrical
pulses, and vice versa. See also Piezoelectric Crystals.

Porosity

Round or oblong discontinuities in weld metal formed as a result

of entrapment of gas during weld metal solidification.

Probe

A device used in Ultrasonic Inspection, consisting of a

Piezoelectric Crystal, which may transmit and/or receive sound
pulses and convert these into either mechanical vibrations or
electrical pulses. See also Piezoelectric Crystal.

Procedure Qualification
Record

A record of welding parameters used to produce a sound weld in

a specified material in accordance with a welding procedure
specification, such that the weld also meets the specified
mechanical property requirements.

Prod Method

In Magnetic Particle Inspection, the method utilizing a prod that

can locate surface and sub-surface indications parallel to the
alignment of the poles of the prod.

Quartz

A material used to create a piezoelectric signal (effect) in

transducers (probes) for ultrasonic inspection. See also Barium
Titanate, Piezoelectric Crystal and Probe.

Radiography Sensitivity

The ease at which images of fine object features can be detected.

Radiographic Technique

The entire Radiographic method used during testing in terms of

radiant energy used, wavelength, Source-to-Film distance, film
used, material and material thickness, etc.

25

Radiography (RT)

A Non-Destructive Testing Method in which radiant energy is used

in the form of either X-rays or Gamma-rays for the volumetric
examination of opaque objects. See also Gamma Radiography,
Radiographic Technique and X-Ray Radiography.

Radioisotope

A naturally or artificially produced isotope that releases ionizing

radiation during its decay. See also Cobalt 60 and Iridium 192.

Root

See preferred term Root of Joint or Root of Weld.

Root Edge

A root face with zero width.

Root Face

Portion of a bevelled edge preparation that is left substantially

perpendicular to the workpiece surface, usually to prevent burn
through.

Root Gap

See preferred term Root Opening.

Root of Joint

The portion of a joint to be welded where the members are closest

to each other. In a cross-section, the root of the joint may be
either a point, a line or an area.

Root of Weld

The points as shown in cross-section at which the back of the

weld intersects the base metal surfaces.

Root Opening

The separation between the members to be joined at the root of

the joint.

Semi-automatic Welding

Welding operation where the filler metal is fed automatically into

the weld pool but a welder holding a gun or torch controls the
travel speed, travel angle and the work angle.

Shear

The type of force that produces an opposite but parallel sliding

motion between two parts in the same plane.

Shear Waves

In Ultrasonic Inspection, sound waves in which the particle motion

or vibration within the test material is perpendicular to the direction
of the propagated wave. See also Wavelength.

Shielding Gas

Gas delivered through a welding gun or torch with the objective of

protecting the arc and the weld metal from atmospheric
contamination.

Single-Welded Joint

In arc or gas welding, any joint welded from one side only.

26

Size of Weld

Groove Weld: The joint penetration (depth of bevel plus root

penetration when specified). The size of a complete penetration
groove weld and its effective throat are one and the same.
Fillet Weld: The leg lengths of the largest right-angle triangle that
can be inscribed within the fillet weld cross-section.

Slag

A glassy substance formed on top of the weld metal as a result of

melting of the flux and its reaction with the weld metal.

Slope

Quantitative measure of the incline of the power source

volt\ampere curve.

Slugging

The act of adding a separate piece, or pieces, of material in a joint

before or during welding that results in a welded joint not
complying with design, drawing or specification requirements.

Source-to-Film Distance

In Radiography, the distance between the source of radiation and

the recording medium (film).

Space Strip

A metal strip or bar prepared for a groove weld, and inserted in

the root of a joint to serve as a backing and to maintain the root
opening during welding. It can also bridge an exceptionally wide
gap due to poor fit-up.

Specification

A document that usually sets forth in some detail the requirements

and/or acceptance criteria demanded by a buyer for a certain
product. It may be, or become the basis of a contractual
agreement between the buyer and the supplier. See also Code
and Standard.

Standard

A document by which a product may be judged. In terms of

welding, a standard generally summarizes the requirements for
processes, procedures, consumables, materials, inspection,
acceptance criteria, etc. See also Code and Specification.

Strain

A measure of the change in dimensions of a body due to the

presence of stress.

Stress

The internal force induced in a material to counter-balance an

externally applied force. Mathematically, it is the applied force
divided by cross-sectional area, and is represented by the Greek
letter sigma, .

Stress/Strain Curve

A graph that plots the stress (y-axis) against the strain

(x-axis) of a material during a tensile test.
27

Surface Tension

Force at the surface of liquid that tries to reduce its surface area
and prevents it from wetting the solid that it is in contact with.

Surfacing

Use of a welding process to deposit a layer of a similar or different

material on the surface of a workpiece to restore dimensions or to
achieve desired properties (corrosion resistance, wear resistance,
etc.).

Tensile Strength

See preferred term Ultimate Tensile Strength.

Tension

The type of force that tends to pull an object, or a surface of an

object, in opposite directions.

Toughness

The ability of a metal to absorb energy and deform plastically

before fracturing.

Transition Curve

A graph that plots the energy value obtained in an impact

toughness test (y-axis) versus specified temperatures (x-axis).

Transition Temperature

The temperature at which the transition curve shows a sharp

change in toughness.

Travel Angle

The angle between the electrode axis and the perpendicular to the
weld axis in a plane defined by the electrode and weld axis.

Ultimate Tensile Strength

The maximum stress from tension that a material can withstand

without fracture. Mathematically, it is the maximum load applied
divided by the original cross-sectional area.

Undercut

A groove or a notch formed in the base metal adjacent to a weld

toe.

Volt\Ampere Curve

A graphical representation of the voltage-current relationship for a

given power source when a steady load is placed on it.

Wavelength

The distance a wave travels through one complete cycle. See

also Longitudinal Waves and Shear Waves.

Weld Bead

A weld deposit resulting from a pass.

Weld Face

The surface of the weld opposite to the root.

Weld Metal

That part of an arc weld that was completely molten at one time
during welding.

28

Weld Pool

The molten metal, prior to its solidification, under and adjacent to

the arc.

Weld Reinforcement

Weld metal in excess of the quantity required to fill a joint.

Weld Root

The region of a weld pass where the underside of a weld bead

meets the base metal.

Welding Head

A part of a completely mechanized welding equipment set-up that

incorporates the gun or the torch, wire feeder and wire spool.

Welding Inspector

A person specially trained in any applicable aspect of welding,

fabrication and inspection of weldable materials in terms of
judging a weldments compliance against a prescribed acceptance
criteria.

Welding Leads

Cables that are part of the electrical circuit and connect the power
source to the electrode (electrode lead) and to the workpiece
(workpiece lead).

Welding Procedure

The details of materials, joint geometry, welding consumables,

welding parameters, preheat/interpass temperature/postweld heat
treatment, etc., and related practices and procedures for the
production of welds.

Weldment

Any fabricated component or unit to which welding has been

applied.

Wetting

Phenomenon that allows liquid weld metal to easily spread over

and fuse with the base metal.

Wire Feed Speed

The rate (length per unit time) at which wire is fed and melted in
welding.

Workpiece

The member that is to be welded.

X-Ray Radiography

Radiographic method in which X-rays are utilized to produce a

permanent image on a recording medium. See also Gamma
Rays and Radioisotope.

X-Rays

In Radiography, a form of relatively high radiant energy, created

by the bombardment of electrons on a material at high voltage.
See also Radiography and X-Ray Radiography.

29

Yield Point

The first point at which a material under load experiences an

increase in strain without an increase in stress. It is the stress
levels at which plastic deformation begins. Not all metals exhibit a
definite yield point. See also Elastic Limit.

Yield Strength

The stress at which the yield point is reached. Mathematically, it

is the load applied at the yield point divided by the original crosssectional area.

Yoke

In Magnetic Particle Inspection, a device used to locate surface

and sub-surface indications transverse to the alignment of the
pole.

Youngs Modulus

See preferred term Modulus of Elasticity.

30

Chapter 2
Welding Codes and Standards

Table of Contents
2.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33

2.2

Purpose of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

2.3

Development of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

2.4

Administration of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38

2.5

CSA Standard W47.1 Certification of Companies for Fusion Welding of Steel . . . . . . . . . . .39
2.5.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39
2.5.2 Company Certification to CSA Standard W47.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40

2.6

CSA Standard W47.2 Certification of Companies for Fusion Welding of Aluminum . . . . . . .43
2.6.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43
2.6.2 Similarity Between W47.1 and W47.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43
2.6.3 Major Differing Provisions Between W47.1 and W47.2 . . . . . . . . . . . . . . . . . . . . . . .44

2.7

CSA Standard W48.01 Filler Metals and Allied Materials for Metal Arc Welding . . . . . . . . . .45

2.8

CSA Standard W59 Welded Steel Construction (Metal Arc Welding) . . . . . . . . . . . . . . . . . .45
2.8.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45
2.8.2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46

2.9

CSA Standard W59.2 Welded Aluminum Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

2.9.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

2.10

CSA Standard S6 Design of Highway Bridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50

2.11

CSA Standard S16-01 Limit States Design of Steel Structures . . . . . . . . . . . . . . . . . . . . . .50

31

2.12

CSA Standard W186 Welding of Reinforcing Bars in Reinforced Concrete Construction . . .51
2.12.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51
2.12.2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51

2.13

CSA Standard W178.1 Qualification Code for Welding Inspection Organizations . . . . . . . . .53
2.13.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53
2.13.2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .54

2.14

CSA Standard W178.2 Certification of Welding Inspectors . . . . . . . . . . . . . . . . . . . . . . . . .55

2.14.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55
2.14.2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56

2.15

National Building Code of Canada (NBC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57

2.15.1 Provincial Building Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57

2.16

CSA Standard Z662 Oil and Gas Pipeline Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57

2.17

ASME - American Society of Mechanical Engineers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .58

2.18

AWS - American Welding Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60

2.19

AWS Codes of D-Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61

2.20

AWS A5 Specifications - Filler Metal Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61

2.21

ANSI/AWS D1.1 Structural Welding Code Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62

2.22

ISO Standards (International Standards Organization) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63

32

2.1

Introduction

30 kN

The heading of this chapter may give the reader an impression that the subject is dry and full of legal
jargons. Not if you are just starting your design career and the senior engineer assigns you a welding
design project, in which one component is as shown in Figure 2.1. You are to calculate the fillet size
required for the Tee-joint given:

For simplicity, it is assumed that the flange welds

resist shear in the X-X direction and web weld
resists shear in the Y-Y direction. The length of
welds are given:

100

S1

E4918 electrode, 1 mm fillet = 0.156 kN/mm

S2 100
X

Flange welds:

60 kN

S1 50-50

S1 =

60 kN
2.0 mm
200 0.156

Web welds:

W 310 x 118 COL.

S2 =

PL 35 x 500 x 500

30 kN
2.0 mm
100 0.156

Figure 2.1

What is the weld size you would indicate on your design sketch 2.0 mm? If you do, you are wrong.
Why? Because the code, CSA W59 or AWS D1.1 and others stipulate that the minimum fillet size for
35 mm thick plate is 8 mm. Why?
The answer to this question involves a lot more consideration than just the simple strength calculation.
It involves welding heat input, cooling rate, welding metallurgy, weld mechanics, weld cracking, and last
but not least, practicality. It is impossible to lay down a fillet weld that small in a structural fabrication
shop. If you do show 2.0 mm fillet on your design sketch, the draftsman will know that you have no
welding design experience.
We need codes and standards to legally protect our professional career and, most importantly, to
ensure public shafety.

33

Note also that construction specifications always specifiy conditions such as the following:
g

The welding fabrication must be done in accordance with CSA W59 or AWS D1.1.

Welders and welding operators shall be qualified by the Canadian Welding Bureau (CWB)
according to CSA W47.1.

Electrodes used on this project shall be of the E49XX series and certified by the CWB.

Inspection shall be done according to CSA W59 by CWB certified companies and
personnel.

As another example, the following clauses are part of the requirements by CAN/CSA S16-01 Limit
States Design of Steel Structures.
16.6.17.3
Fabricators and erectors of welded construction covered by this standard shall be certified by
the Canadian Welding Bureau in Division 1 or Division 2 to the requirements of CSA Standard
W47.1 or W55.3, or both, as applicable. Specific welding procedures for joist fabrication shall
be approved by the Canadian Welding Bureau.
16.9.5

Installation of Steel Deck

16.9.5.2
(a)

The installer of steel deck to be fastened to joists by arc spot welding shall be certified
by the Canadian Welding Bureau to the requirements of CSA Standard W47.1.

(b)

The installation welding procedures shall be approved by the Canadian Welding Bureau.

(c)

The welders shall have current qualifications for arc spot welding issued by the
Canadian Welding Bureau.

24.3

Fabricator and Erector Qualification

Fabricators and erectors responsible for making welds for structures fabricated or erected under
this Standard shall be certified by the Canadian Welding Bureau to the requirements of CSA
Standard W47.1 (Division 1 or Division 2), or CSA Standard W55.3, or both, as applicable. Part
of the work may be sublet to Division 3 fabricators, however, the Division 1 or Division 2
fabricator or erector shall retain responsibility for the sublet work.

34

31.5

Third-Party Welding Inspection

When third-party welding inspection is specified, welding inspection shall be performed by firms
certified to CSA Standard W178.1, except that visual inspection may be performed by persons
certified to Level 2 or 3 of CSA Standard W178.2.
The design engineers responsible for preparing the construction documents must be familiar with all
the relevant clauses in the structural steel design and welding specifications. The fabricators
engineers responsible for welding design and procedures must also be familiar with both the applicable
welding and structural steel specifications to produce a satisfactory structure.
Codes and standards set the level of acceptable quality so that the owner knows what quality product
to expect and the engineers know what specifications govern the design, fabrication and erection. It is
the common technical (and legal) language among owners, contractors, architects and engineers. It
forms an important part of all engineering contractual documents.
Some of the major codes or specifications pertaining to welding fabrication will be mentioned with brief
outlines in the following paragraphs.

2.2

Purpose of Standards

In this time of rapid development of new technologies, standards can be said to fulfill the all-important
task of harnessing these developments into product performance or service-oriented regulatory
constraints, setting levels of acceptable quality while taking a responsible stand on the protection of
public interest.
The ultimate objective of standardization is to build confidence in the user (public), and in so doing,
stimulate production and commercial activity and in turn greatly enhance the economic well being of
the country.
In more specific terms, the functions of standards is:
g

to ensure public safety

to educate by setting general rules for guidance of producers and consumers

to simplify by reducing the range of variations in sizes, processes and hence the
stock and related record-keeping services

to conserve by saving time and materials through ready and official acceptance of
developments permitting the use of more advanced design methods and the attainment of
higher production efficiencies

to certify by serving as hallmarks of quality and value

35

The Canadian structural welding standard system merits special mention. Its uniqueness lies in the
successful combination of advanced material and design standards and their well conceived integration
with Canadas certification standards. Since 1947, the certification standards have established a solid
foundation of technological capability in welding engineering and supervisory personnel for the
Canadian welding industry. The superior performance record of welded steel structures in this country
is a direct result.

2.3

Development of Standards

Most national standards are generated through the voluntary efforts of all segments of society,
government and industry, producers and consumers, institutions and individuals.
In most Western countries, the governments freely relinquish their prerogative to impose control or
direction on the standardization system. However, some feel compelled to design a mechanism to
prevent overlap and duplication in the system, to develop procedural methods to coordinate its output,
and at the same time ensure the widest possible national acceptance of standards.
This particular function is fulfilled:
g

in Canada by the Standards Council of Canada (SCC), established by an Act of Parliament;

in the USA by the American National Standards Institute (ANSI)

Most of the standards of the Western world are voluntary standards. That means:
g

they are developed with the voluntary cooperation of all concerned

the use of the standard by those affected (meaning the specifiers) is also totally voluntary
(unless otherwise mandated)

Such standards, however, become mandatory when so designated by a pertinent regulatory authority
or when specified contractually.
The standards are also consensus standards, meaning that everyone affected by the development or
use of a standard has an opportunity to participate in the development of these standards, either
directly or indirectly through representation or through public review.

36

Since a balanced representation on a committee is of overriding importance, it will serve the purpose to
identify the most common interest groups. These particular groups, or rather the individual members of
each group, are defined by the Standards Council of Canada as follows:
1.

Producer: in the context of a standards-writing committee, a producer is a representative of an

organization involved with the manufacture or promotion of the product, material or service of
concern to that committee.

2.

User: in the context of a standards-writing committee, a user is an individual or a representative

of an organization concerned with the use or application of the product, material or service of
concern to that committee.

3.

General Interest: in the context of a standards-writing committee, a general-interest member is

an individual or a representative of an organization who is not associated with the production,
distribution, direct use or regulation of the product. This category is intended to include
professional and lay people employed by academic and scientific institutions, safety
associations, etc.

4.

Consumer: in the context of a standards-writing committee, a consumer is an individual who

uses goods and services to satisfy his needs and desires, rather than to resell them or produce
other goods with them.

Standards are developed in many different types of organizations:

g
g
g
g

companies
trade associations
governmental agencies
technical and professional societies

Therefore, there are many degrees of consensus involved in developing a voluntary standard. By no
means must all standards originate from the consensus system. Some are intended for use only within
a company, an industry or governmental agency.
However, a standard dealing with a commodity servicing an open market should be developed by a full
consensus procedure. Its balanced interest representation will:
g
g

permit it to attain a high level of credibility

make it eligible for recognition as a national standard

37

2.4

Administration of Standards

The term administration is taken strictly to imply an ongoing activity on the part of an agency, to which
a standard has given a clear mandate to monitor the maintenance of its requirements.
The classical example of such standards are the Canadian Standards Association (CSA) welding
certification standards, whether concerned with products (e.g., electrodes) or organizations (e.g.,
fabricators, inspectors or inspection companies). The Canadian Welding Bureau (CWB), Certification
Division (a Division of the CWB Group Industry Services) is charged with the task of administering
and monitoring these standards. It is the responsibility of CWB to monitor, through an appropriate
mechanism, the manufacturers or the companys adherence to the full conditions of the standard
under which certification has been granted.
There are other administrative models. In Canada, the area of boilers and pressure vessels is under
the jurisdiction of provincial governments. The designated departments of the appropriate ministries
perform a function similar to the responsibility of CWB (to monitor), but the extent of this function is
defined by the applicable Pressure Vessel Act.
In the United States, boiler and pressure vessels are governed by ASME codes (American Society of
Mechanical Engineers).
In some European countries, agencies have been established to fulfill the functions similar to those
performed by the CWB. The assignment of these agencies may also include an inspection function.
In Canada, the CWB certifies welding inspectors and welding inspection organizations.

38

2.5

CSA Standard W47.1

Certification of Companies for Fusion Welding of Steel
2.5.1 General

The CSA W47.1 Standard becomes the foremost qualification code in structural welding in Canada.
The Standard specifies the conditions and personnel qualification requirements that shall be met for a
company to become certified.
Company certification under W47.1 is a unique concept that has been implemented not only by the
Canadian welding industry, but also in many countries around the world. It is administered by the
Canadian Welding Bureau (CWB).
The first edition was published in 1947 as CSA W47, and since then other editions have been published
in 1973, 1983, 1992 and 2003. The CSA W47 Standard is re-numbered as CSA W47.1 to differentiate
it from CSA W47.2, which is the standard for Certification of Companies for Fusion Welding of
Aluminum.
CSA W47.1 Standard is mandated in Canada (and therefore W47.1 company certification) through
various CSA Design Standards such as those for buildings and bridges, which are similarly mandated
by the National Building Code of Canada (NBCC), and by Provincial Building Codes or Bridge Codes.
All major governing design specifications in Canada make certification to W47.1 Standard a mandatory
requirement. For example, as stated in the introduction of CSA S16-01, Limit States Design of Steel
Structures.
CSA W47.1 Standard is interlinked with other welding Standards, e.g., CSA W59 Welded Steel
Construction (Metal Arc Welding), and the CSA W48 Standard Filler Metals and Allied Materials for
Metal Arc Welding.
Although company certification to W47.1 is a mandatory requirement of Canadian structural fabricating
companies, where not mandated in welded manufacturing, it is frequently required as a quality control
measure.
The following is a brief outline of the Standard. Consult the text for full details.
1)

The W47.1 Standard is explicitly concerned with certification of companies. It is not a product
standard and cannot be used to either evaluate or approve products. Consequently, it is not
intended to supersede or encroach on codes or other jurisdictions governing the manufacture of
specific products such as pressure vessels (ASME, API, CSA-B51).

2)

Although the basic provisions of the code can be construed to indirectly constitute some
measure of quality assurance, it must be stressed that it remains the responsibility of the
purchaser (owner) to ensure, through adequate inspection, that the required quality of welded
fabrication is attained.
39

3)

Canadian Welding Bureau (CWB) representatives audit certified companies periodically, the
sole objective being to monitor each companys compliance with the conditions of CSA W47.1
Standard. These CWB audits are not to be construed as inspection of welds or inspection of
welded products, and they do not diminish the need for inspection by others as stated above.

4)

Canadian Welding Bureau is not a government organization, nor a division of CSA. CWB is a
federally incorporated, not for profit organization.

The uniqueness of the certification system under CSA W47.1 lies in the principles for qualification of
the company, including specifically:
g
g
g
g
g

the
the
the
the
the

employment of qualified welders

employment of qualified welding supervisors
employment of qualified welding engineers (full time or retained)
approval of welding procedures
administration of the Standard by a single independent third party (CWB)

2.5.2 Company Certification to CSA Standard W47.1

The following is a brief explanation of the procedural steps a candidate company has to take in order to
acquire the CSA W47.1 certification status:
1)

Make Formal Application

W47.1 specifies Each company applying for certification shall make formal application to the
Bureau. The application is signed by the CEO and applies only to the plant or site identified in
the application.
The company must indicate in which division it wishes to be certified:
g

Division 1: the company shall employ a welding engineer on a full-time basis.

Division 2: the company shall employ a welding engineer on a part-time basis.

Division 3: the company is not required to employ or retain a welding engineer.

It should be noted that a full time qualified welding supervisor must be employed by all
divisions.
2)

CEO Shall Designate Key Personnel

The CEO shall designate engineering, shop and field supervisory and quality control personnel,
giving them the authority to act and be responsible to the company in their respective work
areas. These persons shall be designated on a form signed by the CEO.

40

3)

Resumes for Designated Welding Engineer and Welding Supervisors

Work Experience Required by Welding Engineers: A minimum of five years of welding-related
experience is required. When the engineer responsible is retained, he/she shall report in
writing through the company to the Bureau on his/her effective participation in the companys
welding operations.
Work Experience Required by Welding Supervisors: A certified company (for all divisions) shall
have at least one full time welding supervisor. Each supervisor shall meet the following criteria
for work experience:
g

The welding supervisor shall have a minimum of five years of welding-related experience,
a thorough knowledge of companys welding procedures, be able to read drawings,
interpret welding symbols, know weld faults, quality control and inspection methods.

Educational Requirements Engineering Personnel: Each engineer shall be a member of a

professional engineering association. The educational requirements for the engineering
personnel include the academic background and tangible evidence of additional courses of
study involving examinations on a number of welding-related areas. Briefly, the additional
courses of study would include subjects such as:
g

Weldability of metals; fatigue and brittle fractures; welding procedures and practices;
welded joints and connections; welding processes, equipment and materials; weld faults;
and methods of control of quality.

Additionally, engineers shall have knowledge of the applicable welding codes and
standards.

Educational and Practical Requirements Welding Supervisors

Educational Requirements: Each welding supervisor shall have knowledge of applicable CSA
welding standards (e.g., CSA W47.1 and CSA W59) pertaining to his/her normal work.
Additionally, each supervisor shall have a knowledge of weld faults, quality control and
inspection methods, and be able to read and interpret drawings, all pertaining to his/her normal
work.
Practical Requirements: Each welding supervisor shall have practical welding related
experience; and shall have a thorough knowledge of the companys welding procedure
specifications and related welding procedure data sheets. The supervisor shall be familiar with
the operation of the various types of welding equipment related to his/her work with the
company.

41

4)

List of Welding Equipment and Any Quality Control System Used

The equipment list will define the welding scope of the companys operations, and will assist the
Bureau when assessing the qualifications of designated welding personnel.

5)

Provide a List of Welding Personnel

This list refers to the welders, welding operators and tack welders. Before company certification
can be granted, each welder, welding operator and tack welder shall be qualified for the welding
process(es) and welding position(s) in which he/she welds. (See also Welder Qualification).

6)

Provide a File on Companys Welding Standards

The company shall prepare a set of Welding Standards, including Welding Procedure
Specifications (WPS), and related Welding Procedure Data Sheets (WPDS), which shall be
submitted to the Bureau for acceptance/approval.
Welding procedures, using joints designated as prequalified in the applicable governing
standard or code, and which satisfy procedural stipulations as they apply to the welding
process, shall be accepted by the Bureau.
Welding procedures which do not meet the aforementioned conditions, and for which sufficient
testing information has not been accumulated, shall undergo procedure qualification testing in
accordance with the provisions of the W47.1 Standard. Welding of the test assemblies shall be
witnessed by the Bureaus representative.

7)

Qualification of Welders and Welding Operators (Performance Tests)

W47.1 describes specific plate test assemblies for the qualification of welders, welding
operators and tack welders. There is also a pipe test assembly for welder qualification.
Qualification by welding non-standard test assemblies may be allowed where special welding
conditions exist due to equipment or plant operations. Qualification is governed by
classification, welding process, mode of process application and position of welding (see the
following paragraphs).
The Bureaus representative shall witness all welding personnel qualification tests, and shall
issue a record showing each persons welding qualification. The welding qualification issued for
a welder/operator is subject to special validating conditions, and is recognized only while the
welder/operator or tack welder is employed by a company certified under CSA W47.1.
After certification, the company shall report to the Bureau, showing names and qualifications of
all welding personnel (tack welders, welders and welding operators).

42

8)

Welding Processes
CSA W47.1 addresses the following welding processes:
g
g
g
g
g
g
g
g

2.6

shielded metal arc welding (SMAW)

flux-cored arc welding (FCAW)
metal cored arc welding (MCAW)
gas metal arc welding (GMAW)
submerged arc welding (SAW)
gas tungsten arc welding (GTAW)
electroslag welding (ESW)
electrogas welding (EGW)

CSA Standard W47.2

Certification of Companies for Fusion Welding of Aluminum
2.6.1 General

This standard specifies requirements for certification of companies engaged in fusion welding of
aluminum alloys and erection of aluminum structures. It is similar to the CSA W47.1 Standard.

2.6.2 Similarity Between W47.1 and W47.2

1)

The concept of three divisions with the same distinguishing criteria applicable.

2)

The certification and administration procedures to be followed by the Bureau with identical
obligations on the part of the company for maintaining the condition of certification (reporting of
changes in welding personnel and welding procedures).

3)

The educational and practical experience requirements for the engineering and supervisory
personnel in Divisions 1 and 2, but with the years of experience for the latter increased to 4 in
case of Division 3.

4)

The requirements related to Welding Procedure Specification and Welding Procedure Data
Sheets.

43

2.6.3 Major Differing Provisions Between W47.1 and W47.2

1)

2)

The application of the Standard is restricted to:

i)

commercial fabrication of aluminum structures and their repair - specialized product

fabrication is totally excluded (pressure vessels)

ii)

thickness 3 mm or greater

The welding processes are limited to include only:

gas metal arc welding (GMAW)
gas tungsten arc welding (GTAW)
plasma arc welding (PAW)
arc and capacitor discharge process for stud welding (SW)

g
g
g
g
3)

Essential variables related to each of the welding processes are listed and base metal alloy
groupings as another variable clearly tabulated.

4)

Welding Procedure Qualifications include plate and pipe test assemblies for groove welds and
plate assemblies for fillet welds. In case of pipe assemblies a 6G (inclined 45 to the
horizontal) non-rotating pipe has been introduced. A fracture test has been added to normal
W47.1 procedural tests.

5)

The concept of performance levels has been introduced for welder qualification:
Level I
Level II
Level III

6)

designating fillet welding only

designating welding of groove joints either from both sides or from one side with
backing
designating welding of groove joints from one side without backing for the full
thickness of material

In addition to the performance levels the qualification of welders and welding operators is
governed by:
g
g
g
g

the welding process

mode of process application (semi-automatic, automatic)
type of weld and position
the filler metal alloy group in the case of the GMAW process

7)

Pipe and plate test assemblies are provided for with a fracture test.

8)

While in W47.1 qualification for F, H, V, OH was designated as for example class F

qualification, the W47.2 Standard uses the term category F.

44

2.7

CSA W48-01
Filler Metals and Allied Materials for Metal Arc Welding

The CSA W48 Electrode Standard is a companion to the W59 and W47.1 Standards.
Fabricators undertaking work specified to CSA Standard W59, Welded Steel Construction, are required
to use welding electrodes and filler metals conforming to CSA or equivalent standards.
The W48 Standard covers the specifications of the following types of electrodes:
g
g
g
g
g
g

carbon steel covered electrodes for shielded metal arc welding

chromium and chromium-nickel steel covered electrodes for shielded metal arc welding
low-alloy steel covered electrodes for shielded metal arc welding
solid carbon steel filler metals for gas shielded arc welding
carbon steel electrodes for flux and metal cored arc welding
fluxes and carbon steel electrodes for submerged arc welding

This standard prescribes certification requirements for electrodes for the given individual welding
process.
The CSA W48 Standard is administered by the Canadian Welding Bureau, which is under obligation to
publish lists of certified electrodes at yearly intervals.
The objective of certification under the CSA W48 Standard is to demonstrate the properties of weld
metal deposited in a standard joint under specified and controlled welding conditions.
The CWB Module 6 Electrodes and Consumables, covers this standard in more detail.

2.8

CSA Standard W59

Welded Steel Construction Metal Arc Welding
2.8.1 General

The CSA W59 Standard is considered to be the primary steel welding standard in Canada. As already
pointed out, it is directly linked with the CSA Standards W47.1 and W48 dealing with certification of
companies and filler metals respectively, and because of its stipulations involving the other two, may be
considered greatly responsible for the resounding success of structural welding quality throughout
Canada.
In conjunction with the CSA W178.1 qualification code for welding inspection organizations and
individual inspectors there is in Canada an encompassing framework of standards and an extremely
well-integrated system of certification and qualifications, all geared to provide a reliable measure of
assurance of safe performance of welded structures in service and hence of public safety.

45

W59 Includes:
g
g
g
g
g

workmanship standards and technique

prequalified details of joints and welding processes
inspection procedures and acceptance criteria
design strengths under static and cyclic loadings for welds for allowable stress and limit
states design methods
allowable stress ranges for fatigue loading

It should be noted that Clauses 1 to 10 of the Standard cover the requirements common to all types of
structures and:
g
g

Clause 11 governs welding of statically loaded structures

Clause 12 governs welding of cyclically loaded structures

It should be further noted that in these two clauses, provisions are included for:
g
g

the allowable stress design (ASD) method

the limit states design (LSD) method

This is in view of both methods being used in current engineering practice, although the phasing out of
the ASD method in other governing CSA design standards (S16-01) indicates that the LSD design
approach will eventually become the preferred design method.

2.8.2 Review
A number of provisions in the Standard are of significance in that they clearly define the extent of its
coverage and the specifications mandated upon the designer, the fabricator and the inspection agency.
These are briefly discussed next, occasionally with explanatory background reasoning for their
inclusion into the Standard where feasible or deemed necessary.
1)

Clear statement is made with respect to types of steel structures excluded from the coverage.
Reference is made to the distinct requirements provided by other regulatory authorities
exercising jurisdiction and having specific expertise pertinent to related, given products (e.g.,
water pipes American Water Works Association AWWA, pressure vessels ASME, API).

46

2)

Applicable welding processes are those used in actual fabrication operations:

g
g
g
g
g
g
g
g
g

shielded metal arc welding (SMAW)

flux cored arc welding (FCAW)
metal cored arc welding (MCAW)
gas metal arc welding (GMAW)
submerged arc welding (SAW)
gas tungsten arc welding (GTAW)
electroslag welding (ESW)
electrogas welding (EGW)
stud welding (SW)

A further requirement of the Standard is that the filler metals (consumables) for each process be
approved either in accordance with the provisions of the CSA W48 electrode standard series or
when not applicable in accordance with the pertinent provisions of the CSA W47.1 Standard.
3)

Pre-approved materials, meaning those acceptable without reservation, are identified as steels,
whose specified minimum yield strength does not exceed 700 MPa (100 ksi). A comprehensive
listing of the eligible steels together with their CSA or ASTM designation is provided.

4)

An important requirement of the Standard is directed towards the technological capability of the
fabricator. It requires the fabricator:
g

to be either certified under the provisions of the CSA W47.1 Standard, or

to demonstrate competency to produce welded structures of desired quality and soundness

to the engineer, the professionally qualified, designated representative of the regulatory
authority or of the purchaser, as applicable.

Although this requirement appears optional, the fact is that in almost all types of steel structures
it can be said to be mandatory by virtue of other, governing design standards demanding
certification to the W47.1 standard. As a matter of fact, imposed on this fundamental
requirement is the additional stipulation in the CSA S16-01 Standard Steel Structures for
Buildings that only fabricators certified to Division 1 and 2 under the CSA W47.1 standard are
eligible to undertake work on any steel structures the design of which is governed by S16-01.
5)

With respect to inspection, the Standard stipulates that preferably, organizations certified to
CSA W178 Qualification Code for Welding Inspection Organizations be used. The nonmandatory certification to W47.1 (point 4) and to W178 (point 5) was prompted by no trade
restriction considerations.

47

6)

7)

Fundamental concepts in strength calculations of welded joints and connections are provided.
These concepts are of basic importance in assessing the capacity (ASD) or the resistance
(LSD) of welds for which pertinent formulae are tabulated in Clauses 11 and 12 and which
involve:
g

types of welds, types of groove welds

their minimum and effective sizes together with separate provisions for fillet welds, plug
and slot welds

Requirements governing the workmanship and welding technique of the fabricator are included:
g

conditions for matching filler-base materials for welding of corrosion resistant steels

maximum exposure times and subsequent storage and conditioning of electrodes,

especially those of the low hydrogen type

specified limits of acceptability of planar discontinuities in base material and recommended

action for repair of edge discontinuities

preparation of material for welding with tolerance limits for assembly and fit-up of structural
elements

workmanship tolerances for the preparation and fit-up of groove welded joints

provisions for tack welds, temporary welds and seal welds

details of welding procedures and techniques for each welding process

extensive treatment of stud welding

recommended sequences in assembly and welding aiming at minimizing distortion

and residual stresses

preheat and interpass temperatures as tabulated are related to steel designations

(Carbon equivalent expressing their weldability), thickness of material, and low
hydrogen or non-low hydrogen electrodes or processes.

conditions permitting reduction of preheat temperatures for minimum single pass SAW fillet
sizes

dimensional tolerances for finished welded structural elements

acceptable profiles of fillet and groove welds

corrective action for defective welds

requirements for stress relief, when specified. These refer to temperature, holding time and
rates of heating and cooling
48

8)

Requirements, with respect to welding inspection include:

g

an emphasis on advanced communications between the inspection organizations and the

fabricators as well as timely scheduling of inspection

a clear statement of the obligations of the fabricator in matters arising from results of
inspection

conformance of nondestructive (NDT) procedures to pertinent clauses of the Standard and

applicable ASTM codes

9)

Repair and strengthening of existing structures.

10)

The concept of prequalification applies to SMAW, FCAW, MCAW, GMAW and SAW welding
processes.

11)

As already pointed out, Clause 11 includes all design and construction provisions for statically
loaded structures. Clause 12 covers the requirements for cyclically loaded structures with
particular emphasis on fatigue, the inherent mode of behaviour under fluctuating loads typical of
bridges and crane runways.

2.9

CSA Standard W59.2

Welded Aluminum Construction
2.9.1 General

This standard is similar to CSA W59 (for steel construction) in format but specifies the requirements for
welded aluminum construction for general applications. For special applications such as pressure
vessels, pipelines or the aviation industry, other standards applicable to that specific type of fabrication
should be used. This standard gives provisions for the following welding processes:
g
g
g
g

gas metal arc welding (GMAW)

gas tungsten arc welding (GTAW)
plasma arc welding (PAW)
stud welding (arc and capacitor discharge process) (SW)

The Standard gives guidelines of design of welded connections, joint geometries, filler alloy selection,
filler metal alloy groupings and base metal alloy groupings. Inspection methods and acceptance
criteria are also provided. The appendix lists the physical properties and the mechanical properties of
alloys with various tempers, which is very important information for design engineers.

49

2.10

CSA Standard S6
Design of Highway Bridges

CSA Standard S6 is a comprehensive design standard encompassing all essential engineering

considerations, with separate coverage requirements for steel-reinforced and pre-stressed concrete,
and timber. In the steel section, a separate clause on welds:
g

invokes the provisions of the CSA Standard W59 Welded Steel Construction

reserves fabrication only for companies certified to CSA Standard W47.1 Certification of
Companies for Fusion Welding of Steel

requires inspection to be performed by either the designer or inspection organizations

certified to CSA Standard W178.1 Qualification Code for Welding Inspection
Organizations

Although the mandatory use of electrodes certified to the CSA W48 Standard is inherent in the W59
and the W47.1 Standards, the S6 Standard calls for their certification separately for greater emphasis.

2.11

CAN/CSA-S16-01
Limit States Design of Steel Structures

S16-01 covers a wide scope with rules and requirements for design, fabrication and erection of steel
structures, with considerable attention given to joining and fastening material by welds and bolts.
This is a limit states design Standard for steel structures. The current edition at the time of this writing
was preceded by seven working stress design editions dating back to 1924, and six limit states design
editions beginning with the 1974 edition. The Standard is prepared in SI (metric) units. In accordance
with the provisions in the Standard, the working stress design method has been officially withdrawn.
It should be noted that this National Standard of Canada, CAN/CSA-S16-01, has been adopted by the
National Building Code of Canada as a reference Standard for steel structures, thereby further
reflecting its dominant position among the governing Canadian design standards.
Although primary attention of S16-01 is directed to statically loaded structures, it includes provisions for
the design of fatigue loaded structures. Since the latter provisions appear in a number of other CSA
Standards (e.g., W59, S6), it is of importance to note that though a concerted and coordinated intercommittee effort these provisions have been kept identical in all pertinent standards.

50

The comprehensive design requirements of the Standard are said to incorporate the latest research
recommendations. Exhaustive design coverage is given to beams and girders, open web steel joists,
composite beams and columns, built-up members and connections involving welds and bolts.
With respect to welding, reference is made to the CSA Standard W59 and W55.3 for design and
practice in arc welding and resistance welding respectively, while for the same types of welding
processes, certification of fabricators is required either to the CSA W47.1 (Division 1 and 2 for primary
contractors) or the CSA W55.3 Standards.
Workmanship requirements include fabrication tolerances, erection and inspection.
Appendices provide explanatory background information or complementary information.

2.12

CSA Standard W186-M1990

Welding of Reinforcing Bars in Reinforced Concrete Construction
2.12.1

General

It should be first stressed that the W186 Standard is primarily a combined welding design and
certification standard, which in addition prescribes workmanship and inspection requirements.
Its scope of application expressed in terms of welding is precisely defined by types of joints and
connections, types of welding processes and types of base materials.

2.12.2

Review

With respect to joints and connections, it covers welding of reinforcing bars either directly to one
another or through splice members or to structural steel members used in anchorages in pre-cast or
cast-in-place concrete construction.
The accepted welding processes include the:
g
g
g

shielded metal arc welding (SMAW)

gas metal arc welding (GMAW)
flux cored arc welding (FCAW)

together with:
g
g

pressure gas welding (PGW)

thermit welding (TW)

51

In case of the first three, emphasis is put on low-hydrogen or controlled-hydrogen electrode

classifications. All electrodes for these processes are required to be certified to the CSA W48 Standard
while filler metals for the last two processes are subject to procedure qualification as prescribed in the
Standard.
Base materials are identified by referencing pertinent reinforcing steels of the CSA G30 Standard
series while a number of structural steels are listed under their CSA G40.21 or ASTM designations.
The administration of the Standard and specifically of its certification program is left with the Canadian
Welding Bureau with the usual requirements for the Bureau to check on the maintenance of code
conditions and to publish lists of certified fabricators.
The detailed and comprehensive design provisions of the Standard are based on the limit states (LSD)
principles with the SI (metric) units used throughout. Described are types of bar splices, types of bar to
structural steel anchorage connections together with types of welds used (grooves and fillets). Special
attention is given to flare grooves (welds between two round bars in a longitudinal lap joint and welds
between a round bar and flat plate also in a lap joint). The effective sizes and lengths of all types of
welds are established.
The minimum factored resistances of joints are defined and a number of formulae delivering these
resistances provided. Included are also formulae precluding any other predictable or possible modes
of failure associated with a given joint configuration. A comprehensive tabulation of design applications
relating bar or plate material to electrode classifications for all possible types of welds is provided.
The provisions for workmanship are based on principles of good welding practice. Low temperature
limitations for welding, preparation, assembly requirements and the mandatory use of approved
welding procedures with emphasis on proper application of preheat are clearly specified. Options for
welding of galvanized steel are offered. Storage and conditioning of electrodes and the quality of
welds are covered.
In the part on certification, the Standard has adopted identical requirements to those in the W47.1
Standard. They include:
1)

qualified:
g
g
g

engineering personnel, employed or retained

supervisory personnel
welding personnel

2)

approved welding procedures

3)

adequate welding equipment

52

However, of great significance is the fact that fabricators already certified in Divisions 1 or 2 of CSA
W47.1 are accepted as certified under the W186 Standard with only a few additional requirements
related mainly to welding procedure qualifications and welder qualification for flare groove welding.
Conditions attached to the qualification of the engineering and supervisory personnel include
educational and practical experience requirements. However, in each case as already stated, pertinent
qualification under W47.1 is given almost full recognition under this Standard.
Welding procedure qualifications are covered with appropriate test assemblies and types of tests
together with specified ranges of acceptable test results.
Qualification of welders is given similar comprehensive coverage with validity extended to two years.
Qualification on flare grooves is taken as acceptable for fillet welds while fillet qualifications under the
W47.1 Standard are also considered valid.
For other than visual inspection, the Standard requires the use of inspection organizations certified to
the CSA W178.1 Standard.
In one of the appendices, typical design solutions are provided for guidance of the designer. All
classical types of joints and welds are used to illustrate the calculation procedures.

2.13

CSA Standard W178.1

Qualification Code for Welding Inspection Organizations
2.13.1

General

The W178.1 Standard uses the qualification concept of the W47.1 Standard and hence encompasses
the full organization or its respective division performing welding inspection. As a national standard
administered by the Canadian Welding Bureau, it serves to ensure uniform and reliable inspection
capabilities of companies engaged in quality control of welded structures.
In view of the growing national and international appreciation of quality in the manufacturing process,
the competency of inspection services assumes ever-increasing importance. Design sophistication and
advanced exploration of material capabilities further substantiate this importance.
The W178.1 Standard, although clearly written for certification of independent inspection organizations,
does not preclude its application to the manufacturers or fabricators own inspection systems. Its main
objective is to set basic requirements for obtaining and maintaining certification in any of the clearly
identified inspection service categories.

53

2.13.2

Review

The administration provisions define the responsibilities as well as the extent of authority of the Bureau
at the time of granting of certification and during the subsequent surveillance of the companys
adherence to the conditions of certification.
Included are provisions for specific cases where retention by contract of qualified personnel and
equipment is necessary for a given company to fully meet the requirements of the Standard. Clear
reference is also made to the Bureaus obligation to periodically publish lists of certified companies to
serve the industry. On the other hand, the companies are required to immediately report any changes
in personnel and equipment.
Eleven separate categories of certification are provided for selection by the companies. These relate
to distinctive product or group of products oriented fabrications. The specific requirements for
certification include the following:
1)

Standard inspection procedures providing clear instructions with respect to execution of the
basic and most common inspection functions such as checking the qualification of welding
personnel, determining the availability of approved welding procedures, identifying base and
filler materials, establishing the extent of inspection prior to and during welding together with
acceptance inspection and preparation of reports.

2)

Standard testing procedures covering all destructive and non-destructive methods contemplated
for use by the company.

3)

Inspection personnel who meet the qualification requirements of the Standard

4)

Suitable inspection equipment and test facilities.

The Standard distinguishes between the following three levels of inspection personnel:
g
g
g

certified welding division supervisors

certified welding inspectors
qualified operators for test facilities and equipment

It sets specific educational, training and experience criteria for each level. Mobility of qualified
personnel within certified companies is permitted.
In the case of radiographic and ultrasonic inspection methods (anticipated for use in service by the
company) certification as senior industrial radiographers and senior ultrasonic operators in accordance
with the respective Canadian General Standards Board (CGSB) specifications is required.

54

The very demanding educational requirements for the division supervisor extend to his/her knowledge
of principles and application of welding processes, through understanding of inspection methods, weld
discontinuities and the applicable welding codes and standards. He/she may be required to
demonstrate his/her knowledge by means of examinations. Reference is made to acceptable courses
of study, upon successful completion of which the foregoing requirements may be partly reduced.
With respect to the second personnel level, separate provisions are made for junior and senior welding
inspectors with the former to perform routine inspection but under the supervision of the responsible
senior personnel. Obviously, less demanding educational requirements are set for the junior inspector
level requiring some knowledge related to welding and experience in certain capacities.
In the case of senior inspectors, the Standard stipulates more advanced educational requirements,
extending to a proven ability to interpret drawings and inspection results, to understanding governing
codes, and to demonstrate acceptable familiarity with the qualification system of welders and welding
procedures. A longer period of practical welding fabrication or welding inspection is also stipulated.
These superior qualification requirements for the senior key inspection personnel are thought to add
markedly to their capability to properly carry out their inspection responsibilities.
With respect to operators of other testing equipment like that for the Magnetic Particle or Liquid
Penetrant methods, certification to pertinent CGSB specifications is required, while in the case of other
equipment the qualification is left to the discretion of the Bureau.

2.14

CSA W178.2
Certification of Welding Inspectors
2.14.1

General

The main objective of the W178.2 Standard is to provide further assistance to the industrys efforts to
produce quality products by providing welding inspection personnel certified individually with qualified
skills and capability.
Effectively supporting this main objective is the fact that the prospective applicants need not be
members of an inspection organization. This adds a good measure of flexibility to the setting of the
manufacturers own quality programs.
Recognizing the fact that the integrity of inspection is largely dependent on the theoretical knowledge
and practical experience on the part of those performing it, the W178.2 Standard establishes
appropriate educational and experience criteria considered adequate to ensure the required level of
inspection competence. In its scope it also logically provides a link with the W178.1 Standard.

55

The administration of the Standard is entrusted with the Canadian Welding Bureau, relying on its best
professional judgment in the implementation of all those provisions, where such judgment is necessary.
Although not explicitly required by the Standard, the Bureau publishes a list of certified welding
inspectors at proper time intervals to serve the manufacturing industry.

2.14.2

Review

The Standard sets certification requirements for three levels of inspector personnel (levels 1, 2 and 3)
in an ascending order of competence with some provisions for trainees. It defines the responsibilities
together with the related competency requirements for each inspector level. In the case of
radiographic, ultrasonic, magnetic particle, and liquid penetrant inspection methods, certification to
appropriate Canadian General Standards Board (CGSB) specifications is mandatory.
Experience in welded fabrication or welding inspection in one or more capacities of several listed for
each area, is required. The years of experience increase with each level and are directly related to
educational requirements in a manner that allows for reduction in time with more advanced and
substantial educational backgrounds.
In addition to practical tests on visual detection and identification of faults, open book examinations are
specified on inspection standards for any category of products (10 listed) for which certification is
sought. Closed book examinations are specified for welding, inspection and metallurgy with the
required extent of knowledge duly apportioned for each level.
It is important to note that suitable recognition is given to inspection personnel employed by companies
certified under the W178.1 Standard. Certification in Level 2 is granted to inspectors qualified by the
American Welding Society when certain conditions are met, and AWS provides reciprocal recognition.
Submission of evidence of satisfactory vision is required. Duration of validity is set for 3 years with
procedures for renewal fully described.
Finally, the code of ethics is invoked to further stress the importance that the Standard attaches to the
integrity of welding inspectors.

56

2.15

National Building Code of Canada (NBC)

The National Building Code of Canada is published by the National Research Council. Prepared by
the Canadian Commission on Building and Fire Codes, it comprises nine parts serving as models of
technical requirements with respect to public health and safety in buildings, and is suitable for adoption
by appropriate legislative authorities in Canada. In Canada, under the terms of the Constitution Act,
provincial and territorial governments are responsible for the regulation of buildings, and therefore, the
NBC has become the basis for provincial building codes. It also has been widely accepted in municipal
bylaws. In its subsection on steel in Part 4 Structural Design, the NBC requires buildings and their
structural members to conform to CAN/CSA-S16-01 Standard, Limit States Design of Steel Structures.
Additionally, in its Appendix, a reference identifies the specific clause in S16-01 that requires
fabricators and erectors of welded structural steel to be certified to the requirements of CSA Standard
W47.1, Certification of Companies for Fusion Welding of Steel, in either Division 1 or Division 2.
The Users Guide NBC, Structural Commentaries (Part 4), has detailed coverage on topics such as:
g
g
g
g
g

serviceability criteria for deflections and vibrations

wind loads
snow and rain loads
effects of earthquakes
foundations

2.15.1

Provincial Building Code

Each province or major metropolitan has its own building code which, in large measure, is based on
the National Building Code. There may be minor variations that may be more stringent.

2.16

CSA Standard Z662

Oil and Gas Pipeline Systems

The purpose of CSA Standard Z662 is to establish essential requirements and minimum standards for
the design, construction and operation of oil and gas industry pipeline systems. These requirements
and standards apply to conditions normally encountered (as opposed to abnormal or unusual
conditions) in the oil and gas industry. This Standard is published in SI (metric) units, and its first
publication in 1994 combined and superseded the two CSA Standards Z183 and Z184, Oil Pipeline
Systems and Gas Pipeline Systems respectively.

57

This Standards clause on Joining includes an extensive coverage on welding (arc welding, gas
welding, explosion welding and roll welding). It stipulates its own welding procedures and welder
qualifications together with details of essential variables, test assemblies, preparation of test
specimens, inspection and testing of production welds, and acceptance criteria. Arc welding
consumables shall be in accordance with the requirements of the CSA W48 Standard, which is
administered by the Canadian Welding Bureau Certification Division. Therefore, the welding
consumables shall be certified by the Bureau.
Other clauses in the Standard relate to:
g
g
g
g
g
g
g
g
g

materials
installation
pressure testing
corrosion control
operating, maintenance and upgrading
offshore steel pipelines
gas distribution systems
plastic pipelines
oilfield steam distribution pipeline systems

There are also eight (8) non-mandatory appendices to CSA Standard Z662, including one on Limit
States Design.

2.17

ASME American Society of Mechanical Engineers

ASME, through its Council on Codes and Standards, is recognized worldwide as a major standardssetting organization. Founded in 1880 as an educational and technical society, it continues to pursue
its basic objectives through dissemination of technical information and promotion of economic, reliable
and safe practices in a wide area of product-oriented engineering and manufacturing activities.
One of the avenues used most effectively towards this objective is standards development. ASME
directs this particular activity through its ten Code and Standards Boards, which exercise full jurisdiction
over the standards-generating committees, each responsible for a specific area of standards
development. To ensure full implementation of the standards, it uses accredited companies for
certification of compliance with its codes.
Of primary interest is the Pressure Technology Board and specifically the Pressure Vessel Codes
under the auspices of the ASME Boiler and Pressure Vessel Code Committee. The function of the
committee is to establish rules of safety covering design, fabrication, inspection and testing of boilers,
pressure vessels and associated equipment during original construction. In formulating these rules,
consideration by the committee is given to the needs of the manufacturers, users, inspectors and
regulatory agencies. The objective of the rules is to provide a margin of deterioration in service so as
to ensure a reasonably long and safe period of usefulness. Advances in material technology and new
experience are also considered.

58

The ASME Boiler and Pressure Vessel Code is published every three years. It has been adopted in 46
states in the USA, numerous municipalities, all provinces in Canada (see CSA B.51) and is used in
several other countries. Addenda are published regularly to maintain an updated status of the code.
For general information, the following are the 11 sections of the ANSI/ASME Boiler and Pressure
Vessel Code:
I)

Power Boilers

II)

Material Specifications
Part A Ferrous Materials
Part B Nonferrous Materials
Part C Welding Rods, Electrodes and Filler Metals

III)

Subsection NCA General Requirements for Division 1 and Division 2

Division 1

Subsection NB Class 1 Components

Subsection NC Class 2 Components
Subsection ND Class 3 Components
Subsection NE Class MC Components
Subsection NF Component Supports
Subsection NG Core Support Structures
Appendices

Division 2

Code for Concrete Reactor Vessels and Containments

IV)

Heating Boilers

V)

Nondestructive Examination

VI)

Recommended Rules for Care and Operation of Heating Boilers

VII)

Recommended Rules for Care of Power Boilers

VIII)

Pressure Vessels
Division 1
Division 2 Alternate Rules

IX)

Welding and Brazing Qualifications

X)

Fiberglass-Reinforced Plastic Pressure Vessels

XI)

Rules for Inservice Inspection of Nuclear Power Plant Components

59

Of greater interest is section IX and its part dealing with welding procedures and performance
qualifications. A cursory review of those in comparison with similar requirements in the CSA W47.1
Standard would reveal that, among others, differences exist in:
g

the definition of essential variables, these being more numerous and more restrictive in
W47.1

section IX listing a separate set of variables where impact testing is a part of procedure
qualification

performance qualification, with section IX:

i)
ii)
iii)
iv)

2.18

providing for a 6G assembly (inclined pipe section)

ruling differently on the extent of validity on basis of a test in one given position
not specifying a mandatory time limit for check testing
not permitting mobility for qualified welders

American Welding Society (AWS)

The American Welding Society was founded in 1919. It has gained wide national and international
recognition over the years for its contribution to the transfer of welding technology, a service that is
essential to research, development and application engineering, as well as to manufacturers and
users.
One of the most successful activities is the Societys outstanding input into the formulation of
standards. As an accredited standards-developing organization under ANSI guidelines, the AWS can
publish, within the prescribed rules, American National Standards (ANS) pertaining to welding.
Balanced representation on committees is strictly enforced, and a two level review is provided as
follows:
1)

by the Technical Activities Committee (TAC), for technical content and adherence to rule
of operation, and

2)

by the Technical Council for publication

60

There are 22 technical AWS committees responsible for more than 100 standards covering a wide
range of areas related to welding.
It should be noted that there is a strong and active Canadian participation on the AWS committees.
This participation is dictated by the desire to keep welding standards in both countries at the same
advanced level and without any major differences, to effectively support the high volume of trade
between them.

2.19

AWS Codes of D Series

The AWS Code of D Series numbers from D1.1 to D18.2. The frequently referred ones relevant to
structural fabrication are listed below:
D1.1
D1.2
D1.3
D1.4
D1.5
D1.6
D3.6

2.20

Structural Welding Code Steel

Structural Welding Code Aluminum
Structural Welding Code Sheet Metal
Structural Welding Code Reinforcing Steel
Bridge Welding Code
Structural Welding Code Stainless Steel
Specification for Underwater Welding

AWS A5 Specifications: Filler Metal Specifications

The AWS A5 Specifications consist of 32 specifications which cover a wide range of alloys for various
welding processes. The following specifications are more relevant to fabrication shops:
AWS A5.7-84R

Specification for Copper and Copper Alloy Bare Welding Rods and
Electrodes

AWS A5.9-93

Specification for Bare Stainless Steel Welding Electrodes and

Rods

AWS A5.10:1999

Specification for Bare Aluminum and Aluminum Alloy Welding

Electrodes and Rods

AWS A5.11/A5.11M-97

Specification for Nickel and Nickel Alloy Welding Electrodes for

Shielded Metal Arc Welding

AWS A5.14/A5.14M-97

Specification for Nickel and Nickel Alloy Bare Welding Electrodes

and Rods

AWS A5.22-95

Specification for Stainless Steel Electrodes for Flux Cored Arc

Welding and Stainless Steel Rods for Gas Tungsten Arc Welding
61

2.21

AWS A5.23/A5.23M-97

Specification for Low Alloy Steel Electrodes and Fluxes for

Submerged Arc Welding

AWS A5.28-96

Specification for Low Alloy Steel Electrodes and Rods for Gas
Shielded Arc Welding

AWS A5.29-1998

Specification for Low Alloy Steel Electrodes for Flux Cored Arc
Welding

ANSI/AWS-D1.1
Structural Welding Code Steel

As the most prominent of the D1 series, this Code provides comprehensive rules pertaining to the
construction of welded steel structures. It covers the design and strength of welds, qualification
requirements for welding procedures and welding personnel, workmanship, inspection and quality
acceptance criteria.
The Code consists of eight sections, annexes and commentary as follows:
Section 1
Section 2
Section 3
Section 4
Section 5
Section 6
Section 7
Section 8
Annexes and

General Requirements
Design of Welded Connections
Prequalification of WPSs
Qualification
Fabrication
Inspection
Stud Welding
Strengthening and Repairing Existing Structures
Commentary

The Code covers only arc welding processes and incorporates the concept of prequalified welding
procedures in conjunction with the use of prequalified joint details.
It is comparable to the CSA Standard W59 Welded Steel Construction, except for its extensive
coverage of tubular structures and its requirements for qualification of welding procedures and welding
personnel.
It should be noted that it is a conformance code. It is up to the fabricators to voluntarily adopt and
conform to it and the AWS does not monitor and enforce their adherence as does the Canadian
Welding Bureau on CSA W59 and W47 with certified companies.
The AWS D1.1 Code, with its commentary, has been published on a yearly basis. As an authoritative
document, it enjoys worldwide recognition.

62

2.22

ISO Standards
International Standards Organization

The ISO Standards were originated by European Common Market Countries. They are now adopted
by some North American companies as well. The Standards give guidelines for quality control in
industrial and business operations.
The Standard that covers arc welding operations is ISO 2553. Since standards will eventually be
globalized, we should all know something about ISO 2553. Although the content of the Standard is not
as extensive as CSA W47.1/W59 and AWS D1.1, the Table of Contents is shown below to give an
outline of ISO 2553:
Welded, Brazed & Soldered Joints Symbolic Representation of Drawings
Contents
1.

Scope

2.

Normative References

3.

General

4.

Symbols
4.1
Elementary Symbols
4.2
Combination of Elementary Symbols
4.3
Supplementary Symbols

5.

Positions of Symbols on Drawings

5.1
General
5.2
Relationship Between the Arrow Line and the Joint
5.3
Position of the Arrow Line
5.4
Position of the Reference Line
5.5
Position of the Symbol with Regard to the Reference Line

6.

Dimensioning of Welds
6.1
General Rules
6.2
Main Dimensions to Be Shown

7.

Complimentary Indications
7.1
Peripheral Welds
7.2
Field or Site Welds
7.3
Indication of the Welding Process
7.4
Sequence of the Information In-the-Tail of the Reference Mark

8.

Examples of Application of Spot and Seam Joints

ISO 3834

Quality Systems for Welding

ISO 9606

Welder Qualification Procedure

63

64

Chapter 3
Weld Joints and Welding Symbols

Table of Contents
3.1

Introduction

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67

3.2

Definition of Joint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68

3.2.1 Types of Basic Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69

3.3

Definition of Weld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71

3.3.1 Basic Types of Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71

3.4

Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .73

3.4.1 Single Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74
3.4.2 Double Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75

3.5

Prequalified Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76

3.6

Positions of Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79

3.6.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79
3.6.2 Designation of Welding Positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80
3.6.3 Positions of Groove Welds in Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .81
3.6.4 Positions of Groove Welds in Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .82

3.7

Joint Edge Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83

3.8

Fundamental Concepts of Welding Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87

3.8.1 Weld Symbols, Supplementary Symbols, Welding Symbols . . . . . . . . . . . . . . . . . . . .87

3.9

Basic Weld Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88

65

3.10

Supplementary Weld Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95

3.10.1 Field Weld Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95
3.10.2 Melt-thru Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95
3.10.3 Contour Symbol and Finishing of Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95
3.10.4 All-Around Weld Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97

3.11

Break in Arrow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98

3.12

Combined Weld Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100

3.13

Information in Tail of Welding Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102

3.14

Extent of Welding Denoted by Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103

3.15

Multiple Reference Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103

3.16

Complete Penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105

3.17

Groove
3.17.1
3.17.2
3.17.3
3.17.4
3.17.5
3.17.6

3.18

Fillet Welds
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117
3.18.1 Symbols of Fillet Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118
3.18.2 Size of Fillet Welds - Equal Leg Fillet Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . .122
3.18.3 Minimum and Maximum Fillet Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
3.18.4 Conventional Fillet Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
3.18.5 Size of Fillet Welds - Unequal Leg Fillet Welds . . . . . . . . . . . . . . . . . . . . . . . . . .126
3.18.6 Intermittent Fillet Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127

3.19

Plug Welds
.............
3.19.1 Size of Plug Welds . .
3.19.2 Angle of Countersink .
3.19.3 Depth of Filling . . . . .
3.19.4 Spacing of Plug Welds

Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .107
Location of Dimensions for Single Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . .107
Dimensions for Double Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .108
Depth of Preparation and Groove Weld Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110
Flare-Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
Surface Finish and Contour of Groove Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . .114
Joints with Backing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

66

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

. . .129
. . .129
. . .130
. . .130
. . .132

3.1

Introduction

Welding consists of joining two or more pieces of metal by the application of heat and sometimes
pressure. In electric arc welding, the heat comes from an electric arc and no pressure is employed to
fuse the metal parts. In most applications of arc welding, filler metal is added to the joint which is
specially prepared in certain shapes, like a mold, to receive the molten filler metal. In some
applications, the metal parts are fused together without additional filler metal.
Since welding is related to making joints, the student should first be familiar with the terminology of
welds and joints. Not only must the names of these joints and welds be familiarized, but also the
systems by which they are technically represented. It is through the correct usage of the terminology
that we can communicate with each other in this field in the most effective and exact manner.
This chapter is the abridged version of the following CWB Modules. Students are advised to study
them for more detailed information.
Module 2
Module 3

Engineering Drawings, Basic Joints and Preparation for Welding

Symbols for Welding

67

3.2

Definition of Joint

JOINT: The junction of members or the edges of members which are to be joined or have been joined.
The following figures show various joints and it can be seen that an alternative description of a joint
might refer to the faying surfaces which are in contact. While this is not entirely correct, it will assist
the student in deciding on the joint which is present under certain conditions.
Look at the joint shown in Figure 3.1 and at the same time, consider the definition of the work joint
and also the faying surfaces which are in contact.

The student should realize that

there is only one joint shown in
Figure 3.1, and that joint extends
the whole length of the plate.

Figure 3.1

Now, look at Figure 3.2, the assembly consisting of three plates. Consider the number of joints and
select your answer from the following:
1 joint only?

2 joints?

3 joints?

4 joints?

Check your answer.

68

ANSWERS

COMMENTS ON ANSWERS

1 joint only

No. You are thinking of one assembly which after welding will form one
weldment. A weldment is an assembly whose component parts are joined by
welding.

2 joints

This answer is correct. The three plates form two joints. The actual joint is the
faying area in contact with the centre plate.

3 joints

No. You are considering three plates which form part of the assembly.

4 joints

No. Perhaps you are considering each side of the joint. For example, there are
four sides where fillet welds could be made. However, these are only two areas
of faying surfaces.

Figure 3.2: The faying surfaces of these two joints have been marked by a thick black line.

3.2.1 Types of Basic Joints

There are five basic joints, although many variations of these result from the manner of preparation
and assembly. These five, illustrated in Figure 3.3, are termed butt joint, corner joint, tee joint, lap joint
and edge joint.
The actual joint is shown as a shaded area on the right side of each joint.

69

Figure 3.3: Five basic joints.

70

3.3

Definition of Weld

A localized coalescence of materials (metals or non-metals) produced either by heating the metals to
suitable temperatures, with or without the application of pressure, or by the application of pressure
alone, with or without the use of filler materials.
The word coalescence is used since coalescence is defined as growing together, or growing into one
body. In welding metals, the metallic bond is formed as the weld is being made.

3.3.1 Basic Types of Welds

There are five basic types of welds which are:
1)

groove weld

2)

fillet weld

3)

plug and slot welds

4)

surfacing weld

5)

flanged weld
Figure 3.4: Groove weld.

1) Groove Weld
A groove weld is a weld made in a groove between the workpieces. There are many different shapes
of grooves. Figure 3.4 shows one type of groove weld.

2) Fillet Weld
A fillet weld is a weld of approximately
triangular cross-section joining two surfaces
approximately at right angles to each other in a
lap joint, T-joint or corner joint as shown in
Figure 3.5.

Figure 3.5: Fillet weld.

71

3) Plug Weld and Slot Weld

A plug weld is a weld made in a circular
hole in one member of a joint fusing
that member to another member. A slot
weld is similar to a plug weld except
that the hole is elongated. See Figure
3.6.
In preparation for plug and slot welds,
holes or slots are made in the upper
plate. On relatively thinner material,
such welds can be made without holes
or slots and are called arc spot and arc
seam welds, in which the upper sheet
is melted and fused to the lower sheet.

Figure 3.6: Plug weld and slot weld.

4) Surfacing Welds
All welds are composed of one or
more weld beads. A bead is a single
run or pass of weld metal. A weld
bead or beads may be applied to a
surface, as opposed to making a
joint, to obtain desired properties or
dimensions. Such a weld is called
surfacing welds, as shown in
Figure 3.7.

Figure 3.7: Surfacing weld.

5) Flanged Weld
Flanged weld is a group term
which covers: corner-flange
welds, edge welds and edgeflange welds. As shown in
Figure 3.8, they are apparently
neither groove welds nor fillet
welds. They are not surfacing
welds because these welds are
forming joints along two members.

Figure 3.8: Flange welds.

72

3.4

Groove Weld

A weld made in the groove between two members to be joined.

Figure 3.9 shows the geometries and welding terms for typical groove weld joints. In order to describe
the geometry of a joint, all the numerical data for plate thickness, bevel or groove angle, groove radius
of J-groove, root face and root opening should be given.

Figure 3.9
The above examples are shown on a single groove joint. All the terms are applicable to double groove
joints as well.
Figure 3.10 shows more terms related to welds and joints.
Note: The weld size or effective
throat (x) is defined in sketches A,
B, C and D. Where joint
penetration is complete as in A
and B, the weld size is the
thickness of the plate. Where the
plates differ in thickness as in C,
and joint penetration is complete,
the weld size is the thickness of
the thinner plate. Where joint
penetration is incomplete as in D,
the weld size is the depth of
penetration.

Figure 3.10: Joint and

welding terms.

3-73

3.4.1 Single Groove Welds

The terms Single Weld and Double Weld should be clarified. A square groove, when welded from
one side, is called a single-square-groove weld as shown in Figure 3.11. When welded from both
sides, it is called a double-square-groove weld (see Figure 3.14).
Figure 3.12 shows a bevel-groove weld that is chamfered on one side only, but welded from both
sides. It is commonly considered as a single-bevel-groove weld.

Figure 3.12: Single-bevel groove weld.

Figure 3.11: Square-groove weld.

The following examples (Figure 3.13) are of single-V-groove welds.

Figure 3.13: Single V-groove welds.

74

3.4.2 Double Groove Welds

Double groove welds are shown in Figure 3.14. When welds are made from both sides of a squaregroove joint or when both sides of the joint have been chamfered to form groove welds on both sides,
then the term Double is used.

Figure 3.14: Double groove welds.

75

3.5

Prequalified Joints

There are some groove weld joints that are designed as prequalified weld joints. These joints meet the
requirements of: joint geometries, welding processes, welding positions, base metal and filler metal
specifications.
The objective to designate certain joints as prequalified is to exclude these joints from the
requirements of welding procedure qualification tests. Economy is a major factor for so doing. The
accumulated experience of the welding industry over the years demonstrates that reliable good
performance of these weld joints can be readily achieved under the prescribed conditions. Also,
designers and fabricators are provided with the best tried and proven practice and they do not have to
go through the trial and error process and welding procedure qualification tests. It should be noted that
different welding codes and standards may differ slightly in the designation of prequalified joints.
There are prequalified joint designated in both complete joint penetration joints and partial joint
penetration joints. Sample prequalified joints are shown in the following tables in which CSA W59
(Welded Steel Construction) and AWS D1.1 (Structural Welding Code) are referenced. It should be
noted that certain joints are designated by AWS D1.1 as prequalified joints, but which are not
prequalified in CSA W59. CSA W59 and AWS D1.1 should be consulted for the complete list of
prequalified joints.
The student is reminded that there are other welding standards with prequalified joints that may be
different from CSA W59 and AWS D1.1.

CSA W59
0

S (E) G
0

RF

Welding
Processes

Joint
Designation

Base Metal
Thickness
T (mm)

Root Face
R F (mm)

SMAW

BC-P2b

$13

FCAW

BC-P2-FC

SAW

BC-P2-S

G=0

Groove
Angle

Permitted
Welding
Positions

45#2<60
60
45#2<60
60

F, O
F, V, O

60

Weld Size
(mm)
S-3
S
S-3
S
S

Tolerances are given in CSA W59, Clause 5.

Figure 3.15: Prequalified partial joint penetration groove welds (sample joint).
76

Figure 3.16: Prequalified complete joint penetration groove welds (sample joint).

77

CSA W59

T(T)

G
0

G
T1

T2

Joint
Welding
Process Designation

SMAW

FCAW

SAW

(1)
(2)
(3)
(4)

B-U4b

TC-U4b-FC

TC-U4a-S

Base Metal Thickness

(U = unlimited)
T1

T2

Groove Preparation
Root Opening G Groove Angle
12mm

20

10

30

45

30

45

30

45

10

30

45

16

20

10

30

45

No prequalified joint for GMAW process.

SP - Straight polarity, electrode negative.
RP - Reverse polarity, electrode positive.
Split pass mandatory in root layer.

Figure 3.17

78

Permitted
Gas
Welding Shielding
Positions for FCAW Polarity
F, O
Only
All

Yes

F, H
Only

SP

No

RP

F, H
F
Only

(2)

No

(3)

3.6

Positions of Welding

With metallic arc welding, it is possible to deposit weld metal in any position with some of the welding
processes, so that a welder may make a joint that is below him, in front of him, above him, or at any
intermediate positions between these welding positions.
The following welding positions are defined and frequently referred to by the welding industry:

3.6.1 Definitions
Terminology

Definitions

Flat Welding Position

The welding position used to weld from the upper side of the joint;
the face of the weld is approximately horizontal, Figure 3.18, 1G
and 1F.

Horizontal Welding Position

Fillet Weld The position in which welding is performed on the

upper side on an approximately horizontal surface and against an
approximately vertical surface, Figure 3.18, 2F.
Groove Weld The welding position in which the weld face lies in
an approximately vertical plane and the weld axis at the point of
welding is approximately horizontal. See Figure 3.18, 2G.

Overhead Welding Position

The position in which welding is performed from the under side of

the joint, Figure 3.18, 4G and 4F.

Positioned Weld

A weld made in a joint which has been so placed as to facilitate

making the weld.

Vertical Welding Position

The position of welding in which the axis of the weld is

approximately vertical, Figure 3.18, 3G and 3F.

Positions of Pipe Welding

The position of a pipe joint in which welding is performed in the

horizontal position and the pipe may or may not be rotated.

Horizontal Fixed Welding

Position

The position of a pipe joint in which the axis of the pipe is

approximately horizontal and the pipe is not rotated during
welding.

Horizontal Rolled Welding

Position

The position of a pipe joint in which the axis of the pipe is

approximately horizontal and welding is performed in the flat
position by rotating the pipe.

79

3.6.2 Designation of Welding Positions

This section will give the student a quick view of the welding positions with respect to groove and fillet
welds made on plate material.
A weld is said to be made in the flat position, horizontal position, vertical position or overhead position
depending on the position of the joint in relation to the floor. Welding techniques for the four positions
of welding vary according to the positions the weld metal is deposited. It is possible to deposit weld
layers of considerable volume in the flat and vertical positions but stringer beads are normally used for
horizontal and overhead positions. These positions are better illustrated in Figure 3.19 to augment
some of the definitions given earlier. The number and letter combinations are used to designate each
welding position for quick reference. The letter G stands for groove weld, letter F for fillet weld. The
numbers 1, 2, 3 and 4 correspond to flat, horizontal, vertical and overhead positions respectively, as
shown in Figure 3.18.

Figure 3.18: Positions of welding.

80

3.6.3 Positions of Groove Welds in Plate

Figure 3.18 shows the welding positions in the most exact manner, but in practical shop fabrication,
the welding positions can be in any of the intermediate positions. Figure 3.19 shows the sectors which
are designated as certain welding positions. The sector angles are measured clockwise from the 0
point as shown. Within one sector, the centerline of a groove cross-section can vary from one radius
to the other, and all the groove welds are considered in the same welding position. It is an
approximation with the actual welding techniques considered in different positions.

Figure 3.19: Positions of groove welds.

81

3.6.4 Positions of Groove Welds in Pipe

Positions of welds in pipe may vary from flat to overhead and all the positions in between if the pipe is
not rotated. Also, the axis of the pipe may vary from 0 (horizontal) to 90 (vertical) and all the angles
in between. Figure 3.20 shows the welding positions around the circumference joint for pipe axis from
0 to 90.

Figure 3.20: Welding position diagram for groove welds in pipe.

82

3.7

Joint Edge Preparation

Plate edges to be welded are prepared according to the joint configurations, be it square, bevelled or Jgrooves. The CWB Module 2 Engineering Drawings, Basic Joints and Preparation for Welding, gives
the full description of this subject. The students are recommended to read Module 2 for methods of
preparation. In this chapter, a brief description of the most common methods will be presented.
Oxyfuel cutting is the most common method used in structural steel fabrication shops. Figure 3.21
shows the cutting torch positions for simple or compound cutting. It should be noted that the cutting is
not done by the heat in the flame.
Briefly, the basic principle of oxygen cutting depends upon the simple fact that steel at red heat will
oxidize rapidly or burn where a jet of oxygen is directed onto it. The ordinary cutting torch enables
this to be done by providing both a heating flame and a pure oxygen jet each with its own controls
the heating flame being used chiefly to preheat the steel where the cut is to be started, after which the
oxygen jet does the cutting.
Only a small area needs to be preheated for starting the process since, as soon as oxidation
commences, the combustion of the steel produces very intense local heat. This further preheats the
metal around the oxidation point, enabling the oxygen jet to pierce almost any thickness of steel, or to
make a cut in whichever direction the torch is moved. After the cut has started, the main function of
the heating flame is to keep the oxide fluid (so that it will leave the cut easily) and to compensate for
heat losses, especially at the upper edge. The pressure of the oxygen jet blows away the oxide fluid.
It should be pointed out that the cutting is not done by melting, although it appears that way. The
process depends entirely on the combustion (that is, burning) of the steel in the path of the oxygen jet.
On mild and normal welding quality steels, the process has no detrimental effect on the metal and
there is no need to machine the cut surface before welding.
Smoothness of the cut edge is an important feature and this depends on the proper tip size, tip to work
distance, oxygen pressure, and on the uniformity of speed with which the torch is moved. The
movement may be made with the torch held in the hand (ie., manual cutting) or it can be mechanically
propelled (machine cutting).

83

Figure 3.21: Use of oxygen cutting for preparing square and bevel edges.
(Note - figures in brackets indicate order of the location of torches in the direction of cutting)

84

Another commonly used method is the air carbon arc gouging which is mainly used to make J- or Ugrooves. J- or U-grooves can also be made by machining, which is much more costly than air carbon
arc gouging. Figure 3.22 shows how the joint is prepared.
Compressed air carbon arc, as the name implies, consists of melting the metal to be gouged or cut
with an electric arc and blowing away the molten metal with a high-velocity jet of compressed air
parallel to the electrode. Because it does not depend on oxidation, it works on metals which do not
oxidize readily. The equipment used is a torch that directs a stream of air along the electrode and
external to it. The torch is connected to an arc welding machine and an ordinary compressed-air line
delivering approximately 100 lbs per sq. inch. Since the exact pressure is not critical, normally no
regulator is necessary. The electrode used is a composition of carbon and graphite and is usually
copper clad to increase its life and provide a uniform groove, as well as to reduce radiation heat. The
shape of the electrode may be round or half round. DCRP is used for most applications, but in some
materials DCSP is preferred. An electrode for alternating current is also available and this, when used
with either AC or DCSP, gives improved results on certain applications.

Figure 3.22(a): Manual air carbon arc torch.

Figure 3.22(b): Principle of air

carbon arc process.

85

Figure 3.23: Automatic arc-air gouging machine.

There are also mechanical methods for joint preparation. For square edges, saw cut may be used.
For bevel edges, specially designed edge bevellers are available. They can be mounted and selfpropelled or a portable manual type can also be used as shown in Figure 3.24.

Figure 3.24(b): Portable beveller.

Figure 3.24(a): Rotary shear.

(Photo courtesy of Gullco International)

86

3.8

Fundamental Concepts of Welding Symbols

3.8.1 Weld Symbols, Supplementary Symbols, Welding Symbols

Definitions
In welding symbols terminology, there are several standard terms in common use. A clear
understanding of these terms is very important to have any meaningful dialogue involving welding
symbols.
These terms are:

a)
b)
c)

weld symbols
supplementary symbols
welding symbols

The definition of these terms and their interrelationship are described as follows:
a)

Weld symbol is a term used explicitly to designate a specific type of weld. The pertinent types
of welds considered under the governing AWS A2.4 specification for Symbols for Welding,
Brazing and Nondestructive Examination and the basic weld symbols are shown in Figure
3.25. Weld symbols such as these form an integral part of any typical welding symbol.

GROOVE WELDS
Square

Fillet

Scarf*

Plug
or
Slot

Stud

Bevel

Spot
or
Projection

Seam

Back
or
Backing

Flange
Surfacing

*Used predominantly in brazed joints - see section on Brazing.

Figure 3.25: Basic weld symbols.

87

FlareBevel

Flare-V

Edge

Corner

b)

Supplementary symbol, as the term indicates, is used to provide complementary information to

that given by the basic elements of a typical welding symbol.
Supplementary symbols are always used in conjunction with a welding symbol, and they are
shown in Figure 3.26.

Weld all
around

Field Weld

Backing
Consumable
or
Melt
Insert
Spacer
Through
(Square) (Rectangle)

Contour
Flush
or
Flat

Convex

Concave

Figure 3.26: Supplementary symbols.

c)

Welding symbol, in turn, provides comprehensive information with respect to the geometry of
preparation, fit-up and welding of joints. It is composed of a number of standard elements,
including a weld symbol, and uses any of the applicable supplementary symbols to effectively
complement such information.
All the basic elements of a typical welding symbol, including reference to supplementary
symbols and their respective designated locations, are shown in Figure 3.27.

3.9

Basic Weld Symbols

Reference was already made to the primary purpose of a weld symbol. Its main and specific objective
is to graphically identify each type of weld.
To assist in this identification, the shape of the symbol, whenever possible, is made to conveniently
reflect the relative configuration of the fusion faces as represented by a vertical section through the
joint.

88

89

Figure 3.27: Standard location of elements of a welding symbol.

The following symbols (Figure 3.28) are some examples, with the viewing position indicated for correct
placement:

Single- or DoubleBevel-Grooves

Single- or DoubleJ-Grooves

Single- or DoubleFlare-Bevel-Grooves

Single- or DoubleFillet Welds

Viewing Position of the Reader

Figure 3.28: Correct placement of weld symbols.

Hence, all weld symbols as shown in Figure 3.29, viewed from the same position as in Figure 3.28,
are incorrect.

Figure 3.29: Incorrect placement of weld symbols.

This rule points to the fact that proper attention must be given to the placement of a weld symbol on
the reference line.
Several practical examples are given in the following pages to illustrate the application of basic weld
symbols.
90

EXAMPLE 1
GROOVE WELD IN A BUTT JOINT

Required:

A Single-V-Groove Weld

ALTERNATIVE 1:
1
Arrow Side

Other Side

t
Other Side

Arrow Side

Joint
WELD

SYMBOL 1

SYMBOL 3

Preparation from the same side

could have been obtained
using Symbol 3 as shown

indicates preparation to be made

from the Arrow Side

ALTERNATIVE 2:
1
Arrow Side

Other Side

t
Other Side

Arrow Side

Joint
4
SYMBOL 2

WELD

SYMBOL 4
Preparation from the same side
could have been obtained
using Symbol 4 as shown

indicates preparation
from the Other Side

91

EXAMPLE 2
CRUCIFORM
DOUBLE-T
JOINTS

TYPE 2
Member M1

TYPE 1
Member M1

Member M2

Member M2
Member M3

Member M3

Required Fillet Welds

M1

TYPE 1: - Member M1 is to be
welded to member M2;
TYPE 2: - Member M1 is to be
welded to member M3;
the common areas of contact and,
therefore, the joint must be as
shown.

TYPE 1

TYPE 1

TYPE 2
M1

Joint for
Members 1 and 2

Joint for
Members 2 and 3

Fillet on the
Arrow Side
of Joint

M2

M3

M2

Joint for
Members 1 and 2

TYPE 2

Joint for
Members 1 and 3

M3

Fillet on the
Arrow Side
of Joint

Fillet on the
Other Side
of Joint

Fillet on the
Other Side
of Joint
Each pair of symbols calls correctly for the same weld.

92

EXAMPLE 3
CRUCIFORM JOINTS SIMILAR TO THOSE IN EXAMPLE 2
BUT WITH FILLET SIZES SPECIFIED FOR EACH CORNER

TYPE 1

TYPE 2

3/8 Fillet

5/16 Fillet

3/8 Fillet

5/16 Fillet

1/4 Fillet

5/16 Fillet

1/4 Fillet

5/16 Fillet

Desired Welds
showing identical arrangement for each type

The correct double fillet welding symbols for each type, respectively, will be:

TYPE 1

TYPE 2
or

or

5/16

3/8

1/4

5/16

3/8

5/16

3/8

5/16

or

or

5/16

1/4

3/8

5/16

1/4

5/16

1/4

5/16

or

or
Appropriate Symbols
for the Desired Welds

Either metric or imperial measurement may be applied

93

In this
case
identical

EXAMPLE 4

t
GROOVE WELD IN A T-JOINT

Required a combined
Bevel- and J-Groove Weld
(A hypothetical requirement
considered only for the
purpose of the exercise)

Assuming half a thickness preparation for each type of groove, the alternatives are:
ALTERNATIVE 1:

Other Side

Arrow Side

Arrow Side

Joint

Other Side
Joint

SYMBOL 1

WELD

SYMBOL 2

indicates bevel preparation from

Arrow Side and I preparation
for the Other Side

Identical preparation of the joint

can be obtained using Symbol 2

ALTERNATIVE 2:

Other Side

Arrow Side

Arrow Side

Joint
SYMBOL 3

Other Side
Joint

WELD

indicates J preparation from the

Arrow Side and Bevel preparation
for the Other Side

SYMBOL 4
Identical preparation of the joint
can be obtained using Symbol 4

94

3.10

Supplementary Weld Symbols

As seen from Figure 3.26, supplementary weld symbols are used to convey specific requirements.
One can again observe that their graphic designations bear a very close visual likeness to the effect
they mean to achieve.
3.10.1

Field Weld Symbol

Welding in the field is generally understood to mean welding in a place other than that of initial
construction. The erection phase of welded construction work will most likely involve welding in the
field, or on site, as some may refer to it.
The weld symbol designating welding in the field must show a flag placed above and at right angle to
the reference line at the junction with the arrow. The direction of the arrow is left optional. However,
some may prefer to point it away from the arrow.

or

Figure 3.30

3.10.2

Melt-thru Symbol

Occasionally, there are conditions in a joint in which full or complete joint penetration is required that
permit welding only from one side. As a visible manifestation of such penetration, a reinforcement may
be specified for the other side.
This reinforcement may be conveniently expressed by a melt-thru symbol, together with its required
height, or without it when a specific height is of no significance.

3.10.3 Contour Symbol and Finishing of Welds

Sometimes it is required to specify the contour of the weld surface. It is important for certain structures
or mechanical components to minimize stress concentrations. When mechanical means are intended
to obtain the desired contour, the supplementary contour symbols should be used with the users
preferred mechanical means specified to obtain it.

95

The required contour of welds may be obtained without recourse to mechanical means. However, if
such means are necessary to produce the required finish, the appropriate letter designation assigned
to the following methods of finishing must be added to the contour symbol. It must be understood that
these designations specify the method and not the degree of finish.
C Chipping
M Machining
H Hammering

G Grinding
R Rolling

The following sketches will illustrate these provisions:

EXAMPLE 5

1/8

or

1/8
1/8

Symbol

Desired Weld

EXAMPLE 6

or

Not specified
Reinforcement to be removed flush
by subsequent Machining {M}

M
Symbol

Desired Weld

96

EXAMPLE 7

or
1 mm
Convex Reinforcement of 1mm to
be provided by Grinding {G}

1
Symbol

Desired Weld

3.10.4

All-Around Weld Symbol

The all-around supplementary symbol must be shown in the welding symbol when a circumferential
weld is required and abrupt changes in the direction of welding are involved. This procedure is well
illustrated in the case of a hollow structural section (HSS), rectangular in profile and welded to a base
plate.

Desired Weld

Required Welding Symbol

Figure 3.31

97

EXAMPLE 8

3.11

Break in Arrow

In most cases of groove welds, both members to be joined require some form of preparation. Hence,
the use of an ordinary welding symbol with an ordinary straight-line arrow is entirely satisfactory.
However, such is not the case with groove welds requiring the preparation of only one member in the
joint. If the preparation of one specific member is of importance, the welding symbol must have
positive means to identify this member.
This is conveniently done by a break in the arrow, with the arrow pointing in the direction of the
member to be chamfered.
It should be noted that the arrow need not touch this particular member. The only matter of importance
is the direction in which the arrow points (to the left or to the right).
The arrow need not be broken if the welding symbol is not used to specify which members have to be
prepared. There are, of course, situations in which only one member can be prepared, in which case a
break in the arrow is superfluous.
The following examples will illustrate the point.

98

EXAMPLE 9
DOUBLE-BEVEL-GROOVE WELD IN A BUTT JOINT
WITH THE CHAMFERED MEMBER SPECIFIED

or

Desired Weld

Symbol

Preparation specified on the

right-side member as shown

The arrows point to the right

member to be chamfered

EXAMPLE 10
DOUBLE-J-GROOVE IN A BUTT JOINT WITH THE MEMBER
TO BE CHAMFERED NOT SPECIFIED
4

Same

or

or in this
case

2
Symbols

Desired Weld
Either of the two are acceptable

Pointing the arrow to the left or to the right member as shown

has no significance with regard as to which member is to be
chamfered because they are straight-line arrows without a break.

99

EXAMPLE 11
SINGLE- BEVEL-GROOVE AND A FILLET WELD IN A T-JOINT

or

Symbol
Arrow with a Break

Desired Weld

or

Symbol
Arrow without a Break is acceptable
as only one (the intended) member
can be prepared

[ As a matter of principal, it is recommended to use the arrow with a break ]

3.12

Combined Weld Symbols

Normally, joints will require more than one type of weld symbol. Joints for which one type of weld
symbol is sufficient are represented by welding symbols in Figure 3.32.

Figure 3.32

100

However, joints that require a combination of two or three different types of weld symbols are illustrated
by the following welding symbols (Figure 3.33):

Single-V and Back Weld

symbols

Fillet and Single-Bevel

symbols

Two Fillet and two Single-Bevel

symbols

Figure 3.33

Each of the welds represented by the pertinent weld symbol must appear in the welding symbol, either
in a single or in a combined arrangement.

EXAMPLE 12
A SIMPLE WELD SYMBOL COMBINATION

or

Symbol

Desired Weld

A combination of a V and
Back Weld symbols

101

EXAMPLE 13
A MULTIPLE WELD SYMBOL COMBINATION

or
or

A
Desired Weld

Combined Symbols
A) on Single Reference Line

3.13

B) on Multiple Reference Line

(to specify sequence of welding)
[see next heading]

Information in Tail of Welding Symbol

The tail of the welding symbol allows convenient placement of any type of information that will
effectively complement the information conveyed by the other components of the symbol.
The conventions as used by individual companies and geared to their specific operations will have a
great bearing on the type of information that each of them will consider sufficient in scope for their
needs.
Such information may:
1)
2)
3)

refer to a specific welding process, or

consist of a welding procedure specification number, or
show typical when the required weld is representative of all welds shown in the drawing

102

3.14

Extent of Welding Denoted by Symbols

When the length of a weld is not specified in the symbol, the welded length is the one between abrupt
changes in the direction of welding.
The length of weld may also be designated by a dimensional length of hatching.
As its name implies, the all-around welding symbol specifies the weld all around the joint, regardless of
the number of planes involved.
The section on fillet welds reviews all of the above variations to the extent of welding, as covered by a
welding symbol with or without supporting dimensioning.
When intermittent welds are required in the length of a joint, they should be dimensioned by the
welding symbol in a manner as shown and thoroughly discussed in the section on fillet welds.
Groove welds are normally continuous for the full length of the joint, in which case no reference to
length is needed in the welding symbol.

3.15

Multiple Reference Lines

The very objective of the multiple reference line concept is to give the welding symbol the added
capability to specify the sequence of welding operations as well as to provide additional information
relative to the examination of welds or other operations.
Two or more reference lines are feasible. They should, however, be used with good judgment.
The rule that applies to sequencing is very clear. It states that the operation that is desired first is to be
shown on the reference line closest to the arrow. All subsequent operations are to follow the same
sequencing order as the pertinent reference lines move away from the arrow. The welding symbol
shown in Figure 3.34 explains this principle:

103

3rd Operation

1st Operation

2nd Operation

2nd Operation

1st Operation

3rd Operation

MT = Magnetic Particle
Examination method

MT

Sequence:

1) weld single-V on arrow side,

2) weld back-pass on other side,
3) inspect other side, using MT method.

Supplementary symbols may also be used as applicable.

Figure 3.34

RT

RT = Radiographic
Examination method

Backgouge*
*The CSA Standard W59
preferred term is GTSM
= Gouge To Sound Metal
Sequence:

1)
2)
3)
4)

weld the all-around arrow-side bevel-groove in the field;

back-gouge from the other side (in the field, obviously);
complete the all-around other side bevel-groove weld in the field
inspect, in the field, the whole weld using the radiographic examination method.

Figure 3.35
104

3.16

Complete Penetration

Where complete joint penetration (that is penetration equal to the thickness of material) is to be
obtained with no regard to the type of weld and joint preparation, the letters CJP must be shown in the
tail of the welding symbol.

CJP

Figure 3.36
The definition of complete joint penetration groove welds may vary from one governing design standard
to another. Important observations on this subject are offered in the section on groove welds.
When a sequence is to be specified for a joint defined by the following welding symbol...

Figure 3.37

then these sequencing alternatives may be considered:

(a)

(b)
or

Figure 3.38

105

The specified sequence for each case is shown as follows:

Sequence (a)
1)
2)
3)

Sequence (b)

weld bevel-groove
weld flat filler
weld back pass

3
1

arrow side
arrow side
other side

weld backing pass

weld bevel groove
weld flat fillet

Desired Weld and Sequence

other side
arrow side
arrow side

Desired Weld and Sequence

Figure 3.39
Weld examination requirements may also be placed on the second or third reference line as,
obviously, the first or the first two lines must provide a weld that can then be examined.

106

3.17

Groove Welds
3.17.1

Location of Dimensions for Single-Groove Welds

Although Figure 3.27 shows the location of all the standard elements for any type of weld, including a
groove weld, it will be of advantage to extract from that figure all those elements that specifically apply
to grooves.
Since there are many types of grooves, with each assigned its own weld symbol, as evident from
Figure 3.25, specific and comprehensive information must be provided in the welding symbol to
accurately describe the required preparation and fit-up of the two members in the joint.
Besides the applicable weld symbol for the pertinent type of groove, the additional information will have
to define:
1)

The depth of preparation also described as the depth of chamfer from each side of the joint
(arrow side and other side), and normally designated by the capital letter S.

2)

The angle at which such preparation should be made, also referred to as angle of chamfer, but
officially termed the bevel angle.

3)

The root opening required for proper fitting of the two members in the joint. Its primary
objective is to provide adequate access for welding. It may be used with other related factors in
the joint for greater welding economy.

The location of these elements in the welding symbol for single grooves is shown in Figure 3.40.

1/ 2

1/ 8
60

Depth of
Preparation
(S)

Groove Angle

Root Opening

20
18

Figure 3.40

107

3.17.2

Dimensions for Double-Groove Welds

In line with the general principle, and irrespective of its appearance on one or both sides of the
reference line, each weld symbol must be accompanied in the welding symbol with all the data
necessary for the preparation, fit-up and execution of welding.
The principle applies to all double-groove welds, and it makes no difference whether the data is
identical on each side of the reference line or not. However, the size of the root opening, being
common to both sides, need only appear once.

55
1/8

30
Imperial Units

Metric Units

7/8
7/8

25
25

1/8
30
60
3
60

1/2
7/8
45

16
12

30
3
25

Figure 3.41

The angle with respect to application in the symbol is the groove angle which, for bevel- and Jgrooves, will also be the bevel angle. However, it is the groove angle (the angle contained between
the fusion faces of the groove) that appears in the welding symbol.
The angle at the root of the joint associated exclusively with J- and U-grooves has strong design
implications, as the governing specifications make the applicable effective throat of welds in such
grooves a function of this particular angle.
In conjunction with this angle, it should also be noted that the minimum groove radius for the J- and Ugrooves is given in CSA W59 and AWS D1.1 for prequalified joints.

108

Figure 3.42

There are three basic angles in the joint (Figure 3.42), and they are part of every groove except
the square groove.
1)

BEVEL ANGLE
(official AWS A3.0 term
or
Angle of Preparation
or
Angle of Chamfer

2)

GROOVE ANGLE

3)

ANGLE AT THE
ROOT OF THE GROOVE

For the U-groove, all three angles have different values with 2 being twice that of 1.
For the bevel-groove, however, all the angles have the same value: 1 = 2 = 3.

109

3.17.3

Depth of Preparation and Groove Weld Size

The AWS A2.4 specification has introduced the groove weld size (E) as a quantity to be shown in the
welding symbol for groove welds.
The specification ruled at the same time that this new quantity should:
1)
2)

appear in brackets: (E), and

be located on the right side of the depth of preparation S and so share with it a common
location in the welding symbol

S (E)
S (E)

S1 (E1)
S2 (E2)

Figure 3.43

The term throat of a groove weld, while previously an acceptable term, is now considered a nonstandard term for groove welds size.
Unfortunately, the weld size as defined in the AWS A2.4 specification purely from the point of view of
logical symbol application, need not necessarily be the same and in many cases it is not as the
weld size defined in the governing standards for design application.
Therefore, caution must be exercised and appropriate distinction made when referencing weld sizes
under the jurisdiction of one specification as compared with another.
While the application of the groove weld size concept is quite straightforward and offers a clear
advantage in a number of applications, this advantage is less visible and even confusing in others. A
more elaborate discussion on the subject will follow later.
First, lets explore what is meant by groove weld size according to AWS A2.4 specification, and how
that size relates to depth of preparation. Both are independent quantities. However, the magnitude of
(E) relates strongly to the root geometry of the joint, the welding process and the parameters of the
welding procedure (see Figure 3.44).

110

S(E)

or

S(E)
Desired Weld

Symbol

70
1/2 ( 3/4 )
60

0
60

or

3/4

1/2 5/8

7/8

7/8 ( 1 )

0
7/8 ( 1 )

70

1/2 ( 3/4 )
Desired Weld

60
0
70

Symbol

Figure 3.44

The quantity (E) is measured from the top of the plate to the furthest point where the weld
penetrates the joint. The AWS A3.0 specification defines this as joint penetration.
The value of (E) may be greater than S as shown in Figure 3.44. However, it may also be
smaller than S as shown in Figure 3.45.

111

3/4 ( 5/8 ) 0
50

3/4

5/8

1 1/2

50

or

0
3/4 ( 5/8 )

Desired Weld

Symbol

Figure 3.45

3.17.4

Flare-Groove Welds

There are two basic types of flare-groove

welds:
As shown in Figure 3.46, the weld symbols for
these grooves also reflect the shape of the joint
that contains them.

Figure 3.46

112

50
0

EXAMPLE 14
FLARE-BEVEL-GROOVE WELDS

(a) Single-Flare-Bevel-Groove Weld

Plane of Joint
3/4 (1/4)

1/4
3/4
or

3/4 (1/4)

Desired Weld
Symbol

(b) Double-Flare-Bevel-Groove Weld

18 (6)
18 (8)

Other Side

18

Arrow Side
Plane of Joint

Symbol

8
Desired Weld

(c) Single-Flare-Bevel-Groove with a reinforcing Fillet Weld

3/16

E=

S=

1/2

Plane of Joint

3/16

1/2 (3/16)

Desired Weld

Symbol

3/16

113

3/16

EXAMPLE 15
FLARE-V-GROOVE WELDS
R=

25

10
Horizontal Lap Splice of two round Bars of the same size

25 (10)
25 (10)
Arrow Side

Other Side

Plane of Joint

10
Desired weld

3.17.5

Symbol

Surface Finish and Contour of Groove Welds

The desired contour of groove welds may be obtained naturally or with recourse to mechanical means,
these being left to the discretion of the user.
The series of welding symbols in Figure 3.47 shows the correct application of the required
supplementary symbols.

Flush

Flush

Figure 3.47

114

Convex

The most common contour for groove welds is the flush contour, and this is in view of its expectedly
better performance in service. To further improve that performance, a mechanical finish may be
considered and indicated in Figure 3.48.
G

Chipping

M
Machining

Grinding

Rolling

Chipping / Rolling

Figure 3.48

3.17.6

Joints with Backing

Joints with backing are joints welded from one side. Such welds must be considered where there is no
access for welding from the other side or where, if such access is feasible, the more skill-demanding
and less-productive welding in overhead position, or welder discomfort, will dictate their use.
The welding symbol designating a groove with backing shows the groove and the supplementary
backing symbols (Figure 3.49).

M
S(E)

Figure 3.49

It should be noted that the supplementary symbol for backing is a rectangle while that for consumable
inserts is square in shape.
1.

The type of backing material must be identified, and such identification (by means of an
assigned letter) must appear inside the rectangle of the supplementary backing symbol.

M
Designating a material in general terms

115

2.

The direction to remove the backing, if required, must also be placed in the rectangle, using the
letter R for removal.

MR
3.

The dimensions of the backing must be specified in the tail of the symbol or elsewhere in the
drawing.

Many materials are successfully used as backing materials. As stated, they must be identified in the
symbol. For the purpose of explanatory examples, only steel (S) will be used.
For joints in which there is no fusion into the backing material, the removal of such material need not
be specified in the symbol.
Joints with backing will be joints with complete penetration.

EXAMPLE 16
S

JOINTS WITH BACKING

3/4 (3/4)

45

1/4
45
or

3/4
3/8
45
3/4 (3/4) 1/4
S

1/4
Desired Weld

1 X 3/8

Symbol

SR
40 (40)

30

10
30
or

40
10
40 (40)
10

30
10
SR

Symbol

Desired Weld
(Backing to be removed)

116

25 X 10mm

3.18

Fillet Welds

Fillet welds are the most commonly used type of weld in welding fabrication. It does not require
special joint preparation, like bevel cutting. A fillet weld joins two surfaces, usually, but not always, at
right angles to each other. Fillet welds are used to make lap joints, T-joints or corner joints. The
profiles of fillet welds and the associated terms are shown in Figures 3.50 and 3.51. These are equal
leg fillet welds.

Figure 3.50: Convex Fillet Weld

117

Figure 3.51: Concave Fillet Weld

In Figures 3.50 and 3.51, all the terms are self explanatory. The term Effective Throat is the shortest
distance measured from the root of the weld to its face, less any reinforcement. Also, it should be
noted that the root penetration is only considered as part of the effective throat for fillet welds made by
the submerged arc welding process. This is stated in the CSA W59 Standard (Clause 4.3.2.4). In
some standards or codes, the root penetration is not considered.

3.18.1

Symbols of Fillet Welds

The composition of welding symbols for fillet welds is governed by a number of explicit rules. For
proper application of such welds in welded fabrication these rules require that the following information
be shown at designated locations in the welding symbol unless specific general notes covering
standard dimensions of fillet welds appear elsewhere on the drawing. Also, see Examples 2 and 3
given on pages 3-28 and 3-29 for the correct ways of placing symbols to a cruciform joint.

118

1.

The Fillet Weld Symbol

Rule: The vertical side of the triangle representing the weld symbol must always be on the left
side from the readers viewing position as shown in Figure 3.52.

Arrow Side

Other Side

Both Sides

Readers Viewing Position

Figure 3.52

2.

Location of Fillet Weld Size

Rule: The size must be shown for each weld symbol and must always appear to the left of each
weld symbol as shown in Figure 3.53.

SIZE

SIZE
SIZE
Arrow Side

SIZE
Other Side

Readers Viewing Position

Figure 3.53

119

Both Sides

3)

The Length of the Fillet Weld

Rule: The length must be shown for each weld symbol and must always appear on the right
side of each weld symbol as shown in Figure 3.54. Absence of a specified length designates a
length defined by the side of the joint between two points of abrupt change.

LENGTH
LENGTH
Arrow Side

Other Side

LENGTH
LENGTH
Both Sides

Figure 3.54

Depending on the specific conditions or requirements for a given application, additional use may be
made of the following supplementary symbols:
1.

Weld All Around

Symbol represented by a circle placed at
the junction of arrow line and reference
line (Figure 3.55).

2.

Field Weld
Symbol represented by a flag with its
direction optional but preferably pointing
away from the arrow and placed at the
junction of arrow line and reference line
(Figure 3.55).

120

Figure 3.55

3.

Contours
Contours may be obtained in either one or two ways:
a)

with no application of mechanical means (Figure 3.56)

Concave

Convex

Flat

Figure 3.56

b)

with application of mechanical means (Figure 3.57)

M
M

C
Concave
by Chipping

Convex
by Grinding

Flat
by Machining

Figure 3.57

4.

References
The designated location for reference is the tail of the welding symbol (Figure 3.58).

Reference to
Submerged Arc
Welding Process

SAW

WG3

Figure 3.58

121

Reference to
Specific Welding Procedure
Specification
or Welding Procedure
Data Sheet

3.18.2

Size of Fillet Welds Equal Leg Fillet Welds

The strength of a fillet weld is governed by both the fillet size and the effective throat thickness. The
fillet size is the length of the side of the largest triangle that can be inscribed within the weld crosssection as shown in Figures 3.50 and 3.51. For equal leg convex fillet welds, the measured leg size is
the fillet size as shown in Figure 3.50. For equal leg concave fillet welds, the fillet size is the side of
the inscribed triangle, or the theoretical effective throat multiplied by 1.4 as shown in Figure 3.51.
1)

The Specified Size is the size as it appears in the welding symbol and is designated by
the letter S.

2)

The Effective Size is the size that corresponds to the specified size and is designated by
the expression S effective = Seff.

3)

The Measured Size is the size established on the basis of measurement and is
designated by the term S measured = Sm.

These sizes for the following three types of fillet welds are as shown in Figure 3.59.

Sm
Seff

S = Seff = am

Concave Fillet

Ef
Th fec
ro tive
at

M
e
Th asu
ro re
at d

Ef
Th fec
ro tive
at
Flat Fillet

Sm
am

S = Seff = Sm

Seff

Sm

Sm

S = Sm = Seff
Convex Fillet

Figure 3.59: Size of fillet welds (equal legs).

122

Figure 3.60 shows a few examples of fillet weld sizes and symbols. The fillet size specified in the
design must be the effective size.

3/8

3/8
3/8

or

3/8

Desired Welds

Symbols
TEE (T-) JOINT

Figure 3.60: Fillet weld sizes and symbols.

123

It is of interest to note that in some countries the size of the fillet as it appears in the welding symbol
may be specifying the size of the throat of the weld rather than the size of the leg (Figure 3.61).
The International Standards Organization (ISO) responsible for the formulation of international
standards has recently established its own position on this issue. Sub-committee 7 Graphical Welding
Symbols of its Technical Committee 44 on Welding (ISO/TC44/SC7) has, in recognition of the
entrenched practices in using one or the other system, officially accepted both, leaving it at the
discretion of each country adopting the ISO Standard to opt for the one system it prefers.
However, as a necessary precondition of such compromise, ISO has made it a mandatory requirement
that each system be clearly identified by the following means.
It should be understood that the designations z and a have no other significance except to identify
the system. It is also very important for the student to be aware of the difference in interpretation
attached to the definition of sizes of fillet welds between the AWS and the ISO concepts, both of which
are used internationally. This awareness will be of specific importance to those who, because of
involvement with international contracts, are dealing with foreign drawings. In North America, fillet
welds are specified by leg dimensions.

(a)

Letter z to precede size

to designate LEG size
for the desired weld having

Letter a to precede size

to designate THROAT size
for the desired weld having

(b)

Leg Size = S = 10mm

Throat Size = 10mm

s= 10

t)

oa

r
Th

10

a=
* See preceding figure for

s= 10

effective throats of fillets

with different profiles.

or

or

z10

a10
z10

Figure 3.61: ISO fillet size

designation.

a10

The interrelation of both sizes is expressed by:

z=a

124

or

a=

z
2

3.18.3

Minimum and Maximum Fillet Size

It is advisable to point out that some governing design applications like CSA W59 (Welded Steel
Construction) or AWS D1.1 (Structural Welding Code) stipulate minimum fillet sizes as a function of
material thickness.
On the other hand, maximum fillet sizes may be set either by considerations of balanced design, that
is, by keeping the capacity of the welds reasonably close to that of the parent metal, or by
requirements of good welding practice.
With regard to the latter, specific reference is made to welding against a cut edge where the maximum
recommended fillet size for thicknesses over inch (6mm) is:

1/16
t

(Imperial)

where: S = t - 1/16
or:

(Metric)

S=t-2

Figure 3.62

3.18.4

Conventional Fillet Sizes

The fillet sizes are usually measured in millimeters (mm) in the metric system or in inches (in) in the
imperial system. The smallest dimension is adjusted to the nearest size in mm or in inches, 1/16 of an
inch intervals. The common sizes used are shown as follows:
Metric (mm)

10

12

14

16

18

Imperial (1/16 in)

1/8

3/16

1/4

5/16

3/8

1/2

9/16

5/8

11/16 3/4

125

20

3.18.5

Size of Fillet Welds Unequal Leg Fillet Welds

For unequal leg convex fillet welds, the effective throat (t), as shown in Figure 3.63, is the shorter leg
(a) multiplied by sin .

Figure 3.63: Unequal leg fillet (convex).

For unequal leg concave fillet welds, the effective throat thickness must be obtained by direct
measurement which is the shortest distance from the root point to the weld surface. Its equivalent leg
sizes are the lengths of the sides of the inscribed triangle as shown in Figure 3.64.

Figure 3.64: Unequal leg fillet (concave).

The AWS Standard A2.4 does not define the convention of which leg size is specified first. Where it is
possible to misinterpret the fillet leg sizes, a sketch defining the leg sizes must be shown on the
drawing. In recognition of this shortcoming, AWS A2.4 requires that a sketch of the joint complete with
the desired orientation of the fillet weld be shown on the drawing whenever necessary. Figure 3.65
will illustrate this point.

126

(a)

1/4 x 3/8

The Fillet Weld specified in T-joint (a):

could be interpreted to mean:

(b)

(c)
or
3/8

1/4

1/4

3/8

Figure 3.65: Unequal leg fillet size and symbol.

Consequently, in addition to the basic symbol under (a), an additional sketch showing either fillet weld
(b) or (c) as applicable must be shown on the drawing. Only then will complete information be
provided for an error-free interpretation at the time of welding. Such clarifying sketches may not
always be necessary, although they may be considered.
The effective throats of unequal leg fillets are discussed previously.

3.18.6

Intermittent Fillet Welds

There are three types of intermittent fillet welds, although the first one is inherent in the remaining two:
1)
2)
3)

basic intermittent fillet welds, applicable to a single line of fillet welds (Example 18)
staggered intermittent fillet welds (Example 19)
chain intermittent fillet welds (Example 20)

127

The following sketches will help explain the fundamental concept and the rules governing its
application.

EXAMPLE 17

EXAMPLE 18
(Staggered)

128

EXAMPLE 19

(Chain)

(Imperial Units)
10

Pitch=
[common centres]

10

C
L

3
L=

C
L

3
6

1/4
1/4

C
L

3
6

6-10
6-10

3
6

Under normal circumstances the general tendency is to keep the length of the increments and the size
of the fillets the same on both sides of the joint, if only to simplify fabrication and avoid errors.

3.19

Plug Welds

Information required for the application of plug welds must include the following:
1)
2)
3)
4)
5)

Size of hole
Angle of countersink
Depth of filling
Spacing of welds
Reference to contour and surface finish, if required

3.19.1 Size of Plug Welds

In line with the common principle, the size of the plug weld is expressed by the diameter of the hole
and must be placed on the left side of the weld symbol as shown in Figure 3.66.

129

DIAMETER
DIAMETER

2
100

Weld in the Arrow-Side Member

Weld in the Other-Side Member

Figure 3.66

It should also be noted that the governing design specification may relate the minimum hole diameter
to the thickness of material in which a plug weld has to be made. CSA W59 and AWS D1.1 set this
minimum at
dmin

= t + 8mm

(CSA W59 [metric])

= t + 5/16 in. (CSA W59 and D1.1 [imperial])

3.19.2 Angle of Countersink

In order to facilitate welding and provide easier access to the root, hence greater assurance of
soundness, countersunk holes with circumferentially sloping sides may be considered. When specified
in the symbol, the angle of countersink must be placed outside the horizontal side of the weld symbol.

3.19.3 Depth of Filling

If complete depth of filling is required, it need not be specified in the welding symbol. However, if the
depth of filling required is less than complete, it must be specified and shown inside the weld symbol.

130

EXAMPLE 20
Imperial Units

3/4

a = 45
1 1/4 5/8
45
For users
standard angle,
a = 30.

5/8

or

1 1/4

Location of centre line

to be dimentioned
on drawing.

Desired Weld

1 1/4 5/8
For users standard
angle, a = 45.

Symbol

EXAMPLE 21
Metric Units

70

70
0
For users
standard angle,
a = 30.

50

or

Location of centre line

to be dimentioned
on drawing.

Desired Weld

70
For users standard
angle, a = 0.

Symbol

131

3.19.4 Spacing of Plug Welds

The concept of pitch is used to designate the centre-to-centre spacing of plug welds. As in the case of
other welds, the assigned standard location for length is at the right side of the weld symbol as shown
in Figure 3.67.

SPACING
SPACING

100
6

Spacing for Welds on

the Arrow-Side member

Spacing for Welds on

the Other-Side member

Figure 3.67

132

Chapter 4
Metal Arc Welding Processes
Table of Contents
4.1

Introduction

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .135

4.2

Shielded Metal Arc Welding (SMAW) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136

4.2.1 Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136
4.2.2 Power Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .137
4.2.3 Types of Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .140
4.2.4 Classification of Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .141
4.2.5 Applications and Limitaions of the SMAW Process . . . . . . . . . . . . . . . . . . . . . . . .145
4.2.6 Shielded Metal Arc Welding of Carbon and Low Alloy Steels . . . . . . . . . . . . . . . . .146

4.3

Gas Metal Arc Welding (GMAW) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .148

4.3.1 Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .148
4.3.2 Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151
4.3.3 Metal Transfer Across the Arc in GMAW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .154
4.3.4 Shielding Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .163
4.3.5 Advantages and Limitations of the GMAW Process . . . . . . . . . . . . . . . . . . . . . . . .167
4.3.6 Applications of Gas Metal Arc Welding Process . . . . . . . . . . . . . . . . . . . . . . . . . . .168
4.3.7 Electrode Wires for Gas Metal Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169

4.4

Flux Cored Arc Welding (FCAW) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .170

4.4.1 Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .170
4.4.2 Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .172
4.4.3 Advantages and Applications of the Cored Wire Process . . . . . . . . . . . . . . . . . . . .173
4.4.4 Classification of Cored Wires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175
4.4.5 Shielding Gases for Flux Cored Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .181

4.5

Submerged Arc Welding (SAW) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .182

4.5.1 Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .182
4.5.2 Current Type and Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183
4.5.3 Advantages and Applications of Submerged Arc Welding . . . . . . . . . . . . . . . . . . .184
4.5.4 Multiple Wire Submerged Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .186
4.5.5 Wires and Fluxes for Submerged Arc Welding of Carbon
and Low Alloy Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187
4.5.6 Submerged Arc Welding of Carbon and Low Alloy Steels . . . . . . . . . . . . . . . . . . .192

133

134

4.1

Introduction

There are quite a few arc welding processes. We are concentrating on those processes that are used
in structural fabrication shops. There are four commonly used welding processes in the structural
fabrication shops, and each has its own intrinsic characteristics. See Table 4.1.

Table 4.1

Process

Abbreviation

Electrode Designation
(Example)

Shielded Metal Arc Welding

SMAW

E49XX

Gas Metal Arc Welding

GMAW

ER49S-X

Flux Cored Arc Welding

FCAW

E49XT-X

Submerged Arc Welding

SAW

E49AX-EXXX

Each welding process requires its own equipment, power source and filler metal (with or without gas
shield or flux). Therefore, the manual skills, set-up and deposition rate are all different. Each process
is suitable for certain types of joints and welding positions. This chapter briefly explains each process
in more general terms. There are several other welding processes that are not discussed here, but are
covered in depth in the following CWB Modules:
Module 4
Module 5
Module 6

Welding Processes and Equipment

Power Sources for Welding
Electrodes and Consumables

The types of welding joints and welding positions mentioned above can be found in Clause 10 of CSA
W59 Welded Steel Construction (Metal Arc Welding). The electrode designations and classifications
can be found in the CSA W48-01 standard, which covers electrodes for various welding processes:
*CSA W48.1
*CSA W48.2
*CSA W48.3
*CSA W48.4
*CSA W48.5
*CSA W48.6

Mild Steel Covered Arc Welding Electrodes

Stainless Steel Electrodes
Low-Alloy Steel Covered Arc Welding Electrodes
Solid Mild Steel Filler Metals for Gas Shielded Arc Welding
Mild Steel Electrodes for Flux Covered Arc Welding
Base Mild Steel Electrodes and Fluxes for Submerged Arc Welding

(* indicates previous designations given here for cross reference.)

135

4.2

Shielded Metal Arc Welding (SMAW)

Shielded metal arc welding (SMAW), or manually operated metal arc welding with covered electrodes,
is one of the most commonly used welding processes. It allows the greatest amount of flexibility in
terms of the range of materials and thicknesses that can be joined in all welding positions.
The SMAW process is the first one developed during the experimental stage of arc welding in the early
1920s. The knowledge of joint design, arc action, heat control and metal reaction gained from shielded
metal arc welding has been of great value in developing all other variations of the arc welding process.

4.2.1 Principles of Operation

In shielded metal arc welding, an arc is established between the end of a covered metal electrode and
the workpiece to be welded. The heat of the arc melts the surfaces of the joint as well as the metal
electrode. The filler metal is carried across the arc into the weld joint and mixes with the molten base
metal. As the arc is moved at a suitable travel speed along the joint, the progressive melting of the
metal electrode and the base metal provides a moving pool of molten metal, which cools and solidifies
behind the arc (Figure 4.1).

Electrode Coating
Electrode Wire
Arc

(Straight Steel Rod)

Molten Metal

Protective Gas From

Electrode Coating

Slag
Metal Droplets
Solidified
Metal
Weld
Base

Metal

Figure 4.1: Schematic sketch of the shielded metal arc process.

136

The electrical circuit for shielded metal arc welding is relatively simple and is shown in Figure 4.2. It
comprises a power source with electrical leads connected to the workpiece and the electrode holder.
The arc characteristics, weld bead shape, and weld metal soundness and properties depend on the
selection of the type of power source, electrode, joint design, as well as welding parameters and
welder skill.

Electrode Holder

Power Source
Electrode Lead

On
Off
4 5 6
3
7
2
8
1
9
0 10

120

150

180

90

210

60

240
270

30

Electrode

300

Base Metal

Work Lead

Figure 4.2: Electrical circuit for SMAW.

4.2.2 Power Sources

The power source used for SMAW is a constant current type, i.e., it has a drooping volt-ampere curve
(see Figure 4.4). With such a power source, the welder sets the required current at the power source
and the voltage is controlled by the arc length that the welder maintains during welding. The drooping
power source is preferred because there is a small but continual variation in the arc length due to the
manual nature of the welding process. This is reflected in a continual change in the arc voltage, but
due to the drooping characteristics of the volt-ampere curve the accompanying changes in the arc
current, and therefore the electrode melt off and deposition rates, are small. Figure 4.3 shows typical
SMAW power sources.

137

Different power sources, however,

may have different slope (or incline)
for the volt/ampere curve, and
some machines are designed to
enable some adjustment of the
slope. Figure 4.4 shows that when
the volt-ampere curve is flatter,
there is a greater change in current
for a given change in voltage. The
adjustment of the slope of the
volt/ampere curve enables the
welder to maintain better control of
the weld pool and penetration in
certain situations such as out of
position welding (vertical or
overhead positions) or depositing a
root pass in a pipe over a varying
gap. For example, by adjusting the
volt/ampere curve to be flatter an
intentional increase in the arc
length by a welder pulling the
electrode away increases the arc
voltage and decreases the current
sufficiently to reduce penetration or
risk of burn-through.
Conversely, electrode sticking is
also prevented when the rod is in
near contact with the base metal
and the arc length, and therefore
voltage is reduced, causing the
current to increase sufficiently to
Figure 4.3: Typical SMAW power sources.
increase the burn off rate to
prevent sticking. With a pure
drooping (or vertical) volt/ampere
curve, there would be no change in the current due to change in voltage or arc length. The welder
would have no control over the electrode burn-off rate in this case.

138

100
Note: Lower slope gives a greater
change in welding current for a
given change in arc voltage.

Maximum OCV

Voltage
50

32
27
22

125 A
27 V

Minimum OCV

Long Arc
Normal Arc Length
Short Arc

Current, A

Arc
Voltage

100

200
15
40

Figure 4.4: Change in current due to change in voltage in a constant current power source.

Power sources are available to provide direct current (DC), alternating current (AC), or both. A
transformer or an alternator type of source is used for AC welding, and transformer/rectifier or motor
generator type for DC welding. Some power sources (single phase transformer/rectifier or alternator
rectifier type) can be used for AC and DC welding. Inverters are becoming popular due to their
portability and smooth operating characteristics. Figure 4.5 shows a typical inverter power source.

139

Figure 4.5: Portable inverter type constant current power source.

4.2.3

Types of Electrodes

Electrodes for shielded metal arc welding generally comprise a coated, solid electrode wire (core) of
limited length (300 to 400 mm or 12 to 16 in long). Occasionally, the solid electrode can be replaced by
a metallic sheath containing metal powders with the objective of adding specific alloying elements to
the weld metal.
The covering on the electrodes can be applied either by an extrusion process or by dipping, though
extrusion is far more common. The covering itself contains several ingredients depending on the type
of electrode. The function of these ingredients is generally one of the following:
g provide a gas shield to prevent contamination of the weld metal by atmospheric gases
g provide a slag cover to protect the hot weld metal from atmospheric contamination and
control the bead shape
g scavenge some of the impurities in the weld metal
g stabilize the arc by promoting electrical conduction across the arc; this is especially important
for AC welding where the arc effectively goes out and needs to be re-established after each
current reversal
g provide a means to add alloying elements to enhance mechanical/corrosion properties, and iron
powder to increase deposition rate

140

Each electrode classification produces different amounts of gases and slag to shield the weld metal.
Electrodes that rely on slag to protect the metal can carry higher current and provide a higher
deposition rate. Conversely, electrodes producing a smaller amount of slag and relying on the gas
shield are stable to operate at lower currents and therefore are more suitable for out of position
welding.
All electrodes can be used with direct current, although some are designed for use with AC also. Use
of AC reduces arc blow and voltage drop in welding cables.
Direct current (DC) power has certain advantages:
g
g
g
g

easier arc initiation

better arc stability
good wetting action
ability to maintain a short arc

Direct current is especially useful for applications requiring small diameter electrodes and low currents,
e.g., out of position welding, welding of thin materials, etc. When direct current is used for SMAW,
DCEP (electrode positive) polarity provides deeper penetration and DCEN (electrode negative) polarity
provides a higher electrode melting rate.

4.2.4 Classification of Electrodes

Shielded metal arc welding electrodes are available for welding of carbon and low alloy steels,
stainless steels, cast irons, aluminum, and copper and nickel and their alloys. However, electrodes for
welding carbon and low alloy steels and for stainless steels are of greatest commercial significance,
and systems for their classification as described in CSA Standard W48-01 are summarized here. For
more details, see Module 6 - Electrodes and Consumables of the CWB Modular Learning System
(MLS).

1) Carbon and low alloy steel electrodes

The electrode designation comprises the letter E (for electrode) followed by digits, e.g., E4918 for
metric designation (E7018 for imperial). For metric designation, the first two digits indicate the
minimum tensile strength (in MPa) of the weld metal. For imperial designation, the first two digits
indicate the tensile strength in ksi.

141

The third digit (imperial designation or metric designation) indicates the welding position for which the
electrode is designated with 1 meaning suitable for all welding positions (flat, horizontal, vertical and
overhead), 2 suitable for horizontal fillet and flat positions only and 4 meaning suitable for vertical,
downwards progression only.
The fourth digit (imperial or metric) indicates the usability characteristics of the electrodes (type of
coating, welding current type, etc.). For example, 0 and 1 indicate cellulosic covering, 2 and 3 indicate
covering containing rutile, 8 indicates low hydrogen, iron powder containing covering, etc. These digits
may be followed by additional letters and digits, which are usually indicators of weld metal toughness
or alloy content. Further details about digits, indicating suitable current type and polarity, etc., can be
found in CSA Standard W48-01 or from electrode manufacturers.
The usability characteristics of some of the more commonly used electrode types can be summarized
as follows in metric designation.
EXX10:

high cellulose, sodium compounds for arc stability, DC electrode positive polarity;
deeply penetrating arc; suitable for all welding positions; may be used for welding
from one side with adequate back bead profile; 5 mm or smaller diameter electrodes
used for all position welding;

EXX11:

high cellulose, potassium compounds for arc stability, AC or DC electrode positive

polarity; otherwise similar to EXX10 electrodes;

EXX12:

high titania with sodium compounds, AC or DC electrode negative polarity; medium

penetrating, quiet arc; most often used for single pass, high speed, high current,
horizontal fillet welds;

EXX13:

high titania with potassium compounds, similar to EXX12 type; used for sheet metal
work for vertical down welding; provides better radiographic quality in multipass
welds than EXX12 electrodes;

EXX14:

high titania and iron powder covering; AC or DC either polarity; similar to EXX12 or
13 but with iron powder providing a higher deposition rate;

EXX15 :

basic covering with sodium compounds; DC electrode positive polarity; limestone

and other basic ingredients in the covering provide weld metal with good toughness
and low hydrogen content; also suitable for welding high sulfur steels; usually 4 mm
or smaller diameters are used for all position welding;

EXX16:

basic covering with potassium compounds; AC or DC electrode positive polarity;

otherwise similar to EXX15;

EXX18:

basic, iron powder covering; similar to EXX15 or 16 but with iron powder in the
covering thus providing higher deposition rates; most structural steels are welded
with EXX18 type of electrodes;

142

EXX22:

iron oxide covering; AC or DC either polarity; used for single pass, high speed, high
current flat and horizontal lap and fillet welds in sheet metal;

EXX24:

titania, high iron powder covering; AC or DC either polarity; similar to EXX14

electrodes but restricted to welding in flat and horizontal positions; used mostly for
fillet welds;

EXX28:

basic, high iron powder covering; AC or DC electrode positive polarity; similar to

EXX18 but with higher iron powder content; suitable for welding horizontal fillets and
flat position welds only;

EXX48:

basic, iron powder covering; AC or DC either polarity; also similar to EXX18 but
designed for welding in the vertical position with downward progression.

2) Stainless Steel Electrodes

Requirements for covered electrodes for welding stainless steels are included in CSA Standard W4801. These electrodes are classified based on the chemical composition of the undiluted weld metal,
the welding position and the type of welding current for which the electrode is designed. A typical
designation can be represented as EXXXxx-XX where E represents electrode, and the next three digits
and any letters immediately thereafter (e.g., 309L, E310M) indicate the weld metal composition. The
last two digits are usually 15, 16, 17 or 26, where digit 1 indicates suitability for all position welding for
electrode diameters up to 4 mm. Conversely, digit 2 indicates suitability for flat and horizontal positions
only. The number 5 indicates that the covering contains calcium carbonate (limestone) and sodium
silicate, and that the electrode is suitable for welding using DC electrode positive polarity. The letter 6
indicates the presence of titania and potassium silicate in addition to the calcium carbonate. The
presence of potassium compounds makes the electrode suitable for AC welding. The 7 signifies an
acid flux with a significant amount of silica, which makes the slag more fluid.
The EXXXxx-15 electrodes provide a more penetrating arc, and a convex and coarsely rippled bead.
These electrodes are preferred for out-of-position welding since the slag solidifies quickly. The
EXXXxx-16 electrodes provide a smoother arc, less spatter, and a finely rippled bead. These
electrodes are less popular for out-of-position work because the slag is quite fluid. For more details,
see Module 6 - Electrodes and Consumables of the MLS series.
3) Handling and Storage of Electrodes
The electrodes should be handled with care to ensure that the electrode covering is not chipped off.
Unopened boxes should be stored at 30C 10C with relative humidity less than 50%. Cellulosic
electrodes (EXX10 or 11), however, are supposed to have a certain amount of moisture in the covering
and therefore should be stored in relative humidity of 20 to 70%.

143

Electrodes with basic (low hydrogen) coatings (containing calcium carbonate) are prone to moisture
absorption from the atmosphere and therefore should be packaged in hermetically sealed containers.
Once the container is opened, the electrodes should be removed from their packaging and stored in a
holding oven at a temperature of about 120C. Also, if the basic electrodes for welding carbon and low
alloy steel have been exposed at ambient temperature for 4 hours or more, or if their packaging has
been damaged, they should be rebaked at a temperature (370C to 430C) and for a time (1 to 2
hours) recommended by the electrode manufacturer. Cellulosic electrodes, however, should not be
placed in holding ovens or rebaked.

4) Selection of Electrode Diameter and Current

The classification and size of electrode, and the welding current for a given application are chosen in
light of the thickness of the material to be welded, groove geometry and welding position. Generally,
larger diameter electrodes are used for welding thick materials and in the flat position so that higher
deposition rates can be achieved. Smaller diameter electrodes are generally needed for welding the
root passes in V grooves and for out-of-position welds so that the welder can have better control of the
weld pool and the bead shape.
For prequalified joints, CSA Standard W 59 Welded Steel Construction limits the maximum electrode
size to 4 mm for welding in the vertical position (fillet and groove welds), and to 5 mm for groove welds
in horizontal and overhead positions, and fillet welds in the overhead positions. Larger diameter
electrodes are used for welding in the horizontal and flat positions only.
Table 4.1 shows typical current ranges for satisfactory electrode burn off and stable arc conditions
using steel electrodes of various diameters. However, the complete range of current may not be
suitable for all situations. When welding on thinner material, the lower end of the range might be
applicable. This would also apply when welding in the vertical or overhead positions. For example, 3.2
mm diameter E4310 electrode, according to Table 4.1 has a usable current range of 75 to 125 A. For
joining heavy material in the flat position, it would be logical to use the upper part of the range, 100 to
125 A. But if welding is to be done in the vertical up position, the range might be 90 to 110 A.

144

Table 4.1: Typical Current Ranges (in Amperes) for Electrodes of Different Diameters
(from CSA Standard W48-01)
Electrode
Diameter,
mm
1.6
2.0
2.5
3.2
4.0
5.0
6.0
8.0
Electrode
diameter,
mm
2.5
3.2
4.0
5.0
6.0
8.0

E4 X 00
E4 X 10
E4 X 11

45 85
75 125
110 170
155 235
190 290
275 425
E4915
E4916
70 120
110 150
140 220
200 280
270 350
375 475

E4 X 12

E4X 13

E4 X 22

E4 X 27

E4914

20 40
25 60
40 90
80 140
110 190
155 265
225 360
300 500

110 160
140 190
200 410
380 520
-

125 185
160 240
230 330
270 380
375 475

90 135
110 160
150 210
220 300
295 375
390 500

E4918

20 40
25 60
50 90
80 130
105 180
165 250
225 315
320 430
E4924
E4928

80 110
115 165
150 220
220 350
285 360
375 470

110 160*
140 190
180 250
250 335
300 390
400 525*

80 140
150 220
210 270
-

E4948

* These values do not apply to the E4928 classification.

4.2.5 Applications and Limitations of the SMAW Process

The shielded metal arc welding process can be used to weld most metals and alloys of engineering
significance. It has been extensively used to weld all types of steels (carbon and low alloy steels,
stainless steels, etc.) in the fabrication of pressure vessels, oil and natural gas pipelines, field storage
tanks, bridges, buildings, ships and offshore structures, railway cars, trucks and automobiles, nuclear
power stations and numerous other miscellaneous products including those made from cast iron.
Among the non-ferrous alloys, the shielded metal arc welding process is used for welding nickel and
nickel-based alloys, and to some extent copper alloys, such as bronzes. Though electrodes are
available, it is not popular for welding aluminum alloys. The process is also used for hardsurfacing
various components exposed to wear, impact, corrosion and heat resistant alloys.

145

The shielded metal arc welding process is usually the most appropriate for repair and maintenance
welding since each job is usually a one-time-only situation, the amount of welding required is relatively
small and in-situ locations are most suitable for the shielded metal arc process only. The process is
also frequently the only one in shops where welding constitutes only a small portion of the complete
manufacturing process. The shielded metal arc welding process is also generally the easiest to use in
the field due to the simplicity of the equipment and its tolerance to the normal outdoor environment.
Nonetheless, it is advisable to install protective enclosures when welding in the field to get protection
from rain, wind, etc.
The advantages of the shielded metal arc welding process thus include its applications to a variety of
materials, and the ability to weld in all positions (vertical and overhead as well as flat and horizontal)
and at most locations. As well the equipment required is easily portable and relatively inexpensive. The
main limitation of the SMAW process is the necessity of frequent breaks as each electrode is
consumed to about 50 mm of its original length and a new one used to re-initiate the welding operation.
This frequent change of electrode along with the need to chip off the slag means that duty cycle
(percentage of time that an arc is maintained for the purposes of welding) is less than 20% and the
deposition rate is low. Also, the unusable electrode stubs add to waste and cost of the filler material.

4.2.6 Shielded Metal Arc Welding of Carbon and Low Alloy Steels
Joint Design
For base metal thickness up to about 6 mm, a square groove with suitable root opening may be
employed for a complete penetration groove weld, provided that welding is performed from both sides
and in the flat position. At low current, a skilled welder can weld base metal as thin as 1.6 mm. For
larger thicknesses, the base metal edges must be beveled, and in very thick sections, J- and U-grooves
become more economical by reducing the weld metal volume required. The root gap for groove welds
is typically equal to the electrode diameter to achieve complete penetration, and the groove angle
should be large enough to achieve side wall fusion and minimize slag entrapment. In assembling a
joint for welding, the fit-up should be good enough to maintain the groove geometry within acceptable
tolerances. Thus, too small a root gap or misalignment between the two members to be joined can
locally lead to incomplete joint penetration. Fit-up tolerances and workmanship and some prequalified
joint geometries given in CSA Standard W 59 Welded Steel Construction are shown in Table 4.3 and
Figure 4.6, respectively. For complete prequalified joint design, see Clause 10.2 of CSA Standard
W59.

146

Table 4.3 : Fit-up and Workmanship Tolerances for SMAW Groove Welds

1. Root Face of Joint

2. Root Opening of Joints:
Without Steel Backing
With Steel Backing
3. Groove Angle of Joint

Root Not Gouged

Root Gouged

+ 2 mm

Not limited

+ 2 mm

+ 2 mm
- 3 mm

+ 6 mm, - 2 mm

Not applicable

+ 10E, -5E

+ 10E, - 5E

S
(T)

G =T
min
T
= 10

Backing Strip

ma x

S
T(T)

T
Backing Strip
2

Positions

20
30

12
10

F, O only

45

F, V, O

60

F, V, O

Figure 4.6: Typical prequalified complete joint penetration groove welds

for the shielded metal arc welding process (SMAW).
147

4.3 Gas Metal Arc Welding (GMAW)

4.3.1 Principles of Operation
The gas metal arc welding process is shown schematically in Figure 4.7. Compared to the shielded
metal arc welding process, the metal electrode is bare (without any covering). The coiled wire
electrode is fed continuously through the welding gun. The continuous wire feed improves the
productivity of the process by allowing longer welds to be made without stopping. In contrast, in
SMAW the length of weld that can be deposited is limited by the length of the electrode. The
protection of the weld zone from atmospheric contamination is provided by a continuous stream of
shielding gas or gas mixture.

Solid Wire Electrode

Current Conductor

Shielding Gas In

Travel

Wire Guide and Contact Tube

Shielding Gas
Gas Nozzle

Molten Weld
Metal
Solidified
Metal
Base
Metal

Figure 4.7: Schematic representation of the GMAW process.

An arc is struck between a continuously fed bare consumable wire electrode and the workpiece. The
heat generated by the arc melts the end of the electrode and part of the base metal in the weld area.
The arc transfers the molten metal from the tip of the melting electrode to the workpiece where it
combines with the melted base metal to form the weld deposit.

148

The process was first applied to the welding of aluminum using inert gases for shielding the arc and
the weld pool. The term MIG (Metal Inert Gas) welding has been a popular name for the process.
However, for joining of steels, it is common to have carbon dioxide and/or oxygen present in the
shielding gas mix. These two gases are not inert. Changing the proportions of carbon dioxide and
oxygen in the shielding medium can influence the chemical composition and therefore the properties of
the weld metal. The process therefore is sometimes referred to as Metal Active Gas (MAG) welding in
Europe. In North America, a more generic description Gas Metal Arc Welding (GMAW) has been
adopted.
The equipment arrangement for the GMAW process is shown schematically in Figure 4.8. It
comprises a power source, electrode wire feeder and control system, the welding gun and a supply of
shielding gas.

* Flowmeter
2000
1500

* Wire Feed Speed

Control

2500

1000

3000

500

3500

4000

CWB

* Voltage/Amperage
Output Adjustment

On
Off
4 5
3
2
1

0 10

6
7
8
9

120

150

180

90

* Output Selector (AC,

CC, CV)

210

60

240
270

30
0

300

Gun

* Shielding Gas

Work Lead

* = Variables that must be selected for GMAW

Figure 4.8 Equipment arrangement for GMAW.

149

A constant potential (i.e., a constant voltage) power source and a constant speed wire feeder are
generally used for GMAW welding, and current type is almost exclusively direct current with electrode
positive (DCEP). In such an arrangement, the amperage controls the electrode melting rate (wire feed
speed) and the power source tries to maintain a constant voltage by adjusting current output. Voltage
is closely related to arc length. Should there be a change in the arc length (for example, due to
welding over a tack weld or moving the gun toward or away from the workpiece), the power source
responds by changing current output. Since current affects wire melt-off rate, controlling current to
keep the wire burning off at the same distance from the puddle will maintain an essentially constant arc
length - constant arc voltage. The power source is constantly responding to the changing demands of
the arc, and to fluctuating input power. For the process to operate in a stable manner the power
source must be capable of responding correctly. The following describes a typical power source
response.
In Figure 4.9(a), the preset welding parameters are 400 A and 34 V, and let us assume that the
corresponding wire feed speed is 400 inches per minute (one inch per minute per ampere welding
current). When arc length increases, the power source responds by reducing current output and
thereby slowing wire melt-off rate. Figure 4.9(a) shows that if arc voltage increases to 37 V, the
operating current and wire melt-off rate would be 325 inches per minute. However, the wire feeder will
still keep on feeding wire at 400 inches per minute. Since the wire feed rate rate is greater than the
wire melt-off rate, the electrode extension will increase and the arc length will progressively decrease,
and the operating point will again move towards the initial setting. If the arc length were to shorten
inadvertently, then the adjustment would be just reversed (see Figure 4.9(b)).

56
52
48
44
40
Voltage 36
(Volts) 32
28
24
20
16
12

56
52
48
44
40
36
Voltage
32
(Volts)
28
24
20
16
12

New Operating Point

(37, 325)
Old Point
(34, 400)

100 200 300 400 500 600

Welding Current (Amps)

Old Point
(34, 400)

(31, 475)
New
Operating
Point
0

100 200 300 400

500 600

Welding Current (Amps)

Figure 4.9(a): Shift in operating point

due to an increase in arc length.

Figure 4.9(b): Shift in operating point

due to a decrease in arc length.

150

4.3.2 Equipment
The GMAW process is most often used in the semi-automatic mode, that is, a welder holding the gun
moves and guides it along the weld seam while depositing the weld metal. This also gives him the
flexibility to manipulate the gun to maintain appropriate weld pool shape and wetting and fusion along
the side walls. A typical gun is 250 to 375 mm in length and provides the means to supply current,
continuously feed the wire and provide the shielding gas to protect the arc and the molten weld pool
(see Figure 4.10). Some guns rated for higher currents or higher duty cycles may also have provision
for water cooling.

WaterWater
HosesHose

Steel Liner

Shielding Gas Hose

Gas Diffuser

Protective
Sheath

Gas Nozzle
Copper
Contact Tip

Gun Trigger

Handle
Power Cable

Figure 4.10: Schematic sketch of GMAW gun.

151

The direct current constant potential power sources used for gas metal arc welding can be engine
driven generators, transformer rectifiers or inverters. The latter two types are more common since the
generator type power source responds slowly to changing arc conditions.
It should be noted that though the power source recommended is a constant voltage type, the voltampere curve does have some slope instead of being a flat horizontal line. Also, the electric circuit in
the power source has some inductance, a characteristic that controls the rate at which the current
increases in the case of a short circuit. Slope and inductance together determine the dynamic
characteristics of the power source and are key factors affecting the performance of a GMAW power
source for semi-automatic applications. As a result of different slopes and inductance values, one
power source may operate more smoothly for a given set of welding conditions than another. Some
power sources are available with adjustable slope and inductance, allowing them to provide smooth
operation for a range of wire types and diameters. More information about these features can be
found in Module 5 - Power Sources for Welding.
In a conventional semi-automatic equipment set up, an analog constant speed wire feeder is used in
conjunction with the constant voltage power source. The wire feeders main components are drive
rolls, guide tubes, a gear box, a variable speed motor, wire support and controls and meters. Figure
4.11 shows a four-drive-roll system, which is more dependable than a two-roll system. A grooved roll is
usually combined with a flat roll for feeding solid wires. The groove is usually V-shaped for carbon and
stainless steel, and U-shaped for softer aluminum wires.
With the basic analog wire feeder and standard power source set up, the wire feed is adjusted by using
an incremental dial on the feeder and checking the wire feed speed with a hand held meter.
Alternatively, the dial may be adjusted to obtain a desired current reading. Although connected, the
wire feeder and power source do not communicate with each other in this set up; the power source
simply supplies the necessary power to burn off the wire as fast as it is fed into the arc. As a result, for
the same dial setting, the actual wire feed speed can vary depending on the actual line voltage,
slippage, etc. Therefore, unless the wire feed speed is verified on a regular basis, there will be
variations in welding current and arc characteristics for no apparent reason.

152

Pressure Adjusting
Screws
Drive Rolls
Inlet Guide
Tube

Centre Guide
Tube

Gear Box
Motor

Wire Feed and

Power Cable

Consumable
Wire

Inlet Guide
Assembly

Outlet Guide
Tube
Idler Roll
Drive Rolls

Figure 4.11: Four-drive-roll wire feed system.

153

On digital wire feeders with wire feed speed modulation capabilities, the target wire feed speed can be
set directly. The feeder provides an accurate set-to-actual speed relationship through the use of better
speed control on the feed motors, feedback of actual speed from tacho-generators, etc. Otherwise,
there is no special communication between the feeder and power source. Digital wire feeders,
compared to analog ones, result in better repeatability of procedures, which positively affects the
quality and economy of welding operations.

4.3.3 Metal Transfer Across the Arc in GMAW

The GMAW process is identified with a number of different modes of metal transfer depending on the
following welding parameters: current, electrode size, shielding gas composition, electrode chemistry
and the type of power source.
Once the tip of the electrode melts into a globule of molten metal from the heat of the arc, one of the
forces acting to detach it and propel it across the arc to the weld pool is the electromagnetic pinch
effect. The strength of the magnetic field, and therefore the pinch effect, depends most strongly on the
current density (welding current divided by the cross sectional area of the electrode). Consequently,
the rate and mode of droplet detachment also depends on the current density. The principal droplet
transfer modes of interest in GMAW are: short circuiting, globular, spray and pulsed (where pulsed
transfer is a form of spray transfer).

1)

Short Circuiting Transfer

In the short circuiting mode (Figure 4.12), the current density, i.e., the amperage used in relation to
the wire size, is relatively low. The wire therefore melts at the electrode tip but the pinch force is not
enough to detach it. However, the wire feeder keeps on feeding the wire and therefore the molten
electrode tip comes into contact with the weld pool. When this happens, the constant voltage power
source increases the amperage which in turn increases electrode heating and the magnetic pinch
effect acting at the electrode tip. The magnetic forces pinch off the droplet, which is then drawn into
the weld pool by surface tension forces. The gap between the electrode and the weld pool is then
recreated and the arc is re-established. This process repeats itself very quickly, typically more than
100 times per second, so that the human eye does not notice the short circuits and the arc seems
continuous.

154

Drop

New Arc
Arc

Before
Transfer

No Arc
During
Short
Circuit

After
Transfer

Figure 4.12: Short circuit transfer.

The welding current and voltage for short circuit welding, also called short arc or dip transfer welding,
are relatively low and therefore it is best suited for welding of thin ferrous materials in all welding
positions, and root passes of thicker steels. The short circuiting mode of metal transfer can be difficult
to apply successfully to thicker materials because of the smaller diameters wires (1.2 mm or smaller)
and low currents (less than 200 A for 0.9 mm diameter wire), and the resulting low heat input which can
cause fusion problems (cold welding). Consequently, welding specifications for critical structural
applications such as pressure vessels, bridges, naval vessels, etc, prohibit short circuit transfer mode if
GMAW process is to be used. The shielding gases used for carbon-manganese steels are normally
carbon dioxide (CO2) or 75% argon - 25% CO2.
The short circuiting mode of metal transfer can not be applied to non-ferrous metals and alloys. Cast
irons are mostly welded in the short circuiting mode.

155

Figure 4.13: Globular transfer.

2)

Globular Transfer

Globular transfer occurs as the current and voltage increase beyond those for short circuiting transfer.
In this transfer mode (Figure 4.13), the molten drop of metal at the electrode tip can reach a diameter
1.5 to 3 times the wire diameter. This large drop of metal detaches from the electrode tip due to the
force of gravity. It has an irregular shape, may have a rotational motion and takes an irregular path
across the arc. The glob of molten metal splashes into the weld pool causing expulsion of some liquid
metal (spatter). Globular transfer in GMAW tends to splatter and is usually avoided.
Carbon dioxide as well as argon rich gas mixtures containing CO2 or oxygen can provide globular
transfer. Very good penetration characteristics can be produced at higher current levels. The main
drawbacks of globular transfer are spatter formation, irregular bead shapes and formation of numerous
slag islands.

156

3)

Spray Transfer

Spray transfer (Figure 4.14), also called axial spray, occurs at current and voltage levels above those
for globular transfer, and when an argon rich (85% minimum) shielding gas mixture is used. The
molten metal is transferred across the arc in a continuous stream of fine droplets, and the droplet
diameter is typically less than the wire diameter. The arc is quite stiff so that the drops travel directly
along the centerline of the electrode and into the weld pool, and therefore can be easily directed
without affecting the arc behaviour.

Figure 4.14: Spray transfer.

The transition current (the current at which the mode of transfer changes) for the change from
globular to spray transfer (Figure 4.15) depends on the wire diameter, shielding gas composition,
electrode composition and the electrical extension (the length of wire stick out at the contact tube). At
very high currents, above the range for axial spray, the line of metal drops begins to rotate about the
electrode axis, and there is an increase in spatter.

157

Short
Circuiting
Transfer

Globular
Transfer

Voltage

Spray
Transfer

Transition
Current

Current (amperage)

Figure 4.15: Transition current for globular to spray transfer.

Spray transfer is characterized by:

g minimal spatter;
g a relatively quiet and smooth arc;
g weld beads with good penetration and nice appearance.
However, because of the high current and voltage levels, the weld pool is rather large and difficult to
control for out-of-position welds. Spray transfer is therefore suitable for welding in the flat and
horizontal positions, and welding thick materials.

Pulsed Transfer (GMAW-P)

Pulsed transfer is a form of spray transfer. Its primary benefits are:
g
g
g
g

all-position capability for ferrous and non-ferrous metals

more productive for thin material than GTAW
no spatter even for difficult filler metals
able to use larger diameter electrodes

158

Pulsed spray transfer involves the use of a specially designed power source whose current output
changes or "pulses" between a peak value and a background value at a rapid but controllable rate
(Figure 4.16(a)). Peak current surges to above the transition value for spray transfer then drops to a
background level so that in each pulse a drop of metal is detached and transferred across the arc.
Background current is sufficient to maintain the arc and keep the electrode tip hot and ready to detach
the next droplet during the next pulse.
The average current is generally in the range for globular transfer, well below the spray transition
value, but the bead appearance resembles that obtained with spray transfer. Also, the lower average
current implies a smaller weld pool and lower heat input, thus enabling out-of-position welding and
welding of thin materials.

Pulse
Peak
Current
Spray Transfer
Current Range

Spike
Ip

Current,
A

Globular
Transfer
Current
Range

Ib

Background
Current
Time

Pulsed-Spray Welding Current Characteristics

Drop Formation

Figure 4.16(a): Pulsed spray transfer.

159

Aluminum and other reactive metals are welded with pulsed spray transfer. Larger diameter electrodes
improve feeding and reduce weld pool contamination to significantly reduce the wire surface
incorporated into the deposit. The length of electrode per kilogram(pound) greatly reduces as the
electrode diameter increases. (eg., 1 kg of 0.9 mm diameter aluminum wire is 592 m long. By
comparison, the same wire at 1.1 mm diameter is only 358 m long, a reduction of about 40%.)
An electronically controlled wire feeder with real-time wire feed regulation is used to ensure wire feed
speed always remains close to the set speed.
a) Effects of Pulse Parameters
Electronically controlled pulsed power supplies allow adjustment of a number of pulsing parameters
(Figure 4.16(b)):
g
g
g
g

pulse rate
pulse width
peak amperage
background amperage

b) Pulse Rate
Changes in wire feed speed are accompanied by changes in pulse frequency. As wire speed
increases, the pulse frequency and therefore average current must also increase so that wire feed
speed and burn-off rate are continuously matched. The increase in average current causes an
increase in heat input. Pulse frequency can readily be used to control arc length.
c) Pulse Width
Pulse width is the time at peak amperage. Average amperage and heat input are directly effected by
pulse width- both increase with increasing pulse width. Increasing pulse width also has some effect
on increasing droplet size and widens the arc cone (bead width increases).
d) Peak Amperage
Peak current must be high enough to be above the spray transfer transition. Peak current detaches
droplets and propels them across the arc. Peak current directly affects arc length - arc length
increases with increasing peak current. Some power sources produce a spike to promote droplet
detachment from the electrode tip.
e) Background Amperage
Control of current rise and fall during the pulse cycle is used to control droplet shape and to shape the
electrode end in anticipation of the next droplet detachment.

160

Current
Increase Pulse Rate (pulses per second)

Increasing Pulse Rate:

Current

- increases arc length

- increases heat input

Current

Decrease Pulse Rate (pulses per second)

Increasing Pulse Width:

Current

Increase Pulse Width (pulse peak time)

- increases arc length

- increases heat input
- increases penetration
- increases bead width

Current

Decrease Pulse Rate (pulses per second)

Increasing Peak Amperage:

Current

Increase Peak Amperage

- increases burn-off rate

- increases arc length
- decreases droplet size

Current

Decrease Peak Amperage

Increasing Background
Amperage:

Current

Increase Background Amperage

- increases arc length

- increases heat input
- increases penetration
- increases wetting action

Decrease Background Amperage

Figure 4.16(b): Summary of effects of pulse parameters.

161

Modern power sources allow peak current (lp), background current (lb) and the pulse width (duration or
frequency) to be pre-programmed for a given application (i.e., shielding gas, wire type and diameter)
and changes in wire feed speed are accompanied by changes in pulse frequency. As wire speed
increases, the frequency and thus the average current increases so that wire feed speed and burn-off
rate are continually matched. Some power sources produce a spike to facilitate the droplet
detachment from the electrode tip.
With modern GMAW-P equipment there is a wide variation from one manufacturer to another in arc
control methods and pulse programming. As a result, care must be taken in selecting appropriate
equipment. Procedures that were successful with one equipment package may not be duplicated
successfully on a different package and a certain amount of procedure development may be required
for each case.
Synergic power sources are electronically controlled power sources that can provide a variable pulse
frequency that is proportional to the wire feed speed. Synergic control is a "one knob" system that
changes a number of interrelated variables at one time, simplifying operator control. Synergic power
sources are commonly pre-programmed for 0.9, 1.2 and 1.6 mm diameter mild steel, stainless steel
and aluminum wires. The systems are designed to allow re-programming or "fine tuning" of the prepackaged programs.
The pulsed spray mode of metal transfer can be substituted for any of the three transfer modes
discussed in the preceding paragraphs. When developed and applied correctly, the pulsed spray
transfer mode enables welding in all positions, and helps reduce heat input, distortion and spatter.
The effect of metal transfer mode on weld bead shapes is shown in Figure 4.17(a) and 4.17(b) which
present cross sections of six bead-on-plate welds.

Figure 4.17(a): Short circuit and globular transfer.

162

Figure 4.17(b): Globular and spray transfer.

4.3.4

Shielding Gas

The following shielding gas or gas mixtures are normally used for welding of carbon and alloy steels:
g carbon dioxide
g argon-carbon dioxide
g argon-oxygen
Carbon dioxide is the least expensive of the shielding gases used for gas metal arc welding. Once
ionized, carbon dioxide has a high thermal conductivity, which helps to keep the arc plasma as a small,
dense column under the electrode, and the metal is transferred in either the short circuiting or globular
mode. The arc is less stable and spatters. The deposited weld bead has a rough surface but deep
and round penetration as carbon dioxide transfers the greatest amount of heat to the weld pool (Figure
4.18).

163

Rough Bead
Appearance
Cold,Peaked Bead
From High Surface
Tension

Excessive
Spatter

Steel

Round, Deep
Penetration from Non-axial Transfer

Finger-Like Penetration
From Axial Spray

Argon

CO2

Helium

Figure 4.18: Effect of shielding gas on weld bead shape.

Carbon dioxide is an active gas in the sense that at the arc temperature it dissociates to produce
carbon and oxygen, and the latter can oxidize the weld metal to form slag. The GMAW wires for use
with carbon dioxide shielding gas therefore have sufficient level of deoxidizers like silicon, manganese,
etc. to tie up the oxygen. As a result, the manganese and silicon contents of the weld metal tend to be
lower than those in the wire. Conversely, in the case of stainless steels, the weld metal can pick up
some carbon, which can make stainless welds more prone to corrosion.
Argon has a low ionization potential, which means that arc voltage and therefore the arc length can be
smaller. Also, in the ionized form, argon has a low thermal conductivity. This causes the arc column to
expand and extend upwards above the end of the electrode as the welding current is increased. The
electrons hitting the electrode above the tip cause local heating and tapering of the electrode. This
increases the local current density and the pinch force, causing small droplets to be easily detached
and propelled at a high velocity to the weld pool in the form of a spray. However, the arc tends to be
cold and unstable, and the weld bead formed is peaky with undercut and finger shape penetration
(Figure 4.18). As a result, pure argon is not used for welding of steels. With higher conductivity gases
such as carbon dioxide, the plasma column does not expand as much and therefore the electrons are
restricted to striking the end of the electrode only. Therefore, no preheating of the wire end occurs and
globular or short circuit transfer is promoted.

164

Small additions of active gases like carbon dioxide or oxygen to argon lead to the formation of a small
amount of iron oxide on the surface of the weld pool. The oxide is able to increase arc stability as it is
a better electron emitter, and it also reduces the weld pool surface tension. Lower surface tension
promotes weld pool fluid motion and helps to reduce the tendency toward lack of fusion-type flaws.
Carbon dioxide also transfers more heat to the base material and promotes rounded rather than finger
penetration. Argon-carbon dioxide mixtures contain 75% argon and 25% carbon dioxide, commonly
referred to as C-25 gas. This mixture provides better bead appearance and less spatter than straight
CO2 (Figure 4.19). It is generally used on mild and low alloy steels with short-circuiting or globular
transfer.

98% Argon
2% Oxygen (spray)

95% Argon
5% CO2 (spray)

75% Argon
25% CO2 (globular)

91% Argon
5% CO2
4% Oxygen

Figure 4.19: Effect of argon rich shielding gas mixture on weld bead shape.

Reducing the carbon dioxide content decreases the transition current for spray transfer, and therefore
mixtures containing 15% or less carbon dioxide are more conveniently used for spray transfer. When
the argon content is 85% to 92%, good penetration and smooth bead appearance are obtained. The
current required for spray transfer is still reasonably high and the resulting higher arc energy and good
penetration makes this gas composition range suitable for welding thicker materials.

165

With further increase in the argon content of the gas mixture to 95%, stable spray transfer can be
maintained at a lower voltage. As a result, the arc energy is somewhat lower and therefore these
higher argon containing gases are more suitable for welding thinner material in the flat and horizontal
positions, though they can be used for thicker materials as well.
Since argon is an inert gas, it does not influence the weld metal composition. Therefore, as the
amount of carbon dioxide, an active gas, is reduced in the argon-carbon dioxide mixture, a greater
proportion of manganese and silicon present in the wire will be retained in the weld metal. As well, the
weld metal will have a lower oxygen content and this can help to improve the notch toughness of the
weld metal. Thus, argon-5% carbon dioxide is a commonly used gas mixture for welding high
performance naval steels, where high notch toughness is very desirable.
There are some other gas mixtures that are used for stainless steel, aluminum and other alloys.
Information regarding these gas mixtures can be found in CWB Module 4, Chapter 4, Gas Metal Arc
Welding.

Safety with Gas Cylinders

The shielding gases and gas mixtures are normally supplied as compressed gases in cylinders.
Whether in use or in storage, the cylinders must be secured and handled carefully since knocks or falls
could damage the cylinder or the valve, and could cause a leak or an accident. The following
precautions should be taken in the use of gas cylinders:
g
g
g
g

always properly secure the cylinders

while standing to one side, momentarily open the valve to clear any dirt present before
connecting a regulator
after connecting the regulator, release the pressure-adjusting screw and then slowly open
the cylinder valve to prevent a high-pressure gas surge in the regulator
always shut off the cylinder valve and back off the adjusting screw when the cylinder is
not in use

The cautions given above apply to all shielding gases, whether for use with the GMAW process or
other gas shielded processes (FCAW, MCAW, GTAW) discussed later. For more details on Welding
Safety, reference CSA Standard W117.2 Code for Safety in Welding and Cutting (Requirements for
Welding Operators) or Module 1, Welding Health and Safety, of the MLS series.

166

4.3.5

Advantages and Limitations of the GMAW Process

The main advantages of the GMAW process are its application to a wide variety of materials, higher
deposition rate and productivity compared to the shielded metal arc welding process (Figure 4.20) and
the better quality of the deposited weld metal. As well, with the recently developed advanced welding
power sources and the availability of smaller diameter wires (0.9 and 1.1mm diameter), welding
procedures can be developed to apply the process in all welding positions, both in semi-automatic and
automatic modes.

35
Typical for GMAW
1.1 mm wire;
stickout 19 mm

30
25
Deposition 20
Rate,
15
lb/h

10

Typical SMAW
Range of Sizes
and Types

5
0
0

200

400

600

Welding Current, A
Figure 4.20: Comparative deposition rates of GMAW and SMAW.

The higher deposition rate results from the absence of electrode covering and higher current density
(same current but smaller diameter wire). The higher productivity of this process results from:
g
g
g

a higher duty cycle

the time saved in not having to clean slag or flux from the deposited metal
higher utilization of the filler metal

167

The weld metal deposited using the GMAW process is generally cleaner (fewer non-metallic inclusions)
and in the case of high strength structural steels, weld metal with superior toughness can be obtained
with proper selection of shielding gas (Ar-5% to 15% CO2). Such applications include girth welding of
large diameter natural gas and oil pipelines, submarine hulls, etc. The low hydrogen nature of the
process is an additional important characteristic, especially for welding of high strength steels.
One of the main limitations of the gas metal arc welding process is its sensitivity to the welding
parameters. Seemingly small changes in voltage, electrode extension, etc. can have a significant
influence on the bead shape and penetration, and thus on the incidence of weld flaws such as
incomplete fusion. Faithful reproduction of qualified welding procedures is therefore critical to obtain
sound production welds. In this regard, matching wire feed speed between the qualification procedure
and production situation is a good indicator that the correct procedure has been implemented. One
must also be aware that air drafts can reduce the effectiveness of the shielding medium causing
porosity in the weld metal.

4.3.6

Application of Gas Metal Arc Welding Process

Virtually all weldable materials can be joined with the GMAW process. Nonferrous alloys (aluminum,
magnesium, copper, nickel, titanium and their alloys) are welded using spray or pulsed spray mode,
and successful procedure development depends mainly on selection of the shielding gas and welding
parameters. The filler metal is often designed to somewhat match the base material in composition.
Due to the inert shielding gas, no significant changes in chemistry of the deposited weld metal should
occur. There are exceptions of course; aluminum filler metals are formulated to prevent hot cracking
and do not normally match the base metal chemistry.
Gas metal arc welding of structural steels on the other hand can be more complex. Considerations
include filler metal composition, shielding gas and metal transfer mode, as well as the metal thickness,
joint design and welding position.
Similar joint designs can be employed for gas metal arc welding as for shielded metal arc welding
except that groove angles can be reduced due to the smaller diameter of GMAW wires. GMAW can be
less forgiving than SMAW or FCAW, particularly when using smaller wire diameters. Good penetration
and fusion is easily obtained directly beneath the arc however, in many cases increased oscillation is
required to properly fuse into the sides of the joint, whereas for the same joint FCAW or SMAW can
produce satisfactory results without oscillation. High argon-content shielding gases create a directional
penetration shape (finger penetration), which is prone to incomplete fusion.
Steels in thickness from about 1 to 3 mm can generally be butt welded with square edges in one pass,
provided the gap is less than 3 mm. For steel thickness ranging from 3 to 6 mm, a complete
penetration groove joint can be obtained with square edge preparation by welding from both sides,
provided that there is adequate root gap (1 to 4 mm). Above 6 mm, it is customary to prepare the joint
edges. Thicknesses greater than 6 mm usually require multiple passes.

168

4.3.7

Electrode Wires for Gas Metal Arc Welding

The requirements for filler metals for carbon steels are covered in CSA Standard W48-01, and those for
higher strength steels (ultimate tensile strength greater than 490 MPa) in AWS Specification A5.28.
Gas metal arc welding wires are classified based on their composition and the expected weld metal
tensile strength. A typical gas metal arc welding electrode designation can be written as:

ER XX S-xxx
where

E
R
XX
S
xxx

designates an electrode
designates a rod or wire suitable for processes such as GMAW, GTAW, etc.
represent the minimum weld metal tensile strength (in increments of MPa) in the aswelded condition when deposited in accordance with a specified procedure (W4801). In AWS specifications, only two digits indicate the tensile strength in ksi
indicates a solid electrode
are a one to three digit/alphabet-digit combination indicating the composition of
the wire

For example, a wire designation ER49S-2 means that the as-welded deposit will have a minimum
tensile strength of 490 MPa and that the wire contains nominal amounts of zirconium, titanium and
aluminum for deoxidation purposes in addition to silicon and manganese. Such wires are capable of
producing sound welds in semi-killed and rimmed steels, especially using the short-circuiting mode of
metal transfer. Moreover, these wires can be used to produce acceptable welds even when there is
some rust present at the steel surface. Further guidance on the optimum use of various carbon steel
wires is given in the appendix to CSA Standard W48-01. It should be noted that CSA Standard W4801 certifies GMAW wires based on tests with 100% CO2 shielding gas only. Certified wires are
permitted to be used with argon-rich gas mixtures, but with certain restrictions on the CO2 and O2
contents. Argon rich gas mixtures cause an increase in weld metal manganese and silicon content
while decreasing oxygen content. This occurs becasue of a lack of active gas in the arc atmosphere.
Electrodes certified with CO2 are purposely over-alloyed with manganese and silicon to compensate for
losses in a CO2 arc environment. Retaining these alloys in the weld will increase the yield and tensile
strength properties before deciding to use shielding gas mixtures containing only small amounts of
oxygen and/or carbon dioxide.
Electrode diameters are commercially available in the range of 0.9 to 1.6 mm. The largest diameter
electrode that can be used depends partly on the steel thickness to be welded. It is usually 0.9 mm
diameter for workpiece thickness up to 10 mm, and 1.2 mm for workpiece thickness up to 20 mm.
Gas metal arc welding is treated as a controlled hydrogen welding process as long as due care is
taken to ensure that the electrode and the joint surfaces are clean. The shielding gas used must have
a low moisture content. Moisture content is evaluated by the temperature at which condensation
occurs. Welding grade gases usually have a dew point of -40EC.

169

4.4

Flux Cored Arc Welding (FCAW)

4.4.1 Principles of Operation

The gas shielded flux cored arc welding process combines specific features from both the shielded
metal arc and gas metal arc welding processes. A continuous filler metal electrode is used but it has a
hollow core. The core is filled with flux and other ingredients that perform the same functions as the
covering on the shielded metal arc welding electrodes, to stabilize the arc, generate gases and provide
a slag cover to shield the arc and weld metal from atmospheric contamination, purify the weld metal,
add alloying elements, shape the weld bead, etc. Further protection is provided by an externally
supplied shielding gas.
In operation, an arc is struck between the continuously fed tubular wire containing the fluxing and other
ingredients (flux cored wire) and the workpiece (see Figure 4.21). As in the GMAW process, the heat
generated by the arc melts the end of the electrode (the metal sheath and the ingredients inside) and
part of the base metal at the weld seam. The arc transfers the molten metal from the tip of the melting
electrode to the workpiece where it becomes the deposited metal. The arc travel along the weld seam
can be mechanized (automatic welding) or manual (semi-automatic welding).

Figure 4.21: Schematic representation of the gas shielded

flux cored arc welding process.

170

There are three groups of tubular electrodes are available for common use:
g

The first group of wires are called gas shielded flux cored wires and these are meant to
be used with an external gas shield following the original developments in the 1950s.

The first major variation of the gas shielded flux cored wire was the self-shielded flux
cored wire. With these wires, as the name implies, no external gas shield is used
(Figure 4.22) and instead all the required shielding of the arc and the weld pool is
provided by the gases formed by the break down of flux ingredients in the core and the
slag cover on the weld metal. A certain amount of nitrogen pick-up from the atmosphere
is unavoidable and therefore denitriders or nitrogen fixers such as aluminum are added
to the core ingredients.

Wire Guide and Contact Tube

Solidified
Slag

Tubular Electrode
Powdered Metal, Vapour Forming
Materials, Deoxidizers and Scavengers

Molten
Slag

Arc Shield Composed of Vapourized

and Slag Forming Compounds
Arc and Metal Transfer

Direction of Welding
Weld Metal
Weld Pool

Figure 4.22: Schematic representation of the self-shielded

flux cored arc welding process.

Another group of tubular wires are called metal cored electrodes. These wires combine
features of flux cored and gas metal arc welding wires. The continuously fed wire is cored
but does not contain fluxing ingredients. Instead the core contains only arc stabilizing
compounds, deoxidizers and metal powders. The shielding is therefore provided only by
the externally supplied shielding gas as in GMAW. Metal cored arc welding electrodes
are grouped with flux cored arc welding wires in Canadian Standard CSA W 48-01, but in
the United States, these wires are considered as a variation of gas metal arc welding
process.
171

4.4.2 Equipment
The equipment arrangement for gas shielded FCAW and MCAW is essentially the same as for GMAW
and is described previously (Figure 4.23). It comprises a constant voltage power source, a constant
speed wire feeder and control system, the welding gun and a supply of shielding gas. The power
source and the gun must be rated for the current levels that are likely to be used with the selected
electrode. Since the flux cored arc welding process may involve higher welding currents, guns for
semi-automatic welding can be provided with an attached protective hand shield. Most electrodes are
designed for welding with direct current electrode positive polarity. As in GMAW, the constant voltage
power source and constant speed wire feeder enable a constant arc voltage/arc length to be
maintained.
In the case of self-shielded flux cored arc welding, the guns used are slightly different since there is no
need for an external gas supply. Most self-shielded flux cored arc welding wires are designed for
welding with direct current, electrode negative polarity and with longer electrode extensions. Due to
the latter, an insulated extension guide is attached to the contact tube to ensure that the wire and the
arc are directed at the intended location.

Figure 4.23

172

4.4.3 Advantages and Applications of the Cored Wire Processes

The cored wire processes offer a high quality weld deposit with higher deposition rate and productivity
than the SMAW process. Higher productivity is a result of a high duty cycle, high deposition efficiency
and high travel speeds. Compared to GMAW, the cored wire processes are more tolerant of small
deviations in welding current, voltage, tip to work distance, etc., and therefore are more likely to
provide weld deposits free from incomplete fusion flaws.
Among the three cored wire variations covered here, the self shielded flux cored wires are better able
to tolerate air currents than the others and therefore are a more suitable candidate for field work. In
automatic applications, very high travel speeds are possible with self shielded wires, leading to high
productivity. However, these wires should be properly selected since some formulations are not
designed for multipass welds. Metal cored electrodes produce little if any slag or oxide, similar to the
GMAW process. However, the metal cored wires provide a higher deposition rate than does GMAW,
and also a wider, more rounded bead shape when argon rich gas shielding is used (Figure 4.24).

16

19 mm TTW
85% Ar, 15% CO2 Shielding Gas

14
Metal
Cored

12
10
8

Solid Wire
GMAW

Deposition Rate 6
(lb/h)
4
2
0

160

200

240

280

320

360

400

440

Arc Current, A

Figure 4.24: Comparative deposition rates for GMA and

metal cored wire welding with 1.6 mm diameter wires.
(TTW - tip to work distance)
Compared to GMAW, the main disadvantage of the cored wire processes is the amount of fume
generation. Self shielded tubular electrodes produce the greatest amount of particulate fumes, which
in some cases may be more than covered electrodes. Gas shielded flux cored and metal cored
electrodes normally produce less fume than covered electrodes but more than GMAW, though the
rates can vary significantly from wire to wire. Secondly, there is a need for interpass slag removal with
the flux cored wires. Finally, as for GMAW, the weld quality of gas shielded FCAW and MCAW welds
can be impaired by the presence of air drafts.
173

Tubular electrodes are available for welding several of the commercially significant metals and alloys
such as carbon and low alloy steels, stainless steels, nickel alloys, as well as for hardfacing and
surfacing applications. Depending on the wire size and the type of ingredients in the core, cored wire
processes can be applied for welding in all positions.
The flux cored arc welding process is a more productive substitute for shielded metal arc welding in
most applications. It is commonly used for medium thickness workpieces, which may be considered as
relatively thin for optimum application of the submerged arc welding process and relatively thick for
optimum application of small diameter wire, gas metal arc welding with CO2 gas shield. Such
applications are quite common in the fabrication of construction equipment. General structural steel
and industrial equipment fabrication (e.g., machine tool bases, ladles for the steel industry, etc.) are
also a major user of the process.
More recently, the flux cored arc welding process is being used for pipe welds. Applications in the
pressure vessel industry are also increasing gradually as newer wires provide lower weld metal
hydrogen content, better toughness and better control over excessive strength (Figure 4.25). Still, the
weld metal toughness can be adversely affected by thermal stress relief, and therefore weld tests
should be performed to confirm that weld metal toughness is still adequate after stress relieving.

*Minimum
valuevalue
specified
by CSAby
W48-01
for the
*Minimum
specified
CSA W48-01
E491T-9 Classification
for the E491T-9 Classification

*Maximum
value specified
by CSA W48-01
forW48-01
the
*Maximum
value specified
by CSA
E491T-9 Classification
for the E491T-9 Classification

Figure 4.25: Improved control of weld metal mechanical properties

and hydrogen content in recently developed E491T-9 wires.

174

4.4.4 Classification of Cored Wires

The requirements for carbon steel flux cored arc welding wires (both self-shielded and gas shielded)
are described in CSA Standard W48-01 and with some differences, in AWS Specification A5.20. Those
for low alloy steels, stainless steels, and for surfacing are included in AWS Specifications A5.29, A5.22,
and A5.21, respectively.
Gas shielded carbon steel and low alloy steel flux cored wires are usually classified as rutile or basic,
depending on the flux chemistry. Metal transfer using rutile wires is in the spray mode over a large
operating current range, and for all practical purposes there is no globular-to-spray transition current
(Figure 4.26). The deposited bead is generally smooth with excellent penetration, and out of position
welding capability is achieved by controlling the slag fluidity by suitably designing the core ingredient
mix. Recent improvements in the design of rutile wires include lower weld metal hydrogen content and
better notch toughness by microalloying the weld metal with titanium and boron.

400

19 mm TTW
85% Ar, 15% CO2 Shielding Gas

350
300
Wire Feed
Speed
(in/min)

250
200
150
Usable Operating Range
100
150

190

230

270

310

350

390

Current, (A)

Figure 4.26: Wire feed speed and spray transfer mode with 1.6 mm diameter rutile FCAW wire.

175

Basic flux cored electrodes have core ingredients rich in limestone and fluorspar, similar to the covering
on basic (E4918) electrodes. These electrodes do not readily operate in the spray mode. Metal
transfer occurs in the short circuiting mode at low currents and in globular mode at high currents
(Figure 4.27). While the penetration characteristics are comparable to that of rutile wires, the arc is
less stable, with considerable spatter. More important, the slag is very fluid making it difficult to use
basic wires for out-of-position welding. Due to the basic nature of flux ingredients, the weld deposit
has relatively low hydrogen content and superior notch toughness compared to rutile wires (Figure
4.28).

19 mm TTW
85% Ar, 15% CO2
Shielding Gas
Globular

Spray

400
Short
Circuit
300
Wire Feed
Speed
(in/min)
200

100

Usable Operating Range

180

220

260

300

340

380

420

Current (A)

Figure 4.27: Wire feed speed and metal transfer mode in 1.6 mm diameter basic FCAW wire.

176

9
Basic (E492T-5)

9
Rutile (E491T-9)

Figure 4.28: Comparative weld metal notch toughness and diffusable hydrogen levels
in weld metals deposited by basic and rutile wires.

19 mm TTW
85% Ar, 15% C02
Shielding Gas
Globular

400

300

Wire Feed
Speed
(in/min)

Spray

Short
Circuit

200
Transition
Current
100

Usable Operating Range

180

220

260

300

340

380

420

460

Current (A)

Figure 4.29: Wire feed speed and metal transfer mode in 1.6 mm diameter metal cored wire.
177

Metal transfer in metal cored wires can be in any of the three modes short circuiting, globular or
spray, depending on the welding parameters and shielding gas (Figure 4.29). In practice, spray mode
is used most often. For out of position applications pulsed spray transfer can be used.
The cored wire designation schemes followed in CSA Standard W48-01 and in AWS Specification
A5.20 for classification purposes are shown in Figures 4.30(a) and 4.30(b), respectively.

Electrode

The letter M designates that the

electrode is classified using 75% 80% argon, balance CO2 or that the
electrode is self-shielded.

XMJ

Slag System,
Current, Polarity,
Shielding Gas
Minimum Tensile
Strength
43 = 430 MPa
49 = 490 MPa

Type of Wire:
T = Flux Cored Electrode
C = Metal Cored Electrode

Welding Positions:
1 = All Positions
2 = Flat & Horizontal Fillets

HZ

X X

Optional designator
about controlled
hydrogen, Z indicates
the maximum
diffusable hydrogen
per 100g or deposited
weld metal. Z can be
2, 4, 8 or 16.

The letter J designates that the

electrode meets the requirements for
improved toughness of 27 J at -40C.
Absence of the letter J indicates
normal impact requirements as given
in Table 16.

Figure 4.30(a): Classification scheme for flux cored wires in CSA Standard 48-01.

178

Electrode

XM

Slag System,
Current, Polarity,
Shielding Gas
Minimum Tensile
Strength
6 = 60 ksi
7 = 70 ksi

Tubular Wire

Welding Positions:
0 = Flat & Horizontal
1 = All Positions

HZ

Optional
designators
about controlled
hydrogen.

The letter M designates that the

electrode is classified using 75%
- 80% argon, balance CO2
shielding gas. When the letter
M designator does not appear, it
signifies that either the shielding
gas used for classification is
100% CO2 or that the electrode
is self-shielded.

Figure 4.30(b): Classification scheme for flux cored wires in AWS specification A5.20.

The main differences between the two schemes are:

g

Two digits are used in the CSA scheme to denote the minimum weld metal tensile strength
(in increments of 10 MPa) as opposed to a single digit (equal to ksi/10) used in the AWS
scheme;

For the welding position indicator, CSA Standard uses the digit 2 to indicate suitability for
flat and horizontal positions where as the AWS system uses 0 for the same purpose;

Metal cored wires are included in CSA Standard tables of CSA W48-01 dealing with flux
cored wires whereas in AWS, metal cored wires are included in tables CSA W48-01
dealing with solid wires for gas metal arc welding;

The last digit in the cored wire designation in the classification scheme denotes the slag system,
current polarity and shielding gas are shown in Table 4.4.

179

Table 4.4 : Shielding gas, current, polarity and slag system for
electrodes of different classification.

CSA W48-01
Classification
T-1*
T-2*
T-3
T-4
T-5*
T-6
T-7
T-8
T-9*
T-10
T-11
T-12*
T-13
T-14
T-G
T-GS
C-3*
C-6*
C-G
C-GS

Application

Slag System

Shielding Gas

Current and Polarity

Multiple Pass
Single Pass
Single Pass
Multiple Pass
Multiple Pass
Multiple Pass
Multiple Pass
Multiple Pass
Multiple Pass
Single Pass
Single Pass
Multiple Pass
Single Pass
Single Pass
Multiple Pass
Single Pass
Multiple Pass
Multiple Pass
Multiple Pass
Single Pass

Rutile
Rutile
Fluoride, rutile
Fluoride
Lime, fluoride
Basic oxide
Fluoride
Fluoride
Rutile
Fluoride
Fluoride
Rutile
c)
d)
b)
b)
Not applicable
Not applicable
Not applicable
Not applicable

CO2*
CO2*
None
None
CO2*
None
None
None
CO2*
None
None
CO2*
None
None
a)
a)
CO2*
CO2*
a)
a)

dc, electrode positive

dc, electrode positive
dc, electrode positive
dc, electrode positive
dc, electrode positive
dc, electrode positive
dc, electrode negative
dc, electrode negative
dc, electrode positive
dc, electrode negative
dc, electrode negative
dc, electrode positive
dc, electrode negative
dc, electrode negative
a)
a)
Dc, electrode positive
Dc, electrode positive
a)
a)

*The classification T-1M, T-2M, T-5M, T-9M, T-12M, C-3 and C-6 are possible if the qualification tests are
made with gas mixtures of 75% - 80% argon, balance CO2.
(a)
(b)
(c)
(d)

As agreed upon between supplier and user.

Slag system developed by the manufacturer for specific applications.
Designed for root pass in pipeline girth welds.
Designed for welding of galvanized and aluminized sheet steels.

The classification scheme for low alloy flux cored arc welding wires in AWS 5.29 Specification is similar
to that for carbon steel wires, the main difference being the higher weld metal tensile strength and
additional letters and numbers at the end used to indicate alloying elements present in the weld metal.
Similarly, metal cored wires for low alloy steels have the same classification scheme as low alloy steel
solid wires in AWS A5.28 except that S (denoting solid wire ) is replaced by C (indicating composite or
metal cored wire). In comparison, stainless steel flux cored arc welding wires are classified based
primarily on the weld metal composition and the shielding medium used during welding.

180

4.4.5

Shielding Gases for Flux Cored Electrodes

When required, carbon dioxide is used as the shielding gas for classification purposes. However,
Argon-Carbon Dioxide (Ar-CO2) mixtures are increasingly becoming popular as their use with rutile
wires provides less spatter, smoother beads and better wetting action and puddle control for out-ofposition welding. Similarly, with basic wires, less spatter and smoother beads are obtained. Fewer
fumes are generated when compared with 100% CO2 shielding gas. However, weld penetration is
reduced to some extent. For the reasons just mentioned, the last revision of the Standard CSA W4801 allows for a M9 (Mixed gas) designator in the classification, which allows classification of wires
with gas mixtures having 75% - 80% argon, balance CO2. Metal cored wires are used mostly with ArCO2 mixtures, as welding with 100%CO2 shielding gas is rare.
Since Argon (Ar) is an inert gas, it does not react with elements in the arc. Use of Ar-CO2 mixtures as
a shielding gas causes less oxidation of Manganese (Mn) and Silicon (Si) present in the wire, leading
to higher Mn and Si content in the weld metal. This increases the weld metal tensile strength, and may
also reduce the elongation values (Figure 4.31). Similarly, the amount of hydrogen retained in the
weld metal can be larger compared with the use of CO2 gas. The wires can be designed to avoid
excessive increase in weld metal strength and impairment in elongation, and therefore the
manufacturer should be consulted and/or procedure qualification performed before embarking on the
use of Ar-CO2 mixture with flux cored wires in fabrication. The shielding gas selected does not affect
the deposition rate to any significant extent.

Figure 4.31: Effect of shielding gas on weld metal strength and elongation.

181

4.5

Submerged Arc Welding (SAW)

4.5.1 Principles of Operation

The submerged arc welding process is shown schematically in Figure 4.32. Compared to the shielded
metal arc welding process, the flux to provide shielding is laid in granular form on the unwelded seam
ahead of the bare metal electrode. The electrode is fed continuously from a coil, thus avoiding the
interruptions inherent in the SMAW process to change electrodes. The flux is quite effective in
preventing the atmosphere from contaminating the molten weld metal and no external shielding gas is
required.

Figure 4.32: Schematic representation of the submerged arc welding process.

The arc is struck beneath the flux between the bare electrode and the workpiece, which melts a small
amount of the flux. Although a non-conductor when cold, the flux becomes highly conductive when
molten (about 1300C) providing a current path to sustain the arc between the continuously fed metal
electrode and the workpiece. The heat generated by the arc melts the end of the electrode, the flux,
and part of the base metal at the weld seam. The arc transfers the molten metal from the tip of the
melting electrode to the workpiece, where it becomes the deposited metal. As the molten flux
combines with the molten metal, certain chemical reactions occur that remove some impurities and/or
adjust the chemical composition of the weld metal.
While still molten, the flux, which is lighter than the weld metal, rises to the surface of the weld pool
and protects it from oxidation and contamination. On further cooling, the weld metal solidifies at the
trailing edge of the moving weld pool, and the weld bead usually has a smooth surface due to the
presence of the molten glass-like slag (molten flux resulting from all the chemical reactions) above it.
The slag freezes next and continues to protect the weld metal as it cools. Frozen or solidified slag is
readily removable, sometimes popping off the bead spontaneously. Excess, unmelted flux can be
recovered and reused after proper processing.
182

The complete welding operation takes place beneath the flux without sparks, flash or spatter, and it is
for this reason that the process is called submerged arc welding. As a result, the welding operator
does not normally need a protective shield or helmet.
Since there is a need to lay granular flux along the weld seam and the molten weld pool can be quite
large and fluid, submerged arc welding is best performed in the flat position, and if needed, in the
horizontal position. Also, since the operator can not see the arc or the weld seam, submerged arc
welding is best suited for situations where long welds with little or no geometric variation are to be
made in the flat position. The process can be mechanized or used in a semi-automatic mode.

4.5.2 Current Type and Equipment

The equipment set-up for single wire submerged arc welding is shown in Figure 4.33. In addition to
the power supply, a submerged arc welding system requires a wire feeder to maintain a continuous
feed of the electrode wire through the torch. For single wire submerged arc welding, direct current
electrode positive (DCEP) is used for most applications as it provides better control of bead shape,
ease of arc initiation, and deeper penetration welds with greater resistance to porosity.

Figure 4.33: Equipment set-up for single wire submerged arc welding.

183

Direct current electrode negative (DCEN) polarity is also occasionally used to provide a greater
deposition rate. However, penetration is reduced and there is some increased risk of lack of fusiontype flaws. From a practical point of view, a change from DCEP to DCEN may necessitate an increase
in voltage of about 2 to 3 V if a similar bead shape is to be maintained.
Both constant voltage and constant current (drooping voltage characteristics) power sources can be
used. With constant potential power sources, used in conjunction with constant speed wire feeders,
the arc length self-adjusts to a nearly constant value depending on the voltage, as in GMAW. The wire
feed speed and the electrode diameter control welding current, and the power source controls voltage.
By comparison, a constant current power source tries to simulate a manual welder. Essentially, a
voltage sensitive relay in a variable-speed wire feeder constantly adjusts the wire feed speed to
maintain the target arc voltage and, therefore, a constant arc length. The power source controls
current, and arc voltage depends on wire feed speed and electrode diameter. Modern power sources
are available that operate in either constant voltage or constant current mode.
Power sources are available that can deliver up to 1500 A. However, direct current is usually kept
below 1000 A since there can be excessive arc blow. Alternating current can be used to reduce arc
blow in high current applications and other situations prone to arc blow, e.g., multiwire and narrow gap
welding. Alternating current power sources are usually constant current type with a nearly square wave
output voltage to assist in arc ignition at each polarity reversal. Square wave constant potential power
sources have also become available that provide both voltage and current in square wave form and
therefore have less difficulty in arc re-ignition at polarity reversals. The weld bead penetration obtained
with alternating current is in between that for DCEP and DCEN.
A coil attached to the welding head provides a continuous feed of the metal electrode from the coil
through wire straighteners and a contact tip to the workpiece, and a hopper provides flux in front of the
metal electrode feed. The welding head is usually mounted on a carriage, where it moves at a
predetermined travel speed, thus enabling complete mechanization of the welding process.
Alternatively, the welding head can be fixed and the workpiece moves beneath it at a predetermined
speed.

4.5.3 Advantages and Applications of Submerged Arc Welding

By far, the greatest advantage of the submerged arc welding process is its high productivity, resulting
from high deposition rate and a high duty cycle. The high deposition rate is a consequence of the
mechanized nature of the process as it enables use of higher travel speeds and larger diameter wires
and therefore higher currents than possible with semi-automatic processes. Variations such as the use
of multiple wires, and the addition of a controlled amount of iron powder to weld seams along with the
granular flux can further increase deposition rate.

184

The weld deposit is considered to be a controlled-hydrogen type, provided due care has been taken
in storage and handling of flux and wire. Heated flux storage units, similar to electrode storage ovens,
are often used. Little fume is generated in the process and arc radiation and spatter are generally not
a problem. When the weld joint design is appropriate and welding parameters are chosen correctly,
sound welds with a smooth, uniform finish are easily obtained.
The main limitation of the submerged arc welding process is that it is limited to welding in the flat and
horizontal positions only. The mechanized nature of the process implies more expensive equipment
and greater set up time.
Most submerged arc welding applications are for carbon and low alloy steels. The process is also
used for joining stainless steel and nickel based alloys. However, the fluxes are proprietary in nature
and flux manufacturers must be consulted for optimum flux selection.
Because of the mechanized nature of the process, it is most effectively used when numerous similar
welds are to be made (splicing of plates and panels in shipyards, fabricated structural shapes, welding
longitudinal or spiral seams of large diameter oil and natural gas pipelines (see Figure 4.34) and when
the thickness to be welded is large (circumferential and longitudinal seams in thick wall pressure
vessels). Other applications of submerged arc welding include overlaying (stainless steel overlay on
chromium-molybdenum steels for high temperature, high pressure hydrogen applications) and
rebuilding and hard surfacing.

Figure 4.34(a) - Double submerged arc welding

of spiral seam in large diameter line pipe inside (Welland Pipe Inc.).

185

Figure 4.34(b) - Double submerged arc

welding of spiral seam in large diameter line
pipe - outside (Welland Pipe Inc.).

4.5.4 Multiple Wire Submerged Arc Welding

One of the great advantages of the submerged arc welding process is the ability to use multiple
electrodes fed into the same weld pool thus considerably increasing the deposition rate. Some
configurations (Figure 4.35) for multiple wire submerged arc welding are:
Parallel Electrode Welding: Also called twin wire welding, two electrode wires are connected in
parallel to the same power source. Both electrodes are fed by means of a single wire feeder and
through the same welding head. Welding current is the sum of currents for each electrode and a
single deep penetrating weld pool is obtained.
Multiple Arc Welding: Also called tandem welding, two (or more) electrodes can be connected to
individual power supplies and fed by separate drive rolls through separate contact tips. The lead
electrode in such cases is connected to a DC power source and the trailing electrode to an AC source
to reduce interaction between the magnetic fields of the two arcs. It is important to ensure that the
spacing between the arcs is not too large. The trailing arc is usually positioned close enough to the
leading arc that the slag cover does not solidify between deposits. The total current in multiple wire
welding can be as high as 2000 A, although in most applications it does not exceed 1200 A.
Series Arc Welding: Two electrodes, fed through separate guide tubes, are connected in series.
Separate sets of drive rolls and contact tips, insulated from each other, need to be employed. The
current path is from one electrode to another, through the weld pool. The weld bead has relatively
shallow penetration, making this arrangement useful for overlay welding.

Single Wire

AC
10 mm
Typical

Twin Wire Parallel Electrodes

20 to 75 mm
Typical

Tandem Electrodes

Figure 4.35: Submerged arc welding process.

186

DC
or
AC

Series Arc

4.5.5 Wires and Fluxes for Submerged Arc Welding of Carbon

and Low Alloy Steels
Traditionally, solid wires similar to those for GMAW have been used for submerged arc welding. The
electrode size tends to be larger and the composition may be different, since one must consider the
influence of the flux and the greater dilution from the base metal on the weld metal composition
(Figure 4.36). For this reason consumables for submerged arc welding are selected as a wire-flux
system rather than on an individual basis. More recently, composite wires (tubular wires with alloy
powder and other ingredients in the core) are being used for submerged arc welding. The advantages
of a tubular electrode is the wide range of deposit chemistry possible and the ability to increase travel
speeds.

50% Filler Metal

50% Base Metal
20% Filler Metal
80% Base Metal

70% Filler Metal

30% Base Metal

Figure 4.36: Dilution ratios of some common weld joints.

Fluxes for submerged arc welding can be categorized by method of manufacture or effects on weld
metal composition. There are two types of fluxes: fused fluxes and bonded fluxes. The manufacture
of fused fluxes involves melting together various ingredients to provide a homogeneous mixture, which
is then allowed to solidify by pouring it onto a large chilling block. The glass-like, solidified particles are
crushed, screened for sizing and then packaged for use. The main advantages of fused fluxes are
their chemical uniformity (irrespective of the flux particle size), resistance to moisture absorption and
easy recycling without changes in particle size or composition. The disadvantage of fused fluxes is
that it is difficult to add deoxidizers and ferroalloys because these compounds tend to oxidize during
the melting process.

187

In comparison, bonded fluxes are made by finely grinding the individual components of the flux,
mixing them in appropriate proportions and then adding a binder, typically potassium and/or sodium
silicate. The wet mixture is then baked at a relatively low temperature and ground to size for
packaging. The main advantage of bonded fluxes is that it is easier to add deoxidizers and ferroalloys.
On the negative side, such fluxes are prone to moisture pick up, and to local changes in composition
due to segregation or removal of fine mesh particles.
Fluxes that significantly influence the composition of the weld metal through slag/metal reactions are
termed active fluxes. Typically, these fluxes add manganese, silicon and chromium to the weld metal.
The extent of this addition increases with arc voltage, since higher arc voltage leads to increased flux
consumption (Figure 4.37). Very active fluxes may be used to deposit single or two pass welds only,
since the increase in the Si and Mn content of subsequent passes may be sufficiently large to impair
the weld metal ductility and also make it more prone to hydrogen cracking. Certain active fluxes,
termed alloy fluxes, add elements such as Ni and chromium. Such fluxes enable the welding of
weathering steels (containing chromium, nickel or copper) using carbon steel wires, and compensate
for the loss of chromium from the wire by oxidation when welding stainless steels.

Figure 4.37: Effect of arc voltage on weld metal silicon content for two active fluxes.

Neutral fluxes also participate in slag-metal reactions but the changes in silicon and manganese are
smaller and not dependent on arc voltage (Figure 4.38). There is little build up of elements and such
fluxes are therefore well suited for multipass welds.

188

Figure 4.38: Effect of arc voltage on weld metal silicon content for two neutral fluxes.

Fluxes are also referred to as chemically basic, neutral or acidic. Chemically basic fluxes have
Calcium Oxide (CaO) and Magnesium Oxide (MgO) as the major ingredients. Chemically acidic fluxes
have Silicon Oxide (SiO2) as the main ingredient. When the ratio of basic oxides to acidic oxides
present is greater than 1, the flux is chemically basic and when it is less than 1, it is chemically acidic.
Ratios near 1 imply a chemically neutral flux. Basic fluxes transfer smaller amounts of Si, Mn and
oxygen to the weld metal, and therefore are preferred for critical applications.
Requirements and selection for carbon steel wires and fluxes providing weld metal with minimum
specified ultimate strength of 490 MPa are detailed in CSA Standard W48-01, and for higher strength
weld metals, one can consult AWS Specification A5.23. Submerged arc welding wires are classified
based on their composition, whereas fluxes can only be classified in conjunction with a welding wire
and their classification indicates the weld metal strength and toughness. The classification scheme for
flux-wire combinations is shown in Figure 4.39. Thus, a flux-wire combination conforming to the
designation F49A5-EM12K indicates that: (i) the electrode wire has a medium (M) manganese
content, nominally 0.12% C (12) and is made from a silicon killed steel (K); and, (ii) when used with
the specified flux in a standardized test, will provide weld metal that, in the as-welded condition
(without post-weld heat treatment), will meet the requirements of minimum 490 MPa ultimate tensile
strength, and minimum 27 J Charpy Vee notch impact strength at -50C.

189

Figure 4.39: Classification system for submerged arc welding wires and fluxes.
(as per CSA W48-01)

190

It is important to note that as a result of the above classification scheme, a particular flux can assume a
different designation when used in conjunction with another wire. For example, Lincon weld 882 flux
when used with Lincoln weld LA-71 (EM14K) wire is classed as F49A4-EM14K or F49P5-EM14K but
when used with Lincoln weld L-61 wire, it is classed as F49A5-EM12K. Another consequence of the
joint effect of wire and flux on weld metal properties is that once a specific flux-wire system has been
approved to a particular classification, then no other flux or wire of the same designation but different
trade name may be substituted for it without a complete new series of tests to demonstrate that all the
requirements are still met. For more details on submerged arc welding consumables, see Module 6.

Flux Usage
Following are some of the precautions that should be taken in the storage and use of fluxes:
g

Fluxes can absorb moisture and thus compromise the controlled hydrogen characteristics
of the process. It is therefore important that once a flux bag is opened, it is stored in a
dry environment. If there is any doubt of its condition, the flux should be baked before
use, following the manufacturers recommendations.

Fluxes look alike and therefore if a flux is transferred to a different container for proper
storage, it should be properly identified.

In recovering and reusing flux, it should be ensured that particle size distribution is
maintained. Too many fines in the flux make it difficult to feed, and loss of fines may
change the flux composition, which may change the chemistry of the deposit.

When active or alloy fluxes are used, the specified welding parameters must be followed
diligently otherwise the weld deposit properties will be different from those expected.

Do not use an active or alloy flux where a neutral flux is required, and vice versa.

191

4.5.6 Submerged Arc Welding of Carbon and Low Alloy Steels

Joint Design
Because of the high currents and deep penetration possible with submerged arc welding, steels up to
12 mm may be welded in one pass without any edge preparation. With edge preparation, steels with
thickness up to 25 mm are weldable in one pass. However, it assumes that the joint is suitably
designed to prevent burn-through and that the weld zone mechanical properties achieved are
acceptable. Figure 4.40 shows a typical prequalified joint configuration for submerged arc welding.
Clause 10.2 of CSA W59 lists all the prequalified joint configurations.

Figure 4.40 - Prequalified joint from CSA Standard W59 for

submerged arc welding of carbon and low alloy steels.

192

Welding Procedures
For welded construction in accordance with CSA Standard W59, the following limitations are specified
for pre-qualified joints:
g

fillet welds up to 12 mm may be deposited in a single pass in the flat position; in the
horizontal position, the maximum single pass fillet size is 8 mm; in any case, the current
must not exceed 1000 A for the single electrode and 1200 A for the parallel electrode
variation of the process;

to prevent burn-through, either appropriate backing bars should be used or the root face
must be at least 6 mm; for root face less than 6 mm, a shielded metal arc weld pass
may be manually deposited on the back side;

the largest wire that may be used for submerged arc welding is 6 mm;

in groove welds, current for the root pass should be less than 10 times the groove
angle; this is to control the bead shape and dilution so as to reduce the likelihood of
weld metal solidification (centerline) cracking; for subsequent passes, welding
parameters should be chosen so that in cross section, the depth of the weld bead or its
width at any point along its depth does not exceed the surface width of the weld bead
(Figure 4.41);

Figure 4.41: Depth and width of weld bead.

with a single electrode wire, the layer thickness in groove welds is limited to 6 mm except
for the root and capping passes; this limitation does not apply to welds made with parallel
electrodes; also, split passes are required when the root opening is more than 13 mm, or
when the layer width exceeds 16 mm in multipass welds.

193

The limitations for multiple arc welds are slightly different; a fabricator can design welding
procedures outside the W59 limitations as long as a procedure qualification test is carried out to
demonstrate the adequacy of the welded joint.
As mentioned earlier, the submerged arc welding process is treated as a controlled hydrogen
process subject to proper storage and conditioning of the electrode and flux. Due to its
mechanized nature, it is capable of providing a sound weld deposit of uniform and consistent
properties. In some cases, taking advantage of these two process characteristics allows for the
elimination of preheating. Minimum size fillet welds for steels of different thickness and carbon
equivalent that may be deposited without preheat and without hydrogen induced heat affected cold
cracking are shown in Table 4.5. The minimum fillet size represents a certain minimum heat input
that, depending on the steel thickness and carbon equivalent, is expected to keep the heat
affected zone hardness below a critical level for cold cracking.

Table 4.5 : Minimum single pass submerged arc fillet weld sizes to eliminate preheat .
(from CSA W59)

Plate Thickness (t),

mm
T< 12 welded to t > 40
T > 12 welded to t > 40

0.35
8
8

Carbon Equivalent*
0.45
0.50
8
10
10
10

0.40
8
8

0.55
10
12

0.60
12
16

*Carbon Equivalent = C + (Mn + Si)/6 + (Cr + Mo + V)/5 + (Ni+Cu)/15

When joining quenched and tempered steels, caution must be exercised in using high heat inputs
(high deposition rates). The accompanying slower cooling rate can adversely affect the weld joint
mechanical properties. Table 4.5 is not applicable to quenched and tempered steels.

194

Chapter 5
Welding Metallurgy

Table of Contents
5.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197

5.2

Basic Concepts of Iron and Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .198

5.3

Iron, Cast Iron and Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .199

5.4

Phase Transformation During Heating and Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .200

5.4.1
Phase Diagrams (Iron-Carbon Equilibrium Diagram) . . . . . . . . . . . . . . . . . . . . . . . .202

5.5

Effect of Heating and Cooling on Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203

5.5.1
Slow Cooling of Steel from Above 910C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203
5.5.2
Fast Cooling of Steel from Above 910C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207
5.5.3
How Fast a Cooling Rate is Fast Enough to Form Martensite? . . . . . . . . . . . . . . . . .208
5.5.4
Heat Treatment of Structural Low Alloy and Quenched and Tempered Steel . . . . . . .211

5.6

Alloying Elements in Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213

5.7

How Does Hardness Affect Welding? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .215

5.8

Heat Affected Zone (HAZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216

5.9

Weldability of Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .218

5.9.1
Weld Cooling Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .219

5.10

Solidification Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .226

5.11

Strength and Toughness in the Weld Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .227

195

5.12

Hydrogen Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .229

5.12.1 Factors Affecting the Formation of Hydrogen Cracks . . . . . . . . . . . . . . . . . . . . . . .230
5.12.2 Avoiding Hydrogen Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .231

5.13

Heat Treatment of Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .234

5.13.1 Annealing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235
5.13.2 Normalizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
5.13.3 Quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
5.13.4 Tempering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
5.13.5 Stress Relief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
5.13.6 Concept of Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .237
5.13.7 Ways to Harden Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .238
5.13.8 Cold Work (Mechanical Deformation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239
5.13.9 Heat Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239

5.14

Influence of Welding on Mechanical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .240

5.15

Designation of Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .240

5.15.1 Carbon Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .241
5.15.2 Alloy Steel, Tool Steel and Stainless Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .241

5.16

Classification of Steels (Numbering System) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .241

5.16.1 CSA G40.21 - Canadian Standards Association . . . . . . . . . . . . . . . . . . . . . . . . . .242
5.16.2 SAE - AISI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .242
5.16.3 ASTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .244

196

5.1

Introduction

Metallurgy is an ancient science begun when our ancestors started to use metal tools thousands of
years ago. It has come a long way since cavemen discovered that when stones were heated, dark
lumps were left in the bottom of the fire pit. The lumps were hard and strong and they did not know
why black magic was born. Today, metallurgists explore the atomic structures of metals. Welding
metallurgy is the latest application of metallurgy.
We all know that metallurgy and materials science is a major discipline, as is civil engineering. It is a
formidable task to explain this subject to its full extent in this chapter. Therefore, we will present only
the basic principles of metallurgy with brief explanations that are essential for understanding welding.
To understand what is happening in welding, we must learn some fundamental principles of metallurgy.
With the ever-increasing demand to join vast arrays of materials in all types of manufacturing
industries, it is of the utmost importance to design weld joints to meet loading and environmental
conditions.
The phase diagram is an important tool to explain metallurgical make-up, changes or transformations
of alloy interactions at various temperatures. The students are advised to study the following CWB
modules to supplement this chapter:
Module
Module
Module
Module
Module

9
20
21
22
23

Introduction to Welding Metallurgy

Structure and Properties of Metals
Welding Metallurgy of Steels
Welding Metallurgy of Stainless Steels
Welding Metallurgy of Non-Ferrous Metals and Cast Iron

The development of our modern industrial society is closely related to the development of metallic
materials. In fact, materials like steel have been at the centre of most major industrial breakthroughs.
So as not to overextend our effort to all aspects of welding metallurgy, this chapter concentrates on
welding metallurgy of steels, which every civil engineer will likely be involved with at one time or
another. Once you are familiar with the metallurgy of steels, you should be able to venture into other
metals with the study of related technical references, which are numerous and readily available.

197

5.2

Basic Concepts of Iron and Steel

Elements such as carbon, oxygen, iron or copper have distinctive properties such as:

atomic weight

atomic diameter

density

melting point

boiling point

To create solid structures like metals, atoms

have to be joined strongly. In fact, atoms in
metals are joined in specific patterns. Iron
arranges its atoms in a cube, as shown in
Figure 5.1. This basic arrangement is a cube
with one atom on each corner and one in the
middle. This cube arrangement is the basic cell
or building block of steel (like bricks in a wall).
It is called Body Centered Cubic.
Alloys have properties that may differ greatly
from the parent elements. Adding carbon to
iron changes its properties, producing a new
material from the two elements steel.
Matter can normally be found in three states,
depending on the energy contained in the
atoms. When energy level is low, matter is
solid. As energy or temperature is gradually
increased, matter transforms from solid to liquid
and finally to gas. A good example of this is
water, shown in Figure 5.2. It can be found as
a solid (ice), as a liquid (water) or as a gas
(steam).

Figure 5.1: Body centered cubic

arrangement of iron atoms (BCC).

198

States of Matter
Solid

(e.g.: ice)

Liquid

(e.g.: water)
(e.g.: water vapour)

Gas

Figure 5.2: States of matter.

During welding all three states of matter are present.

Metals to be welded are in the solid state. The heat
generated by the welding arc will melt the metal and
gases will be produced. Metallurgy is a science that
studies these changes of states in metals and the
compounds they form.
Since welding is concerned with solid matter, we will
concentrate our efforts on elements that can be
worked within the solid state. Changes to a material
can happen in the solid state; metals are particularly
useful in this regard. For example, the properties of
steel can be changed while it is solid. One of the best
ways to change properties is to heat and cool the
material. Welding metallurgy not only studies the
weld metal, but it is also used to predict changes in
the base metal that happened due to the welding
heat. Welding locally heat treats the parts being
joined.
Welding metallurgy attempts to predict the effect of
this heat treatment on the structure and properties of
the material.

5.3

Iron, Cast Iron and Steel

Iron alloys can be subdivided into two groups - Steels and Cast Irons. Depending on the amount of
carbon contained in the mixture, the alloy will be called cast iron or steel. Cast iron contains more
carbon than steel.

1)

Cast Iron

Cast Iron has different grades and each has specific properties. However, three properties
characterize all grades of cast iron:
g
g
g

2)

high carbon content (higher than 1.7%)

lower melting points than other iron-carbon alloys (1150C to 1200C)
cannot be forged

Steel

Steel is the major product of iron-carbon alloys. In contrast to cast iron, steel has a carbon content
ranging from 0.01% to 1.7%. Surprisingly, reducing the carbon content in iron-carbon alloys produces
stronger, tougher and harder steels. Weldable grades of steels must keep the carbon content low
usually less than 0.4% by weight.

199

All commercial steel contains four main elements and some impurities:
g
g
g
g

iron
carbon
manganese
silicon

Steel mills spend millions of dollars to remove impurities from the metal while it is still in a liquid state.
As well, certain metallic compounds are added to improve the properties of the steel.

5.4

Phase Transformation During Heating and Cooling

In steel certain constituents may undergo changes in the solid state as temperature rises or decreases.
These changes are called phase transformations. When heated, metals (solids) will gradually
transform into liquids and gases. As stated previously, all matter can be found in these three distinctive
states.
Each state has specific properties that can be summarized as follows:

Gas

g
g
g

fills all space available

can be compressed
number of atoms in a given volume depends on pressure
and temperature

Liquid

g
g

cannot be compressed
atoms are relatively free to move

Solid

g
g

well defined volume

properties specific to a given orientation

Except for mercury (Hg), metals are normally found in the solid state. In solids, atoms are joined by
directional forces that hold them according to specific arrangements. Metallic atoms group themselves
in crystalline patterns (arrangements).
Metals arrange their atoms into three principal cubic patterns, which are shown in Figure 5.3.

200

Body-centered cubic (BCC)

Face-centered cubic (FCC)

Hexagonal-closed packed (HCP)

Figure 5.3: Crystalline structure of metals.

Each crystal-type pattern has specific properties. For instance, an FCC structure is more compact
(dense) and ductile than a BCC. Most metals will have only one pattern, but steel has two BCC and
FCC. In steel, BCC is called ferrite and FCC is called austenite. Common structural steels are BCC at
room temperature and change to FCC when heated above 723C. Welding, of course, deposits liquid
metal into the joint and melts some base material. Therefore, the weld deposit and the area around
the joint go through these changes in arrangement as the temperature rises during welding and falls
while cooling.
This ability to change arrangement (phase) while solid is what makes steel such a popular material
with which to work. While in the solid state, steel can be BCC or FCC. It is through this transformation
that different properties can be created. The material can be purchased in one condition, fabricated
into a useable shape and then have its properties changed completely through heat treatment.
To understand what really happens to steels when heated, metallurgists have developed a diagram
showing the relationship between temperature, structure transformation and chemistry of different
steels. This diagram is called a phase diagram.

201

5.4.1 Phase Diagrams (Iron-Carbon Equilibrium Diagram)

Basically, the welding operation rapidly heats a metal to a temperature higher than its melting point.
During the heating process, atoms absorb energy and expand. When the metal reaches the melting
point, it transforms into a liquid. When the heat source is removed, the process is reversed.
Solidification of the weld puddle (from liquid to solid state) produces the weld bead. Figure 5.4 shows
a simplified iron-carbon diagram. This diagram allows metallurgists to see how adding carbon changes
the response of the steel to temperature changes. Phase diagrams are sometimes called equilibrium
phase diagrams. These diagrams show what structures are most stable at a given composition and
temperature.
Phase diagrams are created by cooling the material very slowly and thereby allowing the most
preferred phases to form. During welding, cooling rates are much faster than the equilibrium diagram.

Figure 5.4: Iron-iron carbide phase diagram.

202

5.5

Effect of Heating and Cooling on Steel

Understanding the effect of heating and cooling on steel is important, not only because these effects
are used to enhance the properties of the steel as mentioned above, but also because the welding
operation involves similar effects, and final properties of the weld and its soundness can depend on the
rate at which the weld cools after the weld metal has been deposited.

5.5.1 Slow Cooling of Steel from Above 910C

From the previous sections, you have learned that:
g

In pure iron at temperatures above 910C, the atoms are arranged in a face centered cubic
(FCC) pattern or lattice. On slow cooling at 910C, the arrangement of the atoms changes
to a body centered cubic (BCC) lattice and stays like that on further cooling to room
temperature (see Figure 5.5).

Figure 5.5: Face centered and body centered cubes.

203

Under microscope, the microstructure of the pure iron at room temperature will show a large number of
grains, which look like soap bubbles viewed against a piece of glass (Figure 5.6). In each grain, the
atoms are arranged in the BCC pattern, but the orientation of the cubes is different in each grain. At
the surface where two grains meet, the orderly arrangement of the atoms is disturbed and this surface
is called a grain boundary (Figure 5.7).

Figure 5.6: Grains and grain boundaries.

Figure 5.7: Grain boundaries are

areas of mismatch.

The term ferrite phase describes metal grains having the BCC lattice structure; the main difference
between these and the soap bubble example is that the boundaries between metal grains are not
always straight.
At temperatures above 910C, iron with the FCC structure can dissolve more than 1 wt% carbon. The
carbon atoms, being smaller than the iron atoms, fit in the spaces between the larger iron atoms as
shown in Figure 5.8, and the overall crystal structure remains as FCC. If one were to examine the
steel at 920C under microscope, there will be no evidence of the carbon in the steel and one will
again see grains similar to pure iron ferrite at room temperature but, because of their different crystal
structure (FCC), these are called the austenite phase.

204

Figure 5.8: Carbon atoms dissolved in austenite.

Compared to the high-temperature austenite phase, the low-temperature ferrite phase can hold very
little carbon, and therefore the carbon has to come out during the slow cooling of the austenite phase.
This is how the layered structure called pearlite is formed.
Looking first at the example of slow cooling and
separating out of carbon in a steel containing 0.8 wt%
carbon, its microstructure at room temperature as seen
in a microscope, will again show a large number of
grains, but within each and every grain there will be
alternate layers (or lamellae) of ferrite (almost pure iron)
and iron carbide (a chemical compound of iron and
carbide, more commonly called cementite). This type of
structure (alternate layers of ferrite and cementite) is
called pearlite (Figure 5.9).
The pearlite phase in this case of 0.8% steel formed
from the austenite at 723C (Figure 5.9). One can
imagine that at this temperature two things occur. First,
the carbon chemically combines with iron to form
cementite (Fe3C) in each grain of austenite and at
various locations within each grain. Second, when all
the carbon is exhausted the remaining face centered
cubic iron (FCC austenite) changes to ferrite, the body
centered cubic (BCC) form.

205

Figure 5.9: Pearlite from slow cooling.

Steels containing intermediate levels of carbon behave in an intermediate manner. For

example, a steel containing 0.15 wt% carbon will have about 19% (0.15 / 0.8 x 100%)
pearlite phase and remaining 81% ferrite phase.

1200
1100
1000

Austenite

860

Austenite & Ferrite

Temperature C

BCC Ferrite 900

800

723

700
600
500
Ferrite & Pearlite
400
300

19 %

200Pearlite

81 % Ferrite

100 %
Pearlite

100
0
0

.20

.40

.60

.80

1.0

1.2

Carbon %
0.15% C
Steel

0.8% C
Steel

Figure 5.10: Structures formed on slow cooling.

The changes also occur over a range of temperatures (Figure 5.10), starting at about 860C (this
temperature will depend on the carbon content) and always finishing at 723C. In between these two
temperatures, an increasing amount of ferrite forms first as the temperature decreases until, at 723C,
there is about 19% austenite and 81% ferrite. Then, at 723C, the austenite changes to pearlite for the
0.8%C steel (Figure 5.10).
In slow-cooled steels (hot rolled or normalized), increasing the amount of carbon increases the amount
of pearlite in the microstructure and increases the tensile strength of steels.
206

5.5.2 Fast Cooling of Steel from Above 910C

You will recall that above 910C, the steels are in the face centered cubic (austenite) form. Now, when
the steel is cooled very fast from this temperature (hot steel quickly put in ice-cold water, i.e.,
quenched), the carbon atoms do not have the time to diffuse and form cementite. But the steel still
tries to change its crystal structure to a body centered cubic form. The result is that the carbon atoms
are trapped in the BCC crystal structure and distort the lattice (Figure 5.11).

Figure 5.11: Fast cooling of

steels from above 910C.

This distorted, body centered cubic phase is called martensite and its properties depend mainly on the
carbon content of the steel. The higher the carbon content of the steel, the more distorted the crystal
structure is and the resulting martensite becomes harder and stronger (higher strength) but also more
brittle. Figure 5.12 shows how the maximum hardness of the martensite changes with the steel
carbon content.

207

At cooling rates that are in

between slow cooling and
quenching in water, various
amounts of ferrite, pearlite,
martensite and some other phases
form. The main point is that
hardness (strength) varies in
between the extreme values for
the slow cooled and quenched
conditions of the steel, and higher
hardness values indicate greater
amounts of martensite present. In
welding we usually experience fast
cooling rates. Therefore carbon
content has a strong effect on
weld zone hardness and
consequently on the weldability of
the steel.

Figure 5.12: Hardness of martensite vs. carbon content.

5.5.3 How Fast a Cooling Rate is Fast Enough to Form Martensite?

This depends on the composition of the steel and can be judged from the hardening curve of the steel.
A hardening curve is a plot of the hardness of the steel when it is cooled at different rates over a wide
range. Figure 5.13 shows the effect of steels carbon content on the hardening curve. The Mn and Si
contents of three steels are assumed to be the same (1.2% Mn and 0.2% Si) and with 0.1% to 0.3% C,
these steels can be considered as typical of weldable structural and pressure vessel steels. It is seen
that an increase in carbon content from 0.1% to 0.3% increases the maximum hardness that is
possible for the steels at very high cooling rates.

208

600

0.3% C
(100% Martensite)

550
550

0.2% C
(90% Martensite)

Hardness, Vickers

450
400

0.1% C
(80% Martensite)

350
300
250

200

150

0%
Martensite

100
1

10

1000

100

Cooling Time Between 800oC and 500oC, (seconds)

High

Cooling Rate

Low

Figure 5.13: Hardening curves for three C-Mn structural steels.

Also, increasing carbon allows a given level of hardness to be achieved at a lower cooling rate. For
example, if one wanted the hardness not to exceed 400 HV, then one would be assured that there is
no danger of reaching or exceeding the allowable hardness level when welding the 0.1% C steel. But,
with increasing carbon content, the steel must cool more slowly in order to not exceed the maximum
hardness requirement.
The effect of other elements when added to a 0.2% C, 1.2% Mn, 0.2% Si steel is shown in Figure
5.14. You can see that these elements do not significantly increase the maximum hardness that is
possible at very high cooling rates. However, the hardening curve can become flatter, and to not
exceed a given level of hardness, the steel must be cooled progressively more slowly. In this regard,
Mo is most effective in increasing the hardening capacity since the hardening curve is the flattest, so
one can say that Mo increases the hardening capacity the most. Nickel on the other hand, does not
change the curve too much and therefore is the least effective in increasing the hardening capacity.

209

600
550

1.0% Mo

2.0% Mn

Hardness, Vickers

500

1.0% Ni

450

1.0% Cr

400
350

1.2% Mn

300
250
200
150
100
1

10

100

1000

Cooling Time Between 800oC and 500oC, (seconds)

High

Cooling Rate

Low

Figure 5.14: Hardening curves for steels with different amounts of Mn, Ni, Cr, Mo.

In the context of welding, martensite can form in the heat affected zone as a result of fast cooling of
the weld. If the steels carbon content is high, the martensite formed will be harder and brittle.
The resulting structure will be more prone to cracking in the presence of hydrogen coming from the
welding arc. Also, if large amounts of alloying elements are present, then martensite will form more
easily unless the cooling rate is controlled to be quite slow. It is in light of this background that the
weldability of the steel is said to decrease when its carbon content or the alloying element content is
high.

210

5.5.4 Heat Treatment of Structural Low Alloy and Quenched and Tempered Steel
Steel plates are always hot rolled in
the still mills. Earlier it was
mentioned that besides the hot
rolled condition, the steels may be
provided with a heat treatment after
normalizing or quenching. Various
heat treatments are illustrated in
Figure 5.15.
In both of these heat treatments, the
steel is first heated to a temperature
where the normal ferrite and pearlite
phases, present at room
temperature, change back to the
austenite phase (FCC structure).
This temperature for structural
steels is typically about 900C and
this part of the heat treatment is
called austenitizing.
If the steel is now taken out from the
Figure 5.15: Heat treatments.
furnace and allowed to cool in the
air, then it is said to have been
normalized. For structural steels, the microstructure of the normalized steels is generally similar to that
of the hot rolled steel (ferrite and pearlite) except that, due to the presence of such elements as
aluminum, vanadium, etc., the grain size of the normalized steel is smaller than that of the hot rolled
steel. Smaller grain size increases the strength and low-temperature toughness of the steel.
If instead, the steel is taken out of the furnace and immediately immersed in cold water, the steel is
said to have been quenched. The objective here is to obtain a hard, strong martensitic structure.
Since this structure also makes the steel brittle, the quenched steel is always tempered, by putting the
steel back into a furnace at about 550C to 650C. The tempering temperature has to be less than
723C to prevent transformation to austenite. If austenite begins to form again the effect of quenching
and tempering will be lost.
During fabrication the steel temperature should not exceed the temperature at which the
steelmaker tempered the steel since this will lead to a lower steel strength, possibly below the
minimum specified requirements.

211

The tempering step allows the trapped

carbon to come out of the distorted BCC
structure in the form of fine, round
cementite particles, leaving behind an
undistorted, fine-grained ferrite matrix
(Figure 5.16). The quenched and
tempered steel obtained is tougher than
the as-quenched martensite, but of lower
hardness and strength. Generally, the
higher the tempering temperature (still
below 723C), the greater the improvement
in toughness and reduction in strength.
Compared to the normalized steel of the
same composition, the quenched and
tempered steel should have higher
strength and toughness.
Figure 5.16: Spheroidized cementite
Remember that for the quenching and
in matrix of ferrite.
tempering treatment to be effective, one
should be able to achieve 100%
martensite as a result of the quenching.
As a result, the steel should have sufficient carbon and/or alloying element content to allow 100%
martensite at the cooling rate achievable during quenching. Since thicker steel will cool at a slower
rate than thinner steel in the water spray, it follows that thicker quenched and tempered steels require
greater amounts of alloying elements. (This is partly true for hot rolled and normalized steels as well,
in that thicker plates cool more slowly in the air and, therefore, have less strength than the thinner
plates. To compensate for this, thicker plates are likely to have slightly higher amounts of carbon or
other alloying elements.) In regards to welding, thicker material of the same designation presents
increased possibility of cracking due to greater hardenability as well as fast cooling.
A third heat treatment is called the annealing treatment. It is similar to normalizing in that the steel is
austenitized first but then cooled in the furnace itself rather than in air. The objective here is to control
the cooling rate to be even slower than cooling in air. Annealing treatment is used for steels that have
relatively high carbon and/or alloy element content so that even cooling in air is fast enough for the
steel to form at least some martensite. Therefore, to get a completely soft, martensite-free structure,
such steels need to be cooled in the furnace, i.e., annealed.

212

5.6

Alloying Elements in Steels

As mentioned before, the steels that you may be asked to

weld at different times are likely to belong to different
specifications, or even to the same specification, and will have
different compositions depending on the steel mill supplying
the steel. Alloy atoms are of a different size than the iron
atoms. Some are smaller, some are larger. In either case
their intended effect is to cause small distortions in the cubic
structure of the steel, as shown in Figure 5.17.
At this stage it is useful to note the intended functions of the
alloying elements, including carbon, that are added to the
steel, and then in later sections, we can consider their effects
on welding.

1)

Figure 5.17: Alloy elements

in FCC cube.

Carbon (C)

Increases the tensile strength of steels by increasing the amount of carbide present. Increases the
hardening capacity of the steel so that it may be effectively quenched and tempered.
Decreases the toughness of the steels. More so when present as lamellar (layered) cementite in
pearlite rather than round (globular/spheroidal) particles.

2)

Silicon (Si)

Added as a deoxidizer during steel melting. Increases strength. Moderate increase in hardening
capacity.

3)

Manganese (Mn)

Present in amounts up to 1.8 wt%. Combines with sulfur to form less harmful manganese sulfide
inclusions in high sulfur steels. Increases the steels strength but less than silicon. Increases the steels
toughness to some extent. Considerably increases the steels hardening capacity.

4)

Nickel (Ni)

Little effect on steels strength and hardening capacity but considerably improves its low temperature
toughness. Also increases the atmospheric corrosion resistance of the steel.

5)

Chromium (Cr)

Little effect on steels strength but increases the steels hardening capacity. Increases the steels
resistance to scale/oxide formation when heated to elevated temperatures. Also, combines with carbon
to form chromium carbides that are more stable than cementite, i.e., they do not break down with time at
elevated temperature applications. Chromium helps to maintain the steels strength and reduces its flow
(creep) at higher temperatures and for longer periods of time.
213

6)

Molybdenum (Mo)

Has a small effect in increasing the steel strength. Increases hardening capacity, slightly more than
chromium. Forms more stable carbide than cementite. Increases the steels resistance to deformation
(creep).

7)

Vanadium (V)

Forms carbides. Added for strength and toughness via grain refinement in as-rolled (control) as well as
normalized steels. Helps retain higher hardness and strength after tempering in quenched and
tempered steels. Also added in some steels meant for elevated temperature applications.

8)

Niobium (Nb)

Forms nitrides and carbides. Added for strength and toughness since a fine dispersion of niobium
carbides promotes grain refinement. It also helps retain fine grain size in the heat affected zones of
welds.

9)

Copper (Cu)

Added to increase the steel strength. The effects on toughness and hardening capacity are small.
Increases the atmospheric corrosion resistance of the steel. Total amounts of copper added are small
to prevent hot shortness.

10)

Boron (B)

Added to relatively low carbon steels in very small amounts to increase the hardening capacity of
steels meant to be quenched and tempered. A very strong strengthening agent when used in
combination with molybdenum, titanium or vanadium.

11)

Nitrogen (N)

Intentionally added only when other elements like vanadium are present so that vanadium nitrides can
improve strength and help refine the grain size.
In summary, and in order of decreasing effectiveness, various alloying elements are added to steel for
the following purpose:
Increased Strength:

C, Si, Cu, Mn, Mo (also Nb and V; their exact effect depends on other
factors such as the rolling temperature and time, amount of carbon and
nitrogen present, etc.)

Hardening Capacity:

C, Mn, Mo, Cr, Ni, Cu, B

Toughness:

Ni, grain refinement (achieved via the presence of Nb, V, Al , Ti)

Elevated Temperature
Properties:

Cr, Mo, V

Atmospheric Corrosion
Resistance:

Cu, Ni
214

5.7

How Does Hardness Effect Welding?

Hardness is a measure of the resistance of the

material to plastic deformation. Hardness is a
comparative measurement that uses a
standardized indenter to create an indentation in
the surface of the material. The size of the
indentation created is measured against a
standardized scale. Softer materials will exhibit a
larger indentation. A hardness test is described in
Figure 5.18.
A weld contracts as it cools. Hot weld metal is
much weaker than the surrounding parent
material. As the temperature of the weld area
drops its volume must decrease. Since it is
prevented from uniformly shrinking in three
dimensions, it must compensate in those
directions which are free to contract (Figure 5.19).
The atoms of the material must move in order for
the contraction to occur in a manner without
cracking.

Figure 5.18: Hardness test.

High hardness prevents the flow of the atoms

past one another increasing the likelihood of
cracking. For this reason we are concerned
with predicting and controlling weld zone
hardness.

Figure 5.19

215

5.8

Heat Affected Zone (HAZ)

Most people think that a weld is only the portion of metal fused during welding. This is a very limited
way of looking at a weld. A more accurate way is to consider a weld as the area affected by the heat
input during welding. According to this definition, a weld is composed of three main parts illustrated in
Figure 5.20:
g
g
g

fused zone (weld metal)

bonding zone (fusion line)
base metal heat affected zone (HAZ)

Weld Metal
A mixture of base metal and filler metal (when
used) combined during the welding process.

Fusion Line
A line or zone where the temperature was just
under the melting point of base metal.

Heat Affected Zone

The area of the base metal next to the weld that
does not melt but is changed by the heat from
the welding process. In a way, this area is heat
treated by the welding process, that is,
its mechanical properties have been altered.
In theory, the HAZ refers to all areas of the base metal
heated to above ambient temperature during welding.
In practice, the term HAZ is used to describe the areas
altered by welding heat input.

Figure 5.20

The width of the HAZ depends primarily on heat input and thermal conductivity (heat dissipation in
base metal). If heat input is decreased or thermal conductivity increased, the HAZ size will decrease.
This means that a weld made with SMAW process will normally produce a narrower HAZ than one
made with FCAW (using a large diameter electrode). Similarly, stainless steels will have a larger HAZ
than carbon steels, since the thermal (heat) conductivity is lower than steel.

216

Example
According to the iron-carbide phase diagram, the width of HAZ in carbon steels will
extend from the weld interface to where the temperature reached 723C. In
practice the HAZ will extend a bit further.

The iron-carbon phase diagram shows that a

phase transformation starts when the temperature
reaches 723C. At that temperature BCC
transforms into FCC. Since weld cooling rates
from temperatures above 723C may be rapid,
hardening of the weld area is common. The heat
affected zone (HAZ) is a very important area
because weld faults may occur in this zone
(Figure 5.21). A weld that contains a crack in the
HAZ is likely to fail in service. Cracks in the HAZ
are often small and difficult to detect.

Figure 5.21: Cracks in the HAZ.

Properties of the HAZ depend on:

g
g
g

type of base metal

welding process
welding procedure

Since different categories of steels behave differently to various heat treatments, the properties of the
HAZ will vary with the type of base metal. The welding procedure will affect the HAZ through the heat
input and cooling rate. Effects of welding on the HAZ are similar to heat treatments involving high
temperatures (as in annealing), and fast cooling rates (as in quenching).

217

As can be seen in Figure 5.22, the

transformations that take place in the HAZ
depend on the highest temperature attained
at each point of the zone during welding.
This figure illustrates what happens to a steel
that has been cold worked before welding.
This is the case for most rolled plates that did
not receive a heat treatment after rolling.
Where the temperature is minimal, grains
(deformed by rolling) will use the heat
provided by the welding process to
recrystallize. Long grains (deformed by
rolling) will transform into several smaller
grains. Areas where the temperature rises
above 723C will show the effects of phase
transformations (BCC FCC BCC). Near
the weld fusion line, where temperatures are
just below the melting point, very large grains
form. This is generally the weakest part of a
weld.
When it is important to limit grain growth in
the HAZ, the welder should be following strict
welding procedures and limit heat input by
using small (stringer) weld beads when
possible. Weaving is commonly used, but
should be limited to plain low-carbon steels
where heat treatments have lesser effects.
Figure 5.22

5.9

Weldability of Metals
Definition

The weldability of the steel is defined as the

ease with which it can be welded without
affecting the performance of the welded joint
in the intended application, that is, with
adequate properties and without harmful
defects.

Figure 5.23: Heat affected zone in fillet weld.

218

Students have previously studied the weld process/technique-related flaws. Their presence does not
really depend on the type of steel being welded and, therefore, these are not discussed further in this
chapter. Other elements of good weldability (mechanical properties and the absence of metallurgical
flaws) do however, depend on the type of steel being welded.
Prevention of hydrogen cracking, one of the potential metallurgical flaws that can be present in the
weld zone, is one of the most important considerations in designing welding procedures. Besides the
steel type/composition, two other factors determine if hydrogen cracking can occur in the weld joint.
These factors are: the rate at which the weld cools once it has been heated by the welding arc; and,
the presence of locked-in stresses.

5.9.1 Weld Cooling Rate

During welding, the steel next to the
molten weld pool (beyond the fusion
line) very nearly reaches the melting
temperature, but not quite. As one
moves further away from the fusion
line, the peak temperature becomes
less and less, until at some large
distance and beyond, no significant
rise in temperature occurs. Thus, at
the fusion line, the temperature
reaches more than 1350C. The
further from the fusion boundary, the
lower the peak temperature reached.
(Figure 5.24).
From the earlier discussion, we
know that there is a change in the
microstructure of any part of the
steel that gets heated above
approximately 700C. This region
next to the fusion boundary is called
the heat affected zone.
Figure 5.24: Heat Affected Zone (HAZ)

219

Also, from Section 5.5, we know that:

g

When the temperature reached in the heat affected zone is more than about 900C, the
steel changes to 100% austenite phase, that is, it is fully transformed. The width of this
region is rather small, from a fraction of a millimeter to a few millimeters depending on the
weld heat and workpiece thickness. Once the welding arc has passed by such a location,
the austenite phase formed changes to phases such as ferrite, pearlite, or martensite
depending on the steel composition (hardening capacity) and the cooling rate. The
properties (strength, hardness, toughness) of this part of the heat affected zone, also called
the supercritical heat affected zone, depend on the microstructure.

The region next to the supercritical heat affected zone that gets heated to a temperature
between about 700C and 900C is called the intercritical heat affected zone or the partially
transformed heat affected zone. The latter term indicates that the temperature did not
exceed about 900C in this region and therefore, the amount of austenite formed was less
than 100%, the other phase present being ferrite. Therefore, this region cannot form 100%
martensite on cooling.

The next region after the partially transformed or the intercritical heat affected zone is the
untransformed (no austenite formed at all) or the subcritical heat affected zone. The
maximum temperature reached in this region is about 700C. The microstructural changes
in this region can be hard to detect with an ordinary microscope. For quenched and
tempered steels, the region of the subcritical HAZ that reaches peak temperature above
the tempering temperature (say 620C) can suffer some reduction in strength. Also, in the
presence of microalloying elements (Nb, V), there is potential for some reduction in notch
toughness in the subcritical heat affected zone.

From the point of view of hydrogen cracking, it is the supercritical, fully transformed heat
affected zone next to the fusion boundary that has the highest hardness and highest tendency
to form hydrogen cracks. Whether hydrogen cracking would indeed occur or not in a given steel
depends partly on the exact microstructure which, in turn, depends on the steel composition and the
local cooling rate. If these two parameters are accurately known, it becomes possible to design a
welding procedure that will prevent hydrogen cracking. Since one generally knows the composition (or
at least the type) of steel being welded, at this stage it is important to understand what factors
determine the weld zone cooling rate.
The cooling rate indicates how fast the weld zone cools. Therefore, it is measured as the average
decrease in temperature of the weld zone (weld metal or the heat affected zone next to the fusion
boundary) in one second. A cooling rate of 70C/s is a much higher (or faster) cooling rate than
10C/s. Conversely, if one looks at the time to cool from 800C to 500C, then smaller cooling times
imply high cooling rate and larger cooling times imply slow cooling.

220

The cooling rate of the weld zone depends on the following three factors: weld heat, the thickness of
the steel and whether the steel has been preheated.
g

Weld Heat: also called the arc

energy, is the amount of electrical
energy (Figure 5.25) that is supplied
to the welding arc over a given weld
length (an inch or a mm). The greater
the weld heat (arc energy) used to
deposit the weld metal, the longer it
takes to remove the heat from the
weld and, therefore, the slower it cools.

Arc energy is calculated as follows:

Arc Energy =

where,

Arc Current x Arc Voltage x 60

Arc Travel Speed x 1000

arc energy is in kJ/mm (kJ/in)

current is in amperes
voltage is in volts
travel speed is in mm/min
(in/min)
(Note that 1kJ/mm = 25 kJ/in)

Figure 5.25: Arc energy or energy input.

In the above equation, Arc Current x Arc Voltage
is the electric energy being supplied to the arc in
one second (J/sec), and when this is divided by
the distance traveled in one second (travel speed in mm or inches per minute divided by 60), one
obtains the arc energy (Joules per mm or Joules per inch; a kJ is simply equal to 1000 J).
For example, if you use the SMAW process (E4918 electrode, 4 mm diameter) for depositing a weld
pass using the following parameters:
Current = 160 A; Voltage = 22 V; Travel Speed = 8 in/min (203 mm/min)

221

To slow down the weld cooling rate, you can increase the arc energy. This can be done by running the
arc hotter, i.e., increasing the welding current, or by decreasing the travel speed.
For example, increasing the current to 180 A will increase the arc energy to

and decreasing the travel speed to 6 in/min (2.5 mm/min) instead will increase the arc energy to:

Note that if you use a weaving technique instead of depositing stringer beads, then the arc travel
speed is reduced and the arc energy increases.
An important factor to note is that if you have the same weld heat or arc energy for the SMAW process
(5 mm diameter electrode, 220 A, 22 V, 6 in/min; arc energy = 48.4 kJ/in = 1.9 kJ/mm) and the SAW
process (3.2 mm diameter wire, 500 A, 30 V, 18.6 in/min travel speed, arc energy = 48.4 kJ/in = 1.9
kJ/mm), the cooling rate will not be the same in the two cases. This is because, in the open arc
processes (SMAW, FCAW, GMAW), some of the weld heat is lost to the surrounding atmosphere
whereas in the SAW process, almost all of the electrical energy gets into the weld zone as heat energy.
The energy that goes into the steel is called heat input and,
Heat Input = Arc Efficiency x Arc Energy
Arc efficiency takes into account the fraction of the arc energy that goes into the workpiece and is not
lost to the surrounding atmosphere. Submerged arc welding process has the highest arc efficiency
and gas tungsten the smallest. Different people use different values and, later on when preheat
requirements are estimated, the arc efficiency will need to be taken into account.
g

Thickness of Steel: the loss of heat from the weld zone to the surrounding steel is much
faster than to the surrounding air, therefore, one can intuitively see that for a fixed arc
energy (heat input), as the steel thickness increases, the heat is sucked out more quickly
and the weld zone cools faster, that is, the cooling rate increases.

Preheat: if the steel has been preheated first, then the cooling rate decreases again
because the hotter surrounding material has a reduced ability to pull heat from the weld
zone. However, the effect is greater in the low temperature range (less than 300C) and
rather small in the higher temperature range (500 to 800C) where the transformed
microstructures form.

222

Because of the several factors that affect the weld cooling rate, its calculation can be complex.
Fortunately, graphs have been developed that help to calculate the cooling rate. For example, Figure
5.26 shows one such scheme developed by Graville1 for the SAW process.

Figure 5.26: Graph to determine cooling rate in bead-on plate for submerged arc process.

Brian A. Graville, The Principles of Cold Cracking Control in Welds, (Dominion Bridge Company Ltd., 1975).

223

Let us consider two examples here:

g

In the first example, let us say that the SAW process is being used to make a butt joint in a
1 thick plate using the following parameters: 600 A, 30 V, 18in/min travel speed, 100C
preheat. Therefore, arc energy = 600 x 30 x 60 / 18 x 1000 = 60 kJ/in. Now, going to
Figure 5.27(a), we start at point A at 60 kJ/in at the bottom line and go up vertically to hit
the line for 1 thickness at point B; next, we go towards the right from point B until we
intersect the 100C preheat line at point C. Now go up vertically again and at point D, read
the cooling rate as about 11C/sec at 540C (1000F). The cooling rate is estimated at
540C because the development of microstructures like ferrite and pearlite occurs in the
temperature range of about 500C to 600C during the relatively fast cooling of welds and it
is the cooling rate in this temperature range that is important.

Figure 5.27(a)

224

In the second example, let us say that the SMAW process is being used to weld 0.5 thick
plate using the following parameters: 170 A, 22 V, no preheat (room temperature) and 8
in/min travel speed. Then, the arc energy =
170 x 22 x 60 / 8 x 1000 = 28 kJ/in
However, due to the open arc, not all this energy goes into the weld pool. Therefore, if
Figure 5.27(b) is to be used to calculate the cooling rate for processes like SMAW, FCAW
and GMAW, then the calculated arc energy for these processes should be multiplied by a
correction factor (arc efficiency) which can be taken as 2/3 for the SMAW process and 4/5
for FCAW and GMAW processes. Therefore, for the example at hand, heat input (= arc
energy x arc efficiency) will be
28 x 2/3 = 18.7 kJ/in
Now following the same procedure, one starts at point A, goes to point B and C and reads
out the cooling rate to be between 40 and 50C/sec at 540C.

Figure 5.27(b)

225

5.10

Solidification Cracking

Consider that the weld puddle solidifies like a small casting. The process of solidification starts with
the formation of several crystals (or dentrites) at the unmelted heat affected zone and continues as
these crystals grow towards the center of the puddle. Where two crystals meet, they form a grain
boundary and a sound weld should result (Figure 5.28).
However, in the presence of such elements as carbon, sulfur and phosphorous in the weld metal, small
amounts of liquid metal enriched in sulfur and phosphorous are trapped between the crystals before
the solidification is completed. As the weld metal shrinks further during cooling, a crack may form in
the region where the liquid was trapped. The liquid that solidifies last, near the grain boundaries, has a
lower melting point because of the impurities such as sulfur and phosphorous.

Figure 5.28: Solidified weld no hot cracking.

Solidification cracks are more common in welds that are deep and narrow (a submerged arc weld
deposited at a high travel speed) because it is easier for the liquid metal to get trapped between the
solidifying crystals (Figure 5.29). These cracks are also called centre-line cracks or hot cracks
because they form near the center of the weld nugget and when the weld metal is still hot. Cracks
seen in craters also form by the same mechanism, and are called crater cracks because of their
location.

226

The steel being welded affects the possibility of solidification cracking when it has high content of such
elements as carbon, sulfur and phosphorous and the welding procedure selected is such that there is
considerable dilution of the weld metal by the base material. Minimizing dilution, controlling weld bead
shape and employing lower arc energy are some of the approaches used to prevent solidification
cracking.

Figure 5.29: Solidification crack.

5.11

Strength and Toughness in the Weld Zone

It is difficult to predict the effect of the welding procedure (arc energy, welding consumable, pass
sequence, welding technique, etc.) on the strength and toughness of the various regions of the welded
joint, namely, the weld metal and heat affected zone. (One cannot easily measure the strength of the
heat affected zone and, therefore, it is more common to talk in terms of the hardness of the heat
affected zone). The fabricator usually performs a procedure qualification test, which demonstrates that
with the selected procedure, the minimum specified properties (heat affected zone and weld metal
toughness, if specified, maximum allowed heat affected zone hardness, strength in a cross-weld tensile
test) are achieved.
In this section, one can only point out that arc energy (more accurately, the heat input) is one of the
most important parameters that determines the properties of the heat affected zone and the weld metal
(Figure 5.30). What effect this will have depends on the composition of the steel or the weld metal.
Generally speaking:
g

at low arc energies, the weld metal and heat affected zone hardness (strength) tend to be
high; as the arc energy increases, the hardness and strength decrease.

227

if at low arc energy the hardness is too high, then the notch toughness of the heat affected
zone tends to be poor. Conversely, if the arc energy is too high, the grain size in the
microstructure becomes too large and this also reduces the toughness. The best level of
notch toughness will be obtained at intermediate levels of arc energy, .

the optimum level of arc energy for maximum notch toughness depends on the chemical
composition; as the hardening capacity increases, the optimum arc energy level will
increase. The preheat and interpass temperatures act in the same direction as the arc
energy but the effect is usually smaller.

Hardness

Strength
or
Hardness

Toughness

Toughness

Low

Energy Input

High

High

Cooling Rate

Low

Increasing
Hardening
Capacity

Toughness

Low

Energy Input

High

High

Cooling Rate

Low

Figure 5.30: Cooling rate effect on properties.

228

5.12

Hydrogen Cracking

It is known that a certain amount of hydrogen is usually present in the weld pool. This comes from the
breakdown of moisture that is generally present in fluxes (electrode coating, submerged flux, filling in
flux cored wires) and that may also be present in shielding gases. Occasionally, high humidity on
certain days can also increase the amount of hydrogen that might be introduced into the weld pool.
At room temperature, hydrogen is known
to affect the properties of steels, basically,
hydrogen embrittles the steel and reduces
its ductility. In the case of welds, the
hydrogen may also lead to the formation
of cracks in the heat affected zone or the
weld metal. If cracks do form, then some
of their typical locations are shown in
Figure 5.31.
Such cracks are completely unacceptable
in welds and all precautions must be
taken to ensure their absence. Their
importance can be judged from the fact
that sometimes weldability is narrowly
defined as the ease with which steels may
be welded without the formation of
hydrogen cracks. (Hydrogen cracking is
also called:
g

cold cracking because cracks

form only when the weld has
cooled down to below about
100C; and

delayed hydrogen cracking

because cracks can form
several hours or days after weld
completion.)

Figure 5.31: Hydrogen embrittlement.

229

Locations of Hydrogen Cracks

Possible Solutions

Reduce sources of hydrogen

Use preheat, check if properly applied and maintained
Increase preheat and/or interpass temperature
Increase welding energy, (soften weld zone)
For multiple-pass welds, increase interpass time while maintaining
interpass temperature
Consider slowing cooling rate after weld completion or postweld heat
for thick welds
Minimize all fit-up gaps (to < 1/16")
Reduce joint rigidity by assembly or weld sequence
For cracks that appear in the weld metal only, consider the use of lowerstrength electrodes, subject to the owner's approval
Ensure tacks incorporated in final weld are proper size and not cracked

Figure 5.32: Cracks in and around welds.

5.12.1 Factors Affecting the Formation of Hydrogen Cracks

The tendency to form hydrogen cracks depends on the following factors:
g

Amount of hydrogen present in the weld pool: The greater the amount of hydrogen
present in the weld pool, the greater the chance of forming hydrogen cracks. The amount
of hydrogen introduced into the pool depends on such factors as the welding process used;
the design of the welding consumable and its storage conditions; and the presence of
moisture, oil, grease, etc. on the workpiece to be welded. Generally, the GTAW and
GMAW introduce the smallest amount of hydrogen into the weld pool and these are called
low-hydrogen processes.

230

The amount of hydrogen introduced in SMAW and FCAW processes will depend on the
electrode designation, electrode manufacturer and the conditions under which the
electrodes have been stored. For example, E4918 electrodes can be provided in
vacuum-sealed packaging and these electrodes should introduce a very low amount of
hydrogen into the weld pool. However, if after opening the package and before use, the
electrodes are allowed to stay in the open in a high humidity environment for a period of a
few hours or days, then the amount of hydrogen getting into the weld pool will be higher
and this will increase the chances of hydrogen cracking.
g

Locked-in stresses present: The higher the magnitude of the locked-in stresses, the
easier it is for the hydrogen cracks to form. Residual stresses (weld zone shrinkage
against the colder steel) are always present in welds. In addition, stresses may be present
due to high restraint (the workpiece is too rigid to move).
Also, if notches are present, then stresses are magnified at these locations and hydrogen
cracks can form more easily. Some such locations include the root pass in a groove weld,
one-sided weld on backing bar or unspliced backing bar for a longitudinal weld (Figure
5.32).

Steel hardness/microstructure: Generally speaking, the harder the heat affected zone,
the greater its tendency to form hydrogen cracks. A harder microstructure means that it
has a smaller proportion of ferrite and more of martensite-like phases. Depending on the
hydrogen content and stress, the hardness above which hydrogen cracking may occur
varies from 300 to 400 HV. Whether a hard heat affected zone forms on welding or not
depends on the steels hardening capacity, which in turn depends on the composition of the
steel (amount of C, Mn, Cr, Ni, Mo, etc.) and the rate at which the weld cools. As
mentioned in the previous Section, the rate at which the weld cools depends on the arc
energy, steel thickness and preheat.

5.12.2 Avoiding Hydrogen Cracking

Once a steel has been selected and purchased for welding, the options available to counter the
possibility of hydrogen cracking include:
g
g
g
g
g
g

minimize weld joint restraint

avoid notches in the area of the weld
use a low hydrogen process
use low hydrogen consumables and ensure their proper storage
use high arc energy to reduce the cooling rate (but this may reduce other properties such
as strength and toughness)
use preheat (and post heat); its main function is to slow down the cooling rate below
100C and give more time for hydrogen to diffuse out.

231

It is obvious that not all of the above factors can be controlled easily and that some of the above steps
may entail additional costs. Therefore, one has to be sure of the need and effectiveness of the above
steps. This depends on the type of the steel being welded and can be judged from the zone diagram
developed by Graville2 shown in Figure 5.33.
In this Figure, the vertical scale is the carbon content of the steel. You will recall that the carbon
content determines the maximum hardness that is possible for the heat affected zone if it is cooled fast
enough. The higher the carbon content, the higher the HAZ hardness possible and the greater the
likelihood of hydrogen crack formation.

0.40

0.30

Zone II

C (wt %)

Zone III

0.20

0.10

Zone I
0.0
0.20

C.E. = C +

0.30

040

0.50

0.60

0.70

Mn + Si
+ Ni + Cu + Cr + Mo + V
6
15
5

Figure 5.33: Zone diagram for classifying steels based on their weldability.

The horizontal axis in the diagram is a steel composition factor (C.E. = C + (Mn+Si)/6 + (Ni+Cu)/15 +
(Cr+Mo+V)/5) called the carbon equivalent. It includes the carbon content of the steel as well as other
elements that affect the hardening capacity of the steel. The higher the carbon equivalent of the steel,
the greater its hardening capacity, and greater the hardening capacity, the easier it becomes to get
high hardness in the heat affected zone at slower cooling rates.

B.A. Graville, The Principles of Cold Cracking Control in Welds (Dominion Bridge Co., 1975).

232

The possibility of hydrogen cracking and the suitable means to avoid it can then be obtained from the
location of the steel on this Zone diagram. For example, for steels that fall within Zone I, the carbon
content is usually less than 0.10 wt% so that even if the hardening capacity is high (high carbon
equivalent), the maximum hardness possible in the heat affected zone, even at fast cooling rates, is
relatively low, typically less than 300 HV. (See the hardening curve for the 0.1% C steel in Figure
5.13.) Therefore, the possibility of hydrogen cracking is small. There should be no need for preheat
(unless the thickness is very large), and good welding practices (control of consumables) should be
enough to prevent hydrogen cracking.
Within Zone II, the carbon content is greater than 0.1 wt% and therefore, the maximum possible
hardness achievable (fast cooling rate) in the HAZ is high. But the addition of alloying elements that
increase the hardening capacity, and therefore the carbon equivalent, is limited (see the hardening
curve for the 0.3% C steel in Figure 5.13). Therefore, if the weld can cool slowly (small workpiece
thickness, high arc energy), then the maximum possible hardness is not achieved. In fact, the
hardness may be sufficiently low for thin plates/high arc energies so that no preheat is required to
prevent hydrogen cracks.
But it should be kept in mind that increasing the arc energy can have undesirable side effects such as
reduced strength and toughness. Also, as the plate becomes thicker, it becomes difficult to slow down
the cooling rate and then one must minimize the hydrogen content and/or use some preheat.
Within Zone III, the carbon content is greater than 0.1 wt% and, in addition, sufficient alloying elements
are present (high carbon equivalent) so that high hardness values are obtained even at slow cooling
rates (see the hardening curve for the 1%Mo steel in Figure 5.14). Consequently, heat input control
cannot be used to prevent hydrogen cracking. Therefore, one has to focus on minimizing the initial
hydrogen content and its removal by preheat and occasionally, by postweld heat (i.e., maintaining a
temperature of about 100C to 200C for a length of time after completion of welding, depending on the
thickness of the weld/steel plate).

233

Figure 5.34: Location of different types of steels in the zone diagram

(reproduced from ASM Handbook, Vol. 6, on Welding, Brazing and Soldering, 1993).

5.13

Heat Treatment of Steels

Heat treatment can be defined as an operation or combination of operations involving the heating and
cooling of a metal or alloy in the solid state. Steel properties can more easily be controlled by heat
treatment than by mechanical work.
By heat treatment, steel can be made strong
and hard, or it can be made soft and ductile.
By varying the carbon and alloy contents, and
the heat treatment of steels, a wide range of
mechanical properties can be produced.
Since alloyed steels are more expensive than
plain carbon steels, they are usually heat
treated to take full advantage of their
properties.
What is a heat treatment? Heat treatments
basically consist of a three-step process:
g

heating the steel to a specific

temperature

maintaining the steel at that

temperature for a certain length of
time

cooling the steel at a specific rate

Figure 5.35: Typical heat treatment cycle.

234

Heat treatments, with a few exceptions, always involve some phases and/or grain transformations.
Heat treatments may be subdivided into two broad categories:
Conventional Heat Treatments
g
g
g
g
g

annealing
normalizing
quenching
tempering
stress relieving

Special Heat Treatments

g
g
g

flame hardening
hot shots
case hardening

Since this is a vast subject, we will concentrate on conventional heat treatments. As mentioned
previously, heat treatments, when applied to heat treatable steels, will modify steel properties to
regenerate some properties or to improve existing ones.

Example
Increase hardness
Softening
Relaxing stresses

5.13.1

Quenching
Annealing
Stress relief

Annealing

Annealing is most often a softening process, where steel is heated to an elevated temperature, held for
a certain time at this temperature, and allowed to cool slowly to room temperature.
In annealing, sufficient time (approx. 1 hour per 25 mm thickness) has to be allowed at the specific
temperature to ensure complete transformation to austenite (FCC). Slow uniform heating and cooling
are desirable. Furnace cooling is typically used.

235

5.13.2

Normalizing

Normalizing is similar to annealing except that the rate of cooling is increased by allowing steel to cool
in the air instead of in a furnace. Normalizing is used to control grain size and lessen residual
stresses. Normalized steels are harder and have higher strengths than steels that have been
annealed.

5.13.3

Quenching

Quenching is probably the most common and well-known heat treatment. Quenching can be described
as an operation that provides for rapid cooling of steel from the austenitic temperature (FCC) to lower
temperatures such as room temperature. If cooling is rapid enough, steel will become much harder
and stronger. Different rates of cooling can be obtained by immersing the piece in air, water, oil, brine
and molten salts or molten metals.
Quenching is particularly useful for tools that must be hard and that must maintain their sharpness
under severe conditions. Note that maximum hardness is generally accompanied by brittleness. To
optimize mechanical properties, applying a subsequent heat treatment is often necessary. The
treatment is called tempering.

5.13.4 Tempering
Quenched steels exhibit a wide range of mechanical properties. Hardness, tensile and yield strength,
and brittleness will be very high. On the other hand, toughness and ductility will be much lower.
Tempering is an operation designed to modify steel properties resulting from quenching. Tempering is
essentially a reheating process and is always done at temperatures where no structure change occurs.
Its usual purpose is to increase toughness, reduce brittleness and alleviate high internal stresses.

5.13.5

Stress Relief

Stress relief is the heating of steel to a temperature below the transformation temperature, as in
tempering, but it is done primarily to relieve internal stresses and to prevent distortion or cracking
during machining.
When a metal is heated, expansion occurs. Upon cooling, the reverse reaction takes place and
contraction is observed. In welding, when a part is heated more at one point than at another, internal
stresses develop. Internal, or residual stresses, are bad because they can generate warping during
machining. To relieve stresses, steel is heated uniformly and cooled slowly to room temperature.

236

5.13.6 Concept of Hardening

Some words that are universally understood are often misused. Hardening is one of these words. We
frequently use the word hard to describe something that is firm or solid. In peoples mind a hard
substance will not wear easily. This is only partially true and to understand what hardness really
means, we will look at how hardness is achieved.
The hardening process is often associated with heat treatments like quenching or aging. The process
is described as the increasing of hardness by suitable treatments, usually involving heating and cooling
or cold working.
Hardness, in fact, is a measure of the resistance of a material to plastic deformation usually by
indentation. Plastic deformation is a change in shape (no matter how small), which will remain
permanent after removal of the force which caused it. The term may also refer to stiffness or temper,
or to resistance to scratching, abrasion or cutting. Indentation hardness may be measured by various
tests such as Brinell (B), Rockwell (HR) and Vickers (HV).
Hardness testing methods measure the size of an indentation made in the surface of a material. The
indentation size made with the same load and indentor is compared (soft material has a large indent,
hard material a small indent).
Hardness is achieved by a hardening process, and the effects of this treatment will depend on the
grade of steel being treated. The response of a given steel to a hardening treatment is called
hardenability.

Hardenability

The relative ability of a ferrous alloy to form martensite

when quenched from high temperatures.

Hardenability is closely related to the formation of a hard microstructure called martensite. Martensite
is the hardest steel microstructure. It is the result of rapid quenching from above the transformation
temperature (723C). As discussed previously, when the transformation from FCC to BCC is forced to
occur quickly, carbon and alloy elements cannot separate from the material to make pearlite; they will
create a distorted body-centered phase called martensite.
Hardness is often considered as a good indicator of wear resistance. This is only partially true, since
wear may take many forms such as grinding wear, sharp particle wear or friction wear. One has to be
very careful to not automatically select the hardest material for a given wear action. Hardness is also
associated with brittleness. Except in a few situations, brittleness normally increases when hardness
increases.

237

Example
Steel with a hardness of 50 HRc is more brittle than a steel with a hardness
of 20 HRc. HRc is a commonly used hardness scale called Rockwell C
Hardness.

In earthmoving equipment, a combination of hardness and toughness is often required. This is

achieved by alloying steels with manganese. Hadfield or manganese steels are very hard on the
surface (martensite) and soft inside (austenite).

5.13.7 Ways to Harden Steel

Three main ways to harden steel are:

g
g

introduction of alloying elements

(Figure 5.36)
mechanical deformation (cold work)
heat treatments

1)

Alloying Elements

The introduction of alloying elements to the crystalline

patterns (such as BCC or FCC) will deform the pattern
and harden the metal. Carbon is one of the main alloying
elements because it is cheap and has a tremendous
impact on hardness and strength. Carbon has a dual
effect on steel, as it fixes the maximum attainable
hardness and contributes substantially to determine the
hardenability.
Several other alloying elements are manganese (Mn),
silicon (Si), chromium (Cr), and nickel (Ni). The most
important function of these elements, in heat treatable
steels, is to increase hardenability, making the hardening
of large sections possible while using moderate quenching
methods.

238

Figure 5.36: Positions of

alloying elements.

5.13.8 Cold Work (Mechanical Deformation)

Cold working is deforming a metal plastically, at a temperature below the recrystallization temperature.
During cold work (such as during the
rolling of the plate), hardening is
produced by severe plastic deformation.
Cold work increases hardness, yield
strength, and tensile strength and
lowers ductility.
Hardness and elongation react
differently to work hardening. As cold
work increases, hardness increases
and elongation decreases down to a
minimum after which the piece will
break. This is what happens when a
wire is broken after repeated bending in
the same place, or when it is formed
over a radius that is too small, as in
brake press work.
Figure 5.37: Plate rolling.

5.13.9

Heat Treatments

Hardening steels by heat treatment can only to accomplished if the steel has a suitable alloy content.
Some steels, like plain low-carbon steels, do not have enough alloying to respond to standard
hardening treatments.
Requirements for hardening steels by heat treatment:
g
g
g

sufficient carbon content in the steel

steel must first be completely austenitized (FCC)
austenitized steel must be cooled rapidly to a temperature range at which hard phases
are formed (before pearlite can form)

Quenching is the most common hardening heat treatment. When steel is quenched, carbon and other
alloying elements are trapped in areas where there is not enough space. This produces a deformed
structure that can be associated with bainite or martensite. These structures are called hard phases or
hard constituents.

239

5.14

Influence of Welding on Mechanical Properties

In previous sections, we mentioned that mechanical properties could be affected by heat treatments
and by welding. Before talking about the effects of welding on mechanical properties, it is necessary to
define what mechanical properties are.
Mechanical properties are the features of a material that show how it responds to force. They are a
good indication of the materials suitability for mechanical applications.

Example
g
g
g
g

tensile strength (ultimate strength)

yield strength
elongation
hardness

Mechanical properties, like hardness, can be changed by mechanical work, with the addition of alloying
elements and by heat treatment. Surprisingly, mechanical properties are often mistaken for physical
properties.
Physical properties are properties of a metal or alloy that are insensitive to structure and can be
measured without the application of force.

5.15

Designation of Steels

There are multiple grades of steels to suit numerous

service demands. They are grouped into four major
categories as shown in Figure 5.38.

STEEL

Steel can be classified according to the

carbon content and the type of alloying
elements added.

Carbon Steel

Alloy Steel

Tool Steel

Figure 5.38

240

Stainless Steel

5.15.1 Carbon Steel

Carbon is the principal alloying element.
low-carbon
medium-carbon
high-carbon

0.01% to 0.30% C
0.30% to 0.45% C
0.45% to 1.70% C

(Medium and high-carbon steels can be heat treated.)

5.15.2

Alloy Steel, Tool Steel and Stainless Steel

To improve specific steel properties, quantities of elements such as chromium, nickel, molybdenum and
vanadium are added. The resulting steel is much stronger than plain carbon steels, but is more
expensive. Stainless steels are alloy steels that exhibit high corrosion resistance. High alloy steels are
often called tool steels.
Having a good understanding of the properties of different steels is important because each category of
steel requires specific welding procedures. Since carbon and low alloy steel represent more than 80%
of all welded steel, we will focus our attention on these.

5.16

Classification of Steels (Numbering System)

Several codes classify steel according to chemical composition, applications and/or mechanical
properties.
The common numbering systems used in North America are:
CSA G40.21
SAE
AISI
ASTM

Canadian Standards Association

Society of Automotive Engineers
American Iron and Steel Institute
American Society for Testing Materials

241

5.16.1

CSA G40.21 Canadian Standards Association

This specification normally refers to structural steels. Eight different types of steel are produced under
this classification.
Table 5.1: Types of steel.

G
W
WT
R
A
AT
Q
QT

General construction steel

Weldable steel
Weldable notch toughness steel
Atmospheric corrosion resistant steel
Atmospheric corrosion resistant weldable steel
Atmospheric corrosion resistant weldable notch toughness steel
Quenched and tempered low alloy steel
Quenched and tempered low alloy notch toughness steel

The eight types and seven strength levels have been combined into twenty-two grades:

Table 5.2: Grades of steel.

Type

G
W
WT
R
A
AT
Q
QT

Yield Strength, MPa (ksi)

230
(33)

260
(38)

300
(44)

350
(50)

400
(60)

480
(70)

700
(100)

230G
--------

-260W
260WT
------

-300W
300WT
------

350G
350W
350WT
350R
350A
350AT
---

400G
400W
400WT
-400A
400AT
---

-480W
480WT
-480A
480AT
---

------700Q
700QT

5.16.2 SAE AISI

The SAE-AISI numbering system normally consists of
four digits. The first two digits (e.g., 86) provide
information about the elements used as alloys. The
last two digits refer to the percentage of carbon in the
steel in hundredths of 1 percent. (e.g., 20 means
0.20% C). For example, AISI 8620:

242

86

20

alloy
content

carbon
content

Table 5.3: Designation system for AISI and SAE steels.

AISI or SAE Number

10xx
11xx
13xx
23xx
25xx
31xx
33xx
40xx
41xx
43xx
46xx
48xx
51xx
61xx
81xx
86xx
87xx
92xx
xx
Mn
B
C
E

Composition
Plain carbon steel
Plain carbon (resulfurized for machinability)
Manganese (1.5% 2.0%)
Nickel (3.25% 3.75%)
Nickel (4.75% 5.25%)
Nickel (1.10% 1.40%), chromium (0.55% 0.90%)
Nickel (3.25% 3.75%), chromium (1.40% 1.75%)
Molybdenum (0.20% 0.30%)
Chromium (0.40% 1.20%), molybdenum (0.08% 0.25%)
Nickel (1.65% 2.00%), chromium (0.40% 0.90%),
molybdenum (0.20% 0.30%)
Nickel (1.40% 2.00%), molybdenum (0.15% 0.30%)
Nickel (3.25% 3.75%), molybdenum (0.20% 0.30%)
Chromium (0.70% 1.20%)
Chromium (0.70% 1.10%), vanadium (0.10%)
Nickel (0.20% 0.40%), chromium (0.30% 0.55%),
molybdenum (0.08% 0.15%)
Nickel (0.30% 0.70%), chromium (0.40% 0.85%),
molybdenum (0.08% 0.25%)
Nickel (0.40% 0.70%), chromium (0.40% 0.60%),
molybdenum (0.20% 0.30%)
Silicon (1.80% 2.20%)

Carbon content, 0.xx%

All steels contain 0.50% + manganese
Prefixed to show bessemer steel
Prefixed to show open-hearth steel
Prefixed to show electric furnace steel

243

5.16.3 ASTM
ASTM classification is widely used for structural and pressure vessel steels. In this classification,
steels are given a reference number, for example

Example

ASTM A-36
ASTM A-285
ASTM A-516
ASTM A-572

The number refers to a set combination of chemical composition and mechanical properties. Some
ASTM steels are comparable to Canadian steels. For instance, Grade G40.21-300W can be used as a
substitute for ASTM A-36. The ASTM number is sometimes followed by a grade number (ex. ASTM A572 Grade 42 or 50). Here, different Canadian grades have to be selected. Grade G40.21-300W can
be considered as equivalent to ASTM A572 Grade 42 and Grade G40.21-350W will be used as an
equivalent to ASTM A572 Grade 50.
The ASTM publishes specifications of various special purpose steels, which are updated regularly.
Detailed information of ASTM designated steel can be found in its individual specification.

244

Table 5.4: Index of steel specifications in welding procedure table.

ASTM
Steels
A27-83
A36-81
A53-83
A105-83
A106-83
A108-81
A120-83

1
2
3
5
6
7
8

A131-82
A134-80
A135-83
A139-74
A148-83
A161-83

9
10
11
12
13
14, 18

A176-83
A178-83

37
15

A179-83

16

A181-83
A182-82

17
18,22,23,
26,28,30,
32,35,36,
37
38
39
40

A184-79
A185-79
A192-83
A199-83
A200-83
A202-82
A203-82
A204-82
A209-83
A210-83
A213-83

A214-83
A216-83
A217-83
A225-82

Item
No.

26,27,28
29,30,32,
35,36
26,27,28,
29,30,32,
35,36
41
42
18
18
43
22,23,25
26,27,28,
30,32,33,
34,35,36
44
45
34,36,37,
46
47

Specification Title
Steel castings, carbon, for general application
Structural steel
Pipe, steel, black and hot-dipped, zinc-coated welded and seamless
Forgings, carbon steel for piping components
Seamless carbon steel pipe for high-temperature service
Steel bars, carbon, cold-finished, standard quality
Pipe steel, black and hot-dipped zinc-coated (galvanized) welded and seamless for
ordinary uses
Structural steel for ships
Pipe steel, electric-fusion (arc) welded steel plate pipe (sizes 16 inch and over)
Electric-resistance-welded steel pipe
Electric-fusion (arc) welded steel pipe (sizes 4 inch and over)
Steel castings, high-strength, for structural purposes
Seamless low-carbon and carbon-molybdenum steel still tubes for refinery service
(2 inch 9 inch outside diameter)
Stainless and heat-resisting chromium steel plate, sheet and strip
Electric-resistance-welded carbon steel boiler tubes (1/2 inch 5 inch outside
diameter)
Seamless cold-drawn low-carbon steel heat-exchanger and condenser tubes (1/8
inch 3 inch outside diameter)
Forgings, carbon steel, for general purpose piping
Forged or rolled alloy-steel pipe flanges, forged or rolled alloy-steel pipe flanges,
forged fittings and valves and parts for high-temperature service
Fabricated deformed steel bar mats for concrete reinforcement
Welded steel wire fabric for concrete reinforcement
Seamless carbon steel boiler tubes for high-pressure service (1/2 inch 7 inch
outside diameter)
Seamless cold-drawn intermediate alloy-steel heat-exchanger and condenser tubes
Seamless intermediate alloy steel still tubes for refinery service
Pressure vessel plates, alloy steel, chromium-manganese-silicon
Pressure vessel plates, alloy steel, nickel
Pressure vessel plates, alloy steel, molybdenum
Seamless carbon-molybdenum alloy-steel boiler and superheater tubes
Seamless medium carbon steel boiler and superheater tubes (1/2 inch 5 inch
outside diameter)
Seamless ferritic and austenitic alloy-steel boiler, superheater and heat-exchanger
tubes
Electric-resistance-welded carbon steel heat-exchanger and condenser tubes
Steel castings, carbon, suitable for fusion welding for high-temperature service
Steel castings, martensitic stainless and alloy for high-temperature service
Pressure vessel plates, alloy steel, manganese-vanadium-nickel

245

ASTM
Steels
A226-83
A234-82

Item
No.
48

A240-83

18,23,26,
28,32,35,
36,49
37

A242-81
A250-83

50
18

A252-82
A266-83
A268-83
A276-83
A283-81
A284-81

51
52
37
37
53
54

A285-82
A299-82
A302-82

55
56
57

A311-79
A321-81
A322-82
A328-81
A331-81
A333-82
A334-83
A335-81

A350-82

58
59
60
61
60
62
63
18,20,22,
23,26,28,
30,32,33,
34,35,36
18,19,23,
26,28,30,
31,32,36,
37
64

A352-83

65

A353-82
A356-83

66
18,21,22,
24,26,28,
67
18,22,23,
26,27,28,
30,32,35,
36,68
69
70

A336-83

A369-79

A372-82
A381-81

Specification Title
Electric-resistance-welded carbon steel boiler and superheater tubes for highpressure service (1/2 inch 5 inch outside diameter)
Piping fittings of wrought carbon steel and alloy steel for moderate and elevated
temperatures
Heat-resisting chromium and chromium-nickel stainless steel plate, sheet and strip
for pressure vessels
High-strength low-alloy structural steel
Electric-resistance-welded carbon-molybdenum alloy steel boiler and superheater
tubes
Welded and seamless steel pipe piles
Forgings, carbon steel, for pressure vessel components
Seamless and welded ferritic stainless steel tubing for general service
Stainless and heat-resisting steel bars and shapes
Low and intermediate tensile strength carbon steel plates, shapes and bars
Low and intermediate tensile strength carbon-silicon steel plates for machine parts
and general construction
Pressure vessel plates, carbon steel, low and intermediate tensile strength
Pressure vessel plates, carbon steel, manganese-silicon
Pressure vessel plates, alloy steel, manganese-molybdenum and manganesemolybdenum-nickel
Stress-relieved, cold-drawn carbon steel bars subject to mechanical properties
Steel bars, carbon, quenched and tempered
Steel bars, alloy, standard grades
Steel sheet piling
Steel bars, alloy, cold-finished
Seamless and welded steel pipe for low-temperature service
Seamless and welded carbon and alloy steel tubes for low-temperature service
Seamless ferritic alloy steel pipe for high-temperature service

Steel forgings, alloy, for pressure and high-temperature parts

Forgings, carbon and low-alloy steel, requiring notch toughness testing for piping
components
Steel castings, ferritic and martensitic, for pressure containing parts suitable for lowtemperature service
Pressure vessel plates, alloy steel, 9% nickel, double-normalized and tempered
Steel castings, carbon and low-alloy, heavy-walled, for steam turbines
Carbon and ferritic alloy steel forged and bored pipe for high-temperature service

Carbon and alloy steel forgings for thin-walled pressure vessels

Metal-arc welded steel pipe for use with high-pressure transmission systems

246

ASTM
Steels
A387-83
A389-83
A405-81
A420-83
A423-83
A426-80

Item
No.
22,23,26,
28,30,32,
35,36
24,26

A486-82
A487-83
A498-68

24
71
72
18,20,22,
23,26,28,
30,32,33,
35,36,37
73
74
75
76
32,35,36,
37
77
37,78
80

A500-82

81

A501-83
A508-81

82
83

A511-79
A512-83
A513-82
A514-82
A515-82

37
84
85
86
87

A516-83
A517-82
A519-82
A521-76
A522-81

88
89
90
91
66

A523-81

92

A524-80
A529-82
A533-82

93
94
95

A537-82
A541-81

96
26,28,97

A434-81
A441-81
A442-82
A455-82
A473-82

Specification Title
Pressure vessel plates, alloy steel, chromium-molybdenum
Steel castings, alloy, specially heat-treated for pressure-containing parts suitable for
high-temperature service
Seamless ferritic alloy-steel pipe specially heat treated for high-temperature service
Pipe fittings of wrought carbon steel and alloy steel for low-temperature service
Seamless and electric-welded low-alloy steel tubes
Centrifugally cast ferritic alloy steel pipe for high-temperature service

Steel bars, alloy, hot-wrought, or cold-finished, quenched and tempered

High-strength low-alloy structural manganese-vanadium steel
Pressure vessel plates, carbon steel, improved transition properties
Pressure vessel plates, carbon steel, high-strength manganese
Stainless and heat-resisting steel forgings
Steel castings for highway bridges
Steel castings suitable for pressure service
Seamless and welded carbon ferritic, and austenitic alloy steel heat exchanger
tubes with integral fins
Cold-formed welded and seamless carbon steel structural tubing in rounds and
shapes
Hot-formed welded and seamless carbon steel structural tubing
Quenched and tempered vacuum-treated carbon and alloy steel forgings for
pressure vessels
Seamless stainless steel mechanical tubing
Cold-drawn butt weld carbon steel mechanical tubing
Electric-resistance-welded carbon and alloy steel mechanical tubing
High-yield-strength, quenched and tempered alloy steel plate, suitable for welding
Pressure vessel plates, carbon steel for intermediate and higher-temperature
service
Pressure vessel plates, carbon steel, for moderate and lower-temperature service
Pressure vessel plates, alloy steel, high-strength, quenched and tempered
Seamless carbon and alloy steel mechanical tubing
Steel, closed-impression die forgings for general industrial use
Forged or rolled 9% nickel alloy steel flanges, fittings, valves and parts for lowtemperature service
Plain end seamless and electric-resistance-welded steel pipe for high pressure
pipe-type cable circuits
Seamless carbon steel pipe for atmospheric and lower temperatures
Structural steel with 42 ksi minimum yield point (1/2 inch maximum thickness)
Pressure vessel plates, alloy steel, quenched and tempered, manganesemolybdenum and manganese-molybdenum-nickel
Pressure vessel plates, heat treated, carbon-manganese-silicon steel
Steel forgings, carbon and alloy, quenched and tempered, for pressure vessel
components

247

ASTM
Steels
A542-82
A543-82

Item
No.
28
98

A553-82
A562-82

66
99

A572-82
A573-81
A575-81
A576-81
A587-83
A588-82
A589-83
A592-74

100
101
102
103
104
105
106
107

A594-69
A595-80
A612-82

108
109
110

A615-82
A618-81
A633-79
A645-82
A656-81
A660-79
A662-82

111
112
113
114
115
45
116

A663-82
A668-83
A671-80
A672-81
A675-82
A678-75
A690-81
A691-83

117
118
119
120
121
122
123
124

A692-83

18

A694-81

125

A696-81

126

A699-77
A706-82
A707-83
A709-81
A710-79

127
128
129
130
131

A714-81
A724-82

132
133

A727-81
A730-81
A734-82

134
135
136

Specification Title
Pressure vessel plates, alloy steel, quenched and tempered chromium-molybdenum
Pressure vessel plates, alloy steel, quenched and tempered nickel-chromiummolybdenum
Pressure vessel plates, alloy steel quenched and tempered 8% and 9% nickel
Pressure vessel plates, carbon steel, manganese-titanium for glass or diffused metallic
coatings
High-strength low-alloy columbium-vanadium steels of structural quality
Structural carbon steel plates of improved toughness
Steel bars, carbon, merchant quality, M-grades
Steel bars, carbon, hot-wrought special quality
Electric-welded low-carbon steel pipe for the chemical industry
High-strength low-alloy structural steel with 50 ksi minimum yield point to 4 inch thick
Seamless and welded carbon steel water-well pipe
High-strength quenched and tempered low-alloy steel forged fittings and parts for
pressure vessels
Carbon steel forgings with special magnetic characteristics
Steel tubes, low carbon tapered, for structural use
Pressure vessel plates, carbon steel, high-strength, for moderate and lowertemperature service
Deformed and plain billet-steel bars for concrete reinforcement
Hot-formed welded and seamless high-strength low-alloy structural tubing
Normalized high-strength low-alloy structural steel
Pressure vessel plates, 5% nickel alloy steel, specially heat treated
Hot-rolled structural steel, high-strength low-alloy plate with improved formability
Centrifugally cast carbon steel pipe for high-temperature service
Pressure vessel plates, carbon-manganese, for moderate and lower temperature
service
Steel bars, carbon, merchant quality, mechanical properties
Steel forgings, carbon and alloy, for general industrial use
Electric-fusion-welded steel pipe for atmospheric and lower temperatures
Electric-fusion-welded steel pipe for high-pressure service at moderate temperatures
Steel bars, carbon, hot-wrought, special quality, mechanical properties
Quenched and tempered carbon steel plates for structural applications
High-strength low-alloy steel H-piles and sheet pilings for use in marine environments
Carbon and alloy steel pipe, electric-fusion-welded for high pressure service at high
temperatures
Seamless medium strength carbon-molybdenum alloy-steel boiler and superheater
tubes
Forgings, carbon and alloy steel, for pipe flanges, fittings, valves and parts for highpressure transmission service
Steel bars, carbon, hot-wrought or cold-finished, special quality, for pressure piping
components
Low-carbon manganese-molybdenum-columbium alloy steel plates, shapes and bars
Low-alloy steel deformed bars for concrete reinforcement
Flanges, forged, carbon and alloy steel for low-temperature service
Structural steel for bridges
Low-carbon age-hardening nickel-copper-chromium-molybdenum-columbium and
nickel-copper-columbium alloy steels
High-strength low-alloy welded and seamless steel pipe
Pressure vessel plates, carbon steel, quenched and tempered, for welded layered
pressure vessels
Forgings, carbon steel, for piping components with inherent notch toughness
Forgings, carbon and alloy steel, for railway use
Pressure vessel plates, alloy steel and high-strength, low-alloy steel, quenched and
tempered

248

Chapter 6
Residual Stress and Distortion

Table of Contents
6.1

Introduction

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .251

6.2

Expansion and Contraction of Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .252

6.3

Coefficient of Thermal Expansion and Thermal Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . .254

6.4

Residual Stresses
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
6.4.1 Residual Stresses Induced by Thermal Process . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
6.4.2 Residual Stress Induced by Arc Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .257
6.4.3 Residual Stress of Universal Mill Plates with As-Rolled Edges . . . . . . . . . . . . . . . . .260
6.4.4 Residual Stress Induced by Flame Cutting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .260
6.4.5 Residual Stress in Welded Wide Flange Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . .261
6.4.6 Residual Stress in Universal Mill Rolled Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . .264
6.4.7 Estimation of Shrinkages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .265

6.5

Distortion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
6.5.1 Distortion Caused by Oxyfuel Gas Cutting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270
6.5.2 Distortions Caused by Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271
6.5.3 Transverse Contraction (Shrinkage) - Angular Distortion . . . . . . . . . . . . . . . . . . . . . .273
6.5.4 Longitudinal Expansion and Contraction (Shrinkage) . . . . . . . . . . . . . . . . . . . . . . . .274
6.5.5 Other Causes of Welding Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .276

249

6.6

Welding Procedure and Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .278

6.6.1 Welding Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279
6.6.1.1 Weld Pass - Single Pass, Multiple, or Small Pass . . . . . . . . . . . . . . . . . . .279
6.6.1.2 Travel Speed of Welding Arc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279
6.6.1.3 Uniformity of Heat Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .280
6.6.1.4 Joint Design, Preparation and Fit-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . .280
6.6.1.5 Welding Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .280
6.6.1.6 Seam Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283
6.6.1.7 Non-Continuous Fillet Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .286
6.6.1.8 Built-Up Structures - Neutral Axis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .286
6.6.1.9 Complicated Weldments - Accurate Assembly . . . . . . . . . . . . . . . . . . . . . .288

6.7

Control and Correction of Distortions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .289

6.7.1 Control of Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .289
6.7.2 Correction of Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .293

250

6.1

Introduction

The objective of this chapter is to discuss the phenomena of residual stress and distortion, explain their
causes, behaviour, magnitude, how to avoid or minimize them and finally, how to rectify welding
distortions when they occur beyond allowance. Residual stress is neither visible nor readily
measurable, where distortion is both. The latter is always the manifestation of the former. The relation
between residual stress and distortion will be discussed. In dealing with distortion problems, the
adherence to established welding procedures and manufacturing plans is very important. Years of
shop fabrication experience is still the best assurance. Knowledge of the fundamental theory and
equations will help you grasp the nature of the problem but precise control is not always achievable.
Due to the many variables involved, experience and theory are the best tools to avoid distortion.
It is often difficult to establish an exact, satisfactory welding procedure for an unsatisfactory design.
The following lessons are of equal importance to both the designer and supervisor. The designers
work is not finished on the drawing board or on the computer, since the designer and supervisor must
share the responsibility for the final product.
The development of welding procedures should begin with the understanding that the heat of welding
will produce expansion, contraction and stress, and consequently their major objectives should be to:
1.
2.
3.
4.

Produce sound weldments

Maintain dimensions by controlling distortion
Reduce and balance internal residual welding stresses
Be easily accessible and economical

Obviously, welding procedures will involve the welding process, base metals, joint design and
preparation, filler metals, power source, current and voltage, welding technique, heat treatment, etc.
Even more important, however, is the pattern of heat input to the work as determined by the sequence
of assembly and the sequence of welding.
To control distortion and residual stresses, the effort of each of these factors must be thoroughly
understood and the welding procedure should be planned accordingly. The welding procedure, once
planned, should be checked by trial run and modified when required. It should be clearly laid out and
purposely followed by all personnel. It forms an important part of the shops quality control system.
For further study the following CWB Modules provide more detailed discussions and practical
examples:
Module 7
Module 39

Residual Stress and Distortion

Weld Mechanics

251

6.2 Expansion and Contraction of Metals

By nature, metals expand volumetrically when heat is applied. When the heat source is removed the
metal contracts during cooling, also volumetrically. The expansion and conctraction movement of
heated metal can be illustrated by considering the movement of a metal ball. Figure 6.1 shows a
metal ball expanding freely under heat and contracting freely to its original volume and shape after
cooling.
Figure 6.2 shows a metal ball that is fitted snugly between two rigid stops and heated, then cooled.
Figure 6.2B shows that the ball is expanding in the open direction and restricted in the other direction
by the stops. During this process the metal grains have undergone adjustment under force to expand
to open space. This is called restricted expansion. When the metal is cooled it contracts again.
Consequently, it leaves gaps between the stops as shown in Figure 6.2C. This is called free
contraction. Another explanation is that metal becomes plastic at high temperatures and can be
moulded and then retains that shape when cooled.

Figure 6.1: Free expansion and contraction.

252

A steel ball is heated between two barriers which CANNOT MOVE:

20EC

500EC

20EC

Before

Expanded by Heating

After

The grains (and atoms) in the material have rearranged themselves.

Figure 6.2: Restricted expansion and contraction.

We have illustrated free expansion and free contraction, and restricted expansion and free contraction.
Let us now illustrate the third condition; restricted expansion and restricted contraction, as shown in
Figure 6.3. The ends of a round metal bar are rigidly gripped between two solid stops. Heat is applied
at any point on the bar causing it to expand, but it is not allowed to expand lengthwise. Therefore,
during heating all the expansion takes place in the diameter of the heated portion because this part has
to absorb all the volume of metal. The prevented expansion produces the same effect as if the bar
were allowed to expand lengthwise and then compressed back to its original length. The upsetting
(i.e., swelling of the heated part) is known as permanent deformation, that is, it will not disappear after
the bar cools. Therefore, the bar is in compression during heating (Figure 6.3A) and in tension after
cooling (Figure 6.3B). During heating, the metal is softened and forced to upsetting in diameter.
During cooling, the bar is stretched by the rigid stops. If the bar is sufficiently elastic, tensile stress will
be set up in the bar. If not, the bar will break as shown in Figure 6.3B.

253

Figure 6.3: Behaviour of metal bar when heated and cooled

while expansion and contraction are prevented.

6.3 Coefficient of Thermal Expansion and Thermal Stress

In the foregoing discussion we know that
metals expand when the temperature is
raised. In study of various metals, it is found
that given the same temperature rise the
amount of expansion differs for different
metals. Table 6.1 shows the coefficient of
expansion of some common metals. The unit

Metal
Mild Steel
Stainless Steel
Austenitic
Martensitic
Nickel
Copper
Aluminum
Magnesium
Lead
Zinc

-6

of the coefficient is in micrometre (10 metre)

per metre per degree (EK or EC). The
coefficient is not a constant and can be seen
in Figure 6.4.

Coefficient of
Expansion m/m EK
11.8
14.5
9.5
13.3
16.5
23.1
27.1
29.3
39.7

Table 6.1: Values for the coefficient of thermal

expansion for a number of metals at room
temperature. (Note: values are the same in units of
micro inches per inch per C)
254

Figure 6.4: Typical values of the coefficient of thermal expansion for mild steel
as a function of temperature.

Thermal stress is the stress induced by restricted thermal expansion or contraction. In Figure 6.3A for
example, assume that the entire bar is heated uniformly and the expansion per unit length (i.e., thermal
strain) can be calculated:
Thermal strain = .
Total expansion L = L . = L . . T
When expansion is prevented, the metal bar is shortened by the same strain. In other words, the metal
must be under compression. From the stress and strain relation:
Stress = (MPa) = - E
where E = Youngs modules of elasticity, for steel E = 200,000 MPa
= Temperature increase (C)
- Negative sign indicates a compressive state.
At room temperature (20EC) if the yield stress of the steel bar is 350 MPa, the temperature rise ()
required to reach yielding in compression can be calculated:

)T =

350
F
= 148EC (above room temperature)
=
-6
E"
200 000 x 11.8 x 10

255

In this example, when temperature rises () and is higher than 148EC, upset will occur. When the
steel bar is cooled to room temperature, residual tensile stress will be induced if the bar is not allowed
to contract.
6.4 Residual Stresses
The term residual stress means that some internal stress is created and stays inside the metal after the
manufacturing processes are completed. These processes can include thermal cutting or heating,
welding, mechanical forming or metallurgical changes such as heat treating. In this book, our
discussion deals mainly with the first two processes.
6.4.1 Residual Stresses Induced by Thermal Process
We have discussed thermal effect on metals in the foregoing paragraphs. We also explained how
stress may be set up when expansion and/or contraction is restricted. Previously we discussed
heating a metal bar gripped at both ends (see Figure 6.3). To bring this analogy one step further,
consider that a large square steel plate is spot heated (a small round area) at the centre of the plate as
shown in Figure 6.5. At the heated area the metal becomes upset due to restricted expansion by the
surrounding, relative cold, metal mass. After cooling, the upset remains and the contraction induces
tension around the heated area. This tensile stress stays inside the plate if nothing else is done to the
plate. This is why it is called residual stress, to distinguish it from other stresses created by external
loading.

Prior to
heating
Upset
During
heating
restricted
expansion

Compression

After
cooling
tensile
residual
stress is
included

Tension

Contraction
EXAGGERATED

HEAT SPOT: The temperature is relatively the same throughout the thickness
of the material but is localized.
Figure 6.5: Residual stress.

256

It may help the student to understand the practical meaning and effect of residual stresses as they are
regarded as internal compression and tension in the metal. For example, tensile forces are developed
across a butt weld when the weld metal is unable to contract freely. The residual stresses are static
and balanced, i.e., the overall tensile stressed areas are balanced by the compressive stressed areas
and no movement results once the balance is attained. But, in the process of balancing, while the
metal cools movement may happen. Then, distortion occurs. This is another important subject which
will be discussed.

6.4.2 Residual Stress Induced by Arc Welding

Next we shall investigate what happens when welding heat is applied to join two plates together as
shown in Figure 6.6. Two large, thick, rectangular plates of same size are welded together along their
long sides. During welding the long edges are under intensive heat (actually melted) and go through
thermal expansion. But the areas a short distance from the edges are relatively cool, and do not
expand at the same rate, or hardly expand because of the very steep thermal gradient. In other words,
the expansion is restricted by the plates themselves. Following the same reasoning, when the weld is
cooling down it goes through restricted contraction and sets up high tensile stresses along the weld
line. This high tensile stress stays with the plates if nothing else is done to them. This is how residual
stress is induced by welding. The residual stress in the longitudinal direction may be as high as the
yield stress of the plate (see Figure 6.6B). As explained previously, the thermal expansion and
concentration are in all directions (volumetric). Therefore, there is residual stress transverse to the weld
line, as shown in Figure 6.6A.

(A)

Figure 6.6: Typical residual stress pattern in a weld in a flat plate. Transverse stresses are not high
except at the ends where they are compressive. The most important residual stresses are the high
longitudinal stresses along the length of the weld and heat affected zone.

257

Further explanation of Figure 6.6 is illustrated with the aid of Figures 6.7 and 6.8. Figure 6.7 shows
that the plate edges along the weld joint undergo expansion during welding and the plates tend to bow
outwards. As the weld cools, shown in Figure 6.8A, the plates tend to bow in opposite direction
because the plate edges contract with the weld. Since the weld holds the plates together, the middle
part of the plates will be under tension perpendicular to the weld line. In the end regions compression
is induced to balance the tension region in the middle part. This is the transverse residual stress
pattern shown in Figure 6.6A.
Figure 6.8B illustrates the formation of longitudinal residual stresses that occur because the length of
a welds undergoes changes. Imagine how the weld metal stretches longer to fit the plate edges as
they expand outwards, and then contracts as the weld cools. This will result in tensile stress in the weld
metal and part of the adjacent plate, for width b on either side of the weld (Figure 6.6B). This tension
region must be balanced by compression regions outside width b on either side of the weld. These are
the longitudinal residual stress patterns shown in Figure 6.6B.

Figure 6.7: When the weld is deposited the edge of the plates get hot, expand, and tend to bow the
plates. Yielding occurs along the edges of the plates.

258

Figure 6.8: On cooling, the plates bow in the other direction but are held by the solidifying weld metal.
Residual stresses, equivalent to a bending moment applied to the plate ends,
result from the attempt to restrain the bow.

259

6.4.3

Residual Stress in Universal Mill Plates with As-Rolled Edges

Figure 6.9 shows the pattern of residual stresses in a universal mill plate with as-rolled edges. It
shows compressive residual stress at the edges and tensile residual stress in the middle of the plate.
A comparison of rolled edges with flame-cut edges (see Figure 6.10) shows a distinctive contrast. The
flame-cut edges have tensile residual stress at the edges whereas at the as-rolled edges the residual
stresses are compressive. From the discussion of the effect of heating and cooling we know that the
rolled edges cool faster than the middle part of the plate. Therefore, when the whole plate is cooled to
room temperature, the edges are under compression.

Figure 6.9: Residual stress in plate with rolled edges.

6.4.4 Residual Stress Induced by Flame Cutting

In oxyfuel gas cutting of steel, the temperature along the cutting surfaces can reach over 1000EC
(1800EF). The rapid heating and subsequent cooling will induce residual stress. When a plate is cut
with two torches simultaneously, the residual stress in the cutting edges is tensile. This is, of course,
because of the restraining effect of the relatively cool areas adjacent to the cutting edges. As a result,
the adjacent areas are in compression. The distribution of the longitudinal stresses across the width of
the plate is shown in Figure 6.10.

260

Figure 6.10: Residual stress in plate with flame cut edges.

6.4.5 Residual Stress in Welded Wide Flange Shapes

From the previous discussions on residual stresses in as-rolled universal plates, flame-cut plates and
the influence of welding, we should be able to visualize the residual stress patterns in two welded wide
flange shapes. Figure 6.11 and Figure 6.12 show the built-up shapes with as-rolled and flame-cut
stress patterns.
The residual stress patterns of a welded wide flange with cover plates is shown in Figure 6.13. A large
tensile residual stress is induced at the flange tips because of the high welding heat.
Residual stress in the welded box section is shown in Figure 6.14. Applying the same principles, the
corner areas cool slower and are in tension.

261

+
Figure 6.11: Longitudinal residual
stresses in welded built-up column with
as-rolled flange plates (Flange edges in
compression).

Plate flame cut

after welding

Plate flame cut

before welding

Large tensile
stress
Small compressive
or tensile stress

Figure 6.12: Welded wide flanges with flame cut edges.

262

Figure 6.13: Residual stress

patterns in a welded wide flange
with cover plate.

Cover plate

Large tensile
stress

+
-

Figure 6.14: Longitudinal residual stresses in welded box column.

263

6.4.6 Residual Stress in Universal Mill Rolled Shapes

After the discussion of residual stresses in plates and welded wide flanges, you would expect these
forces to be present in hot rolled shapes.
From the discussion earlier, we recognize that residual stress is induced due to uneven heating and
cooling. Residual stresses are induced in hot-rolled I-shapes for the same reason. As shown in
Figure 6.15, the parts that cool first (or faster) are the toes of flanges and the centre part of the web
wherein compressive residual stress is formed. The parts that cool last (or slower) are the flange and
web junctions, which are still contracting. The contraction is restrained by the parts that cooled first,
and tensile residual stress is formed. Therefore, the pattern of residual stress is as shown in Figure
6.15.

Figure 6.15: Residual stress in hot-rolled I-shape.

264

6.4.7 Estimation of Shrinkages

Formulas are available for calculating the amount of contraction or shrinkage of welds. The exact
amount of shrinkage is not always calculable because all the variables cannot be exactly controlled,
but these formulas do provide some indicators of which variable or variables exercise the most
influence on shrinkage. In other words, use these formulas as a guide in practical shop fabrication and
keep distortions within the code allowance.
1)

Transverse Shrinkage of Butt Welds

The following formula is applicable to carbon and alloy steels

S = k Aw + 0.05 d
t
Where:
S
Aw
t
d
k
k

=
=
=
=
=
=

transverse shrinkage, mm or inch

cross-sectional area of weld metal, mm2 or square inch
thickness of plate, mm or inch
root opening between plates edges, mm or inch
0.18 for 6 mm < t < 25 mm (1/4 < t < 1)
0.20 for t > 25 mm (t > 1)

The graph in Figure 6.16 shows the relationship between the plate thickness and transverse shrinkage
of 60E groove angle, single and double V-groove joints. It can be observed that single V-groove
contracts more than double V-groove of same thickness.

Figure 6.16: Transverse shrinkage of single and double vee welds.

265

Figure 6.17 shows that in joints with the same thickness of plates, the greater the weld metal area the
greater the transverse shrinkage. This graph shows that to reduce transverse shrinkage a joint should
be designed with double grooves of minimum groove angles as allowed by the welding standard for the
applicable welding process.

Figure 6.17: Proportion of transverse weld shrinkage produced

by various types of butt joint preparations.

2)

Longitudinal Shrinkage of Butt Joints

L = Aw x 0.025 L
Ar
where: (see Figure 6.18)
L =
L =
Aw =

Total longitudinal shrinkage, mm or inch

Length of weld joint, mm or inch
Cross-sectional area of weld

Ar =

metal, mm2 or square inch

Cross-sectional area of
restraining plates, mm2 or square
inch

Figure 6.18: Longitudinal

shrinkage of butt joints.

266

Due to restraint, this formula loses accuracy if the cross-sectional area of the plate is greater than 20
times that of the weld. In such cases the chart shown in Figure 6.19 may be used. It should be
observed that in the curves in Figure 6.19 the shrinkage, which attains high values for small resisting
sections, falls extremely rapidly as the section increases. The shrinkage tends to become constant
when the resisting section exceeds a certain value. The form these curves take need not surprise us.
In fact, the resistance to shrinkage offered by the resisting section increases very rapidly because the
effect of shrinkage is a maximum in a relatively narrow band symmetrical with respect to the axis of the
weld. Outside this band, only rather low temperatures are reached during welding and the metal offers
a rapidly increasing resistance to the shrinkage arising from the hot parts. The resisting section, once
it has exceeded slightly from the section corresponding to the hot parts of the assembly, exerts
essentially its maximum resistance. Further increase in resisting section has scarcely any effect on
shrinkage.
The following observation makes
this phenomenon more significant.
When the cross-sectional area of
the weld is increased, the highly
heated transverse portion, which
is acted upon by shrinkage, is
larger, and the resisting section
necessary to completely prevent
the effects of shrinkage also is
larger. This is what the curves
show. The dotted curve in Figure
6.19 shows the resisting section at
which shrinkage becomes
practically constant.

6.0

5.5

LONGITUDINAL SHRINKAGE - THOUSANDTHS IN. PER IN. OF WELD

5.0

4.5

4.0

3.5
Aw = TRANSVERSE WELD
CROSS - SECTION - SQ. IN.
3.0

A
w
0.

80

2.5

0.

90

0.

70

0.

60

2.0

0.5

0.

40

1.5

0.2

0.

15

1.0

0.3

0.2

0.

10

0.

05

0.5

0
0

10

11

12

13

14

TRANSVERSE CROSS-SECTIONAL AREA OF PLATES JOINED - SQ. IN.

267

15

Figure 6.19: Each curve

represents the variation of unit
longitudinal shrinkage as a
function of the transverse crosssectional area of the welded
assembly for a given cross
section. (The shrinkage tends to
become stabilized when the
sectional area of the assembly
exceeds a certain value, which is
indicated by the dotted line).

3) Transverse Shrinkage of Fillet Welds of Tee Joint

Figure 6.20: Angular distortion in a flange due to the two fillet welds.

A flange may suffer angular distortion as a result of the web-to-flange welds. The extent varies with the
flange thickness, since a thicker flange bends less. A suggested formula for this distortion is:
1.3

= 0.193 W /t2
Where

is the displacement as defined in Figure 6.20, mm

is the width of the flange, mm
is the fillet leg size, mm
is the plate thickness, mm

For the majority of practical cases, flange distortion predicted by this formula is within the tolerance
specified in CSA W59 or AWS D1.1.

268

6.5 Distortion
What causes distortion? Stress. The stresses we are concerned with here are residual stresses
caused by the thermal forces of the welding process. The reason a product distorts is due to the
residual stresses induced during fabrication, which are somewhat reduced in the distorted state. In
other words, a product distorts to reach an equilibrium state. For example, a distorted product, say a
bowed welded tee-shape, is put in jig, supported at both ends and some force is applied in the middle
to push back the bow and keep it straight (Figure 6.21). The tee is not in equilibrium by itself because
as soon as the force is removed the bow comes back again. That is why we say the distorted state is
an equilibrium state, the least energy state.

Figure 6.21: Bending distortion of welded tee-section.

The phenomena of distortion can only be fully understood with thorough knowledge of the behaviour of
residual stresses. To complete the picture of the stress situation it is also necessary to point out that in
the steel mill heating, rolling and cooling, some residual stresses are already present in plates and
shapes before any welding or other work is attempted. Normally these stresses are also in equilibrium
(otherwise, distortion occurs). For example, each flange of an I-beam processes residual rolling
stresses (also from heating and cooling) but they are balanced by the equal stresses in the other
flange.
However, when the balance of residual stresses is disturbed, distortion may occur. As an additional
example, it may be mentioned that the application of heat to one flange of an I-beam may cause
distortion solely because the residual stresses in that flange are reduced; i.e., the balance of stresses
is upset.
This is one reason why distortion sometimes occurs in a welded structure despite the use of all the
normal precautions.

269

6.5.1 Distortion Caused by Oxyfuel Gas Cutting

Oxyfuel gas cutting, commonly called flame cutting, is one of the major causes of distortion as the
result of improper application of a thermal process. From the discussion of residual stresses, we know
that flame-cutting induces residual stress and in turn causes distortion.
Figure 6.22 shows the mode of distortion that can happen in flame-cutting. By now you should be
able to explain why the plate bows along its cutting edge. To avoid this type of bowing (bending)
distortion, two torches must be used to cut simultaneously, as shown in Figure 6.23. When cutting a
long strip of steel plate of any thickness, for example 3 mm to 300 mm, two torches should be used to
apply heat along both edges to keep the plate straight.

Figure 6.22: Effect of cutting a flat plate

(with one torch).

Figure 6.23: Effect of cutting a flat plate

(with two torches).

In Figure 6.22, when one torch is used and heat is applied to one edge only, bowing is inevitable after
cooling. Bowing is caused by the tensile residual stress which is induced by the heat of cutting. When
two torches are used, as shown in Figure 6.23, the plate stays straight after cutting because the
residual stresses along the edges are balanced to each other and the resultant residual stress is
coincident along the centerline, or neutral axis of the plate.

270

In flame-cutting shapes from plate, the work piece must be kept with the remaining large plate until the
last severance cut. This will prevent the work piece from moving away from the large plate due to the
thermal expansion. A good example is shown in Figure 6.24 in cutting a round plate out of a large
plate. The cutting operation is controlled either numerically or by computer and the cutting torch
traverses a perfect circle regardless of the expansion movement. Due to expansion, when the
workpiece moves it will end up slightly oval, and the torch will not return to the starting point unless the
starting point of the cut and sequence of cutting are preplanned. In Figure 6.24, the cut should be
started at point A, never point B, when proceeding in counterclockwise direction. Similarly, you can
start at point B and proceed in a clockwise direction.
Another practical example is shown in Figure 6.25. In cutting a ring flange plate from a large plate the
first cut, second cut (removal of the centre piece), and the final (third) cut are shown. This is different
from Figure 6.24, as the cut is initiated by piercing inside the plate, not the edge. Even so, the cutting
directions must be followed. Remember that the width of the cutting kerf also provides room for
expansion. It should be noted that the centre piece (scrap) should never be removed first or the inside
diameter will change (pull inward) and the width of the ring will vary.

Figure 6.24: Method of cutting out a circle

near the corner of plate.

Figure 6.25: Method of cutting out ring

flanges near edge of plate.

6.5.2 Distortions Caused by Welding

As shown in Figure 6.6 and in the discussion of residual stress we have learned that welding heat
causes residual stress and distortion. The frequently seen types of welding distortions are shown in
Figure 6.26. It should be recognized that when distortion occurs it is not always in the simple form of
distortion as shown. Quite often distortion occurs in compounded forms, such as bending and twisting
or angular and bending and any combinations of the simple forms.

271

Shrinkage
g longitudinal shrinkage and
transverse shrinkage

Angular Distortion
g caused by transverse
shrinkage

Bending Distortion
g caused by longitudinal shrinkage

Buckling
g Caused by longitudinal shrinkage
(also to a minor degree, by transverse
shrinkage); most often when welding
large, thin plates or sheets

Twisting
g caused by high longitudinal
shrinkage; more likely in thin metal
Figure 6.26: Types of distortion caused by welding.

272

6.5.3 Transverse Contraction (Shrinkage) - Angular Distortion

Consider a V groove joint as shown in Figure 6.27A, which is unrestricted, i.e., free to move as
required by weld contraction. After welding, this joint will tend to assume the shape shown in B. The
angular distortion results from the non-uniform contraction of weld metal due to the greater width of the
top of the weld compared with the root of the Vee. If the weld metal could be deposited to form a more
uniform section between the edges, as shown at C and D, there would (in theory) be no angular
deformation, only uniform contraction across the joint.
Likewise it will be appreciated that in fillet welds the distortion resulting from contraction will be as
shown in Figure 6.27 F and G for a joint initially set up as shown in Figure 6.27E.

Figure 6.27: Distortion of butt and fillet joints due to weld metal contraction.

273

6.5.4 Longitudinal Expansion and Contraction (Shrinkage)

When we consider movements along this joint, the effect of expansion and contraction of the joint
edges becomes important because these movements are resisted by the comparatively cool metal
surrounding the weld point. Under this restraint considerable stress is set up in the metal.
This is illustrated in Figure 6.28. With reference to Figure 6.28A, if we assume that a portion of one
edge has been rapidly heated, the result is the production of an effect similar to that described in
conjunction with Figure 6.28B. In this case the expansion of the heated zone is prevented by the
comparatively cool metal; the result is that the increased volume of metal in the heated zone is
absorbed by a slight thickening or upsetting of the plate edge. Then, when cooling, contraction takes
place, the edge shortens, producing the shape shown in Figure 6.28B. This is exactly what is
happening to any joint edge or surface during welding, and the magnitude of the cooling effect
depends upon the size of the heated zone in relation to the size of the plate.

Figure 6.28: A and B show how

heating and cooling cause
distortion of plate edge.
C shows how contraction
causes plates to take the shape
shown by dotted lines.

274

If the edges are restrained this effort to contract will, instead of causing distortion, set up stresses
between the heated area (the weld) and the plate. This will happen if the parts being joined are
massive and rigid, or if rigidly clamped or rigidly tacked in place, restricting movement.
The effect of both the transverse and longitudinal contraction (shrinkage) of a butt joint where the plate
is not rigid is shown in Figure 6.28C. The important point, which should be very clearly understood, is
that local heating always produces contraction during cooling of the base metal, which, with the
additional contraction of the weld metal, causes concave bending, i.e., shortening of the weld side of
the joint both transversely and longitudinally.
Another example shows the plate edge movement during welding in Figure 6.29. Figure 6.29A shows
the far end of joint moving closer during welding with the shielded metal arc welding process (SMAW).
This is the result of the low heat input and low travel speed, which allows the plate edges to contract.
To prevent this from happening, a wedge block is inserted at the far end to keep a constant root
opening. Figure 6.29B shows the far end of the joint moving apart during welding with the submerged
arc welding process (SAW). Contrary to SMAW, submerged arc welding employs a high heat input
and a fast travel speed, which keeps the plate edges in an expanding state ahead of the welding arc
during the welding process. In this case, a heavy tack weld, or a tack-welded metal bar, at the far end
must be used to maintain the constant root opening. Submerged arc welding can produce 3 times the
heat input at 5 times the travel speed of SMAW.

Figure 6.29: Contraction of two butt-welded plates - effect of travel speed.

275

6.5.5 Other Causes of Welding Distortion

We have discussed distortion caused by residual stress, but residual stress alone does not cause
distortion, such as bending or angular distortions. When the distribution of residual stresses is
symmetrical about the neutral axes of the shape, bending or angular distortion will not occur, although
longitudinal shortening will always exist.
The neutral axis of some common section profiles are shown in Figure 6.30. The neutral axis is
located through the centre of gravity of the cross-section of a shape. When residual stress is in
symmetry about the neutral axis of a member it produces axial stress (tension or compression) only.
When the residual stress is not in symmetry about the neutral axis, a moment is created (Figures 6.31
and 6.32), equal to force P times e (eccentricity, distance between the resultant of residual stress and
the neutral axis). When the moment is large enough a visible or rejectable distortion will result.

Figure 6.30: Neutral axis of various sections.

276

Figure 6.31: Bending distortion due to eccentricity.

For complicated built-up shape such as the one shown in Figure 6.32, point A indicates the centre of
gravity of the built-up shape and point B is the centre of gravity of the weld areas, through which the
apparent shrinkage force acts. The distance between A and B is the eccentricity.

Figure 6.32: Bending distortion results

when the net longitudinal shrinkage force
of the welds acts in a line displaced from
the neutral axis of the assembly. The line
of action of the net apparent shrinkage
force is approximately at the centre of
gravity of the welds.

277

From the previous discussion, we can conclude that there are five types of distortions:
1.
2.
3.
4.
5.

Longitudinal distortion - shortening in length

Bending distortion - unbalanced residual stresses
Angular distortion - transverse contraction
Buckling distortion - longitudinal plus transverse
Twisting distortion - longitudinal contraction likely in thin plates or sheet metal

6.6 Welding Procedure and Distortion

When a welding arc is passing along the surface of a steel plate it creates a very drastic change in
temperature variations, called a thermal gradient as shown in Figure 6.33. Observe that within a few
millimetres of the welding arc the temperature may drop by 1000EC. The magnitude of the
temperature drop in a given material is proportional to heat input and travel speed. In a large
assembly, distortion occurs because of the uneven heating and rapid cooling of welding.
In previous paragraphs, we have already shown several modes of distortions caused by welding. To
control welding distortion we must fully understand the relationship between distortion and welding
procedures, joint design, preparation and fit-up.

C
1700
1500
1300
1100
900
700
500

2
3
4

HAZ WM HAZ

Subcritical HAZ
Intercritical or Partially
Transformed HAZ
Super Critical HAZ
Figure 6.33: Thermal gradient of welding arc.
278

6.6.1

Welding Procedures
6.6.1.1

Weld Pass: SIngle Pass, Multipass or Small Pass

Generally speaking, multi-pass welding increases the angular distortion, i.e., a large number of small
passes causes more distortion than a few large passes. The first pass forms a hinge point about which
the contraction of subsequent passes takes place. Transverse shrinkage will also be greater because
each pass will increase the number of upset areas along the plate edge. Therefore, the greater the
number of passes the greater the distortion tendency.
In some cases, however, the number of passes should be increased rather than decreased. This
occurs when the distortion in the longitudinal direction is more critical. In this case, the smaller the
cross section of a bead the less contraction force it can exert against the rigidity of the plates and the
more it will stretch. This apparently paradoxical relationship is a function of the thickness of the plate
and its natural resistance to distortion. There is inherent rigidity against the longitudinal bending or
shortening of a plate, providing the plate is thick enough. Light gauge sheets have little rigidity in this
direction and, therefore, will buckle easily. Unless the two plates to be welded are restrained, there is
virtually no lateral rigidity; since each of the two plates is free to move with relation to the other, out-ofplane distortion is more common.

6.6.1.2

Travel Speed of Welding Arc

The distortion of a joint will be affected by the rate of welding (travel speed). As the arc travels along
the joint the heat fans out in all directions from the weld point, as indicated in Figure 6.34. Any heat
that travels ahead of the weld point will distort the free joint edges and must, therefore, be kept to a
minimum. The slower the rate of travel, the more time there is for the heat to spread ahead of the weld
point, as shown in Figure 6.34A; the faster the travel the less heat spread that will occur ahead of the
weld point as shown in Figure 6.34B.

Figure 6.34: Arc travel speed and temperature distribution.

279

6.6.1.3

Uniformity of Heat Input

Expansion and contraction of the metal in the heat zone is further complicated by the fact that the heat
input to the joint is not uniform, but, as shown in Figure 6.6 is in the form of a concentrated zone (the
weld point) which travels along the joint as the weld progresses. At the weld point the heated joint
edge is expanding and upsetting (as previously described) and the weld metal is deposited in the fully
expanded condition. Behind the weld point the joint edges and weld metal are cooling and contracting.
In front of the weld point the joint edges are relatively cold and not yet subjected to expansion.
Obviously a better effect could be secured if the heat could be applied to the joint uniformly and
simultaneously throughout the whole length. Although this is not practical in structional fabrication
shops, preheat of work prior to welding does reduce the thermal gradient during cooling, in turn
reducing distortion. By the same reasoning, postheat also reduces distortion.

6.6.1.4

Joint Design, Preparation and Fit-Up

It has already been noted in reference to Figure 6.3 that the more symmetrical the weld section and
the more balanced the transverse contraction movements, the less angular distortion will be. Joint
design should, therefore, be as symmetrical as possible about the longitudinal centre line. Joint D of
Figure 6.27 is preferable from the viewpoint to Joint B. Similarly a U groove preparation is better than
a V groove.
Since the weld metal shrinkage is proportional to the amount of weld metal, it follows that the smaller
the weld the better. It is therefore the responsibility of the designer to detail weld sizes matching the
calculated strength requirements, and for the operator to make welds no greater than shown by the
drawings.
A large fillet will give more angular distortion than a smaller fillet and a wide Vee groove more than a
narrow groove since the contraction at the top will be greater (see Figure 6.27).
Therefore Vee grooves should be designed for a minimum bevel, consistent with accessibility, and
should be carefully prepared to see that this bevel is not exceeded.

6.6.1.5

Welding Sequence

Welding sequence is an essential part of any welding procedure. For example as shown in Figures
6.35 and 6.36 for the same double V groove joint, the sequence of weld metal deposited affects the
outcome of distortion.

280

1) Groove Welds
Figure 6.35 shows a symmetrical double-V butt joint preparation. In Figure 6.35A one side is welded
completely and the joint is distorted as shown in Figure 6.35 A2. Then, when the other side is welded,
the final joint geometry is distorted as shown in Figure 6.35 A4. This welding sequence cannot
eliminate the distortion that occurred in step A2 because the joint is locked rigidly. Figure 6.35B shows
an alternate sequence. The numbering of the weld passes shows that at step B2 the distortion caused
by pass 1 is partially eliminated. At step B3 the joint bends slightly upward. At step B4 the joint is
recovered to straight position. This is a satisfactory welding sequence. Notice that the plate assembly
has to be turned over and back a few times to achieve the final weld.
Figure 6.36 shows a double-V groove joint with unequal depths. Figure 6.36A shows welding without
root gouge and Figure 6.36B with root gouge. The one without a root gouge shows angular distortion.
The one with a root gouge ends up straight. It should be noted that root gouge is always done on the
shallow groove side for reduction of angular distortion. Again, in the example the work has to be
turned over once for downhand welding.

Figure 6.35: Symmetrical double-V butt joint preparation showing effect of welding procedure:
balanced welding (right) prevents distortion.

281

(A)

(B)
1

A1

B1

1
2
1
2

A2

B2
2

B3
A3

3
1
2

Root gouging
3
2

A4

1
2

4
4

B5

3 4
5

B6

23

A5

B4

3
1
4
5

Figure 6.36: Asymmetrical double-V butt joint preparation showing how gouging prevents distortion.

It should be noted that these two figures (Figures 6.35 and 6.36) do not show the welding position,
although the positions of the weld passes are shown. All the weld passes are deposited in a flat
position.
The sequence in which welds are carried out should be studied from the viewpoint of avoiding
complete restraint, which will inevitably introduce residual stresses in joints and, when severe, result in
distortion or cracking.
Figure 6.37 illustrates the welding sequence necessary to avoid restraint when welding structures
consisting of plates and stiffeners. The welding sequence is given as follows:
1)
2)
3)
4)
5)
6)

weld transverse fillets; this allows plate A to shrink without restraint

weld butts in plate A; plate is free to move.
butts in stiffener may now be welded while it is free to move
stiffener may now be welded to vertical plate;
brackets may be welded to vertical plate; bracket plate is free to move along stiffener.
bracket may now be welded to stiffener.
282

Vertical Plate
Bracket

Scallops

Plate A

Angle Stiffeners

Figure 6.37: An example of welding sequence in a structure combining plating and stiffeners.

6.6.1.6

Seam Welding

Seam welding is normally required when building ships or large fuel tanks where multiple plates are
welded along the seams (horizontal and vertical seams) to form the hull or tank wall. Correct welding
procedures or sequence is necessary if smooth surfaces and joint geometry are to be maintained.
The simplest form of distortion control is exemplified by
the well-known method for welding a longitudinal seam
of starting the weld some distance in from the end of
the joint and making a short weld first, as shown in
Figure 6.38. In this way the first weld pre-sets the joint
edges and prevents the closing in of the joint as the
main weld proceeds (compare Figure 6.29).
It has already been mentioned that distortion control
involves applying the proper pattern of heat distribution.
We have seen how this principle may be applied by
welding equal and opposite welds. Also it has been
noted that it would be desirable to apply heat uniformly
and simultaneously throughout the entire length of a
joint. As this is obviously not possible in arc welding,
the next best thing is to weld at spaced intervals along
the joint.
283

Figure 6.38: Simple welding sequence.

Figure 6.39 shows several sequences that apply this principle. A simple back-stepping method is
shown at A. This consists of starting a weld a short distance from the end of a seam - the distance
being the length of bead deposited by one electrode (SMAW). The next weld is then started a similar
distance from the first weld and is fused in to the previous starting point, and so on, until the joint is
completed. B is a minor variation of A, leaving one unwelded space in consecutive steps, called backstep and skip welding. On long joints the welder works outwards from a central point as shown in C
and E. This is an important principle to follow.
Still more elaborate variations of this procedure are the staggered or wandering sequence shown in
D and E. These procedures consist of leaving spaces between each weld bead, progressing along the
seam in this manner and then completing the unwelded spaces.

Figure 6.39: Seam welding techniques.

With a large area of plating, as shown in Figure 6.40, the welding should start at a central point and
proceed outwards, keeping the progress of welding as symmetrical about the centre as possible as
shown by the numerical order. The principle is to arrange in a way to allow for each joint to have
freedom of movement for the maximum time interval.

284

Figure 6.40: Suggested sequence for plate welding.

At the junction of seam weld (horizontal) and butt weld (vertical) the welding sequence is shown in
Figure 6.41. The seam weld adjacent to the butt weld should be left unwelded for a length of 300 to
380 mm on each side and then completed after the vertical butt is welded. This sequence allows
contraction of the butt weld and avoids high rigidity.

Figure 6.41: Sequence for seams and butts.

285

6.6.1.7

Non-Continuous Fillet Welds

The seam welding technique as shown in Figure 6.39 may be used for both butt and tee joints, but in
the latter case the welds may be staggered on both sides of the joint as shown in Figure 6.42. The
main advantage of non-continuous fillet welds (or intermittent fillet welds) is that the heat input to the
joint is considerably less and thereby distortion and shrinkage stress are reduced.
It will, of course, be essential to make sure that a non-continuous weld will give the required joint
strength. Quite often the minimum practicle size of the fillet provides more strength than that required
by design calculations; in such cases non-continuous welds may very well be used. On the other
hand, if a complete joint seal is required, non-continuous welding cannot be adopted.
Another advantage is that the heat is more uniformly distributed than it would be in the case of a
continuous weld. Moreover, the longitudinal weld shrinkage and, therefore, overall distortion, is only a
small fraction of that produced by continuous welding. It has, in fact, been found that the reduction in
these factors is far greater than would appear to be represented by the proportion of intermittent to
continuous weld.

6.6.1.8 Built-up Structures - Neutral Axis

The advantage of equal and opposite welding
about neutral axis has already been noted in
Figure 6.27, B and D and is also shown in
Figure 6.43. The neutral axis always passes
through the centre of gravity and is usually
defined as the line on which there will be
neither tension nor compression when the
piece is flexed or bent.
In the case of a piece of plate, the neutral axis
coincides with the centre plane of the plate
(see A in Figure 6.30); similarly, in the case of
an I beam of channel the neutral axis
coincides with the centre of the web, see B
and E. In the case of a Tee or single section
member, arranged as shown at C and D, the
neutral axis is not in the centre of the depth,
but is near the flange.

Figure 6.42: Intermittent fillet welds.

A clear understanding of the position and

function of the neutral axis is necessary if the
effects of welding either a plate or section, or a
complete weldment, are to be visualized.

286

As previously mentioned, the simple deposition of a bead of weld metal on the surface of a plate will
cause that plate to bend with concavity on the welded side. This is simply due to the fact that the
contraction of the weld metal exercises a shrinkage force that is offset from the neutral axis of the
plate. If, on the other hand, beads were deposited simultaneously on opposite sides of the plate, the
contractions of the two welds would be balanced about the neutral axis and there would be no bending
or distortion.
This balancing of welds about the neutral axis of a built-up
section or structure is the most important fundamental point
in reducing distortion. A further example is shown in Figure
6.43 where various welds are arranged around the neutral
axis of a built-up section, the sequence in which the welds
should be made being indicated by numbers.
Emphasis so far has been laid on the importance of welding
equally about the neutral axis to maintain alignment. This
assumes that the structure is true to begin with. In some
cases this may not be so and welding unequally about the
axis may be used as a means of straightening.
A case in point is the construction of a beam from plate
sections where the web plate has a curvature as received
from the mill. This might be as much as 10 mm in 1500 mm.
The following procedure may then be used to produce a
straight beam. (See Figure 6.44).

Figure 6.43: Balancing the sequences

of welds about the neutral axis of a
section.

Figure 6.44: Operations in welding a built-up I beam with curved web.

287

The flange plate f1 is laid down on a slab and the web plate (with convex side down) is set up vertically
on it as in A. The flange is then pulled up to the web plate and strongly tacked as in B. The welding of
this flange and the web is then carried out until the web is not only straightened but slightly bent in the
opposite direction as shown in C. The second flange f2 is now fitted to the web and tacked securely as
in D. The welding is then completed, preferably using two welders on opposite sides of the web and
working in the same direction. With such a sequence the beam should be reasonably straight on
completion. Welding the first flange to the web before the second flange has been tacked to the latter
results in a considerable bending effect due to the shortening of the weld in as much as the beam is
not strong or stable without the second flange. If in doing such welding the beam is slightly over bent,
the welding on the second flange, when completed, ought to be just sufficient to pull the beam back to
the straight position, since due to greater rigidity the shrinkage effect will not be as great as under the
conditions in which the first flange was welded.

6.6.1.9

Complicated Weldments - Accurate Assembly

In the case of complicated assemblies, the accuracy of preparing the various components requires
careful consideration to enable dimensional tolerances to be kept to a minimum. An accumulation of
tolerances over a number of components may create costly post-welding difficulties. Obviously the
more generous the tolerances, the greater the fit-up gaps, and an excessive amount of weld metal will
be necessitated, resulting in greater distortion than would otherwise be involved.
To avoid this, it may be desirable to machine components to size to obtain close tolerances and
increase the accuracy of the final weldment. It is also often possible to arrange the assembly of
components in such a way that cumulative tolerances can be controlled and prevented from adversely
affecting the final accuracy of the structure. (See Figure 6.45).
Where accurate location points are essential, the assembly arrangement of the structure should
provide for some allowance in case the various sub-assembly allowances do not work out to the
degree of accuracy expected. For example, in the case of built-up I-beams, the accumulated
longitudinal contraction of the flange to web welds and the transverse contraction of the stiffener welds,
will result in appreciable shortening of the beam, and it is usual to leave the flange and web plates
overlength so that they may be finished to size after welding.
Similarly, for machine structures such as bedplates, engine frames, etc., those points that must be
located to close tolerances should be fixed only by the last weld that affects their location.
With tolerance of " 1.6 mm on plates X and Y, assembly A would necessitate a tolerance of " 3 mm
whereas the accuracy of B could be " 0.8 mm.

288

Figure 6.45: Arranging components to ensure finished accuracy.

6.7 Control and Correction of Distortions

We have discussed the causes and types of distortions. What happens when the weldment is
distorted beyond the allowance referenced by the applicable codes or standards? What are the
common measures used by the welding fabrication shops to prevent distortions? What corrective
actions can be taken to eliminate distortions once occurred? A brief discussion will be given in the
following paragraphs.

6.7.1 Control of Distortion

In previous discussions of welding and distortion, several ways of preventing distortion have already
been mentioned. The following is a summary of control of distortion by welding procedure control:
1.
2.
3.
4.
5.
6.
7.
8.

Accurate joint preparation and fit-up. This is one way to maintain minimum weld metal for the joint.
The use of back-stopping or a skip technique.
Welding progresses outwards from a central point.
Balancing welds on either side of a centre line, central point or about the neutral axis of a section.
Welding butts (groove joints) before fillets to allow large contraction to take place first.
Using intermittent fillets instead of continuous fillets when code allows.
Arranging the weld sequence so that each joint has the maximum freedom of movement for the
longest possible period.
Dividing a weldment into sub-assemblies to reduce cumulative distortions or shrinkage,
especially lengthwise.

289

In addition to the above procedures, which are aimed at reducing distortion, the following points should
be given attention since they are particularly concerned with the production of an accurate weldment:
(a) Applying the welding so as to counteract plate edge curvature if any
(b) Arrange components so as to avoid accumulation of errors due to tolerances on plate width
(c) Where a high degree of overall accuracy is required, prepare components accurately to
reduce fit-up tolerances
(d) Arrange the sequence of welding so that location points necessitating a high degree of
accuracy are assembled and welded last
(e) Allow for weld metal shrinkage
(f) Arrange for some latitude in assembly dimensions so that a weldment can be machined
to size if shrinkage and other allowances do not work out as expected
Other means of distortion control besides the welding procedures are:
1. Preheating - reduces shrinkage because it provides more uniform heating and cooling
2. Peening - reduces shrinkage because it stretches the weld metal
3. Restraint - any degree of restraint, external or self weight, may be expected to reduce the
amount of shrinkage and such restraint may be applied in any of the following ways:
a)
b)
c)
d)

clamping
rigid tacking
maintain minimum or zero root opening (reduce transverse shrinkage)
cooling between weld passes (reduce the restraining required)

Mechanical Control
a)
b)
c)
d)

Presetting to allow recovery of angle or longitudinal distortion (see Figure 6.46)

Use of temporary stiffeners
Use of strongbacks or special jigs or fixtures
Artificial cooling

290

Figure 6.46: Presetting of joint members to allow for contraction of weld metal.

It has been already noted that distortion may be reduced by fixing components either by tacking,
clamping, or by assembling in jigs. Complete rigidity in this respect is however, contrary to the abovementioned principle of minimizing stresses. Therefore, unless the weld metal can be permitted to
contract freely (e.g., as in a preset joint), a balance must be found between the extremes of free
movement and complete rigidity so that both distortion and stresses may be kept to a minimum.
Accurate edge preparation and joint fit-up has considerable influence on the production of stress-free
joints. A variable and unnecessarily wide joint causes considerable heat concentrations at the wide
places, thus creating excessive lock-up, that is, residual stresses in the assembly.
Another preparation fault is excessive root face, particularly if accompanied by a tight fitting joint. Not
only is complete fusion of the joint difficult (if not impossible) to achieve but shrinkage of the deposited
metal will be prevented. The result will be high shrinkage stresses that are very likely to cause
cracking in service if the weld does not crack before it is completed.
Rigid alignment and complete restraint of joints by strongbacks, clamps and such devices should be
avoided. Figure 6.47 shows several methods commonly used to align joints. In A the joint is made
rigid and the method is entirely incorrect. In B and C the joint is free to contract and the methods are
suitable, while D is correct if the jack is removed after tacking and before final welding.

291

Figure 6.47: Methods of joint alignment.

Alternatively, weldments may be rigidly clamped to heavy slabs or bases during the period of welding;
also they may be mounted in rigid fixtures or assembled and rigidly tacked for welding. Strongbacks
and temporary stiffeners may be used to align and rigidly maintain edges and joints.
Heavy slabs and fixtures will not only hold assemblies rigidly, but will withdraw the heat of welding from
the weldments, further reducing distortion. A similar effect can be obtained by immersing assemblies in
water, or by spraying.
However, none of the methods of restraint can be expected to fully retain alignment. Some springing
and distortion will usually follow release from such superimposed control.
Further, the greater the restraint against contraction the greater will be the residual stresses induced
and the more likelihood that cracking will result as in Figure 6.47C.
Heavy weldments of heavy plate may in themselves offer great rigidity and restraint to welds.
Figure 6.48 shows cover plates welded to H sections. The fillet welds will have a tendency to shorten
due to their longitudinal contraction. This contraction will cause bending and a shortening of the
sections. If they are tacked or clamped together as shown, this bending tendency in each will be
counteracted by the other. The procedure should be to start welding in short increments outward from
the centre, alternating from one section to the other so that equal and opposite welds are made
alternatively and thus counterbalance each other.

292

It is understandable that distortion will be increased in large assemblies where the welds are long. It
therefore follows that if the job is broken down into a number of smaller weldments or sub-assemblies,
the distortion in each will be less and can more easily be controlled and corrected. If necessary, each
sub-assembly can be straightened or machined before final fitting and welding. Therefore final
fabrication from sub-assemblies is to be recommended and the designer should bear this requisite in
mind. Sub-assemblies further make for easier and more efficient handling and reduce the
accumulation of additive residual stresses.

Figure 6.48: Eliminating distortion

by balancing weld contractions of
two similar weldments clamped or
tacked back to back.

Experience has shown that control of distortion and reduction of welded-in-stresses can be achieved
by carefully planning the welding procedure.

6.7.2

Correction of Distortion

Although the foregoing suggestions for control or minimizing distortion provide some assurance of final
products, it should be appreciated that, despite the observance of all reasonable precautions, distortion
may still occur. Any such distortion will, however, be much less severe than it would have been had no
precautions been taken. When the distortion is greater than the code allowance corrective measures
are necessary. There are two common methods which are available to the fabrication shops:
1. Mechanical straightening
Use mechanical device, such as jacks, presses or specially designed straightener as shown
in Figure 6.49.

293

Figure 6.49: Specially designed straightener. (Courtesy of Canron Inc.)

2. Application of heat
Use the principle of resisted expansion during heating and subsequent contraction on cooling.
Several examples are given later to illustrate how the principle is applied.
It should be pointed out that each method is suited for certain applications. Most mechanical
straighteners are suited for minor straightening. Heavy components require specially built
straighteners, which are not available in small fabrication shops.
By far, flame straightening or flame forming is more common and readily available in all fabrication
shops, and is especially suited for large assemblies that cannot be corrected by mechanical
straighteners.
For example, a piece of Tee section bent as shown in Figure 6.50A may be straightened by heating
and cooling the area XYZ. The basic principle, which has already been studied, is that the expansion
of the metal in the heated zone is resisted by the cool surrounding metal. It therefore upsets and
remains so on cooling (Figure 6.50B) resulting in a reduction in the distance XY, straightening the
member as shown at right.
For the successful application of this principle both heating and cooling should be as rapid as possible
and the dimensions of the heated area should be at a maximum where most contraction is desired. In
the example shown at A, B and C in Figure 6.50 rectangular areas are heated. Even large, built-up Ibeams can be straightened by successively heating and cooling along the convex side of the beam as
shown at D. This principle can be applied also to the correction of distortion or buckling on plates or a
combination of plating and stiffeners.

294

N.A.

Figure 6.50: Eliminating distortion by heating and cooling.

295

The first example (Figure 6.51) shows a simple cylindrical vessel with a flat welded end or bottom
plate, which may be in any thickness of plate usually encountered in what is called the light or medium
tank work field. The weld is a corner weld inside and out round of the flat end plate, thus causing a
bulge in the centre. This condition can easily be corrected by application of heat in local spots as
shown. To achieve the best result, the spots should be evenly spaced and symmetrical over the
bottom. It must be noted that it is possible to overdo the application of local heat and undo much of
the good that may have been done. Overheating can produce buckles as bad as those it is desired to
eliminate. Therefore, in the first place, spots 1 to 5 should be tried, spots 6, 7, 8 and 9 being tried if
the first prove inadequate. The heat should be applied in the form of spots about 2 inches in diameter
and the plate brought to a cherry red colour. Care should be taken not to overheat or the effect can be
nullified.
The heat, of course, is applied by means of an oxy-acetylene flame. In this connection a word or two
about nozzle sizes may be helpful. A nozzle or tip for approximately 9 cubic feet per hour gas flow in a
standard torch is the best for use for anything up to 16 mm thick plates. For thicker plates, a nozzle for
approximately 23 cubic feet per hour glass flow in a heavy duty welding torch should be used.

Figure 6.51: Application of heat in local spots, evenly spaced and symmetrical.

296

The next example illustrates how considerable trouble can be experienced with rectangular tanks.
Figure 6.52 shows diagrammatically the top flange of a rectangular tank, indicating how the welding
contractions have pulled it out of square. A simple way of correcting this is to apply heat to the corners
of the flange as shown by the arrows. Evenly heat the area a few inches round the corners of the
flange on the outside at the extremities of the long diagonal and round the inside of the flange at the
extremities of the short diagonal. Correction can be assisted by inserting a prop, with a jack at one
end, across the short diagonal and applying pressure to stretch. Alternatively, bars attached to either
end of the long diagonal can be used with a turn buckle to draw in the tank in this direction. In some
cases it may be necessary to torch cut the flange at each end of the long diagonal and possibly
remove a triangular section to permit the necessary movement. Further assistance can be given in
stretching the material by peening the metal adjacent to the welds down each corner of the tank. Care
should be taken not to cause excessive indentations on the plate surface.

Figure 6.52: Distorted top flange of rectangular tanks. Apply heat to the corners of the flange.

Another example (Figure 6.53) of a large rectangular tank shows how the heavy type flange or curb
can be distorted by the weld that joins it to the top of the tank body. Again the cure is comparatively
simple; it consists of the local application of heat to the spots indicated by the arrows, and the heat in
this case is applied across the face of the flange in a V shape, the wider part of the V matching the
side that is being shrunk. This is particularly necessary if the flange is a heavy one.
Finally, large rectangular tanks with heavy stiffeners can bulge appreciably in the panels made by
positioning of the stiffeners; this bulging can be eliminated by heat applied to spots in the middle of the
panels as described previously. The sketch (Figure 6.54) indicates the manner in which this can be
done, and a tank in a comparatively bad state can be brought into good shape, with almost complete
flatness in the panels.

297

Figure 6.53: Another example of distorted flange of a large rectangular tank.

Local application of heat as shown.

Figure 6.54: Bulged panels of a rectangular tank. Apply heat to spots in the middle of the panels.

298

Chapter 7
Fracture and Fatigue of Welded Structures

Table of Contents
7.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .301

7.2

Stress-Strain Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .302

7.3

Fracture of Steel Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .303

7.4

Fracture Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .304

7.5

Cleavage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .305

7.6

Grain Size Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .306

7.7

Transition Temperature and Brittle Fracture

7.7.1
Design Considerations . . . . . . . .
7.7.2
Material Selection . . . . . . . . . . . .
7.7.3
CSA G40.21 Steels . . . . . . . . . . .
7.7.4
Stress Concentrations . . . . . . . . .
7.7.5
Net Section Yielding . . . . . . . . . .
7.7.6
Effect on Brittle Fracture . . . . . . .
7.7.7
Effect of Temperature . . . . . . . . .
7.7.8
Plain Specimen . . . . . . . . . . . . . .
7.7.9
Notched Specimen . . . . . . . . . . .
7.7.10 Transition Behaviour . . . . . . . . . .

7.8

Effect of Strain Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .319

7.9

Fracture Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .321

7.10

Stress State of Crack Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .322

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

299

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.306
.309
.310
.312
.314
.314
.315
.315
.316
.316
.318

7.11

Stress Intensity Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .324

7.12

Fatigue
7.12.1
7.12.2
7.12.3
7.12.4
7.12.5
7.12.6
7.12.7

and Fatigue Cracks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .326

Stress Range Categories and S/N Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .327
Cumulative Damage Formula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .339
Fatigue Life Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .341
Toe Grinding Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .341
Prudent Design Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .345
Prohibited Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349
Alternate Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349

300

7.1 Introduction
To engineers who design mechanical components, crane runways, highway bridges and sesmic
resistant structures, fractures and fatigue of structural members and connections are not strange
subjects. Because these structures carry moving and variable loads, the stresses in the members
fluctuate and sometimes reverse. The strength of the member and connection under these conditions
are much lower than those under static loads. It is a rather complex process to predict the member
strength because it involves load magnitude, range of fluctuation and frequency, member configuration,
service temperature, manufacturing process and metallurgical compositions. Therefore, the breaking
strength cannot be simply formulated as in the case under static loads.
Another factor that complicates the problem is fatigue of metals. To fracture in fatigue a crack must be
first initiated. The crack may be initiated under fluctuating loads, or are existing, such as inclusions or
porosities due to welding. It may be superficial or internal. It is difficult to decide when a crack is
initiated. There is always a time lag between the initiation of the crack and the detection of the crack.
The time required for a crack to grow (propogate) under each loading cycle until fracture is always an
educated guess (statistical). Therefore, the study of fatigue and fracture relies heavily on experiment
results and statistical interpretation. A lot of documented data is available to aid the design engineer.
For instance, various welded structural joints are grouped and categorized. The fatigue resistance of a
member or a detail can be calculated accordingly.
It should be pointed out that fracture and fatigue are two different subjects. Fracture can happen under
static loads alone, such as in laboratory tensile tests without cyclic loads (fatigue). The result of fatigue
always leads to fracture.
Fracture mechanics is an important tool in analysing and designing welded joints. It is also used to
predict fracture behaviour and fatigue life. Only brief outlines will be discussed to assist the
understanding of the fracture phenomena. The students are encouraged to study the following CWB
Modules to get further insight on this subject:
Module
Module
Module
Module
Module

35
36
37
38
39

Fracture Fundamentals
Fracture Applications
Fatigue Fundamentals
Fatigue Applications
Weld Mechanics

301

7.2 Stress-Strain Relationship

We are all familiar with the stress-strain relationship curves of steels as shown in Figure 7.1. The
steep, straight portion of the line shows the steel before it reaches the yielding point. The curve to the
right of the straight line shows the steel passing the yielding point into plastic deformation, before
breaking. This is a typical ductile structural steel property that designers depend on. The striking
characteristics of the tensile test sample is shown in Figure 7.2B showing necking and a large amount
of strain before breaking.
It should be noted that the curves shown in Figure 7.1 and breaking feature in Figure 7.2B are
generated under certain extrinsic conditions, such as monotonic loading and room temperature. Some
metals with ductile behaviour under a given set of extrinsic conditions appear to lose their ductility
under another set of conditions and become brittle as shown in Figure 7.2A.

Stress-Strain Curves Show Properties

100
High-Strength
Quenched and
Tempered Steel

Stress, 1,000 lb/in

HSLA Steel

46
32
Mild Steel

0.04

0.08

0.12

0.16

0.20

0.24

0.28

Strain, in/in
Figure 7.1: Ductility is the mainstay of structural steels that designers depend upon.

302

(A) Brittle Fracture

(B) Ductile Fracture

Figure 7.2: (A) Brittle fracture occurs with negligible deformation or elongation,
(B) ductile failures act in an opposite manner.

7.3 Fracture of Steel Components

Numerous studies and analysis of fractured steel comonents conclude that there are generally two
modes of fracture: ductile fracture and brittle fracture. Ductile fracture surfaces show large deformation
(strain) and shearing characteristics, which are inclined to loading (See Figure 7.3 & 7.4). Brittle
fracture surfaces, on the other hand, show little deformation (strain) and a flat surface which is
perpendicular to loading (Figure 7.4). In between these two extremes, there is a range of mixed
modes of fractures.
Although brittle fractures seldom occur in everyday structures, when designing for low temperatures
under fluctuating loads, engineers must consider this possibility. We all know through study or
experience that steel is a ductile metal, but the service temperature and the state of stresses can
change the ductility. For instance, when the ambient temperature is low and/or the shearing stress is
restricted by biaxial or triaxial stresses, the ductility will be reduced, as at stress concentrations of a
member with sharp changes in the cross-sectional areas. A combination of low service temperatures
and restricted stresses can actually reduce the ductility to near zero. The member can fracture in a
brittle manner, like cast iron.

303

7.4

Fracture Surface

The first obvious clue as to the type of fracture that

has occurred is the angle and shape of the fracture
surface. Consider a thin plate (strip) loaded in
tension until it breaks (Figure 7.3). Plastic
deformation occurs in a plane at an angle to the
direction of loading because that is where the shear
stresses causing slip are highest. After the material
has necked down and failed, the fracture surface
shows a characteristic slant.

Figure 7.3: Plastic deformation occurs on planes

of high shear stress, at an angle to the
principal stress. Ductile failures show
characteristic slant fracture surface.

A brittle fracture, on the other hand, runs more or less normal to the maximum tensile stress and shows
a characteristic flat, normal surface (Figure 7.4).

Figure 7.4: Brittle fracture (left) shows flat surface, normal to principal stress.
Ductile fracture (right) shows slant.

304

From the study of previous chapters, we have learned that the atomic structure of steel in room
temperature is body-centered cubic, due to the presence of an atom at the centre of the cube. Shear
failure of steel crystals occurs by sliding on a plane along the diagonal of the cube that contains the
centre atom. Brittle failure of steel crystals occurs by cleavage, separation between the faces of the
cube. Figure 7.5 shows a brittle fracture by cleavage.

Figure 7.5: Surface of brittle fracture.

The shiny facets result from the cleavage
of the individual grains giving
a crystalline appearance.

7.5

Figure 7.6: Characteristic river pattern on the

surface of a cleavage fracture observed
under high magnification.

Cleavage

In some metals, notably iron and steel at low temperatures, fracture may occur by cleavage, as
individual grains crack along specific crystallographic planes, with negligible plastic deformation. The
cleavage of grains appears as shiny facets on the fracture surface as Figure 7.5 illustrates. The
energy required for cleavage fracture is very small. A fast-running brittle fracture can continue to run
under a stress as small as 35 MPa (5 ksi).
As a cleavage crack advances, the stress field at the tip of the crack causes cleavage to occur in the
grains just ahead of the main crack. These may cleave on different planes, which may be at slightly
differing angles. The cracks subsequently link up. Under a high power microscope (Figure 7.6) the
cleavage surface is flat but not completely smooth, showing instead characteristic river patterns,
where the crack jumps from one to another parallel crystallographic plane. The small cleavage cracks
on separate planes may be linked by tearing i.e., a plastic deformation process. The energy
associated with such crack propagation is increased because of the energy required in plastic tearing,
and it is higher for a smaller grain size metal where many more tears must form. This fracture surface
is referred to as quasi-cleavage.

305

7.6

Grain Size Effect

One of the most important microstructural

features affecting brittle fracture is grain
size. A grain boundary provides an
obstacle to cleavage crack propagation
since adjacent grains will have their
cleavage planes at varying angles. A metal
with a small grain size, and hence a large
number of grain boundaries, has a greater
resistance to cleavage fracture. A small
grain also means that a cleavage crack in
a single grain is shorter and less likely to
initiate a crack in an adjacent grain (Figure
7.7).
Figure 7.7: Cleavage crack propagation through
polycrystalline metal. Small grain size increases the
resistance to cleavage.

7.7

Transition Temperature and Brittle Fracture

Brittle fracture is not common in most structures, occuring far less frequently than fatigue. If brittle
fracture does occur, it can be catastrophic. A brittle crack propagates through the material at the speed
of sound.
A brittle fracture starts with little warning: it fails suddenly with extremely little deformation (Figure
7.2A), in direct contrast to the ductile failure shown in Figure 7.2B.
Above a certain temperature, a given steel behaves in a ductile manner, while below this temperature,
the same steel behaves in a brittle manner. This transition occurs at the transition temperature (Figure
7.8). Steels having good resistance to fracture are said to be tough.
The most common method used to establish the transition temperature is the Charpy V-notch impact
test (Figure 7.9). Standard-sized specimens are subject to an impact load over a range of
temperatures. The absorbed energy is then plotted against temperature, the results being represented
in a typical curve.
The transition temperature is often defined in design specifications as the temperature that
corresponds to an energy level of 15 ft-lb. However, the whole process of impact testing is somewhat
arbitrary, as the Charpy V-notch test does not produce a definite value that can be directly included in
design calculations. Rather, impact values are primarily used to facilitate material selection or verify
contract specifications.

306

Charpy values indicate which of two grades of steel is the tougher, however, the values will not, of
themselves, predict if the grade is adequate. The impact tests are done on small, standardized
specimens, which do not correlate well to the much larger sections of a given engineered structure.
When the Charpy test was first introduced, most steels had a ferritic microstructure. Modern steels,
being very clean and thus very ductile, may give extraordinary high energy absorbtion values. There
are several other tests that were developed to gauge the material toughness. See CWB Modules 35 to
38 for detailed descriptions of testing set-up and procedures.

Figure 7.8: A given grade of steel will behave in a ductile manner if above its transition
temperature, but becomes brittle if below this value.

307

Scale
Pointer

Pendulum

Standard
Striking
Edge
Anvil

Specimen

8 mm

0.25 mm rad.
2mm

L/2
10 mm
55 mm
(2.165 in)

10 mm

45
V notch

2 mm
5mm

5 mm
Saw cut
1.6 mm or less
Keyhole notch

2 mm
U notch

Figure 7.9: Charpy V-notch testing.

308

Where there is a history of past performance, the Charpy impact test can be used to establish a
meaningful value for inclusion in the engineering specification. The first and somewhat classic
example was the investigation into the large number of brittle fractures associated with welded ships
during World War II (Figure 7.10).
Extensive research established that those vessels with Charpy values over 15 ft-lb at the normal
operating temperature were almost totally free of brittle fractures.

Figure 7.10: Fracture of Liberty ships during World War II.

7.7.1

Design Considerations

Let us look at the critical factors associated with brittle fracture. It is important to recognize that no
fracture can occur unless the following three conditions exist simultaneously:
1.

Temperature is below the transition point, indicating that the material is in a brittle state.

2.

Presence of a notch or severe stress concentration, often a welding defect.

3.

Presence of tensile stress. (The residual tensile stresses from welding are generally at the
yield point.)

If a notch, or weld defect, is subject to a high tensile stress while the steel is in a brittle state, a running
crack will propagate from the notch. Thus, the designer must assess the probability of these three
factors occurring during operation. If the factors will likely occur, the structure should be designed with
no obvious stress raisers and the material selection should be reviewed.
309

7.7.2

Material Selection

Special material considerations are not normally necessary, since only where there exists the likelihood
that all three of the above-noted conditions will occur simultaneously will brittle fracture occur. A
number of other factors should also be noted:
g

A high rate of strain will have the effect of raising the transition temperature. This fact also
points out one weakness of the Charpy test: the actual rate of strain imposed on the test
specimen greatly exceeds what a structure can be expected to resist in practice.

Thicker sections are more susceptible to brittle fracture (Figure 7.11), as plane strain rather
than plane stress governs the fracture mechanism. Triaxial stresses have a similar effect
(see Fracture Mechanics later in the Chapter).

Figure 7.11: Material toughness decreases as the material thickness is increased.

A material having a fine grain size is most beneficial (Figure 7.12): such grades are usually
controlled rolled and fully killed (deoxidized). These two manufacturing features represent the
only methods currently available that result in greater toughness without lowering the
strength. Normalizing and cross rolling are also effective.

310

Figure 7.12: Modern steels (CSA G40.21 Series) that are fine grain and/or controlled rolled,
and fully killed, offer excellent toughness.
311

7.7.3

CSA G40.21 Steels

General Structural steel should be ordered to the CSA G40.21 Specification. All grades ending in WT
(44WT or 350WT) are weldable, with improved impact properties. The WT steels are available in five
impact categories, one of which must be set by the customer. All grades are fine grained, fully killed.
(a)

Fine Grain Practice

A fine grain practice simply means that elements such as aluminum or silicon have been added in
sufficient quantity to raise the coarsening temperature. This results in a critical range that makes it
easy to obtain a fine grain with controlled heating and/or controlled hot working. Since neither the
austenitic grain size test nor an aluminum analysis is necessarily performed, the actual grain size
or aluminum content is not certified or reported. Regular rolling practices apply.
(b)

Fine Grain Steel

The term fine grain steel indicates that the steel has a carburized austenitic grain size of 5 or finer
when subjected to the McQuaid-EHN test (ASTM Standard E112). If aluminum is used for grain
size control, a product analysis showing a minimum of 0.010% acid soluble aluminum is acceptable
as an alternative to a McQuaid-EHN test result. In this case, the controlled rolling practice is used,
forcing recrystallization.
Thin plates may achieve a fine grain structure when only semi-killed, or partially deoxidized.
Thick plates, being subject to plane strain, must be fully killed.
(c)

Static Strength

The overwhelming tonnage of steel used is the common carbon or HSLA grades. Those specified
under CSA G40.21 have three common features that make them particularly suited to structures:
(1)
(2)
(3)

Strength, yields of 44 and 50 ksi

Ductility, approximately 22% standard elongation
Good weldability

Structural steels are often locally stressed beyond the yield point; grades with good ductility (Figure
7.1) can readily redistribute the stresses and maintain equilibrium without fracture. Higher strength
steels sacrifice ductility.
(d)

Weldability

Weldability is used to describe the ease with which a steel may be joined by the arc welding
process. A convenient approach in assessing weldability, and the effect of carbon and the alloying
elements, is the carbon equivalent (CE) formula. This formula expresses the relative influence of
the various elements in terms of carbon. Table 7.1 shows the CE values for G40.21 steels. Note
the rating index at the bottom.

312

Carbon Equivalent (CE) Values and Weldability Ratings

CE Values

(1)(2)(3)

Standard

Grade

Thickness
(mm)

Mean

Max

Weldability rating based on

carbon equivalent values

CAN3-G40.21M

230G
260W
300W
350W
400W
480W
260WT
300WT
350WT
400WT
480WT
350R
350A
400A
480A
350AT
400AT
480AT

25
25
25
25
20
10
25
25
25
20
10
14
25
25
19
25
25
19

0.33
0.40
0.41
0.40
0.40
0.38
0.39
0.41
0.43
0.39
0.38
0.40
0.44
0.45
0.48
0.44
0.45
0.48

0.36
0.44
0.44
0.44
0.43
0.40
0.43
0.44
0.46
0.42
0.40
0.46
0.50
0.51
0.54
0.50
0.52
0.54

excellent
excellent to good
good
excellent to good
excellent to good
excellent
excellent to good
good
good to fair
excellent to good
excellent
good to fair
good to fair
good to fair
fair to poor
good to fair
good to fair
fair to poor

ASTM A283

A
B
C
D

25
25
25
25

0.19
0.22
0.29
0.34

0.23
0.26
0.33
0.38

excellent
excellent
excellent
excellent

25

0.31

0.35

excellent

ASTM A36
ASTM A572

42
50
60
65

25
25
25
12.7

0.40
0.40
0.38
0.40

0.44
0.44
0.40
0.44

excellent to good
excellent to good
excellent
excellent to good

ASTM A588

25

0.44

0.50

good to fair

ASTM A242

1
2

12.7
25

0.40
0.44

0.46
0.50

excellent to fair
good to fair

ASTM A441

25

0.40

0.44

excellent to good

Stelco Wearwell

25

0.59

0.65

poor

(1)

(2)
(3)

Carbon equivalent is calculated using the formula in CAN3-G40.21-M81:

CE = C + 1/6 Mn + 1/5 (Cr + Mo + V) 1/15 (Ni + Cu)
Weldability rating: CE to 0.40 incl.
- excellent
CE 0.41 to 0.45 incl.
- good
CE 0.46 to 0.52 incl.
- fair
CE over 0.52
- poor
The mean and maximum carbon equivlalent (CE) values were determined using the average and
maximum chemistries normally applied by Stelco for plate in the thicknesses shown.
Lower carbon equivalent values are normally applicable for light thicknesses while higher carbon
equivalent values may be applicable for heavier thicknesses.

Table 7.1: Carbon equivalent (CE) values provide an index to weldability.

313

7.7.4

Stress Concentration

To illustrate the effect of stress concentration, suppose a plate contains a circular hole (Figure 7.13).
The cross section area in the plane containing the hole is less than the gross section but must still
carry the same load. Since no load can be transmitted across the hole, the stresses in some parts are
therefore increased. It is sometimes useful to think of stress in terms of continuous lines of force within
the material. In a reduced section the lines must bunch up, representing an increase in stress. The
presence of the hole consequently produces a stress concentration, which for a round hole raises the
stress by a factor of three. It does not matter how large the hole is, the stress concentration factor for
a round hole is never larger than three.

Figure 7.13: Stress concentration due to the presence of a round hole in the plate. In the vicinity
of the hole the stress is raised by a factor of three, regardless of the diameter of the hole.

7.7.5

Net Section Yielding

As the load on the plate increases (illustrated in Figure 7.14) the material at the edge of the hole
reaches the yield point and starts to yield while the average stress across the specimen is only one
third of yield. With the load continuing to increase, the yielded region cannot support stresses higher
than yield, so the load redistributes itself with higher stresses being supported in the elastic region
away from the hole. The plastic yielded zone spreads across the specimen until the entire specimen
has yielded. This is termed net section yielding.
If, instead of being elastic/perfectly plastic the material work hardens, stresses higher than yield could
be supported in the yielded region. If, in addition, the hole is small, stresses in the net section will rise
above the yield stress until yielding in the gross section occurs.
You can see from this example that, in a ductile material, general yield in the presence of a small hole
occurs at a load equal to the general yield load of a plain specimen. The presence of small defects or
stress concentrations has essentially no effect on the overall yield strength of a ductile material.
314

It is for this reason that a designer traditionally determines the load-carrying capacity of a member on
the basis of average stresses across a section and does not take into account the effects of section
changes, surface imperfections and other stress concentrations that affect the stress locally.

7.7.6

Effect on Brittle Fracture

Stress concentrations, however, have a significant effect on brittle fracture. We may illustrate this by
considering the fracture of steel specimens at various temperatures. Steel is brittle at very low
temperatures but is ductile at higher temperatures, so a specimen may be made to fracture in a certain
manner simply by changing the test temperature.

Figure 7.14: Yielding in a plate containing a round hole. Yielding starts first at the edge of the
hole, then spreads out. A small amount of yielding at the edge of the hole does not affect the
overall linear behaviour of the specimen. Eventually the net section yields.

7.7.7

Effect of Temperature

Let us consider the behaviour of a smooth steel specimen subject to tension tests over a range of
temperatures. At room temperature the specimen shows considerable plastic deformation after
passing the yield stress, finally reaching the ultimate tensile strength (UTS) before necking down and
breaking. As the temperature is lowered, this behaviour is retained, except that the yield and ultimate
strengths increase. Figure 7.15 illustrates this behaviour.

315

Figure 7.15: Behaviour of plain specimens of steel tested

in tension over a range of temperature.

7.7.8

Plain Specimen

At a sufficiently low temperature the behaviour changes. Failure now occurs with little plastic
deformation, and at very low temperatures fracture occurs as soon as the yield stress is reached. Note
that with a smooth, defect-free specimen fracture does not occur below the yield stress even at very
low temperatures. This has been verified by measuring the yield stress in compression. It is the same
as the fracture stress in tension at very low temperatures. We may conclude from this that at least
some plastic deformation is required even for a completely brittle fracture.

7.7.9

Notched Specimen

Now imagine the same tests on specimens containing a small notch. At room temperature there is
very little difference in behaviour. As we have seen, the load at yield is about the same as for the
smooth specimen, and there is substantial deformation before failure. At lower temperatures, however,
the behaviour is markedly different. Fracture now occurs at an average stress well below the yield
strength of the material (Figure 7.16). The reason, is of course, that the notch provides a stressconcentrating effect, raising the stress at the tip of the notch beyond the yield stress, while the average
stress is still well below the yield. The specimen fails by fracturing from the notch.

316

Figure 7.16: Notched steel specimen shows a transition in fracture load with decreasing
temperature. Above the transition range the specimen behaves in a similar manner to a plain
specimen, showing similar strength. Below the transition the fracture load is well below the
general yield load.

When the tests are conducted on specimens containing sharp cracks rather than notches, the fracture
stresses at low temperatures are further reduced while behaviour at higher temperatures does not
change very much (Figure 7.17).

Figure 7.17: Increasing the sharpness or size of the notch lowers the fracture stress
at low temperatures but has little effect at high temperatures.
317

The three important conclusions from these observations are:

g
g
g

for brittle fractures to occur at average stresses less than yield, defects (stress
concentrations) must be present
the yield strength is not very sensitive to the presence of defects when the material is ductile
the fracture stress is very sensitive to the size and sharpness of defects when the material is
brittle

7.7.10

Transition Behaviour

Let us explore more fully the transition in fracture of steel with changing temperature. In Figure 7.18
we consider the fracture stress and the yield stress independently, but both as functions of
temperature. We see that at high temperatures the yield stress is reached before the fracture stress,
and the specimen deforms plastically. In effect, yielding intervenes to prevent fracture. At low
temperatures the fracture stress is reached before general yield takes place, the specimen failing by
fracture without much deformation. We can conclude that the temperature dependence of the yield
stress in steel is a basic cause of transition behaviour.

Figure 7.18: Fracture

transition behaviour in
steel. At low temperatures
the brittle fracture load is
reached before general
yielding occurs. At high
temperatures yielding
occurs first.

318

Anything that raises the effective yield stress, without similarly raising the fracture stress, results in an
increase in the transition temperature (Figure 7.19) Yield stress can be raised in many ways. As we
have seen, a deep notch in thick material raises the effective yield stress because yielding is
constrained by the triaxial stress state.

Figure 7.19: Factors that increase the yield stress have the effect of
increasing the transition temperature.

7.8

Effect of Strain Rate

A second factor that raises the yield strength, particularly in low- and medium-strength steels, is the
strain rate. High strain rates experienced, for example during impact loading, increase the yield stress
and cause a corresponding shift in the transition temperature. In fact, one reason for performing the
Charpy test under impact loading is to artificially raise the temperature to reveal brittle behaviour.
Figure 7.20 shows the effect of strain on yield stress for a typical mild steel, and Figure 7.21 shows
the approximate shift in transition temperature of a structural steel that follows from this effect. The
effect of strain rate diminishes as the strength of the steel gets higher, virtually disappearing for steels
of more than 1000 MPa (150 ksi) yield strength. The shift in transition temperature, therefore,
decreases roughly linearly with yield strength (Figure 7.22).

319

Figure 7.20: Effect of strain rate on the yield stress for a typical mild steel.

Figure 7.21: Effect of

temperature on the fracture
toughness of a structural
steel. Note the increase in
the transition temperature
with increasing strain rate.

320

Figure 7.22: The shift in transition

temperature on increasing the
strain rate from slow to fast
(impact) progressively decreases
with an increase in steel strength.
For steels with a strength greater
than about 1000 MPa (150 ksi)
there is no effect of strain rate on
fracture behaviour.

7.9

Fracture Mechanics

Fracture mechanics is the science of

fracture behaviour quantitatively in
relation between applied loads and
crack sizes, and defines the
toughness of materials against brittle
cracking. Figure 7.23 shows the
possible modes of crack tip
deformation. In this chapter, we
concentrate our attention on Mode I,
which is the most relevant to our
interest.

Figure 7.23: Crack tip deformation modes.

321

7.10

Stress State of Crack Tips

Figure 7.24 illustrates the different behaviours of thin and thick plates with cracks. It explains the
physical meaning of plain strain and plain stress.

Figure 7.24: Plain strain and plain stress.

322

Figure 7.25 shows the different stress states of crack tips of thin and thick steel plates. In thin plates,
the Z-direction restraint is negligible and therefore no stress is induced. The stress is two dimensional
(plain stress). In thick plates, high restraint in the Z-direction prevents contraction (plain strain,
Poissons effect) from taking place and consequently tensile stress is induced in that direction. Figure
7.25 (b) illustrates the triaxial stress state at the tip of the thick plate.
The stress in the Y-direction, (, near the crack tip is very high because of the stress concentration
effect due to shape change in the cross section. The stress in the X-direction, x, is zero at the crack
surface and rises to a maximum and then drops flat at that distance away from the crack tip. The
stress in the Z-direction is, obviously, zero on the plate surface. The stress curve of z is shown at the
centre of the plate thickness. The biaxial stress state near the crack tip is shown in the thin plate and
the triaxial stress state is shown in the thicker plate. This explains why fractures occur even when the
average stress is low. Also, the brittle cleavage fracture occurs because the triaxial stress state
prevents fracture in shear.
It should be noted that in the triaxial stress state, yielding in the Y-direction is inhibited and y
continues to rise. The maximum y may reach three times the uniaxial yielding stress.

Figure 7.25: State of stress at the root of a notch under uniaxial load.

323

7.11

Stress Intensity Factor

Figure 7.26: A crack in an infinitely wide plate.

The stress equation using polar coordinates with 2=0 (see Figure 7.26) is:

yy =

a
KI
=
2r
2r

KI is called the stress intensity factor, which determines the stress gradient near the crack tip. For
practical application, KI is expressed in the following equation for tensile loading:
KI2 = Q F2 Ba

Q - Shape Factor

crack through an infinite plate

Q=1

internal circular crack

Q = 4 / B2

internal elliptical crack

Q = 1 / N2

long surface crack (shallow)

Q = 1.2

elliptical surface crack

Q = 1.2 / N2
324

1/ 2

b2 a 2 2
sin d
= 1
2
0
b

where 2 is the angular coordinate

a = crack radius for circular cracks
a = semi-minor axis for elliptical cracks
b = semi-major axis
KI is affected by the same factors that effect fractures, such as:
g
g
g
g

service temperature
loading rate
component thickness and geometry
fabrication and material composition

The stress equation is often expressed in the following form:

K I = 2a
and the unit of KI is MPa m , (m is metre in metric units), and kip in in imperial units. So, when a
is expressed in millimetres it must be divided by 1000 to convert it to metres. Also note that a is half
the length of internal through-cracks, and is the crack length for edge cracks. For surface cracks other
correction factors are used to convert a to equivalent through-crack lengths. See Figure 7.27.

Figure 7.27: KI values for various crack geometries (infinitely wide plates).

325

The above equation shows that there is a critical crack size (or flaw size) for each nominal stress (F).
When Qa exceeds the KI of the member, crack size will grow.
It should be noted that the general equation is only applicable to certain crack sizes with negligible
plasticity at the crack tip. See CWB Modules 35 and 36 for practical examples.

7.12

Fatigue and Fatigue Cracks

As mentioned in the introduction, fatigue is caused

under fluctuating loads. For fatigue cracks to
happen, there are usually locations with stress
concentrations or intrinsic flaws. Fatigue crack
growth originating from welded details are observed
in structures and laboratory experiments. Fracture
mechanics is applied to account for the behaviour of
cracks in structural components.
Figure 7.28 shows that the fatigue crack surface is
striated, corresponding to loading cycles. Fatigue
cracks originate from an initiation point that can be
clearly seen in Figure 7.29. The initiating point can
be internal (internal flaw) or external.
Figure 7.28: Microscopic fatigue striations
electron microscopy. (6500X)

Crack Origin

Clam Shell Mark

The crack growth always shows clam-shell

markings, which indicates the initiating point.
As the crack grows or propagates the net
section area is gradually reduced, and finally
fractures when it cannot support the internal
stress.
Striation changes directions at grain
boundaries to suit crystalline orientation.

Figure 7.29: Microscopic view of fatigue

fracture surface.
326

7.12.1

Stress Range Categories and S/N Criteria

Through numerous experiments and observations it is found that the stress range is the most dominant
factor affecting the fatigue life of a component. The stress range is defined as the algebraic difference
between maximum and minimum nominal stresses as shown in Figure 7.30. From Figure 7.30 it can
be seen that under static loading the stress range is zero. Stress range should not be confused with
maximum stress. Figure 7.31 shows the same stress range but with different Smax and Smin.

Figure 7.30: Stress range.

327

Figure 7.31: These stress ranges, while looking different, are equal.

328

The relationship between stress range (S) and number of loading cycles (N) can be used as an
indicator of fatigue characteristics of a member or a welded joint. Figure 7.32 shows the S/N curves
plotted to log-log scale. It covers the variables of stress range, number of cycles and design
categories. The stress range of the horizontal portion of each curve represents the endurance limit of
that category. It means that if the component is in or under that stress range,no fatigue will occur.
Figure 7.33 illustrates a butt joint and a fillet weld Tee-joint. Under the same stress range the fatigue
life (NI) of the CJPG butt joint is much longer than the fillet Tee-joint (N).

Figure 7.32: CSA W59 and S16-01 base fatigue life calculations on S-N Diagrams.
Laboratory data is plotted to log-log scale.
Stress range, number of cycles, and design category make up the variables.

329

Figure 7.33: Different welded connections produce different stress concentrations,

and thus have different fatigue life profiles.

The stress range of each category is given in Table 7.2 (Table 12-4 of CSA W59) for different fatigue
lives (N). For detailed illustrations of stress range categories, see Table 7.3 (Table 12-4 of CSA W59).

Category
A
B
C
D
E
F
W

For 100,000
Cycles
120
110
97
76
55
40
40

Allowable Range of Stress, F (MPa)

For 500,000
For 2,000,000
Over 2,000,000
Cycles
Cycles
Cycles
95
75
65
85
65
50
70
52
40
50
35
25
35
25
15
27
20
12
27
20
12

Table 7.2 CSA W59 provides design values for various categories.

330

General
Condition

Situation
S-No.

Stress
Range
Category

Kind of
Stress

Description

Illustrative
Example (See
Figure 12-1)

(See Table 12-3)

Plain
Material

S1

Base metal with rolled or cleaned

surfaces. Flame-cut edges with a
surface roughness not exceeding
1000 as defined by CSA Standard
B95, Surface Texture (Roughness,
Waviness and Lay).

1,2

Built-Up
Members

S2

Base metal and weld metal in

members without attachments, builtup of plates of shapes connected by
continuous complete or partial
penetration groove welds or by
continuous fillet welds parallel to the
direction of applied stress.

3,4,5

S3

Base metal and weld metal along the

length of horizontal stiffeners and
cover plates connected by continuous
complete or partial penetration
groove welds or by continuous fillet
welds parallel to the direction of
applied stress.

Tension
or
Reversal

S4

Base metal at toe of transverse

stiffener welds on girder webs or
flanges subjected to calculated
flexural stress.

S5

Base metal at end of longitudinal

stiffeners.

S6

Base metal at end of partial length

welded cover plates narrower than
the flange, having square or tapered
ends, with or without welds across
the ends.
Flange thickness < inch
Flange thickness > inch

E
F

S7

Base metal at end of partial length

cover plates wider than the flange
having square ends with welds
across the ends.
Flange thickness < inch
Flange thickness > inch

E
F

Table 7.3: Stress range categories for various applications.

331

* Either RT or UT to meet the quality requirements of Clause 12.5.4.4 applicable to welds subject to tensile loads.

Table 7.3 Continued

332

General
Condition

Fillet
Welded
Connections

Situation
S-No.

Stress
Range
Category

Kind of
Stress

Description

Illustrative
Example (See
Figure 12-1)

(See Table 12-3)

S12

Base metal at intermittent fillet welds

S13

Base metal adjacent to fillet welded

attachments where the length L of the
attachment in the direction of stress
is less than 2 inches and stud-type
shear connectors.

6,13,14,15

S14

Base metal at details attached by

fillet welds subject to longitudinal
loading only when the detail length, L,
in direction of stress is between 2
inches and 12 times the plate
thickness, but less than 4 inches, and
the transition radius R is less than 2
inches.

14

See
Tabulation in
Example 12,
Figure 12-1.

12

Tension
or
Reversal

S15

For base metal at details attached to

webs by fillet welds subjected to
transverse and/or longitudinal loading
regardless of detail length the
stress range categories shall be as
shown in Figure 12-1 in the tabulation
for the same Example.
Shear stress on the throat of fillet
welds shall be governed by stress
range category W.

S16

Except for cover plates (S6, S7) and

details attached to webs (S15) base
metal at end of details 4 inch or
longer attached by fillet welds where
the length of weld is in the direction of
stress.

16

Fillet Welds

S17

Shear stress on throat of fillet welds.

16

Stud-Type
Shear
Connectors

S18

Shear stress on the nominal area of

stud shear connectors.

13

Table 7.3 Concluded

333

Shear

Figure 7.34: Illustrative examples of various details representing stress range categories.
(CSA W59)

334

Figure 7.34: Continued

(CSA W59)
335

The log-log plot of the S/N data is a straight line, and can be represented by a simple formula. The
general form is given by:
N = C/Sm
Where,
N = number of load cycles
S = stress range
C = constant (different for each stress range category)
m = slope of S/N line
Thus, accurate calculations are now possible; instead of trying to pick off values from a small log-log
graph or using Table 12-3 of CSA W59 where values between 500,000 cycles and 2,000,000 cycles
are not listed.
The constants for the basic equation are tabulated below; a sample calculation follows.

Table 7.4

Thus, for Category C, at N = 2,000,000 cycles, one can find the permissible stress range.
N = C/Sm
2 x 106 = 101 x 108 / S3.33
S3.33 = 50.5 x 102
S = 12.95 ksi (compare with Table 12-3: S = 13 ksi)

336

The equation can be rearranged:

N1/N2 = (S2/S1)m
where,
N1 = Original number of cycles
N2 = Revised number of cycles
S1 = Original stress range
S2 = Revised stress range
Now the effect of changes in stress range or number of cycles can be evaluated. For example, if
the stress range for a CAT.C detail is increased by 25%, its fatigue life will be reduced by 50%. For
example:
S2 = 1.25 S1
N1/N2 = 1.253.27 = 2.0744155
N2/N1 = 1.0 / 2.0744155 = 0.48
The stress range in the fatigue equation is raised to the 3rd power ; thus, small changes in stress
cause large changes in fatigue life (Figure 7.35). When investigating an existing structure subject to
fatigue, it is imperative to use the actual state of stress, rather than the assumed (used for original
design).
In conjunction with the values given in Table 7.2 and
the rearranged S/N formula, the constant C can be
calculated for each stress range. See Example 1 for
the calculation of the equation constant for Category B.

Figure 7.35: Because the equation for the S/N

curve is raised to the 3rd power, small increases
in load result in rather large reductions in fatigue
life.

337

S/N Curve Equation

Because tables of fatigue data have only selected values of S for widely spaced values of N, and
since reading directly from curves may be imprecise, it is often more useful to work from the
logarithmic equation. In this example the equation for the S/N curve representing a specific
welded joint detail - designated category B - is determined from the values given in Table 12-3
from CSA W59.

Category
A
B
C
D
E
F
W

For 100,000
Cycles
415
310
220
185
145
110
115

S)
Allowable Range of Stress, MPa (
For 500,000
For 2,000,000
Over 2,000,000
Cycles
Cycles
Cycles
250
165
165
190
125
110
130
90
70*
110
70
48
85
55
32
65
40
18
85
65
48

* See notes in W59.

The equation is N = CS-m

We have

First determine m.

N2/N1 = (S1/S2)m

From the table above take

Therefore

2,000,000 310
=

100,000
125

N1 = 100,000

S1 = 310

N2 = 2,000,000

S2 = 125

20 = (2.48)m
i.e. m = log 20/log 2.48 =

3.3

Next determine the constant C

C = N x Sm = 2 x 106 x 1253.3
The equation is therefore

S)-3.3
N = 1.663 x 1013 (

1.663 x 1013

S in MPa

Example 1: Determination of the constant of an S/N curve equation.

338

7.12.2

Cumulative Damage Formula

The S/N graph is for a constant amplitude stress range, while many applications are subject to a
variable stress range. In a practical application, if one analyses a crane runway girder for the
maximum stress range only, a severe economic penalty is paid.
A crane runway sees a wide spectrum of loading:
g
g
g
g

crane full
crane empty
crane with empty ladle
the above loads, but with load located at different distances from the girder

By using the Miners Rule, an equivalent stress can be calculated, with subsequent savings. For each
stress range, S, the corresponding design life, N (number of cycles) is determined directly from a
standard S/N diagram.
Miners Rule:
Stress range

S1

S2

S3

Number of actual cycles

n1

n2

n3

Number of allowable cycles

for stress range, S

N1

N2

N3

Fraction of endurance used

n
--N

n1
--N1

n2
--N2

n3
--N3

To avoid failure, the sum of the fractions must be less than 1.0.
That is to say:

n/N < 1.0

See Example 2 for detailed calculation.

339

Miners Rule

A member containing a category B detail is subject to three sources of repeated loads. These
produce the following stress ranges and number of cycles during the life of the component.

1)

S1 = 150

n1 = 250,000

2)

S2 = 125

n2 = 500,000

3)

S3 = 115

n3 = 800,000

Miners Rule is applied to determine whether the component has adequate fatigue life under the
combined stresses. The first step is to calculate N, the total fatigue life, for each of the stress
range components from the equation for a category B detail found in the example on page 338.

N = 1.663 x 1013 (S)-3.3

The next step is to determine the fraction of life (n/N) used by each component. The total of these
fractions is then found. If the total is less than one, the component is considered to have
adequate fatigue life. The results of the calculations are given in the table below.

1)

2)

3)

150

125

115

250,000

500,000

800,000

1,096,000

2,000,000

2,634,000

0.228

0.250

0.304

n/N
n/N

0.782 < 1

O.K.

Example 2: Cumulative damage example using Miners Rule.

340

7.12.3

Fatigue Life Enhancement

The British and Danish Codes have provisions for enhancing the fatigue life of new or existing
structures by a factor of 2. The economic benefits are real and meaningful, especially on existing
structures.
Research at The Welding Institute in Cambridge, England, identified an acute line of intrusions along
the toes of all welds made by the arc process, with the exception of GTAW. All processes, however,
produce some degree to undercut at the toe, notwithstanding good looking weld profiles. The practical
implication is that all welds have a pre-existing defect, in the form of either microscopic undercut or
slag intrusions, or both (Figure 7.36).
This is the basic reason that welded connections have a much lower fatigue life than equivalent plain
materials. Weldment life is primarily one of propagation, while plain materials experience a crack
initiation stage. See Figure 7.37.

7.12.4

Toe Grinding Method

A burr grinder is lightly run over the weld toes, always moving parallel to the weld, as shown in Figure
7.38. Grinding to a depth of 1/32 (0.8mm) below the point of undercut is the easiest, quickest and
most effective enhancement. The small pre-existing defects are either removed or the sharp openings
dulled. The resulting profile modification also complements the overall effectiveness.
In Figure 7.39 the toe dressing on the left was ineffective because grinding was restricted to the weld
face; on the opposite side, the dressing was effective because grinding was directed at the weld toe.
A bit of base material must be removed.
On new designs, toe grinding at critical points can be specified right on the engineering drawing.
When these pre-existing toe defects are perpendicular to the applied stress, fatigue crack propagation
is accelerated.

341

Figure 7.36(a): Stress concentrations

arise from overall geometry, weld
profile, and weld defects.

Figure 7.36(b): The arc welding process is prone to lines of intrusions and severe undercut along
the toe of the weld. They are not detectable by normal NDE.

342

Figure 7.37(a): These toe defects

greatly contribute to the reduced
fatigue life of a welded specimen.

Ni

Crack Initiation
Np

Crack Propagation

S/N

Sp

S1

Stress Range S

Failure

Cu

rve

Cycles N

Weld
Fatigue Life

Plain Material
Fatigue Life
Figure 7.37(b): The fatigue life of a weldment is essentially one of crack propagation because of
inherent defects, while plain material has a crack initiation phase.

343

Figure 7.38: Light grinding

along the weld toe with a burr
grinder will enhance
fatigue life of an existing
weldment.

Web or brace

Ineffective
grinding

Effective
grinding

Defect

Depth of grinding
should be 0.5 mm
below bottom of
any visible undercut

Flange or
chord

Figure 7.39: To be effective, the grinding must be along toe (not weld face),
removing material 0.5mm below bottom of visible undercut.

344

There is little advantage to placing an attachment parallel to the stress (Figure 7.40(b)), because the
end of the weld is category E in both cases. Much time and money have often been expended on
such a task, even after the fact.
Another method of fatigue life enhancement is to use a bolted connection (ASTM A325, fully torqued,
friction tight); the existing crane runway girder in Figure 7.41 was modified, changing the controlling
fatigue category from E to B.
Shot peening is also an effective fatigue enhancement method, especially for irregularly shaped
components that dont readily lend themselves to grinding (the benefits are about equal).
Good Details
In Figure 7.42 a number of preferred details are shown. In addition, there are a number of prudent
design steps that will tend to avoid fatigue problems.

7.12.5

Prudent Design Measures

Certain prudent design measures can be taken routinely. They do not add cost and just may keep you
out of trouble when the unexpected occurs. The following points are especially appropriate for
miscellaneous attachments, doubly so if they must be field welded:
a)
b)
c)
d)
e)
f)
g)
h)
i)
j)

Use smooth shapes and transitions. Avoid notches.

Locate welds in areas of low stress, e.g., weld to the web rather than the flange.
Locate member splices in areas of lower stress.
Do not weld on the edge of flanges unnecessarily. Keep welds approximately 1/2 in from
the edge of plates.
Avoid intermittent welds. Consider smaller continuous welds.
Avoid intersecting welds. Cope the ends of stiffeners.
Interrupt fillet welds at corners.
Show and locate welds clearly for drafting office to follow.
Specify that all temporary welds, if removed, be done carefully and the area ground smooth.
This includes such items as lifting lugs, scaffolding lugs, etc.
Remove tack welds and grind properly.

345

Figure 7.40: Stress applied perpendicular to the toe defect tends to open the defect, while
stress applied parallel to the toe has little benefit. The orientation of the two
welded attachments are both Category E.

346

Figure 7.41: Friction tight ASTM A-325 H.T. bolts raised the fatigue category from E to B.

347

Figure 7.42: Fatigue design tips.

348

7.12.6

Prohibited Welds

There are a number of welds that are prohibited under fatigue loading by CSA W59 (Clause 12.4.14).
Some of the more common ones are:
1)
2)
3)
4)

intermittent welds
partial joint penetration groove welds in tension
welds with backing bar
plug and slot welds

7.12.7

Alternate Codes

Another problem associated with fatigue calculations is that no single Code contains all the
categories. Even the allowable stresses vary within some specifications. Thus, the designer should
be aware of alternate codes. A list is given below for reference.

Standards Providing Fatigue Design Guidance

Standard

Comments

CSA W59-2003

Follows CAN/CSA S16-01

Ontario Highway Bridge Design Code

Includes aluminum

CAN/CSA S16-01

Includes mechanical fasteners

AISC Handbook of Steel Construction

Includes Tee joints

British Standard BS-5400

Excellent written descriptions accompany

diagrams. Shows where cracking is likely
to occur.

German Standard DIN 4132

Includes concentrated loads (e.g. wheel

loads) on web to flange welds

AISE Std No 13

Covers industrial mill buildings, crane

runway girders

AASHTO requirements for highway bridges

Excellent diagrams

AWS D1.1-2004 Structural Welding Code

Covers hollow structural sections

Note that each standard or code may give different designations to the same or similar details.

349

350

Chapter 8
Welding Design

Table of Contents
8.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .353

8.2

Scope and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .354

8.2.1
Accessibility for Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .356
8.2.2
Formula for Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .356

8.3

Design Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .357

8.3.1
Allowable Stress Design (ASD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .357
8.3.2
Limit States Design (LSD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362

8.4

Shear Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .365

8.5

Fillet Weld Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .370

8.6

Fillet Weld Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .375

8.7

Restrained Members and Moment Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .382

8.7.1
Panel Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .384

8.8

Welding
8.8.1
8.8.2
8.8.3
8.8.4
8.8.5

8.9

Design Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405

8.10

Sizing Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .406

8.10.1 PJPG Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .408
8.10.2 CJPG Welds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .408

of Hollow Structural Sections (HSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .397

CIDECT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .397
Typical Joint Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .398
Possible Failure Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .398
Joint Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .401
CIDECT Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .401

351

352

8.1

Introduction

The subject of welding design can cover a wide range of products. The design method of each
application may be different, because of the configuration, type of base material and the nature of
loading.
Welding design involves steels and ferrous alloys, all classes of stainless steels, aluminum, copper,
cobalt, titanium and other metals and their alloys. Welding design also involves non-metallic materials
such as plastics, ceramics and composite materials.
The following list gives some idea of the extent of the welding industry:
g
g
g
g
g
g
g
g
g
g
g
g
g
g
g
g

medical and bioengineering equipment

electronics
household utensils
automobiles and all transportation equipment
pressure vessels and the energy industry
power generation
petroleum industry (including off-shore oil drilling platforms)
underwater welding
pulp and paper industries
plastics
machinery
buildings
vrane girder and runways
bridges
ship building
aerospace industry

It should be pointed out that in the above list, in addition to buildings and bridges, weldments in the
aerospace industry, ship building, pressure vessels and other industries are designed and fabricated
under the supervision of civil engineers who are predisposed to design disciplines. In other words, civil
engineers are not limiting their playing field to buildings and bridges. Any engineered product or
structure, excluding engines and mechanical components, can be analyzed and designed by civil
engineers if they choose to do so.

353

8.2

Scope and Objectives

The objective of this chapter is to focus on all elements necessary for making a sound and economical
welding design decision. Most of the factors that bear on an engineering judgment will be identified.
In addition, design guidance is provided by due consideration of established good welding practices,
and specifically to the related welding design rules of the governing standards, such as CSA S16-01,
CSA W59 and AWS D1.1.
There are two fundamental welding design-related considerations requiring individual and
knowledgeable engineering attention:
First for the welds to fulfill the exact design function assigned to them in the structure or the product
and to reliably maintain their integrity under the anticipated handling, shipping and ultimately service
loads.
Second for the welded joints to fully satisfy the requirements of optimum economy in their execution
and adequate access for inspection.
Since welding design involves all kinds of fabricated products, it is beyond the scope of this chapter to
cover every aspect of welding application. The following discussion is closely related to building
structures.
It is not intended to describe how the entire structure is analyzed. Instead, when all the external forces
are given, the welds at the joint will be judged accordingly to the internal stresses.
The following CWB Modules are suggested for those who would like to pursue further studies:
Module
Module
Module
Module
Module

30
31
32
33
34

General Design Considerations for Welding

Design of Flexible Connections
Design of Moment Connections
Welded Trusswork
Miscellaneous Structural Welding Design

The following paragraphs are valuable guidelines when dealing with welding design and fabrication. A
good welding design must be good for fabrication. Welding design is the starting point of the whole
fabrication process.
Although the design effort makes only about 5% of a products total cost, it usually determines
more then 70% of a products manufacturing cost.
Real economies will not necessarily be achieved on the shop floor, but in the design office and in the
drawing office. The cost of a joint is largely determined before it reaches the shop floor. Quality and
profit begin when the designer first puts pen to paper and all else follows. Those first decisions
predetermine the rest of the job.

354

The designer or the draftsman must specify the most economical joint design and welding process.
Should the designer use a full strength weld? a 100% butt weld? a partial butt weld? a pair of fillets?
continuous or intermittent welding? Should the joint be gapped? Should the designer use a single V
or a double V? a bevel or a V joint? backgouged or not?
Overwelding can be more detrimental than underwelding. Design values provide a safety factor that
will cover a considerable degree of underwelding error, but do not provide for some of the distortions
and stress raisers that could result from overwelding. Designers should recognize that overwelding
can be as serious as underwelding.

What Does a Weld Do?

1.

If it is to provide a path for the transfer of forces, a welded design is justified as are all the
calculations made to determine stresses and weld sizes.

2.

If it is simply to hold parts together, continuous welds are invariably wasteful, and a few
intermittent welds will prove to be more efficient.

Excessive Welding Results When:

1.

Designers call for 100% butt welds or continuous all-around fillets because:
a)
b)
c)
d)
e)
f)

2.

It looks solid
Loads are not given, so make it 100%
Designing the welds is too much trouble
The design basis is not known or might change
The customer expects it and will not challenge the adequacy
Two pieces of steel are in contact

Designers increase fillet weld sizes because they suggest that:

a)
b)
c)
d)

The shop might underweld

The basic design assumptions are questionable
The calculations are approximate only
They are only asking for visual inspection so they have to be sure

355

A Responsible Designer Will:

1. Size the weld to suit the load not the member
2. Use stitch welds when minimum size fillets govern the weld size
3. Use stitch welds and give an alternate continuous size for automatic welding when not
restricted by minimum fillet size
4. Remember that even at an allowable stress of 925 lbs per inch per 16th, a fillet weld still has
a safety factor of 2.9 to 4.0.
5. Remember that welders invariably tend to overweld rather than underweld
6. Use partial penetration butt welds whenever conditions permit
7. Always use a balanced weld design (developing the connected part 100%) as an upper limit
for the amount of weld specified

A Responsible Designer Will Not:

1.
2.
3.
4.
5.

Cause welding to be done where it offers no benefit

Require a consistency in welder performance that is difficult to maintain
Forget the practicalities of production
Uncritically repeat the way it has been done in the past
Specify more welding than is necessary

It is important to set down hard and fast rules for joint design. However, in this chapter we shall review
definite steps, on several topics, that should lower your costs through effective design.

8.2.1 Accessibility for Welding

Figure 8.1 shows the requirement of welding access. A practical design engineer will never design a
weld which is hard to reach by the welder. If it cannot be reached or does not have minimum access,
it cannot be welded. Or, the welder manages to get it welded, but the quality or soundness of the weld
may be questionable. If it is difficult to weld because of access problems, it will be difficult to inspect it.
Therefore, good access is essential for good quality welds.

8.2.2 Formula for Success

The formula for success is quite straight forward. Although the process starts with engineering, there
must be open communications and genuine cooperation with all involved personnel, particularly those
in the shop and field.

356

8.3

Design Principles

There are two major design

principles, i.e., Allowable Stress
Design Principle (ASD) and Limit
States Design Principle (LSD). The
former design principle has been in
use over one hundred years. The
latter design principle has emerged
since the 1970s and it is replacing
the former in Canada, Europe and
other countries. In the United
States at present, both design
principles are acceptable. At
present in Canada, the official
design principle is Limit States
Design. Therefore, in this chapter,
both Limit States Design (in metric
and imperial units) and Allowable
Stress Design (in imperial units) are
given for cross reference.

Figure 8.1: Minimum access requirement for welding.

8.3.1 Allowable Stress Design (ASD)

In allowable stress design the actual loads are used to calculate the stresses in a weld joint.
Table11.2(a) of CSA W59, gives the allowable design stresses for various types of welds. This table is
for statically loaded structures. For dynamically loaded structures, see Table 12.2(a) of CSA W59.
Columns 4 and 5 in both tables give the allowable stress and joint capacity for matching conditions.
Columns 7 and 8 give the non-matching conditions. Matching or non-matching means that the
strengths of electrodes and base metals are equal or unequal respectively.

357

358

Table 11.2(a):
Allowable Stresses for Welds and Joint Capacities for Statically Loaded Structures
(extracted from CSA W59-03 - see Clause 11.3.4)

359

Table 11.2(a): Continued

360

Table 12.2(a):
Allowable Stresses for Welds and Joint Capacities for Cyclically Loaded Structures
(extracted from CSA W59-03 - see Clause 12.3.4)

361

Table 12.2(a): Continued

8.3.2 Limit States Design (LSD)

The basic concept of the Limit States Design method is briefly presented by means of a number of
statements and definitions. It should be noted that the design method as such is outside the scope of
this book. Only general information and specific reference pertaining to welding have been briefly
provided. Of particular interest should be the calibration of the method to yield results comparable to
those obtained with the ASD method. It should be further pointed out that the LSD method has been
accepted by the International Standards Organizations (ISO). Canadian engineers continue to
contribute significantly to the development of ISO standards backed by the expertise and experience
gained with this method nationally.
Principle: All buildings must be designed to prevent with sufficiently small probability the
occurrence of various types of collapse and unserviceability.
Limit States are those conditions that correspond to the onset of various types of collapse or
unserviceability.
Ultimate Limit States are conditions associated with collapse.
Serviceability Limit States are conditions associated with unserviceability.
Ultimate Limit States are concerned with strength and stability. For these states the structure
must retain its capacity up to the factored load levels.
Serviceability Limit States are concerned with satisfactory performance of the structure at
specified loads and impose requirements on maximum deflections, permanent deformations,
fatigue cracking and so forth.
Load Factor (") is applied to specific loads to take account of loads higher than anticipated and
of shortcomings of methods of analysis.
Resistance Factor (N) is applied to resistances (R) or strength of members and takes account
of variations in material properties, dimensions, workmanship and uncertainty in the prediction
of the resistance.
Resistance Factor (N) for the base metal is taken to be 0.9 to maintain uniformity and simplicity
in design, with adjustments made to resistance formulae for other types of member failures than
that by yielding (buckling).
The Resistance Factor (N) for weld metal has been set at 0.67 in line with its value for other
fasteners (bolts) to ensure that connector failure will not occur prior to general failure of the
member as a whole.

362

Load Factor (")

Consistent probabilities of failure have been determined for Dead-to-Live load ratios.
Dead 1.25
=
Live 1.50

In order to take full advantage of the proven, successful application of the ASD method, the LSD
method has been calibrated to yield reasonably comparable results.

Live 1.5
=
= 1.67

0 .9
This is equal to the factor of safety for conventional Allowable Stress Design.

8.3.2.1

Load Combinations

The load combinations (not including earthquake) is expressed as follows:

D D + ( L L + wW + T T )

(CSA-S16-01)

Load factors, , shall be taken as follows:

D
D
L
W
T

= 1.25 except that

= 0.85 when the dead load resists overturning, uplift, or load reversal effects
= 1.50
= 1.50 for wind
= 1.25

The load combination factor, , shall be taken as follows:

= 1.00 when only one of L, W, and T acts
= 0.70 when two of L, W, and T act
= 0.60 when all of L, W, and T act
The most unfavourable effect shall be determined by considering L, W. and T acting along with
= 1.00, or in combination with = 0.70 or 0.60

363

The importance factor, , shall be not less than 1.00 except for those structures where it can be shown
that collapse is not likely to cause injury or other serious consequences, it shall be not less than 0.80.
For load combinations including earthquake, the effect of factored loads (in force units), is the structural
effect due to the factored load combinations taken as follows:
(a)
(b)
(c)

1.0D + (1.0E); and either

1.0D + (1.0L + 1.0E) for storage and assembly occupancies; or
1.0D + (0.5L + 1.0E) for all other occupancies.

The AISC load factors and combinations are similar to CSA S16-01, but with minor differences, such as
the factors for live load and dead load:
Loads, Load Factors, and Load Combinations
(excerpts from AISC Manual of Steel Construction)
The following nominal loads are to be considered:
D: dead load due to the weight of the structural elements and the permanent features on the
structure
L: live load due to occupancy and moveable equipment
Lr: roof live load
W: wind load
S: snow load
E: earthquake load determined in accordance with Part I of the AISC Seismic Provisions for
Structural Steel Buildings
R: load due to initial rainwater or ice exclusive of the ponding contribution
The required strength of the structure and its elements must be determined from the appropriate critical
combination of factored loads. The most critical effect may occur when one or more loads are not
acting. The following load combinations and the corresponding load factors shall be investigated:
1.4D
1.2D + 1.6L + 0.5 (Lr or S or R)
1.2D + 1.6 (Lr or S or R) + (0.5L or 0.8W)
1.2D + 1.3W + 0.5L + 0.5 (Lr or S or R)
1.2D 1.0E + 0.5L = 0.2S
0.9D (1.3W or 1.0E)

(A4-1)
(A4-2)
(A4-3)
(A4-4)
(A4-5)
(A4-6)

364

In this chapter, our concerns are mainly the factors applicable to weld strength and welding design.
The following formulas for weld strength are based on the CSA W59 Standard, Limit States Design.
Table 11.2(b) Factored Resistance of Welds for Statically Loaded Structures and Table 12.2(b)
Factored Resistance of Welds for Dynamically Loaded Structures are attached for reference. The
following highlights the factored resistances of limit states design.

8.4

Shear Resistance

(A)

Complete and Partial Joint Penetration Groove Welds, Plug and Slot Welds:
The factored shear resistance shall be the lesser of:
(a) Base metal

Vr = 0.67 w Am Fu

(b) Weld metal

Vr = 0.67 w Aw Xu

or

where w = 0.67
Am shear area of effective fusion face of base metal
Aw area of effective weld throat, plug or slot weld
(B)

Fillet Welds (static load):

The factored resistance for tension or compression-induced shear shall be taken as the lesser
of:
(a) Base metal

Vr = 0.67 w Am Fu

(b) Weld metal

Vr = 0.67 w Aw Xu (1.00 + 0.5 sin1.5 )

or

where w = 0.67
= angle of axis of weld with the line of action of force (0 in parallel, 90 in transverse)
It should be noted that when = 0
Vr = 0.67 x w Xu
when = 90

Vr = 0.67 x w Xu (1.5)

It indicates that a transverse weld line is 1.5 times stronger than a parallel one. This fact is
verified by testing and also shows in the design tables in the Handbook of Steel Construction
(CISC or AISC).
365

366

Table 11.2(b):
Factored Resistances of Welds for Statically Loaded Structures
(extracted from CSA W59 - see Clause 11.3.5)

367

Table 11.2(b): Continued

368

Table 12.2(b):
Factored Resistances of Welds for Cyclically Loaded Structures
(extracted from CSA W59 - see Clause 12.3.5)

369

Table 12.2(b): Continued

8.5

Fillet Weld Strength

Before we get into the actual design, we should be familiar with the weld strengths and how they are
derived. Fillet welds are the most commonly used type of weld. The following explanation shows the
geometry and the derivation of its shear strength in both LSD and ASD methods. Figure 8.2 illustrates
the shear condition in which the applied load is parallel to the weld A ( = 0)

Figure 8.2: Shear condition in which applied load is parallel to the weld.

For weld B, under static loading in limit states design:

weld metal

Vr = 0.67 x Nw Aw Xu (1.00 + 0.5 sin1.5 )

370

Fillet Weld Strength

Limit States Design (LSD) metric units

1 mm

Electrodes: E4918, Xu = 490 MPa

Steels:G40.21-350W, Fu = 450 MPa
(see strengths given in CAN/CSA S16-01)

Aw

1 mm

Am
P

45
1 mm

See CSA W59, Table 11.2(b) Fillet Welds

Weld shear strength at throat, = 0:

Vrw = 0.67 w Aw Xu
= 0.67 x 0.67 x 490 x Aw = 220 Aw (N)
Shear on the faying surface: Am = 1.414 Aw

Vrm = 0.67 w Am Fu
= 0.67 x 0.67 x 450 x Am = 286 Aw > Vrw
therefore, weld strength governs.
The weld shear strength on the faying surface to balance strength at throat will be:

Vrw =

220A w
= 156 MPa
1.414A w

A fillet weld is usually designed according to the leg size. The factored shear resistance of various fillet
sizes is shown in Table 3-24 and Table 3-25 of CISC Handbook of Steel Construction.
Example:
Calculate the strength of 8 mm fillet weld:
Vr = 8 x 156/1000 = 1.24 kN/mm

371

Fillet Weld Strength

Limit States Design (LSD) - Imperial Units
Electrodes: E70XX, Xu = 71.0 ksi
Steels:
G40.21-350W, Fu = 65.0 ksi
(see strengths given in CAN/CSA S16-01)
Aw
Am

See CSA W59, Table 11.2(b) Fillet Welds

Weld shear strength at throat:
Vrw = 0.67 w Aw Xu
= 0.67 x 0.67 x 71 x Aw = 31.9 Aw (kips)
Shear on faying surface:

Am = 1.414 Aw

Vrw = 0.67 w Am Fu
= 0.67 x 0.67 x 65 x Am = 41.3 Aw > Vrw
therefore, weld strength governs.
The weld shear strength on faying surface to balance strength at throat will be:

Vrw =

31.9A w
= 22.6 ksi
1.414A w

Fillet weld is usually designed according to the leg size. The factored shear resistance of various fillet
sizes is shown in Table 3-24 and Table 3-25 of the CISC Handbook of Steel Construction (in KN/mm).
In imperial units, the fillet weld strength of 1/16 per inch is:
Example: Calculate the strength of 1/4 in fillet welds:
Vr = 4 x 1.41 = 5.64 kips/in

372

22.6
= 1.41 K
1
16
16

in

Fillet Weld Strength

Allowable Stress Design (ASD) - Imperial Units
Electrodes: E70XX, Xu = 71.0 ksi
Steels:
CSA G40.21 350W or
ASTM A572 Grade 50
Fy = 50 ksi
Aw
Am

See CSA W59, Table 11-2(a) Fillet Welds

Weld shear strength at throat:
Vrw = 0.3 Aw Xu
= 0.3 x 71.0 x Aw = 21.3 Aw (kips)
Shear on faying surface:

Am = 1.414 Aw

Vrw = 0.4 Am Fu
= 0.4 x 50 x 1.414 Aw = 28.28 Aw > Vrw
Therefore, weld strength governs.
The shear stress on faying surface in the weld to balance that in the throat will be:
21.3
15.06
= 15.06 ksi, and
= 0.941
1.414
16

1
16

in.

Example: Calculate the strength of 1/4 in fillet by ASD method.

Vr = 4 x 0.941 = 3.76 kips/in

373

It should be noted that in sizing the fillet weld, the designer should be aware that the smaller size and
longer fillet is more economical than the larger size and shorter fillet. Figure 8.3 illustrates this fact by
comparing different fillet sizes and lengths of the same weight of weld metal.

in (12.7 mm)
Legs

in (6.35 mm)
Legs
12 in
(304.8 mm)

6 in
(152.4 mm)

Decreasing the leg size and increasing the length of

fillet welds can slash weld metal requirements. These two fillets
have equal strength, but the one on the right uses half the weld metal.

ASD*

Each weld weighs 1 lb or 0.454 kg

Leg size
3/8 in
(10 mm)

5/16 in
(8 mm)

in
(6 mm)

LSD*

(0.928 kip/in/1/16)

(152 N/mm/mm)

280 kips/lb

3878 kN/kg

336 kips/lb

4846 kN/kg

420 kips/lb

6460 kN/kg

560 kips/lb

7755 kN/kg

50.3 in
(1157 mm)

72.4 in
(1807 mm)

113.1 in
(3213 mm)

3/16 in
(5 mm)

201.0 in
(4626 mm)

Maximizing fillet length within stress limitations saves weld metal.

The long 3/16 in (4.76 mm) fillet resists twice as much force per pound
of weld metal as the 3/8 in (9.52 mm) fillet does.
Note: The metric and imperial sizes are the preferred sizes, not exactly equal.
* Based on Xu = 70 ksi or 480 MPa

Figure 8.3: Smaller, longer welds can reduce weld deposit by 50%.
374

8.6 Fillet Weld Groups

Design tables of fillet weld groups are given in both the CISC Handbook of Steel Construction and
AISC Manual of Steel Construction. It should be noted that these coefficients in the tables were
derived by testing under ultimate loading. Therefore, they should only be used in similar weld
configurations and loading directions. The loadings in the CISC Tables are given in parallel to one
weld line of each weld group. The loadings in the AISC Tables are given at 15 angle intervals to one
weld line of each weld group. If the actual loading is in between the angles (for example 23 is
between 15 and 30), the coefficients in the lower angle (15) should be used. Otherwise, direct
analysis should be carried out. Straight line interpolation may give unsafe results. Figure 8.4 shows
the weld group configurations.

Figure 8.4: Weld group configurations.

375

In designing weld joints with fillet welds, the designer should first inspect the member thickness. There
is a minimum required fillet size for the thickest member connected which is given by Table 4-4 in CSA
W59. The reason for this requirement is to prevent weld cracking due to fast cooling and restraint.

Table 4-4: Minimum Fillet Size

Material Thickness, t of
Thicker Part Joined (mm)

Minimum Size of Fillet Weld (mm)

T#12
12 < t #20
20 < t

5
6
8

Single pass
welds must be used.

Note: The minimum fillet sizes in Table 4-4 need not apply if welding procedures have been
established to prevent cracking as provided in Clause 5.7.

The following example illustrates the use of the fillet weld group tables in the CISC Handbook, 7th
Edition. The weld strengths are also based on CISC Handbook, 7th Edition.

376

Example 1
400 kN

Given:

400

A column bracket as shown in the

figure. The fillet weld group is
rectangular shaped. E4918
electrodes and CSA G40.21, 350W
steels. Find the fillet weld size
required.

120

Solution: Use fillet weld group Table 3-30:

400

l= 400 mm
al = 420

120
= 360 mm
2

= 0.9

tf = 28 mm

kl= 120 mm

K=

120

No weld across
flange thickness,
top and bottom

120
= 0.3
400

From Table

C = 0.187

Weld Size

D=

400
= 5.35 mm
0.187 400

Use 8 mm fillet to satisfy tf = 28 mm. This is the minimum size required to meet Table 4.4, CSA W59.

377

Example 2: Truss Gusset Plate Connection

Figure 8.5: Truss joint with gusset plate.

A typical to chord joint with gusset plate is shown in Figure 8.5(a). Figure 8.5(b) shows all the forces
acting on the gusset plate. The gusset plate to chord may be welded by either CJPJ or double fillets.
Moment on section m-m:

h
M m = C12 Pxe,
2
or

where C12 = C2 - C1,

Mm = V1 x e1 + V2 x e2

Assume C2 > C1

(1)
(2)

Use of equation (1) is rather straightforward, no need to resolve P1 and P2 into H and V components.
Direct forces,

V = 0

P = V1 - V2

(3)

Shear forces,

H = 0

C12 = H1 + H2

(4)

378

Calculate the maximum stress in gusset plate along m-m:

Mm
1 2
Lt
6

bending

fb =

direct

fc =

P
Lt

MPa

shear

fv =

C12
Lt

MPa

resultant stress

MPa

f r = ( fb + fc )2 + ( fv )2

MPa

The above formulae are applicable when the gusset plate is welded by CJPG. When double fillet
welds are used:

Mm
1 2
L
3

bending

fb =

direct

fc =

P
2L

kN/mm/weld

shear

fv =

C12
2L

kN/mm/weld

select fillet
size for

f r ( fb + fc )2 + ( fv )2

kN/mm/weld

kN/mm/weld

379

Example 3:

45

45

294
219
WT 200 x 66

250

kN

433
Gusset Plate

kN

48.9

CJPG

433kN
kN

kN

24.4

c Column

2L 89 x 76 x 7.9
1000
577

kN

73.3
kN

27

Gusset Plate
c

43

433

170

c of column 250 kN
W 310 x 97

6 @ 80

80

43

NT 200 x 66

50

250

ao
kN

1000

24.4

30 577

2L

kN

50

S8

48.9

0k

170

294
164 55

9x

kN

73.3

7.6

x7

.9

kN

10

35.7

350
(a)

350
(b)

Figure 8.6: Typical truss end to column connection.

The detail of truss end connection is shown in Figure 8.6(a). The header angles are bolted to the
truss end. This type of connection is normally used because it provides more room for erection
adjustment and reaming or short slotted holes can be used.
Find the forces applied at each bolt:

250
= 35.7 kN
7

Vertical force

Rv =

Horizontal force

M = 250 219 = 54750 kN/mm

on each bolt

(a)

Let the force in the first bolt above and below the centroid of the bolt group (the fourth bolt) be PH,
then, the second and third bolts above and below the centroid will be 2PH and 3PH respectively.
Find PH by taking moment about the centroid of the bolt group:
M = PH (1 x 160 + 2 x 320 + 3 x 480) = 2240 PH

380

(b)

kN

Since (a) = (b)

PH =

54750
= 24.4 kN
2240

2PH = 48.9kN
3PH = 73.3kN
The bolt loads are shown in Figure 8.6(b).
Check gusset plate along line A-A, use CJPG weld:
Shear:

V = 433 + 24.4 24.4 48.9 73.5 = 310.8kN

fv =

310.8 103
= 68.3 MPa
350 13

Direct tension:

ft =

250 103
= 55.0 MPa
350 13

Bending:

take moment about the centre of the gusset plate

M = 250 x 45 + 24.4 x 27 24.4 x 187 48.9 x 267 73.3 x 347 = 41270kN/mm

fb =

Resultant stress:

41270 10 3
= 155.5 MPa
1 13 350 2
6

f r = (68.3) 2 + (55.0 + 155.5) 2 = 221.3 MPa < 0.9Fy

Other design check: Shear yielding and rupture, and tension yielding and rupture, should all be
checked.

381

8.7 Restrained Members and Moment Connections

When beams, girders or trusses are subject to both reaction shear and end moment due to full or
partial end restraint or to continuous or cantilever construction, their connections shall be designed for
the combined effect of shear, bending and axial load.
When beams are rigidly framed to the flange of an I-shaped column, stiffeners shall be provided on the
column web if the following bearing and tensile resistances of the column flange are exceeded:
(a)

opposite the compression flange of the beam when:

Br = wc(tb + 10tc )Fyc <

Mf
db

except that for members with Class 3 or 4 webs,

Br =
(b)

640000
w c (t b + 10t c )
(h c /w c ) 2

opposite the tension flange of the beam when:

2

Tr = 7 t c Fyc <

Mf
db

where

wc = thickness of column web

tb = thickness of beam flange
k = distance from outer face of column flange to web-toe of fillet, or to web-toe of flange-to-web
weld in a welded column

Fyc = specified yield point of column

db = depth of beam
hc = clear depth of column web
tc = thickness of column flange
The stiffener or pair of stiffeners opposite either beam flange must develop a force equal to:

Fst =

Mf
Br
db

382

Stiffener shall also be provided on the web of columns, beams or girders if Vr calculated from Clause
13 is exceeded, in which case the stiffener or stiffeners must transfer a shear force equal to:

Vst = Vf 0.55 wdFy

In all cases, the stiffeners shall be connected so that the force in the stiffener is transferred through the
stiffener connection. When beams frame to one side of the column only, the stiffeners need not be
longer than one-half of the depth of the column.
Note: The factored shear resistance of column web: Vr = 0.55 wdFy.
In the case of direct beam-to-column moment connections, the full moment capacity can be developed
with flanges fully welded and the beam web is either bolted or welded to a single shear plate
connection. The column stiffeners, when required, are usually of same size and thickness for both
tension and compression beam flanges.

Figure 8.7: Plated moment connection.

However, in the case of plated rigid moment connections, the top and bottom plates are usually of
different thicknesses since downhand welding considerations will dictate a narrower, thicker top plate
and wider, thinner bottom plate (Figure 8.7).
When beams are framed to both column flanges, their flanges are not always on the same elevation. If
the difference in elevation is not more than 50 mm (2 in) Graham, Sherbourne and Khabbaz suggest
that one horizontal stiffener may be used provided that its thickness is increased by a factor of 1.7.

383

In the case that the beam is framed only to one flange of the column, the stiffeners, if required, need
only be extended to within half of the column depth. However, the welds connecting the stiffener to the
column web must be sufficient to develop the force P:

P = Ast Fy
In the case of built-up column sections the web-to-flange welds may require strengthening before these
provisions are applied.

8.7.1

Panel Zone

In addition to the stiffeners opposite the tension and compression flanges for transferring the flange
forces, the column web between the horizontal flange stiffeners, the panel zone (Figure 8.8), may
also require stiffening if the plastic shear capacity of the web, 0.55wdFy, is exceeded. This is based on
the Huber-Henckey-von Mises criterion in consideration of the coexistence of axial load in the column
web.
In the case of built-up columns where there is a large differential in beam moments causing high
longitudinal shear stresses between the column web and flange, larger weld capacity (hence larger
weld size) may be required in the connection region. Increased fillet size or partial grooves with
superimposed fillets may be appropriately used here. Figure 8.8 illustrates the basic requirements for
the welds holding the component plates in a built-up column section together. Further explanation can
be itemized as follows:
1.

The entire length of the column must have sufficient welds to withstand any longitudinal
shear between floors resulting from the floor load or other external loads with or without
earthquake loads.

2.

Within the region of beam connection to the column, the longitudinal shear is much higher
because of the abrupt change in flexural stresses within the depth of the beam.

3.

The regions of column flange in contact with beam flanges also transfer the direct forces
(tension or compression), through a portion of the web-to-flange welds.

Summing up these conditions, heavier weld is usually required in the connection region. There are
several ways in which different types of welds can be combined when fabricating built-up columns to
satisfy the above requirements, which are shown in Figure 8.9.

384

Figure 8.8: Welding requirements for built-up columns.

Typical analysis and design procedures as they apply to the basic types of two-way, one-way and
square-knee rigid connections follow next. Some degree of repetition will be encountered in these
general solutions. However, the intent is to present a complete design procedure in each individual
case. Numerical examples illustrating these procedures are included at the end of this book.

385

Double bevelled
entire length

Case 2
The web plate is bevelled to the proper depth on
all four edges along the entire length. The groove
weld is first made along the entire length.
Second, a fillet weld is made over the groove
weld within the connection region to bring it up
to the proper size.

Case 1
If the weld sizes are not too large, the
column may first be fillet welded along its
entire length. Second, additional passes
are made in the connection region to bring
the fillet weld up to the proper size.

Additional bevelling in
region of beam to
column connection

Bevelled only within

connection region

Bevelled entire length

Case 3
The web plate is bevelled to the proper depth
along short lengths within the connection
region. First, a groove weld is made flush
with the surface within the connection region.
Second, a fillet weld is made along the entire
length of the column.

Case 4
The web plate is bevelled to the proper depth on
all four edges along the entire length. Within the
connection region the web is further bevelled to
a deeper depth. First, a groove weld is made
within the connection region until the plate edge
is built up to the height of the first bevel. Second,
a groove weld is made along the entire length.

Figure 8.9: Alternate methods for making welds in a built-up column

at the point of beam framing.

386

Example 1

Two-Way Rigid Beam-Column Connection

Given:
Sections as shown in Figure 8.10
Steel designation: CSA G40.21 350W, Fy = 350 MPa.
Problem:
Investigate column stiffening
requirements for the tension and
compression flanges of beams.
Investigate column web for shear.
K = 36 k1 = 21

Figure 8.10: Data for Example 1.

Compression Flanges:
Since no moment values are given, the maximum moment capacities will be connected. From Beam
Selection Table of CISCs Handbook of Steel Construction both beams are Class 1 sections (Table 51), and maximum moment is Mr = Mp.
From Beam Selection Tables of the Handbook Mp can be found under the Heading Mr.
for W360 x 51, Mr = 281 kNCm
W360 x 64, Mr = 359 kNCm
or Mp may be calculated as Z Fy
for W360 x 51,

Mp = 0.9 x 894 x 103 mm3 x 350 MPa/106

= 281 kNCm

W360 x 64,

Mp = 0.9 x 1140 x 103 mm3 x 350 MPa/106

= 359 kNCm

(Divided by 106 converts newton-millimetres to kilonewton-metres)

387

Computing flange forces as M/d

for W360 x 51,

281 kN m 103 mm/m

C1 =
= 792 kN
355 mm
for W360 x 64,

C2 =
(a)

359 kN m 103 mm/m

= 1035 kN
347 mm

Considering W360 x 64,

Minimum Column web thickness not requiring horizontal stiffening opposite compression flange:

C2
1035 kN 10 3 N/kN
= 16.6 mm
wc =
=
(t 2 + 10 t c )Fyc 0.9(13.5 + 10 19.6) 350 MPa
Since 16.6 mm > 11.9 mm stiffeners are required.
Column web capacity:
Br = Wc (tb2 + 10 x tc) Fy
= 0.9 x 11.9 x (13.5 + 10 x 19.6) x 350 x 10-3
= 785 kN < 1035 kN
Force to be carried by stiffeners, Cs = C2 Cw
= 1035 785 kN
= 250 kN
(b)

Considering W360 x 51
Minimum column web thickness not requiring horizontal stiffening opposite compression flange:

wc =

C1
( t b1 + 10 t c )Fyc

N
kN
=
0.9(11.6 + 10 19.6 ) 350 MPa
792 kN 10 3

= 12.1 mm > 11.9 mm, therefore stiffeners are required.

388

Force to be carried by stiffener,

Cs = C1 Br = C1 - wc (tbl + 10tc) Fyc
= 792 kN 0.9 x 11.9 mm (11.6 mm + 10 x 19.6 mm) x
= 14 kN

350 MPa
N
10 3
kN

The stiffener must be designed for the larger of the two forces, 291 kN.
Area of stiffeners required.

N
C
kN = 921 mm 2
As = s =
0.9 350 MPa
Fy
250 kN 10 3

Width of stiffeners
a)
b)

W360 x 64 beam flange width = 203 mm

W250 x 101 column flange width = 257 mm

Maximum width = (257 mm-11.9 mm)/2 = 122 mm

Minimum effective width = 203/2 22 = 79 mm
Area of one stiffener = 921/2 = 460 mm2
Maximum b/t =

145
= 7 .8
Fy

Try two plates 12 x 85:

85/12 = 7.1 < 7.8, OK

effective width bnet = 85 21 + 11.9/2 = 70 mm

net area Anet = 70 x 12 = 840 mm2 > 460 mm2, OK.
Note: Usually the total width of stiffeners needs not be wider than beam flange, but narrow
stiffeners result in larger fillet size at the stiffener end. This will be seen later in the design of
stiffener welds. The AISC Specification states that the minimum one stiffener width shall not be
less than one-third of beam flange, and the minimum thickness shall not be less than one-half
of beam flange thickness.

389

Tension Flange:
Capacity of column flange opposite tension flange
Tr = 7tc2 Fy
= 0.9 x 7 x 19.62 mm x 350 MPa/103 N/kN
= 847 kN
Force to be carried by stiffener Ts = T2 Tr
therefore,

Ts = 1035 847
= 188 kN < 250 kN used to design stiffener opposite compression flanges.
Therefore, use same stiffeners as for compression flange.

Shear in Column Web:

Minimum web thickness not requiring reinforcement (equation 24) is

(1035 kN 792 kN )103 N

C 2 C1
kN = 5.3 mm < 11.9 mm
=
wc =
0.55d c Fy 0.9 0.55 264 mm 350 MPa
Web does not need reinforcement.

Plate:
12 mm

K1 = 21

Figure 8.12: Detail of stiffener ends.

Figure 8.11: Detail of compression stiffener.

390

Welding of Stiffeners:
The shear force in the pair of stiffeners to be transferred into the column web is taken as
compression flange
tension flange

= 1035 - 785 = 250 kN

= 1035 - 847 = 188 kN < 250 kN

Although the tensile force to be resisted by the stiffeners is less than the compression
same size stiffener is usually used for both tension and compression stiffeners.

force, the

Design fillet welds for 250 kN, stiffeners to column web.

Try four 5 mm fillet welds, E49XX electrode
(0.778 kN/mm from Table 3-25, CISC Handbook, = 0)
length of each weld

250
= 80 mm
4 0.778

web length = 264 - 2 x 37 = 190 > 80

OK

The general shop practice is to weld the full length, otherwise, excessive unweld length should
be checked for compression from beam flange.
Design welds for end of stiffener to column flange:
previously calculated the affective width = 70 mm
fillet size

250
= 0.893 kN/mm
4 70

use four 6 mm fillet welds (1.21 kN/mm, = 90, see Table 3-25, CISC Handbook)
According to Table 4-4, CSA W59, minimum fillet size required for tc = 19.6 mm is 6 mm.
If the force is too large, complete penetration welds are usually used to connect the stiffener ends to
column flanges.

391

Example 2

One-Way Rigid Beam-Column Connection

Given
Steel: G40.21 350w
Sections
Depth
Flange thickness
Flange width
Web thickness
K
K1
Moment resistance
Class

Electrode: E49XX
Beam
Column
W460 x 67
W310 x 86
454 mm
310 mm
12.7 mm
16.3 mm
190 mm
254 mm
8.5 mm
9.1 mm
36 mm
23 mm
456 kN@m
1
1

Figure 8.13: Connection for Example 2.

Problem:
Investigate the requirements for column stiffeners and column web reinforcement for the moment
connection and sizes given in Figure 8.13.
(a) Opposite compression flange (Clause 21.3(a) of CSA S16-01):
Br = wc (tb + 10tc) Fyc
= 0.9 x 9.1 (12.7 + 10 x 16.3) x 350/103
= 504 kN

M r 456 103
=
= 1004 kN > 504 kN
db
454
Stiffener required for the force of 1004 - 504 = 500 kN
stiffener width according to beam flange

= (190 - 9.1)/2 = 90 mm

stiffener width according to column flange

= (254 - 9.1)/2 = 122 mm

Note that AISC Specification allows minimum stiffener width to equal one-third of beam flange.
392

Net area of one stiffener

500 103
A s (net) =
= 794 mm 2
2 0.9 350

Gross area of one stiffener

As (gross) = 794 + (K1 - wc/2) x ts

= 794 + (23 - 9.1/2) x ts = 794 + 18.5 ts
= 794 + 18.5 x 12 = 1016 mm2 > 794 mm2

Try 12 mm thick plate

As (gross)

width required

bs = 1016/12 = 85 mm, use bs = 90 mm

b s 90
145
=
= 7 .5 <
t s 12
Fy

OK

(b) Opposite tension flange (Clause 21.3(b) - CSA S16-01):

Tr = 7 (tc)2 Fyc
= 7 x 0.9 x 16.32 x 350 x 10-3
= 586 kN < 1004 kN

stiffener required

Design stiffener: force = (1004 - 586)/2 = 209 kN

Net area required:

209 103
As(net) =
= 663 mm 2 < 794 mm 2
0.9 350

compression stiffener

Use same size as compression stiffeners, 2 plates 20 x 90.

Check shear in column web:
Total shear force = 1004 kN
(Neglect the storey shear, i.e., the shear forces in column above and below the connection)

(c) Shear capacity of column web: Vr = 0.55 x 0.9 x 9.1 x 310 x 350 x 10-3
= 489 kN < 1004 kN
Therefore, stiffeners are required. Use diagonal or web doubler plate.

393

(d) Diagonal Stiffeners:

The force is carried by stiffeners

Cs = 1004 - 489 = 515 kN

This is the horizontal component of the shear force which will be carried by a pair of
diagonal stiffeners. The resultant compressive force in the diagonal stiffeners will be:

Cs
515
=
= 913 kN
cos 0.564
where

cos =

dc
2

dc + db

310
(310) 2 + ( 454) 2

= 0.564

913 10 2
= 2898 mm 2
0.9 350

Net area of stiffeners =

Net area of one stiffener:

= 2898/2 = 1449 mm2

Gross area of one stiffener:

As (gross) = 1449 + 18.5 ts

Try 18 mm thick plate:

As (gross) = 1449 + 18.5 x 18 = 1782 mm2

width of stiffener:

bs =

Check:

145
b s 100
=
= 5.6 <
= 7.8
ts
18
Fy

1782
= 99 mm , use 18 x 100 plate
18

As the tension and compression stiffeners selected were only 90mm wide and the diagonal is
100mm wide, all the stiffeners should be made 100mm wide.
(e) Welding of stiffeners:
Welding the ends of stiffeners
Tension stiffener load = 209 kN
End weld length L = 100 - 18.5 = 82 mm
fillet size required

S1 =

209
= 1.274 kN/mm
2 82

use double fillet 8 mm size 1.62 kN/mm, see Table 3-25, = 90, CISC Handbook

394

Stiffeners to column web weld size

S2 =

209
2 (dc 2k )
209
= 0.439 kN/mm
2 (310 2 36)

Try 5 mm double fillets, 0.778 kN/mm, from Table 3-25, = 90, CISC Handbook
Length required

209
= 134 mm
2 0.778

Use 5 mm x 150 mm for S2.

Compression stiffener load = 500/2 = 250 kN
End weld length

L = 82 mm

Double fillet size

S3 =

250
= 1.524 kN/mm
2 82

Use double fillet 8 mm size 1.62 kN/mm, = 90. In this instance, double fillet is preferable.
Stiffener to column web weld

S4 =

250
= 0.525 kN/mm
2 (310 2 36)

Use double fillet size 5 mm, 0.778 kN/mm, = 0. Since this size has higher strength than
calculated, it is advantageous to check the weld length required:

L=

dc 310
250
= 160 mm >
=
= 155 mm
2
2
2 0.778

Diagonal stiffener load = 913/2 = 457 kN

End weld load:

457
= 2.786 kN/mm
2 82

Use CJPG at both ends. It is too big for fillet weld.

The diagonal stiffeners are under compressive load at both ends as compression struts. The weld
required to column web is nominal to preclude buckling only. Use minimum size fillet, 6 mm (Ss)
for ts = 18 mm.
Note: Web doubler plate can also be used instead of diagonal stiffeners. The AISC Manual of
Steel Construction, LRFD Volumn II - Connections, should be consulted. See Figures 8.14
and 8.15 for final details.
395

S2
8
8

CJP

S1

18

S5
S4
8
8

S3, fillets or CJP

Figure 8.14: Details of welds for Example 3.

PJP

Figure 8.15: Alternative for welding diagonals and top stiffeners (detail A of Figure 8.14).

396

8.8

Welding of Hollow Structural Section (HSS)

8.8.1 CIDECT

The hollow tube is the most efficient compression element. Because of the difficulty in connecting
tubular members, however, their popularity as structural elements has only been fully realized as a
direct result of the development of welding. Hollow structural sections have recently established a
significant hold in the construction marketplace because of their efficiency and aesthetics. Their
connections however, have offered a particular challenge to engineers: a challenge which has been
met in particular by the international research organization CIDECT (Comit International pour le
Developpement et lEtude de la Construction Tubulaire). The CIDECT work has been performed on
statically loaded joints.
Historically, the circular tube or pipe found favour in Europe in relatively light structures, often involving
space frames. Joining the circular shapes however, required complex contour cutting at the ends,
which was eventually simplified by the introduction of automatic cutting machines.
Square and rectangular sections then
began to appear on the market which
offered a new scope because the
end connections were greatly
simplified as they only involved flat
cuts. In fact, HSS trusses can be
successfully fabricated with different
combinations of hollow sections and
open sections. Figure 8.16 identifies
several of these combinations. It is
the square and rectangular sections,
member types RR, which have
become very popular in Canada.
Although massive tubular sections
are often seen in large offshore
structures, the Hollow Structural
Section found in more conventional
structures is defined by CIDECT as
being up to 508 mm in diameter
maximum if round, 400 x 400 mm
maximum if square and 500 x 300
mm maximum if rectangular. It is
this range in size of sections which
we will consider in this book.

Figure 8.16: Combinations of sections

in tubular trusswork.

397

8.8.2 Typical Joint Configurations

All typical joint configurations are possible with HSS (Figure 8.17). The N joint, a Pratt configuration,
the K joint, a Warren configuration and the KT joint are the most relevant in truss design.
In an ideal truss, the centroids of the members at a joint intersect (Figure 8.18). However, in HSS
trusses it may not be practical or desirable to create a zero eccentricity joint. Instead, it may be
essential to build in a certain amount of positive or negative eccentricity through the use of gapped or
overlapped joints (Figure 8.19 (b), (c) & (d)).

8.8.3 Possible Failure Modes

Traditionally, truss joints have been designed on elastic principles such as plane sections remaining
plane to ensure adequate strength through the application of an allowable stress concept. The HSS
joint, however, differs from the traditional concepts because of the flexibility of the tube walls within the
joint. Several different failure modes of the joints in a truss with rectangular hollow sections are
depicted in Figure 8.19. Combinations of these modes are also possible. The HSS joint differs
conceptually from the joints in other trusses because of this flexibility. The manner in which the joint is
welded may not therefore be the critical factor in its performance.
Because of the flexibility of the plate elements in the joint and the resulting redistribution of stresses,
the strength of joints in structural hollow section trusses can best be assessed using Limit States
concepts. Considerable experimental and theoretical work has been undertaken over the years, by
CIDECT in particular, to establish the fitness of a particular combination of joint parameters and loads
for their intended use.

398

Figure 8.18: Eccentricities in

HSS truss joints.

Figure 8.17: Typical joint configurations

in HSS trusswork.

399

Mode F: Local buckling of the

chord face

Figure 8.19: Possible failure modes for HSS truss joints.

400

8.8.4 Joint Capacity

Essentially, the effort has been aimed at establishing the Ultimate Limit State, or maximum load
capacity of a joint. This maximum load capacity may have been reached as a result of instability within
the joint, rupture of a member, transformation of the joint into a mechanism, excessive deformation,
excessive creep or cracking.
Additionally, Serviceability Limit States must be established relating to satisfactory performance under
normal use. Serviceability criteria are associated with excessive deformation, premature cracking,
excessive vibrations or excessive displacements without loss of equilibrium.
The performance of the HSS joint is thus directly dependent upon the geometry, size, wall thickness
and configuration of the various members framing into the joint. The selection of the members in the
truss by a designer to carry the principal axial forces, and the performance of the resulting joint, are
thus closely linked. The truss design engineer must also assume the responsibility for ensuring the
adequate performance of the resulting joint configuration. Unlike many other structures, the fabricator
may have no opportunity for substitution when dealing with an HSS truss.
In 1986, CIDECT produced Monograph 6 The Strength and Behaviour of Statically Loaded Welded
Connections in Structural Hollow Sections. The Monograph summarizes the work of CIDECT since its
inception in 1963. Recommendations, strength equations, and supporting commentary are provided in
the Monograph for the multitude of configurations possible with HSS. The rather voluminous work is
recognized as being incomplete, however, and research continues.
It is not practical to attempt to duplicate the work of CIDECT in this book. For Canadian engineers
familiar with the K and N joints on rectangular and square chord members, design aids based on the
requirements of S16-01 have been presented by Dr. J.A. Packer.

8.8.5 CIDECT Recommendations

Section 3 of Monograph No. 6 has also listed a series of considerations which lead to the development
of efficient hollow section trusses:
1)

Chords should be the most compact section commensurate with requirements for
stability and economy (i.e., keep do/to or ho/to and bo/to as small as possible where do =
diameter of chord, ho = depth of chord, to = thickness of chord, and bo = width of chord).

2)

For gap joints, web to chord wall thickness ratio (tl/to) should be as small as possible
(i.e., keep dl/do as large as possible), where tl and dl are the thickness and width of the
web members respectively. In general, the minimum gap for welding at toes, if fillet
welding is used, should be four times the average thickness of the web members.

3)

For overlap joints, smaller, thicker wall sections are often more suitable.

401

4)

When selecting chord and web members always keep in mind the geometry of the
connection.

5)

Fillet welds should be used to connect web members to chords whenever possible, as
they are more economical than full penetration groove welds.

6)

With thin wall webs and square or rectangular chords with large corner radii, full width
webs should be avoided because difficulties could arise in forming the weld between the
web and the corner radius (Figure 8.20(a)). In some circumstances however, the corner
geometry of the chord can be used with advantage to enable a groove weld to be made
with the minimum of preparation.

Figure 8.20: Side welds in web/chord joints.

Figure 8.21: Insufficient gap between webs.

402

7)

When webs of less than full width of the chord are used, sufficient land should be
allowed to ensure that an adequate size of fillet weld can be obtained (Figure 8.20(b)).
As a rule of thumb, for HSS having a width of 101.6 mm or less, the web should be 2030 mm narrower than the chord (x = 10 to 15). For widths greater than 101.6 mm, x
should be approximately 25 mm.

8)

The intersection of web member centre lines should be kept as near as possible to the
chord centre line. However, the gap or overlap of the web members must meet the joint
strength requirements. Where centre line noding gives unacceptable gap (Figure 8.21)
or slight overlap conditions, the webs should be moved apart to give sufficient gap
(Figure 8.22) or moved together to give a reasonable overlap (Figure 8.23). If this
results in an eccentricity of noding outside the range
e
0.55 <
< 0.25
ho
then the moments generated by this eccentricity must be taken into account in the
design of the connections. Any moments should always be taken into account in the
design of the members.

9)

In lap joints, the tension web members can often be smaller than the compression web
members and this will facilitate fillet welding. Item 7 above also applies.

Figure 8.22: Satisfactory gap joint detail

(not used where fatigue governs the
design).

Figure 8.23: 50% overlap joint detail.

403

10)

In lap joints, overlaps from 30% to 100% can be used. Figure 8.23 shows a joint with
50% overlap.
When using circular webs, or when tension and compression bracings are of
significantly different size, consideration should be given to providing a cross plate as
shown in Figure 8.24 to facilitate fit-up and welding.

11)

Negative eccentricities are preferred to positive eccentricities since negative joint

eccentricities tend to cancel secondary moments.

12)

When overlapping is neither sufficient nor possible, a chord local stiffening plate may be
used as shown in Figure 8.25. Welding, all around, must be used to attach the
plate to the chord.

Figure 8.24: Cross plate to facilitate fit-up of overlapping webs.

Figure 8.25: Gap joint with stiffener plate.

404

8.9

Design Procedures

The challenge of hollow section joint design is to satisfy all the above rules simultaneously. The
procedure is not as difficult as it may at first seem and, with a little practice, the process becomes
automatic. A flow chart for the suggested design procedure is outlined in Figure 8.26. The CISC has
in fact developed a computer program for use in personal computers to assist the design process.
Consideration (5) of CIDECT
Recommendations advises
that fillet welds should be used
to connect web members
whenever possible because
they are more economical than
full penetration groove welds.
In fact, the limited amount of
testing performed by CIDECT
to investigate the actual
influence of welding on the
joints has suggested that
groove welds generally give a
slightly lower ultimate strength
than comparable fillet welds.
In general, the joint strength
criteria developed by CIDECT,
which reflects the failure
modes depicted in Figure 8.19
have, except for perhaps Mode
C, assumed that the welds
would not be the failure mode.

Figure 8.26: Flow chart for HSS truss design procedure.

405

8.10

Sizing Welds

When sizing the fillet welds it must be remembered that nonuniform stress distributions develop in the joints as shown in
Figure 8.27. High stress concentrations develop in the corners
of the tubes. When fitting members together, tack welds should
never be placed at the corners. Finish welds should be
established around the entire perimeter of the joint and never
start or finish in the region of the corners. A suggested welding
sequence is shown in Figure 8.28.
Figure 8.29 illustrates several fillet weld configurations and the
minimum recommended leg size for truss joints to ensure
adequate deformation and rotation capacity in the joint. In many
cases these weld sizes in shear will develop the full member
capacity in tension or compression. These sizes should be
adhered to when the maximum joint strength is required. The
weld sizes could, however, be reduced accordingly as the joint
strength requirements are reduced. However, the structural code
requires a minimum weld size to give a strength not less than
50% of the maximum capacity of the member connected.

Figure 8.27: Stress distribution

in transverse weld under applied
tensile load.

Figure 8.28: Welding sequence for rectangular web-to-chord face.

406

Minimum fillet leg size (S) suggested for HSS truss joints
Fy = 350 MPa and matching E480 electrodes
Web
member
wall
thickness = 60
(t1) mm

Minimum fillet leg size for truss joints, (mm)

= 65

= 70

= 75

= 80

= 85

= 90

= 95

= 100 = 105 = 110 = 115 = 120

3.18

3.81

4.78

10

10

10

10

10

10

10

6.35

10

10

10

10

12

12

12

12

14

14

14

14

7.95

10

12

12

12

14

14

14

16

16

16

16

18

18

9.53

12

14

14

14

16

16

18

18

18

20

20

20

20

11.13

14

16

16

18

18

20

20

20

22

22

22

24

24

12.7

16

18

18

20

20

22

22

24

24

27

27

27

27

Figure 8.29: Fillet weld configurations and minimum recommended

leg sizes for HSS truss joints.

407

8.10.1

PJPG Welds

In tubular joints, because the welding is done from one side (outside), partial penetration groove welds
(PJPG) are normally achieved if special conditions are imposed. These special conditions are required
for complete joint penetration groove welds (CJPG). Figure 8.30(a) shows the common PJPG welds
in tubular connections. To meet the prequalified joint details for PJPG, the joint preparations, fit-up and
groove angles must be within the ranges of one of the details shown.

8.10.2

CJPG Welds

Figure 8.30(b) shows the prequalified joint details for CJPG welds. CSA Standard W59 and AWS D1.1
should be consulted to verify all of the relevant requirements. It should be noted that certain welding
processes are excluded from use on these prequalified joint connections. Also, the welders making the
CJPG welds must be qualified to the T classification requirement in accordance with CSA Standard
W47.1.
When overlapping web members, it is good engineering practice to position the weaker member on top
of the stronger member. In partially overlapped joints, the toe of the overlapped member is not usually
welded (Figure 8.31). In fully overlapped members the preparation and welding detail at the toe of the
overlapped member is particularly important. The toe must be welded to the chord (Figure 8.32).

408

Toe zone

= 90E- 75E

side or heel

Figure 8.30(a): Prequalified joint details for partial joint penetration groove welds
in circular tubular joints.
409

Transition from C to D

Figure 8.30(b): Prequalified joint details for complete joint penetration groove welds in T, Y or K
connections (Standard flat profile for limited thickness).
(From Supplement N:1-M1989 to W59-M1989).
410

Figure 8.31: Partially overlapped

joint.

Figure 8.32: Welds in a fully

overlapped joint.

While HSS trusses are light, strong and graceful, bad fit-up of the structural members can significantly
increase welding and rectification costs. While it is not necessary to have machine fits, time spent to
ensure proper preparation and assembly of the trusses is extremely important.

411

Table S1: Prequalified Joint Dimensions and Groove Angles for Complete Joint Penetration.
Groove Welds for Tubular T, Y, or K Connections Made by Shielded Metal Arc, Flux-Cored Arc,
and Gas Metal Arc (Short-Circuiting Transfer) Welding (1)

End
preparation
( )
Fitup or root
opening (R)

Max
Min
Max

Completed
Weld

FCAW

GMAW

SMAWn FCAW-S
5 mm
5 mm

Min

Joint included
angle ()

Detail A
= 180E to 135E

2 mm
No min
for
> 90E

Max
Min
tw

Detail B
= 150E to 50E
90E

Detail C
= 75E to 30E*

10E or 45E for > 105E

FCAW GMAW

10E

SMAWn
6 mm

FCAW-S 3 mm

FCAW-S**
6 mm for
> 45E
8 mm for
< 45E
2 mm

2mm
No min
for
>120E
90E

2 mm

45E

40E if less use C

> tb for > 90E
> tb/sin for < 90E

> tb

60E for < 105E

W Kmax

SMAWn
GMAW
FCAW**

5 mm

3 mm
6 mm
9 mm
12 mm
40E, if more use
B

>tb/sin but need
not exceed 1.75
tb Weld may be
built up to meet
this

Detail D
= 40E to 15E

25E to 40E
15E to 25E

30E to 40E
25E to 30E
20E to 25E
15E to 20E

>2tb

> tb/sin but need

not exceed 1.75 tb

*Not prequalified for groove angles () under 30E.

tOtherwise as needed to obtain required ().
KInitial passes of back-up weld discounted until width of groove (W) is sufficient to ensure wound welding; the
necessary width of weld groove (W) provided by back-up weld.
nThese root details apply to SMAW and FCAW (self-shielded), qualified in accordance with Table S1 in W47.1-S.
**These root details apply to GMAW (short circuiting transfer) and to FCAW (gas shielded), qualified in
accordance with Table S1 in W47.1-S.
Note: (1) For GMAW see Clause S4.3.3. These details are not intended for GMAW (spray transfer).

Table S3: Joint Detail Application

Detail*
A
B
C
D

Applicable range of local dihedral angle

180E to 135E
150E to 50E
75E to 30E Not prequalified for groove
40E to 15E angles under 30E

* The angle and dimensional ranges given in Details A, B, C or D include maximum allowable tolerances.
Notes: (1) The applicable joint detail (A, B, C or D) for a particular part of the connection is determined by
the local dihedral angle, y, which changes continuously in progressing around the branch member.
(2) Local dihedral angle is the angle measured in a plane perpendicular to the line of the weld between
tangents to the outside surfaces of the tubes being joined at the weld.
412

Chapter 9
Weld Faults and Inspection

Table of Contents
9.1

Introduction

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .415

9.2

Weld
9.2.1
9.2.2
9.2.3
9.2.4

9.3

Distortion or Warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .420

9.3.1 Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .420

9.4

Dimensional Faults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .422

9.4.1 Incorrect Weld Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .422
9.4.2 Incorrect Profile and Size of Lap Weld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .432
9.4.3 Out of Line Weld Beads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .433

9.5

Structural Faults in the Weld Zone .

9.5.1
Gas Inclusions . . . . . . . . .
9.5.2
Causes of Porosity . . . . . .
9.5.3
Moisture . . . . . . . . . . . . . .
9.5.4
Parent Material . . . . . . . . .
9.5.5
Surface Contaminations . .
9.5.6
Insufficient Flux Coverage
9.5.7
Slag Residue . . . . . . . . . .
9.5.8
Shielding Gas . . . . . . . . . .
9.5.9
Welding Techniques . . . . .
9.5.10 Slag Inclusions . . . . . . . . .
9.5.11 Tungsten Inclusions . . . . .
9.5.12 Copper Inclusions . . . . . .
9.5.13 Oxidation . . . . . . . . . . . . .

Fault Classifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .416

Dimensional Faults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .416
Dimenstional Faults Prior to Welding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .416
Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .417
Incorrect Joint Preparation and Fit Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .418

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

413

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.434
.434
.435
.436
.436
.436
.437
.437
.437
.438
.438
.440
.440
.440

9.6

Fusion Faults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .441

9.6.1 Incomplete Fusion (Lack of Fusion) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .441
9.6.2 Incomplete Penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .444

9.7

Cracking
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .445
9.7.1 Solidification Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .445
9.7.2 Hydrogen Induced Cold Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .446

9.8

Surface Defects (Irregularities) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .450

9.9

Defective Properties (Weld Metal and Joint) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .452

9.10

Summary of Weld Faults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .452

9.11

Welding Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .453

9.12

Methods of Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .455

9.12.1 Visual Welding Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .455
9.12.2 Liquid Penetrant Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .462
9.12.3 Magnetic Particle Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .464
9.12.4 Radiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .467
9.12.5 Ultrasonic Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .475

414

9.1

Introduction

Before we know what weld faults are, we should know what a good weld is. A good weld has complete
fusion to base metal, good profile, no porosity or inclusions, no cracks, with adequate mechanical and
metallurgical properties. If you know how a weld is made, you will realize that it takes a concerted
effort of all parties of the welding operation to meet all these conditions and achieve a good weld.
Welding involves base metal, filler metal, machinery, electricity and human dexterity. When all the
other factors are at an optimum, the human skill plays a decisive role. Welding is a relatively refined
manufacturing process, not the same as concrete construction, which can tolerate more in dimension
and handling. Remember that weld metal solidifies almost instantaneously. Concrete takes a few
hours to set.
Therefore, when you read the words weld faults, do not be deterred. After all, a weld dimension may
be only from a few millimetres (fillet weld, for instance) to a few centimetres (butt joint of thick plate),
and the load it carries can be from a few kilonewtons (a few 100 lbs), to thousands of tonnes. The
load path is much more critical than concrete structures and so is the design analysis, but a properly
trained and certified welding fabrication shop can always produce sound weldments. Weld fault means
repair. Codes or standards stipulate the tolerance and seriousness of different types of weld faults.
This chapter outlines the types of faults, causes and methods of detection (inspection). Inspection is
another branch of technology. It requires special training, both in theory and practice, to be a qualified
inspector. The following pages will provide the reader with an introduction to this subject. Further
reading can be obtained in the following Modules:
Module
Module
Module
Module
Module
Module
Module

10
11
12
16
17
18
19

Weld Faults and Causes

Basic Inspection Technology
Mechanical Testing of Welds
Techniques of Visual Inspection
Surface Inspection
Radiographic Inspection
Ultrasonic Inspection

415

9.2

Weld Faults Classifications

Faults in welding may range from inadequate metallurgical properties to such physical imperfections as
cracks, porosity, slag inclusions, incomplete fusion, undercut, incomplete penetration, dimensional
defects, etc. The importance of weld defects, however, both as to type and quantity, is relative to the
type of weldment and the service required; an imperfection harmful in one case need not be so where
other types of welded work are concerned. It is, therefore, a difficult task to assess their relative
importance, since the extent and type of a weld fault needs to be analyzed in relation to the function of
a given weldment. Where existing experience is inadequate, this should be done by experimental
research including necessary tests to establish standards of acceptability.
There are certain types of defects that may occur in arc welding and they are of three general classes:
1.
2.
3.

Dimensional defects
Structural discontinuities in the weld
Defective properties (weld metal and joint)

These classes of defects can be subdivided under many headings, but since it is impossible to state
rules where by an inspector can identify all the factors likely to cause defects in the welds, this chapter
will describe only some of them briefly. An inspector will be better fitted to judge the chances of
obtaining welds which are satisfactory for a particular service if he has a thorough knowledge of the
limitations of a given welding process and an understanding of those conditions that are likely to cause
the formation of defects.

9.2.1 Dimensional Faults

The production of satisfactory weldments depends upon, among other things, the maintenance of
specified dimensions, whether it be size and shape of welds or the finished dimensions of an
assembly. Requirements of this nature will be found in the drawings and specifications. Departure
from the requirements in any respect should be regarded as a dimensional defect that must be
corrected before final acceptance of the weldment. The more common types of these defects are
discussed as follows.

9.2.2 Dimensional Faults Prior to Welding

a)
b)
c)
d)
e)
f)

Incorrect bevel angles

Incorrect J-groove radii
Incorrect root face
Incorrect fit up (mismatch)
Incorrect root openings
Irregularities in the surface of the joint preparation

416

9.2.3 Causes
Faults described previously (and illustrated in Figure 9.1) are the direct result of poor workmanship in
operations leading up to the point at which the assembly is to be welded.
Faults of this nature indicate a lack of quality control and should be reported by the welding inspector
so that corrective action may be initiated. Figure 9.1 illustrates some dimensional faults.

Figure 9.1

417

As illustrated in Table 9.1, codes and specifications provide for tolerances on bevel angle, root face
thickness and root openings, and should be followed accordingly, e.g., a groove angle specified on the
drawing as 60 should be between 55 and 70.

Table 9.1: CSA Standard W59 and AWS D1.1

(not applicable to electroslag or electrogas welding).
Root Not Gouged
1. Root Face of Joint

Root Gouged

1.6 mm (1/16)

Not limited

- Without Steel Backing

1.6 mm (1/16)

1.6 mm (1/16)
3 mm (1/8)

- With Steel Backing

6 mm (1/4)
1.6 mm (1/16)

Not applicable

+ 10 degrees
- 5 degrees

+ 10 degrees
- 5 degrees

2. Root Opening of Joints:

3. Groove Angle of Joint

Incorrect joint fit up also represents difficulties in producing sound weld deposits. Care should be taken
to meet the fit-up tolerances to avoid the following weld faults:
Insufficient Root Openings
g
g
g

lack of penetration
lack of fusion
slag entrapment

Excessive Root Openings

g
g
g
g

porosity
slag entrapment
excessive weld reinforcement
additional distortion

9.2.4 Incorrect Joint Preparation and Fit Up

Good welding practice requires proper joint dimensions and preparation. Improper joint preparation
makes it exceedingly difficult for the operator to make a sound weld and greatly increases the tendency
to produce structural discontinuities in the weld. Therefore, it is important that the joint preparation
meet the applicable welding standards within specified limits.
418

Incorrect joint preparation could be caused by one, or a combination of the following:

g
g
g

improper bevel angles or J-groove radius

improper root face
irregularities in the finished surface

Irregularities in the finished surface to be welded may also lead to various weld faults and defects. The
method or preparation usually determines the type of weld fault that may be experienced as illustrated
below.

Sheared Surfaces
Depending on the condition of the shear blades and lubricants used, various undesirable foreign
materials may be entrapped, leading to porosity, slag entrapment and lack of fusion.

Flame Cut Surfaces

When oxygen is used in flame cutting, notches and irregularities may occur. Quite often, slag may
adhere to these notches and surfaces, and if it is not removed prior to welding, such faults as porosity,
lack of fusion, slag entrapment and chemistry composition defects may occur.
Codes, standards and specifications often limit surface irregularities and should be followed
accordingly. For example, when welding in accordance to CSA Standard W59 specifications (as in
Figure 9.2), the following conditions and limitations are to be applied.

Preparation of Material
Surfaces and edges to be welded shall be smooth, uniform, and free from fins, tears, cracks and
other defects which would adversely affect the quality of strength of the weld. Surfaces to be welded
shall also be free, within 50 mm (2 in) of any weld location, from loose or thick scale (except for
tightly adhering small islands of scale), slag, rust, paint, grease, moisture and other foreign material
that will prevent proper welding or produce objectionable fumes.
Occasional notches, not more than 5 mm (3/16 in) deep, on otherwise satisfactory surfaces shall be
removed by machining or grinding. Occasional notches, exceeding 5 mm (3/16 in) and less than
10 mm (7/16 in) deep, in oxygen cut edges of plate up to 100 mm (4 in) thick, not to be welded, may,
with the Engineers approval be repaired by welding. For material 4 in thick or over, the depth of the
notch shall not exceed 15 mm (5/8 in).
Such repairs shall be made by suitably preparing the defective area, welding with basic electrodes to
an approved procedure and grinding the completed weld smooth and flush with the adjacent surface
to produce a workmanlike finish.

Figure 9.2: Excerpt from CSA Standard W59.

419

Dimensional Faults After Welding

9.3

a)

Distortion

b)

Incorrect weld profile such as:

i)
Convexity
ii)
Concavity
iii)
Insufficient throat
iv)
Insufficient leg
v)
Excessive reinforcement
vi)
Undercutting (internal & external)
vii) Overlap
viii) Out-of-line weld beads

Distortion or Warpage
9.3.1 Causes

The welding operation involves the application of heat and the fusion of metal in localized sections in
the weldment. Stresses of sufficient magnitude may be induced (due to thermal expansions and
contractions), which will cause distortion of the structure.
Distortion may have a number of contributing causes such as:
g
g
g
g
g
g

lack of control of heat input

inadequate control of weld pass sequencing
inaccurate preparation of the joint
inadequate control of the fit up
incorrect joint design
over-welding

Various codes and specifications provide dimensional tolerances as illustrated in Figure 9.3. It should
be noted that other codes, such as AWS D1.1, for example, have similar specifications.

420

CSA W59
Dimensional Tolerances
Unless otherwise specified in the applicable design Code or Standard, the dimensions
of welded structural members shall be within the following special tolerances:
Imperial
Lengths of 45 feet and under:
3
1 Number of feet of test length
but not over "
"
8
10
8

Lengths over 45 feet:

3 1 (Number of feet of test length 45)
"+ "
10
8 8

Metric
Length of 14 mm and under:
L
mm but not over 10 mm

1000

Length over 14 mm:

10 + [ (L - 14000) / 1000] mm
L = mm in test length

Figure 9.3: Dimensional Tolerances

421

9.4

Dimensional Faults
9.4.1 Incorrect Weld Profiles

Weld deficiencies related to weld profiles are illustrated in Figures 9.4 & 9.5.

Figure 9.4

422

Figure 9.5: Weld profiles - acceptable and defective with various faults illustrated
in accordance with CSA Standard W59 (AWS D1.1 is similar).

423

Overlap is a condition where an excess of weld metal exists at the toe of a weld beyond the limits of
fusion and is illustrated in Figures 9.6 & 9.7. This condition produces notches, which are harmful due
to a resultant stress concentration under load, and, in the case of a fillet weld, may actually reduce its
effective size.

Figure 9.7: Overlap in groove weld.

Figure 9.6: Overlap in fillet weld.

If the nature of overlap is examined, it will be found that there is a mass of weld metal that is not fused
to the parent metal. Overlap is common in fillet and groove welds with various processes, and typical
causes of overlap are as follows:
Improper Technique, including
g
g
g

travel speed is too slow

improper electrode angles
improper weave techniques

Essential Variables, including

g
g

insufficient electrode diameter

improper amperage and voltage settings

Joint Preparation Contaminants, including

g
g
g
g

oil
paint
rust
mill scale

424

Excessive convexity tends to produce notch effects in multipass welds, and may lead to other weld
faults, such as slag inclusions, lack of fusion and porosity when depositing subsequent passes.
The term convexity normally refers to the profile of a fillet weld (Figure 9.8) whereas excessive weld
reinforcement refers to the profile of a groove weld, as illustrated in Figure 9.9.

Figure 9.8: Excessive convexity in

a fillet weld.

Figure 9.9: Excessive reinforcement

in a groove weld.

Excessive convexity, like overlap, may be caused by inhibited weld metal fluidity. Typical causes of
convexity may be one, or a combination of the following:

Improper Techniques
g
g
g

travel speed too slow

incorrect electrode angles
incorrect weave techniques

Essential Variables
g
g

insufficient electrode diameter

insufficient amperage and voltage

Joint Preparation Contaminants

g
g
g
g

oil
paint
rust
mill scale
425

Excessive welding reinforcement is associated with groove welds and is undesirable since it tends to
stiffen the section at that point as well as establish notches. This condition results from improper
welding technique, or insufficient welding current, and is shown in Figures 9.10 & 9.11. This fault is
often connected with some irregularity in weld profile and is illustrated in Figure 9.9. The opposite of
this defect is insufficient reinforcement in the groove weld like that shown in Figure 9.12.

Figure 9.10: Excessive weld reinforcement.

Figure 9.11: Excessive root reinforcement

in a single V butt.

Codes, specifications and standards limit the amount of reinforcement on groove welds and should be
followed accordingly. The maximum reinforcement permitted by CSA Standard W59 is 1/8 (3 mm) for
groove welds (see Figure 9.5).
Insufficient weld reinforcement, also associated with groove welds, is considered undesirable. The
effective load capacity is reduced considerably if not properly corrected. As illustrated in Figure 9.13,
additional passes should be added to bring the weld to a proper size, but care should be taken when
applying additional passes to:
g
g
g
g

maintain proper profiles

not exceed reinforcement requirements
blend passes into base material
not create additional weld faults

426

Figure 9.12: Insufficient face reinforcement

(underfill).

Figure 9.13: Correction for insufficient

reinforcement.

Undercut
This term describes the melting away of the parent metal during the welding process. If undercutting is
not corrected, it may be detrimental to the component and is, therefore, a fault. Undercut will produce
notches and result in stress risers, which can be harmful under load. Limitations for undercut are
specified in governing codes and standards and are based on the type of loading that the weld is
subjected to (i.e., static, dynamic or cyclic).
Undercut can occur at any stage of the welding process, for example:
g

root undercut in a singe V butt weld without back welding (Figure 9.14)

undercutting of the sidewall of a welding groove at the edge of a layer or bead, thus
forming a sharp recess in the sidewall at a point where the next layer or bead must fuse
(Figure 9.15)

reduction in base metal thickness at the line where the last bead is fused to the surface
(Figure 9.16 external undercut)

427

Figure 9.14

Figure 9.15

Figure 9.16

Undercutting of the side walls of a groove does not affect the completed weld if sufficient care is taken
to correct the condition before depositing the next bead. Failure to correct the condition may lead to
slag being trapped in the cavity during the welding of the next pass. Surface undercutting, both
internal and external, should be corrected. However, some construction codes and standards allow
limited amounts of undercut to remain in the weld. For example, CSA Standard W59 and AWS D1.1
state that undercut for cyclically loaded structures shall not be more than 0.010 inch (0.25 mm) deep
when the weld is transverse to the primary stress in the part that is undercut. They further state that
undercut shall be no more than 1/32 inch (1 mm) deep when the weld is parallel to the primary stress
in the part that is undercut.

428

On the other hand, the designer or specifier may specifically state in a product specification that
undercut in any degree is not allowed.
Some of the probable causes of undercut are as follows:

Operator Technique
g

too much current on too long an arc may increase the tendency to undercut

Electrode
g

different types of electrodes show varying characteristics in this respect

Joint Accessibility and Position

g

with some electrodes the most skilled operator may be unable to avoid undercut under
certain conditions such as accessibility and position

Joint Preparation
g

inadequate root face may cause excessive internal undercut (Figure 9.17)

Figure 9.17

Excessive concavity may occur in the root pass of a groove weld (Figure 9.18), but is more often
associated with fillet welds (Figure 9.20). It should be noted that drawings may call for concave fillet
welds, in which case it would not be considered a weld fault. The size of a concave fillet weld is
determined by its throat size, not the actual measurement of its leg length. A concave fillet profile is
dependent on service conditions.
As illustrated in Figures 9.19 & 9.20, note that an excessively concave weld profile gives a deceptive
appearance as to its actual size. Figure 9.21 indicates corrective action for concave fillet welds.

429

Figure 9.18: Root concavity.

Figure 9.19: Concave fillets.

Figure 9.20: Excessive concavity.

Lack of root penetration.

Figure 9.21: Corrective passes for

a concave fillet.

Insufficient weld reinforcement may be caused by any one or a combination of the following operator
manipulation techniques:
g
g
g
g

travel speed too fast

insufficient passes or layers
incorrect weave techniques
excessive included groove angles

430

Weld deficiencies due to insufficient or excessive size and poor profile may be detected by visual
examination, or by the use of suitable gauges as illustrated in Figure 9.22.

Figure 9.22: Multi-purpose welding gauge.

Typical causes of concavity can be divided into the following categories:

a)
b)
c)
d)
e)
f)

incorrect operator manipulation

change in the essential variables
inadequate joint geometry
position of welding
process behaviour
material type

Each of these categories may be sources contributing to concavity either individually, or in

combination.

431

9.4.2 Incorrect Profile and Size of Lap Weld

The exposed corner of the upper plate is melted off along the length of the weld, reducing the length of
the vertical leg and consequently the designed throat size of the weld, as illustrated in Figures 9.23 &
9.24.

Figure 9.23: Section showing actual weld size with reduced throat thickness.

Meltdown
This edge should
be visible
Correct size,
Fig. 9.5 (b)

Reduced Throat Thickness

Compare Fig. 9.5 (c)

Figure 9.24: Reduction in throat thickness.

432

The upper edge should just remain visible, or,

failing this, the weld fault should be corrected by
the addition of another weld pass as illustrated in
Figure 9.25.
Melting the upper edge may be caused by:
g

g
g
g
g
g

inadequate operator manipulation

(slow travel speed, wrong electrode
angle)
process behaviour
essential variables (excessive current
or voltage)
material type
position of weld
inadequate joint geometry

Figure 9.25: Restoration to correct size

by addition of a weld pass.

Proper manipulation by the operator is usually

determined by the above mentioned categories and
may be the cause of the weld fault. Each of the above categories may contribute to the operator
technique.

9.4.3 Out-of-Line Weld Beads

Causes
The following causes can lead to misalignment of
the weld (Figure 9.26):
g
g
g
g

insufficient care in positioning

automatic welding machines
incorrect bead placement by the welder
incorrect edge preparation
careless chipping out of the back side
of welds
Figure 9.26

433

9.5

Structural Faults in the Weld Zone

Gas Inclusions (Porosity)
g
g
g
g
g
g
g

isolated gas holes

worm holes (elongated gas holes)
piping
hollow root (suck back)
scattered porosity
grouped porosity
christmas tree porosity

Inclusions
g
g
g
g
g
g
g

isolated slag inclusions

slag lines
slag entrapment behind backing strip
slag inclusions missed by back gouging
(double-V weld)
tungsten inclusions
copper inclusions (from carbon arc-air
operations)
slag from laminations in parent material

9.5.1 Gas Inclusions

The term porosity is used to describe gas pockets trapped in the solidifying weld metal. Porosity may
manifest itself in a variety of patterns, sizes, shapes and quantities. Porosity may be present in any
position in the deposited weld metal. Some porosity may appear on the surface of a weld, and
therefore, can be detected visually. However, when the porosity is sub-surface, special testing such as
radiography or ultrasonics is necessary to disclose it. Examples of porosity are illustrated in Figures
9.27 9.31.

Figure 9.27: Sever surface porosity

(sulphur or moisture).

Figure 9.28: Porosity at root of the joint.

434

Figure 9.29: Severe piping at lap joint.

Figure 9.30: Radiographic image of

piping porosity resulting from use
of wet basic electrodes.

Figure 9.31: Radiographic image of porosity in aluminum weld.

9.5.2 Causes of Porosity

In multi-pass welding, the location of porosity in relation to the depth over the cross-sectional area of
the weld may assist in determining the probable cause.
In many cases porosity is cumulative as subsequent passes are deposited. To preclude building up
the density of the porosity to a point where a completed weld would be unacceptable, the porosity
should be removed entirely prior to the addition of further passes.

435

The probable causes may be categorized as follows:

a)
b)
c)
d)
e)

Moisture
Chemistry and structure of the parent material
Surface impurities and contaminants
Faulty electrodes, fluxes, shielding gases or slag
Operator techniques

9.5.3 Moisture
Moisture pick-up in flux-coated electrodes or on the surface of flux-coated wire will cause porosity. The
same situation pertains to externally applied flux in welding processes such as submerged arc and
electroslag.
To avoid moisture, the consumables, including fluxes, should be stored under controlled conditions.
Various codes and standards may require procedures for the proper storage of weld consumables,
such as CSA Standard W59 and AWS D1.1. Storage conditions will be governed by the type of flux,
with basic fluxing systems requiring storage temperatures above 250F (120C). This ensures
moisture levels are kept at an acceptable level to produce a weld deposit with a low hydrogen
designation. For details, see Module 6.

9.5.4 Parent Material

It is important to select the proper filler metal to match the chemistry of the material to be welded. In
cases of relatively high sulphur content, porosity is commonly encountered. Other elements, such as
zinc in galvanized steels, may also create excessive porosity after welding.
Materials with dense oxides, such as aluminum, should be carefully cleaned. Dense oxide layers can
become contaminated with moisture or oils, etc. and cause porosity. Laminations in plate may also be
a source of porosity in the welding operation.

9.5.5 Surface Contaminations

When fabricating metals, the surfaces may be in contact with certain contaminants that can cause
porosity. Some of these contaminants are as follows:
g
g
g
g

oil
grease
paint
oxide

Rust and mill scale can also absorb contaminants and become a source of porosity.
436

The method of preparing material for welding often introduces contaminants, and some of these
methods are:
g
g
g
g
g

shears
band saws
abrasive grinding wheels
mechanical nibblers
oxy-fuel apparatus

Malfunctioning tools, such as air grinders, air chipping tools or air scaling guns may deposit films of oil,
grease or moisture on the surfaces to be welded.
Where aluminum is being welded, tools used on other materials, such as steels, may introduce
contaminants that will cause porosity. Care should be taken to use tools designated for aluminum only.
These tools, such as stainless steel wire brushes, must be kept clean and separated from general use
to reduce the chances of contamination.
Carbon steel wire brushes used on stainless steels may also be a cause of porosity.
Contaminants that cause porosity may be picked up during recovery operations of unused flux during
or after submerged arc welding operations.

9.5.6 Insufficient Flux Coverage

Insufficient flux covering in submerged arc welding may be a cause of scattered surface porosity.

9.5.7 Slag Residue

Slag left on the surface of tack welds or internal weld beads may cause porosity.

9.5.8 Shielding Gas

Porosity associated with shielding gases is often caused by poor distribution within the arc and
surrounding areas, insufficient or excessive shielding gas flows, or impurities collected in the gas
through hoses, connections and the torch or gun assemblies.
When setting up for GMAW of aluminum using pure argon, hoses, regulators and cables (which have
been used exclusively for welding steel with CO2 or CO2 mixtures) will be a possible source of
contamination and subsequent porosity if they are not replaced or cleaned properly.
Loose fittings and connections may allow atmospheric gases to enter the gas hoses and assemblies
and cause porosity.
437

As illustrated in Figure 9.32, upon welding an outside corner with GMAW, .035 E70S6 and Ar CO2
shielding, a gas nozzle inner diameter less than 13 mm (1/2 ) could cause porosity.

Figure 9.32: Effective shielding for outside corner joint.

If the distance the cup is held from the work is incorrect, it may cause porosity (i.e., 20 mm () is
acceptable and 40 mm (1 ) is not).

9.5.9

Welding Techniques

In manual welding applications, the following may cause porosity:

g
g
g
g

faulty manipulation of the electrode

excessive arc voltages
incorrect electrode angle
incorrect weave techniques

9.5.10 Slag Inclusions

This term is used to describe oxides and other non-metallic solids that are sometimes found as
elongated or multifaceted inclusions in welds.
Slags are always produced when welding with covered electrodes, and they serve as scavengers of
impurities in the molten metal pool. In addition, they form a blanket over the weld to control the cooling
rates and exclude atmospheric oxygen from the hot metal surface.

438

During the welding process, fluxes form slag that is forced below the surface of the molten metal by the
stirring action of the arc. Slag may also flow ahead of the arc causing the metal to be deposited over
it. In any case, it tends to rise to the surface because of its lower density.
A number of factors may prevent its release and result in the slag being trapped in the weld metal.
Some of these factors are:
g
g
g
g
g

high viscosity (thick) weld metal

rapid solidification
too low a temperature
improper manipulation of the electrode
undercut on previous passes

One other potential cause of slag is foreign material entrapped in laminations in the joint preparation.
In multi-pass welding, insufficient cleaning between weld passes can leave portions of the slag coating
in place which is then covered by subsequent passes. Such slag inclusions are often characterized by
their location at the edge of the underlying metal deposits, where they tend to extend longitudinally
along the weld.
In making a root pass the electrode may be so large that the arc strikes the side of the groove instead
of the root. The slag will roll down into the root opening and become trapped under the root layer
because the arc failed to heat the root area to a sufficiently high temperature to allow the slag to float
to the surface.
Slag lines can be either intermittent
or continuous. If the prior pass
produces a bead that is too convex,
or if the arc has undercut the joint
surface, it will be difficult to remove
the slag between the surface of the
groove and the deposited metal.
When the slag is left in place it is
covered by subsequent passes
(Figure 9.33).

Figure 9.33

Figure 9.34: Elongated slag inclusions.

439

The majority of slag inclusions may be prevented by proper preparation of the groove before each
bead is deposited (including sufficient cleaning), and using care to correct the contour that would be
difficult to penetrate fully with the arc.

9.5.11

Tungsten Inclusions

Tungsten inclusions are characteristic of the inert atmosphere welding methods. If the tungsten
electrode comes into contact with the weld metal, tungsten particles can be trapped in the deposited
metal. These may be in the form of small pieces of the tungsten wire. Due to its high melting point,
fusion of the tungsten to the deposited weld metal does not occur.

9.5.12

Copper Inclusions

This type of inclusion occurs when pieces of the copper sheath of a carbon arc-air electrode fall into
the groove and are subsequently welded over. Continuous electrode processes use copper contact
tips and copper alloy nozzles. If these parts contact the weld pool, copper inclusions can be created.
Another cause of copper inclusions may occur during magnetic particle testing of welds. The current
supplied to create the magnetic field may be passed through copper conductors (prods). If there is
poor contact of the prods to the steel when the current is applied, sparking will occur, and copper
particles may be melted into the structure. This type of problem should be carefully controlled due to
the propensity for crack propagation from the embedded inclusions.

9.5.13

Oxidation

In pipe and tube welding of components for critical service (i.e., nuclear plant) some specifications
forbid the presence of oxides on the internal surface of the welds. In these cases, the internal surface
of the pipe/tube is purged with a constant supply of inert gas. If the gas flow is inadequate, oxides will
form and cause the weld to be rejected. Control of the gas supply is, therefore, an essential operation
to produce sound welding.

440

9.6

Fusion Faults
g
g
g

incomplete fusion (inert)

incomplete sidewall fusion
incomplete root fusion

g
g
g

incomplete fusion in fillet welds

underbead non-fusion/cold lap
incomplete penetration

9.6.1 Incomplete Fusion (Lack of Fusion)

The term incomplete fusion is used to describe the failure to fuse weld metal to the base material, or
adjacent layers of weld metal to each other. Failure to effect fusion may occur at any point in the
welding groove or fillet weld as illustrated in Figures 9.35 to 9.42.

Figure 9.35: Incomplete fusion at root

and along joint face.

Figure 9.37: Incomplete fusion at root.

Figure 9.36: Incomplete fusion along joint face.

Figure 9.38: X-ray of incomplete fusion at root.

441

Figure 9.39

Figure 9.40

Figure 9.41: Incomplete fusion at root of fillet.

Figure 9.42: Incomplete fusion at root of

J-groove weld in thick section.

Incomplete fusion may be caused by a number of factors, either singly or in combination. Some of
these factors are listed below:
a)

Using too large an electrode for a narrow preparation

b)

Using the wrong type of electrode

c)

Insufficient welding current, resulting in failure to raise the temperature of an adequate

amount of base material to the melting point

d)

Improper manipulation of the electrode

e)

Failure to dissolve, by proper fluxing, the oxides or other foreign materials on the
surfaces to which the weld metal must fuse

f)

Poor joint design. As an example, a narrow Vee groove in a thick plate would limit
manipulation of the electrode. This would increase the probability of non-fusion of the
weld metal to the parent metal (Figure 9.43)

g)

Inadequate shielding gas (if used)

442

Figure 9.43

Figure 9.44: Operator techniques.

443

9.6.2 Incomplete Penetration

The term incomplete penetration describes the failure of the deposited weld metal to fuse integrally
with the parent material at the root of the weld joint (see Figure 9.45).

Figure 9.45

It must be noted that incomplete penetration is not necessarily a weld fault. Some welded connections
are designed with partial penetration welds. Incomplete penetration becomes a weld fault when the
codes, specifications and designs require complete penetration.

45

D = 12 mm min.

25 mm

ETT = 9 mm min.
D = 12 mm min.

ETT - Effective Throat Thickness

Figure 9.46: Partial penetration groove weld.

444

The causes of incomplete penetration are very similar to those causing lack of fusion, and are listed
below:
a)
b)
c)
d)
e)
f)

9.7

Using too large an electrode for a narrow joint

Using the wrong type of electrode
Insufficient welding current, resulting in failure to penetrate and fuse the root faces
Improper manipulation of the electrode (Figure 9.44)
Poor joint design
Poor fit up causing inadequate gap between the root faces.

Cracking
9.7.1 Solidification Cracking

During the solidification of weld metal,

grains begin to grow from the fusion
boundary towards the central region of
the weld pool.
Some alloying elements and impurities
are rejected ahead of the growing
crystals. Their presence lowers the
freezing temperature substantially below
that of the first liquid to solidify. As
solidification takes place, the weld and
surrounding material are progressively
cooling, and this gives rise to contraction
strains across the weld.
When solidification is almost complete,
and grains begin to meet, the low
melting point liquid may lead to such low
ductility that the contraction strains
produce cracking.

Figure 9.47: Solidification cracks.

Hot Cracks
The development of hot cracks in welds results from the combined effects of metallurgical and
mechanical factors. Some metals are prone to hot cracking, e.g., high temperature alloys and high
sulphur steels.

445

9.7.2 Hydrogen-Induced Cold Cracking

Cold cracks may occur in the weld metal or in the heat affected zone. Cold cracks in the weld metal
may occur in any orientation with respect to the weld axis, but the commonly observed positions are
illustrated in Figure 9.48.

1
2

1. Transverse crack in weld metal

2. Transverse crack in heat affected

zone

BUTT JOINT

3. Toe crack
4. Weld metal crack

1
5
2

5. Root crack

4
3

6. Underbead crack

TEE JOINT
Figure 9.48: Commonly observed positions of cold cracks in butt and fillet welds.

Transverse cracks in the weld metal may extend into the heat affected zone of the parent plate and
beyond.
HAZ cracks are usually longitudinal and most often occur at the root or the toes of the welds. Under
some conditions, longitudinal cracks may be very long, sometimes running the entire length of the
weld.
Cold cracking may also manifest itself as fine micro-cracks that are difficult to detect by normal
inspection and non-destructive test methods. The presence of micro-cracks may be symptomatic of a
more serious condition (such as high hydrogen level) that could lead to more serious cracking.
Figures 9.49 & 9.50 show typical cold cracking and commonly observed positions. As the name
implies, the cracks form at low temperatures generally below 200C (390F). In most cases, cracks
occur at room temperature when the weld has completely cooled.

446

Cold cracks are often delayed. Even after the joint has cooled to room temperature there may be a
further lapse of time before cracking occurs. This may be a few minutes or several hours, although in
some extreme cases cracks have been observed to form several weeks after welding is complete.

Figure 9.49: Typical cold crack in

the heat affected zone.

Figure 9.50: Typical cold crack

in the weld metal.

The causes of hydrogen-induced cold cracking are complex and cannot be fully documented in this
text, however, a brief list of some of the causes is as follows:
g
g
g

hydrogen from coated electrodes

hydrogen from external sources in the base material, i.e., hydrogen sulphide
insufficient pre- and post-weld heat treatment

Tack welds left for inclusion in the completed weld may be the cause of cracks. If the tack weld is
made on a cold surface of a large mass compared to the size of the tack, the result is a rapid quench.
If the tack is badly made or of insufficient size, a crack may readily occur.
Crater cracks occurring during solidification are more likely to form in a long crater, Figure 9.51 (a),
where the columnar crystals form from each side of the joint, at right angles to the axis of the weld.
This leaves a plane of juncture subject to cleavage as the metal shrinks. A short crater displays
columnar crystals radiating from the centre as shown in Figure 9.51 (b).
Figures 9.52 to 9.59 illustrate a variety of weld and HAZ cracks.

447

Figure 9.51

Figure 9.52: Longitudinal crack in butt joint.

Figure 9.53: Crack in fillet

lack of penetration.

Figure 9.54: Hot crack in deep

penetration fillet.

Figure 9.55: Crater cracks.

448

Figure 9.56: Root crack in first

pass of double-V butt.

Figure 9.57: Root crack in thick U butt.

Figure 9.58: Solidification crack.

Figure 9.59: Fusion line crack in low-alloy

steel (underbead crack).

Some of the causes of cracking are listed below:

1)
2)
3)
4)
5)
6)
7)
8)
9)
10)

Hydrogen content of electrode

Hydrogen impregnation of the parent material
High sulphur or phosphorous content of the base material
High carbon content of the base material
High restraint on the joint
Rapid cooling of hardenable and brittle material
Welds too small for the size, rigidity and quenching effect of the parts joined
Poor joint fit up
Unsuitable electrodes
Secondary faults such as lack of penetration, porosity, elongated craters, etc.
449

9.8

Surface Defects (Irregularities)

Sometimes conditions are encountered during the

welding which result in holes in the surface of the
deposit (Figure 9.60). This is generally considered
the result of a highly reducing atmosphere. Such a
condition is most likely to be encountered at the
bottom of a narrow groove where the air is
completely excluded and no normal reaction takes
place between the arc atmosphere and the
surrounding air. The base metal being welded can
be a factor (sulphur, moisture in both base metal and
electrode, as mentioned before) but leaving this out
of consideration, improvement is usually obtained by
changing the electrical conditions such as current
and polarity. Often an increase of arc length will
correct this condition, but it may be necessary to
change the type of electrode.

Figure 9.60: Severe surface porosity

(sulphur or moisture).

Unsatisfactory Surface Appearance and Spatter

The following illustrated surface irregularities should be noted:
Figure
Figure
Figure
Figure

9.61
9.62
9.63
9.64

badly shaped surface ripples

badly shaped ripples and excessive spatter
an inadequately filled crater
(a) and (b) stray flash (accidental striking of arc on plate, adjacent to weld)

The operator is usually directly responsible for these defects as a result of incorrect technique or
improper machine settings. Sound welding finished in a poor manner should not be excused even
though the adequacy of the joint may be beyond doubt. The ability and integrity of the welder must be
questioned.
In some cases, faulty or wet electrodes and unsuitable base material (high sulphur, for example) may
cause similar defects and unsatisfactory weld appearance.
Bead requirements are defects in as much as they constitute an abrupt change of section. Spatter in
itself is not necessarily a defect, but is quite likely indicative of improper welding and the likelihood of
other associated faults.
Stray arc flashes either with the electrode or holder are more serious than might at first be expected.
They create a quenched and brittle condition in alloy steels and are inadvisable even on mild steel,
where high static or normal fatigue stresses may be encountered. The repair of such damage may be
difficult and costly, involving chipping and probably preheating in the case of alloys.

450

Unsatisfactory Surface Appearance, Spatter, Stray Arc Flash

Figure 9.61: Badly shaped ripples.

Figure 9.62: Excessive spatter.

Danger: cracks in low alloy steels and

stress raiser under fatigue loading.
Figure 9.63: Inadequately filled crater.

Figure 9.64(a): Electrode holder stray flash

cross section.

Figure 9.64(b): Electrode holder stray flash

accidental striking of arc on plate,
adjacent to weld.
451

9.9

Defective Properties (Weld Metal and Joint)

Specific mechanical and chemical properties are required of all welds made in any given weldment.
These requirements depend on the codes or specifications involved and departure from specified
requirements is considered a defect. These properties are generally determined with specially
prepared test plates but may be made on sample weldments taken from production. Where test plates
are used, the inspector should see that specified procedures are followed, otherwise the results
obtained will not necessarily indicate the actual properties of the weldments.
Mechanical properties that may be defective are tensile strength, yield strength, ductility, hardness and
impact. Chemical properties may be deficient because of incorrect weld metal composition or welding
procedure. Both may result in lack of corrosion resistance.
Not all these defects are due to improper welding conditions since many such difficulties are caused by
the base metal. Properties of the base metal that may not meet the requirements are chemical
composition, internal conditions (laminations and stringers), surface conditions (mill scale, grease,
paint, oil, etc.), mechanical properties and dimensions. All these factors should be kept in mind when
considering the causes of welding difficulties.

9.10

Summary of Weld Faults

A good inspector,
supervisor or welder can,
and will assist greatly in
preventing faults in
welding. As an aid to
competent inspection,
Table 9.2 lists a number
of defects that are likely
to occur under certain
welding conditions.

Table 9.2
Welding Conditions
Cold weather

Effects Likely to Occur

Cracks in weld or base metal

Thick or rigid assemblies

Cracks in weld or fusion line

Hardenable materials

Cracks in either weld or base metal

Base material known or suspected to be

high in sulphur content (free machining
steels for example)

Cracks and porosity in weld

Rusty, oily or greasy joints

Porosity, inclusions and lack of fusion

Limited access to joint

Undercut, particularly if welding in vertical

or overhead position.

Welding in corners, at ends of welds and

first passes in deep grooves

Arc blow, resulting in poor fusion, porosity

incomplete penetration, spatter and poor
appearance.

452

9.11

Welding Inspection

Practically all areas of welding fabrication involve some form of inspection. To begin with, the welders
or welding operators, before the welding operation, will cursorily examine the fit up. During welding,
experienced welders would sense if the welding operation was progressing normally. After depositing
each weld pass, the welder examines it before laying the next weld pass. After completion of a weld
joint, the welding inspector goes over the weldment again. Therefore, the first step of inspection is
visual inspection. Other inspection methods will follow depending on the type of joints and type of
welds. For example, complete joint penetration of groove weld requires ultrasonic or radiographic
examination to see the inside of the weld. Visual or surface examination alone is inadequate. The
following inspection methods will be explained:
1.
2.
3.
4.
5.

Visual inspection
Liquid penetrant inspection
Magnetic particle inspection
Radiographic inspection
Ultrasonic inspection

Inspection of assemblies fabricated by arc welding involves a great many factors that cannot be
covered in any code or specification. These factors include not only the fundamental principles of the
actual operation of welding and a knowledge of common weld faults, but also related subjects
associated with the process, such as basic properties of welds and parent metal, testing methods and
interpretation of drawings and specifications.
An inspector acts as a responsible representative of an organization, which may be either the
manufacturer, the purchaser or some outside agency. His/her decisions are governed by some form
of written lists of requirements that others have drawn up, but which he/she must be able to interpret
both as to limitations and intent.
Because of the variety of welded structures requiring inspection, no one class of inspector is expected
to be proficient in all types of inspection. A shop inspector employed by a fabricator makes routine
checks of materials, dimensions, workmanship, finish and other details to ensure that design
requirements are met. Such inspection is as much a part of production as the necessary welding and
machine operations.
Independent inspections are on a basis different from that of routine shop inspection. An inspector
cannot possibly check all details of each part of the product. Any such checks must be limited to spot
checks or random samples, unless the inspector has reason to believe that code or specification
requirements are not being met.
As is the case with all other inspection and testing activities, welding inspection should form a part of
the planned operations that are set in place to produce a completed fabrication. With the foregoing in
mind, it is logical to organize the inspection portions of the fabrication sequence into rational
procedures or checklists, which will serve not only to standardize inspection but which will also provide
a format for the documentation of inspection activities. The following examples illustrate the foregoing
points.
453

Assume that a pressure vessel is being fabricated and that the first operations consist of preparing,
rolling and welding a plate to form a part of the vessel shell. A probable checklist for the inspector
would be:

INSPECTION ACTIVITY

ACCEPTABLE

NOT ACCEPTABLE

1) Verify that raw material conforms with that

which is specified on the drawing.
2) Inspect joint preparations to ensure that they
conform to drawing requirements.
3) Inspect joint preparations to ensure they
are free from laminations, cracks and other
discontinuities which would cause welding
problems.
4) After rolling, check for dimensional
accuracy.
5) Check the weld joint fit-up for gap and
misalignment (hi-lo).
6) Sign-off acceptance of assembly for welding
operations to continue.
Before welding is initiated, the inspector
should verify the following:
a) Welder/operator qualifications.
b) Check the welder/operator familiarity with
the approved procedure.
c) Verify the consumables in accordance with
the requirements of the procedure and the
applicable standards.
..............................................................
Inspectors Signature

454

9.12

Methods of Testing

The methods commonly used in testing and inspecting welds for the defects previously listed are of
two types non-destructive and destructive. The terms in themselves are descriptive and it is obvious
that non-destructive testing would include visual, radiographic, ultrasonic, etc. The term destructive
might be interpreted erroneously as destructive of the whole weld fabrication by means of an overload
test. This, however, is not so, and the word is commonly used to mean some form of mechanical test
applied to a typical sample of a weld, or to a section cut from a weld.

9.12.1

Visual Welding Inspection

Visual welding inspection is of great importance because it constitutes the principal basis of
acceptance for many types of weldments.
It is one of the most extensively used methods of inspection because it is easy to apply, fast, relatively
inexpensive and, provided the inspection report format is properly organized, gives very important
information with regard to the welding operator, the weld, and the general conformity of the weldment
to specification requirements.
Visual welding inspection should begin prior to the actual fabrication operations. The inspector should
examine the drawings, specifications, welding procedures and consumables, condition of the welding
equipment, and weld operator qualifications. He should also ensure that the parent metal to be joined
conforms to the specification requirements, and that it is free from such defects as laminations, laps,
seams, scale or other harmful surface conditions.
When the components are assembled for welding, the inspector should note incorrect root openings,
improper edge preparation and alignment, and other features of joint preparation that may affect the
quality of the welded joint. The inspector should check the details of the work while welding is in
progress. These details may include maintaining pre-heat conditions, welding speed and deposition
rate, welding current, etc. In short, the inspector must ensure that the welding operator is working in
accordance with an approved welding procedure.
Inspection after welding and heat treatment (where required) completes the inspection cycle. At this
time the inspector checks the finished weldment for weld width, bead or weave appearance, surface
defects such as crater cracks, porosity, longitudinal and transverse cracks, non-fusion problems and
undercut. Reinforcements should be also checked to ensure adherence to specification requirements.
Dimensional checks of the welded component should be carried out at this stage to make sure that the
welding process has not distorted the finished assembly beyond drawing tolerances.
Visual inspection requires some simple tools, such as measuring tapes, callipers, try squares, plumb
bob and fillet weld gauges (see Figures 9.65 to 9.70).

455

Ensure hook is
not damaged

Figure 9.65: The steel tape.

Figure 9.66: Measuring the thickness

of a plate with callipers.

Tightening screw

Moving member
Fixed

Vernier scale

0.66

Zero gap

1
0123456789

Figure 9.67: Callipers showing the principle of the vernier scale.

This simple device increases the precision of measuring instruments.

456

Figure 9.68(a): Using the inside

edges of a try square to check
squareness and wall flatness on a
hollow structural section (HSS).

Figure 9.68(b): Application of try

square. Note the corner is clipped
to clear the fillet weld.

457

Symbol
S (t)

GTSM
G

(gouge to sound metal)

groove angle
bevel
angle
weld face
toe

depth of fusion

weld beads
(passes)

S = depth of
preparation

face reinforcement

layers
t = thickness
= weld size

root face
toe

root reinforcement

heat affected zone

G = root opening
back weld (done after
welding prepared side)

Figure 9.69: Terminology for groove welds.

Figure 9.70: Typical gauges for measuring fillet weld sizes.

458

Fillet weld sizes are measured with welding

gauges such as the ones illustrated in Figure
9.70, and the sketch on page 461 shows the
method of use. Weld profiles are more difficult to
measure, but simple convexity and concavity can
be determined from the throat measurement
(Figure 9.71). Acceptable and unacceptable fillet
weld profiles as required by CSA W59 are shown
in Figures 9.72 and 9.73. Unacceptable weld
profiles may be corrected by grinding or depositing
additional weld metal as shown in Figure 9.74.

Figure 9.71: This gauge can measure the weld

throat, convexity and concavity in a fillet weld.

Size

Size
45
Size

Size

Size

According to CSA W59, convexity,

C, of a weld or individual surface
bead shall not exceed 0.07 times the
actual width of the weld or individual
bead plus 1.5 mm.

Size

Size

Size
C

Figure 9.72: Fillet welds considered acceptable to CSA W59.

459

Size

Size

Insufficient throat

Size

Excessive convexity

Size

Overlap

Size

Insufficient leg

Inadequate penetration

Figure 9.73: Examples of fillet welds considered unacceptable to W59.

Gouge or grind
to sound metal
and re-weld
Figure 9.74: Repairing overlap.

460

Measuring fillet welds

For CONVEX or FLAT fillet welds use

gauge to measure leg length.

This gauge incorrectly

measures the longer
leg length

This gauge correctly

measures the shorter
leg length

For fillet welds with unequal leg

sizes (where an equal leg fillet
was specified) always measure
the shorter leg length

For CONCAVE fillet welds use

gauge to measure the throat.

For concave fillet welds the gauge

should touch both sides. For welds of
unequal leg size, a concave fillet gauge
may give false indication of size. In this
case, if equal leg size had been
specified, use the special throat gauge
shown in Figure 9.71.

461

9.12.2 Liquid Penetrant Inspection

Liquid penetrant inspection (LPI, also called Penetrant Testing, PT) is a versatile method capable of
locating cracks, porosity, laps and folds that are open to the surface. The method is based on the
ability of the penetrating liquid to be drawn into a discontinuity. When the object is wiped clean, the
liquid remains in the discontinuity but can be drawn out by adding a developer, such as a fine powder,
which acts as a blotter. The penetrant shows up against the developer indicating the discontinuity on
the surface.
There are six basic steps involved in performing LPI, which are illustrated in Figure 9.75.
1.

Prepare the surface of the part to

be inspected by cleaning and
degreasing.

2.

Apply the penetrant to the

surface.

3.

Allow a period of time for it to be

drawn into any discontinuities.

4.

Remove the excess penetrant in

a manner that ensures retention
of the penetrant in any
discontinuities.

5.

Apply a developer to draw the

penetrant liquid from the
discontinuities out to the surface
and thereby provide an enhanced
indication of the discontinuities.

6.

Examine and assess the

discontinuities visually under
appropriate viewing conditions.
Clean the part and, if necessary,
apply a corrosion preventative.

Figure 9.75: The principle of liquid

penetrant inspection.

It is a simple, inexpensive method, but a good understanding of how it operates and the correct
procedures are necessary to get the best results.

462

The penetrants used in inspection are commercially available liquids containing visible dyes that have
been carefully formulated to combine a large number of desirable properties. The foremost
requirement is, of course, the ability to penetrate very small openings, which depends on the surface
tension and wetting ability of the penetrant. The rate at which the liquid flows and penetrates openings
is influenced by the viscosity. High viscosity liquids penetrate slowly but low viscosity liquids may drain
away too rapidly, with a tendency to drain out of shallow defects.
Penetrant should essentially be nonvolatile, although a small amount of evaporation at the defect helps
to intensify the dye content and prevent excessive spreading of indications. Rapid evaporation of
volatile solvents could imbalance the formula of the penetrant, decrease the ability to spread and cause
the penetrant to dry up.
A further desirable property of a penetrant is that it not lead to corrosion of the part being tested. The
compatibility of the inspection materials with the metal under test should be checked, particularly when
dealing with special alloys (e.g., titanium and nickel alloys) that could be sensitive to specific elements
such as sulphur or chlorides (halogens).
There are several methods by which the penetrant principle is used in inspection, and the standards
such as ASTM E-165 group them in various ways. Penetrant inspection methods can be classified
according to the:
g
g
g

type of dye
method of excess penetrant removal
form of developer

For the penetrant to contrast with the developer to reveal the presence of a defect, the penetrant
contains a dye. Two types of dyes are used:
g
g

fluorescent
nonfluorescent or visible

Properties of an Ideal Penetrant

g
g
g
g
g
g
g
g
g
g
g
g

penetrates very fine openings

remains in coarse openings
resists evaporation
removed easily from the surface
has the mobility to re-appear from openings quickly
spreads in very thin films
resists colour fading
non-corrosive
non-flammable
stable under storage
non-toxic
inexpensive

463

9.12.3

Magnetic Particle Inspection

Another method of testing, Magnetic Particle Inspection (MPI), utilizes a magnetic field, which is
induced in steel or other magnetic ferrous alloys.
If this field is interrupted by a discontinuity, such as a crack in the material, the field will become
distorted at the point, and a north and south pole will form at each point of material separation.

Flux leakage

powder collects at
increased flux density

Figure 9.76: Flux leakage at a discontinuity.

(a) AC/DC Yoke

(b) Cracks detected by MPI

Figure 9.77: Magnetic lines of force follow a path around surface defect; leakage field formed
at surface of weld attracts and holds magnetic powder in sharp, well-defined build-up.
464

If fine magnetic particles are applied by spraying or dusting onto the test object, the north and south
poles of the crack faces will attract the particles, which will form a visible bridge across the gap. See
Figure 9.76 and 9.77.
The magnetic field can be induced into the part in a number of ways, depending upon the form and
finish of the part.
The most common way to induce the field in the testing of welds is by the prod method using direct
current or alternating current.
When direct current is used it is possible to detect surface defects as well as linear defects, which are
sub-surface and do not break to the surface of the weld.
Alternating current is normally used where defects break the surface of the weld or component under
test. The prods are positioned in such a manner that the magnetic field intersects with the defect. For
this reason the prods should be placed in two positions to detect linear defects parallel to the direction
of weld, as well as linear defects transverse to the direction of weld. See Figure 9.78.
The surface roughness of the weld
will cause some loss of
effectiveness and sensitivity, but
even in these circumstances it is
an excellent tool to disclose major
weld defects, which otherwise
would have been missed.
The equipment used in magnetic
particle testing of welds can be
extremely bulky and awkward to
handle, or it can be light and
portable. The heavy equipment is
used to generate high DC current,
which is used where a sub-surface
linear defect must be disclosed.

Leg of
AC Yoke

The lighter equipment, such as AC

Yokes or permanent magnets, is
Indications in this
used when defects penetrate to the
direction are detected
surface of the weld, but are still too
tightly structured to be detected by
Figure 9.78: AC yoke used to detect linear discontinuities.
normal visual inspection. Magnetic
particle testing can be used in all
positions including overhead, and
semi-skilled persons can be used to
perform the actual test. However, the test results must be interpreted by a skilled and qualified
magnetic particle operator. Figure 9.79 illustrates a portable magnetic particle inspection unit.
465

Figure 9.79: Portable magnetic particle inspection unit.

Because the test is a magnetic method it is not applicable to materials that are non-magnetic, such as
aluminum, brass, bronze and austenitic stainless steels.
The magnetic particles used as the detection medium are fine iron powders of various colours (to
provide a contrast against a given background).
Another type of detecting media is a solution of iron powder and a suspension medium. The iron
particles in this method are coated with a fluorescent material that is most brilliant when viewed under
a near ultra-violet light (black light). The fluorescent method is not recommended for use on material in
the as-welded condition because random fluorescent materials may be retained in the crevices of the
weld bead and this may confuse the inspector.

466

9.12.4 Radiography
Radiography is the most commonly used non-destructive method for the detection of sub-surface
volumetric discontinuities in welds.
The radiographic method can be applied to most welded joints, but is largely confined to butt and
corner joints.
As with all testing methods, radiography has certain limitations and it is incumbent upon the welding
inspector to have some knowledge of both the methods of radiography used in the inspection of welds:
X-radiography and gamma radiography. Both X-rays and gamma rays have extremely short wave
lengths and it is this characteristic that enables them to penetrate objects opaque to ordinary light. The
two types of radiation can affect sensitized photographic film.
X-rays are generated in an X-ray
tube by propelling (at high speed) a
stream of electrons against a target,
constructed of materials with high
atomic numbers and high melting
points such as tungsten. The
electron stream interacts with the
atomic structure of the target
material, temporarily dislodging
electrons. Energy is generated from
this dislodgment action, 99% of
which is heat and 1% X-rays. The
heat is dissipated by the copper
anode and cooling media in the tube
housing, and the X-rays are
projected (in a cone from the target
material) against the weld
undergoing test. See Figure 9.80.

Figure 9.80: Basic circuit of self-rectified X-ray apparatus.

The density of the electron stream is

controlled by the current (M/A) input,
and the wave length (and therefore
the penetrating ability of the X-rays)
is controlled by the high voltage input
(kV) across the tube.

The system, as can be seen from the foregoing, is electrical, and no ionizing radiation is generated or
retained in the system when it is switched off. In addition, the wave length, and therefore the
penetrating power, is adjustable.

467

The equipment is cumbersome and even the units designed for portability are relatively heavy and
awkward to manoeuvre. Figure 9.81 shows the control box and X-ray unit positioned to radiograph a
weld.
Gamma rays, as used in radiography, emanate from a radioisotope.

Control Panel

Figure 9.81

X-Ray Tube

Small amounts of material, such as Cobalt 59 or Iridium 191, are placed in a nuclear reactor and
subjected to neutron bombardment. During this time the atomic structure of the Co 59 or Ir 191
captures a neutron. The material is then called Cobalt 60 or Iridium 192 and is in an unstable
condition.
In the natural order of things, the structure constantly strives to return to a stable condition and in so
doing releases energy in the form of gamma rays.

468

The wave lengths of the gamma rays are

fixed by the type of isotope (Co 60 or Ir
192) that emits them. Co 60 has wave
lengths that are much shorter than Ir 192,
and therefore have more penetrating ability.
The sources of radiation are constantly
emitting ionizing radiation and cannot be
shut off. The sources, therefore, are
shielded in a protective casing
manufactured from extremely dense
material such as lead or tungsten. The
material absorbs the radiation and protects
the operator from exposure (See Figure
9.82).
When the source is to be used, it is
Figure 9.82: Cutaway view of a
remotely handled. A drive cable is
typical isotope camera.
connected to the source pigtail and
projected through a hose called a nose
tube to a position from which the radiation
passes through the weld and onto a film. The drive cable is long enough to allow the radiographer to
stay a safe distance away from the source while it is out of the shielded position.
Figure 9.83 shows typical equipment for gamma radiography. Figure 9.84 shows a typical set-up for
gamma radiography. The source is cranked out of the safe position and is deployed through the nose
tube to a position inside the tubular weldment. The source remains in position for a period of time
calculated to produce the best image on the film.

Figure 9.83: Gamma radiography equipment.

Figure 9.84: Gamma radiography set-up.

(Photo courtesy of Canspec).

(Photo courtesy of Babcock & Wilcox Canada).

469

As X- or gamma radiation is directed at the weld, a certain amount will be absorbed by the structure of
the metal and the remainder will pass through onto a film that has been placed into position. The
amount of radiation absorption depends upon the material type and thickness; each material (steel,
aluminum, copper, etc.) having a different coefficient of absorption.
When a weld has internal discontinuities such as slag or gas holes, more radiation will reach the film
under these areas than in an adjacent area, which has no discontinuities and is therefore absorbing
more radiation. In this manner, differential amounts of radiation reach the film and react on the
sensitized emulsion in varying degrees. These differences in radiation absorption through a defective
weld appear on the developed film as shadows and are interpreted by the shape and density as to
what they represent (slag, porosity, gas holes, etc.).
When a defect in a weld does not constitute a relatively substantial difference in the total cross section
of the weld, the difference of radiation absorption will be less and therefore the image on the film will
not be pronounced. With certain defects such as tight cracks, cold lap, lack of side wall fusion, etc.,
there is a strong likelihood that no discernable image will appear on the film. Where the beam of
radiation is not directed into the plane of defect, Figure 9.85, the defect can be completely missed.
When a discontinuity in a weld does not constitute a relatively substantial difference in the total cross
section of the weld, the difference of radiation absorption may not be detectable by the density
difference it creates. (The image on the film will not be pronounced.) Certain flaws such as tight
cracks, cold lap, incomplete fusion, etc., can present little difference in the amount of radiation
absorbed such that no discernable image will appear on the film. Where the beam of radiation is not
directed into the plane of defect (Figure 9.85), the defect can be completely missed. This is the main
limitation of radiography.

Source Position A
Source Position B

Source Position C

Tight Sidewall Non-Fusion

Angled Crack

Film
Source Position A - No Shadows on Film
Source Position B - Shadow of Sidewall - Non-Fusion on Film
Source Position C - Shadow of Angled Crack on Film

Figure 9.85: Radioisotope projected into position by remote control.

470

In general, radiographs made with X-rays are superior to those made with gamma rays mainly due to
the fact that the penetrating power of the radiation emitted from X-ray equipment can be adjusted. In
steel up to 50 mm (2), this fact is demonstrated by superior radiographic sensitivity and clarity of the
defect image. Above 50 mm (2), the wavelengths necessary to penetrate the steel are of the same
order of magnitude to those of gamma radiation so that the radiographic superiority is reduced
somewhat.
Construction codes such as ASME Boiler & Pressure Vessel Code, Section VIII, CSA Standard W59 &
AWS D1.1, etc. dictate that the way in which a radiograph is produced (procedure) will meet specific
requirements. One of these requirements which is important to a welding inspector is the use of a
device known as a penetrameter.
The penetrameter is used to assess the quality of the radiographic technique how good is this film.
The penetrameter is an object of known shape size and geometric features. Seeing the image of the
penetrameter gives the viewer a sense of how clearly the image of a discontinuity would appear.
There are different types of penetrameters. Some have wires of different diameters while others have
holes or slots of specific dimensions machined into them. The sizes of the holes are related to the
thickness of the penetrameter. A wire type penetrameter is shown in Figure 9.86(b).

(b) Wire Type

Penetrameter
(a) Hole Type Penetremeters
Figure 9.86
471

The penetrameter, also known as the image quality indicator (IQI), Figures 9.87 & 9.88, is placed on
the parent material adjacent to the area of the weld that is to be radiographed, usually on the side of
the material that is closest to the source of radiation. Figure 9.86(a) indicates the size and placement
of ASME Boiler & Presure Vessel Code penetrameters as shown in Figure 9.88. Figures 9.89(a) &
(b) indicate the placement of penetrameters in accordance with CSA W59 requirements.
In general, construction standards will dictate that a penetrameter thickness shall be 2% of the
thickness of the material to be radiographed. Although some schools of thought equate the size of the
smallest defect discernable in the weld with the smallest hole discernable in the penetrameter, this is
open to serious dispute.
A penetrameter is usefule because the image of a discontinuity is affected by the radiographic technique.
That is, the geometric set-up of the source in relation to a) the object being examined directly and b) the
film, affects the image created on the film. For example, Figure 9.87 shows the effect of a reduction of
source to film distance (D) to image distortion. Penetrameters allow inspectors to factor out the effects
of distortion, because the image must be clear enough to enable the interpreter to see the outline and
the qualifying holes, and hence factor the image distortion into his/her judgement.

Source Radiates from All Sides

D1
D2
Penetrameter

Penetrameter

Flaw
Film

Penumbra

Penumbra (Image Distortion)

Figure 9.87: Effect of source position.

According to the ASME Boiler and Pressure Vessel Codes, at least one penetrameter shall be used for
each exposure, to be placed at one end of the exposed length, parallel and adjacent to the weld seam.
Where the source is placed to radiograph a circumferential seam with one 360 exposure, three
penetrameters located at 120 intervals shall be used.
472

Weld Identification

W5

40

Penetrameter

Weld

Lead Location Deltas and Numbers

Figure 9.88

The thickness of the penetrameter shall not be more than 2% of the thickness of the plate. In each
penetrameter there shall be three holes of diameter equal to one, two and four times the penetrameter
thickness, but in no case less than 0.010 . Tables in the codes indicate which of the holes must
appear as images on the radiographs. Each penetrameter has numbers affixed to it identifying the
material and minimum thickness of the plate. Figures 9.86, 9.87 & 9.88 illustrate the ASME Code
penetrameter and placement.
In other codes and standards, penetrameter size and hole dimensions may be different, as in CSA
Standard W59 (Figure 9.89).

Figure 9.89 (a)

Radiograph Identification and Penetrameter Location on Approximately Equal Thickness Joints
(CSA W59)

473

Figure 9.89 (b)

Radiograph Identification and Penetrameter Location on Transition Joints
(CSA W59)

Standards of acceptability for welds subject to radiographic examination have been established by
various code committees and are detailed in codes and standards such as ASME Sections I, III and
VIII, API 650, CSA Standard W59, ANSI B31.1, etc.
Some fabricators use radiography as a quality assurance tool even when not required by the contract.
In this event, acceptance standards are set in an arbitrary manner by company management although
usually in reference to a known industry criteria.
When the radiographic report specifies that indications of discontinuities are unacceptable the defective
area of the weld is removed and the cavity re-welded. The repair is then radiographed in accordance
with the original procedure.
Radiography is the most successful and reliable method for non-destructive testing of welds. There
are limitations to the test method, some of which have been pointed out. Where the limits of the
method have been defined, it is probable that an alternative test, such as ultrasonic, will serve to
complement radiography.

474

9.12.5

Ultrasonic Testing

For many years railroad wheels and similar items were subjected to hammer tests. The pitch of the
sound emanating from the wheel indicated to the inspector whether the wheel was flawed or not. The
sound waves emitted during these tests were of frequencies up to 20 kHz, (20,000 cycles per second)
that is, they were audible to the human ear.
Ultrasonic examination utilizes sound frequencies between 20 kHz to approximately 10 MHz
(10,000,000 cycles per second). When examining welds it is common to use frequencies in the range
of 2.5 to 5 MHz.
Ultrasonic sound waves are generated by applying electric pulses to piezoelectric crystals such as
quartz or barium titanate. These crystals vibrate, and electrical energy is transferred into mechanical
energy. In effect, when the crystal is placed on a material (i.e., steel) the pulses turn the crystal into a
hammer.
The two main ultrasonic beam modes used in weld testing are;
g
g

longitudinal waves
shear waves

Longitudinal waves are propagated as pressure waves, that is, the particles of the material under test
oscillate in the direction travelled by the ultrasonic waves as shown in Figure 9.90.
As can be seen in Figure 9.91 when shear waves are propagated, the particle motion is transverse to
the wave direction.

Wave Direction
Particle Motion

Wave Direction

Particle Motion

Figure 9.90: Longitudinal wave propagation.

Figure 9.91: Shear wave propagation.

The mechanical energy is transmitted through the particles of the material undergoing test. The
velocity at which the ultrasonic beam moves through material is constant for that specific material and
for the wave mode (longitudinal wave or shear wave). For example, in carbon steel a longitudinal
wave beam moves at a velocity of 0.585 centimetres per microsecond and a shear wave beam at a
velocity of 0.323 centimetres per microsecond.

475

When a change occurs in the material (such as a void caused by a defect), the velocity of the beam
changes and what is known as acoustic mismatch occurs. When this happens, part of the sound beam
is reflected back to the crystal, transformed back into electrical energy and projected onto a screen.
Any flaw in the material will cause acoustic mismatch and reflect the ultrasonic beam. The complete
cycle of this action is shown in Figure 9.92.
This phenomena of reflection due to acoustic mismatch is the basis of all ultrasonic testing.

Indication of Steel to Air Interface (B)

Oscilloscope

Flaw Indication (A)

Transmission
Pulse
Electrical
Pulse Out

Electrical
Pulse Return

Mechanical Energy

(B) Steel to Air (Acoustic Mismatch)

Sound Wave Reflects to the Crystal

Crystal
Crystal Housing

(A) Flaw in the Steel (Acoustic Mismatch)

Sound reflects to Crystal

Figure 9.92

Weld testing using ultrasonic methods requires very precise procedures both for calibrating the test
equipment and locating and evaluating discontinuities. The capability and qualification of the ultrasonic
technician is critical to the accuracy of the test. As indicated previously, both longitudinal and shear
wave modes can be generated. In weld testing both modes are used. Figure 9.93 briefly illustrates
the different modes.

476

Longitudinal Beam

Angle Beam (Shear Wave)

Figure 9.93

The ultrasonic examination system can be used for testing welds in almost any thickness except that
under approximately 3 mm (1/8). Application of ultrasonic methods and interpretation of test results
requires special techniques for thin materials.
The system used for weld testing is extremely sensitive, and provided that the wavelength is short
enough, can detect extremely small discontinuities.
The system will detect all types of discontinuities. To provide a maximum indication on the
oscilloscope, the ultrasonic beam should strike the major face of the defect at 90. For this reason
evaluation of a weld defect is usually done using more than one angle (i.e., 45 and 60). Its main
difficulty is detecting isolated discontinuities such as a single pore or inclusion.
Before an ultrasonic operator can start to test a weld certain information is needed to generate the
correct technique. The following describes a few examples:

1. The Welding Method

Reason:
Knowledge of the welding method is important, particularly when the operator is evaluating
the type of defect. For example, slag inclusions would not occur in a weld deposited with the
gas metal arc process. Tungsten inclusions would not be found in a weld made with the
shielded metal arc process.

2. The Type of Material and Condition of Heat Treatment

Reason:
The operator must know the material velocity because it will affect the distance and angle
calculations. Heat treatment in some materials affects sound wave velocity slightly and
compensates for equipment in set up.

477

3. The Weld Joint Design

Reason:
The operator must know the angles of the weld preparation because one of the angle beams
selected must strike the bevel as closely as possible to 90, to detect incomplete sidewall
fusion. For example, if the weld bevel is 30 one of the probes will be 60 (60 + 30 = 90)
and if the bevel is angle 35, the probe selection would be 55 (See Figure 9.95).
Where the weld joint configuration is such that the ultrasonic beam cannot be manipulated so that it will
interact with the bevel or other areas of interest at 90o, an alternate method to the single transducer
technique may be used (See Figures 9.96 and 9.97). This method is known as the pitch and catch
technique. The ultrasonic beam is projected by the transmitting transducer against the reflecting
surface, and is reflected by that surface to a receiving transducer.
If an ultrasonic test operator does not have detailed knowledge of the test subject, he/she cannot
perform an adequate test.
Most of the ultrasonic test systems
used in shop or field conditions do not
provide for a permanent record of test
results. Permanent records can be
produced through the use of fully
automated, electronically controlled
systems. Figure 9.94 shows a fully
automatic system for checking plate
quality prior to fabrication and welding.
The equipment comprises a series of
probes mounted on a motorized fixture.
The X and Y axis of movement is
controlled by the technologist and a 3dimensional "map" of the object under
test is created and stored in a
computer.

Figure 9.94: Fully automatic ultrasonic testing machine.

(Photo Courtesy of Babcock & Wilcox Canada)

The interpretation is usually made by

the operator. From this it can be seen
that the inspector must be reliable and a person of integrity, as well as being highly trained, and that
his written reports must be accurate and reliable.
Once a technique has been detailed, the actual test is completed in a relatively short time.
The nature, size and orientation of a specific fault is not easy to plot. This evaluation process requires
considerable experience and judgment.

478

Radiography can be used to complement the ultrasonic tests by first using ultrasonics to rapidly locate
the fault, and then radiographing the area to both define the problem and have a permanent record.
Conversely, ultrasonics complements radiography due to its ability to define tight incomplete fusion and
cracks much more satisfactorily than radiography. It can also be used to locate the depth of a defect
disclosed by radiography. This information is extremely valuable in welds that are accessible from two
sides, to determine from which side a repair should be made.
Figures 9.95, 9.96 and 9.97 illustrate a number of techniques used to detect straight sided defects,
both perpendicular, and angled.

35

30

55

60

Figure 9.95: Single V groove weld,

lack of side wall fusion - probe selection predicated on bevel angle.

Figure 9.96: Double J groove weld,

incomplete fusion - side wall - pitch and catch method weld preparation almost perpendicular to sound path.

Figure 9.97: Double V groove weld in butt joint,

incomplete penetration - pitch and catch method reflecting surface near perpendicular to sound path.
479

Figure 9.98 illustrates an ultrasonic unit

with a single probe, Figure 9.99 shows a
technique using an angle probe to detect
lack of penetration in butt welds welded
from one side.

Figure 9.98: Portable ultrasonic instrument.

(Photo Courtesy of Babcock & Wilcox Canada)

A
Corner
Lack of
Penetration Reflector

B
Penetration

Corner
Reflector
Transmission
Pulse

Signal From
Corner Reflector

No Echo

Indications from
Cases B and D

Indications from
Cases A and C

C
Lack of
Penetration Corner Reflector

Figure 9.99: Incomplete penetration

indicated by reflected signal on
ultrasonic oxcilloscope, schematic
diagram.

Technique Used for

Thin Sections Below
3/8 Thick

D
Penetration

No Corner
Reflector

480

Chapter 10
Weld Cost Estimating

Table of Contents
10.1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .483

10.2

Consistent Application of Welding Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .483

10.3

Cross-Sectional Area of Weld (At) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .484

10.4

Excess Weld (X) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .484

10.5

Unit Weight of Weld (M) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .486

10.6

Weight of Weld Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .486

10.7

Weld Metal Deposition Rate (D) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .487

10.8

Shielding Gas (G) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .488

10.9

Flux for SAW Process (F) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .488

10.10

Process Deposition Factor (Dp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .488

10.11

Welder/Operator Work Efficiency Factor (Dw) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .489

10.11.1 Operating Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .489

10.12

Weld Cost Estimating Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .491

10.12.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .493

10.13

Computer Estimating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .499

481

482

10.1

Introduction

The costs of any industrial process can be accurately estimated. A process such as welding will
involve factors such as:
g
g
g

labour
welding consumables
overhead

A cost analysis is needed prior to bidding on a contract involving welding, and it may be done at any
time during the course of a project for verification of actual cost. The following methods may be used
to determine the welding costs of any welding process, but only the most common processes are used
as examples.
Weld cost estimating involves many aspects of the fabrication process. Formulas have been
developed to calculate welding costs that contain factors that can vary widely depending upon the
welding process, the manner in which the process is applied, and shop practices and efficiencies. The
process of evaluating welding costs begins with an assessment of How much weld is to be made?
and What type of welds are to be made? All other decisions flow from these first assessments. The
rate at which weld metal is deposited varies significantly from process to process. The manner in
which that process is applied (welding procedure) effects the potential deposition rate.
Mathematical formulas have long been in place that predict the time necessary to complete a weld.
These formulas are relatively simple and are based upon the ability of a welding process to produce
metal at a certain rate. The variation in the use of these formulas comes from two areas:
g
g

10.2

consistent application of welding methods

operating factor

Consistent Application of Welding Methods

The costs of welding are based upon the amount of weld metal that must be deposited. Knowing the
amount of weld metal to be made, we then select the process that best suits the application. Each
process in a given application has the potential to produce a certain number of kilogram/hour
(pounds/hour). The critical issue for the estimator is to know shop practices and welding procedures.
Properly documented and applied welding procedures (recipes) offer the estimator hard data on
potential deposition rate.

483

10.3

Cross-Sectional Area of Weld (At)

To determine the amount of weld metal that must be made we start with calculating the welds crosssectional area. The areas of welds are usually simple geometric shapes either alone or in
combinations. Their cross-sectional area is calculated from standard formulas. Tables have been
provided here to aid in area determination.

Figure 10.1: Joint Volume

The actual area of the weld will be the total of the theoretical joint volume plus the area of
reinforcement.
At = A

(100 + X)
100

A = Theoretical cross-section area of weld size

At = Total cross-section area of deposited weld (includes excess weld)
Table 10.8 lists the joint areas for many common joint designs. The column under single bevel, 45
bevel angle can be used for determining the volume of fillet welds.

10.4

Excess Weld (X)

The amount of weld metal deposited will exceed the theoretical amount due to oversize welds,
additional weld surface reinforcement, and fit-up tolerances. This amount is called overwelding and
must be considered in predicting realistic cost estimates. It is largely due to two reasons.

484

The first reason involves errors in weld size judgment by the welder (or intentional overwelding). Most
welding is done either manually or semi-automatically. An experienced welder will try to slightly overweld, knowing how difficult it is to add a small amount of metal to a slightly undersize weld. The
important issue is to control overwelding to within reasonable limits and to account for these costs in
the estimating process. In some operations it has been found to represent 30% of the total welding
cost. Obviously, this is an area where immediate savings can be made.
The second reason for overwelding is related to the joint itself. The fit-up of the joint influences the
amount of welding necessary to complete it. The accuracy of the fabrication processes tremendously
influences welding costs. Obviously joints that have wide gaps require more metal to fill. They also
are difficult to fit and cause large amounts of distortion, which may make the assembly unsuitable for
service.
In the case of complete penetration weld joints that are to be backgouged, the root pass technique has
a tremendous effect on the cost of producing the joint.

Figure 10.2: Effect of improper root bead welding.

In Figure 10.2, the root pass in Sample 1 penetrates only partly through the joint and will require a
large amount of gouging to reach sound metal. In contrast, Sample 2 requires only minimal clean up
to reach sound metal. Therefore the total cost of producing identical joints will vary significantly.

485

10.5

Unit Weight of Weld (M)

For mild steel welds the density equals:

M = 7850 kg/m3
M = 0.283 lb/in3
For copper welds the density equals:
M = 8925 kg/m3
M = 0.34 lb/in3
For stainless steel welds the density equals:
M = 7880 kg/m3
M = 0.286 lb/in3
For magnesium welds the density equals:
M = 1740 kg/m3
M = 0.063 lb/in3
For aluminum welds the density equals:
M = 2700 kg/m3
M = 0.098 lb/in3

10.6

Weight of Weld Metal

W

= At x L x M
=VxM

W = weight of deposited weld metal (total weight for length L)

At = total cross-section area of deposited weld (includes excess weld)
L = length of weld (or total length of similar welds under consideration)
V = volume of deposited weld metal (includes excess weld)
M = unit weight of weld metal (per unit volume)

486

10.7

Weld Metal Deposition Rate (D)

The weld metal deposition rate is defined as the weight of deposited weld metal per welding arc hour.
The deposition rate is very much dependent upon the welding procedure, i.e., welding process,
electrode size and welding current. If the welding process is not known when the estimate is made,
then certain assumptions must be made by the estimator. These assumptions become very important
and will have an influence on the weld cost estimate.
If it is known that a fabricators specific welding procedure data sheets will be used, they should be
used as a basis for the estimators information.
The following examples explain how widely the weld metal deposition rate can vary:
SMAW Process
E4910
E4912
E4914
E4918
E4924
E4924

x
x
x
x
x
x

3.2
3.2
4.0
4.0
3.2
4.0

mm
mm
mm
mm
mm
mm

@
@
@
@
@
@

110 A ...
110 A ...
180 A ...
180 A ...
140 A ...
210 A ...

...
...
...
...
...
...

...
...
...
...
...
...

1.27
1.00
1.82
1.82
1.41
2.09

kg/hr
kg/hr
kg/hr
kg/hr
kg/hr
kg/hr

(2.8
(2.2
(4.0
(4.0
(3.1
(4.6

lb/hr)
lb/hr)
lb/hr)
lb/hr)
lb/hr)
lb/hr)

@
@

275 A ...
400 A ...

... ... 5.00 kg/hr (11 lb/hr)

... ... 5.46 kg/hr (12 lb/hr)

@
@
@
@

175 A ...
175 A ...
225 A ...
270 A ...

...
...
...
...

FCAW Process
E491T-9-CH x 1.6 mm
E491T-9-CH x 2.4 mm
GMAW Process
ER49S-6
ER49S-6
ER49S-6
ER49S-6

x
x
x
x

0.9
1.2
1.2
1.2

mm
mm
mm
mm

...
...
...
...

2.05
1.68
2.36
3.73

kg/hr
kg/hr
kg/hr
kg/hr

(4.5
(3.7
(5.2
(8.2

lb/hr)
lb/hr)
lb/hr)
lb/hr)

SAW Process
Deposition Rate (lb/hr)

Wire Diameter
(mm)

Amperage

Current Type

4.0
4.0

400
600

DC+
DC+

7.9
14.7

9.2
16.7

4.8
4.8

500
500

AC
DC+

10.4
9.4

4.8
4.8

700
700

AC
DC+

16.7
15.0

4.8
4.8

400
400

AC
DC+

7.1
6.4

487

Electrode Stick-Out

10.8

Shielding Gas (G)

Shielding gases are used with the FCAW and GMAW welding processes, except when a FCAW
electrode is self shielding and does not require shielding gas coverage. Self shielding electrodes are
usually employed in field welding. Shielding gas costs are specific to the company and are usually a
reflection of the total amount of gas used by the company. These costs will have to be taken from
actual accounting information.
It is important to note that control of the amount of shielding gas consumed is in the hands of the
welder. Many welders do not realize that high shielding gas flow rates do not necessarily improve the
soundness of the weld. It is common to find wide variances in flow rate settings, even through the use
of pressure gauges in place of flow meters.
Pressure gauges do not control flow; rather they supply gas at a given pressure not flow rate.
Education of the welder and welding supervision is the key to controlling excessive gas consumption.
For Estimating Shielding Gas Consumption: cubic foot per hour (cfh), or liter per minute (l/min)
Steel: 30c fh 40 cfh or 15 R/min - 18 R/min
Aluminum: 40 cfh 55 cfh or 18 R/min 25 R/min

10.9

Flux for SAW Process (F)

SAW flux is consumed by a portion of it being melted in the welding arc, then fused to form slag over
the SAW weld. The unfused flux is usually vacuumed and returned to the flux hopper. Literature
suggests that the weight of flux consumed is approximately the same as the weight of SAW weld
deposited. The cost per pound of flux can be found in actual accounting information.

10.10

Process Deposition Factor (Dp)

The weight of electrode consumed is always greater than the weight of weld metal deposited. The
electrode loss is caused by weld spatter, formation of flux and electrode end loss (stub ends in
SMAW). Each electrode manufacturer publishes deposition efficiencies for their products so this
information is readily available.
Following are typical values for deposition rates for the stated welding processes:
Dp
Dp
Dp
Dp
Dp

(SMAW)
(FCAW)
(FCAW
(GMAW)
(SAW)

=
=
=
=
=

65%
85% (Not for metal core wire)
95% (Metal core wire)
95%
98% - 100%

488

10.11

Welder/Operator Work Efficiency Factor (Dw)

This is a factor for the average arc time per welder manhour while the welder works on welding and
welding related functions. In other words, this is the total number of minutes out of an hour during
which welding is actually taking place. This is an important factor, which is often difficult to establish. It
will have a significant influence on the weld cost estimate so one should make a careful decision on
selecting Dw. The Dw factor is weld-shop specific, influenced by shop supervision, shop layout, and
the welders ability to work efficiently.

10.11.1

Operator Factor

In any operation there are down times. Any time the arc is not operating, the joining process is not in
progress. This can be attributed to many reasons; changing electrodes, fitting parts, replacing
components in a jig, turning assemblies over, personal down time, equipment repair or maintenance
and so on. Each company operates somewhat differently. However, these factors can be brought
together and accounted for in an estimate. Evidence for this comes from the ability of experienced
supervisors to accurately predict the time to complete a job. The supervisor knows from first-hand
observation how long it will take his welders to complete the task. The estimator can often draw on
this type of data from historical information.
Welding procedures provide the starting point in establishing welding costs. Since the welding
procedure is a recipe for making a weld, it will include all the information necessary to determine the
rate at which metal can be made. Lets take an example of a typical welding procedure for making a
fillet weld with the FCAW process.
6 mm Horizontal Fillet Weld Using FCAW 1.6 mm diameter
Layer

Pass

Wire Feed Speed

Current

Voltage

Travel Speed

5.9 m/min (230 in/min)

300

26

430 mm/min (17 in/min)

From the information above we see that the welder, when following this procedure completes the 6 mm
fillet weld at 430 mm/min (17 in/min). Logically then, if the part contains 10 m (395 in) of 6 mm fillet
weld it should take about 23 minutes to complete.
10 m divided by travel speed of 0.43 m/min = 23.3 minutes
An experienced estimator will know not to bid on this basis. There are great differences between what
is theoretically possible and actual performance. These differences are related to real arc-on time
(operating factor) and whether the procedure is actually used by the welder.

489

We know that a welder following the procedure needs no less than 23 minutes of arc-on time to make
this much metal. If the welder sets the equipment to lower parameters the weld will take longer. If the
equipment is in poor condition, time will be lost to continuously adjusting the setup in an attempt to
improve the operation of the process. If the welder is assigned other tasks than just continuously
welding, operating factor is reduced again. There are many reasons why 100% operating factor is
never achieved, even when using robots.
The data taken from a welding procedure is used to represent 100% operating factor. The welding
procedure data represents how much weld metal can be made when using the procedure, if welding
was continuous.
The following are examples of typical Operating Factors for the most common welding processes.
They should be used as a baseline from which to start, and compared against the experiences of the
shop.
a)

SMAW Process
g

b)

c)

Dw could typically be 10% - 20%. This will provide 6 to 12 minutes arc time per hour.

FCAW Process
g

This is a continuous wire feed process so the efficiency factor will be higher than for SMAW.

If applied semi-automatically on long joints, the welder will stop every 2 to 3 minutes to
change position. When applied to short welds, the stops will be more frequent.

Assume the welder stops each minute for minute, and takes 5 minute breaks each hour.

Dw could typically be 20% - 40%. This will provide 12 to 24 minutes arc time per hour.

GMAW Process

This process is similar to the FCAW, with the exception that there is no slag covering the weld.
g
d)

Dw could typically be 20% - 40%. This will provide 12 to 24 minutes arc time per hour.

SAW Process

SAW process is usually employed on long weld runs, and often on a production basis. The longer
and/or thicker the joints, the less material handling is involved and higher operating factors can be
expected. For these reasons, the range of operating factor is very wide.
g

Dw Could typically be 30% - 70%.

Note: The above Dw factors are approximations, and the estimator should consult the fabricator for
realistic values before proceeding with the estimate. The estimator can correlate welds of similar type
and welding process, then choose a Dw factor that is suited to the specific shop operation.
490

10.12

Weld Cost Estimating Procedure

Summary of symbols used throughout this discussion and weld estimate formulae:
A
X
At
L
V
W
M
D
Dp

= Theoretical cross-section area of weld size

= Excess weld due to oversize and weld surface reinforcement (in %)
= Total cross-section area of deposited weld (includes excess weld)
= Length of weld (or total length of similar welds under consideration)
= Volume of deposited weld metal (includes excess weld)
= Weight of deposited weld metal (total weight for length L)
= Unit weight of weld metal (per unit volume)
= Weld metal deposition rate (weight per hour)
= Weld process deposition factor {(wt. of metal deposited) / (wt. of electrode used)}
(expressed in decimal format)
Dw = Welder/Operator work efficiency factor (arc time per hour expressed in decimal format)
E = Weight of electrode used for length L
Tw = Person hours unfactored to weld length L
Tt = Total person hours (factored) to weld length L (Tt will be greater than Tw)
G = Shielding gas consumed (cubic content)
F = Flux consumed (weight)
OH = Overhead cost on labour

1.

Assume joint geometry for each type of joint.

2.

Make a weld take-off of the various types of welds.

3.

For each type of weld, choose the excess weld factor (X%). (One may wish to group welds by
size).

4.

Determine the Weld Metal Deposition Rate (D) by knowing the welding process, electrode, and
some average current range (knowing welding procedures will help).

5.

Determine the Welder/Operator Work Efficiency Factor (Dw) (this factor can be weld-shop
specific).

6.

Determine Weld Process Deposition Factor (Dp).

7.

Calculate total cross-section area for the various welds:

At = A

(100 + X)
100

491

8.

Calculate volume of weld deposited:

V = At L

9.

Calculate weight of weld:

W = V M

10.

Calculate unfactored person-hours (for gas consumption if required)

Tw =

11.

Calculated factored person-hours:

Tt =

12.

W
D Dw

Calculate electrode weight:

E=

13.

W
D

W
Dp

Calculate gas consumption:

G = Tw x (cu volume/hour)*
*Approximately 35 ft3/hr or 1000 R3/hr
14.

Calculate weight of SAW flux.

F=W

492

10.12.1

Summary

At = A

Tw =

(100 + x)
;
100

V = At x L;
W=VxM

W
;
D

Tt =

W
;
D Dw

E=

W
Dp

G = Tw x (volume/hour);
F=W
Labour Cost:
Direct Labour

Tt x Rate

= _______________

Overhead

Tt x OH

= _______________

Total

= _______________

Material Costs:
SMAW Electrode

W (SMAW)

$/Unit Wt.

_____________

FCAW Electrode

W (FCAW)

$/Unit Wt.

_____________

GMAW Electrode

W (GMAW)

$/Unit Wt.

_____________

SAW Electrode

W (SAW)

$/Unit Wt.

_____________

Flux

W (Flux)

$/Unit Wt.

_____________

Gas

Volume

$/Unit Vol.

_____________

Total

493

____________

Table 10.1
Fillet Welds
Assume:
a)

1/8 and 3/16 welds

increased by 1/64 in
size

b)

Welds > 3/16

increase 1/32 in size

c)

Weld reinforcing is
convex by 1/16
(parabolic)

Fillet

Actual Fillet

Size
(in)

Area
(sq. in)

Size
(in)

Area
(sq. in)

Reinforcement
Area (1/16)
(sq. in)

%
Excess
Weld Area

1/8

1
1
2

.0078
.0313
.1250
.2813
.5000
1.125
2.0000

9/64
9/32
17/32
25/32
1 1/32
1 17/32
2 1/32

.0099
.0396
.1411
.3052
.5317
1.1724
2.0630

.0083
.0166
.0313
.0460
.0608
.0902
.1197

133
80
38
25
19
12
9

Table 10.2
45 Bevel Weld with Backing:
Assume:
a)

Weld reinforcement of
1/16 and 1/8
(parabolic)

b)

Weld reinforcement
overlaps the top edge
of the joint by 1/8 each
side.

c)

Root opening =

Thickness T
(in)

Area
(sq. in)

3/16

1
1
1
1
2

.0645
.0938
.2500
.4688
.7500
1.0938
1.5000
1.9688
2.5000

494

Reinforcement
(Area sq. in)

% Excess Weld

Size
(in)

Area
(sq. in)

1/16
Reinf.

1/8
Reinf.

Avg.

.0286
.0313
.0417
.0521
.0625
.0729
.0833
.0938
.1042

.0573
.0625
.0833
.1042
.1250
.1458
.1667
.1875
.2083

44
33
17
11
8
7
6
5
4

89
67
33
22
17
13
11
10
8

67
50
25
17
13
10
9
8
6

Table 10.3
45 V-Grooves:
Assume:
a)

Weld metal to root of

groove (zero root
opening)

b)

Reinforcing overlaps
top edge of groove by
1/8 each side

c)

Calculate for 1/16

and 1/8 reinforcing

45-V
Groove
Depth
(in)

Area
(sq. in)

1/8

1
1
1
1
2

.006
.026
.104
.233
.414
.647
.932
1.268
1.657

Reinforcement
(Area sq. in)

% Excess Weld

1/16
Reinf.

1/8
Reinf.

1/16
Reinf.

1/8
Reinf.

Avg.

.015
.019
.028
.037
.045
.054
.063
.071
.080

.029
.038
.056
.073
.090
.107
.124
.142
.159

250%
73%
27%
16%
11%
8%
7%
6%
5%

483%
146%
53%
31%
22%
17%
13%
11%
9%

367%
110%
40%
24%
17%
13%
10%
9%
7%

Table 10.4
60 V-Groove:
Assume:
a)

Weld metal to root of

groove (zero root
opening)

b)

Reinforcing overlaps
edges of groove by 1/8
each side

c)

Calculate for 1/16 and

1/8 reinforcing

60-V
Groove
Depth
(in)

Area
(sq. in)

1/8

1
1
1
1
2

.009
.036
.145
.325
.577
.902
1.299
1.768
2.310

495

Reinforcement
(Area sq. in)

% Excess Weld

1/16
Reinf.

1/8
Reinf.

1/16
Reinf.

1/8
Reinf.

Avg.

.017
.023
.035
.047
.059
.071
.083
.095
.108

.033
.045
.069
.093
.117
.141
.165
.189
.213

188%
64%
24%
14%
10%
8%
6%
5%
5%

367
125
48
29
20
16
13
11
9

278
94
36
22
15
10
10
8
7

Table 10.5
45 Bevel Groove:
Assume:
a)

Weld metal to root of

groove (zero root
opening)

b)

Reinforcing overlaps
edges of groove by
1/8 each side

c)

Calculate for 1/16 and

1/8 reinforcing

45
Bevel
Depth
(in)

Area
(sq. in)

1/8

1
1
1
1
2

.008
.031
.125
.281
.500
.781
1.125
1.531
2.000

Reinforcement
(Area sq. in)

% Excess Weld

1/16
Reinf.

1/8
Reinf.

1/16
Reinf.

1/8
Reinf.

Avg.

.016
.021
.032
.042
.053
.063
.074
.084
.095

.031
.042
.063
.083
.104
.125
.146
.167
.188

200
68
26
15
11
8
7
5
5

388
135
50
30
21
16
13
11
9

294
102
38
23
16
12
10
8
7

Table 10.6
60 Bevel Groove
Assume:
a)

Weld metal to root of

groove (zero root
opening)

b)

Reinforcing overlaps edge

of groove by 1/8 each
side

c)

Calculate for 1/16 and

1/8 reinforcing

60
Bevel
Depth
(in)

Area
(sq. in)

1/8

1
1
1
1
2

.014
.054
.217
.487
.866
1.353
1.949
2.652
3.464

496

Reinforcement
(Area sq. in)

% Excess Weld

1/16
Reinf.

1/8
Reinf.

1/16
Reinf.

1/8
Reinf.

Avg.

.020
.029
.047
.065
.083
.101
.120
.138
.156

.039
.057
.093
.129
.165
.201
.237
.273
.310

142
54
22
13
10
7
6
5
5

279
106
43
26
19
15
12
10
9

211
80
33
20
15
11
9
8
7

Table 10.7
Backgouged Groove - 45 V-Groove:
Assume:
a)
b)
c)
d)
e)

45 groove angle
3/16 root radius
use 50% of root circle
use 3/8 wedge in middle
two side triangles

Areas
(sq. in)
Gouged Depth
Total
(T)

1
1
1
1
2

Reinforcement
(Area sq. in)

% Extra Weld

R=3/16

3/8xD

[email protected]

Total

1/16
Reinf.

1/8
Reinf.

1/16
Reinf.

1/8
Reinf.

Avg.

.055
.055
.055
.055
.055
.055
.055
.055

.023
.117
.211
.305
.398
.492
.586
.680

.002
.040
.131
.273
.468
.714
1.011
1.361

.080
.212
.397
.633
.921
1.288
1.652
2.096

.028
.037
.045
.054
.063
.071
.080
.089

.056
.074
.091
.108
.125
.143
.160
.177

35
17
11
9
7
6
5
4

70
35
23
17
14
11
10
8

53
26
17
13
11
9
8
6

D=T-R
1/16
5/16
9/16
13/16
17/16
21/16
25/16
29/32

497

Table 10.8: Cross-section Areas for Joint Size S

Butt
Size
S

5/16 (.31)
3/8
7/16 (.44)

9/16 (.56)
5/8
11/16 (.69)

13/16 (.94)
7/8
15/16 (.94)
1
1 1/8
1
1 3/8
1
1 5/8
1
1 7/8
2
2
2
2

.036
.056
.081
.111
.145
.183
.226
.273
.325
.379
.442
.510
.588
.731
.902
1.092
1.299
1.525
1.768
2.030
2.310
2.923
3.609
4.367

.026
.040
.058
.080
.104
.130
.162
.197
.233
.272
.317
.366
.414
.524
.647
.783
.932
1.094
1.268
1.456
1.657
2.097
2.589
3.132

.017
.026
.038
.052
.067
.084
.105
.127
.151
.176
.205
.237
.268
.339
.418
.506
.603
.708
.820
.942
1.072
1.356
1.675
2.026

.063
.078
.094
.110
.125
.140
.156
.173
.188
.203
.219
.235
.250
.281
.313
.344
.375
.406
.438
.469
.500
.563
.625
.688

.031
.048
.070
.097
.125
.157
.195
.238
.281
.328
.383
.442
.500
.633
.781
.945
1.125
1.320
1.531
1.758
2.000
2.531
3.125
3.781

.056
.072
.091
.113
.137
.162
.189
.221
.255
.289
.365
.361
.546
.650
.762
.884
1.015
1.155
1.461
1.804
2.183

5.197

3.728

2.411

.750

4.500

498

.018
.028

Table 10.9: Weight Per Length of Electrode

Wire Diameter

Inches Per Pound of Filler Alloy

Decimal Fraction
Copper
Aluminum
Inches
Inches
(deox.)

Nickel

Carbon Stainless
Magnesium
Steel
Steel

Silicon
Bronze

0.020
0.025
0.030
0.035
0.040
0.045
0.062
0.078
0.093
0.125
0.156
0.187
0.250

9900
6820
4400
3240
2480
1960
1030
647
460
252
162
115
63

11,100
7,680
4,960
3,650
2,790
2,210
1,160
730
519
284
182
130
71

10,300
7,100
4,600
3,380
2,580
2,040
1,070
675
510
263
169
120
66

10.13

3/64
1/16
5/64
3/32
1/8
5/32
3/16

32,400
22,300
14,420
10,600
8,120
6,410
3,382
2,120
1,510
825
530
377
206

9800
6750
4360
3200
2450
1940
1020
640
455
249
160
114
62

10,950
7,550
4,880
3,590
2,750
2,170
1,140
718
510
279
179
127
70

50,500
34,700
22,400
16,500
12,600
9,990
5,270
3,300
2,350
1,280
825
587
320

Computer Estimating

There are several computer software programs on weld cost estimating. Knowing all the factors
involved, you will be in better control when using computers to do the estimating for you.
Weld_IT is an excellent program designed by welding experts of the Canadian Welding Bureau. It can
be obtained through the CWB office.

499

500

Welding for Design Engineers

Index
A
accessibility for welding, 8-6
AISI, 5-47
allowable stress design (ASD), 2-16, 8-7
alloy elements in steel, 5-19, 4-4
angle of bevel, 1-17
annealing, 5-41
anode drop zone, 1-10
API, 2-9
arc blow, 1-17
arc efficiency, 5-27
arc force, 1-17
arc plasma, 1-17
arc radiation, 1-16
arc voltage, 1-17
arc welding, 1-4
ASME, 2-9, 2-28
ASTM, 2-17, 2-22
autogenous weld, 1-17
AWS A5 Specification, 2-31
AWS D1.1, 2-3, 2-30, 2-32

B
back gouge, 1-18
backing ring, 1-18
backing strip, 1-18
backing weld, 1-18
bare electrode, 1-18
barium titanate, 1-18
base metal, 1-18
beam angle, 1-18
bevel angle, 1-18
BHN, 1-18
body-centered cubic (BCC), 5-4, 5-7, 5-9, 5-13
boron, 5-20
brittle fracture, 7-5, 7-6
built-up column, 8-35
butt joint, 1-18

C
carbon, 5-6, 5-19
carbon equivalent (C.E.), 7-15
case hardening, 5-41
cast iron, 5-5
cementite, spheroidized, 5-18
CGSB certification, 2-25
Charpy V-notch testing, 7-10
chromium, 5-19
CIDECT recommendations, 8-51
CJPG welds, 8-16, 8-18, 8-58
cleavage fracture, 7-7
cobolt 60 (Co60), 1-18, 9-57
cold crack, 1-19
cold work, 5-45

D
deposition rate, 1-19
deposition efficiency, 1-20
depth of fusion, 1-20
destructive testing, 1-20
developer, 1-20
direct current electrode negative (DCEN), 1-20
direct current electrode positive (DCEP), 1-20
distortion, 6-1, 9-8
bonding distortion, 6-21
angular distortion, 6-20
caused by flame cutting, 6-22, 6-23
caused by welding, 6-23 to 6-34
correction of distortion, 6-45 to 6-50
drag angle, 1-20
ductile fracture, 7-6
ductility, 7-4
duty cycle, 1-20
dwell time, 1-20

Index 1

effective throat, 1-20

electrical shock, 1-16
electrodes, classification of (SMAW), 4-9
electrode (wires) for gas metal arc welding, 4-37
electrodes (wires) for flux cored carc welding, 4-44
electrode extension, 1-21
electrode extension, effect of, 1-15, 1-20
electrogas welding, 1-5
electron beam welding, 1-5
electroslag welding, 1-5
essential variables, 1-21

gamma rays, 1-22

gap joint, 8-52 to 8-54
gas inclusions, 9-22
gas metal arc welding, 4-16 to 4-38
gas pipeline system, 2-27
gouge to sound metal (GTSM), 1-22
grain boundaries, 5-10
grain size effect on fracture, 7-8
groove angle, 1-22
groove radius, 1-22
groove weld, 3-5, 3-9 to 3-11
gusset plate connection, truss, 8-28, 8-30

F
face-centered cubic (FCC), 5-7, 5-9, 5-13
fatigue, 1-21
fatigue cracks, 7-28
fatigue failure, 1-21
fatigue fracture, 7-28
fatigue life of weldments, 7-45
fatigue strength, 1-21, 7-28 to 7-45
fatigue striation (fracture surface), 7-28
ferrous alloy, 1-21
filler metal, 1-21
fillet size, minimum, 8-26
fillet weld groups, 8-21 to 8-25
fillet weld strength, 8-20
fire hazards, 1-16
flat position welding, 1-21
flame hardening, 5-41
flame straightening, 6-46 to 6-50
flux, active, 1-17
fluxes for submerged arc welding, 4-56
flux cored arc welding, 4-39 to 4-50
fracture and fatigue, 7-1
fracture of ships, 7-11
fracture mechanics, 7-23 to 7-28
fumes, welding, 1-16

H
hardenability, 5-43
hardening curves, 5-15, 5-16
hardness, 1-22, 5-21
heat affected zone, 1-22, 5-22, 5-24, 5-25
heat input, 5-28
heat treatment, 1-22, 5-17, 5-40, 5-45
health and safety, welding, 1-16
horizontal position, welding, 1-22
hexagonal-closed packed (HCP), 5-7
hydrogen cracking, 5-35, 9-34
hydrogen in weld metal, 1-16

I
incomplete fusion, 1-23, 9-29
incomplete penetration, 9-32
inert gas, 1-23
inspection cycle, 1-23
inservice inspection, 2-30
inspection, welding, 9-1
ionizing radiation, 1-23
iradium 192 (Ir 192), 1-23, 9-57
iron, 5-6
iron-iron carbide phase diagrams, 5-8
ISO Standards, 2-33

Index 2

joint build-up sequence, 1-23

joint design, 1-23
joint (butt, corner, tee, lap, edge), 3-6
joint, definition, 3-5
joint edge preparation, 3-19 to 3-22
joint penetration, 1-23
joints, prequalified, 3-12
joints, types of basic, 3-5

National Building Code (NBC), 2-27

nickel, 5-19
noibium, 5-20
nitrogen, 5-20
nondestructive testing, 1-24, 9-41 to 9-68
normalizing, 5-41

O
oil pipeline system, 2-27
open circuit voltage, 1-24
overhead position welding, 1-24
overlap joint, 8-53
oxy-acetylene welding, 1-4

K
K, stress intensity factor, 7-26
KI, stress intensity factor, mode I, 7-26

L
laser welding, 1-5
layer, weld, 1-23
limit states design (LSD), 2-16, 8-12
liquid penetrant inspection, 9-50
load combinations, 8-13
load factors, 8-13
longitudinal wave, ultrasonic inspection, 1-24, 9-63

M
magnetic field, 1-11, 1-12
magnetic particle inspection, 9-52
manganese, 5-6, 5-19
manual welding, 1-24
material toughness, 7-12
martensite, 5-14
melting rate, 1-24
metal transfer, 4-22, to 4-27
Miners Rule, 7-41
moisture, porosity, 9-24
molybdenum, 5-20
moment connections, 8-32

P
panel zone, 8-34
partial penetration joint, 1-24, 8-16, 8-18
pearlite, 5-11
penetrameter, 1-24, 9-59
penetrant, liquid penetrant inspection, 1-25
pezoelectric crystal, 1-25
phase transformation, 5-6
pinch effect, 1-13
plain strain, 7-24, 7-25
plain stress, 7-24, 7-25
plasma arc welding, 1-5
plug weld, 8-17, 8-19
poisson effect, 7-24, 7-25
polarity, effect of, 1-14
porosity, weld, 1-25, 9-23
power boiler, 2-29
preheat, 5-28
prequalified joint details, 8-59, 8-60
pressure vessel (ASME), 2-28
prod method (MPI), 1-25, 9-52
procedure qualification record, 1-25

Index 3

Q, shape factor of crack, 7-26

qualification of welders and welding operators, 2-12
qualification, welding, 2-29
quenching, 5-41

SAE, 5-47
semi-automatic welding, 1-26
shear resistance, 8-15
shear wave, 1-26
shielded metal arc welding, 4-4
shielding gas, 1-26, 4-31
silicon, 5-6, 5-19
size of weld, 1-27
slag, 1-27
slag inclusion, 9-26
slot weld, 8-17, 8-19
S-N diagram (fatigue), 7-31
solidification cracking, 5-32, 9-33
steel, 5-5
steel, classification of, 5-47 to 5-54
steel, fine grain, 7-14
strain rate, effect of, 7-21
stress concentration, 7-16, 7-17
stress range, 7-29 to 7-37
stress relieving, 5-41
submerged arc welding, 4-51 to 4-63
multiple electrode, 4-55
wires and fluxes, 4-56
welding procedures, 4-62
surfacing, weld, 1-28
symbols, welding, 3-23 to 3-68

R
resistance welding, 1-4
radiography sensitivity, 1-25
radiographic technique, 1-25
radiography (RT), 1-25, 9-55
radioisotope, 1-25
residual stress, 6-1, 6-8
transverse residual stress, 6-9
longitudinal residual stress, 6-9
residual stress in plate, 6-12, 6-13
residual stress in built-up column, 6-14, 6-15
residual stress in rolled I-shape, 6-16
root, weld joint, 1-25
root edge, 1-25
root face, 1-25
root opening, 1-25

T
tempering, 5-41
thermit welding, 1-5
thermal expansion, coefficient of, 6-6
toughness, 1-28
transition curve, 1-28
transition temperature, 1-28, 7-8, 7-9
transition behaviour, 7-20
travel angle, 1-28
TTW (tip to work distance), 4-42
tungsten inclusion, 9-28

Index 4

U
ultimate tensile strength, 1-28
ultrasonic inspection, 9-63
undercut, weld, 1-28

V
Vickers (hardness), 5-14, 5-15
vanadium, 5-20
visual welding inspection, 9-41
volt-ampere curve, 1-28

W
weld bead, 1-28
weld, basic types of, 3-7
weld cooling rate, 5-25
weld cost estimating, 10-1
weldability of metals, 2-11, 5-24, 7-14
welding design, 8-1
weld heat, 5-27
welding of hollow structural sections, 8-47 to 8-51
welding inspection, 9-41
welding inspector, 1-29
welding metallurgy, 5-1
weld pool, 1-29
welding procedure, 1-29
welding procedure specification, 2-12
welding processes, 2-13, 2-17
weld profiles, incorrect, 9-11, to 9-21
welding qualification, 2-29
weld root, 1-29
welding symbols, 3-23 to 3-68
wetting, weld metal to base metal, 1-29
wire feed speed, 1-29

X
X-ray, radiography, 9-55

Y
yield point, 1-30
yield strength, 1-30
yoke (magnetic particle inspection), 1-30, 9-52

Index 5

Index 6

Additional Resources
Welding Health and Safety
CWB/Gooderham Centre - Module 1, Canadian Welding Bureau, 7250 West Credit Ave., Mississauga,
ON, Canada, L5N 5N1.
CAN/CSA - W117.2. Safety in Welding, Cutting, and Allied Processes, Canadian Standards
Association, 178 Rexdale Blvd., Rexdale, ON, Canada, M9W 1R3. Page 12 of this standard lists other
CSA standards relevant to safety in welding.
CAN/CSA-Z94.2. Hearing Protectors, Canadian Standards Association, 178 Rexdale Blvd., Rexdale,
ON, Canada, M9W 1R3.
ANSI/ASC Z49.1-94. Safety in Welding and Cutting, American Welding Society, 550 N.W. LeJeune
Rd., Miami, FL 33135, U.S.A.
ANSI/AWS F4.1. Recommended Safe Practices for the Preparation for Welding and Cutting of
Containers That Have Held Hazardous Substances, (AWS publishes a Safety and Health Information
Packet that includes these two standards).
Structure and Properties of Metals
CWB/Gooderham Centre - Modules 8, 20, Canadian Welding Bureau, 7250 West Credit Ave.,
Mississauga, ON, Canada, L5N 5N1
Physical Metallurgy Principles, Reed-Hill, R.E., D.Van Nostrand Company, Inc.
Welding Handbook, Eighth Edition, Vol 1, American Welding Society, 550 N.W. LeJeune Rd., Miami, FL
33126, U.S.A.
Metals Handbook, Tenth Edition, Vol. 6, American Society for Metals.
Welding Metallurgy
CWB/Gooderham Centre - Modules 8, 9, 12, 20-23, Canadian Welding Bureau, 7250 West Credit Ave.,
Mississauga, ON, Canada, L5N 5N1.
Linnert, G.E., Welding Metallurgy of Carbon and Alloy Steels, Volumes 1 and 2 (1965 and 1967),
American Welding Society, 550 N.W. LeJeune Rd., Miami, FL 33126, U.S.A.
Stout, R.D., Weldability of Steels, 1987. American Welding Society, 550 N.W. LeJeune Rd., Miami, FL
33126, U.S.A.

Lancaster, J.F., Metallurgy of Welding, 1980, George Allen and Unwin, London.
AWS D1.1, Structural Welding Code, American Welding Society, 550 N.W. LeJeune Rd., Miami, FL
33126, U.S.A.
Welding Handbook, Eighth Edition, Volume 4, American Welding Society, 550 N.W. LeJeune Rd.,
Miami, FL 33126, U.S.A.
Welding Design
CWB/Gooderham Centre - Modules 30-39, Canadian Welding Bureau, 7250 West Credit Ave.,
Mississauga, ON, Canada, L5N 5N1.
Canadian Standards Association, CAN/CSA S16-01, Steel Structures for Building (Limit States Design),
Canadian Standards Association, 178 Rexdale Blvd., Rexdale, ON, Canada
M9W 1R3
Handbook of Steel Construction, Latest Edition, Canadian Institute of Steel Construction, Toronto.
Haung, J.S., Chen, W.F., and Beedle, L.S., Behaviour and Design of Steel Beam-to-Column Moment
Connections, Welding Research Council Bulletin 188, October 1973.
Blodgett, O.W., Design of Welded Structures, The James F. Lincoln Arc Welding Foundation,
Cleveland, Ohio, 1966.
Salmon, C.G., and Johnson, J.E., Steel Structures - Design and Behaviour, Harper & Row, New York,
NY, 1980.
Kennedy, D.J.L., and Kriviak, Strength of Fillet Welds under Longitudinal and Transverse Shear - A
Paradox, Canadian Journal of Civil Engineering, Vol. 12, No. 1, Mar. 1985.
Kulak, G.L., Adams, P.F., and Gilmore, M.I., Limit States Design in Structural Steel, Canadian Institute
of Steel Construction, Toronto, Sept. 1985.
Manual of Steel Construction, Load and Resistance Factor Design, Vol. II Connections, Seventh
Edition, American Institute of Steel Construction.
Gaylord and Gaylord, Design of Steel Structures, Third Edition, McGraw-Hill Ryerson, 1991.
Enaelhardt, M.D., Design of Reduced Beam Section Moment Connections, North American Steel
Construction Conference Proceedings, AISC, 1999.
Seismic Design Provisions, Uniform Building Code (UBC), U.S.A.

ComitJ International pour le Developpement et lEtude de la Construction Tubulaire (CIDECT), The

Strength and Behaviour of Statically Loaded Welded Connections in Structural Hollow Sections, British
Steel Corporation, Tubes Division, Technical Centre, Corby, Northants, N17 IUA, Great Britain.
Packer, J.A., Design Examples for HSS Trusses, Canadian Journal of Civil Engineering, Volume 13,
Number 4, August 1986, pp 460.
Packer, J.A., and Henderson, J.E., Hollow Structural Section - Connections and Trusses, Canadian
Institute of Steel Construction, 201 Consumers Rd., Willowdale, ON Canada, M2J 4G8.
Fracture and Fatigue Application
CWB/Gooderham Centre - Modules 35-38, Canadian Welding Bureau, 7250 West Credit Ave.,
Mississauga, ON, Canada, L5N 5N1.
Principles of Structural Integrity Technology, Pellini, W., Item ADA-391 PTG., 1976, National Technical
Information Service, US Department of Commerce, Springfield, VA, 22161, U.S.A.
Fracture and Fatigue Control in Structures - Applications of Fracture Mechanics, Rolfe, S.T., and
Barsom, J.M., Prentice Hall, Englewood Cliffs, NJ, 1977.
State of Art in CTOD Testing and Analysis, Harrison, J.D., (3 parts) Metal Construction, Vol. 12, 1980;
(9) Sept. pp 415-422; (10) Oct. pp 524-529; (11) Nov. pp 600-605.
Brittle Fracture of Welded Plate, Hall, W.J., Kihara, H., Soete, W., and Wells, A.A., Prentice Hall,
Englewood Cliffs, NJ, 1967.
Guidelines for Fracture-Safe and Fatigue-Reliable Design of Welded Structures, Pellini, W., The
Welding Institute, Cambridge, England, 1983.
Fatigue and Fracture in Steel Bridges, J.W., Fisher, Wiley, NY 1984.
Fatigue of Welded Structures, T.R. Gurney, 2nd Edition (1979), Cambridge University Press.
A Fatigue Primer for Structural Engineers, J.W., Fisher, G.L. Kulak, Ian F.C. Smith, ATLSS Report,
No. 97-11.