Pipe Welding Procedures

PIPE WELDING
PROCEDURES
Second Edition
by Hoobasar Rampaul
Industrial Press
New York, New York
Library of Congress Cataloging-in-Publication Data

Rampaul, Hoobasar.
Pipe welding procedures I by Hoobasar Rampaul.2nd ed.
p. cm.
ISBN 0831131411
1. Pipe--Weldinge. I. Title.
TS280 .R27 2002
62 I.8'672-dc2 I
2002033938
Industrial Press Inc.

200 Madison Avenue
New York, NY 100164078
Second Edition
Pipe Welding Procedures
Copyright 2003. Printed in the United States of America.

All rights reserved.
This book or parts thereof may not be reproduced, stored in a retrieval
system, or transmitted in any form without the permission of the publishers.
2345678910
Contents
Forward
iv
Preface
1. Introduction
2. Essentials of Shielded Metal-Arc Welding Technology
10
3. Heat Input and Distribution
14
4. Reperation of the Pipe Joint
21
5. Uphill Welding the Root Bead on Heavy-WallPipe

(5GPosition)
31
6. Welding the Rood Bead by the Gas Tungsten Arc

Welding Process
57
7. The Intermediate and Cover Passes
81
8. Welding Thin-Wall Pipe
100
9. Horizontal Pipe Welding (20)
136
10. Welding Complicated Pipe Joints
158
11. Introduction to Welding Metallurgy
170
12. Distortion in Pipe Welding
198
13. Pipe Welding Defects
207
14. Fitting-UP Pipe
222
15. Qualification of the Welding Procedure & the Welder 237

I6.General Welding Safety
243
Index
247
iii
Foreward from First Edition

Today, most pipe is joined by welding, but this was not always
true. for until about 35 years ago pipe was joined by screw threads
or by bolted flanged joints. Welding, now considered the best
method for connecting pipe, is used even in the most demanding
high-pressure and high-temperature service.
Much of the credit for this progress must be given to the welding industry; however, credit must also be given to the many
welders who. through trial and conscientious effort, developed
the welding procedures which consistently produce reliable weld
joints. A great many welders contributed their efforts toward
this end and it is unfortunate that their names have not been recorded.
One person who has contributed greatly to the improvement of
general pipe welding procedures is Hoobasar Rampaul. He has
worked as a pipe welder in refinery construction and maintenance,
as a welding inspector, and as a teacher at the Hobart School of
Welding Technology where I have been closely associated with
him and have often witnessed his outstanding skill as a pipe
welder. His success is due to his fine technical knowledge of welding as well as to his outstanding manipulative skill. By blending
these he has been able to develop and qualify procedures that consistently result in pipe welds of the highest quality.
In order to make this skill available to many more welders I
have encouraged Mr. Rampaul to write this book. It is a major
contribution to the art and technology of pipe welding and I
highly recommend it to all those craftsmen who have an interest
in this field.
HOWARD
B.
CARY
Preface
Industries, such as power plants, oil refineries, chemical plants,
food processing plants, and those that operate cross-country pipelines, have created a great demand for welders who are capable
of producing high-quality pipe welds consistently.
This book is intended to help meet this need by assisting in the

training of pipe welders in trade schools and to help those welders
now working at the trade to enlarge their knowledge and improve
their skills.
The chapters that treat the various welding procedures describe
in detail the correct welding techniques that should be used as the
weld progresses around the pipe joint. Possible mistakes are
pointed out, and the reasons for using the recommended procedures are given. For these reasons, the welder who uses this book
on the job, or while training, will find it to be a very useful reference aid.
In addition to the manipulative procedures, a welder should
acquire as much technical and theoretical information related to
welding as possible. This knowledge is most helpful in learning
the craft of pipe welding and in progressing upward on the job.
Therefore, there are informative chapters on welding metallurgy,
on recognizing and correcting welding defects, on distortion in
pipe joints, on fitting-up pipe joints, and on welding complicated
pipe joints.
The author has gained much valuable experience as a pipe
welder in various industries, not only in the United States but in
the Caribbean; and he has been associated with the Hobart Brothers Company as an instructor in pipe welding at the Hobart School
of Welding Technology. He would like to thank Mr. Howard B.
Cary, Vice President, Hobart Brothers Company, for providing
the inspiration to write this book and for his encouragement in its
preparation.
The author would also like to take this opportunity to acknowledge, among many others, the help given in the preparation of
this book by James Hannahs, Wade'Troyer, Rudy Mohler, Denny
Deweese, Nan Kidder, Helen Wilt, Marilyn Tarcea, Judy Parrish,
and Lana Shelkon. He would also like to express appreciation to
the American Welding Society and the following firms for their
assistance in providing illustrations: The Hobart Brothers Company, the Tube Turns Division of the Chemetron Corporation,
and the H & M Pipe Beveling Company.
V
Finally, the author would like to gratefully acknowledge the

help and advice given by Karl Hans Moltrecht, Technical and
Vocational Editor. Industrial Press Inc., especially for his contribution in the preparation of Chapters 11 and 12.
Acknowledgment by the Publisher
Industrial Press wishes to express its sincere appreciation to Robert
O'eon for his invaluable assistance in the preparation of the second
edition of this book.
vi
CHAPTER
Introduction
Only high quality pipe welds are acceptable in modern industry;
for the failure of a pipe weld not only can disrupt the operation of a
plant, it can be the cause of a serious accident with the possible loss
of life and property. For this reason, a pipe welder must be a
thoroughly qualified person in his craft.
The objective of this book is to describe the techniques that will
result in a successful pipe weld, which must be sound throughout as
well as look good. The pipe welder will be provided with the related
information necessary for him to do his job correctly. To be a
successful pipe welder and achieve high quality pipe welds, such as
shown in Fig. 1-1, requires practice in welding pipe. It cannot be
Fig. I-I. Example of a high quality pipe weld.
learned by reading a book alone; however, if incorrect techniques

are repeated, practice alone will never lead to successful pipe welding. Those who will take the time and effort to read this book will
learn the correct techniques which, if practiced, will result in obtaining the skills required to be a successful pipe welder.
Before starting to learn pipe welding, a person should be proficient in welding in the four basic positions: 1. flat; 2. horizontal; 3.
Chapter 1
vertical; and 4. overhead. All of these positions are used to weld
pipe. Since the pipe has a round shape, there is usually a gradual
transition from one position to another.
The welding positions are defined by standard symbols which are
shown in Fig. 1-2. It is important for the welder to learn to identify
FLAT POSIT ION
HORIZONTAL posmON
ZG
liE RTlCAL posmON

3G
PLATES AND AX IS
OF PIPE VERTICAL
PLATES VERTICAL AND

AXIS OF PIPE VERTICAL
PLATES A'lO A)<IS

OF PIPE HORIZONTAL
TEST POSITION
HoR IZONTAL 2 G
11.1
>
HORIZOIHAL
FiXED
OVEIlHEAD POSITION
4G
PLAT ES HOIIllONTAL
TEST
~G
POSITION
66
80<
o
-e ) }AXIS OF PIPE
VERTlCAL
PIPE SHALL NOT BE

TURNED Of! ROLL EO
WHILE WELOING
Count'syo!th.. Nobart Broth"" Cu.
Fig. ] -2. Standard symbols designating the welding positions.
these positions by their symbols (lG. 2G, etc.). These symbols will
be used in this book to identify the various welding positions.
When making the weld, the welder is confronted with two primary tasks. First he must prepare to make the weld, and second he
must concentrate his entire attention on the welding operation.
In preparing to make the weld, the welder is concerned with the
following matters:
I.
2.
3.
4.
The type of metal to be welded

The selection of the correct welding electrode
The preparation and cleaning of the edge, or weld joint
The fit-up of the pipes to obtain the correct alignment.
Introduction
After all of the preparations have been made, the welder must
give his complete attention to making the weld. He must strike the
arc and manipulate the electrode correctly in order to deposit a
sound bead. He must watch the molten puddle of metal and, when
welding a root bead, he must watch the keyhole (see Chapter 5, Fig.
5-4) constantly. Ever alert to notice small changes that may affect
the quality of the weld, he must be prepared to make instantaneous
adjustments in his welding technique when required. In other words,
when the weld is in progress, the welder should see and think of
nothing outside the area of the weld.
Basic Pipe Welding Procedures
When the pipe is in the SG position, with its axis horizontal as in

Fig. 1-3, positions on the pipe can readily be identified by their
likeness to the numbers on the face of a clock. Thus, the top of the
pipe is in the 12 o'clock position and the bottom is the 6 o'clock
position.
Two different welding procedures are used when the pipe is in the
horizontal position: downhill and uphill pipe welding. The choice of
the method is not affected by the diameter of the pipe; it depends
primarily on the wall thickness and the alloy content of the pipe, as
explained in the following section.
II
(2
8
6
Fig. 1-3. The identification of the welding positions around the pipe joint by the
numbers on the face of a clock.
Downhill Pipe Welding. Regardless of the method used. the pipes

must first be tack welded together. For downhill welding (Fig. 1-4),
the weld is started in the 12 o'clock position and the bead is welded
Chapter 1
progressively downward around the pipe until the 6 o'clock position
is reached. Starting again at the 12 o'clock position, the bead is
welded around the other side of the pipe to close with the first bead
at the 6 o'clock position.
Downhill pipe welding is used primarily to weld thin-wall mild
steel pipe having a wall thickness of lis to %{1 inch. The relatively
thin wall of the pipe retains the heat longer than thick metal would.
This causes the metal in the area of the weld to cool slowly jf the
speed of welding and the heat input are the same. Slow cooling is
START
STOP
STOP
START
Fig. 1-4. (Left) General procedure for downhill pipe welding. Fig.
General procedure for uphill pipe welding.
1~5.
(Right)
desirable because a softer and more ductile grain structure is then

obtained in the metal in the area of the weld.
When welding mild steel pipe, the s~ower cooling rate of the
thin-walled pipe does make it possible to deposit the weld at a faster
rate without harmful effects to the welded joint. For this reason,
downhill welding is preferred when welding thin-wall mild steel pipe.
The ductility of the metal in the weld and in the surrounding area
can be further improved by depositing several beads around the
weld. Each succeeding bead heats the previous bead, which cools
relatively slowly.
The fabrication of cross-country transmission pipelines and other
low pressure storage vessels are examples where the downhill method
of arc welding (SMAW) is used. Since such fabrications will be of
materials less than 3/8 of an inch thick, the downhill technique allows
faster welding speeds with less tendency to bum through the root of
the joint. In contrast, thicker materials and many of the alloy steel
materials require the uphill welding method.
Downhill welding requires the use of fast-freezing, lightly coated
electrodes such as 6010, 6011, 6012, 7010, and 7014 that produce
minimum slag. As the electrode is moved down along the joint the
molten puddle and it's slag covering will tend to smother the arc causing porosity and slag inclusions in the weld. The use of the proper
Introduction
electrodes, proper rod angle, and adequate travel speed, keeping ahead
of the molten slag will insure sound welds. In some instances the use
of straight polarity (electrode negative) along with the downhill
method will eliminate burn though when poorly fitted joints are
encountered.
The use of the heavier coated electrodes such as 7024 and the iron
powder low hydrogen types are not suitable in downhill welding as the
problems of slag entrapment, porosity, and cold lapping become insurmountable. Such electrodes will also require higher operating currents
thus increasing the chance of bum though.
Early in the twentieth century the oxy-acetylene welding (OAW)
process was used to join the low and medium carbon steel pipe material used at the time. Later, in the 1930's, when electric arc welding
(SMAW) came into general use in pipe fabrication, the higher welding
temperatures generated by the arc caused cracking problems, particularly in the root or first stringer pass.
Since that time pipe manufacturers have refined and improved the
metallurgy of their products to take advantage of the newer welding
processes and electrodes that were becoming available.
In recent years as the demand for larger diameter pipe with thicker
Fast moving welding crew.
Chapter 1
walls for transporting grade oil over very long distances and natural
gas at higher pressures, the pipe industry has improved the mechani~
cal properties of the pipe by additions of manganese and silicon, along
with more rigidly controlled amounts of carbon. Also small amounts
of columbium and vanadium have been used, usually in the range of
0.200.26 percent. Such materials are known as "low alloy" piping
because attempting to increase strength by excessive alloying causes
additional problems in welding such as cracking and embrittlement of
the heat effect zone (HAZ).
Increasing the carbon, manganese, and silicon content to achieve
these higher level mechanical properties may seem reasonable.
However, it will not be in the interest of joining pipe edges by welding without risking faulty welds. Even small~quantity increases in car~
bon can have great effects, increasing both the tensile strength and
hardness. Likewise, manganese will increase both toughness and duc
tility, but will suffer from not having the necessary value to resist
cracking.
Today's pipelines, and all of the Lx60 and Lx65 grade pipe materials, are being made by alloying techniques, other than simply
increasing carbon and manganese levels, to the limits appropriate for
carbon steel. In most instances, columbium or vanadium are added to
a steel containing 0.20~to-O.26 percent carbon and 1.0-to-I.35 percent
manganese,and. a "hot rol!" practice within the critical range is
responsible for grain refinement and adequate mechanical properties.
Pipelines welders are on the go all day. They are provided with
welder helpers who are responsible for grinding and wire brushing
each layer of weld deposit. They are also responsible for adjusting the
welding current as instructed by the welders, keeping the welder supplied with electrodes, and handing the welders an electrode each time
the one in use is consumed. The welder's helpers playa very supportive role because of their knowledge and experience in grinding and
brushing welds complements the welder's effort in making perfect
welds.
The welder helper's responsibility goes even further. After welds
have been brushed and grinding is completed, the welder helper is
often the first to discover defects such as surface porosity, poor fusion,
poor tie-in, and inadequate filter metal on the weld groove before capping. The helper then informs the welder, who will provide instruction
about what additional steps are needed to correct such defects.
Because of the many variables on pipeline construction, especially
with welding procedures and standards, welder helpers should be
given training, or brought up to date, before pipeline construction
Introduction
commences.
The majority of pipelines constructed today use low alloy grade
materials, although it is very possible higher strength alloy pipe material will be introduced in the future. Meanwhile, because of their rote
with preheating, interpass temperature, screening welds from wind and
rain, and observing surface defects, welder helpers should be formally
trained. Note: Chapter 8 continues the discussion of pipeline welding.
Uphill Pipe Welding. After the pipe has been tack welded, the weld
is started at the lowest spot on the pipe or the 6 o'clock position
(Fig. 1-5), and the bead is deposited upward around the pipe until
the 12 o'clock position is reached. The second half of the pipe is
then welded by again starting at the 6 o'clock position and welding
upward around the other side to the 12 o'clock position, where the
joint is closed.
This method is preferred for welding heavy-wall pipe and pipe
made of alloy steel. The thicker pipe wall acts as a "heat sink" by
withdrawing the heat more rapidly from the weld area than does a
thin-waH pipe. The faster cooling rate causes the metal in the weld
area to become more brittle in mild steel pipe. In alloy steel pipe the
tendency toward brittleness is greatly increased.
Cross country pipeline laying.
Chapter 1
To overcome this tendency the cooling rate in the weld area must
be reduced. This can be accomplished by decreasing the welding
speed and by depositing a heavier bead. Both of these objectivesslower welding speed and a heavier bead - are achieved by welding
the pipe in the uphill direction.
Horizontal Pipe Welding. When the pipe is in the 2G position, with

its axis vertical, the weld joint connecting the two pipes is in a
horizontal plane, and a horizontal position (2G) weld must be made
around the pipe. Horizontal pipe welding will be treated in detail in
a later chapter.
There are, of course, cases where the weld must be made in still
other positions, such as the 6G, or inclined, position. Welding in
these positions is usually done by using one of the methods just
described. Sometimes a combination of procedures is required and
the welder must exercise good judgment in selecting the best one.
PIPE WELDING STANDARDS ANO COOES
Welded joints in pipes playa vital part in industry. They are used
in oil refineries (see Fig. 1-6), chemical plants, power generating
stations, and food processing plants. Welded joints in pipes also play
a vital role in the transportation of liquids and gases, as exemplified
Introduction
by cross~country pipelines. In the construction industry pipe welded

members are used to support heavy loads.
. Since public health and safety are involved in almost all of these
cases, codes and standards have been established to protect the
public interest. The following organizations are primarily responsible for these codes:
American National Standards Institute (ANSI)
The American Welding Society, Inc. (AWS)
American Society of Mechanical Engineers (ASME)
American Petroleum Institute (API)
It is not possible to list here all of the standards and codes that
apply to the design and construction of pipe welded joints. The
welder should, however, be aware of those affecting the job on
which he is working.
CHAPTER
Essentials of Shielded Metal-Arc

Welding Technology
The objective of this chapter is to provide a review of shielded
metal-arc welding technology. By learning the principles treated
here, the welder will have a better understanding of how and why to
make adjustments to the machine and in his welding technique.
Basic Concept of Electric Arc Welding. In shielded metal-arc welding, an electrical circuit is established between the workpiece and the
welding machine. The current in this circuit may be either alternating current (Ae) or direct current (IX), although OC welding is
preferred for welding pipe.
When using OC current, the polarity may be straight or reversed.
Straight polarity means that the electrode is negative and the work is
positive. The current in this case flows from the electrode to the
work. For reversed polarity the above conditions are reversed~ the
work is negative, the electrode is positive, and the current flows from
the work to the electrode.
Current flow is measured in amperes, and the term "amperage" is
sometimes used to refer to current flow. Voltage refers to electrical
pressure, which is measured in volts and which can be likened to
hydrauliC pressure. Voltage is the force that causes the current to
flow. It is important to realize that a force, as well as a voltage,
cannot exist unless there is a resistance to overcome. Feel the force
built up in your arm when you press against a wall and then try to do
this by "pressing" into the open space around you.
Open circuit voltage is a term encountered in electric arc welding.
It is the voltage existing when there is no contact between the
electrode and the work and when there is no arc. No current can
flow in this case because the resistance is too great for the voltage or
electrical pressure to overcome. However, the electric generator is
trying to overcome this resistance by generating the maximum
voltage of which it is capable.
When the electrode touches the workpiece. the resistance to the
current flow is lowered and the current will flow in the circuit.
Because there is less resistance, the electrical pressure required to
"push" the current is less; Le., the voltage is much lower than the
10
Essentials of Shielded Metal-Arc Welding Technology

open circuit voltage. In this case so much current will flow that if the
electrode is stuck to the workpiece, it will overheat.
However, if the electrode is backed away slightly from the workpiece to form the are, less current will flow because there is increased
resistance to the flow of current caused by the gap. Since there is
more resistance to the flow of current, more voltage will be generated to overcome this resistance.
When the arc is long and the gap is larger, more voltage is
generated and less current will flow than when the arc is short, say
normal, for welding. The increased current flow encountered when a
normal arc length exists causes a greater temperature rise iu the
workpiece and deeper penetration, as compared to a long arc length.
When current flows, it encounters a resistance to this flow in the
workpiece which causes its temperature to rise. An even greater
resistance is encountered in "jumping" the gap, which creates much
heat within the arc. The effect is to cause the metal in the workpiece
and at the end of the electrode to melt.
The metal at the end of the electrode forms a droplet or globule
which is transferred across the arc to the workpiece. If a molten pool
of liquid metal exists on the workpiece, this globule or filler metal
will mix with the metal in the molten pool, or base metal, forming an
alloy. consisting of filler metal and base metal.
When the arc moves on, the puddle of molten metal is maintained
by melting additional base metal ahead of the arc and adding filler
metal from the electrode. However, some of the metal behind the arc
solidifies to form the bead.
Metal Transfer. Of particular importance is that the pipe welder

should understand how the metal is transferred from the electrode to
the workpiece.
The droplet or globule is transferred from the electrode to the
base metal by the propellant force of the arc and by an attractive
force exerted on the globule by the base metal. In passing from the
electrode to the puddle, the force of gravity also acts on the globule.
When welding in the flat position the transfer of the globule is
assisted by gravity. However. in the overhead position it opposes this
transfer.
With the normal arc length used in overhead welding, the com
bined effect of the propellant force and the attractive force is large
enough to overcome the pull of gravity and the filler metal will be
deposited on the workpiece. When a long arc is maintained in
overhead welding, the globule has a longer path to travel and more
time is available for the pull of gravity to act, causing the globule to
slow down. In this case, the globule will sometimes not reach the
11
Chapter 2
base metal and if it does, it will be moving at a slower speed. If. as is
usually the case when the arc is long, the workpiece is not hot
enough. the slow moving globule will not attach itself to the work
piece and win drop to the ground. Therefore, when a long arc is used
in overhead welding, no filler metal is transferred to the workpiece.
This is important when striking an arc to start a bead in the
overhead position.
Electrode Coating, Almost aU modern electrodes are coated. The

function of the coating is to form a gaseous shield over the molten
metal to protect it from the effect of the atmosphere (see Fig. 2~ 1).
,...---ELECTROOE CORD WIRE
COATING ----;~
GASEOUS SHIELD
MOLTEN WELD METAL

Fig. 21. Shielded metal-arc welding.
Certain ingredients in the coating enter the molten puddle to deoxidize the metal. The compounds thus formed are lighter than the
molten metal and rise to the top of the puddle to form a slag coating
when the metal solidifies. This coating, while it is still hot, protects
the solidified metal from the atmosphere.
In some electrodes there are additional ingredients. certain Oi
which are included to stabilize the arc. Others may be added in the
form of powdered metals to provide alloying elements or additiona
iron for the puddle.
In welding, care must be exercised to manipulate the puddle ot
liquid metal so as not to trap the slag within the metal as it solidifies.
It must be allowed to rise to the surface of the weld bead. Also. for
the same reason, the slag coating or crust must be completely
removed before a second bead is welded. over the first bead.
Entrapped slag can have serious harmful effects on the quality of the
weld.
Arc Length. Arc length is the gap or distance between the electrode
tip and the surface of the puddle. The correct arc length is primarily
12
Essentials of Shielded Metal-Arc Welding Technology

dependent upon the type of electrode used and upon the environmental conditions in which the welding is done.
When a heavily coated and highly alloyed electrode is used, a
short arc length must be maintained because the higher alloyed filler
and base metal are very sensitive to porosity. Holding the short arc
protects the puddle from the atmosphere, which is the cause of
porosity.
Since the cooling rate is faster with lightly coated electrodes, the
arc length should not be choked. By maintaining a longer arc (approximately %2 to V8 inch) the voltage will increase slightly, causing the puddle to spread out or enlarge. This, in turn, will allow the
fiUer metal to flow across the entire puddle, as shown in FilL 2-2A.
PROPER ARC LENGTH
SHORT ARC
Fig. 2-2. The effect of lhe arc length. A. Correcl arc length; B. Short arc length.
If the arc is too short (approximately lf16 inch), the size of the puddle is reduced considerably and the filler metal has a limited area in
which to be deposited. This causes the filler metal to rise, which can
result in incomplete fusion along the edge of the bead, as shown in
Fig.2-2B.
The arc length also depends upon the environmental conditions,
whether in a closed shop or in the open atmosphere, as in the case of
an oil line or a refinery. Generally. a shorter arc is required when
welding outdoors; and for heavily coated and highly alloyed electrodes a short arc is used in all situations.
13
CHAPTER
Heat Input and Distribution

The intense heat of the arc increases the temperature of the metal
beneath the are, which therefore reaches the melting point very
quickly. As soon as a temperature difference exists, the heat energy
starts to flow out of the weld area and heats the surrounding metal.
In welding, the welder must use skill to control the input and
distribution of heat.
Heat Input
The heat input is controlled primarily by the current setting. To

some extent the heat input can also be controlled by the arc length.
When welding, the heat input into the puddle is often controlled by
the welding speed and by the weave or whip, as discussed further on
in this book. However, more correctly, this is a matter of heat
distribution.
Unfortunately, it is not possible to provide exact recommendations for the current settings to be used in each case for the following
reasons:
1. The adjusting devices on all welding machines are not the
same. One machine set at 90 amps may liberate the same amount of
heat as another machine set at 130 amps.
2. The age and condition of the machine will affect the heat
obtained at a given current setting.
3. Electrodes made to the same classification but manufactured
by different companies may require different current settings to
liberate the same amount of heat.
4. The mass or thickness of the pipe must be considered because
a larger mass will absorb more heat away from the welded area.
5. The length of the cable in the circuit will affect the heat in the
arc. For a given machine setting, increasing the length of the cable
will result in less heat at the point of welding.
6. Environmental conditions affect the current setting to be used.
For example, a current setting made outdoors on a windy and very
cold day may be different from the setting made indoors in a shop.
14

Since there are so many reasons why current settings vary, it is
common practice for welders to evaluate current settings by depositing a weld or a bead on a piece of scrap metaL Conducting this
evaluation beforehand is a very important step in making a perfect
weld.
For pipe welding, this evaluation should be conducted on plates in
the vertical (30) position. Weld the test plate uphill or downhill,
depending upon the direction to be used to weld the pipe. The plate
should be free from dirt and rust. Rust, when in contact with the
molten pool of metal, will cause undercuts at the edges of the weld.
In conducting the evaluation, the welder should tryout different arc
lengths. He must be cautioned to remember that after a few deposits
are made on a small plate, the plate will become very hot and will, in
many instances, give misleading results. Before this c::curs, a different plate should be used to make the evaluation.
As a result of the evaluation, the welder should be able to
determine the current setting that will give the best results for the
conditions at hand.
The effect of the arc length, welding current, and the welding
speed can best be understood by studying Fig. 3-1.
Heal Distribution
The distribution of the heat and the rate at which it is withdrawn

from the weld zone are dependent on the following factors:
1.
2.
3.
4.
Conductivity of the work material

The mass of the metal surrounding the weld zone
The paths available for heat conduction
The use of the weave, or whipping, technique.
When compared to most other materials, ali of the metals are

good conductors of heat. However, all metals arc not equal in their
ability to conduct heat. For example, aluminum is a better conductor of heat than stainless steel, as shown in Fig. 3-2. The metal in the
area of the weld will cool more rapidly in the case of aluminum than
in stainless steel. Moreover, the heat will disperse throughout an
aluminum plate more rapidly.
A large mass of metal adjacent to the weld zone will tend to
withdraw heat from the weld zone more rapidly than a small mass.
For example, heavy plates and thick-wall pipes will tend to cool the
weld more rapidly than thin plates or thin-wall pipes. As another
example, the corner fillet in Fig. 3-3 can withdraw heat more rapidly
15
Chapter 3
EVEN
WELD QUALITY INSPECTION
PENETRATION,
UNIFOIlM
CROSS
SECTION
A GOOD lEAD
SMOOTH, WELl FOIlMEO BEAD
COMMON WELDINO MISTAKES

waDING
CUIlRENT
TOO lOW
TOO MUCH
PILING UP OF
WELD METAl
O\llllAPPING
WElDING CURReNT
TOO HIGH
UNDEllCUTTING
ALONG EDGES
~JOINT
TOO MUCH SPATTBl

lEAD, HAS POOR TO BE CLEANED OfF
BEAD
PENETRATION
WIDE, FlAT
COlD AT EDGES
ARC TOO LONG

lvoltage too high,
BEAD VERY
IRREGUlAR
POOR PENETRATION
WELD METAL NOT
PIIOPERlY SHIELDED
WEUNG SPEED
TOO FAST
WELDING SPEED
TOOSlO.
BEAD TOO SMAlL,

SHAPE IllllEGOLAll
NOT ENOUGH
WElD METAl IN
CIlIOSS SECTlCW
TOO MUCH
PILING UP Of
WElD METAl
TOO MUCH
PENETRATION
POOIl PENETRATION
WASTES TIME
.. - - -
AND
ElECTllOOES
Fig. 3-1. Weld quality inspection.
than the ordinary fillet. The bevel butt joint has the least ability to
withdraw the heat.
Another factor to consider in estimating the rate at which heat
will be withdrawn is the number of paths available along which the
heat can flow. In Fig. 3-4 the plates have the same thickness, but the
heat will be withdrawn more rapidly from the lap joint than from the
edge joint because the lap joint provides two paths or directions of
heat flow as compared to one direction for the edge joint. The paths
along which heat can flow for a few typical weld joints are shown in
Fig. 3-5.
16
STAINLESS STEEL
AA
STAINLESS STEEL
ALUMINUM
88
Fig. 3-2. The temperature distribution farJm aluminum plate and a steel plate at
two sections (AA and BB) of a plate while welding. At AA the melting point of
stainless steel is shown by the higher temperature of the molten metal in the
puddle. At. BB the temperature of the entire welded plate has increased in both
cases but the temperature of the aluminum plate is more equal throughout.
17
Chapter 3
CORNER FILLET
ORDINARY FILLET
BEVEL BUTT JOINT
Fig. 3-3. The corner fiUet will cause the heat to disperse from the weld more
rapidly than the ordinary fillet. The ordinary fillet will disperse heat more
rapidly than the butt joint.
11
LAP WELD
EDGE WELD
Fig. 3-4. The edge weld has fewer paths than the lap weld, along which the heat
can flow from the welded joint.
18
SOCKET OR SLEEVE JOINT
FACE
a BACK WELDED FLANGE
BORE 8 BACK WELDED FLANGE
WELDED NECK FLANGE
Fig, 3-5. Paths of heat flow found in some typical pipe weld joints,
19
Chapter 3
The distribution of heat in the weld zone is affected by the
welding technique. While welding, the electrode and the arc are
sometimes deliberately moved in and out of the puddle of molten
metal, in a uniform pattern, to reduce the temperature of the molten
metal and to preheat the metal ahead of the weld. This technique is
called "whipping:' When the puddle of molten metal is moved back
and forth in a uniform manner across the weld joint, the heat in the
area close to the weld is spread out; this technique is called "weaving." It is used to spread the heat and to obtain a wider weld bead
without maintaining an excessively large puddle of molten metal.
Whipping and weaving will be treated in greater detail in later
chapters..
In summary, the welder must analyze the weld joint and estimate
the directions in which the heat will disperse. He can then adjust the
current setting on the machine to provide an adequate amount of
heat to maintain the molten puddle of metal and to obtain the
desired welding speed. If the weld joint causes the heat to be
withdrawn rapidly, the current setting is increased and possibly, the
welding speed will have to be slowed down. On the other hand, if
heat will not disperse readily from the weld joint, a lower current
setting is used and it may be necessary to use the whipping technique.
20
CHAPTER
Preparation of the Pipe Joint

The preparation of the pipe joint is an essential part of pipe
welding, as the quality of the weld is affected by the care used in
preparing the joint. Indeed, in many instances the failure of the pipe
joint can be attributed to faulty joint preparation. The pipe welder
must understand and then practice those skills required to prepare
the joint properly for welding. This is the first step in making a
successful pipe weld.
The preparation of the joint consists of four separate steps:
I.
2.
3.
4.
Prepare the edges

Clean the joint surfaces
Fit-up the pipes
Tack weld the pipes together.
Each step will be treated separately in this chapter. However, first

there will be a brief discussion on how to practice pipe welding.
Short lengths of pipe, or pipe nipples, should be used to practice
weld to economize on the cost of the pipe. However, if the pipes are
too short, the cooling rate of the weld may be affected to the extent
that the weld will not correctly simulate conditions for welding
longer pipe lengths. For this reason, a length of 7 inches is recommended for each of the two pipes to be welded. Experience has
shown that this length does not affect the cooling rate of the weld
significantly.
As mentioned in Chapter 1, two different basic procedures are
used to weld pipe. For thin-wall pipe, the downhill method is used;
and for thick-wall pipe, the uphill method is used. Except for details
involving the size of the joint and welding the tack weld, the
procedures for preparing the joint are the same, whether the pipe is
thin-wall or thickwall.
Preparing the Pipe Edges. Single-vee butt joints are usually used to
weld sections of pipe together. The elements or parts of the joint
have been given standard names, which are shown in Fig. 4-1, These
names are used so frequently in pipe welding that they should be
committed to memory.
21
Chapter 4
NCLUDED ANGLE
IEvn
ANGLE
ROO' FA~j
\CJ
..... ROOT OPENING
WilD JOINT DEFINITIONS

Cut<rlt'.'.l"
vIIIle Hobun BrcJIher.< C",
Fig. 4- t. Weld joint definitions.
Figure 4-2 shows the standard joint specifications for thick-wall

pipes that are to be welded together by the shielded metal-arc
welding method. The included angle of this pipe joint is 75 degrees.
In order to obtain this angle when the pipes are brought together for
welding, the bevel angle on the end of each pipe should be equal to
one-half of the included angle, or 37% degrees in this case.
27%~i/
~(
3/32" to
~ I.- 3/32" to 1/8"

Cmalt'.<) (,(Ihe I/"b"" 8rorhen C",
Fig. 4-2. Standard joint specifications for thick-wall pipes.
Care should be exercised in making the bevel to leave the correct

width of the root face. In Fig. 4-2, the width of the root face for
heavy-wall pipe is shown to be 3/32 to % inch.
Usually the pipes and the pipe fittings are beveled by machining in
a shop before they are sent out on the job and no further preparations are required before they are fit-up, other than cleaning. Sometimes, however, the pipes are not beveled and this operation must be
performed on the job. In this case the bevel is cut with an oxyacetylene cutting torch and finished by grinding with a hand grinder.
Often the bevel is cut manually, using an ordinary hand oxyacetylene cutting torch. This requires care and skill. When available, a
pipe-beveling machine is used. such as shown in Fig. 4-3. This
machine is fastened to the end of the pipe and then is carefully
adjusted to cut the bevel. Next, the oxyacetylene cutting torch is fed
22
Courtesy of Ihe N-& M Pipe Bevelill/i Macl1me Coo, fne.
Fig. 4-3. Pipe-cutting machine_
around the pipe to cut the beveL Depending on the design of the
machine, the feed may be accomplished by a hand crank or by a
self-contained power feed.
The surface produced by the oxyacetylene cutting torch will have
a tightly adhering oxide film. If not removed, this film is very
detrimental to the quality of the weld. For this reason, the beveled
surface should be entirely free of the film before any welding is done
over it. The oxyacetylene cutting torch also produces a rather rough
surface that is difficult to weld over.
Grinding the bevel following the oxyacetylene cutting operation is
done to remove the oxide film and to obtain a smooth, flat contoured
surface over which to weld. On very thin-wall pipe (VB in.) the entire
bevel is sometimes ground with a hand grinder.
Cleaning the Joint Surfaces. Contaminants such as grease, oil, scale,
or rust will have a harmful effect on the quality of the weld. As
already explained, all traces of the oxide film produced by the
oxyacetylene torch must be removed. The welder must also make
certain that any other contaminants are removed before starting to
weld. This can usually be accomplished by the vigorous application
of a wire brush.
Fitting-up the Pipe. The two pipes to be welded together must be
accurately aligned prior to welding. The inside surfaces of the pipes
23
Chapter 4
must blend together smoothly, as should the outside surfaces. In
many instances the pipes must be aligned so that the finished piping
system will be in the correct location. Considerable skill is required
to fit-up two pipes in preparation for welding. This is an essential
part of the pipe welder's craft.
Because an understanding of how to fit-up pipe is so important,
Chapter 14 is entirely devoted to this subject. The present chapter
will treat only the method of fitting-up two short pipe nipples in
preparation for practice welding and the reader is referred to Chapter 14 for further details on this subject.
Short pipe nipples are not only used for practice welding, but also
for jobs in the shop and in the field. The pipe nipples recommended
for practice welding are 7 inches long. They are made from 8-inch
Schedule 60 mild steel pipe, for which the actual outside diameter is
8.625 inches and the wall thickness is .406 inch.
~~~~-SPACER
WIRE
-lr
PROPER ALIGNMENT
C9
1l16
"
C
MAXIMUM MISAUGNMENT
C(Jurles)" o{II1<' Hob<irl Bror/,,",\ ('",
Fig. 4-4. Method of spacing two pipe nipples in preparation for tack welding (a);
Correct alignment of pipe nipples in preparation for welding (b): and maximum
allowable misalignment (c).
24

To fit-up the pipe nipples, first determine the required width of the
root opening. Next. find a piece of wire having a diameter equal to
the width of the root opening and bend it into a vee-shape, as shown
in Fig. 4-4 (a). Place one of the nipples on end on the welding table
(b) and place the other nipple on top of it with the wire between
them to act as a spacer. With the spacer wire in place, align the two
pipes with a straight edge (c). The maximum misalignment allowed
(d) is 1/16 inch, which is specified by the ASME Code. When properly
aligned, the two pipe nipples are ready to be tack welded together.
Tack Welding. After the pipes are aligned, four tack welds, evenly
spaced around the pipe, are made in the root of the weld using a If,,
inch E6010 electrode in this case. Each tack weld should be about
% inch long. Since the tack welds will remain as part of the root
bead, they must be sound throughout. They must be strong, and they
should penetrate to the root of the weld from start to finish.
With the pipe nipples properly aligned, one tack weld is made in
the root of the joint, as shown in Fig. 4-5A. The spacer wire is then
moved so that only the bent end is between the nipples, as shown in
Fig. 4-58. Then the second tack weld is made on the side opposite
from the first tack weld.
PENETRATION
FIRST TACK
WELD
SPACE TOO
NARROW
-",=,=~'"T,jj-FIRSTTACK
--f~~~~~~
,;;;;
SECOND TACK
ONE Of
TWO TACKS
--==='=='==-
(oufle.n' "!" 111, H"bun Bmlh,',", Co.
Fig. 4-5. A. First tack weld with wire spacers in place; B. Second lack weld
made with wire spacer partially withdrawn; C. Method of tapping on table 10
equalize root opening.
25
Chapter 4
Remove the spacer wire entirely from the joint and inspect the
root opening. If the opening on one side is only slightly wider than
the other, weld the wide side next. The shrinkage of the tack weld
will equalize the spacing. If the space is too wide to correct by
welding, bump the pipes on the table as shown in Fig. 4-5C until the
openings are equalized. Then weld the third and fourth tack weld 90
degrees from the first two tack welds.
In most cases the ends of the tack welds should be ground to a
feather edge to facilitate the tie-in with the root bead. This is not,
however, always done. Sometimes a grinder is not available on the
job and the welder must be able to make a tie-in on tack welds and
other welds that have not been ground to a feather edge. If he can do
this, he will have no difficulty in making a good tie-in on weld beads
that have been ground. For this reason, when learning to weld, it is
recommended that ends of the tack welds not be ground.
The correct procedure for welding the tack welds will be
described in the following paragraphs. It will be assumed that the
correct welding rod has been selected and that the welding machine
setting is also correct.
It is important to start any weld correctly, whether tack welding
or welding a longer bead. When the arc is struck, it should not be
shortened immediately. Time should be allowed to stabilize the arc
and to allow the gaseous shield to form.
If an unstable arc is brought close to the root face, the electrode
may stick or small globules of filler metal will be deposited on the
beveled edge which can restrict the manipulation of the arc. The
gaseous shield protects the molten metal from oxidation. In the
absence of the gaseous shield, the molten metal in the puddle will
combine with oxygen in the air to form oxides and thus will not
readily flow into the root opening. When this occurs, the first deposit
is usually a large lump of metal that is seldom properly fused, and in
,;f ..
;,:.i ~.
stAAT 'OF
'1;:, ~
1'itCI(WEUl
'"
~ "'.: ........
- I.
Fig. 4-6. A. Method ofslriking the arc ahead of the tack weld; B. The keyhole.
26

almost all cases the penetration into the root opening is insufficient.
The result is a defective deposit that must be removed from the weld.
For the tack weld, the correct procedure is to strike the arc ahead
of the weld, or that part of the groove in which the weld bead is to be
deposited, as seen in Fig. 4-6A. Maintain a long arc and move the
electrode along the groove to the position where the weld bead is to
start. This serves to preheat this surface as well as to provide time for
the arc to stabilize and for the gaseous shield to form. During this
time the globules of metal transferred from the electrode will be
deposited as spatter outside of the weld. On larger pipes, where it is
often' necessary to deposit a tack weld in the overhead position, no
filler metal will be transferred when a long arc is maintained in this
position.
The electrode shOUld be held at the correct electrode angle, as
shown in Fig. 4-7, A and B. It is held over the starting point until the
edges of the groove begin to melt and then is shortened to obtain the
correct arc length for welding. When the keyhole (Fig. 4-6B) has
CD~ n
5-8 0
TOP VIEW
Fig, 4-7. Correct electrode angle for welding the lack weld.
formed, the weld bead is made, using a slight whipping technique.

This is described further on in this Chapter for welding in the 2G
position and in Chapter 5 for welding in the 5G position.
The Keyhole
The keyhole is an essential part in welding a root bead. It should

be about 1% times the diameter across the electrode coating in size.
The keyhole helps to insure that the deposited bead will penetrate to
the root as evidenced by the small crown at the back of the bead.
This crown, Fig. 4-8, should rise about J/lI, inch above the inside
surface of the pipe. While welding, the welder should watch the
27
Chapter 4
~ 1116"
Fig. 4-8. The root bead. (Note the crown at the base.)
keyhole constantly. If it becomes enlarged, overpenetration will

result. This can be corrected by reducing the welding current or, in
some cases, by increasing the welding speed. If the keyhole is too
small, the penetration of the root bead will be insufficient. While
slowing the speed of welding will help in some cases, usually it is
necessary to increase the welding current to correct this condition.
The Whipping Technique
When the pipe nipples are tack welded in the 2G position, the
weld is horizontal and it is necessary to use a slight whipping motion
to control the molten metal in the puddle. The whipping motion,
shown in Fig. 4-9, is a quick movement of the arc away from the
puddle, in the direction of welding, after which it is returned to the
puddle. The arc must be kept within the groove made by the weld
joint and it should not be carried outside of this area. The length of
this movement should be short, about 1% electrode diameters, in
order to maintain a gaseous shield over the molten metal at aU times.
After each whip the electrode should pause over the top edge of the
keyhole, where the molten metal joins the solidified bead, to deposit
filler metal and to keep the puddle fluid. The whipping motion
should be made by moving the wrist, not the entire forearm. The
purpose of whipping is to decrease the fluidity of the molten metal
puddle, allowing it to become somewhat mushy but not to solidify
completely.
Stopping the Weld
The procedure that is used for laying down the root bead is
continued until the bead is about % inch long. At this point the weld
28
Fig, 4-9, Welder's view when welding the lack weld in the horizontal position. A
slight whip should be used. as shown, to prevent the molten metal in the puddle
from dripping.
CorJrlt!sy ,4 tilt! Hob"'l 8mthas Co.

Fig. 4-10. Quenching the arc by a stab through the keyhole when welding a root
bead.
is stopped. The procedure for stopping the weld is to push the

electrode through the keyhole by a quick movement, as shown in
Fig. 4-10. After the arc is extinguished, it is quickly withdrawn. This
method of stopping a root bead helps to obtain full root penetration
when restarting the weld at this point and when making a tie-in.
29
Chapter 4
The tack welds should be cleaned and inspected before welding
the remainder of the root bead. AU of the slag coating must be
removed with a chipping hammer, and the surface of the bead
should be brushed vigorously with a wire brush. Any defects should
be removed by grinding or with a hammer and a chiseL The
defective areas should be rewelded, although this may be done after
the remainder of the root bead has been welded. A perfect tack
weld is shown in Fig. 4-1 1.
Fig. 411. View of a perfect tack weld.
30
CHAPTER
Uphill Welding the Root Bead on

Heavy-Wall Pipe (SG Position)
The Root Bead
The root bead is the foundation of a successful pipe weld and

must be made as perfectly as possible. An illustration of a perfect
root bead is shown in Fig. 5-1.
Fig, 5-1. A. Top view ofa perfect root bead; B. View of section through the root
bead.
Perhaps the most difficult position in which to weld pipe is in the

horizontal, or 50, position. Once this is mastered, welding pipe in
other positions is less difficult to learn. For this reason, it is best to
start by learning how to weld in the 50 position.
For practice welding in the 50 position, a convenient method of
holding the pipes is to clamp them in a fixture or pipe stand as shown
in Fig. 5-2A. Two previously tack welded pipe nipples are shown
clamped in position on the pipe stand in Fig. 5-28. When clamping
the pipe nipples in the fixture, they should be positioned so that the
tack welds are in the positions shown in Fig. 5-2C.
31
Chapter 5
(U~
CouYln'}' {Jfl"~ Hobart 8ru,h~r.\ (0.
Fig. 5-2. Pipe welding stand. A. Unloaded stand; B. Pipe clamped in place;
C Position of lack welds.
Figure 5-3 gives a view of the details of the pipe stand. While all
of the details are shown, not all of the dimensions are given because
they depend on the size of the pipe to be welded and must be
specified accordingly. From this illustration it should be possible to
design a pipe stand on which the practice welds can be made.
General Procedure for Uphill Root Bead Welding
The procedures given in this chapter are recommended for welding the root bead on all heavy-wall pipe. In the following discussion
8-inch Schedule 60 mild steel pipe will be used as the example of
welding in the 50 position. The actual outside diameter is 8.625
inches and the wall thickness is .406 inch. For welding this pipe a
lfg-inch E6010 electrode should be used. The pipe should be
clamped on the fixture with the tack welds in the 2, 5, 8, and 11
o'clock positions, as shown in Fig. 5-2C. Since the pipes to be
welded are heavy-wall pipes, the uphill welding method will be used.
The general procedure is to start at the 6:30 o'clock position and to
weld up to the 12 o'clock position around one side of the pipe, and
again, from the 6: 30 o'clock to the 12 o'clock positions around the
other side of the pipe, to close the weld.
Three basic welding positions are used to weld the root bead when
the pipe is in the horizontal, or 50, position. The weld is started by
welding in the overhead (4G) position; then gradually there is a
32
Uphill Welding the Root Bead on Heavy-Wall Pipe
1/8
~_8
__
~~~E~B_L'l'
._. .
__
lj/16D_ :I10l IN
3H
3,CENTER 3" OVER HOLE
AND WELD_
It OR1LL
6A
6B
.; _ .0:
./8
I,
* CNTE~ Nul OVER HOlE" A.N:D- WELD
1/8
~~2
._1/,
2
2\120
COlU/esy o/Ihe Hobarl 8m/hers Co.
Fig. 5-3. Details of the pipe stand.
change to the vertical (30) position, welding in the uphill direction.

From this position there is another gradual transition to the flat (I G)
position.
When welding around the pipe from the bottom to the top, the
transitions from one position to another are gradual. This requires
the welder to be alert in order to maintain the correct electrode angle
at all times and to change his welding technique as necessary.
33
Chapter 5
Welding in each of the positions will be described separately. However, the welder must develop the skill to recognize whjch welding
method to use and when a change must be made.
The Keyhole
Before going further, it is necessary to say a word about the

keyhole. As seen in Fig. 5-4, the keyhole is a teardrop or pear-
3/16" OIA.
KnHOlE
Courtny of til" Uohart Bmther,'
c".
Fig. 5-4. The keyhole.
shaped enlargement of the root opening ahead of the bead. When

welding a root bead, the keyhole is necessary in order to obtain the
required weld penetration.
The keyhole is the basic guide to the welder as he welds the root
bead. When starting to weld, the welder should make certain that the
keyhole has formed. The correct amount of penetration is obtained
when the keyhole is about 1% times the diameter across the electrode coating, or slightly less. While welding the root bead, the
welder must pay careful attention to the keyhole and watch for
changes in its size. If it becomes enlarged, excessive penetration,
burn-through, or internal undercut will result. If it is too small, the
penetration of the root bead will be insufficient to produce a satisfactory weld. These defects are shown later in Fig. 5-14.
Starting the Root Bead
Since the weld is started at the lowest position on the pipe, it is

usually necessary for the welder to be in a kneeling or crouching
position. Whether kneeling or crouching, he should be situated
comfortably in order to avoid any unsteadiness in manipulating the
welding electrode. Getting into a comfortable position, then, is the
first step in welding a root bead (see Fig. 5-5).
With the machine setting and all of the electrical connections
made, the welder starts by striking an arc between the 6 and 6:30
34
Fig. 5-5. Starting to weld the pipe joint.
4-0'ClOCK POSITION
6:30-0'ClOCK POSITION
Fig. 5-6. Correct electrode angle for welding in the 6:30 position.
35
Chapter 5
position on the pipe joint. The arc should be struck in the joint and
never on a tack weld. A long arc length should be maintained for a
short period after the arc has been struck. During this time he should
weave the electrode backward and forward to preheat the bevel
ahead of the weld.
In addition to preheating the bevel. maintaining the long arc
length stabilizes the arc and allows the gaseous shield to form. No
filler metal is transformed from the electrode to the workpiece when
the long arc is maintained in the overhead position.
After the arc has been stabilized and the gaseous shield has
formed, the electrode is brought back to the 6: 30 position, which is
the actual starting position for the root bead. With the electrode held
at the correct angle (see Fig. 5-6), it is carefully moved toward the
root bead until the correct arc length is established. If necessary, the
electrode is held momentarily in this position, long enough to form
the keyhole and the puddle of molten metal. Then the electrode is
advanced at a slow, steady pace to form the weld bead.
Uphill Welding the Root Bead
With the keyhole and the puddle established and with the electrode held at the correct angle. the electrode is advanced around the
bottom of the pipe. There is no back-and-forth motion across the
pipe joint, only a linear, or straight, movement along the joint in the
direction of welding. The movement of the electrode should be slow
and uniform. Maintaining the correct arc length. the end of the
Fig. 5-7. Welder's view when welding the rool bead at the bOllom oflhe pipe.
36

electrode should be kept near the top of the keyhole, which is the
part of the keyhole adjacent to the deposited weld bead. The
position of the electrode is shown in Fig. 5-7. If the current setting is
correct, it should not be necessary to resort to whipping in order to
control the puddle and the size of the keyhole.
While welding, the welder should pay careful attention to the
puddle and to the keyhole. As the electrode moves along at a steady
pace it melts the edges of the bevel in front of the arc and the molten
metal flows toward the back of the arc. There it enters the puddle
and flows into the root opening. As the electrode moves on, the
molten metal in the puddle that is left behind solidifies to form the
weld bead. Since the bead will be deposited in a circular pattern
around the pipe, the tilt of the electrode must be changed gradually
to maintain the correct electrode angle.
If the keyhole size becomes too small, the electrode angle should
be decreased slightly by pointing the electrode more directly toward
the keyhole. Also, the welding speed may be decreased slightly.
Should the keyhole become enlarged, the remedy is to increase the
electrode angle slightly. If this is not effective, welding should be
stopped and the current setting reduced before proceeding again.
Whipping is sometimes used in overhead welding to reduce the
keyhole size and to control the puddle. However, this practice is not
recommended and should not be used unnecessarily in the overhead
position. When excessive Whipping appears to be necessary, it is best
to reduce the current setting.
Stop and Restart. When welding the root bead, the electrode is
consumed and will need replacement. This necessarily involves
breaking the arc. Welding is also stopped after a tie-in has been
made with another weld, such as a tack weld; however, the procedure for making a tie-in will be treated later on in this chapter.
To stop the weld, the arc is broken by making a quick stab
through the keyhole with the electrode and then by withdrawing it
quickly to clear the work. By this procedure a full size keyhole is left
so that complete penetration can be obtained when the weld is
started again.
Before restarting the weld, the slag coating of the weld bead
should be chipped off with a chipping hammer and the bead should
be wire brushed for a distance of 1 inch, or more, from the keyhole.
All traces of the slag coating should be removed to eliminate the
danger of any of it being trapped in the molten metal when restarting.
To restart the weld, the arc should be struck on the part of the
bead that has been cleaned. Maintaining a slightly longer than
37
Chapter 5
normal are, the electrode should be brought forward to the edge of
the keyhole and held there momentarily to allow the arc to stabilize
and the gaseous shield to form. It is also held at this length in this
position to allow time for the liquid puddle to form at the edge of the
keyhole. When sufficient liquid metal appears at the edges of the
keyhole, but not before, the arc can be shortened to its normal
length and the electrode manipulation can be started to resume the
weld,
Vertical Uphill Welding of the Root Bead
The nature of the welding process changes gradually from overhead welding to vertical uphill welding as the bead progresses from
the 5 o'clock to the 4 o'clock position. As the weld moves toward
the vertical position, it becomes apparent that the liquid metal will
tend to flow downward at a faster rate than when welding in the
overhead position.
When the continuous application of heat resulting from the slow
steady movement of the electrode starts to cause an overflow of the
molten metal, the remedy is to resort to the whipping procedure. The
whipping procedure must always be used when welding in the
vertical uphill portion of the pipe. It is continued until the weld is
stopped at the 12 o'clock position.
c='--~-7
Fig. 5-8. Correct electrode angle for vertical uphill welding of the root bead of a
pipe joint.
The electrode must be held at the correct angle, as shown in Fig.
5-8, and the electrode must follow the path shown in Fig. 5-9A. It is
moved upward about one electrode diameter, Fig. 5-9B, and then
returned to the face of the keyhole, which should be about one and
one-half the electrode diameter.
The electrode should then pause at the face of the keyhole, which
is the lower part of the keyhole adjacent to the deposited weld bead.
In this area the intense heat of the arc can be absorbed by the metal,
and the filler metal from the electrode is deposited here in order to
build up the bead progressively. The electrode should not pause
38
Counes)-
or lhe Jloharl
Hro/hc;-.\' ("0.
Fig. 59. A Welder's view when whipping, showing the path of the electrode. B.
The length of the stroke when whipping should be approximately one electrode
diameter hut should not exceed 1h electrode diameters.
directly over the keyhole because the initial heat in this area will
cause the intense heat of the arc to melt the edges around the
keyhole. This will result in excessive penetration and possible burnthrough.
Whipping should be done by a precise wrist movement and not by
moving the entire forearm. This procedure can be described as a
repeated "whip and pause."
The objective of whipping is to allow the molten pool of metal to
cool sufficiently to lose some of its fluidity. When the molten metal
in the puddle is somewhat mushy, a further deposit of filler metal
from the electrode will not cause it to overflow.
The length of the stroke when whipping should not be excessive.
If it is excessive, the hot liquid metal in the puddle will be exposed to
the atmosphere as a result of the removal of the gaseous shield.
Rapid oxidation will result, which leads to porosity in the weld,
Excessive whipping can also cause slag entrapment in the weld.
Some welders have a tendency to use a current setting that is too
high. Then, to prevent overflowing, they resort to whipping, even in
the overhead position which soon becomes excessive. In such cases,
better results are obtained if the welding current is reduced and
whipping is kept to a minimum. The length of stroke when whipping
should not exceed one and one-half electrode diameters and preferably be less, in order to minimize the effect of uncovering the gaseous
shield from the weld and to prevent slag entrapment.
39
Chapter 5
Flat Welding the Root Bead
Conditions approaching flat welding occur as the weld progresses

to the vicinity of the I o'clock position. When this occurs, the molten
pool of metal becomes even more difficult to control and the
whipping procedure must be continued until the weld is stopped in
the 12 o'clock position.
When welding a single-vee groove, open butt joint in the flat (l G)
position, the molten metal will tend to drip through the opening
causing the bottom of the weld to build up and form a high crown.
This is excessive penetration. In more severe cases, burn-through
will result, especially if the current setting is too high. In some shops,
back-up plates are positioned at the bottom of the joint to prevent
this from happening. However, this is not necessary when the weld is
made by a skilled welder, using the whipping procedure and the
correct current setting.
From what has been said above, it is evident that the welder must
watch the puddle and the keyhole for signs of excessive penetration.
He must continue the whipping procedure to control the pool of
molten metal. If signs of excessive penetration occur, he may
increase the electrode angle somewhat as shown in Fig. 5-IOB. The
Fig. 5-10. Correct electrode angle for flat welding at the top of the pipe.
normal electrode angle for welding in this position is shown in Fig.

5-IOA. If this does not help, the weld should be stopped and the
current setting reduced. Excessive whipping should not be used to
correct this situation.
40
Uphill Welding the Root Bead on
Heavy~Wall
Pipe
If the current setting is correct and the correct whipping procedure is used, a perfect root bead can be welded in the flat position on
top of the pipe joint. When the weld bead has reached the 12 o'clock
position, the weld should be stopped by stabbing the electrode
quickly through the keyhole and withdrawing it when the arc is
quenched. After the weld has cooled sufficiently, the slag coating is
removed in preparation for welding the second half of the pipe joint.
Welding the Second Half of the Pipe Joint. The procedures used for
welding the second half of the pipe joint are identical to those used
in welding the first half.
To start the second half of the pipe joint, a restart must be made
in the 6: 30 position against the starting point of the first root bead. It
is advisable to restart with a new electrode so that a continuous bead
can be deposited, if possible, until a tie-in is made to the tack weld in
the vicinity of the 8 o'clock position. Then, after making a restart at
the other end of the tack, the weld is made around the pipe, stopping
when necessary to replace the electrode and to make the tie-in with
the remaining tack weld, until the final tie-in is made at the 12
o'clock position to close the weld. The procedure for making a good
tie-in, which must be mastered, is given in the next section.
The Tie*ln Procedure
The process of joining a bead, in this case the root bead, is called
a tie-in. The previous weld may be a ta~k weld or it may be the first
half of the root bead. In either case, the two welds must be brought
together smoothly and without discontinuities.
Making a tie-in requires extra care in welding. It is easier to make
a good tie-in if the edges of the existing weld are ground to a feather
edge, as shown in Fig. 5-11. This is done with a hand grinder, using
a thin grinding wheel. When the ends of the existing bead are ground
to a thin edge, the metal in the bead will heat up more rapidly than
..,..,~'"
GRIND SIDES TO SMOOTH
ROUND EDGE
GRIND ENDS to SHARP
\ ", .'
,,"
~~
/.~
FEATHERED EDGE
%
~
Courtesy of the Hobart Brothers Co.
Fig. 5-]1. Edges of a tack weld ground to a feather edge in preparation for
making a tie-in.
41
Chapter 5
in the case of an unground edge, where there is a. relatively large
bulk of metal.
Sometimes facilities are not available for grinding the edges of the
existing bead, and the tie-in must be made to an underground edge,
which is a more difficult job. Since this situation does occur, the
beginner should first practice making the tie-in to an unground edge.
Having mastered this technique, he will have no difficulty in making
a tie-in to a feathered edge.
Sometimes the tie-in must be made on approaching the keyhole,
while at other times it must be made by approaching the opposite,
heavy end, of the bead. These conditions are shown in Fig. 5-12.
Two different welding techniques must be used in these cases.
KEYHOLE
KEYHOLE
APPROACHING
HEAVY END OF
TACK WELD
APPROACHING
KEYHOLE SIDE
OF TACK WELD
Fig. 5-12. A. Tie-in being made to heavy end of bead; B. Tie-in being made to
keyhole end of bead.
When the root bead approaches the keyhole, maintain the same
speed of welding used to deposit the bead. Weld toward the previous
weld at this speed and then gradually close the keyhole. As the
keyhole is beginning to close up, watch the liquid puddle. When the
liqUid puddle appears to have joined the previous weld in a smooth
pattern, withdraw the electrode by simultaneously reversing the
electrode beyond the point of the tie-in and lengthening the arc
somewhat; then break the arc with a sudden movement
Sometimes the keyhole will become enlarged as the arc
approaches the point of the tie-in. In this event, the whipping
technique is used to cool the metal and to avoid excessive penetration at the tie-in. If necessary, there should be no hesitancy in using
the whipping technique; however, the welder must be sure that there
is penetration to the bottom of the root face at the tie-in. A perfect
tie-in is shown in Fig. 5-13.
42
WELD QUALITY INSPECTION

A GOOD WELD
/WELL FUSED
--SLIGHT
CROWN
SLIGHT CROWN
EVEN RIPPLE
END
TOP
BOTTOM
COMMON WELDING MISTAKES

POOR APPEARANCE
1. Welding current too high
2, Foully joint preparation
3. lack of arc control
INSUFFICIENT PENETRATION
1. Welding current too low
2. Travel speed too fast
3. Root opening too narrow
EXCESSIVE PENETRAliON
,. Root opening too wide
2. Welding current {oc high
TriNe! spe:ed 100 S frr~'
3~
BAD TIE-IN
1. Current
100 ~ow
2. Wrong electrode angle

3. Wrong start~ng H~chruque
'1"1'f
"";,,t f ,~J(
I
'
1
.....
I~~i:{:'
1i.
~.
:-.~ ~lf~
~
''i/i!i
SUCK-IN
1. Welding cum:!rH too high
2. Travel speed too slow
3. Root opel1i 119 too VI ido
BURN-THROUGH
1. 'Welrlrng cllnenl"I'IV JI~~;jl
2. Travel speed too

3, Rool
opon~nu:
too
Sl{IW
Wld(J
Courtesy of the Hobarf Brofhers Co.
Fig, 514, Weld quality inspection showing good weld and common welding
defects.
43
Chapter 5
Fig. 5-13. (Left) A perfect tie-in to the keyhole end: A. Tic-in; B. Restart.
(R ighll A perfect tie-in to the heavy end.
When the tie-in is made at the heavy edge of the weld, the speed
of welding should be decreased somewhat, a short distance (approximately % inch) away from the edge of the previous weld. This is
done to allow time for the thick edge to heat up. If the arc is too
short and the welding speed is too fast when approaching the tie-in,
insufficient penetration will result. The welding speed should be slow
and the arc length normal, until the puddle joins the thick edge of
the previous weld. When the pUddle is tied-in smoothly with the
previous weld, the electrode is moved slightly in the reverse direction
while at the same time lengthening the arc slightly; the arc is then
broken with a quick movement.
Inspect the Weld
After the entire root bead has been completed, it should be

thoroughly inspected for visible welding defects. Figure 5-14 illustrates such defects, which are discussed in detail in Chapter 11. Any
major defect should be removed before the second bead is deposited.
A portable grinder equipped with a thin edge composition grinding
wheel can be used to remove the defect.
If it is necessary to remove a large section or a deeply penetrating
section of the root bead to correct a defect, this part of the bead must
44

be rewelded before the second bead is deposited. However, if the
correct welding technique described in these pages has been used
with skill and care, very few or no defects should appear on the
weld. The welding techniques described so far have been proved to
result in a successful pipe weld by countless applications in the field.
Poor Fit-up, Wide Root Opening. With careful workmanship it

should be possible to obtain a good fit-up at all times in preparation
for welding pipes together. A good fit-up is not uncommon in pipe
welding. However, poor fit-up cannot be ignored because, on occasion, welding must be done under these circumstances. Frequent
causes of poor fit-up are manual flame cutting of the bevel by
inexperienced operators, or mistakes in measurements.
One condition encountered as the result of poor edge preparation
is a wide root opening. Wide root openings are difficult to weld and
in such cases a decision must be made whether to attempt the weld
or to replace the entire length of pipe. Replacing the entire pipe can
be very expensive and sometimes cannot be afforded. Usually this
decision depends on the type of job. For example, a poor fitup job
would be unacceptable in a chemical or an atomic plant
If the root opening is too wide, the first step is to join the pipes
together with tack welds. If the tack welds must be made with the
pipes in the horizontal, or 5G, position, they should be welded in the
12, 3, 6, and 9 o'clock positions around the pipe. If the root
openings are normal, the tack welds are made in the 2, 5, 8, and 11
o'clock positions. When tack welds are made in the 6 o'clock
position, the root bead is started against a tack weld, contrary to the
usual procedure.
In starting the tack weld it is first necessary to build a bridge of
weld metal across the wide gap at the root. This is done by welding a
number of small nuggets of weld metal on the root faces until finally
-i 7'\3-
T\
}-
-{
l\
-+ /\ t
Fig. 5-15. Bridging a wide root opening by depositing a series of nuggets.
45
Chapter 5
one of the nuggets joins the two faces together. This procedure is
shown in Fig. 5-15.
The welding current should be reduced slightly to weld the
nuggets. After the arc is struck it is held just long enough to become
stabilized and then it is shortened. A small nugget of weld metal is
deposited on the root face after which the arc is quenched. It need
not be fused perfectly, the principal objective being to deposit a
small lump of metal. The same procedure is then used to deposit a
small nugget of metal on the opposite root face. After cleaning each
nugget, the same procedure is used to deposit one nugget on top of
another until the bridge is buill.
The purpose of the bridge is to form a metallic path across the
wide root opening that can be used to maintain the arc when starting
to weld the bead. These nuggets are imperfectly fused. Because the
gaseous shield may not have formed completely, they are likely to be
porous and lumpy. For this reason they must be removed after a
strong bead of perfectly welded metal has been deposited. Often
this is done by grinding after the first half of the weld has been made
around the pipe joint. The entire bridge and about liz inch of
additional metal along the bead should be removed with a grinding
wheel or with a hammer and a chisel.
After the bridge has been built across the root opening, the
remainder of the tack weld can be deposited. A slightly reduced
current setting should be used in this case. As usual, the bridge
should be cleaned before striking the arc. The arc is struck in the
joint ahead of the bridge and a long arc is held until the arc is
stabilized and the gaseous shield has formed. It is then brought over
the bridge and shortened. The arc length should be slightly shorter
than normal. It is moved slowly across the bridge once or twice until
some liquid metal appears; the weld bead can then be deposited.
The weld bead is deposited by using a U-weave, as shown in Fig.
5-16. It is necessary to use this weave in order to bridge the wide
root opening. Because the root opening is already wide, a full
keyhole may not form; however, the edges of the root face must be
melted at the edges of the weld zone.
One danger encountered in welding across a wide root opening is
excessive penetration resulting from overheating. For this reason the
arc length should be somewhat shorter than normal and the current
setting should be somewhat lower than normal.
In making the V-weave, Fig. 5-16, the electrode should be manipulated to bring the arc all the way out of the molten puddle. The arc
should be moved along the face of the bevel and kept away from the
edges of the bevel. The molten pool of metal should be allowed to
46
(ol.<rlesy oj till' Hobart Brothers Co.
Fig. 5-]6. Welder's view of the correct procedure for making a V-weave.
solidify completely before the arc is returned to the weld zone. This
requires the use of a long V-weave. Then, where the arc is returned
to the weld zone, remelting of the solidified metal takes place to form
a somewhat smaller and less fluid pool of liquid metal. When the
filler metal from the electrode is deposited in this puddle, the molten
metal will not overflow the root opening.
If a slight amount of overheating does occur, the electrode angle
should be increased. If this does not prevent overheating, it may be
necessary to discontinue the arc at short intervals to allow the weld
to cool before welding is again resumed. This is indeed a slow
process which calls for patience and smooth electrode movement.
When the tack weld is about % inch long, the arc is quenched in the
usual manner.
When welding a wide root opening, the root bead is started at the
6 o'clock position, against the end of the tack weld. The arc is struck
on the tack weld and a long arc is maintained over the end of the
tack weld in order to preheat this area. As soon as some molten
metal appears at the edges of the tack, the arc is shortened and
welding is commenced.
47
Chapter 5
The root bead is welded as described above, using a short arc and
a long V-weave. Care must be exercised to avoid overheating and
excessive penetration. The V-weave is used to weld entirely around
the joint. Before welding the second half of the pipe joint, the bridge
used to start the weld and about liz inch of the tack weld must be
removed.
Poor Fit-up; Narrow Root Opening. The pipe welder will encounter
situations where the root opening is smaller than the recommended
size; sometimes the root may be closed entirely. This situation may
be corrected by recutting the edge of the pipe with an oxyacetylene
cutting torch or by grinding the edge with a hand grinder.
Often, however, the joint is simply welded together without
making any alteration to the root opening. To do this. the same
techniques for welding a normal size root opening are used with only
slight modifications.
When the root opening is too small or nonexistent, the welding
current flow must be increased, usually by as much as 10 to 20
amps. The heavier current flow increases the penetration of the
weld. Also, it is usually best to decrease the arc length when welding.
Sometimes a near "drag" arc length can be used.
The welder must give close attention to the amount of penetration
obtained while welding the narrow root opening. If the penetration
is too deep or overpenetration occurs. a slight whipping motion may
avoid this condition. Increasing the speed of welding may also be
helpful. If these measures do not succeed in correcting the overpenetration, the current setting must be reduced. Underpenetration can
be corrected by increasing the welding current and by using a
somewhat slower welding speed.
When confronted with a narrow root opening, it is always good
practice to make some test welds on pieces of scrap metal first. The
plates should be beveled and set apart a distance equal to the narrow
root opening. In this way the current setting and the welding
procedure can be adjusted before welding the pipe; and defective
welds. which must later be removed, can be avoided on the pipe.
Poor Fitup; Wide and Narrow Root Openings. Sometimes a combination of narrow and wide root openings is encountered in a pipe
joint. This can be the result of the ends of the pipe being cut
incorrectly or of the pipes being misaligned - that is, oriented at a
slight angle to each other.
Welding a combination narrow and wide root opening requires
the application of both of the methods previously described. The
method used obviously depends upon the root opening encountered.
48

It is usually necessary to adjust the welding current setting several
times when welding around the pipe.
First of all, of course, the four tack welds should be welded in
place. To prevent restraint cracking, the first tack welds should be
made in the region of the narrow root openings. For the same reason
and because it is less difficult to obtain a start, it is desirable to weld
the narrow root opening fir~t, when welding the root bead. This is
not always possible, however, because the basic method of welding
the pipe from the bottom toward the top should not be abandoned.
For this reason it may be necessary to start the root bead in a region
having a wide root opening.
Poor Fit-up; Root Face Too Wide. When the root face is too wide, it
may be possible to correct this condition by recutting the end of the
pipe. Normally, however, the pipes are welded together in the usual
manner, using a higher current setting to obtain complete penetration.
Poor Fit-up; Root Face Too Thin. In this case the edge, at the root,
will melt very rapidly while the rest of the weld is relatively cold.
The method of overcoming this difficulty is to reduce the current
flow and to use a U-weave for depositing the bead. By reducing the
current setting less heat is generated by the arc and the possibility of
burning through the thin edges is diminished. The U-weave preheats
the bevel ahead of the edge, thereby insuring that the weld deposit
will fuse properly with the parent metaL As before, when using the
U-weave, it may be necessary to allow the puddle to become mushy
before the arc is returned each time.
Welding the Root Bead with Low-Hydrogen Electrodes. Root beads
are seldom welded with low-hydrogen electrodes because very highly
skilled welders are required to make welds that are free from defects.
High-pressure pipe joints are usually welded by welding the root
bead with a deeply penetrating type electrode, such as E60 I0, or by
the Gas Tungsten Arc-Weld process (GTAW, described in Chapter
6), and the remainder of the joint with a low-hydrogen electrode.
However, with skill and care it is possible to weld root beads with
low-hydrogen electrodes. The following instructions describe how
this may be done.
As shown in Fig. 5-17, the diameter across the coating of the
Va-inch E7018 low-hydrogen electrode is larger than that of the
Va-inch E6010 electrode. The heavier coating of the E7018 electrode does not allow the arc to be taken close enough to the root
face, thereby making it difficult to establish the correct arc length.
Welding with an arc that is too long can cause suck-in (see Fig.
49
Chapter 5
1F
::
220lCOATINGl
!I
.125 (WIREI
6010 (APPRoxl
E 7018 (LOW HYDROGEN)
L::::J1~,~
II
-1r"
.125 {COATINGl
-,r-093{WIRE)
E 7018 (lOW HYDROGEN)
Fig. 5-17. A. Diameters across the coatings of .125 in. 60 I0 and 70 1~

low-hydrogen electrodes; B. Diameter across the coating of a .093 in. E7018
low-hydrogen electrode.
5-14). Furthermore, the heavier coating interferes with the manipulation of the electrode when making a weave. For this reason a
smaller size (.093 inch) hydrogen electrode should be used to weld
the root bead. It has a diameter across the coating that is approximately equal to that of the %l-inch 6010 electrode.
When welding with low-hydrogen electrodes, pinholes can be
caused by incorrect arc striking, chipped flux coating, moisture in
the weld joints, or wet electrodes. To avoid pinholes while striking
the arc, strike just ahead of the starting point and shorten the arc as
quickly as possible to the proper length. Then back-up the arc to the
starting point and proceed to weld as soon as the molten pool of
metal has formed. Chipped spots on the electrode coating will cause
the arc to be erratic at that point, resulting in pinholes and a hard
zone in the weld. For this reason electrode containers should be
handled with care and damaged electrodes discarded. Wet joints
should be heated with an oxyacetylene torch to drive off all moisture.
The electrode flux coating is sensitive to moisture and must be
kept dry. Open containers should be stored in a "dry box" or
controlled-humidity storage oven, where the electrodes are kept at a
temperature of 300 to 400F. Electrodes that have been exposed to
moisture can be reconditioned by drying for one hour at a temperature of 600 to 800F, the exact temperature depending on the make
of the electrode.
A higher current setting is almost always used for welding with
low-hydrogen electrodes and, therefore. more heat is liberated. The
arc characteristic is also different. The low-hydrogen electrode produces an arc that is relatively smooth but lacks the penetrating
power of the more lightly coated electrodes.
The heavy electrode coating will form a heavy blanket of slag
over the liquid puddle of molten metal, which causes the cooling rate
to be slower and the metal to remain liquid for a longer period. The
viscosity of the molten slag and weld metal is lower; i.e., it will flow
50

more readily than a liquid having a higher viscosity. The combined
effect of the slower cooling rate and the lower viscosity of the liquid
will cause the molten puddle to drip readily. For this reason,
difficulty is experienced when welding in the overhead and vertical
positions with low~hydrogen electrodes.
The general procedure for welding the root bead with lowhydrogen electrodes is the same as before. Tack welds are made
around the pipe, after which both sides of the joint are welded from
the bottom to the top of the pipe. The welding technique. however, ;s
different.
A very short arc should be used at aU times when welding with
low-hydrogen electrodes. The electrode must be kept very close to
the root face in order to obtain adequate penetration. When the arc
is struck. it should be shortened immediately, as explained before.
The whipping procedure should never be used when welding with
low-hydrogen electrodes.
A V-shaped weave, Fig. 5-18, must be used for welding the root
bead entirely around the pipe when using the low-hydrogen electrode. The objective of this weave is to allow the molten slag and
metal in the puddle to cool and to lose some of its fluidity (or to
increase its viscosity) in order to prevent dripping. Using this weave
also preheats the metal ahead of the weld.
The weave should be made by smooth and precise movements of
the wrist. The arc should be brought out of the puddle and up along
the bevel with a quick movement. Allowing just enough time for the
puddle to lose some of its fluidity by slowing down the return
movement, the arc is returned to the puddle and held there for a
short pause. This movement is then repeated up and along the other
bevel.
The liquid metal must never be allowed to solidify during the
weave. Some of the slag may also solidify and, if this occurs along
Fig. 5-l g Welding

the root bead with
a low-hydrogen
electrode using a
V-weave.
51
Chapter 5
the edges of the weld, it may be difficult to dissolve the heavy slag
entirely - resulting in a lack of fusion. If the puddle is always kept
liquid, the slag will seldom be trapped because it has a lower melting
point than the weld metal. Since the arc is smooth and lacks deep
penetrating power, the movements must be smooth and precise to
avoid solidification of the puddle.
Summary of Root Bead Welding
A perfect root bead should be free from undercuts, porosity,

incomplete fusion, insufficient penetration, and excessive penetration (see Fig. 5-14). All of these defects can be avoided by learning
and practicing the correct welding procedures.
It should be kept in mind that these defects are the responsibility
of the welder. Porosity cannot be blamed on the equipment, but
rather on not cleaning the weld sufficiently when grease, oil, and rust
are present prior to welding. Other causes of porosity are defective
electrode coatings such as chipped coatings, flaking of the coating,
and coatings containing an ex.cessive amount of moisture. All of
these electrode defects can be detected by the welder before he
strikes the arc.
Proper root opening and edge penetration help to attain complete
penetration. The composition of the weld metal is also affected by
the edge preparation. When the spacing and the edge preparation
are correct, the welder will be able to manipulate the electrode
comfortably so that there will be a better intermixing of the base
metal and the filler metal.
Restraint cracking occurs when small welds are made on thick
metal sections, such as heavy-wall pipe. This subject is discussed at
length in Chapter 13. Suffice to say here that the size of the root
bead must be large enough to withstand the shrinkage stresses
without cracking. This calls for careful attention to the condition of
the joint before welding.
The presence of hydrogen in the weld or in the base metal heataffected zone becomes dangerous ifthe microstructure in both ofthese
areas becomes martensite, with hardness exceeding 30 Rockwell. The
danger increases as the carbon content increases. The base metal heataffected zone needs particularly close attention to prevent cold crack~
ing. Hydrogen alone cannot be blamed for under bead cracking; other
contributing variables in the immediate vicinity of the weld area must
be considered as well. The following welding procedures can also con-
52
Manifold - Welds been x-ray-(Perfect) and hydrostaticly tested

OK for receuvubg: station. (crude oil)
53
Chapter 5
tribute to cold cracking: I) reaction stress from restraint of the base
member; 2) residual stress from the unequal stress caused by contraction of the base metal and weld metal; 3) the development of short
range stress due to the transformation; and 4) structure stress resulting
from the austenite4o-martensite transformation. This transformation is
influenced by the alloying elements, especially the carbon content in
respect to higher stress levels; it is likely to increase further from the
melting of the base metal and inducement of filler metal into the weld.
The use of premium electrodes carrying the prefix LC (Low Carbon)
is strongly recommended.
Multiaxial stress develops in those alloys with limited toughness and
ductility. There are limited options for accommodating this high stress.
If hydrogen is present in such a weld, then very little energy will be
needed to promote cracking. Hydrogen in a weld of hardenable steel is
a powerful promoter of cracking.
Carbon steel generally contains less than 0.30 percent carbon. When
joined alloy steel containing 0.10 percent carbon will usually caB for
countermeasures against the influence of hydrogen. Even the addition
of manganese to a level of about 1.25 percent in a carbon steel that
contains 030 percent carbon might need precautions against hydrogen
when welding heavy sections.
It is not difficult to decide when countermeasures are needed against
hydrogen if consideration is given to the principle of anticipated
microstructure, as discussed earlier. The reasoning for such apprehension is clear. Such steel in heavy sections with tend to form martensite
in the heat-effective zone, unless welding was conducted with controlled heat input to secure a relatively slow cooling rate.
Welding procedure are not just for information, but must be implemented. The specifics cannot be ignored during welding of an alloy
material, nor can the welder fail to follow these written instruction. For
example, if preheating and the interpass temperature are ignored, the
metal experiences a cooling rate that would likely lead to a martensite
structure with poor ductility and toughness. Steel has a tendency to
develop a higher level of hardness with an increasing cooling rate.
Therefore, it can be difficult to detennine if a service failure is due to
defective welding or simply weld induced brittleness.
It is important to control cracking in medium or higher carbon steel,
and in steel which incorporates any alloying element to achieve qualities such as strength and hardness, corrosion resistance, impact resistance and toughness, and toughness at sub~zero temperature.
Sometimes adding a single alloying element will work; in other cases
multiple elements are more effective. For instance, it is uncommon to
54

find iron without carbon as a strengthening additive. Carbon is more
powerful and effective than any other element when added in small
quantities. Its effectiveness increases as it is increased for tensile
strength and hardness.
Carbon, as an alloying element, is soluble in high-temperature iron
to an appreciable extent, but is soluble in the low-temperature form
only to a limited extent. With the addition of alloy to a point where its
high temperature crystalline form exists and the alloying element
enters the lattices to form a solid solution. Upon cooling the alloy to
the point where the crystalline transformation occurs, the alloying element suddenly experiences the full effects of limited solution (solid),
and precipitation takes place. This results in embrittlement and
reduced ductility in the HAZ.
The cooling rate must be regulated to produce a fine dispersion of
precipitation throughout the metal so that it has high ductility and a
fine grain structure. Otherwise, the structural condition will result in a
marked decrease in ductility. Transfonnation hardening is the principal mechanism used to increase the hardness and tensile strength of
carbon and alloy steel. Tensile strength moves to higher levels when
the carbon and manganese content increase. In the alloy steels, moderate addition of carbon also increase these mechanical properties, but
reduces ductility and toughness.
Depending upon the welding process employed, the conditions
under which steel hardens to its martensite structure varies. Also,
residual stress, reaction stress, and hydrogen can cause delayed cracking during extended service, particularly in large heavy weldments.
Alloying steel along with steel making practices in recent years has led
to improvements in terms of toughness. This is especially true when
welding high strength, low alloy material and medium-high alloys. A
minute fissure or void in such a weld can or will be accepted, depending on the, the joint design, and the severity and variation of load carrying capacity.
In recent years, toughness has received much more attention within
the microstructure and the heat affected zone, along with the fusion
line. Molten alloy exhibits much more of the stronger carbide forming
elements; it has a slower dissolving rate from its higher temperatures,
and higher preheating, requirements, approximately to 4500 F. Caution
may be necessary to control grain growth by carefully preheating and
the use of temp sticks to indicate when the appropriate preheat has
been reached. The success in welding carbon and alloy steel is determined by the familiarity with the microstructural characteristics of
each particular type of steel, and avoiding the development of an
55
Chapter 5
unsuitable structure (low toughness) in or adjacent to the weld joint. In
merging the chemical composition of carbon steel with low and high
alloy, the addition of an alloying element from a filler metal can
improve a particular quality or may unexpectedly appear as a hindrance to improving toughness. Other important factors include the
microstructural changes involving the transformation of austenite, ferrite, and martensite. Grain size, cooling rate, residual elements, and
multiple alloy may also bring about synergistic effects, though little is
yet known about such synergy. Preheating and postheating can also
help attain the required level of toughness while avoiding cracking and
other metallurgical difficulties.
Emphasis must be given to the type ofthe base metal (chemical composition) to be welded, steel making practices, wall thickness, joint
design, and, in most cases, the work function of such a welded component. In addition, the welding engineer and metallurgist need to
determine the effects of microstructures of a weld based on a hydrogen
free weld, or how to maintain a hydrogen-free condition. Preheating,
interpass temperature, and post weld heat treatment are very important.
Even considering which electrode to use for a weld is important to the
welding engineer in tenus of dilution, pick-Up, and recovery. The
welding engineer must consider all these facts and write an operation
sheet which adequately reflects the above concerns. A qualified welder
can prevent defects such as porosity, slag inclusion, incomplete fusion,
excessive penetration, incomplete penetration, undercutting, root bead
cracking during welding, and distortion caused by shrinkage. Of all of
these, cracking seem to be the most unwanted defect. Defects in the
heat affected zone can be caused by the lack of adherence to proper
procedure in terms of preheating and interpass temperatures.
56
CHAPTER
Welding the Root Bead by the

Gas Tungsten Arc Welding
Process (GTAW)
In the literature of welding the Gas Tungsten Arc Welding process is usually shortened to GTAW. It is also sometimes called TIG
(Tungsten Inert Gas) welding.
When root beads of exceptional quality must be made, the
GTAW process is very frequently used. Entire welds are seldom
made by this process except in situations where unusually stringent
requirements must be met, such as in the case of space vehicles.
Usually, only the root bead is welded by the GTAW process.
However, sometimes the second pass is also made by this process
because GTAW welded root beads tcnd to be somewhat thin.
Stainless steel and high-alloy steel pipes, as well as mild steel pipes,
are welded by the GTAW process, especially for high-pressure pipe
joints that require high-quality welds.
The outstanding features of the GTAW process arc:
1. Welds of exceptional quality can be made in almost aU metals
used by industry
2. Practically no post-weld cleaning is required
3. The arc and the pool of molten metal are readily visible by the
welder
4. No filler metal is transported across the arc stream; thus there
is no spatter
5. Welding is possible in all positions
6. There is no slag which might be trapped in the weld.
When weld deposits made by the Shielded Metal-Arc (Consum~
able Electrode process) and the GTAW processes are compared, the
GTAW deposit is cleaner because there is no slag deposit, incidentally eliminating any chance of weld defects caused by slag inclusions. Moreover, the outside surface of the weld (Fig. 6-1 left), and,
on root beads, the inside surface as well, are very smooth as shown
in Fig. 6-] (right). This eliminates the necessity for deslagging,
grindingo,()f chipping after the root bead has been completed. The
57
Chapter 6
Fig. 6-1. (Left) Outside of a root bead welded by the GT AW process; (Right)
Inside surface of the same root bead.
inside smoothness of the weld significantly reduces turbulence in any

flowing substance inside of the piping system. In many cases thi'i is a
very important factor, such as in nuclear power plant piping systems.
A root bead made by the GTAW process is metallurgically sound
throughout By choosing the correct rod to deposit the filler metal, it
is possible to obtain a weld having the same chemical, metallurgical,
and physical properties as the base metal in the pipe. Defects such as
oxidation are eliminated because the blanket of inert gas covers the
weld and whipping is not used to deposit the bead.
The GTAW Process
Gas tungsten arc welding is a process whereby the base metal and
the filler metal are melted by the intense heat of an arc that is
maintained between the work and a non-melting tungsten electrode.
An inert gas, or mixture of inert gases, is used to shield the hot metal
INERT GAS SHIELD ENVELOPE
HIGHLY
CONCENTRATED
HEAT ARC
FILLER
ROD
.-"'""'~--=-'"
--(ourtcsy olthe Hoban Brothers ('0.
Fig. 6-2. Elements of the Gas Tungsten Arc Welding process <GTAW).
58
Welding the Root Bead by GTAW

from the effects of the atmosphere. Filler metal, supplied by a rod, is
used to supplement the base metal except when welding very thin
sheets of metal, for which no filler metal is used.
This process is shown in Fig. 6~2. The tungsten electrode is held
within a welding torch, which also supplies the inert gas that is
expelled from the end of the gas nozzle at a rate of 15 to 30 cubic
feet per hour to form a shield over the hot metal. Right-handed
welders hold the GTAW welding torch in the right hand while the
rod that supplies the filler metal is held in the left hand. For
left-handed welders these positions are usually reversed. In either
case, both hands are used for GTAW welding.
Equipment
The major components of the GTAW process are:

t. The welding machine
2. The shielding ga'i and the gas controls
3. The GTAW welding torch
4. The tungsten electrode.
W~lding Machine. The welding machines used for the GTAW
process are specially designed for this purpose. Those welding
machines designed for consumable arc welding, either AC or DC,
can also be used if they are equipped with a special high-frequency
atlachment; however, the best welds are obtained by using machines
designed specifically for the GTAW process.
GTAW machines are available in the form of AC/DC rectifiers Or
as DC generators driven by an electric motor or an engine. Either
straight or reversed polarity can be used with direct current. Highfrequency current is used only for starting the welding arc when
using DC current; it is always used with AC current. The welding
current is turned on by a foot or a hand control.
The current characteristic used depends upon the type of metal
to be welded. Specific recommendations arc given in Table 6-1.
I,!!-Shje{tiL,,!g Ga..".:~ To prevent oxygen and nitrogen in the air from

contaminating the weld, either argon Or helium, or a mixture of
both, is used as a shielding gas. Argon is more widely used since it is
easier to obtain and because it is a heavier gas, thus providing better
protection, or shielding, at a lower flow rate. A gas flow of 15 to 30
cubic feet per hour (CFH) is normally used.
The gas is stored in a cylinder, Fig. 6-3, and control of its flow is
similar to that used to control the flow of oxygen by a gas gage.
59
Table 61. Gas Tungsten Arc Welding (GTAW) Selection Chart

.__,_ _
~~.t,,!!.al
Welding Currenttt
Welding Procedure
A.C.H.F. medium p~"etralion

.__,
(. S P (j eep penetrullon
.
D.....
Sheets, plates, castings

ALUMINUM
I' Ie c ast'wgs.)
(sheeI s, pas.
--------~-I
CARBON STEEL
(sheels ~nd plates)
Th-ICk matenal
. on I y
T;;i~-m"l;l~;~1 only
-..
00'_'_._
Sheds" nd pluks
D.CR-P. ~hall"w p~netrat",n

------.-
D.C.S.P. d"ep P'-'Il"lplion
IThin sheets only

. Mm.__
IDo nor lIghtly jig) I ACH.F. me<)ium
Shieldin!: Gas
Argon or argon an~ hehum
argon ~nd helium for deep"r
penetration and faster t~aveJ
Argon and
or argon
and
hehum -_.
argon
helium
preferred
..
~~~~~:~d~
Argon
Argoll or argon una neflum
argon and helium for extra
deep penetration on heavy
p~~-;~~~tio~1~:;:m -~---_---
Tungsten Electrode
Pure or ZlrCODlum
ZirconJUffi - X-ray
quality welds
Th onale
- d-
Thonalcd ~.'
--
Argon or ~rgi)l1al1dneni.rm-=
D.CS.P. dcep p,'nelrallon
Thoriakd -,*
argon and helium preferred
-h""=-=--.o"====--~
tor hc"VXJ'.:.:la"l:ce-'-'ri"'a'-'-l
A.CH.F. medium !,~:nl't.'::':j(l~ I~gon
Ar-g-on-o---r,-,,---g-on-~n--rr-c-Y' j~U---n-,_---1f-"'==='--..=:.:=:...:.::='-"'--~
argon "nd helium preferred
D.CS,P. d~ep pc'net",r!on
Thoflaled -*
lor hC<Jvy matcdaJ -I;'XCt:pl
...
COPPERt
hhcels and plalcs)
Sheets 'lOti plates

V cry lhin mater i"l only
0')
Malcri"llhickcr Ihan 0.050

COPPER ALLOYS
MAUNESIUM
(shn'ts, phtes. "astillgs)
Malerwl thmncr than ().1l3lf.--.-~-+---:---.:

....-----:-._
'BerylliUm "opper illl rhi"knes""._ I A.CH,r. mniluffi p"netratlon
Shce!>, pial"" '''''ling,
fThi " shcl'lsonly
-~-I
btT Ihum c''.l.'!'!:!__ ~_ ..

Argon
A.CH.f'. mcdium penctration
I Argon
D.CRP. shall"w pem'Hatio"
t- Argon
_m
lPurcilrLtrcon lurn
Zirconium - long"r lasting

Pure or Zirconrum
Ztrconium - X-ray
~i.!Y.welds
-_
_._
j T~~~i~,~~3'_:"o-"r.-*~~~~-~
-----mO:EL
MONEL
INCONEL
STAINLESS STEEL
All Ihick T)cs>es

,Sheets. plalc,. ,aslinll:s
(sheets, pbtes, Ci1s,tingsJ
Thin
Thorj~ted
D.CS.P. decp PCllcI,ati...n
Argon
D.CS.P. de,'p p"nl'llllli"n
Atgoo--{ir"arg.", andhdium
argon and helium for extra
s;';~;;~;;;-i;"'----'--- ~:C.II.F m~e~.d~iu~I;;;;;~~;i-;;;;t:~;,~~retra~~n-lh'e~-=--=='==.-:-''=-
-'
Thorialcd - '
TITAN lU M
Lt....ne'scs ...=~_:=':~=-~
IT"yP:U;:cl' llC.".etr"iToiim. ''i'l\!;:il'.:..
----I-;:rlJ:::r!J-W~tC'eT_..:::;r-""-'-" .......~~'-"'<.irintl end 10 pOInt m "ca' pomt.
"Usc with balled end. Slowly increasc welding>:urrcnt ulllli h~1I forms.
tu,c hr~ljnj!, flux on ,." or thicker.
tt A.CII-F. Alternating current -- high frequency;
[l( '. RY.. - Direct cUrrenl -, reveISed polarity:
fH'sr.
Direct current- straight polarity.
()
::::T
-...
l
"0
CD
0')

A.C. OR D.C. WELDER
!
FILLER
ROD
TORCH
GAS IN-
\\
WATER IN WATER OUT

)
.c=~'r- _..
I
WORK
GAS OUT _
FOOT~llL--~l~~:f~~~~~~~
'-----------,L-'---I----------'-;-'
CURRENT
WORK CABLE
COil rt en
"l'iI,,,
ffohtlrl 8ml hers Co.
Fig. 6-3. Major equipment components for Gf AW welding.
--c:():J- u.-~-=r=B
NdzzLE
COlET NUT
COLLET
I
ELECTRODE
-~
CAP
HANDLECOl1rrt'sy 'J Iht' Hobtlrl Brotlwrs Co.

Fig. 6-4. A. The OTAW welding torch. B. Components of the GT A W welding
torch.
Beyond this gage, the gas to the torch and the weld is controlled
either by a switch mounted on the torch or by a foot pedal. When the
gas leaves the tank it is fed through an electrical control valve that is
actuated by the switch which allows the gas to flow only when the
welding current is turned on. The gas can be made to flow continuously by means of manual control but this can be very costly.
61
Chapter 6
GT~.~.Torch:... The GTAW torch, Fig. 6-4A, holds the tungsten
electrode and directs the welding current to the arc. Torches can be
water-cooled or air-cooled, depending on the welding current
amperage. For root bead welding, which is normally deposited using
a current range of 75 to 200 amps, water-cooling is not required.
The parts of a torch are shown in Fig. 6-48. A nozzle surrounds
the electrode, which is held in place by a coUet chuck. Different
collets, ranging in size from .020 to .250 inch, are available to
accommodate the same range of electrode diameters. To insert an
electrode, the cap is removed and the correct-size collet is inserted in
the torch. Insert the electrode in the torch and push it about Vz inch
beyond the end of the nozzle, using the wrench furnished for this
purpose to adjust the collet. Attach the cap to the torch and tighten
it lightly. Then adjust the electrode so that it extends beyond the
nozzle the correct distance, which is usually about I liz to 2 times the
electrode diameter, and finger tighten the cap. The torch is then
ready to be used.
Two different types of nozzles are available. One type is made of
ceramic, which is not transparent. Another is of glass, which offers
better visibility of the pool of molten metal when welding.
El~clrodes .. The electrodes used with the GTAW process are made
from tungsten alloys. They have a very high melting point (6900F)
and are practically non-consumable. When properly positioned, the
electrode is located over the puddle and the intense heat of the arc
keeps the puddle liquid. The electrode, which must be kept clean at
all times, must never touch the molten metal to avoid the possibility
of contaminating its tip with metal from the puddle. Should it
become contaminated, the electrode must be removed from the
torch and the contaminant removed either by grinding or by break
ing off the end of the electrode to remove the contaminated portion.
There are three types of electrodes:

I. Pure tungsten
2. I or 2 percent thoriated tungsten
3. Zirconiated tungsten.
Recommendations for the applications of the types of electrodes are
provided in Table 6-1. Thoriated tungsten electrodes are used for
most pipe welding applications, including mild steel pipe.
The shape of the electrode tip has a marked influence on the
contour, penetration, and width of the face of the weld deposit. It is
62
especially important to shape the electrode tip correctly, similar to a

sharpened pencil, when it is to be used to weld a root bead on a pipe
joint.
The working end, or tip, of the electrode is shaped by grinding it
with a very fine grinding wheel that should be used only for this
purpose. A fine-grained wheel should be used in order to obtam a
very smooth surface finish on the tip of the electrode, which is
helpful in maintaining a more stabilized arc. For this purpose, a 60
grain, 0 to M grade silicon carbide grinding wheel should be used,
such as a C-60-M-V wheel. If the electrode becomes contaminated
with metal, it should be reground immediately.
Exact specifications for the shape of the electrode tip are given in
Fig. 6-5. The included angle of the point should be about 22 to 23
degrees, or the length should be ground back to a distance equal to
about 21jz electrode diameters. It is important to blunt the tip slightly
by grinding it flat at the end for a distance slightly less than V64 inch
from the point.
Shielding the Weld Metal. In all welding processes, the very hot
metal in the region of the puddle is protected by some kind of
shielding. A gas formed by heating the coating of the electrode
protects the top of the weld in the shielded metal-arc process. In
addition, the molten metal is protected by a coating of liquid slag
that rises to the surface. On open butt joints, this coating protects the
top of the weld and some of the slag flows through the liquid metal
to the bottom of the joint where it also protects the exposed surface.
TUNGSTEN
ELECTRODE
Fig. 65. Grinding the lhoriated tungslen electrode ltp for rOOI bead welding
Point angle ~pecification is shown as 22 to 23
0
63
Chapter 6
Actually, there is some uncertainty about the amount of protection
provided to the exposed surface at the bottom of the joint However,
when welding mild steel pipe with the Consumable Electrode
Method, this presents no real problem.
Mild steel pipe can also be welded by the GTAW process without
special precautions. A sufficient quantity of the inert gas reaches the
bottom of the joint to provide adequate protection and the top of the
weld is covered with a heavy blanket of inert gas. Highly alloyed
steel pipe, however, when welded with the GTAW process, will
require extra protection at the bottom of the joint. This is done by
filling the inside of the pipes, in the region of the pipe joint, with an
inert gas.
Several methods are used for containing the inert gas in the pipes;
two of these are shown in Fig. 6-6. Two "pistons" having rubber
seals are inserted on each side of the pipe joint (see Fig. 6-6A). Tlle
pipe joint itself is taped shut to prevent the inert gas, which is blown
into the pipe at a very low pressure, from escaping. A small portion
of the joint is left open to allow the air to escape. When welding, the
sealing tape around the pipe is removed in sections just ahead of the
weld, and additional inert gas is blown inside of the pipe to make up
for the gas lost through this opening. This method can be used when
welding short lengths of pipe; however, it is somewhat awkward to
use on longer pipe lengths and on larger-diameter pipes.
I-<--- GAS
GAS
Fig. 6-6. Methods of containing the inert gas shield inside a pipe. A. Two pistons
used for short, smalJer-diameter pipes. B. Plastic gas bags used in conjunction
with longer pipe lengths.
In field conditions, where the pipe diameters and lengths are

frequently large, the joint can be sealed by inflating two plastic
balloons in the pipe, as shown in Fig. 6-6B. They must be positioned
far enough from the weld joint so that the heat from the weld will
not burst them. The location of the balloons must also be marked on
64

the outside of the pipe, for when the root bead is finished, just
sufficient heat is applied by an oxyacetylene torch at these markings
to burst the balloons. The joint itself is sealed with tape, as before,
and the inert gas is blown into the pipe through a rubber hose or a
length of small-diameter pipe or tubing. Another method of sealing
the inside of the pipe is to tape a wall of polystyrene to the inside of
the pipe. The advantage of these two sealing methods is that the
l1aterials used to make the seal can be blown out by compressed air
or a stream of water after the pipe joint is finished.
Preparation ofthe Weld Joint. The procedure for preparing the weld
joint is the same as for preparing the joint when welding' by the
Shielded Metal-Arc process. The principal difference is in the
dimensions of the joint. The width of the root face is reduced to V16
inch and the root opening is narrowed to lft6 to 3/32 inch, as shown in
Fig. 6-7.
The pipe joint should be carefully fitted together to obtain an
accurate alignment and the correct width of root opening. If alloy
steel pipes are being welded, an inert gas should be blown into the
inside of the pipe as described in the previous section. Highly alloyed
~750y
'/l6"to 3/32'~
I/l6"
Fig. 6-7. Weld joint specification for GTAW welding of the rool bead.
steel pipes should be preheated at tlie pipe joint from 200 to 300F
prior to welding the tack welds and the remainder of the root bead.
When all of the preparations have been made, four evenly spaced
tack welds are made around the pipe joints. Since these welds will
become a part of the root bead, they must be made with the same
care and craftsmanship as the remainder of the root bead.
Procedure for Welding the Root Bead. Even before the tack welds
are made, the GTAW welding torch must be adjusted and the
welding machine must be set to provide the correct current amperage and gas flow.
65
Chapter 6
To weld 8 in., Schedule 60, mild steel pipe nipples, a :1hz-inch
diameter thoriated tungsten electrode and a lis-inch diameter mild
steel filler rod are used. The electrode should be ground to a point
with a fine grinding wheel. In order to hold the electrode, a Ylz-inch
collet is placed in the welding torch, after which the electrode can be
inserted.
For welding a root bead, the electrode must be positioned in the
torch so that it will extend the correct distance beyond the end of the
nozzle. This is done by placing the torch in an upright position with
the nozzle resting on the bevel of the weld joint, as shown in Fig.
6-8. While this is being done, the welding current should, of course,
be shut off. The electrode is then adjusted so that it is positioned
II
II
I I
GLASS OR
CERAMIC
1 I
II
I I
I
II
I I
Fig. 6-8. Position of GTAW torch and tungsten electrode when adjusting the
electrode for root bead welding.
with the end almost flush with the bottom of the joint or the inside
surfaces of the pipe.
Reference to Table 6-1 shows that the welding current characteristic should be direct current, straight polarity (DCSP), when welding mild carbon steel pipe. Table 6-2 provides the range of the
welding current amperage to be used. Because of the many variables
encountered. it is not possible to give a more precise recommendation. In this case, using a 3hz-inch electrode and a direct current,
straight polarity current characteristic, the welding current flow
should be 150 to 225 amps. The gas supply should be adjusted to
provide a flow rate of about 20 cubic feet per hour.
When the pipe is in the 50 position, the tack welds should be
made in the 8:30, 4:30, 1:30, and 11:30 positions. The regular root
bead is started in the 6 o'clock position and the weld is made by
uphill welding around one side of the pipe to the 12 o'clock position.
66

Then the other half of the pipe is welded, starting again at the 6
o'clock position and welding uphill.
There is no difference in the welding procedure used, whether
welding in the 5G or the 2G positions. For this reason, the procedure that is recommended in the following paragraphs can be used
in either position. In both cases (5G or 2G positions), the root bead
should not be started against a tack weld. It is somewhat difficult to
start a weld in the 6 o'clock position because of the stance that the
welder must take; however, the procedure that is used to start the
weld will be the same, regardless of the position of the weld.
The arc should not be struck in a manner similar to the procedure
used for Shielded Metal-Arc Welding with consumable electrodes as
this procedure would seriously damage the tungsten electrode.
To start the weld, the current and the inert gas supply to the
welding torch must be shut off. The GT A W welding torch is held in
the right hand (if left-handed, in the left hand) and the filler rod is
Table 6-2. Current Range Chart for GTAW Welding
Electrode
Diameter
(Inches)
.020
.040
Pure lungsten -- green tip
**Zirconium - brown tip

tThoriateJ tungsten -- 2{1( red tip. I {Jf yellow tip
67
Chapter 6
held in the other hand. The torch is brought into posItion for
welding by carefully placing the nozzle against the two beveled
surfaces of the weld joint, which acts like a V-block to locate the
electrode in the weld groove. In order to position the electrode in the
center of the joint opening, the side angle of the electrode and the
welding torch must be at zero degrees; Le., the electrode must be
perpendicular to the pipes in this direction.
However, the electrode angle must be steep, as shown in Fig.
69A and B. As shown, the welding torch to start with, should be
positioned to hold the electrode at an angle of approximately 55
Fig. 6-9. A. Electrode angle and filler rod angle in preparation for root bead
welding. The welding current is off when the electrode is in this position. B.
Welder ready to start a bead.
Azo._~_
Fig. 6-10. Correct an es of the tungsten electrode and the electrode when
welding.
68

degrees. At the same time, the filler rod should be positioned close to
the weld in preparation for insertion in the weld after the arc is
established and the molten pool of metal has formed. It should not
be positioned in the ,arc area when the arc is first established;
however. the welder should position the filler rod so that he can dip
it into the molten puddle without any disturbing body movements
that might affect the smooth manipulation of the electrode. When
welding, the filler rod should be held at an angle that is between 45
and 70 degrees, as shown in Fig. 6-9A and Fig. 6-10.
With the welding torch and the filler rod held in the starting
positions, the welder must then close his face mask and switch on the
welding current using the foot switch or the switch on the welding
torch. Slowly and carefully he will straighten the welding torch
thereby reducing the electrode angle and bringing the tip of the
electrode into the groove. In doing this the welding torch is pivoted
around the end of the nozzle, which is held in position on the bevels
of the weld joint.
While the welding torch is pivoted, the arc will be formed and
become established. As shown in Fig. 6-10, the welding torch is
pivoted until the electrode angle has reached 20 degrees, the correct
angle for welding. The filler rod is held at an angle of 45 to 70
degrees. As the welding current is started, it also causes the flow of
the inert gas to start to form a protective shield over the weld area.
After the arc has been established, the welding torch is held in
place until a puddle of molten metal has appeared between the two
edges of the pipe. After the puddle is established, the filler rod is
dipped into the molten metaL The welder then begins to oscillate the
welding torch from side to side, as shown in Fig. 6-11, allowing the
arc to travel slightly beyond the edges of the pipe.
(\ )
~
\
'"
I
~t~
Fig. 6-11. Electrode oscillating procedure for welding the rool bead with a
GTAW torch.
69
Chapter 6
Fig. 6-12. Welding a root bead. (Left) Welder in welding posillon: (Right)
Welder's view.
Welding a root bead is shown in Fig. 6-12. When the puddle has
been built up to the required size by the addition of filler metal, the
welding torch is then moved on by slowly and steadily gliding it over
the bevel while at the same time the torch is oscillated from side to
side. Filler is added continuously to the molten puddle from the end
of the rod which is held in the puddle.
While welding, the welder must watch the puddle at all times to
see that it is built up to a uniform height and forms a smooth bead
with only traces of a ripple. As with most welding processes when
welding pipe in the 5G position, there is some tendency for the
puddle to drip, causing excessive penetration. When this occurs,
corrective action should be taken by slightly increasing the speed of
travel and the speed of the oscillating movement. If possible, more
filler metal than usual should be added to the puddle which tends to
chill the puddle and to reduce its size.
The welder must be careful at all times to prevent the electrode
from making contact with the sides of the joint and the molten
puddle. This would not only contaminate the end of the electrode
but cause the arc to become unsteady. Should this occur, the weld
should be stopped and the tip of the electrode reground before
restarting the weld.
One problem that is occasionally encountered is that the nozzle
will tend to stick or grab as it is lightly dragged over the bevel during
welding. The end of the nozzle gets very hot, a factor which
promotes its sticking'1o the heated bevel. As shown in Fig. 6-13A,
the edge of the nozzle is square, and sticking can be avoided by
grinding a radius on the edge of the nozzle as shown in Fig. 6-13B.
Stop and Restart. The weld should never be stopped unnecessarily.

However, there are occasions when it must be stopped, such as whc:n
70

welding a tack weld or when the 12 o'clock position is reached. The
weld must also be stopped and restarted when making a tie-in with a
tack weld which will be discussed in the next section.
To stop the weld, the welding current is simply switched off. If the
filler rod is to be replaced or if the weld bead is finished, the filler rod
should be withdrawn beforehand. However, if the bead is to be
continued, the filler rod is left in the puddle to solidify against the
end of the bead.
The procedure for restarting the weld against a layer of weld
metal is similar to the starting procedure previously described. With
the welding current off, the welding torch is positioned on the bevel
~
GRIND RAOlUS
"\
I
/
1
I
I
A
Fig. 6-13. A. Standard nozzle with square edge: R. Nozzle edge rounded to
facilitate dragging the CiT AW torch when welding a roO[ bead.
Fig. 6-14. A perfect tie-in on a root bead, deposited by the GTAW welding
process.
71
Chapter 6
adjacent to the weld bead. This time, when positioning the welding
torch, it should be held at the angle used for welding; Le., the
electrode angle should be about 20 degrees. The position of the
electrode is carefully adjusted so that it does not quite touch the weld
bead. The welding torch is then rotated about the nozzle until the
electrode angle is about 55 degrees, as was shown in Fig. 6-9. After
the fiUer rod is in position, the welding current is switched on and the
welding torch is slowly rotated to the 20-degree position in preparation for welding, with care being taken not to allow the electrode to
touch either the bead or the sides of the joint. When the arc has been
stabilized and a sufficient quantity of molten metal has appeared, the
filler rod is placed in the puddle and the electrode is oscillated, as
before, and the weld is continued.
Making a Tie-in. When the weld approaches another bead, the

normal GTAW welding procedure is continued until the weld is
approximately %6 inch from the bead. At this point the filler rod is
withdrawn, but the welding torch is slowly moved on; it is tilted as
necessary to prevent the electrode from coming in contact with the
weld metal. The welder must watch the filling up of the gap between
the beads. When a smooth blend between the beads has been
obtained the current is switched off. A perfect tie-in, such as that
shown in Fig. 6-14, will result if this procedure is followed with care
and attention.
Second Pass. When a second pass with the GTAW torch must be
made, the procedure is the same as that used in welding the root
bead. However, it is necessary to readjust the position of the elec~
trode in the GTAW welding torch. The length that the electrode
protrudes from the end of the nozzle must be shortened so that it will
not come in contact with the root bead or the molten pool of metal.
The end of the electrode is ground to a blunted point, as before.
While welding, the electrode and the filler rod are held at the same
angles as before. The welding torch is advanced by slowly and
uniformly gliding and oscillating the welding torch with the end of
the nozzle resting in the V-shaped groove of the weld joint. By
welding slowly and carefully, the second layer will be deposited
uniformly and with only a trace of a ripple.
Root Bead-Pulse Current-Gas Tungsten Arc Welding

Gas tungsten arc welding has been used to great advantage in all
kinds of pipe welding applications. Besides affording the welder
excellent visibility due to the lack of smoke and spatter, there is no
porosity due to slag entrapment, and the deposits are essentially hydro-
72

gen free. The GTAW process is particularly useful, due to more precise heat control, in welding thin wall pipe and in many cases heavy
wall pipe may have it's first root pass done with GTAW while subsequent fill and cap passes are completed with conventional stick welding. In such cases where root penetration must be as smooth as possible, a consumable insert is placed between the mating pipe edges and
fused into the bottom of the root opening.
However as pipe welding is most often done in various "out of position" situations, the GTAW process needed even more precise heat
control to avoid weld pool sagging and bum through. The ]960's saw
the introduction of "pulsed current" welding control in which both
high and low values of current could be preset on the power supply.
During the "high" portion of the pulse cycle, actual fusion would take
place. Then as the torch moves away, the current automatically
switches to a lower value allowing the puddle to partially solidify
before the next high current pulse begins. The length of time for each
pulse is also adjustable making possible optimum welding conditions
over a wide variety of pipe welding applications.
The pulse current mode is shown in Fig 6-15.
Pulse control offers a number of advantages over the inconsistencies
that are experienced during the time the root bead is deposited. These
advantages of using pulse current-gas tungsten arc welding include:
(a)
control over puddle size and fluidity
(b)
ability to increase and maintain adequate penetration
ability to maintain unifonnity of weld deposit
(c)
(d)
ability to preset the pulse mode to any condition, as needed by
the welder.
(e)
ability to position the torch in its rightful position before the
high pulse is activated
Comparing shielded metal arc welding with pulse arc welding for
depositing the root bead on thinwall pipe, the latter shows absolute
Pulse in Seconds: High

Hi h
Pulse in Seconds: Low
Figure 6-15
73
Chapter 6
control of the puddle size, prevention of excessive penetration, com~
plete protection from oxidation, and a clean inner surface of the root
bead. Therefore, pulse arc welding is often preferable for welding pipe
of both medium alloy and high altoy steel. Pulse arc welding is a use
ful innovation for thinwall pipe, including when thinwall pipe needs
two passes to be completed.
Before using pulse arc welding for heavywall pipe, the welder must
be fully acquainted with the appropriate variables. Heavywall pipe
offers an excellent heatsink. The size of the tungsten electrode must be
considered. If a slight oscillation will be used while depositing the root
bead, it is important to know where the tungsten tip will be located and
when the current mode changes to low. The tip should be placed so
that welding occurs under a blanket of inert gas; the hot weld will
then be protected from oxidation. The position of the tungsten tip
should be one sixteenth of an inch on the weld, where there is a chance
of only a slight fissure or tiny void. It is from that position that the
welder should begin when the high pulse is enacted.
The filler metal should be placed between the root opening where the
tip of the rod is in contact with the apex of the groove. The rod must
not be allowed too close to the starting point. When the high pulse is
initiated, the rod should be inserted, making contact where the rod
touches the root bead.
Because the thickness of the pipe makes an excellent heatsink, the
high pulse mode duration should be four times that of the low mode.
When the low mode is inactive, the tungsten-tip should be raised one
sixteenth of an inch. It should then be advanced forward when the high
mode is enacted, with a slight oscillation to remove cumulative grains
that may have fonn during the cooling pattern, and to assure that the
tie-in to the root is properly fused. The slight backstep becomes necessary because the heat-sink, which is so effective, can cause a shrinkage crack. This depends on the frequency of changes between the
modes. The high mode should be enacted while the cooling weld is
still in a semi-solid state.
The power source design for pulse current welding calls for adjusting both the high mode current and low current pulse based on time
duration. Adjusting both depends on wall thickness, preheating, fix or
roll position, and the rootface thickness.
The internal surface of the pipe edges to be welded should be clean
of oxide (rust) and other impurities.
74
Table 6
Prequalified Base Metal-Filler Metal Combinations for Matching StrengthS (see 3,3)
Sleel Speci Iication Requirement.
G
u
p
Sleel Specification I.i

ASTMA364
ASTM A53
ASTMAI06
ASTMAI3l
ASTMAI39
ASTMA38J
ASTMA500
ASTMA501
ASTMA516
---.J
0'1
ASTMA524
I
ASTM A529
ASTMAS70
ASTMA573
ASTMA709
APl5L
ABS
Grade B
Grade B
Grades A, B, CS, D, OS, E
Grade B
Grade Y35
Grade A
Grade B
Grade 55
Grade 60
Grade 1
Grade n
Grade 30
Grade 33
Grade 36
Grode 40
Grade 45
Grade 50
Grnde 65
Grade 58
Grade 36'
Grade B
Grade X42
Grades A, B. 0, CS, DS
Grade ISl
ksi
MPa
36
35
35
34
35
35
33
42
36
30
32
35
30
42
30
33
36
40
45
50
35
32
250
240
240
235
241
240
228
290
250
205
220
240
205
290
205
230
250
275
310
345
240
220
250
240
290
36
35
42
Minimum Yield
PoinllStrength
Tensile
Range
Mmimum Yield
PointlStrenglh
r
()
!'tHer Metal Requirement,
ksi
MPa
58-80 400-550
60 min 415 min
60 min 415 min
58-71 400-490
60 min 414min
60 min 415min
45 min 3JOmin
58 min 400 min
58 min 400 min
55-75 38Cl--515
60-80 415-550
60-85 415-586
55-80 38Cl--550
60-85 415-585
49 min 340 min
52 min 360 min
"53 min 365 min
55 min 380 min
60 min 415 min
u.s min 450 min
65-77 45Cl--530
58-71 400-490
5&--80 400-550
60
415
60
415
58-71 400-490
58-71 400-490
(continued)
Electrode Speci ficallon)' 6
SMAW
AWS A5.! or AS.5'
E60XX
E7UXX
E70XX-X
ksi
MPa
Tensile Strength
Range
ksi
MFa
~
50
345
62 mju
42~
60
415
57
390
12 min
70 min
495
430
a:
S'
...
(Q
:::.J
(1)
SAW
AWS A5.17 or A5.23'
F6XX-EXXX
F7XX-EXXX or
:D
48
58
330
400
60-80 415-550
7Cl--95 485-<i60
...
OJ
(I)
F7XXEXXXX
OMAW
AWS A5.I8
liR70SX
o
o
~
~
60
415
12 min
495
G)
~
FCAW
AWS A5.20
E6XT-X
E7XT-X
(Except -2. -3. -10. -GS)
AWS AS.29 r
50
60
345
415
62 min
nmin
E7XTX-XX
58
400
70-90 49Cl--620
425
495
Table 6 continued
Steel Specification Requirements
r
0
u
p
Steel SpceillC3tion 1,2

ASTMAI31
ASTMA441
ASTMA516
ASTMA537
ASTMA572
ASTMA572
ASTMA588~
ASTMA595
""-J
0>
ASTMA6Q@
ASTMAOO7
Il
ASTMA6J8
ASTM A633
ASTMA709
ASTM A710
ASTMA808
API2H6
API5L
ASS
Grades AH32, DH32, EH32

Grades AH36. DH36, EH36
Grade 65
Grade 70
Class 1
Grade 42
Grade 50
(4 in. and under)
Grade A
Grades Band C
Grade 45
Grade 50
Grade 55
Grades Ib, II, III
Grade A
OradesC,D
(21/2 in. and under)
Grade 50
Grade SOW
Grade A, Class 2 >2 in.
(2112 In. and Ilnder)
Grade 42
Grade 50
Grade X52
Grades AH32. DH32, EH32
Grades AH36, DH36, EH36$
Filler Metal Requirements
Minimum Yield
Point/Strength
ksi
MPa
315
51
350
40-50 275-345
35
240
260
38
45-50 310-345
42
290
345
50
345
50
55
380
60
415
45-50 310-340
45
310
345
50
55
380
46-50 315-345
42
290
345
SO
46
50
50
55
42
42
SO
52
45.5
51
345
345
380
290
290
345
360
315
350
Minimum Yield
Point/Strength
Tensile
Range
ksi
MPa
68-85
71-90
60-70
65-85
70-90
65-90
470-585
490-620
415-485
450-585
485--620
450-620
415 min
450 min
485 min
450 min
480 min
450 min
410 min
450 min
480 min
450 min
430-570
485--620
6Omio
65 min
70 min
65 min
70 min
65 min
60 min
65 min
70 min
65 min
63-83
70-90
65 min
70 min
65 min
60 min
62-80
70 min
450 min
485 min
450 min
415 min
430-550
485 min
66-72 455-495
71-90 490-620
71-90 490-620
(continued)
Electrode Specificalion3. s
Tensile Strength
Range
ksi
MPa
bi
MPa
60
415
72 mm
495
57
390
70 min
480
SMAW
AWS A5.l or A5,5'
E7015,E7016
E7018, 7028
E70IS-X, E7016X
E7018X
SAW
AWS A5.17 or A5.23
'
F7XXEXXX or P7XXEXXXX
()
:J'"
58
400
70-95
485-660
s
'"C
(i)
~
GMAW
AWS A5.18
ER70S-X
FCAW
AWSA5.20
E7XT-X
(E~cepl 2, 3. 10. -OS)
AWS A5.291
E7XTXX
0>
60
415
72 min
495
60
415
72 min
495
58
400
7(}...90
490-620
Table 6 continued
Filler Melal Requiremcnts
Sleel Specification Requirements
Minimum Yield
PointlStrenglh
r
(>
u
p
Sleel Specification!' 1
k$j
MPa
Minimum Yield
Point/Strength
Tensile
Range
hi
MPa
Electrode Specification}'"
SMAW AWS A551
B8015-X. ESOI6-X
ksi
67
E8018-X
ASTM A572
III
ASTMAS37
ASTMA633
ASTMA710
ASTMA710
::j
Grade 60
Grade 65
Class 2j
Grade B'
Grade A, Class 2 '52 in.
Grade A, Class 3 > 2 in.
60
65
46--60
55-<iQ
60-65
60-65
415
450
31S-4IS
380-415
415-450
41S-450
75min
80 min
80-100
75-100
72 min
70 min
515min
550 min
550-690
515-690
495 min
48S min
SAW
AWSA5.23 7
F8XX-EXX-XX
GMAW
AWS A5.28 7
ER80S-X
FCAW AWS A5.29 7
E8XTXX
Tcnsile Strcngth
Range
MPa
ksi
MP.
460
SOmm
550
::J
(0
68
470
80-100
55~90
:r-
CD
68
470
80 min
68
470
8(HOO 550-690
550
Notes:
l. In joinls involving b_ mel4l. of differen, group'. eith.r of the following fiUer metals moy be usoo: (I) that whkh malches Ihe higher strength base me!~I, m (2) thal which matches the lowe' s!length
base lmtallJJd producel .10Whydrogen deposil. Preheating shaJl be in confo""""ce wjlh the require"",nls applicable 10 the highcr .lre_gth group.
2. Match API standard 28 (fabricaled lUbes) .eco'ding 10 .1Ce1 used.
J. When welds are (0 be Sl"".relieved, the depo.ilcd weld ",.1.1 ,Ihall not ..teed 0.05 percenl vanadium.
4. Only low-hydrogen electrod.s shall be used when welding A36 ,,, A709 Grade 36 steel more Ihan 1 in. (25.4 rom) Ihick for cyclically loade<l 'Inlctufe>.
Spcc~aJ wehling materials- and WPS (e.g., E80XXX low~a1f.o'Y eleCfrodes.) may be required tQ match (he nOlch toughness of base metal (for applications in'lolving lmpa~t loading or low LcmpcraWI"l:}, or
for atmospheric corrosion and weathering chiiU"acleristics (see 3.7,3)
6 The designalion afER1OSIB h.. been redas,if,oo as ER8OSD2 11\ A5,18 79.l'r"'-Jualifiect WPS, preparl prior 10 1991 and ,pecifying AWS AS.I g, ER70SIB, may now"" AWS AS.2R79 ER80SD~
when welding stccls in Groups I ""d II.
7. Filler metals of alloy group 1l3, B3l, B4, B4L, lJS. BSL. !l6, B6L, B7. B7L, B8. B8L. Of B9 in ANSUAWS AS.5, AS.23. AS.2B, or A529 Ofe not prequ'lified for use in the as welded COlldltion.
8. Soo Tables 2.3 and 2.5 far al1owal>le slress require men Is fo' m.atching filler ",e",1,
:5.
:E
CD
9:
,.
JJ
CD
(!)
cr
'<
G)
~
~
Chapter 6
Structural Steel Plates
Sueng:h
Toughl1~ss
Gro"p
CIa"
C
----!11
fI
II
Yield S"en8.h
S~jtic-~~iol1 &:.
AsTM A36
ASTM A 131
ASTM AUI
ASThl A5?3
ASTM A709
ASTM A 131
ASTM Al~2
ASTM A572
ASThl A5n
ASTM A588
ASTM A709
ASTM A!3l
AsTM A l) I
ASTM A808
ASTM A516
API Spec 2H
Gr3dc
(to 1 in. Ihick)

Grade A (to II! in Ihick)
Orades 8. D
Gr.de 65
Gr.de 3612
Orade, CS. E
(10 III in. (hick)
Gf1lde 42 (10 2 In. ,hick)
Or.de 50 (to 1/2 in Ihitk)'
(~in. and "ndor)
Grode<
Orode AH32
Grode AHJ6
(slrength varics wilh thickne,,}
Grade 65
Grade 42
Grade 50 (102112 in thick)
son. son
(D\'er 2]f2 .in_ ihti'y.)
API Spec 1W Grode 42
It
(101 in.lhick)
("'er I in thick)
Grode 50 (10 1 in. 'hick)
(o\'er 1 in. thick)
Grade SOT (10 I in. Ihick)
(ovcr I in.1hick)
API Spec 2Y Grade 41 110 I in. Ihick)
(o\'~r 1 in. Ihick)
Grade 50 (10 t in. thick}
(tr\,:er 1 in. thick)
III
nr
GrJde SOT (10 1 in. Ihkj()

(0'0' I in. thiCk)
ASTM A131 GrJdes DH)2. EHJ2
Grodes DH36. EHJ6
ASThl A537 Closs I ('0 2112 in. Ihick)
ASTM A633 Grade A
Grades C, D
ASTM A678 GrodoA
ASTM A633 Grade E
(10 2-112 in. thick)
ASTM A537 Class H
ASTM A678 Grode B
API Spec 1W Gradc 60 (10 I in. In iokl
(o,er I in. ,hick)
API Spec 1Y Grade 60 (to I in. Ihkk)
(ove, I in. ,jilek)
ASTM A7 to Grade A CI." J
(quenched .and pncipit.ltion heat I~C.<HC'd)
~n. ~o4
in.
over 4 in.
IIjTa
:::!
IV
ASTM ,0.514
AsTM ASI7
ASTM A514
ASTM AS 17
3~
34
35
36
34
50
41
50
50
50
45.5
51
42-50.
35
42
50
47
42-67
41-62
50-75
50-70
50-80
50-75
41-67
41-62
50-75
50-70
50-80
50-75
45.5
51
50
42
50
50
60
60
60
60-90
60-85
60-90
60-85
75
6$
60
90
90
100
thru 2 in.
2:
'si
36
(ave, 2-111 in Ihick)

(over 2-112 in> Ihick)
(102112 in. thick)
(to 21/2 in. thic~)
100
in. Thi-ck for Type i or "2 Kmd. Fine G.rain Practicr-
Not; S~ Ibr iii Rl: rcreliced Spedfica(~'\)n'5 for !ulllitl-e-s of the above.
78
MP.
250
~35
135
NO
250
235
3-15
290
345
345
3-15
315
350
290~345
240
290
.::~s
325
290-462
190-427
345-517
345 -483
345-521
345-517
290-461
290-462
345-517
345-483
345-572
345-517
315
350
345
290
345
345
415
415
415
414-611
414-586
414-621
414-S86
515
450
415
620
610
690
690
TensH~
KS.
58-SO
58-71
58-71
65-77
58-80
58-71
70
60
65
70 min
65
68-85
71-90
60-65
65-85
62-80
70-90
70-90
62
61
SrrengEh
MPo
400-550
440-490
400-490
450-550
400-550
400-490
480
415
450
485 min
450
470-585
490-610
415-450
450-585
430-550
4SJ-610
483-610
..;27
427
65
.ws
65
70
70
62
62
65
65
70
70
68-85
71-90
70-90
63-83
70-90
70-90
-:..IS
SO-lOll
80-100
80-100
75
75
75
75
85
75
70
110-130
110-130
1I0-1)0
110-130
483
483
427
~27
448
"'-18
483
483
470-585
490-620
485-620
43S-570
485-610
485 - 610
550-690
550-690
550-690
517
517
517
517
58S
SIS
485
760-890
760~896
760 895
760-895
Structural Steel Pipe and Tubular Shapes
-'-
---'-'
Yield Strength
Gronp
Spe.::iflcaliol1 & Grade
Cla~s
-,-------_._-API Spec 5L
ASTMA53
ASTM 139
ASTM A500
C
Grade B'
Grade B
Grade B
Grade A (round)
(shaped)
ASTM A500 Grade B (round}
(shaped
ASTM A501 (round and shaped)
API Spec 5L Grade X42 (2% max. cold expansion)
ASTM AI06 Grade B

ASTM A524 Grade 1
Grnde 11
(o(}fmalized)
(through 3/8 in. w.L)
(over 318 in. w.l.)
Tensile Strength
hi
MPa
ksi
MPa
35
35
35
33
39
42
46
36
42
240
60
240
60
60
45
45
58
58
58
415
415
415
240
230
270
290
320
250
290
310
310
400
400
400
415
60
60
35
35
30
240
240
205
55-80
415
415
380-550
60
ASTM A333 Grade 6

ASTM A334 Grade 6
35
35
240
240
60
60
415
II
API Spec 5L Grade X42 (2% male cold expansion)

ASTMA61&
52
50
360
345
66
70
455
485
II
API Spec 5L Grade X 52 wi\h SRS. SR6. or SR8
52
360
66
455
ASTM A595 Grade A (tapered shapes)

ASTM A595 Grades Band C (tapered shapes)
55
380
410
65
70
4S0
480
III
60
415
Seamless or WJtb longttudinal seam wdds
Notes;
I. Su list of Referenced Spe<;ifications for full fjfles of lhe .hove.
.
2. SI",elu",1 pipe m.y also be f.brieafed in .<cord.nce wi,h API Spec 2B, ASTM A119+, ASTM A252+. or ASTM A671 u.mg
grade!i of ~ plate listed in Exhibit t except that hydrosl3tic Iesting may be omined.
+ witlllongitudinal 'Weld~ and ciT<'umferential butt welds.
Table C2.6
Structural Steel Shapes (see C2.42.2)
Yield Strenglh
Group
II
II
Class
Speclfication & Grad,e
A36 (to 2 in. thick)

ADI Grade A (to 1/2 in. thick)
ksi
MPa
Tensile Streng1h
ksi
MP.
36
250
34
235
58-80
58-30
400-550
A709 Grade 36T2
36
250
58-80
400-550
AS72 Grade 42 (to 2 in. thick)
42
290
60
A572 Grade 50 (10 1/2 in. thick)

A53& (10 2 in. thick)
50
50
345
65
415
480
345
70
4SS
A700 Grades 50Tl, 50T3

ADl Grade AH32
AUl GradeAH36
50
46
51
345
320
360
65
450
470-S85
4'lO-620
68-&5
71-90
400-550
"To 2 in. Thick for Type I or 2 Killed. Fin' Gillin P,-oclice

Note; This IIIbl, is pan of ~ cornrnenUlry on '""ghnt'S eo>nsid'rations for 'ubular st"''''nre' (or composites of lubul.... and olhcr ~
f(lf o>!Tsh.".. platforms. It is IlOI in,'nded fa imply lha, unlisted shapes IlI'C unsuil.bk for athor applicalians.
79
Chapter 6
Caution
Earlier the importance of maintaining the proper tungsten electrode
tip to work distance was stated. Welders will often employ the technique of "walking" the cup along the sides of the groove in order to
help maintain that critical electrode tip to work distance. There is however a danger of losing effective shielding gas coverage because of the
tendency to lay the gas cup over so that it is more parallel to the joint
rather than pointing down into the groove. This occurs due to the curve
of the basic torch path as it circumnavigates the pipe and as the welder
tries to extend his vision before finally stopping and repositioning the
torch. The welder must realize just how critical inert gas shielding is
to the soundness of the weld.
The inert gas or gasses which are conveyed through the gas cup,
must be so directed that both areas, the molten pool and the area which
is in the solidification stage be exposed to inert blanket of protective
gases which is supplied through the gas cup. Therefore, the angle in
which it is held should be able to provide ample back wash of inert gas
to prevent oxidation and other reactions within the grain boundries.
From a welder's standpoint, the indication ofthese drawbacks will be
the appearance of the bead surfaces, which may be is dark and dulL A
weld which was fully protected from oxidation at high temperature is
a weld with a sheen and show no unusual amount of discolloration.
The seriousness of oxidation as described here is based on the fact
that temperature ofthe weld metal one hundred degrees below the critical range is weak, and with gains boundaries further apart as compaired to much lower temperatures. Without proper protection from
the atmosphere, oxygen may be absorbed within the grainboundries
and combining with other residual elements fonn compounds that will
weaken the grain structure as temperature decreases, with construction
taking place simultaneously, which leads to separation along grain
boundaries.
80
CHAPTER
The Intermediate and Cover Passes

Multilayer welds are used to weld aU thick-wall and most thinwall pipes. The multilayer weld is more ductile and free from defects
than a large single-layer weld would be. A multilayer weld is also
easier to make because it is more difficult to control the large puddle
of molten metal of a single-layer weld.
Since a large weld bead cools more slowly than a small bead,
excessive grain growth occurs, which causes a decrease in the
ductility of the weld. In the case of a weld consisting of normal-size
beads, the ductility is improved because of the heat treating effect
that each succeeding bead has upon the others. When a large bead
cools slowly, the hot metal is exposed to the atmosphere longer
causing oxides to form and this can lead to harmful effects.
This chapter is concerned with welding the intermediate and the
cover passes On heavy-wall pipe. As before, each bead is started at
the lowest part of the pipe and. welding uphill, the bead is welded to
the top of the pipe. Since the joint is now closed by the root bead, a
different welding technique is used. This is a slant weave, which is
used at all positions around the pipe.
The Multiple Layers. The first layer, or bead. in a multiple-layer

butt joint is the root bead. This is followed by one or more
intermediate layers (also called filler layers). The final layer is called
the cover pass. Each intermediate layer or bead is identified by the
order in which it was welded, starting with the second pass.
As shown in Fig. 7-1, the number of beads or passes that are
welded around the pipe joint depends upon the thickness of the pipe
wall. Heavy beads should be avoided for reasons given earlier in this
chapter. For this reason it is obvious that thick-wall pipes will
require more beads than will thin-wall pipes.
The recommended placement of the various layers or beads
across the section of the pipe wall is shown in Fig. 7 J. This
illustration can be used as a guide when deciding how to place the
beads.
Each bead must be firmly fused to the metal with which it is in
contact, whether it is another bead or the base metal at the walls of
the joint In effect, the entire weld should be a solid body of metal,
free from voids and other defects. The joint must possess strength
81
Chapter 7
I sr r~}:- (1;.0]-0
lift" JM
il;\l
~<lJ ,HIll"
.?~~~,
Jl3::"JiJ l,(tw
~ho:" :<I-U)' ql,l..,juy 1Ii~t-iJ '.1 -n"".-rl~,t ;p-lil,j

rC'l<A .""~J
C-.e pro-[l('f T<"~fMl"'!; p,(J~"cl'~1(" I:it"C p~g< 1i~~
!1... t.Ir"~\:ldrdMlr
IJpJlIll"5"C
waIITi)i~lirle.".
)7~
'"
~%
l~t
!'";.IS.I f-.-eOPJ
',;flll,J
m f'rti'a,,,,
J~ ~~H.l,IH'
1;15" .h~
~nc..! ~'}
oIrll f'~-;<;eS
l\t' I,'S" ,h~, VJW
U;" "'H3'~~1," "; l!)1: I,'k.\~ ~J-j
"r. l'a~ ... ~"
HnJrQ1ll;"" de;;noJ~
UphIll S(,
~~t f'~$1 H,OIO

1=:}\"\I>"
2"J Co- 4lh P.a~,<;~~
,,~~ ]"J~"
l~:.i. ~!wn .11-:
->.-n.J
~m{'><)~!'1 "OL'>~""{-"[
I "Yw< 'H~dPi.J","fl ~l~"u"'~n

J,,"(J'U ;U~ iU("':~ L)-:': In.t ,:ujf~"."."f
"',::,
n-.~ P>VC
{)I 1'1:'"
L~, .... II:t,.!I'-'f<:n
ell","t,,'H1r
Will~.l,m:'ht~ ~ ...3t~
<0
..,.
;~t r~"
!:(,[JH!
li:";",111t
~nd
w_ Sill
P<lo~,
~':>': l:W'"r
HO(lwlltal
t.
K","~p -Lo)w-B)i ..t~" 't"kdroo:.!l: i'f"t~ :fOf

\>It 1'(;;'<.1:, M'I.J t!',,~,~:rur.:
~"""""~ '.lCll~lm~ fhfl~ rn~'l! r-dorc ",-~I,:,~,
wwHyJr(l;1:trJ
,j,...
~G
so:."
~kdr(nk
1$1 t(l ~n.i p3h<.....

'-<"" I::-r,OI<J riP.
R~4cj .~,rUTH1Jlh"" CH'I ~<)va ~~.\~
l)1'!
~hl'",,~H
PljXi...:epaf(''Ij/
L.li~. er~.lro.J"
Ullhil150
P":>'5 \{x' \b~
l~l
~rtd
r'Es
~nJ ~I'lf.!
I, tv: rropel ~rl' ko"\glll {~.p~~e l ! J

,', \bJl'lt~~I: S1l1~tt> ek~lr{l<jl' nl./if"f,'IlLaW'''.
t:~"If'~
4th
f'~~,
J. ("1nll .... J"h ""p;, Of":pv'H'
~i:>l"m5in"
pf,,"pn~}
a.tJ,dt"<.'frodt~
ld tc,
~rJ P~'M~
l/1f' :in<,
ckdi<Jd~
$<:-: I.'l/l~'" 1.14 (<lr .... 'il"~~'r ;P4Hnl' ~:mJ

.:!o1-clrV<,1t:jlI'l!lk,
:. Ckaf) U.dl p~'S>. pt'()~,!}.
J. Avu..;! .;!r;KlSJ(''''1!! ;.:m:n P~$~ ""110:1'l ""~ ~~

v"Iyh.... '
1~1 :il:n<J ~rt<llJls:sc.'"
Sl'...,h
"Wlw
~r8"dj.;}. ~I.(dfodt
)f1J ~rn! 4lll p.a,~.e\\
jh~ >llf(lmJ~I,O" (:fn
\1~ f/S"'or5i31"
.r3'oJ, "k<:lrooo:-,
P<l# 150
..,tr ....w:d ..
jlfojJ<tdy
~ltm8ef },t'i(h
~, \fu~ ~.;.dl 13)'~1
1 "hjl1t~ln prpt:"( al:~ l~n~tl "n4 ~:PC('ci
Fig. 7-1. Recommended placement of weld layers on multiple-pass pipe joints

using low-hydrogen andlor E60 10 electrodes,
and ductility sufficient to withstand any load to which it might be

subjected.
The intermediate beads should fill the joint and be flush with the
outside surfaces of the pipes. This is a very important condition prior
to welding the cover pass. It is very difficult to weld a good cover
pass unless the metal over which it is welded is flush with the outer
surfaces of the pipes. The cover pass should be slightly crowned and
the sides should blend smoothly into each pipe without undercut.
Preparation of the Root Bead. A root bead that is perfectly formed

internally can be very rough on the outside. as shown in Fig. 7-2A
and B. Depending on the skill and care used by the welder, it may
82

have humps and undercuts, especially where restarts occur. To clean
and to prepare a surface such as that shown in Fig. 7-2 requires
more than simple chipping and wire brushing.
The root bead should be prepared by grinding out aU of the
undercuts, rough spots, and humps to avoid the risk of having an
invisible defect after the second pass is finished. Visible defects can
be corrected, but this is costly.
Under ordinary circumstances, root bead grinding can be somewhat overemphasized and removing an unnecessary amount of weld
Fig. 7-2. Root bead that is perfect internally, but rough externally. k Outside
surface; B. Underside of root bead.
metal should also be avoided. Deep penetrating type electrodes, such

as E6010, E6011, and E7010, will easily melt the metal in any
defective areas and fill the voids at undercuts. When these electrodes
are used, the smallest humps can be removed by melting; only the
most serious defects will require grinding to correct them.
When shallow penetration type electrodes, such as low-hydrogen
electrodes, are used to weld the joint, much greater care must be
exercised in preparing the joint. Since these electrodes are used to
weld high-pressure pipe joints, extra work and care are required in
the preparation of these joints.
Because of their lack of penetrating power, it is much more
difficult to melt humps and to fill-in undercuts with weld metal when
using low-hydrogen electrodes. To maintain the short arc used with
these electrodes, the surface over which the weld is made should be
smooth.
For these reasons, the surface of the root bead must be ground to
a flat contour, as shown in Fig. 7-3. The root bead surface is
widened by grinding to remove undercuts at the fusion line between
83
Chapter 7
Fig. 7-3. Root bead ground lo flat contour in preparation for welding second
pass with a low-hydrogen electrode.
the weld metal and the base metal. Since the low-hydrogen electrodes have a heavier coating, widening the root bead also provides
more space for electrode manipulation and the flat contour will
enable the welder to maintain a constant arc length.
In summary, the degree of preparation of the root bead for
welding the second pass depends upon the type of electrode being
used. When high-pressure joints are to be welded there can be no
compromise in preparing the joint. In other cases, the welder must
exercise a certain amount of judgment; he must decide which defects
are serious enough to affect the quality of the second pass, and
remove them leaving only those minor defects that can be corrected
on the second pass.
The same conditions prevail when preparing the joint for welding
the beads that will be made following the second pass. When welding
with low-hydrogen electrodes, the surface over which the weld is to
be made must be smooth and free of all defects. When welding with
deeply penetrating electrodes, however, only major defects must be
removed.
The Current Setting. Because of the varying conditions of the job

location, the machine, and other factors detailed in Chapter 3, it is
not possible to give an exact current setting that can be used in all
cases. In order to provide a relative idea of the current setting to use
for the intermediate passes, the current setting for the root bead will
be used for comparison.
84

For the second pass and for subsequent passes, the current setting
will depend primarily on the temperature of the pipe joint. If the
pipe is still very warm (about 200 to 300F) when starting to weld the
second bead, it is possible to use the same current setting that was
used to weld the root bead. At this temperature the pipe joint is too
hot to touch with the bare hand. If the temperature of the pipe joint
has d~opped significantly below this temperature, the current setting
must be increased to provide more heat to melt the parent metal and
to maintain the puddle.
When the welding operation has been suspended for a longer
period of time, such as overnight, the weld joint will be at the
temperature f)f the surrounding air. In this event it is advisable to
preheat the joint with an oxyacetylene torch. In many instances,
such as when welding highly alloyed pipe, preheating is mandatory.
This reduces the thermal shock on the joint when welding is
resumed, and also reduces the severity of the residual stresses locked
in the weld when it has cooled to the temperature of its surround~
ings.
When several layers of weld beads have been deposited, the heat
will transfer out of the weld at a slower rate. In other words, they
tend to store heat, at least for a short time. When this occurs, there is
no need to change the welding current. However, if these layers have
been allowed to cool to the temperature of the surrounding air, they
will act as a heat sink by rapidly withdrawing heat from the weld. In
some cases, the cooling rate of the weld may be so fast that it will act
like a quench. Moreover, more heat is required to raise the temperature of the layers of cold weld metal, often more heat than can be
supplied by the arc.
For this reason, preheating with an oxyacetylene torch is highly
recommended when several layers have been deposited and which
have been allowed to get cold. If several layers have been deposited
in pipe having a wall thickness of % inch or more, preheating is
almost mandatory.
In summary, the current setting depends on the temperature of the
pipe joint. If it is warm (about 200 to 300F), less current can be used
than if it is at the temperature of the surrounding air. If the pipe joint
has been allowed to cool, preheating is very advisable and often
mandatory in certain cases.
Electrode Angle. The correct electrode angles for welding the
intermediate and cover passes around the pipe joint are shown in
Fig. 7-4A and B. The side angle should be such that the electrode is
perpendicular to the surface of the pipe.
85
Chapter 7
Figure 7-4C shows one error that is sometimes made, even by
experienced welders, when welding from the 9 o'clock [0 the 12
o'clock positions. The electrode angle in this case is too great. By
facing upward, the arc will cause the puddle to be less fluid causing
slag inclusions and improper fusion at the edges of the weld.
When the molten pool of metal is mushy, it will not fuse properly
-1
~5."I~
Fig. 7-4. Correct and incorrect electrode angles for welding intermediate and
cover passes, A. Correct electrode angle for 6 to 9 o'clock positions; B. Correct
dectrode angle for 9 to 12 o'clock positions; C.lncorrect electrode angle for 9 to
12 o'clock positions.
nor will it flow readily. The slag, having a lower melting point than
the metal, will be very fluid and it will flow swiftly to the edges of the
weld where it will become trapped between the base metal and the
weld metal. Therefore, it is necessary to prevent the molten pool of
metal from becoming mushy while welding. While several factors
affect the fluidity of the puddle, one that must not be ignored by the
welder is the electrode angle; it should always be as shown in Fig.
7-4A and B.
Welding the Intermediate Passes. Before each pass is welded, the

joint must be cleaned and prepared as described previously in this
chapter. Each layer should start in a different position on the joint;
i.e., no two layers should be started in the same position. For
example, the root bead was started in the 6: 30 position; thus the
second layer should then not start in this position but in the 6
o'clock position.
The arc should be struck on the bead over which the weld is to be
made. It should be struck ahead of the weld to preheat the metal
over which the first part of the weld is to be made. As usual, a long
arc should be held in this position until it has stabilized and the
gaseous shield has formed. It is then held at the 6 o'clock position
and shortened. After the puddle of molten metal has formed, the
electrode is moved on to form the bead.
86
ELECTRODE MOVEMENT
PAUSE and WEAVE
Courte,,, o/rhe Hob(lY/ Bro/hers Co.
Fig. 75. Electrode movemenl for making slant weave used to weld all
intermediate and cover beads.
All of the intermediate beads are made by using a slant weave,

Fig. 7-5. This weave is used in all positions around the circumference of the pipe joint. The arc shoUld be moved along at a smooth
and steady pace by operating the electrode with a wrist movement.
At the end of each weave there should be a slight pause, after which
the direction of the arc is reversed. Because the surface of the pipe is
curved, care should be exercised to maintain a uniform arc length
and electrode angle. Varying the arc length can have a significant
effect on the depth of fusion and the size of the puddle. By maintaining a steady movement and a uniform arc length the electrode is
seldom away from the edge of the puddle which results in a uniform
depth of fusion and a bead with a neat appearance.
It is important to pause momentarily at the end of each weave.
This allows the filler and base metals to mix properly, and it also
allows the bead to fuse properly to the side of the joint. Any slag that
is trapped in the corner of the weld will be remelted and will flow
toward the rest of the metal where it will not interfere with the fusion
of the puddle and the side of the joint. The molten metal in the
puddle will also fill any undercuts in the corners, and will solidify
when the arc is moved on,
The puddle must be maintained at all times and should not be
allowed to become mushy, or small particles of slag may become
87
Chapter 7
entrapped in undercuts at the corners or in cavities. The welding
speed should be adjusted to maintain a fluid puddle, yet care must be
exercised not to allow the puddle to sag when welding from the 6
o'clock to the 3 o'clock positions. As the weld advances beyond the
3 o'clock position, the tendency for the puddle to sag decreases and,
for this reason, the speed of welding can be gradually decreased to
maintain a uniform build-up of weld metal.
For light coated electrodes (E601O, E6011, E701l), the arc
length should be about %2 to % inch. This will provide enough heat
so that the puddle of molten metal win be large enough to accept the
globules of filler metal without excessive build-up. If a short arc (1/16
inch) is used, the size of the puddle of molten metal is decreased
considerably. In this case the size of the pool of molten metal in
which the filler metal can be deposited is limited. When the globule
enters the smaller body of liquid metal, it will rise and. at the same
time, cool more rapidly causing the bead to have a high crown (Fig.
7-6) and, perhaps, lack good edge fusion.
The use of an excessive amount of current must also be avoided.
When this occurs the welder will increase the welding speed and the
rate of electrode manipUlation to keep the molten metal from
sagging or from overflowing. He will tend to become erratic in the
manipulation of the rod, varying both the arc length and the weave.
To control the puddle he will often resort to a U-weave which
periodically will remove the gaseous shield from the molten metal
and produce a harmful effect on the quality and appearance of the
weld.
Fig, 7-6. Bead wi,.. mgh crown and lack of edge fusion caused by welding with
a short arc.
Fig. 7-7. Width of root bead, second pass, and third pass.
88

The third pass should be made wider than the second pass and it
should fill the groove from side to side, as shown in Fig. 7-7. This
pass is made by using the same slant weave as before, pausing at the
end of each weave. The width of subsequent weaves will depend on
the thickness of the pipe and upon the judgment of the welder.The
manner of placing these beads in relation to each other was shown in
Fig. 7 -1. A bead may be deposited by making the center of the
electrode pause over the edge of the bead below, or the weave may
be widened by making the electrode pause one electrode diameter
beyond the edge of the bead below. The appearance of each bead
should be smooth and without undercut.
Stop and Restart. When the cnd of a bead is reached or when the
electrode is consumed, the weld must be stopped. This is done by
simply reversing the direction of the electrode travel for a short
distance and then breaking the arc with a quick movement. The
molten metal in the puddle should solidify in the form of a crater, as
shown in Fig. 7 -8.
Before restarting, the end of the weld should be chipped and wire
brushed for a distance of at least % inch beyond the crater. All
traces of the slag coating must be removed from this area.
To resume welding, the arc should be struck ahead of the crater
and a long arc maintained, as shown in Fig. 7-9. When the arc has
been stabilized and the gaseous shield has formed, it is brought into
the crater and shortened to the normal length for welding. The arc
should be moved slowly, from side to side in the crater, until a pool
of molten metal has formed; only then can the bead be continued by
resuming the slant weave.
Fig. 7-8. View of bead that has been stopped to change electrodes, showing the
crater.
89
Chapter 7
The arc should not be struck in the crater. This practice would
cause the first few globules of filler metal from the electrode to build
up like humps because the pool of molten metal had not formed
when they were deposited. It has also been found that porosity
results in the restart zone when the arc is struck directly in the crater.
This is probably caused by the absence of an adequate gaseous
shield to protect the hot metal, and the absence of enough slag from
.tEI--STRIKE ARC HERE
Fig. 7-9. Restarting the weld bead.
Fig. 7-10. Viewofagood tie-in.
90

the electrode coating, in time to thoroughly deoxidize the molten
metal.
Tie-in. It is not difficult to make a good tie-in (Fig. 7-10) when

welding intermediate and cover beads if care and attention are given
to this matter during welding. When approaching a tie-in, the weave
may be slowed down slightly to allow the molten metal to build up
somewhat. Then. when the built-up pool of molten metal reaches the
other bead, the welder must watch to see that the molten metal fills
the crater and blends smoothly with the other bead; if necessary, he
may have to run the arc a very short distance (1/]6 inch) over the
other bead in order to achieve a smooth tie-in. When the smooth
blend of the two beads occurs, the arc is broken with a sudden
movement.
The Cover Pass. The cover pass is the final pass or bead to complete
the weld. It must have a smooth and uniform appearance. As shown
Fig. 7-11. Perfect cover passes. (Top) Horizontal ,2G) weld made with E6010
electrode. (Bottom) 5G position cover passes, with joint in center made with
E70 18 low-hydrogen electrode and the two end welds made with E601O.
in Fig. 7-11, it should have a slight crown. When the pipes are in the
5G position, as shown by the lower pipe in Fig. 7-11, a single bead is
used to make the final cover pass. Since this is difficult to do,
however, when the pipe is in the 2G position, as shown by the upper
pipe in Fig. 7-11, the cover pass a'many will require three separate
beads to be made. Welding in the 2G position will be treated later, in
91
Chapter 7
Chapter 8~ this chapter will treat welding of the cover pass in the 5G
position only.
Welding the last intermediate layer, or layers, is a preparation for
welding the cover pass. It is important that they be welded flush with
the surfaces of the pipes. Undercut and poor edge fusion should be
avoided; if they occur, they should be removed and the damaged
areas rewelded before welding the cover pass.
In preparation for welding the cover pass, the weld joint must be
thoroughly cleaned. If low-hydrogen electrodes are used, any humps
on the surfaces of the beads over which the weld is to be made must
be removed. Normally, the cover bead should overlap the edges of
the bevel for approximately V1G inch.
Since the weld must be uniform and have a good appearance, the
welder should always be in the most comfortable position that
circumstances will allow in order to be able to manipulate the
electrode smoothly and accurately. Whenever the welder senses that
the manipulation of the electrode is impaired, he should stop welding and change position.
The arc should be struck ahead of the weld and brought back to
the 6 o'clock position while holding a long arc in order to stabilize
the arc and form the gaseous shield. It is then shortened to a normal
arc length. The welding process begins by weaving the electrode two
or three times at a slow rate in order to form the molten pool of
metal and to obtain good fusion.
As shown in Fig. 7-12, a slant weave should be used to make the
entire cover pass. Close attention to the job and careful workman-
Courtes\.' of llr!' Hohar-t Brothers Co.
Fig. 7-12. SIan! weave used to make the cover pass.
92

ship are required in order to obtain a neat appearing bead. The
electrode angle must be correct and a uniform arc length maintained.
When making the weave. the welder should watch the edges of the
underlying bead. They serve as a guide in order to maintain a
uniform width of the cover bead around the entire pipe. At the end
of each weave, the electrode should pause to prevent undercut and
to obtain good edge fusion.
The welder must also watch the puddle and maintain a welding
speed so that the crown (also called the bead reinforcement) of the
solidified bead remains between Vat and lf16 inch above the surface of
the pipe. Care must be exercised not to permit the crown to build up
too high, especially in the vertical and overhead welding pusitions.
Difficulties can be encountered in welding the cover pass for the
following reasons:
1. The pipe is too hot
2. The joint was not cleaned properly prior to welding
3. The intermediate beads have not filled the pipe joint completely
4. There is undercut and poor edge fusion of the previous bead,
If the pipe is too hot the molten meta] will solidify more slowly
and it will be very difficult to control the puddle. As a result there
will be a rough-looking weld which is unsatisfactory as a cover pass.
The remedy would be to allow the pipe to cool until it is warm
(about 200 to 300F) before starting to weld. Sometimes it is possible
to weld a hot pipe satisfactorily by reducing the welding current.
Cleaning the joint properly prior to welding is the welder's responsibility. Undercut, poor edge fusion, and improperly filled joints
are also the welder's responsibility. He should not start to weld the
cover pass until these matters are corrected. Better yet, he should
exercise greater care when welding the intermediate passes so that
these conditions do not occur.
Low-Hydrogen Electrodes. As already explained, low-hydrogen electrodes are not a deep penetrating type of electrode and for this
reason the surface over which they weld must be smooth. Before
each pass, the surfaces of the joint should be cleaned, have all
defects removed, and be ground smooth. Proper preparation is an
important part in welding successfully with low-hydrogen electrodes.
93
Chapter 7
The general procedure for welding the cover bead with lowhydrogen electrodes is the same as for more deeply penetrating
electrodes; however, there are some differences. When the arc is
struck it should be shortened immediately. The whipping procedure
should never be used to control the puddle. The weld should be
made by using the slant weave described in this chapter. At the end
of each weave the welder must be sure to pause in order to avoid
undercutting. It is very important to maintain a short arc at all times
in order to avoid sagging and to prevent pinholes from occurring in
the weld.
The electrode coating must be kept dry and electrodes with a
chipped coating should not be used. Wet joints must be preheated
with an oxyacetylene torch. Welding should never be done in the
rain when using low-hydrogen electrodes. If the welding area
becomes wet, the electric arc will break down the water, forming
hydrogen, which will enter the weld metal with very harmful effect.
Shielded Metal Arc Welding-Electrode

The shielded metal-arc welding process has been widely used for
many years to weld pipe material. It is still used, even in the most
modem industries. The process is used with various types of alloy
pipe materials in constructing power plants, oil refineries, chemical
plants, nuclear power plants, and oil and gas transmission lines.
Specifically it is used to make or join welds that meet the requirement
of the industries' respective codes and specifications for piping systems and materials. These requirements cover various conditions such
as high strength at elevated temperatures, corrosion and oxidation
resistance, steels for cryogenic services, and high alloy steel that must
resist chemical attack or require high tensile strength.
The ability of the steel to achieve its needed mechanical properties
in the various environments, depends on proper steel making practices
such as degassing, removal of impurities. hot rolling within the critical range, grain control, and even controlling the cooling rate and heat
treatment. When the welding is applied, the filler metal on the electrode must be formulated so that the completed weld meets the
required strength and other properties for which the system was
designed.
Shielded metal-arc covered electrodes are formulated in a way analogous to process metallurgy. However, the challenge of formulating
94
electrode coatings involves knowing that the tiller metal will be transferred in the intense arc heated column as fine particles to the base
metal molten pool. Many of the materials used to make slag, in con
nection with refining metal in a furnace, are also used in fluxes for
welding. Limestone, flourspar and rutile are used make steel. There are
many other suitable minerals compounded to make a flux. For example, certain precautions must be taken to ensure protection during the
transfer of the alloying element, and to guard against the pick-up of
elements that could affect weld properties. Furthermore, weld metal is
susceptible to oxidation when the metal is exposed to oxygen. Other
problems include porosity and blow holes caused by the effusion of
gases. These are some of the same problems that are experienced with
ingots and casting during steel making.
Most flux-bearing arc welding electrodes contain additional elements
to function as deoxiders. They are supplemented by adding alloying
elements in the electrode coating. The molten pool is protected by
deoxiders and fluxing agents that remove the oxide formation of the
hot meta] ahead of the arc. Numerous types of coated electrodes are
available for welding various pipe alloy.
Electrode coatings have many functions, including:
(I)
maintaining a stable arc
(2)
controlling depth of penetration
(3)
adding bulk to the weld deposit
(4)
increasing the ionization potential, and stabilizing the arc
(5) depositing welds with a uniform ripple
(6)
production suitability
(7)
durability and storage
(8)
slag that is easily removed
Mild steel material has a low carbon content as well as small amounts
of manganese and silicon. Welding such material offers little chal~
lenge to its microstructure; in fact, the microstructure is basically per~
lite, which is soft and ductile.
The appropriate electrode for welding this grade of material is the E60 I0 electrode. There are many variations of this basic electrode type.
For example, the covering of the -6020 electrode has a high ironoxide content, as much as 30 percent. The amount varies substantially
in the series, with little or no gas shielding provided by the covering.
The primary purpose ofthis covering formulation is to provide a heavy
layer on the metal, thereby obtaining a satisfactory operation with high
current. The welder is then able to obtain a deep penetration in the butt
and fillet welds in the horizontal and flat position. E-6020 electrodes
95
Chapter 7
are used most often on relatively heavy sections where deep penetration and high metal deposit rate is sought.
The E~6010 electrode has a thin layer of (flux) coating that barely
covers the entire weld. This thin layer offers a higher viscosity than the
heavy covering on other electrodes. It thus makes the E-60 10 an electrode most suited for depositing root bead on pipe without dripping
either inside or outside, once the root bead technique is adapted. The
E-6010 electrode does have limits where it can be applied, as follows:
It is effectively used for deposited root bead on carbon steeluphill and downhill welding.
(2)
[t is limited to welding on low alloy steels because of its low
content in alloying elements. However, it is sometimes used
for depositing the root bead, which is then filled and capped
with the low-hydrogen electrode, E-70 18.
(3)
Welding thinwall and low-to-medium carbon steel pipe with
E-60 10 is acceptable.But if the wall thickness exceeds half an
inch, only the root bead is deposited by the E-6010, with fill
ing and capping by the E-7018 electrode.
(4)
E-6010 is considered a high cellulose electrode. Therefore, it
generates a high level of hydrogen not accepted for welding
high carbon and medium alloy steel.
(5)
As a last resort, the E-60 10 electrode can be used for
depositing a root bead on high carbon steel, but not without
preheating and maintaining interpass temperatures.
(6)
Because E-6010 is considered a high cellulose electrode, the
final gaseous product from the volatilization and combustion
of the cellulose results in a covering of carbon monoxide, car
bon dioxide, water vapor, and hydrogen. [n tum, because of
the high inducement of hydrogen, E-6010 is not used for
depositing root bead on either medium alloy steel or steel with
high tensile strength.
The electrode E-7047 has a low hydrogen type of coating. Because
it is lightly coated, this electrode can be used for downhill welding, of
root beads.
There are instances when a low or medium alloy pipe needs to be
welded, and the root bead becomes a matter of concern because no
other process is available and no electrode available other than E-60 10
is available for depositing the root bead. Pick-up, also known as dilu~
tion, becomes a factor. Pick-up relates to the welding of an alloy steel
with a filler metal of a lower alloy content than the base metal content.
In a multilayer weld, the first pass (root bead) naturally picks up the
(1)
96

greater amount of alloying elements from the base metal. When, for
example, welding a chrome alloy pipe. The root pass deposit made
with the available E6010 electrode will gain as much as 50 percent
chromium due to dilution or pick up from the base metal.
The viability of this proceedure must be proven by the welder, as the
root bead alone is deposited by this type of electrode. The welder must
then prepare the surface of the root bead deposit by grinding and
removing one third of its thickness, with smooth fusion edges on both
sides of the root bead. The welder's next objective is to use one e1ec~
trode of reasonable size in diameter, and an adjustment in the current
allowing penetratation of the root bead as deep as possible, but avoidw
ing bum through the root bead.
A small crater should be apparent after the first electrode is consumed. The edge fusion will give the welder an indication as to the
depth of penetration. The root bead can then certainly be enriched by
the deposit from the second pass, using the correct electrode based on
specifications. This procedure involves applying preheating temperature and maintaining an interpass temperature. knowing that the base
metal and the filler are both of alloy type. An experienced supervisor
or experienced welder should help with this procedure. This points out
that solving variations in welding is not always based on the procedure
itself, but often on the skill of a well-trained welder who understands
all that is required.
The electrode coating may carry supplementary alloying elements
that are needed; the core wire is not being of the same composition as
the base metal. Thus, it has become a practice to use a supplementary
addition to the coating when manufacturing electrodes. In addition, the
electrode coating has other functions. These include protecting the
transfer of filler metal within the arc stream by a blanket of gases that
keeps the oxygen out or dissolves the oxidizing gas to a less active one
like Co2 or Co, which can also serve as a protective atmosphere.
The electrode coating is made up of compounds and elements that are
responsible in their own way to protect the molten weld metal acting
as either slag, deoxider, or fluxes.
The slag must be less dense than the weld metal for obvious reasons.
It must not freeze at a higher temperature than the weld metal; otherw
wise, the slag will remain dormant or stagnate on the face ofthe bevel,
disturbing the welding process. In addition. the viscosity ofthe slag (its
ability to flow) has an important bearing on the quality of the weld.
When the viscosity is low, the slag is watery. It will flow, leaving the
hot metal exposed to the atmosphere, and thus to become oxide. If the
viscosity is high, the slag will be very sluggish. This in tum becomes
97
Chapter 7
a challenge to the welder, especially, when making welds in certain
positions. When slag is fonned on the weld, it will be impregnable to
the gases from the atmosphere. Therefore, the weld metal will be protected. However, manufactures of electrodes have taken these variables
into consideration allowing the welder to choose the optimum electrode
for the job at hand.
A deoxider is a oxygen setter. It serves in molten metal to dispose of
oxygen and oxygen-bearing compounds or it remains in the weld pool
as a safeguard in case oxygen enters. When oxygen does enters the
molten weld pool, it reacts with manganese to fonn manganese oxide
(MnO). As long as manganese is present, this process will continue
rather than the fonnation of iron oxide. In tum, manganese oxide reacts
with the carbon to form the Co bubbles that cause porosity (blow
holes). If aluminum and silicon are in the steel or the weld pool, there
will be no blow holes. The silicon and the aluminum should react with
the oxygen to form oxide preferable to FeO. Because it has lower density than the metal, the oxygen bearingslag wi II rise to the surface, joining the bulk of the slag.
During welding, the surface of the joint to be welded can develop an
invisible oxide coating, even though it was brushed twenty-four hours
earlier. While the weld is in progress, the heated metal ahead of the arc
will be exposed to the atmosphere and the oxidation in progress. The
object, then, is to protect the molten pool and to dispose of the oxide
ahead of the arc by reducing them.
The method by which a flux deals with the oxide is by mixing or comingling with the oxide to fonn a slag that has a more favorable melt
ing point and viscosity. Therefore, the electrode coating is also a slag
that forms a blanket on the weld surface, shielding the metal from oxidation and disposing of oxide that forms from inefficient shielding
caused by poor manipulation or other reasons.
There are many individual components in the electrode coating. There
are some that combine with other compounds and become chemically
active shielding the molten puddle and display of oxide. For example,
limestone is a component that can help produce Co and Co2 gases that
serve as a protection to the arc area and the arc stream; it produces 40
percent of its weight in gaseous fonn; the rest becomes calcined, that is,
it changes to a white powdery lime. This latter product of the welding
process is an example of a fluxing agent. It is secured from limestone
after calcination. Other components that could be added to this list are
flourspar, sodium oxide, feldspar, ferromanganese, graphite, ferrosalinium, ferrosilison, chloride, and fluoride salt.
98

These agents melt and gather as a liquid blanket on all parts of the
weld puddle except the arc spot. The flow of the fluxing agent can be
seen as a slight watery substance, a blanket extending the edges of the
puddle and dissolving the oxide. The principle function of the flux is
to dissolve or dispose of the high melting oxide that may form despite
the protection afforded by the gaseous shield around the arc.
To obtain high quality welds the SMAW process requires careful
storage and handling of the electrodes. An experienced supervisor can
determine whether or not the quality of the weld is at risk based on the
condition under which the electrodes are maintained in the work place.
Too often the welder draws the conclusion that using high current is
the best way to address the tiny holes that appear on the surface of the
previous pass (weld). Yet when the surface is removed by using a
grinder, the holes are in fact larger in diameter. Therefore, high current
is not a remedy for such a defect. Gases such as Co and Co2, by themselves or in combination with others, are very soluble in molten metal
during welding. As the metal begins to freeze, gases in the weld seek
to escape from the weld. As the heat dissipates or is conducted into the
walls of the base metal, it also loses heat from the surface of the weld.
If effusion of the gases from the weld is not completed before the weld
puddle becomes solid, the gases will rise within the semi-molten weld,
leaving what are called wonn holes. Because of its viscosity at that
stage, the metal will not be able to flow to fill these holes, which will
remain as voids. If the surface of the weld is in a semi-molten stage, a
portion of the gas will escape, and there will be the appearance of pin
holes on the surface of the weld, known as porosity.
The coatings of properly baked electrodes are brittle. This coating
can fracture quite easily during transportation to the work site, or when
the electrodes are carelessly stored among tools such as chipping hammers and chisels. The cracks on the electrode coating can be so tiny
that they cannot be seen by the naked eye. Often a welder will bend an
electrode, to gain acess to a joint in constricted area. The coating then
may be fracted or actually break-away from the core, As this damaged
electrode is used the electrode coating can flake and fall into the weld,
as well as to the ground. The problem then is that the electrode can be
depleted of its alloying ingredient, which in tum will lead to inade~
quate mechanical properties, oxidation and loss of corrosion resistance, especially if the steel happens to be subjected to high temperature service. In addition, the weld will suffer from not having sufficient
oxidizers, which is likely to cause porosity in the weld. Cracks due to
rapid cooling shrinkage are another possibility.
99
CHAPTER
Welding Thin-Wall Pipe

Thin~wall pipes are categorized as having a wall thickness less
than %6 inch. The diameter of thin-wall pipes may be as small as 1
inch and as large as 42 inches, or larger, such as that used for
cross-country pipelines.
When the pipeline is horizontal, or the pipes are in the 5G
position, either the uphill or the downhill welding procedure can be
used. Because it is a faster method, downhill welding is preferred for
welding cross-country pipelines, where the pipes are usually made of
low-carbon mild steel and this method of pipe welding will be
emphasized in this chapter.
Not only the wall size but the diameter of the pipe affect the
welding method that can be employed. Small-diameter pipes with
thin walls present a problem because the heat resulting from the
weld does not dissipate rapidly, thereby causing the heat to build up
in the weld zone. In this case the heat input into the weld must be
controlled.
To control the heat input, for welding small-diameter pipe the
current setting is less, and a smaner-diameter electrode is used as
compared to that used for large-diameter pipe having the same wall
thickness. While welding small-diameter pipe, the welder must pay
careful attention to the electrode angle and to the arc length. The
electrode angle is more difficult to control on small-diameter pipe
because the position of the electrode changes rapidly in order to
maintain a uniform electrode angle while welding around the pipe.
A drastic change in the electrode angle can have an adverse effect on
the quality of the weld. While watching the electrode angle, the
welder must also prevent the arc length from becoming too large as
(l) this will increase the heat input into the weld, and (2) if excessive,
it will decrease the weld penetration. Considerable practice is
usually required of the welder before the technique of welding
small-diameter pipe is mastered.
Uphill Welding
Very highquality pipe joints can be made using the uphill weld
ing method to weld both large- and small~diameter thin-wall pipe. It
100

is recommended for welding high-alloyed pipe and in other cases
where only the highest-quality welded joint will be accepted.
For large-diameter pipe, the procedure is exactly the same as that
described in Chapters 5, 6, and 7. The root bead may be welded by
the shielded metal~arc-welding process or by the gas tungsten arcwelding (GTAW) process. The procedure for welding the intermediate and cover passes is the same as previously described.
When welding sman~diameter thin-wall pipe, the root bead and
intermediate passes are welded as previously described; however, a
slightly different procedure must be used to weld the cover pass.
Because more heat is retained in the weld zone when welding
small-diameter thin-wall pipe, the temperature of the weld is often
very high when the cover pass is deposited.
In this case, a semicircular weave, shown in Fig. 8-1, is used. The
current setting should be slightly decreased when using this weave
pattern. It is made with a continuous motion, there being little or no
pause at the end of each weave. This allows the area that is seen
encircled in Fig. 8-1 to cool and lose a considerable amount of
fluidity when the arc is on the opposite side. When the arc is returned
Fig. 8-1. Semicircular weave used to weld the cover pass when welding smalldiameter pipes by the uphill-welding method.
to the encircled area and the filler metal is added, the metal will not
drip as a result of overheating. The electrode angle remains constant
when welding around the pipe, as shown in Fig. 8-2.
When this weave pattern is to be used, it is very important to clean
the weld joint thoroughly beforehand. Since there is little or no
hesitation at the end of the weave, some danger exists that slag may
become entrapped, causing incomplete fusion to occur in this area.
The welder, therefore, must take a position which enables him to
manipulate the electrode comfortably and precisely. This is especially important when welding small-diameter pipe because of the
continuous change in the electrode position. A perfect cover pass is
shown in Fig. 8-3.
101
Chapter 8
DIRECTION OF WELOING
Fig. 8-2. Correct electrode angle when welding the cover pass on small-diameter
pipe by the uphill-welding method.
Downhill Welding
Thin-waH pipe can be welded by the downhill weld method.

Downhill welding is very fast and economical; consequently, it is
used extensively to weld cross~country pipelines. In the majority of
cases, the wall thickness of the pipes used for cross-country pipelines
is within the range of thickness that can be welded by this method.
High-quality welds can be made on thin-wall pipe by the Downhill Weld method. No laxity in the quality of the pipe joint is
permissible when the pipeline must transport crude oil, natural gas,
and other fuels. Leaks and other defects can present a danger to life
and property, as well as be the cause of environmental damage.
Most cross-country pipelines are buried underground for almost
their entire length. In order to be certain that they are sound, it is
normal procedure to test over 50 percent of the welds.
The procedure for downhill welding is different from that for
uphill welding. The molten puddle of metal tends to roll down the
pipe in the same direction as the arc is traveling. Moreover, the fluid
slag in the molten metal also flows' in this direction and, unless this
can be controlled, there is a danger of slag inclusions occurring.
In order to deposit a sound layer of weld metal, the arc must
constantly be kept ahead of the molten pool of metal. This is done
by using a high current setting and a fast speed of travel. As a result,
the layers of weld metal are thinner, when compared to uphill
welding.
102
Welding
Thin~Wall
Pipe
Fig, 8-3, A perfect cover pass deposited on small-diameter pipe by uphill

welding,
Fig. 8-4, Typical disposition of weld beads on a cross-country pipeline joint.
Figure 8-4 illustrates a typical disposition of the weld beads in a

joint when welding a cross-country pipeline. As usual, fOUf tack
welds are made to hold the pipes in place, following which the root
bead is welded, starting from the top of the pipe and welding to the
bottom. After the root bead has been welded entirely around the
pipe, the second pass, called the "hot" pass, is welded. The primary
objective of this hot pass is to correct any defects in the root bead;
a relatively small amount of metal is deposited by this pass. A
third, intermediate, bead is then deposited, which is followed by the
cover pass to finish the weld.
For practice-welding, two 7-in. long, 8-in. diameter, Schedule 30
mild steel pipes are recommended. The actual outside diameter of
these pipes is 8.625 inches and the wall thickness is .277 inch. The
103
Chapter 8
instructions to follow apply to welding larger-diameter pipes, such as
those encountered on cross-country pipelines, as well as the nipples.
Horizontal (2G) Welding. The procedure for welding thin-wan pipe

in the horizontal (2G) position is exactly the same as for welding
thick-wall pipe. For this reason, this subject will not be treated here
and the reader should refer to Chapter 9 for information on horizontal pipe welding.
Outdoor Pipeline Welding. Much of the downhill pipe welding is
done outdoors on cross-country pipelines. For this reason it is
necessary to mention a few precautions that should be taken when
welding under these conditions before treating the actual welding
methods.
('"urte.')" i'( tl". H&M Pipe Bel'ellIl!! .Hueh"",
(0. /17(,
Fig. 8-5. Typical line-up clamps used to clamp and align cross-country pipeline
joint.
Weld cracking can be a problem when welding pipes outdoors.

particularly when welding large-diameter pipes having a higher alloy
content. Several different grades of steel are used in pipeline construction and are designated by the API Standard as X42, X52,
X56, X60, and X65. The alloy content increases in these pipes with
an increase in number designation. The tensile strength and other
mechanical properties also increase in the higher numbered pipes.
This progressive increase in properties is the result of the greater
alloy contents, principally carbon, manganese, and, perhaps, silicon.
Line-up clamps, such as shown in Fig. 8-5, are used to align the
pipes and to hold them in place while welding. The root bead is
somewhat thin when welded by the downhill method and the pipe
joint is weak, until sufficient metal has been deposited in the joint.
Short tack welds alone would be apt to crack. For this reason, the
clamps are left in place on the pipes until 50 to 100 percent of the
root bead has been deposited.
104
Another form of cracking that occurs, particularly on pipe having

a somewhat higher alloy content, is under bead cracking. A more
detailed discussion of the nature of this form of cracking is given in
Chapter 13; in the following paragraphs some of the precautions
that must be taken to prevent its occurrence when welding outdoors
are discussed.
Underbead cracking can occur in pipes having a higher alloy
content when the weld is cooled too rapidly and when moisture is
present. Therefore, the weld joint should not be allowed to cool to
the temperature of the surrounding air because the heat retained in
the weld will slow down the cooling rate of the next bead to be
welded.
In the field, this means that the second layer should be welded as
soon as possible after the previous layer is finished, preferably within
five to seven minutes. On large-diameter pipes it is not possible for a
single welder to weld fast enough to maintain the desired interpass
temperature in the weld joint. In order to maintain this temperature,
it is not unusual to see as many as two to four welders working on a
single weld joint simultaneously.
High winds can also cause the weld to cool too rapidly to
maintain the desired interpass temperature, especially if the air is
cold. Unless a wind shield is placed around the welding area,
welding should be stopped if the wind velocity exceeds 25 to 30
miles per hour.
Since moisture can cause underbead cracking, welding should be
stopped if it rains or snow is falling. Of course, if a canopy is placed
over the weld area to protect the weld joint from rain and snow,
welding can be continued. However, the weld joint should be thor~
oughly dried with an oxyacetylene torch before welding and great
precaution should be taken to keep the electrodes dry.
Ingredients in the coating of the electrodes used for downhill
welding can contribute to underbead cracking if the weld cools too
fast. This is another reason for retaining a fairly high interpass
temperature. Ideally, low-hydrogen electrodes should be used
because the coating on these electrodes does not contain the harmful
ingredients. However, it is not possible to weld downhill with lowhydrogen electrodes because the very fluid blanket of liquid slag
obtained will interfere with the weld and will result in a defective
bead.
Preparation of the Pipe Joint. The standard joint specifications for

thin-wall pipes are given in Fig. 8-6. In the field, the sections of pipe
to be welded together have usually been beveled in a shop before
105
Chapter 8
~o~~r/
~(
~~ 1/16"
Courre5rofrhe Huban 8mrhen Co.
Fig. 86. Standard joint specification of a thin-wall pipe joint.
they arrive on the job. Should this not be the case, then it will be
necessary to cut the bevel with an oxyacetylene cutting torch and a
grinding wheel. The surface of the bevel must be ground smooth and
all traces of the tightly adhering oxide coating left by the oxyacetylene cutting torch must be removed.
Before any welding is attempted, the pipe joint must be thoroughly cleaned to remove all foreign matter, such as oil, grease, rust,
paint, dirt, etc. On pipeline construction in the field, this work is
usually done by a two-man crew working ahead of the welders.
On some jobs a second crew is sent to work ahead of the welders
in order to fit-up the pipes in preparation for welding. Line-up
clamps, such as are shown in Fig. 85, are used to align the pipes and
to hold them in place during welding. As already mentioned, these
clamps are not removed until a sufficient amount (50 to 100 percent) of the root bead is deposited, to avoid cracking.
For practice-welding, the short pipe nipples are aligned in the
same manner as for welding thick-waH pipes, as described in Chapter 4. However, in this case, the diameter of the spacing wire should
be lf16 inch, which is equal to the width of the root opening.
Welding the Tacks
When the pipe nipples are correctly spaced and aligned, they are
ready to be welded together (see Fig. 8-7). For welding the tack
welds and the root bead, an E601O, E60IOIP, or E701OA electrode should be used. When the wall thickness of the pipe is lJ8
inch, or less, a I/dnch electrode is used; for thicker-walled pipe a
%2-inch electrode is used. Low-hydrogen electrodes cannot be used
for downhill pipe welding. A very high current setting should be
106
Fig. 8-7. Aligning and tack welding two thin-waH pipe nipples using a piece of
bent wire as a spacer to obtain the correct root opening.
used when welding the root bead by the downhill method. As an

estimate, it should be between 100 and 140 amps DC for the nipples
used in practice welding.
The tack welding operation on the pipe nipples is shown in Fig.
8-8. Also shown in this illustration is the correct electrode angle that
should be used, 10 to 15 degrees. The arc should be struck in the
joint ahead of the weld and immediately lengthened to stabilize it
and to form the gaseous shield. As shown in Fig. 8-9, when this has
occurred, the arc is buried into the weld joint, meaning that it should
be pushed into the joint, with the end of the rod held against the joint
with a light prcjsure. Due to the very high current setting used to
deposit the tack weld and the root bead, the arc will continue to burn
even when the electrode is in contact with the base metal of the pipe.
Holding the electrode at the correct angle while it is buried in the
joint, the welder must observe the formation of the puddle and the
keyhole. As soon as sufficient metal has been built up to bridge the
gap opening, he must begin to mOve the electrode by sliding its end
107
Chapter 8
OF WELDING
TOP VIEW
Fig. 8-8. A. Welding the tack weld; B. Top view of pipe nipples showing the
correct electrode angle for tack welding.
in the groove, constantly maintaining a slight pressure on the end of

the electrode to keep it buried. The electrode may have a tendency
to stick as it is dragged along in the groove, and if this occurs the
electrode holder should be wiggled slightly, while keeping the end of
the electrode buried in the groove. A small keyhole will form behind
the electrode; and this should be watched as the weld progresses, as
well as the molten pool of metaL When the bead is about % inch
long, the arc is quenched by a quick flip of the electrode away from
the keyhole.
108

PUSH ELECTRODE
INTO JOINT
CoUrft\"i)' of lh(~ Uohar{ Brothers Co.
Fig. 8-9. Position of the electrode for downhill welding a root bead. The
electrode is buried in the pipe joint by holding it in the weld wi,.th a Jig"t
pressure. The high current setting allows the arc to be maintained when the
electrode is in contact with the pipe.
It is rather difficult for a beginner to master this method of

welding; therefore, this procedure should be practiced by first welding some flat plates before any attempt is made to weld pipe. It is
best to begin practicing with the plates in the vertical, or 30,
position as in Fig. 8-10. Before starting to practice-weld on flat
plates, the following section, "Welding the Root Bead," should be
read.
Courll'SF (4Ihl...) lJohdrr 8roou..',.-.\- <:
Oc
Fig. 8-10. Set.up for practice-welding root beads by dragging the electrode in a
downward direction. A. Side view; B. Welder's view.
On cross-country pipelines, the position of the tack welds and the

procedute used to make them depends upon many factors.
Moreover, different crews may vary the procedure. When the pipe is
lying in a horizontal position or on a uniformly rising grade, it is
109
Chapter 8
lifted upward slightly by a crane to allow the welder to reach the
bottom of the pipe. The first tack weld then is made in the 6 o'clock
position. After the pipe is lowered, the second tack weld is made in
the 12 o'clock position, followed by two additional tacks in the 3
and 9 o'clock positions. If the ground on which the pipes are lying
curves upward or downward. the first tack weld is sometimes made
in the 12 o'clock position, followed by a tack at the 6 o'clock and
then at the 3 and 9 o'clock positions. For practice welding, the pipe
nipples should be clamped in the stand in the position shown in Fig.
8-11, with the tacks in the 2,5, 8, and 11 o'clock positions.
Welding the Root Bead. In preparation for welding the root bead the
tack welds should be deslagged and cleaned thoroughly with a wire
brush. The ends of the tack weld should be ground to a sharp feather
edge with a grinding wheel, as shown in Fig. 8-12.
The procedure for welding the root bead is essentially the same as
that for welding the tack welds. A high current setting is used,
together with the same types and sizes of electrodes. The electrode
CLAMP
~
~
TACK
TACK
LOCATIONS
COU,ln.\ of th" Jl"hart B'OII"',.1 Cu,
Fig. g1 1. Method of clamping the pipe nipples in the pipe stand.
should be held at the angle shown (10 to 15 degrees) in the

illustration, Fig, 8-13.
To start the root bead, the arc is struck in the 12 o'clock position.
It should be struck ahead of the weld in order to preheat the part of
the joint in which the bead is to be deposited. A long arc is held until
the arc has stabilized and the gaseous shield has formed and then the
electrode is buried in the welding joint and kept in place with a light
pressure.
The welder must closely observe the formation of the puddle of
molten metal and the keyhole. When the liquid base metal and the
filler metal from the electrode have formed a bridge of molten metal
across the gap, the welder then will start to move the electrode by
dragging it in the groove. A small, crescent-shaped keyhole, shown
in Fig. 8-14, will appear at the top side of the electrode.
110

The keyhole is a necessary element in the formation of the weld
bead. It is formed when the faces of the weld joint have melted and it
provides the space for the molten metal to flow to the bottom of the
weld joint, thereby assuring that the weld bead will penetrate to the
root. The solid weld bead should extend below the surface to form a
slight crown that must not exceed V16 inch in height (see Fig. 8-15).
The arc force will cause Some of the molten metal to flow upward
to build the weld bead and to retard its downward movement along
the pipe joint. By watching the action of the molten metal and the
GRIND SIDES TO
SMOOTH ROUND EDGE
GRIND ENDS TO SHARP

FEATHERED EDGE
APl'ROX. 1/16"
COU'I"')' of rhe Hoban Bymhers Co.
Fig. 8-12. Tack weld ground to facilitate making a tie-in.
VERTICAL AXIS---t
Courre.'J" of rhe Hobo,t Brothn.' Co.
Fig. 8-13. Correct electrode angle for welding the root bead using the dragging,
or buried arc. method.
111
Chapter 8
keyhole, the welder can determine the speed of travel at which the
electrode is advanced.
Certain difficulties are encountered when this welding procedure
is used. For example, there will always be a slight amount of
undercutting at the edges of the weld, as can be seen in Fig. 8-15.
However. the undercut is not harmful to the final quality of the weld
.. $MAU IC;IYHQlI SHOUlD $HOW ON
to; '510
Of UfcnODf
"'0 &l[TA,- FlOWING

~l'.,JID
~JPW,Il,RO
SIIlA,U,C'*SC'tu,s.o-IAF'{D
l(UHOl.~
t:1.fCTlI(lf)f
~N()
COIlr/t'JY of lilt' Hobart Bralhers C",
Fig. 8-14. A small crescent-shaped keyhole that appears at the top side of the
electrode when downhill welding the root bead by dragging the electrode in the
weld joint.
FLUSH TO 1/16"
\~~
UNDERCUTTING
ROOT PENETRAnON
Courtesy of the Hobart Brothas Co.
Fig. 8-15. Undercutting of rooI bead caused by dragging the electrode to deposit
lhe bead.
in this case, because it will be removed by the second. or hot, pass, as

explained later.
The high current setting together with the very short arc that is
used with this welding procedure will cause more current to flow
through the electrode than when a longer arc is held, because the
resistance of the arc to the flow of the current is less. This will cause
the electrode to get very hot Often the heat will increase to such an
extent that the electrode coating starts to break down beyond the
end of the electrode. As a result, the arc will tend to blow and the
electrode will have a tendency to stick instead of gliding smoothly
112

along in the groove. If this occurs, the electrode should be oscillated
slightly from side to side, as shown in Fig. 8-16, left and right.
Slight variations in the root opening often occur on pipe joints
that have been fit-up properly. In some places the root opening may
Fig. 8-16. Sidewise wiggle of the electrode, to overcome its sticking when
downhill welding the root bead.
be widened very slightly while in others it is slightly narrow. If the

root opening is slightly narrow, the speed of travel and the electrode
angle also should be reduced slightly. In places where a slightly
widened root opening is encountered, the speed of travel should be
increased slightly.
Increasing the speed of travel, however, can cause pinholes to
form behind the are, usually at a distance of about lis inch. When
this occurs, the welder must swiftly tilt the electrode holder in the
direction of travel and then go back to the original position as in Fig.
8-17. Tilting the electrode causes the arc force to push some of the
weld metal into the void. The welder must be alert to recognize
pinholes and then immediately take the corrective action. Obviously,
if the weld has progressed too far beyond the pinholes, they cannot
be eliminated by this procedure.
As the weld progresses the welder must focus his entire attention
on his work. He must hold the electrode at the correct electrode
angle. He must watch the puddle and the build-up of the weld bead.
At all times he must pay careful attention to the keyhole. Should the
keyhole decrease in size, the welder must slow down the speed of
welding and decrease the electrode angle slightly. If this maneuver
does not enlarge the keyhole, welding should be stopped and the
current setting increased before restarting the weld.
The keyhole must not be allowed to become greatly enlarged as
this will result in excessive penetration or, possibly, bum-through, as
113
Chapter 8
Fig. 8-17. Method of closing pinholes that can develop behind the arc when it is
necessary to increase the welding speed. Swiftly tilt the electrode in the direction
of welding and back, as shown,
Co"rtesy of Iile f{ob"rl 8roll,er,1 ('0,
Fig. 8-18. Burn-through resulting from excessive current flow when dragwelding the root bead.
shown in Fig. 8-18, If burn-through occurs, the defective section

should be removed before continuing to weld the root bead. The
welding current setting should be reduced slightly when restarting
the weld to prevent a recurrence of burn-through.
Stop and Restart. When it is necessary to stop the weld before the
bead is completed, the arc is broken by quickly flicking the end of
114
THIS PREVENTS
SLAG FROM MIXING WITH
UNSOLIDIFIED WELD
METAL (SLAG INCLUSION}.
COHrJt'-'.r (~( thl! Uo!Jarl Brv{hi'r.'i ('0.
Fig. 8-! 9. Method of breaking the arc by quickly flicking the electrode downward. away from the crater.
the electrode downward, away from the keyhole, as shown in Fig.

8-19. This will prevent slag from mixing with unsolidified weld
metal to form slag inclusions.
Before restarting to weld the bead, the slag must be removed from
the bead and also from aboLlt 2 inches of the weld bead away from
the keyhole.
To restart the weld, the arc is struck at a distance of about % inch
back on the bead, and moved toward the keyhole, with a long arc
length. After the arc has stabilized and the gaseous shield has
formed, it is buried in the weld joint at the end of the previous bead.
When a pool of molten metal reappears, the weld is continued as
before.
Tie-ins. When the edges of the tack welds are feathered there is no
difficulty in making the tie-in. The electrode is dragged along in the
joint in the normal manner. As it approaches the end of the bead to
which the tie-in is to be made, the electrode is moved up the sloping
sides of the feathered edge. The welder must watch the molten pool
of metal, and when it blends smoothly between the beads, he must
reverse the direction of travel for a very short distance and then
break the arc by quickly withdrawing the electrode away from the
work.
Preparation for the Hot Pass. A finished root bead is shown in Fig.
8-20. The edges of the outside surface are undercut and particles of
slag arc buried along these edges. A high crown is usually produced
on the top of the root bead when it is welded by the downhill
method. This is caused by the temperature difference in the liquid
metal, which is cooler at the edge of the bead than it is in the center.
Filler metal that has not fused at the edge of the weld will tend to
115
Chapter 8
Fig. 8-20. A finished root bead deposited by downhill welding with the eleClrode
buried in the weldjoint. A. Outside of weld joint; B. Inside surface at the bottom
of the joint.
flow toward the hotter region of the pool, which is at the center.
Thus, the metal tends to gather in the center of the weld to form a
crown.
The slag coating must first be removed from the root bead. The
weld joint should then be ground to partially reduce the high crown
and clean out the worst areas of undercut and slag inclusion at the
edge of the weld to prepare these areas so that they can readily be
filled by the next pass, as shown in Fig. 821. The root bead should
not be ground excessively, and too much time should not be spent on
this operation.
After the root bead has been ground, it should be thoroughly
cleaned with a rotary wire brush to remove as much of the entrapped
slag as possible. Any defects that remain will be corrected by the hot
pass.
The Hot Pass. The primary objective of the second pass, usually
called the hot pass in downhill pipe welding, is to burn out the
remaining slag and to complete the edge fusion between the root and
the base metal. Only a small amount of metal is added by this pass.
As mentioned earlier in this chapter, the weld joint should be warm
when this and all subsequent passes are deposited. The remaining
passes should be made as soon as possible after the previous passes
have been completed.
The current setting used to deposit the hot pass should be slightly
116
Welding ThinWall Pipe
Fig. 8-21. A ground root ready for welding the hot pass
higher than for welding the root bead. While an exact recommendation cannot be given, the current setting should be in the range of
110 to 150 amps DC. The same type of electrodes used to weld the
root are used here, namely, lfs to %2 in., E601O, E6010IP, or
E7010-A. The electrode angle at which the electrode is held should
be 10 to 15 degrees in all positions around the pipe.
In order to make a sound weld joint, each bead should start and
end at a different position from the bead over which it is deposited.
This means that the hot pass should not be started exactly at the 12
o'clock position, but at about 1 inch away, on either side of this
position.
To start the weld, the arc is struck in the usual manner by holding
a long arc until it has stabilized and the gaseous shield has formed. It
should be struck ahead of the weld and then brought back to the
starting point in order to preheat the metal over which it is to be
deposited. At the starting point the arc is carefully shortened to the
normal arc length, which, for the hot pass, is :~12 to 1/8 inch. At this
point the electrode should be held at the correct angle, or 10 to 15
degrees.
The normal arc is held at the starting point for a few moments to
allow the molten pool of metal to form. The weld is then started by
immediately starting to "whip" the electrode as shown in Fig. 8-22.
117
Chapter 8
The purpose of the up-and-down whipping motion is to control the
weld puddle and to force some of the liquid metal to flow into the
corner of the weld, thereby filling the undercut. To reduce the
fluidity of the puddle and to prevent overflowing, the whipping
motion should be about I % electrode diameters long and made in
~\AUQSE
AT OOTS
'" \',\
\
~~
--'---.---L
~
\\
'~, \ '\
'\
\'~,
~,\
'"
DOWN 1h ... \',\ \

ELECTRODE \~, \ ,\
BACK UP
ONE
DIAMETERS~lECTRODE
DIAMETER
ELECTRODE
Courl6Y of Ih~ Hoharl 8rvIJr~'$ Cu,
Fig. 8-22. Whipping procedure used to weld the hot pass.
the direction of welding without a change in the arc length. While

whipping, the arc should be made to pause in order to deposit filler
metal in the crater.
The entire hot pass is deposited using the whipping procedure. As
the bead is being deposited, the welder must maintain the correct
electrode angle in all positions around the pipe. He must watch the
formation of the bead to see that the edges are well filled with
deposited metaL In this step he has the assistance of the higher
current flow which burns out the slag and promotes edge fusion.
When necessary, the arc is quenched by quickly flicking the
electrode downward, away from the weld. Before restarting the
weld, all of the slag must be removed from the crater and from the
second pass for a distance of about 1 inch behind the crater. The arc
should be struck on the root bead about 112 inch ahead of the crater,
as shown in Fig. 8-23. After the arc has stabilized and the gaseous
shield has formed, the arc is brought into the crater and shortened to
the normal arc length. Here it is held until the crater is wen filled
with molten metal. The weld then can be continued by whipping the
electrode as before.
There should be no difficulty in making the tie-in at the bottom of
the joint. The first half of the deposited layer should be deslagged
and cleaned before the weld on the second half is begun. As the weld
deposit approaches the first layer at the bottom of the joint, the
welder will not slow down the speed of welding. He must watch to
118
Welding
Thin~Wall
Pipe
CRATER
STRIKE ARC HERE
CO'lrr,'.'}' of [he f/oo",rr Bm/her", Cu.
Fig. 8-23. Restarting procedure for the second, or hot, pass.
see that the molten pool of metal fills up to blend together neatly
with the first layer and he can then break the arc by quickly flicking
the electrode away from the pipe.
The Intermediate Passes. After the hot pass has been deposited, one
or more intermediate layers are deposited to fill the weld joint with
sound weld metal. The last intermediate layer should be flush, or %2
inch below the top of the weld joint.
Since the objective of making intermediate layers is to fill the weld
joint, slightly larger electrodes should be used, the actual size
depending on the thickness of the pipe wall. For the practice pipe, a
5/32 -inch electrode is recommended as more current can be used with
the larger electrode. The recommended setting is approximately 130
to 150 amps DC. Again, the electrode angle is 10 to 15 degrees,
which should be maintained at all times. The arc length should be
maintained at about one .electrode diameter.
Before each new layer is deposited, the slag coating must be
removed from the previous layer which should also be cleaned with
a wire brush. When starting each intermediate layer, care must be
taken to begin in a different place from the start of the layer below.
While the same welding procedures are used to weld all of the
intermediate layers, the procedure does change in different positions
around the pipe. These positions are; fiat, 11 to 1:30 o'clock;
vertical, 1:30 to 5 o'clock; and overhead, 5 to 7 o'clock. An
especially difficult area to weld on the pipe is that between the 2:30
and 4 o'clock positions. On this portion of pipe, a different tech~
nique must be used to deposit the bead.
Welding in the fiat position at the top of the pipe presents no great
difficulty for an experienced welder. As shown in Fig. 8-24, the bead
is deposited by manipulating the electrode in a loop-shaped weave.
119
Chapter 8
PAUSE AT DOTS --tlf1"A~""~I\~
Courtesy of the Ifubart Bra/has Co.
Fig. 8-24. Loop-shaped weave used to weld the intermediate layer at the top of
the pipe.
The "diameter" of the loop should be approximately 1112 to 2 times

the diameter of the electrode so that a gaseous shield is maintained
over the molten metal at all times. When the arc is in the puddle, the
electrode should pause to allow time for the addition of filler metal.
Manipulating the electrode smoothly, this procedure is continued
until the bead has reached the position where the weld starts to
assume the characteristic of a vertical (3G) weld. At this time the
weaving procedure must be changed to a slant weave, as shown in
Fig. 8-25. The weave should advance downward about one electrode diameter per stroke, and the electrode should pause at the end
of each stroke in order to obtain good fusion at the edges of the
weld.
When the weld is at approximately the 2:30 position, the molten
metal has a tendency to ron downward at a faster rate than before.
To counteract this tendency, the speed of travel must be increased
Courtesy oj' the Hobart Brothers Co.
Fig. 8-25. Slant weave used to weld the intermediate layer at the side ofthe pipe.
120
Fig, 8-26, Electrode angle used to counteract the tendency of the mollen metal
[0 roll at the side of the pipe,
slightly while weaving is continued as before, and the electrode angle

must be increased to 20 to 25 degrees, as shown in Fig. 8-26.
Changing the electrode angle is very important in order to control
the'puddle. By holding the electrode at the steeper angle and keeping
the arc at the edge of the crater, the arc force helps to support the
puddle by forcing molten metal near the arc to flow upward. It is
also an important practice to maintain a sm09th and even motion
while weaving, keeping the arc ahead of the downward-moving
puddle,
When the tendency of the puddle to roU downward lessens noticeably, the welding speed can be slowed down and the electrode
angle changed back to the normal 10 to 15 degrees. Usually, this
occurs near the 4 o'clock position. The slant weave is continued until
overhead welding starts near the 5 o'clock position.
On reaching the overhead welding position the molten metal will
naturally have a tendency to drip downward out of the weld. It is
held in the weld joint by surface tension, assisted by the arc force.
However, if the puddle is permitted to get too large and the molten
metal too fluid, it can drip. To overcome this tendency and to
control the puddle, a horseshoe weave, Fig. 8-27, is employed.
The objective of this weave is to reduce the size of the puddle and
to decrease the fluidity of the molten metal, while at the same time
promoting good edge fusion. The electrode is moved in the semicircular pattern shown in Fig. 8-27, with a pause at the end of each
121
Chapter 8
COlU({J,~y pI lhi' l/obuYt Brolhen (().

Fig. 8-27. Horseshoe weave used to weld the bottom of the intermediate layer.
weave to melt the edges of the joint and to deposit filler metal there
in order to obtain good edge fusion. While pausing at one edge of
the weld, the molten metal at the other edge solidifies, and at the
center of the weld it either solidifies or becomes very mushy, thereby
losing its fluidity. A smooth wrist motion should be used and the
electrode advanced about one electrode diameter for each weave.
In the regions of the joint roughly bounded by the 2 to 4 o'clock
and the 8 to 10 o'clock positions, the speed of travel must usually be
faster than at the top and bottom of the weld, in order to control the
puddle. Usually, for this reason, less metal will be deposited in this
region. Thus, after depositing the next to the last layer, the weld joint
will not be as thick in these regions as it is on the top and bottom of
the weld.
The joint must be evenly filled with weld metal before the cover
pass is made. Therefore, additional beads, called "stripper" beads,
will frequently have to be deposited on the sides of the joint, as
shown in Fig. 8-28. The same welding procedure is used to deposit
the stripper beads as that described previously for welding in the 2
to 4 o'clock positions.
Cover Pass. The cover pass must be a bead with a neat appearance
and having a slight crown (about 1!l6 in.) to reinforce the weld joint.
The electrode, arc length, and electrode angle will be the same as
those used to weld the intermediate passes.
Two slightly different weaves, shown in Fig. 8-29, are used to
122
Welding ThinWall Pipe

weld the cover pass. In each case, the weave must be wider than the
previous beads in order to cover the weld joint. At the end of each
weave the electrode should pause when it is centered over the edge of
the previous layer which will provide the correct bead width as well
as fuse the bead with the edge of the joint without undercut
As shown in Fig. 829A, from the 12 to the 5 o'clock positions, a
slant weave is used. For overhead welding from the 5 to 6 o'clock
positions a semicircular weave (Fig. 8-29B) is used. In each case the
speed of travel should be such that the weld metal will build up to
form the !fIG-inch crown. To avoid undercutting and gas pockets the
correct arc length, electrode angle, and speed of travel should be
maintained while the electrode is manipulated with a smooth movement.
12-0'CLOCK
--+If-
DO NOT ADD
A STRIPPER
BEAD WHERE
IT IS NOT
NEEDED
END VIEW
POSSIBLE STRIPPER BEAD WELD AREA
('mutt's}' of Ihe llobi1Y/ Brulh",s Co,
Fig. 8-28. Stripper beads are used to fill in the weld joint where faster welding
speed has resulted in the incomplete filling of the weld joint.
A
ELECTRODE MOTION
FOR 12-TO SOCLOCK
POSITIONS
B
elECTRODE MOTION
FOR i-TO 60ClOCK
POSITIONS
('olln"J)"
or 'h" Hohilr! Brotl"'fJ C",
Fig. 8-29. Weaves used to weld the cover pass. A. Slant weave used for 12 to 5
o'clock positions. B. Horseshoe weave used for 5 to 6 o'clock positions.
123
Chapter 8
Poor Fit-up. When a poor fit-up is encountered, the necessary
corrective action to be taken is similar to that used when welding by
the downhill method. If the root opening is too wide, nuggets are
deposited on both edges of the weld until a bridge is built across the
wide gap, over which the arc can travel to tie the two edges together.
The welding current should be reduced slightly when welding ovcrspaced joints. An important difference in the welding method occurs
in welding the root bead. In order to be able to carry the arc across
the wide root opening, the root bead should be started against a tack
weld and the bead should be deposited by using a loop-shaped
weave, as shown before in Fig. 8-24. After the root bead has filled
the root opening with sound weld metal, the intermediate beads and
the cover bead are deposited by using the same procedures as used
over a normal size root opening.
When the root opening is too narrow, the tack welds and the root
bead are made by dragging the electrode as previously described;
however, a higher current setting is necessary to obtain the required
amount of weld penetration. Usually the current setting should be
about 15 amps more than for a correctly spaced joint. During
welding, the welder should pay attention to the penetration
obtained. If overpenetration occurs, the electrode angle should be
decreased slightly and the speed of travel can be increased. When
insufficient penetration occurs, the speed of travel should be decreased. slightly and, if this does not help, the current setting must
be increased. Again, the remaining layers are welded in the manner
described for a correctly spaced root opening.
When the root face is too wide, the root bead is deposited by
dragging the electrode, using a sufficiently higher current setting to
obtain the required amount of penetration. For a narrow root face,
the electrode is dragged to deposit the root bead, using a current
setting that is about 10 amps less than normal to prevent over
penetration. If excessive penetration or burn-through occurs, the
welding current is further decreased.
Pipe Axis at 45-Degree (60) Position. On some jobs it is necessary
to weld a pipe joint with the pipes positioned at an approximately
45-degree angle. While the welding procedure used in this case is
similar to welding in the 5G position, there are still some differences
that will be described in the following paragraphs.
As shown in Fig. 8-30, the root bead is deposited by dragging the
electrode from the top to the bottom of the pipe, as before. Depositing the hot pass is also done by using the same method as when the
pipe is in the 5G position.
124
/
Courtesy
~r [he
'NOVIEI
Hobart Brothers Co.
Fig. 8-30. Method of downhill welding the root bead when the pipe is positioned
at 45 degrees. The electrode is buried in the pipe joint and dragged along the
pipe joint to deposit the bead.
Courtesy oflhe Hoban Brothers Co.
Fig. 8-3 I. Electrode angles for depositing the intermediate and cover layers
when downhill welding with the pipes positioned at 45 degrees.
Intermediate beads are deposited in a slightly different manner

than in the case where the pipes are in the 5G position. First, the
electrode must be held so that it points toward the high side of the
joint, as shown in Fig. 8-31. With respect to a perpendicular line
from the surface of the weld, the electrode angle in a plane cutting
through the weld joint should be 10 to 15 degrees, as before. The
side angle should be 25 to 30 degrees, as shown; as it is in respect to
a plane passing through the weld joint. The current setting and the
125
Chapter 8
PAUSE AT DOTS
TRAVEL
Courtesy of the Hobart Brother! Co.
Fig. 832. Oval weaving motion us~d to de~osit the ,intermediate and ~over
passes when downhill-welding WIth the pIpes posilloned at 45 degrees.
arc length (one electrode diameter) are the same as used when the
pipes are in the 5G position.
In the flat welding position on top of the pipe, a loop or ovalshaped weave, as shown in Fig. 8-32, is used. The electrode should
pause where the dots are shown at the upper edge in order to deposit
sufficient metal there and to obtain good fusion against this edge. If
possible, this weave should be continued around the side of the weld
to the 5 o'clock position; however, in the vertical welding position
between 2 and 4 o'clock, the puddle may start to roJl downward
excessively. If this happens, the electrode angle should be increa'ied
to 20 to 25 degrees and the weave pattern changed to a slant weave
(see Fig. 8-25). It may also be necessary to increase the welding
speed slightly, in order to keep the electrode ahead of the puddle.
The arc force should be used as effectively as possible to hold back
the molten metal, also the electrode movement should be kept slow
and steady. When the overhead position is reached, a horseshoe
weave, similar to that shown before, in Fig. 8-27, is used to finish the
weld.
If necessary, stripper beads should be deposited on the sides of the
weld before welding the cover bead. The same angles as in Fig. 8-31
are used to hold the electrode for depositing the cover bead. This
layer is deposited entirely by using an oval weave, illustrated in Fig.
8-32. At each side of the bead, the center of the electrode should
pass over the edge of the previous layer, pausing at the upper edge to
obtain good edge fusion.
Downhill Pipe Welding-Heavy-wall and Large Diameter

The author originally planned to add a new chapter about downhill
pipe welding, addressing the changes that have taken place over the
126

years in the construction of cross-country pipelines, including the
needed changes in welding procedures. After considering ways to
incorporate the original text on downhill welding, it seemed the best
approach was to complement existing material with this section,
The alloys used in pipelines today have led to larger diameters and
increased wall thickness. Pipelines are expensive to construct. Longtenn planning indicates the necessity for future capacity, and it is possible that the pressure at which larger volumes of natural gas and crude
oil flow will also demand higher strength, ductility, and toughness of
the pipeline material.
The pipe materials most encountered today for pipeline construction
are the Lx60 and Lx65, as classified by the API standard. Tensile
strength, hardness, ductility, and toughness have become the focal
point of changes in the welding procedure. The piping materials are
classified as low alloy materials. Attaining this higher strength, ductility, and toughness will be risky in tenns of higher carbon and man
ganese content. Under bead cracking has been experienced when
working with Lx56, which contains carbon and manganese near maximum acceptable limits. This pipe grade has a lower strength than the
Lx65 and Lx60. Therefore, the composition of the steel used for
pipelines must not only factor in weldability under hydrogen-induced
conditions, but also the influence on toughness when the carbon and
manganese content increase.
Certain steels, and all of the higher strength steels, are produced by
some manufacturers by adding small amounts of columbium and vanadium to gain the required strength. These take the place of an increase
in the carbon content, knowing that carbon has a detrimental effect
upon fracture toughness. Fracture toughness is a much needed
mechanical property that helps prevent running cracks from developing. This property is quite important in pipeline steel, where the initia
tion of a running crack can lead to destruction of miles of pipeline.
Adding columbium or vanadium, or both, along with a relatively low,
control finishing temperature in hot rolling, produces a fine-grain,
stronger steel with adequate toughness.
The principle guide for welding pipelines is the API Standard 1t 04,
a document written by representatives of the API, AGA, PLeA, AWS,
and SNT. This standard provides the requirements for obtaining weld
joints of adequate quality for gas and crude oil transmission pipe lines,
Or other high pressure services, using skilled welders and commercially-available material and equipment. API I 104 also outlines the methods of testing and the test requirement for qualifyied welding procedures, as well as materials for a specific kind and size of pipe joint.
127
Chapter 8
When a contract is awarded. the contractor is required to prepare a
document called a welding procedure. In this document, the contractor
stipulates all variables, methods, and an overall plan for executing the
work in order to achieve the objective of the client and responsible
agencies. The following items must be stipulated:
(I)
Material to be welded--ehemical or alloying ingredients
(2)
Welding process
(3)
Diameter-group wall thickness. group combination
(4)
Position-roll or fix position welding
(5)
Direction of welding
(6)
Number of welders
(7)
Time laps between passes
(8)
Preheating temperature
(9)
lnterpass temperature
(10) Filler metal classification
(II) Cleaning procedures (for the weld) and equipment used for
such purposes
All these items and many more must be substantiated by the contractor in accordance to the code and specifications related to the job, in
API 1104.
Depositing Root Bead-Heavy-WaU Pipe,

Large Diameter Pipe
These pipes are known for their higher strength. These low alloy,
higher strength pipe require the use of certain higher strength electrodes that include E-80 I0, E-70 10, and E-7048. The last of these is a
low-hydrogen electrode used for downhill welding. In all cases preheating to specified temperature of interpass temperatures is required.
The technique to weld the root pass with the E-801 0 electrode is similar to using the E-60 I0 electrode, except the movement of the electrode from side to side is slightly limited. However, the speed oftravel is directly in the hands of the welder. Therefore, the arc should not
be allowed to get far ahead of where the weld deposit is or more than
one sixteenth of an inch from where the fusion of the root openings is
taking place.
On-the-job, welders are equipped with grinders that use thin blades
(discs). Sometimes welders do not like the look of the edges of the
root bead. When changing electrodes and restarting the root pass it is
necessary to use the grinder to feather the leading edge of the crater.
They can also use the grinder for quick touch-ups. then resume welding. The method for cleaning, wire brushing, and grinding is the same
128

as previously described. After brushing, the welder should inspect the
finished root bead, looking along the fusion line on both sides, and
especially the extent of undercut. If undercuts along the fusion line
exceed IO percent of the weld, the welder may have to adjust the per~
sonal welding technique, as well as reduce the current setting slightly,
and slow the speed of travel when making future welds. Welder
helpers should be able to help ensure that the welder's need for higher or lower current is instantly met
Controlling the amount of undercut is important because, like the
electrode, the base metal has a high tensile strength. The undercut at
the fusion line raises the stress. If the (preheating) temperature is
allowed to drop, the stress in the areas of the undercut increases; the
weld is likely to crack because it fails to withstand the applied load.
Such stress also points to why preheating is so essential. Normally, if
such a condition presented itself during the welding of a carbon steel
pipe (undercut) with the same dimensions, it would not pose a prob~
lem, because the steel is ofa lower carbon content with high ductility.
Running or depositing root beads on today's pipe materials is quite
different. On pipelines where the undercut must be very limited, preheating must be applied and maintained during welding. Considering
the composition ofthe low alloy steel and its carbon content, there are
reasons to believe that ,without preheating and maintaining the
required temperature, the unfinished weld is likely to crack. Grinding
the surface of the root bead in preparation for depositing the hot pass
compounds the problem.
Depositing: The Hot pass

The hot pass follows the root bead after the root bead is prepared.
This weld layer is deposited with utmost care not to create an undercut, but a weld with proper fusion along the fusion line on both sides.
The third pass is deposited in a similar fashion, using a controlled arc
length and a very slight weave.
Filler Passes and Their Sequences

Filler passes that are deposited on heavy-wall pipes used in crosscountry pipelines are welded by using stringer bead sequences. On the
next page the diagram shows that the first three passes are single layer
deposits, each extending across the bevel. The second diagram shows
all the completed sequences.
Pipeline welds that use the stringer sequences often lack fusion on the
bevel surface. Actually, this is often because welders fail to maintain
the interpass temperature, which assists in having a fluid puddle.
129
Chapter 8
Figure 8-33A
Figure 8-34B
Figure 8-35C
Figure 8-360
Using the Proper Angle To Avoid A Lack of Fusion

B Slag Inclusion
C Porosity
However, the direction is downwards, and the puddle has the tendencyto run or sag in a downward direction. Therefore, welders have to use
their wrists, not their entire anns. They should tum their wrists so that
they are pointing the electrode as shown in Fig. 8-33A. Then they will
be able to have the puddle wash-up on the face ofthe bevel more effectively each time the stringer bead is deposited, as shown in Fig. 8-34B.
This technique also removes any possibilities of having deep
crevices. Such crevices are not easily corrected, even by depositing a
bead with a very high current. The proper method for correcting such
a problem is to use a grinder to slightly widen or open the crevice, then
deposit a stringer bead, with the electrode slightly pointing toward the
face of the bevel. This technique is very important for avoiding deep
crevice altogether, See Fig 8-35 C above.
The tenn stringer bead has different meanings in different situations.
For example, it may be used with a bead on plate test, when a stringer
bead is deposited on the plate as straight as possible. Or, when hardfacing certain steel, the alloying element that is used in the core wire or
the electrode coating should enter the weld and remain as hard particles, with limited dilution to the base metal. The hard particles become
densely clustered, offering the effects of a hard surface for the purpose
for which it was intended. By applying this metal using a stringer
bead-with a heat value slightly above the brazing temperature-the
welding is a straight line deposit, with no indication of a free puddle
washing up at the edges from side to side. However depositing the
stringer bead in pipe welding, the welder must see a definite indication
of the puddle washing up to the sides of the groove, and a gradual tran
sition along the edges of the weld, as shown in Fig. 8-36D. Using the
130

stringer bead, welders need to use a sl ight oscillation when traveling
downhill in order to achieve a homogeneous blending at the edges, and
prevent peaks and valleys in situations when a few stringers are
deposited side by side in such cases, the welding procedure can be
adjusted to provide a higher heat input to the joint or preheat the joint
area to a safe temperature.
The welder's position

is gradually
changing as he goes
around the pipe.
Therefore, his need~
at any point in
time is met by his
welder helper and
the attendent who
must be trained.
This transmission pipe line material is considered to be of low alloy high strength steel which
entails a welding procedure as folJows...Preheating, maintaining interpass temperatures, depositing stringer beads and low hydrogen conditions. Yet the welding crew is fast moving and very
well trained to meet the requirements of API.] 104.
131
Chapter 8
Welders who are properly trained will be able to complement the procedure. Therefore, they must be aware that the many adjustments they
make instantaneously control many variables. Failure of a pipe weld is
not always imminent, but success depends on the welder's ability to
make adjustments at the right time.
Pipe welding crews move quickly, and have very little room for
errors. But consideration must be given to the added activities that surround welding of higher strength piping, in accordance with the API1104 Code. Procedure must also conform to the involvement of preheating and interpass temperature. Good quality welds are not the only
requirement; preheating and interpass temperature are equally important to achieving the required toughness.
The high-strength, low-alloy pipe for gas and all pipelines requires
preheating before welding is started. Preheating must be exercised uniformly where heat is applied by two heating burners, one on either
side, and operated manually around the circumference, allowing the
outer and inner walls to be brought up to the required temperature
evenly in order to avoid stress.
Cover Pass
When the pipe joint is filled almost to the surface, a pipe weld is
ready to be capped, using the downhill technique and stringer beads.
The bevel edge should be visible on both sides to serve as a guide for
depositing straight stringer beads side by side, as the weld is completed. When the first stringer bead has been completed using the bevel
edge as a guide, the second stringer should use the inner edge of the
first as it's guide. The electrode should be placed directly over the
edge, allowing the puddle to wash up on the first. All other passes
should follow the same pattern.
Welding large diameter and heavy-wall pipe requires two welders,
one on either side of the pipe. They need to coordinate their starting of
the weld at the top ofthe pipe. The tie-in at that point should be smooth
and staggered two inches from the previous tie-in, as shown Fig 8-37.
An allowance should be made for the same procedure, staggering the
stringer beads on all filler passes or when depositing stringers for capping. (n addition, tie-in on the overhead can be difficult under certain
conditions. For example, when the filler pass lacks the proper level in
preparation for capping the weld. As the prceeding fill pass should be
slightly below the surface. Therefore, the welder is able to maintain a
constant speed of travel without having to worry about filling and capping at the same time.
132

As for the tie-in on the overhead when capping, the same practices
are applied at the top as on the bottom. Regarding the overhead position, (a) welders can slightly lower their current setting, but maintain a
smooth advancing arc at a continuous speed of travel, (b) one welder
should reach the stop point, four inches ahead of the second welder, and
(c) welders' view of the overhead position should be at 45 degrees; as
the arc moves away from them, and they should maintain the oscillation
process or movement. This allows for the slight wash-up at the edges of
the puddle. The focus from that angle of viewing allows a welder to
keep track of the build-up of the stringer surface, and to monitor the
edges of the deposit. It is now easier for the welder to keep a uniform
weld. Getting directly under the pipe with a clear view of the arc and the
puddle ahead actually can add to confusion. However, by looking from
the recommended angle, the welder can control the speed of travel,
while maintaining a clear view of the puddle edges. The build-up and
solidification pattem determines the speed of travel or any changes
made during the welding process.
This chapter has covered several important steps, among them filling
the pipe almost to the surface, and maintaining the straight bevel edges
oftne pipe. The straight edges can effectively be used as a guide line for
making or depositing the first stringer bead, keeping the cap as straight
as possible and uniform in width. The inner surface (edge) of the first
stringer can be used as a guide for additional stringers and filler. The
welder in training must continually strive for perfection, which is not
easily achieved. Important considerations include:
(a) The instructor should continually help welders improve their
efforts.
(b) The welders on both sides of the pipe must concentrate on making
proper tie-in from the start of their training.
(c) A complete root bead should have little or no undercut; ifundercut
does occur, it is extremely important to maintain the preheating
temperature.
(d) In some instances, porosity can be avoided by proper manipulative
practices.
(e) Wire brushing is the first stage of preparing the root bead surface.
Next, grinding removes the humps and other regularities on the
surface of the root bead. Overheating due to friction of the grind
ing disc on the weld surface should be avoided; it is indicated by
discoloration on the surface of the weld (black, blue, or brown)
caused by oxidation, which can lead to porosity when the hot pass
is deposited.
133
Chapter 8
Composition~based unweldability influences toughness. The steel has
a tendency to develop higher levels of hardness when the cooling rate is
faster then recommended. The heat~affected zone may not develop an
entire martensite structure that would have the highest susceptibility to
hydrogen embrittlement cracking. Nevertheless, this mix structure can
develop or display cracking.
Attention should also be given to welding at ambient temperature.
Circumstances often arise when the work is to be joined at a low tem~
perature. Naturally, the question should be raised whether satisfactory
results can be obtained by applying regular procedures. In addition to
the metallurgical effects of welding at ambient temperature, there are
other aspects of the joining operation that can affect weld quality under
these conditions.
Suppose a weld is to be made applying the downhill technique with
ambient temperature, or very low preheating temperature, on pipe
material that has .25 percent carbon, .90 percent manganese, and small
quantities of other alloying elements. The welder can make a valuable
contribution to a weld that is defect free, and adequate both in ductility and toughness. The procedure in this case is not to ignore preheating, but to use another format that will have a preheating effect. When
the first three passes are deposited, all of which are thin layers, the temperature built~up will be higher or equivalent to the required level.
This technique starts with two or three welders who expertly deposit
the root bead with minimum time lost and with no defects. On completion, the welder helpers immediately grind the restarts areas. These
two steps should be completed within 3-to-4 minutes. The hot pass
then begins with two welders working on either side, starting at the 3
o'clock position, downwards to the overhead position. The welder at
the top extends the hot pass from 12 o'clock to 2, and from 12 o'clock
to 10.
After the welders have completed their respective sides to the over
head, where the tie~in has been made, the welder helpers brush that
area. Time is important in completing those three passes, raising the
temperature above the preheating temperature within ten minutes. Care
must be taken that the root bead have no undercut The hot pass should
be properly deposited, fusing both sides of the bevel; likewise the third
pass.
Porosity can develop from inadequate heat input on the bevel surface.
Fortunately, this is not the case at this point. What is involved from
here on is that when the stringer bead is deposited on the bevel face,
the welder must supply ample heat input by angling the electrode
towards the bevel surface, and then oscillating the electrode in order to
134

have the fluid puddle wash-up on the third pass with adequate fusion,
as shown in the figure on page 130. Angling the electrode also supplies
sufficient heat to the bevel surface so that there is no cold surface in the
close vicinity. Because of this, the slag that enters the puddle will
quickly dissolve to form the weld surface coating. A fluid puddle indicates that there is fusion. The gases that enter the molten puddle will
be able to escape, avoiding porosity.
The success of this approach requires (a), taking advantage of time, (b)
using highly skilled welders, and (c) applying techniques that induce
heat input to maintain the interpass temperature effectively. In short,
angling the electrode by turning the wrist on the bevel is effective in
eliminating defects, even when the pipe is preheated.
When preheating is stipulated for a procedure, the temperature must be
maintained at the start of depositing the root bead, and continues even
when the root bead is completed, as well as the time when the root bead
is being brushed and grinded in preparation for depositing the hot pass.
Effective preheating is based on understanding variables such as lapse of
time between heating and starting the weld, the drop due to conduction,
known as heat sink must also be taken into consideration. Furthennore,
the wall thickness of the pipe also affects the drop in temperature. Thus,
the actual preheating temperature specified in the procedure is likely to
be higher when taking these variable into account.
135
CHAPTER
Horizontal Pipe Welding (2G)

When the pipes are in a vertical position, the welding procedure
used is essentially that which is used for horizontal welding and for
this reason this chapter is entitled, "Horizontal Pipe Welding."
Horizontal pipe welding presents a few difficulties that are characteristic of welding in the 2G position. The molten metal in the
puddle constantly tends to drip downward. The edge of the upper
pipe, when in a fluid state, also flows downward causing undercuts
and the development of a very large keyhole when welding the root
bead. Therefore, certain correct welding practices, as described in
this chapter, must be used to overcome this difficulty.
Fig. 9-1. Perfect pipe weld made in the horizontal l2G) position.
An illustration of a good horizontal weld made on a thickwall

pipe is shown in Fig. 9-1. The width of each bead or layer of weld
metal should not exceed approximately three times the diameter of
the electrode in order to avoid excessive weaving. For this reason
more layers of weld metal are required when welding pipes in the
horizontal position. After the root bead and the second pass are
136
Horizontal Pipe Welding

welded, the following beads are deposited in the manner of a
bricklayer laying bricks in constructing a wall. A foundation layer
is first welded in place and following layers are welded on top of it.
Instead of having a single bead, the cover layer is built up in this
fashion, using several beads.
In order to make high-quality welds, the welder must learn the
correct technique and then apply it with consistency. The weave
motion must be unchanging as the welder welds around the pipe.
The same arc length, speed of welding, and electrode angle must be
maintained. While welding the root bead the keyhole must be
watched and during welding of all beads the welder must watch the
pool of molten metal.
Preparationfor Welding. Most of the procedures used to prepare the
pipes for welding have been described in detail in Chapter 4 and
need not be repeated here. The reader should review these procedures before going on.
For practice welding, it is best to use two 7-inch-long pipe nipples.
The bevel must be cut to standard specifications and thoroughly
cleaned. The pipes are then placed on end on the welding table with
a bent wire spacer between them to obtain the correct root opening.
After they are aligned, four evenly spaced tack welds are welded in
the root of the welds as before,
HORIZONTAL WELDING OF THE ROOT BEAD
It would be possible to weld the root bead with the pipe nipples
placed on end on a welding table. However, it may prove awkward
to weld in this position, especially if the welding table is large. A
better method would be to clamp the pipe nipples onto a welding
stand, as shown in Fig. 9-2.
Fig. 9-2. Method of clamping pipe nipples in pipe stand for welding in the
horizontal position.
137
Chapter 9
The current setting should be adjusted so that it is just sufficient to
provide good fusion of the base metal and the finer metal at the root
opening. Usually, slightly less current is used when the weld is ,a
horizontal weld (2G position)' than when the pipe is in the 5G
position. Too much current can cause serious trouble in welding by
making the molten metal in the puddle and on the upper edge of the
weld difficult to control. A safe procedure before starting to weld the
pipe joint is to make a test weld on two pieces of scrap metal that
have been beveled, to determine the best current setting.
The electrode angle for welding the root bead in the horizontal (2G) position is shown in Fig. 9-3. This angle should be
ELECTRODE
Fig. 9-3. Correct electrode angle for welding the root bead in the horizontal
position.
Fig. 9-4. Correct arc length for welding in the horizontal position.
maintained when welding around the entire pipe. The side angle of
the electrode should not deviate more than 5 degrees from the
horizontal plane. Any further deviation can result in undercutting,
which, in turn, can cause cracking, especially on heavy-wall pipe.
For welding the root bead, the correct arc length is about %2 inch
above the edge of the root face (Fig. 9-4), which is shorter than that
used to weld the root beadin the 5G position. The weld should not
138

be started against a tack weld; usually it is started about 2 inches
away from a tack weld.
The arc should be struck in the joint ahead of the weld. As usual,
a long arc is maintained until it has stabilized and the gaseous shield
has formed. It is then shortened to the normal arc length (1132 inch
above the edge of the root face) and held in place until the keyhole is
formed. When this occurs the weld is started, using the whipping
procedure.
Figure 9-5 illustrates the whipping procedure used to weld the
ELECTRODE
TOP VIEW
Fig. 95. Whipping procedure used 10 weld the roo I bead in the horizontal
position.
root bead in the horizontal position. It is the same whipping procedure used when welding pipe in the 50 position. The arc is moved
out of the puddle momentarily to allow the molten metal in the
puddle to lose some of its fluidity. It is then returned to the edge of
the keyhole, or that part of the keyhole that is adjacent to the weld
bead. The arc is held in this position for a short period of time to
allow filler metal to be transferred from the electrode to the puddle
and to maintain the puddle in a liquid state. This action can be
described as repeated "whip and pause." The arc should not be held
directly over the keyhole, however, as this will cause excessive
penetration, or possible burn-through.
The object of whipping is, of course, to control the pool of molten
metal and prevent it from sagging. The length of stroke should be
about one or two electrode diameters. An excessive stroke length
should not be used as this will cause the gaseous shield to be
removed from the liquid metal with a resultant harmful effect on the
quality of the weld.
While welding the root bead attention must be paid to the keyhole
and the molten metal. If the keyhole shows signs of increasing, the
speed of welding should be increased slightly and the electrode angle
decreased slightly. By increasing the speed of welding the heat is not
139
Chapter 9
retained in one position as long as when welding at a slower rate and
the heat build-up in the weld is decreased somewhat. Slightly
decreasing the electrode angle (see Fig. 9-3) allows some of the heat
from the arc to escape through the root opening during whipping.
To control the keyhole. the welder should maintain a short (%2-inch)
arc length at all times. By not using the prescribed shorter arc length
the keyhole may become enlarged.
These measures. together with the whipping procedure, also help
to control the size of the molten pool of metal. If they fail to control
the keyhole and if sagging of the molten metal occurs, the weld
should be stopped and the current setting lowered before continuing
the weld.
The welder must avoid fatigue. If his arms become tired he is apt
to slow the speed of welding and he will have greater difficulty in
controlling the arc length. Fatigue can cause the welder to become
erratic in his manipulation of the electrode. If fatigue should occur,
it is best to stop work and rest for a few moments, rather than to
continue and make a poor weld.
With practice, and by following the instructions given here, a
good root bead can be made. Figure 9-6 shows the outer and inner
surfaces of a good root bead.
Stop and Restart. It is necessary to stop and restart the weld several
times when welding around the pipe joint. When welding a root bead
the arc is quenched by a quick stab through the keyhole. For other
beads the arc is reversed a short distance and then quenched by a
quick withdrawal from the weld, leaving a crater behind.
Before restarting the weld. all of the slag coating should be
removed from the end for a distance of about % inch. When welding
the root bead, the arc is struck in the weld joint % inch in back of the
keyhole. A long arc is maintained until it has stabilized and the
gaseous shield has formed. The arc is then brought to the end of the
bead and shortened. The welder then watches the development of
the puddle of liquid metaL As soon as it is large enough. and
certainly when it shows signs of sagging, he will start the whipping
procedure and continue to weld as before.
When welding the intermediate and cover layers, the arc should
be struck ahead of the crater of the bead being welded. After the arc
is stabilized and the gaseous shield has formed, the arc is brought
into the crater and shortened. The arc is slowly moved from side to
side within the crater several times until a pool of liquid metal has
formed. When the molten metal shows signs of sagging, the welder
must start to manipulate the electrode, using the weave pattern
described further on in this chapter.
140
Fig. 9-6. A perfect root bead deposited in the horizontal welding position. A.
Outside surface. B. Inside surface.
Making a Tie-in. Tie-ins must be made while welding toward the

keyhole and toward the heavy end of a root bead. The procedures
that are used in these cases are essentially the same as those for
welding in the 5G position, described in Chapter 5.
When welding toward the keyhole, the weld can be continued at
the normal speed of welding, using the whipping procedure. When
the keyhole begins to close, the whipping motion is carried over the
end of the bead to which the tie-in is being made. The welder must
watch the molten metal as it fills the space between the two beads
and when this space is neatly filled with metal the arc must be
quenched by a sudden movement of the electrode away from the
weld.
The procedure for welding toward the heavy end of a bead is very
similar, except that the speed of welding must be reduced somewhat.
This allows time for the heavy end of the weld to get hot enough to
obtain good fusion with the oncoming molten metal. As soon as the
bead that is being deposited gets close enough, the whipping motion
should be directed toward the heavy end of the bead to which the
tie-in is being made. It may even be advisable to pause momentarily
a few times when the arc is over the heavy end. Again, as the weld
metal begins to fill up the space between the beads, the welder will
continue the whipping procedure. When the weld metal forms a
good blend between the beads the arc is quenched.
Poor Fit-up. Poor fit-ups are also encountered when welding in the
horizontal (20) position. The welding procedures used when a poor
fit-up is encountered are nearly the same as those used in the 50
position. These procedures are explained in detail in Chapter 5.
When a wide root opening occurs, a bridge across this opening
must first be built by depositing several nuggets of weld metal on the
root face. These nuggets need not be welded perfectly; their sole
purpose is to form a bridge across which the arc can be carried to
141
Chapter 9
start the weld. Before the second half of the pipe joint is welded, the
bridge of weld nuggets and a short length (about % inch) of the
adjacent root bead must be removed by grinding or with a hammer
and chisel.
The bridge must be built in order to weld the tack welds in place.
When welding a bead across a wide root opening (which may be a
part of the tack weld or the remainder of the root bead), the current
setting should be reduced somewhat and a V~weave must be used.
The tack weld is started at the bridge. Contrary to the procedure
used for a normal root opening, when the root opening is wide, the
regular root bead is started at the end of a tack weld.
The V-weave used to weld the bead should be long enough to
carry the arc completely out of the puddle of molten metal. The
puddle must be allowed sufficient time to solidify completely before
the electrode is reversed and returned to the weld to deposit additional filler metal. When the arc is brought out of the puddle it
should be made to travel up the face of the bevel, away from the
edge of the bevel at the root face. If the arc is concentrated on the
edge of the bevel at the root face, the edge can easily be melted
away, thereby widening the root opening even more. As shown in
Fig. 9-7, the electrode should be held so that it points directly at the
pipe and the arc length must be kept short.
Difficulties can be encountered when welding the root bead across
the wide root opening. Often they are the result of unsteadiness in
manipulating the electrode while making the long V-weave. The arc
length may have been irregular or the pipe bevel around the puddle
DIRECTION OF TRAVEL
TOP VIEW
Fig. 9-7. A. U-weave used to weld a horizontal root bead when the root opening
is too wide. B. Top view of electrode angle used when welding with aU-weave.
142

may have overheated. It is then necessary to discontinue welding
for a short time to permit the weld to cool.
When the root opening is too narrow or is closed entirely, the root
bead can be welded by using a higher current setting and a normal
length arc. This will increase the heat input into the weld and the
penetration. The puddle must not be allowed to become too large or
it will sag. To prevent sagging, it may be nece.<;sary to resort to the
whipping procedure.
A narrow root face can be welded by reducing the current setting
and by using the U-weave to weld the bead. When the root face is
too wide, the current setting should be increased slightly. If the
puddle tends to sag, it may be necessary to resort to the whipping
procedure or to the use of the U-weaye.
Root Bead with Low-Hydrogen Electrodes. Although low-hydrogen

electrodes are seldom used to weld root beads, they can be employed
if the correct technique is used expertly. An entirely different procedure is used as compared to that described for welding the root bead
with E60 I 0 and similar electrodes.
First of all, poorly fitted joints must be avoided. The heavy flux
coating and the slow cooling rate resulting from the heavy blanket
will cause the weld metal to sag if the root opening is too wide.
The electrode coating must be dry and it must not be chipped.
Figure 9-8 shows the position at which the electrode must be held. It
is held at an angle of approximately 10 degrees with respect to the
horizontal plane, so that the end of the electrode points toward the
upper pipe. When welding, the end of the electrode may be lightly
dragged along the edges of the pipe. By positioning the electrode in
this manner, the deposit of weld metal may droop slightly; however,
it will very nearly have the proportion of a perfect root bead.
Low-hydrogen electrodes are not deeply penetrating and a short
arc must be maintained at all times. For this reason a slightly higher
current setting should be used. Moreover, the higher current setting
prevents the electrode from sticking as it is being dragged along the
edges of the pipe.
The arc should be struck ahead of the starting point and shortened as quickly as possible. It is then brought back to the starting
point, where the root bead is started as soon as enough molten metal
appears to form a puddle. A short arc must be used at all times but
the whipping procedure must not be used. The bead is welded by
lightly dragging the electrode along the edges of the joint while
holding the electrode as shown in Fig. 9-8. By following these
recommendations, an exceptionally good root bead can be welded.
143
Chapter 9
HORIZONTAL
(2G)
Fig. 98 Position of electrode

when welding a root bead in
the horizontal position
with a low-hydrogen electrode.
10 0 APPROX.
DIRECTION OF TRAVEL
The Second Pass. Before each layer of weld metal is deposited, the
surface of the weld must be deslagged and thoroughly cleaned. All
defective portions of the weld and extremely high humps should also
be removed.
Again, the tendency of the molten metal to sag under the
influence of gravity must be considered. The puddle must not be
allowed to get too large. To achieve a good result, the welder must
use the correct electrode angle, arc length, weave pattern, and speed
of travel.
Figure 9-9 illustrates the correct welding technique that is used to
weld the second pass. As before, a short arc length should be used to
Fig. 99 Correct welding

procedurefor welding the
second pass of a horizontal
weld. Note the slanted "loop"
weaving technique.
fiRST PASS
SECOND PASS
DJRECTION OF TRAVEL
>PAUSE
~~jt
SIDE VIEW
WEAVE TECHNIOUE

weld in this position. Instead of pointing toward the upper pipe, the
electrode should be held in a horizontal position, with an electrode
angle of 5 to 10 degrees. The current setting should be higher than
that used in the 5G position.
A slanted looped weave, shown in Fig. 9-9, is used to weld the
second layer. This weave resembles a handwritten letter "I" that
leans slightly toward the left. The electrode should be advanced
about one-half of its diameter for each weave.
When this weave is used, the metal has a tendency to flow in two
directions; the molten metal just deposited tends to trail the arc while
that portion of the puddle not in the arc vicinity tends to flow in a
straight downward direction. This combined action causes the
puddle to flow sluggishly in any direction and sagging can be
prevented if the electrode is advanced one-half of its diameter at
every weave. The weave, then, is an important factor in controlling
the puddle and in preventing it from sagging.
The width of the bead for the second pass is controlled by
observing the edges of the root bead. The molten metal should be
allowed to penetrate into the beveled surfaces of both pipes to obtain
good fusion and to prevent undercut. For heavy-wall pipe (% inch,
and larger) the width of each layer of weld metal should not exceed
three times the diameter of the electrode, when welding in the
horizontal (2G) position.
On low carbon steel thin-wall pipe it is possible to use a somewhat
larger weave, as shown in Fig. 9-10. In order to keep the puddle
from sagging, the arc is moved downward, at a slant, and back to the
upper edge of the pipe by weaving the electrode completely out of
the puddle. When the arc is returned to the upper edge it is held in
this position momentarily to allow the puddle to reform and the
molten metal to flow into the corner of the joint. Because the heat is
retained in the thin-wall pipe, the puddle usually does not freeze
completely and reforms very rapidly.
This technique cannot be used on heavy-wall pipe because the
Fig. 9-10. Longer weave used 0 weld second pass on low-carbon-steel thin-wall
pipe.
145
Chapter 9
heat is withdrawn from the weld more rapidly by the thicker metal
and the puddle cannot be maintained. Using this weave on heavywall pipe will result in incomplete fusion of the weld metal with the
base metal and the metal in the adjacent weld beads. One other
difficulty occurs when using this technique; when the electrode is
completely out of the puddle, the weld metal is left exposed to the
atmosphere. causing oxidation and porosity. This occurs very readily in high-alloy steel pipe and, for this reason, a large weave should
never be used to weld such pipes.
Third and Fourth Passes. The third pass is deposited on the bevel of
the lower pipe, as shown in Fig. 9-1 IA. A deep crevice is formed
above the third bead in which the fourth bead is deposited, as shown
m Fig. 9-11 R
3rd.
SIDE VIEW
Fig. 9-11 Procedure for weldingthird and fourth passes in the horizontal position.
146

The end of the electrode should point slightly (about 5 to 10
degrees) toward the lower edge when welding the third layer as
shown at the top of Fig. 9- 11 A. The electrode angle can be 5 to 10
degrees, as before.
The same slanted loop weave used to weld the second bead is used
to weld the third bead. In fact, this weave is also used to weld all of
the remaining layers because it enables the welder to control the
pool of molten metal while at the same time depositing a sound
bead.
When welding the third pass, care must be taken to prevent the
third pass from being deposited close to the upper pipe. An open
crevice must be left in which the fourth pass can be deposited. If this
crevice is too narrow, it may not be possible to weld a sound fourth
pass, as it may lack good fusion and have slag inclusions trapped in
the weld.
Precautions- maintaining uniform preheating practices and maintaining for interpass temperatures during welding. ( See page 140 for bead sequences. )
147
Chapter 9
When the third and fourth passes are being deposited, the arc
should be kept away from the extreme edges of the pipe bevel as they
are easily burnt away by the intense heat of the arc, resulting in an
undercut. The molten pool of metal should be allowed to wash up to
these edges when the electrode is brought close to the edges, but not
quite onto the edges. In this manner good fusion can be obtained at
the edges without undercut.
The fourth pass is deposited within the deep crevice that is located
over the third pass. As shown in Fig. 9-11 A, the side angle (50 to
7) of the electrode is very important when welding the fourth pass
because the force of the arc, or arc~force, must be used to help
maintain the puddle of molten metal in the desired place within the
crevice. If left to itself, the molten metal would spill downward, and
out of this crevice. The arc~force pushes the molten metal upward,
holding it in place.
The size of the puddle should be kept relatively small, about two
or three times the diameter of the electrode. Then, by using the
slanted "loop" weave with a steady speed of travel, a good-looking
and metallurgically sound bead can be deposited.
Fifth, Sixth, and Seventh Passes (Cover Passes). These passes act as
a cover pass when welding in the horiwntal (2G) position. On very
heavy-wall pipe the fifth, sixth, and seventh passes may be intermediate or filler passes, in which case they are welded in a manner similar
to that used to weld the third and fourth pass. However, in the
discussion to follow, it will be assumed that the beads to be deposited
are cover passes.
The cover passes must be neat in appearance and they should
form a crown, or reinforcement, of about 1J8 inch. As before, they
are deposited by building from the bottom up; Le., the fifth pass is
deposited over the third pass as shown in Fig. 9-12. This pass should
extend beyond the third pass about one electrode diameter at the
lower edge of the pipe. Although not desirable, a certain amount of
drooping of this weld will probably be unavoidable. It is important,
however, that there is no undercut into the pipe when welding this
bead.
The sixth and seventh passes are built up on top of the fifth pass,
as shown. When welding all of these passes, the side angle of the
electrode should be approximately as shown in Fig. 9-13. Again the
slanted "loop" weave is used to control tne puddle, which should be
two or three electrode diameters in size, and a short arc length
should be used. Since these beads must be sound and have a good
appearance, the welder should get into the most comfortable pasi-
148
Fig, 9-12. (Left) Disposition of intermediate and cover passes when welding
in the horizontal position. (Right) View of weld from inside.
Courresy of Hobarr Brothers Co,
Fig. 9-J 3. Correct side angle for welding the intermediate and cover passes
shown in Fig. 9-12.
tion possible under the circumstances, Manipulating the arc

smoothly and with consistency, the cover passes can be welded to
fulfill all of these requirements.
The technique just described for depositing the filler passes and the
cap, or cover pass, is the same for many types of welding, such as:
(I)
(2)
(3)
(4)
(5)
(6)
Carbon steel of all composition and heavy-wall pipe

Low~alloy, high-strength heavy-wall and thin-wall pipe
Medium alloy, both heavy-wall and thin-wall pipe
Heat-resistance steel (low- and medium-alloy steel pipe)
Steel for very low temperature services (cryogenic servicespipe)
High alloy steel (stainless steel pipe)
149
Chapter 9
Caution is needed at all times, It starts first with the preparation of the
base metal, in terms of oxyacetylene cutting and beveling, and second,
with the tacking (short weld) of the pipe to be welded.
Many times, when preparing edges of heavy-wall pipe by flame cutting, the fabricator either forgets or ignores preheating the heavy-walls
of low- and medium-alloy pipe. Without preheating, the effects of the
alloy on high carbon pipe leads to a hardening of the bevel surface.
This surface may then develop microcracks due to the very fast cooling rate. This effect is similar to other materials that are welded with
out preheating, though they should be preheated. With some pipe
alloys the HAZ{HeatAffected Zone) may not be affected ifno preheat
is used. Yet, it is important to use a grinder, and remove the oxide and
surface metal approximately .014 inches deep in order to avoid a hardened bevel surface.
The inside surface of the pipe edges at the root should be addressed
first, before tacking the pipe together A file or stiff wire may be used
to remove scale and rust. This will ensure proper fusion in the root.
Tacking the pipe together for welding must not be undertaken as a temporary weld. Instead, it will be part of the entire weld or root bead. It
should be given all the attention that allows it to be part of a perfect
weld.
High carbon pipe materials, as well as those of the alloy type with
the propensity to harden, should be preheated to avoid cracking of
those short tack welds and to avoid shrinkage stress cracking. Before
starting, the welder should examine both edges of the tacks to be certain there is no initial porosity or tear at the edges where the tack was
discontinued.
These elongated tack welds, besides being part of the initial root pass,
also serve to maintain the root opening as the root pass is subsequently completed. It is important for the welder not start or discontinue the
rest of this root pass on one of these tacks. Not only is there a good
chance for a cold lap to take place but the key hole penetration or reinforcement inside the pipe may not be continuously fused into that of
the tack.
When welding in the 20 position, weld the pipe joint by using the
stringer bead method, because stringer bead is smaller and is deposited at a faster rate. If the pipe material being welded has high hardenability, faster travel speed around the pipe ispossible so it will be
important to detennine if any stipulated preheat is adequate. When
alloy elements capable of inducing hardenability are used in two different welded joints-with one weld being made in the 5G position at
a speed of 6 inches per minute, and the other being welded in the 2G
150

position at 11 inches per minute using the stringer bead method-there
is a difference in the heat buildup in the joint.
When preheating is applied initially on hardenable steel, the interpass temperature is much higher. It should be maintained during weld
ing to prevent a hardened microstructure and a weld with a high residual stress, which can lead to cracking. Preheating and the interpass
temperature during welding slow down heat dissipation from the weld
and the surrounding area, thus providing both an allowance for hydrogen diffusion and a weld with the required toughness.
It is well known that hydrogen gas, above a certain level in the welding electrode coating, and through other ways of inducement, can be
observed in its atomic fonn in the molten metal during welding. In
molten metal, its solubility is high. But as solidification or freezing
begins, its solubility decreases as the temperature begins to drop. This
process is slow; most likely hydrogen in its atomic fonn will become
trapped. The second possibility would be for hydrogen to escape into
the heat affected zone during welding. The heat affected zone is within the high and low critical range, and is itself a spongy reserve waiting the inducement of (atomic) hydrogen. The problem, however, is
that hydrogen gas can find its way into micro fissures, slag pockets,
inclusions, and other type of voids. The accumulation of this gas in its
atomic form recombines to form molecules, and both the incubation
and the diffusion phenomenon begins to exert tremendous pressures to
produce cracking. Hydrogen and its activities alone may not cause
fai lures in welds, but can contribute to cracking when the structure of
a weld and its heat affected zone consist of hardened structure, high
residual stress, and low toughness.
Low hydrogen electrodes are baked at high temperatures to reduce
their moisture content. The material and clay used to manufacture
electrodes are analyzed for their moisture levels before they are
processed. Limestone is used in forming low hydrogen electrode coating; it gives out 40 percent of its weight in gaseous form, which protects the molten pool and the metal transfer with the arc stream. The
rest of its content forms a white powdery dust, which helps form slag
as well as being a fluxing agent.
The E-60 I0 electrode used for pipe welding is restricted to root bead
on certain low-alloy steels. However, its use on carbon steel is quite
popular; it can be used on any diameter size in the carbon steel category for completing the weld. Still, it is recommended that when the
thickness of the pipe wall exceeds half an inch, its root bead should be
deposited by the -60 I0 electrode, then filled and capped by the low
hydrogen E-7018 electrode. The reason is that a low hydrogen weld
N
151
Chapter 9
deposit is ductile. The E7018 electrode has a low hydrogen content
and its heavy coating is an excellent insulation to oxidation. In tenns
of low-alloy steel, with respect to the E-60 I0 electrode, the root bead
dillution will depend on the percentage of alloying element in the particular steel. The dilution of the weld deposit is given consideration
accordingly.
Parts of this text have described the effects of carbon and other alloying elements on steel. The discussion of carbon steel has considered
steel containing as much as .10 to .25 percent carbon. Above this
range, there is little doubt that precaution must be taken to plan a welding procedure that avoids cracking, brittleness, and hardening in the
heat affected zone.
The properties that obviously exert the greatest influence on the weld
are the propensity to harden when heated to a high temperature and
quickly cooled, the manner in which the hardness in the heat-affected
zone is controlled by the carbon content, and the ability to harden when
cooling was controlled by the carbon, manganese, and silicon content.
The fonnation of the martensite structure is also a factor.
The carbon range over which steel appears to face the greatest
changes appears to 00.25 to.50 percent. Below this range, there appears
to be little cause for concern about the hardenability of the steel, under
bead cracking, or a brittle heat affected zone, unless small quantities of
other carbide forming elements, such as columbium, vanadium,
chromium, and molybdenum, are present. When alloying elements are
added, they influence hardenability.
Higher strength, toughness, and other properties are achieved by
adding alloying elements and heat treatment. The welding engineer
must continually seek ways of welding steel without risk of cracking
and without impairing ductility and toughness. There are no simple
procedures or systems to predict the behavior of a steel or alloy pipe
during welding, although some progress has been made. Whatever the
reason for their presence, the addition of alloying elements should be
considered in any appraisal of the steel's composition from the standpoint of hardenability.
There is a comprehensive fonnula that takes into consideration the
carbon content and its influence on hardenability as well as the martensite structure that contributes to short range stress and becomes intensified as the carbon content increases. The formula also factors in other
alloying elements that add to hardenability, such as manganese, nickel, and silicon. This carbon equivalent (CE) formula is:
CE = %C+(MI) + (NI) + (Cr) + (eLI) + (MQ) + (V)
20
10
40
50
10
When CE exceeds .40, cracking is possible.

Welding most high carbon pipe material, and those that incorporate
alloying elements in the low~alloy and medium steel, requires contro)~
ling the cooling rate. As welding progresses, welds cool at a rate known
as the critical cooling rate, increasing the hardness of the heat affected
zone. The structure known as martensite is known for its hardness, poor
ductility, toughness, and short range stress, eventually leading to cold
cracking.
The application of supplementary heat to the joint before welding is
an important part of the welding procedure. It elevates the temperature
of the pipe immediately before welding, and is called preheating.
Preheating lowers the cooling rate as the welding arc moves along. For
a given set of welding conditions such as current setting, speed oftrav~
el, and material thickness, the cooling rate will be faster for a weld
made without preheating than with preheating. If the preheating is high,
the cooling rate will be slower. When a piece of metal heated to 11 OOF,
its terminal conductivity is half its capacity at room temperature. When
preheating is applied to a joint to be welded, the heat dissipation by
dilution from the heat affected zone and the surrounding area is lower.
Therefore, the slow drop in temperature allows for transformation, by
diffusion, to a final structure called perlite instead of the hardened
structure called martensite, which is more likely to form if the joint was
not preheated.
This treatment in the welding operation is due to a number of reasons,
the principal ones being as follows (a) to avoid cracking in the heat
affected zone, (b) to increase the toughness and improve the ability to
withstand adverse conditions involving impact loading at low temperatures, (c) to reduce residual stress and internal stress from shrinkage,
phase transformation and reaction to restraint, (d) to minimize shrinkage and distortion, (e) to prevent cooling at elevated temperatures that
would allow a ductile structure that is softer with higher toughness, and
(f) to slow down the cooling rate in order to give time for hydrogen to
diffuse from the hot weld, and thus avoid cracking.
When a welding procedure is stipulated as requiring preheating, it is
complemented by a stated interpass temperature as welL When work~
ing with an alloy steel, the interpass temperature has an effect on grain
refinement of the weld metal. The time between beads affects the extent
to which the grain size is refined. Depositing weld metal immediately
after another may result in no grain refinement of the weld metal.
Allowing the previous bead to cool to room temperature before
depositing the next bead will yield less grain refinement than deposit~
ing the next while the previous one is hot, but below the critical range.
153
Chapter 9
The significance of refining successive weld beads is reflected in the
notch toughness impact value.
An important question to consider asks what interpass temperature
should be maintained when welding an alloy pipe. Establishing appro
priate preheating and interpass temperatures depends on certain factors: (a) the carbon content and the alloying element in the particular
steel, (b) the temperature at which martensite structure begins, and (c)
the temperature at which 90 percent of the martensite structure has
formed. An interpass temperature should be maintained at a higher
temperature than that at which the martensite begins.
With regard to hardenability, a steel that is considered to be shallow
hardened will have a faster cooling rate, and a steel that is considered
to be deep hardened will cool at a slower rate. The alloying elements
in steel, starting with those most effective in increasing the hardenability, are carbon, manganese, molybdenum, chromium, nickel, vanadium, and silicon. These elements in various quantities will determine
the cooling rate, applicable to the particular alloy. In addition, heat
treatment plays an important role in complementing the needed
microstructure. Many formulas and booklets published by steel manufacturers and agencies establish other important criteria.
While the maniplitive techniques used when welding mild
steel are similar to those required when welding the alloy steels, the
welder must be aware of the unique problems that can appear as the
alloy steels are subjected to the heat of welding and then the subsequent cooling down. The job operation sheet as developed by the
welding engineer must be carefully followed. Not only in regard to
the proper joint preparation, but also in respect to preheating and the
maintaining of inter-pass temperatures.
In most cases, the root bead on medium alloy piping and other alloy
materials are welded by the gas tungsten arc welding process. This
process is considered hydrogen free. It protects the molten metal from
oxidation by the inert gas used for shielding. In most instances when
thin-wall pipe (alloy) is being welded-which requires just one or two
passes-the entire joint is welded by this process, Heavy-wall pipe is
most likely to have the root bead deposited by the gas tungsten arc
welding process, while the rest of the weld is completed by the shielded metal-arc welding process.
Alloy Steel in High Temperature Services

Welding at high temperatures demands that the weldment be made of
alloy steel because of two property limitations of iron and carbon steel,
154

which grow weaker as temperature rises. These are a decrease in
strength and an increase in corrosion and oxidation.
Chromium and molychrom steel, also known as chrom-moly, were
originally developed for elevated temperature services. They have been
extensively used in power plants, pressure vessel, heat exchangers, and
other applications where very high temperatures are involved.
Molychrom steel is used because it maintains strength at high temperatures and also resists creep, which means that it does not stretch or
deform under load after long periods of use with high pressure and temperatures. Carbon steels, on the other hand, tend to stretch at high temperature service; they will become brittle in time.
A number of compoSitions have become popular. The molychrome
steel is hardenable; therefore, the welding procedure must includes
postheat treatment. The preheating temperature depends on the specific
steel composition, thickness of material, and compacted joint designs.
Preheating and postheat treatment are important part of any welding
procedure. But welding in itself brings out certain important factors. For
example, molychrome has high strength and creep resistance.
Therefore, welds made on this alloy should not contain any defects such
as undercut, poor restart, or humps along the fusion line and edges,
where a smooth transition into the edges of the base metal is needed. [n
addition, after the weld is completed, the slag covering on the finish
weld should be properly cleaned by using a wire brush, preferably stainless steel. The reason for this is that chromium in steel combines with
oxygen to develop an oxide on the surface of the weld. As a protective
coating to prevent further oxidation and scaling, a completed weld must
be clean of its slag and wire brushed so that, within a given period of
time, the oxygen in the atmosphere will react with the weld metal and
the surrounding area to develop such a protection.
A critical aspect of the SMAW process is the proper angle of the electrode in relation to the work. The electrode angle should never be more
than 10 to 15 degrees off the perpendicular tangent to the pipe surface
as it travels around the pipe. It is this requirement among all others that
make the art of pipe welding so difficult to master. If the electrode
angle is excessive, the arc energy is not focused properly on the work.
In addition, and most importantly in the welding of the alloy steels,
excessive electrode angle causes erratic burning of the flux coating
which is often the cause of surface porosity and difficulty in maintaining the proper arc length. This occurs because as the electrode bums,
an extreme angle to the work causes unconsummed coating to hang
over the tip of the electrode much like a finger nail and actually drop
155
Chapter 9
off into the weld puddle. As 50 to 70% of the weight of an alloy electrode consists of supplementary alloying material contained in the
coating, flakes of the electrode coating falling into the weld puddle can
remain as permanent slag inclusions rather than enhancing the weld
metal as it was intended to do.
Welding Steel For Low Temperature Service (Cryogenic
Services)
Steelmaking plays an important role in securing the properties needed for cryogenic services. It produces steels with features such as (a)
fully deoxide steel (fully killed steel), (b) fine grain, (c) heat treatment,
quench and tempered (d) alloy with nickel, (e) minimal use of carbon,
and (f) reduction in the sulfur and phosphorus content to a lower level
than will normally be found in other alloys used at room temperature.
Steel that contains nickel, with controlled grain size, heat treatment,
and other features mentioned above, can improve fracture toughness.
Steel Nickel Content
Lower Service Temperature
2%
~75
3%
-150 F
5%
-200 F
9%
~20F
The increasingly strong demand for cryogenic service has focused

on the need for strength, toughness, and assurance against brittle failure when services temperatures decrease toward absolute zero or
lower. Construction welds for cryogenic services has become a specialized science.
When making welded pipe joints on the three category of alloys,
attention must be given to the preparation, which is similar to that for
material or alloys discussed for high temperature and pressure in service. The root bead must be uniform all around. Both pipes to be weld
ed should be properly aligned, with the internal alignment consistent in
terms of the internal walls. During the welding, all efforts must be
made to avoid undercut, poor restart, and deep crevices along the fusion
line closest to the bevel, In addition, the cover pass should not have a
high crown with a sharp transition into the base metal, but instead a
gradual transition. This will depend on the size of the electrode being
156

used and the technique. The root pass should be made with the GTAW
process along with pulsed current control. There should be no internal
defects. Any sign of undercut on the surface of the filler should be
ground away.
Nickel Steel of the 9% Grade

This grade of steel was initially developed to provide strength and
extreme toughness at very low temperatures. It would provide an alloy
with the required toughness and strength at very low temperature, 262F,
needed for tank that store liquified natural gas. These tanks are heat
treated and supplied as quench and tempered steel, with a maximum
thickness of two inches. The welds are not subjected to any heat treatment. Preheating is very seldom applied unless the temperature has
dropped to temperatures lower than 70F.
The electrodes used for welding this type of material are high nickelchrome iron electrodes specified as AWS-E-NIC, Fe-2 and E-NiCrFe-3.
The SMAW, combined with the electrode specified above, requires less
heat and the application of stringer beads. The flow of the puddle is not
as fluid as those from either low alloys or high alloys such as stainless
steel. In addition, the puddle will not wash up to the sides to any appreciable degree. Therefore, the welder will have to deliver the weld metal
in the area where it is needed by manipulation of the electrode. If the
electrode size is large, it may compromise the quality of the weld.
Supervisors should be cautious finding the appropriate size electrode to
use; a large size can end up requiring a lot of time grinding.
157
CHAPTER
10
Welding Complicated Pipe Joints

The principles of pipe welding have been treated in the last six
chapters. Much emphasis has been placed on small, but important,
details, such as the preparation of the pipe joint, striking the are, the
arc length. the electrode angle, and controlling the molten pool of
metal. Procedures for welding root beads. intermediate layers, and
cover passes have also been explained in detail. These same principles and procedures are used when welding more complicated pipe
joints, with some slight differences.
Welding Variations for Complicated Pipe Joints
In this chapter, emphasis will be placed on slight variations in the

procedures, which are reqUired for welding more complicated pipe
joints, as those illustrated in Fig. 10-1. This does not mean that
details such as the preparation of the pipe joint, striking the are, etc.
can be neglected. On the contrary. they are just as important as
before, and if the weld is to be sound throughout, they must be
performed in a craftsmanlike manner.
Root Beads. The foundation of any successful pipe weld is the root
bead. When welding more complicated pipe joints the same holds
true and procedures arc exactly the same as those described in
previous chapters.
The welder must determine whether to weld the root bead uphill
or downhill. Usually, uphill welding is preferred. By this method it is
easier to control the weld metal and to obtain a good weld, especially when consideration is given to the various directions in which
the weld must be deposited on some of the pipe joints. Moreover.
root beads deposited by the downhill method must be ground and
grinding certain parts of some of the more complicated weld joints,
such as in sharp corners, is difficult.
When welding a root bead on a complicated pipe joint, the welder
must also decide in advance what procedure to use. Usually. he
158
CourJexy of Ihe Hohart 8mrherJ Cu.
Fig. lOl. A variety of perfectly welded pipe joints.
should start to weld at the lowest part of the pipe Jomt. If this
requires overhead welding, the electrode should be advanced at a
slow, steady pace, with or without whipping the electrode. When the
joint requires welding in the vertical uphill or in the flat position, the
whipping procedure is used to control the puddle. Whipping is also
used to weld in the horizontal, or nearly horizontal, welding position; here the arc length must be shortened somewhat.
159
Chapter 10
This is but a brief review of root-bead welding. For more details.
the reader should refer to previous chapters. In the remainder of this
chapter, root bead welding is not discussed in detail; however, when
actually welding a more complicated weld joint this part of the weld
must be given as much attention as welding the filler layers. Failure
to do so will most certainly result in a defective pipe joint.
Intermediate and Cover Passes. There are also great similarities
between the procedures already described and those used to weld the
intermediate and cover layers on more complicated joints. A welder
who has mastered the principles and procedures treated up to this
point should have no great difficulty in learning how to weld the
filler passes on the pipe joints to be treated further on.
The intermediate and cover layers on the more complicated pipe
joints might be called weave layers because a weave is usually
necessary to control the pool of molten metal and to fill the weld
joint with sound metal. The weaves to be described in th~ following
pages have all beer: used to make higt--quality weld joints. Other
techniques have also been used, but they are not likely to produce
better welds. The beginner should first master the procedures
described in this chapter; then. if he chooses. he can experiment with
other techniques.
There are exceptions when the following I'rocedures should not
be used. These exceptions will be treated at the end of this chapter.
However, when these procedures are used. the welder must not
neglect the following details:
I. Remove the slag from the previous layers and clean the weld
joint thoroughly.
2. Supply adequate heat input by having the correct current
setting for the weld to be made.
3. Strike the arc in the joint, allow it to stabilize, then allow time
for the gaseous shield to form before the arc is shortened; continue
to maintain the correct arc length at all times.
4. Maintain the correct electrode angle.
S. Manipulate the electrode smoothly and maintain the correct
speed of travel.
Some of the joints will require welding in two or three directions.

In such cases the welder must be alert to change the electrode
manipulation and angle as required. He determines when these
changes are required by the behavior of the molten pool of metal. He
must avoid undercuts, sagging. and a bead of poor appearance.
160

The recommended procedures to follow are described by illustrations which show the weave pattern to be used as in Fig. 10-2. In
studying the illustrations of the weave pattern on the following
pages, the lines that describe these patterns have the following
meaning;
1. Heavy lines indicate a normal speed of travel, where filler
metal is, therefore, deposited.
2. Light lines indicate a faster speed of travel to avoid any
regular deposit of filler metal and, at the same time, to temporarily
remove the arc from the puddle in order to cool the molten metal
and prevent it from sagging.
'--....J-_ REGULAR DEPOSIT
Fig. 10-2. Designation of weave pattern symbols. Heavy lines: normal welding
speed; lighllines: faster speed of travel; dots: pauses.
3. Dots indicate a pause in the movement of the electrode in

order to heat the metal and to deposit filler metal. Often this is done
to prevent undercutting.
Uphill Welding with Pipes in an AngUlar (60) Position
The procedures shown in Fig. 10-3 are used when the pipes to be
welded are at an angle, or approximately in the 6G position. In this
case the joint is a butt joint. The bead is started by overhead welding
at the lowest part of the pipe. As the bead progresses around the
pipe, the welding procedure changes slightly. The welder must pay
careful attention to the electrode angle, which is 10 to 15 degrees.
and to the side angle, which is 15 degrees, as shown in Fig. 1O3B.
By poin ting the electrode toward the upper edge of the joint, the arc
force assists in preventing the puddle from drooping downward.
Also, by pausing when the arc is against this edge, sufficient metal is
deposited to assure good fusion and to prevent undercutting.
161
Chapter 10
Starting at the 6 o'clock position, the bead is welded by using a
slanted "loop" weave, as shown inside the circle in Fig. 10-3A. This
Fig. 10-3. Weave patterns used for uphill welding of pipejoints in angular (6G)
position.
weave allows the puddle to cool slightly, thereby preventing the

molten metal from dripping. A pause is made against the upper
edge.
Along the side of the joint, the weld will assume the characteristic
of an uphill vertical weld. As shown in Fig. IO-3A, a slightly
modified slant weave should be used here, while pausing at both
edges of the joint.
When the tendency of the metal to sag diminishes near the top of
the joint, the welder should revert back to the slanted "loop" weave,
pausing only at the upper edge. Both sides of the pipe are welded in
this manner.
If the wall thickness of the pipe exceeds % inch, a weave should
not be used. In this case, the joint should be filled by welding a series
of stringer beads, made by moving the electrode upward, at a slow,
steady pace without extensive weaving, as in Fig. ]0-4.
Full Lateral Joint
When two pipes are joined together as shown in Fig. 10-5, the
edges of the joint are beveled and a root bead is first deposited in the
usual way to close the joint. The intermediate layers and the cover
layers, however, assume the characteristics of a finet weld.
162

In effect, four separate joints, two on each side, must be welded.
Fig. 10-4. Stringer beadS deposited when welding heavy-wall pipe in the 6G, or
similar, position.
FULL LATERAL
FULL LATERAL
B
Fig. 10-5. Weave patterns used to weld a lateral pipe joint A. With intersecting
pipe on top of horizontal pipe; B. With both pipes in a horizontal position.
163
Chapter 10
The weld is started at the bottom of the joint and proceeds toward
the top of the horizontal pipe, where the tie-in is made.
In Fig. 105A, the intersecting pipe is located on the top of the
horizontal pipe. The weld going up the steeper joint is made by using
a slant weave, pausing at both edges. A slanted "loop" weave is used
to weld up the other joint. The puddle will have a tendency to droop
which is overcome by the slanted "loop" weave and by pausing at
the upper edge of the joint.
In Fig. 1O-5B, both pipes are in a horizontal plane. The bottom of
the pipe involves overhead welding which gradualiy changes to
vertical uphill welding. At the top of the joint the welding procedure
becomes essentially flat welding,
To start the weld in the overhead position, a slanted "loop" weave
is used to prevent the puddle from dripping. The electrode should
pause at the upper edge of the joint. When the tendency of the
puddle to droop diminishes, the weave pattern should be changed to
a modified slant weave, which is continued to the top of the joint.
TOP VIEW
DIRECTION 01" WELDING
B
Fig. 10-6. A. Weave pattern for welding aT-joint with intersecting pipe on
top: B. Electrode angle for welding the T.joint.
164

The welder must be alert when welding around the 3 o'clock
position 01;1 both sides of the joint. He must pay close att~ntion to the
electrode manipulation and, if possible, he should not stop the bead
in this position.
T..Joint with Intersection on Top
A typical T-joint is shown in Fig. 10-6. The joint must be very

carefully prepared by beveling the edges so that a good fit-up is
obtained. After the root bead is welded, additional filler beads are
deposited to fill the joint and give it more strength.
On both sides, the weld is started at the lowest part of the joint.
The tie-in of the beads is made at the two highest parts of the joint.
While welding around the joint, it is important to maintain the
correct electrode angles, which are shown in Fig. 10-6.
A slanted "loop" bead is used to weld the lowest part of the joint
in order to prevent the puddle from drooping. The electrode should
point slightly upward (see Fig. lO-6A) to allow the arc-force to assist
in controlling the puddle. To allow the bottom of the joint to cool
slightly, thereby reducing the tendency to droop, the electrode
should pause momentarily at the top of the weld joint. This also
provides additional metal at this part of the joint.
When the tendency of the puddle to droop decreases, the weave
pattern is changed to a slant weave and the electrode pauses again at
both edges of the weld. The slant pattern is continued until the bead
reaches the top of the joint.
Fig. 10-1. Weave pattern for welding T-joint with intersecting pipe below
horizontal pipe.
165
Chapter 10
T-Joint with Intersection on Bottom
The procedure for welding the T -joint shown in Fig. 10-7 is

similar to welding the previous T -joint, where the intersection is at
the top of the horizontal pipe. First a root bead is welded, starting at
the lowest part of the joint. This is followed by several additional
beads that also start near the lowest part of the joint; however, all of
these beads must overlap each other. A bead should never be started
at the same place where the next lower bead was started.
Starting the filler bead in the lower region of the joint, a slanted
"loop" weave is used since this portion of the joint is in the overhead
position. A slight pause is made at the upper part of the joint when
making this weave.
As the weld progresses around the joint, and when the molten
metal loses its tendency to drip, the weave is changed to a modified
slant weave pattern, as shown in Fig. 10-7. The electrode should
pause at both edges of the joint when this pattern is started. This
pattern is used until the layer is stopped at the top of the joint.
Fig. 10-8. Weave pattern for welding T-joint with intersecting pipe located at
the side of the vertical pipe.
T-Joint with Intersection at the Side
Another common orientation at which T-joints must be welded is

shown in Fig. 10-8. In this case, the weld must be made in three
welding positions, namely: overhead, vertical uphill, and flat, in the
166

order given. The overhead weld is made using a slanted "loop"
weave while for the other two welding positions, a slant weave is
used.
YJoint
A Y-joint is illustrated in Fig. 10-9. The pipes are first joined

together by a root bead after which the filler beads are added.
Welding is done by working from the bottom toward the top of the
joint. Three tie-ins must be made. One tie-in is made at the highest
part of the joint and two are made at the sides of the joint (one on
each side) at the point of intersection of the three pipes.
It is best not to start a bead at the point of intersection of the three
pipes because the heat will be withdrawn more rapidly at this point
than anywhere else. For this reason, one of the beads should be
continuously welded past this part of the weld. This bead should be
Fig. 10-9. Weave pattern for welding a V-joint
deposited first. The bead below the other pipe then can be welded up
to this point, where the tie-in is made.
This joint requires welding to be done in three basic welding
positions. A slanted "loop" weave is used to weld in two places
where overhead welding is required. The parts of the joint that
require vertical uphill and flat welding are welded by using a slant
weave, as shown in Fig. 10-9.
167
Chapter 10
Special Precautions
Portions of the pipe joints that require welding in the horizontal

(2G) position, in some instances, should not be welded by using a
weave, While the recommendations given in the preceding pages
usually can be used, they must sometimes be modified when it is
necessary to weld a bead in the 2G position.
When the weaving procedure cannot be used in the horizontal
welding position, the weld is made by depositing a series of
"stringer" beads in this portion of the joint. Stringer beads are
deposited by moving the electrode at a slow, steady pace, with very
little weaving. The situations where a weave pattern cannot be used
in the 2G position are listed below:
1. When welding high-alloy pipe joints
2. When using low-hydrogen electrodes
3. When welding very heavy-wall pipe joints.
Using a wide weave pattern to weld in the horizontal position will
momentarily expose the hot metal to the atmosphere. When welding
pipe having a higher alloy content, this pattern will result in oxidation and porosity in the weld metal. Welding stringer beads provides
better protection against oxidation and porosity as the gaseous shield
will be maintained over the hot metal. Therefore, stringer beads
instead of a weave should be used to deposit the bead when welding
higher alloy pipe in the horizontal welding position.
The weld metal deposited by heavily-coated, low-hydrogen electrodes is very fluid and cools very slowly. When the deposit is in the
horizontal welding position, it will tend to sag and entrap slag. Also,
the temporary removal of the gaseous shield resulting from a wide
weave pattern will cause the slowly cooling metal to absorb oxygen,
forming oxides. For these reasons, stringer beads should be used to
weld in the horizontal position when welding with low-hydrogen
electrodes.
Heat will dissipate more rapidly and the molten metal will solidify
more rapidly in thick-wall pipe than in thin-wall pipe. Moreover,
more heat is required to melt the metal in thick-wall pipes because a
larger amount of heat from the arc is lost to the walls of the pipe. To
overcome this difficulty the current setting is increased to provide
more heat input, which creates a larger pool of molten metal. In the
horizontal welding position the large pool of molten metal will tend
to sag and, if a wide weave is used, difficulty will be experienced in
preventing cold lap and undercutting at the upper edge of the weld
bead. Therefore, heavy-wall pipe should be welded with stringer
beads when in the horizontal, or similar, positions.
169
CHAPTER
11
Introduction to Welding Metallurgy

The materials used by pipe welders are primarily metals. It is,
therefore, logical that a better understanding of the properties and
the nature of metal will be useful in pipe welding and the objective
of this chapter is to enlarge upon this subject.
Metals have distinguishing characteristics that are important to
know. In the solid state they exist in the form of crystals and they
can be deformed plastically. All metals are good conductors of heat
and electricity, and have a metallic luster that readily reflects light.
A few metals, such as copper and aluminum, are used commercially in their pure form. Most commercial metals, however, are not
pure metals; they are usually a mixture of metals called alloys. For
example, brass is an alloy of copper and zinc, and the so-called silver
brazing alloys are mixtures of silver, copper, and zinc, with amounts
of cadmium and tin sometimes added. Most of the aluminum used in
industry is in the form of aluminum alloys containing manganese,
magnesium, silicon, zinc, or copper. Plain carbon steel is an alloy of
iron and carbon. Alloy steels contain additions of other alloying
elements (chromium, tungsten, vanadium, nickel, etc.) to enhance
their properties. The properties of complex metal alloys can best be
learned by first studying the basic concepts of metallurgy.
Properties of Metals
In the following paragraphs the basic properties of metals are

defined and discussed. However, the testing procedure used to obtain
values of these properties is not treated. Information on these
procedures is available elsewhere.
A characteristic of metals is their ability to resist large external
loads. Although, perhaps, it is not correct to speak of internal loads
in metals, it is a useful idea to consider stress an internal load. In this
sense, an extcrnalload applied at one point in a metal will result in
an internal load at many points inside the metal. It is really the
ability to resist these internal loads, or stresses, that provides metal
with the ability to resist externally applied loads.
170

Stress. If in Fig. 11- JA, a load of 1,000 lbs is acting downward on
the rectangular column, and if the area at the end of the column is 2
sq in., the stress on the bearing plate on which the column rests is
500 Ibs per sq in. (1,000 lbs -:- 2 sq in.).
Stress is defined as the load per unit area.

In countries using the English system of measurement stress is
always given in pounds per square inch (Ibs per sq in.). Where the
metric system is used, stress is given in newtons per square millimeter (N/mm2 ), where I lb is 4.44822 newtons.
There are several ways in which stresses act inside a piece of
metal. In Fig. 1I-I A, the stress is a compressive stress and, in
addition to the bearing plate, every particle of the metal inside the
column is subjected to a compressive stress. Usually other kinds of
stresses in combination are also present; however, for simplification,
it can be said that the stress in the column is compressive.
IOOOl8S.
lOOOLBS.
1000 LBS.
1000 LBS.
\....-)
1000 LBS.
IOOOl8S.
(01-1000
r1f\
. \ d7
D
Fig. I I-I. Stresses in metals. A. Compressive; B. Tensile; C. Shear; D. Torsional
(shear).
Chapter 11
In Fig. 11-1 B, a round rod is shown being pulled at each end by a
1,000 Ib load. If the cross-sectional area of the rod is 1 sq in., the
stress in the rod is 1,000 Ibs per sq in., and it is in tension. If we can
imagine a disc-shaped slice anywhere in the rod, this slice is being
pulled by the metal to which it is attached on each side. The stress on
the slice (any slice in the rod) is 1,000 Ibs per sq in. tension.
Figure 11-1 C shows another kind of stress, a "shear" stress. Each
section of the metal in the area of the punch and the die is resisting
the severance of the metal by the internal shear stresses that arc set
up by the load. When the internal shear strength of the metal is
exceeded, it will fail or shear off.
When a shaft, such as the crankshaft in Fig. 11-1 D, is subjected to
a torsional load, internal torsional stresses are set up to resist the
external load. Torsional stresses are not distributed uniformly across
the entire cross section of the part carrying the load. They are
greatest at the outside and are zero at the center of the round bar.
Actually, the torsional stresses are shear stresses; however, the
torsional strength is often reported separately and it is important in
design.
Strain. All metals behave elastically, like a rubber band, up to a
certain limit of stress. That is, they deform slightly when the load is
applied and when the load is released they snap back to their
original length. Technically, strain is the distance each unit length of
the metal is changed as a load is applied. In the English system it is
given in terms of inch elongation per inch length and in the metric
system it is millimeter per millimeter. Strain then alludes to the
elastic movement of the metal when a load is applied, whereas stress
alludes to the resistance to this movement.
Strength. The strength of the metal from which a part is made
depends on the load it must carry, its size, and its shape. Each size
and shape of a given metal can carry a different maximum load;
-therefore a standard size and shape are necessary in order to be able
to compare the strengths of different metals. Thus, a standard size
is established for test pieces and the strength of these test pieces is
re{X)rted in Ibs per sq in. stress. This value can then, in many
instances, be used to calculate loads non-standard sizes and shapes
will stand.
The strength of the metal is usually given in tables and often is
shown in stress-strain diagrams, such as Fig. 11-2. One must be
careful in reading strength values in handbooks because there are
two different values that can be reported, the ultimate strength and
the yield strength.
172

TENSiLE STRENGTH
ULTIMATE TENSILE
STRENGTH
YIELD STRENGTH
CURVE A
g
g:;
a.
<Il
~~L--ST-:-R~A:-:-IN:-:-:(:':':"jN""'.~P:::R:-:l:-:-N.:-l- - ; - -
t:i
REGION OF
~
PLASTIC DEFORMATION
REGION OF ELASTIC DEFORMATION
Fig. 11-2. Stress-strain diagram. Curve A is for ductile materials amI curve B is
for brittle materials.
Ultimate Tensile Strength. Most published figures on the tensile

strength of metals refer to the ultimate tensile strength. It is the
maximum tensile strength that the metal can withstand before failure
by rupture occurs. One difficulty is that this stress occurs in the region
of plastic deformation (see Fig. 11-2) and when the metal reaches
this stress level, it has already been permanently deformed. For this
reason an adequate factor of safety is required when using the
ultimate tensile strength in design calculations.
Yield Strength. In machines and structures the metal must not
deform permanently when subjected to a load. For this reason, the
yield strength is often reported. The yield strength is the stress to
which a metal can be subjected without permanent deformation. In
ductile materials the yield strength is at a lower stress level than the
~ltimate tensile strength, as shown for curve A in Fig. 11-2.
A typical plot of the stress-strain characteristic of a very hard and
brittle metal, such as hardened tool steel, is shown by curve B in Fig.
11-2. Such metals exhibit little or no permanent deformation before
breaking. For this reason. the yield strength has no meaning for very
hard brittle materials and the ultimate strength or breaking strength
is used.
Hardness. Hardness is defined as the resistance to penetration. It is
also sometimes defined as the resistance of the metal to plastic
deformation. Both definitions are essentially correct. Most methods
of measuring hardness do so by measuring the indentation made
when a specified load is placed on a small penetrator. This measurement is then converted to a suitable hardness scale or reading.
173
Chapter 11
Hardness is related closely to strength. Both properties involve the
ability of a metal to resist permanent deformation beyond the elastic
range.
Hardenability. This property is related to the depth below the
surface to which a metal can be hardened. It does not relate to the
maximum hardness that can be achieved in a given metal. There are
several standard tests for hardenability with, perhaps, the Jominy
End Quench Hardenability Test being the most popular.
Ductility. The relative amount that a metal will deform without
breaking is what is meant by ductility. As shown by curve A in Fig.
11-2, ductile metals tend to stretch considerably before they break.
This property enables metals to be bent. twisted, drawn out, or
otherwise changed in shape without breaking.
Ductility is generally defined as percent elongation or percent
reduction in area. However, these values provide only a rough
indication of ductility and cannot be used in design calculations.
Another rough measure of ductility is the bend test, whereby a
specimen is bent, either to fracture or through a complete
I 80degree arc.
Toughness. The ability of a metal to withstand a sudden shock is
called toughness. This property is determined by the energy
absorbed when a notched specimen is struck by a hammer blow
delivered by a swinging pendulum.
Brittleness. This property refers to the ease with which a metal will
crack or break without appreciable deformation. Brittleness is
related to hardness. As a metal gets harder its brittleness also
increases, and as the metal is made softer, its brittleness decreases.
An example of the stress-strain curve of a brittle metal is shown in
curve B, Fig. 11-2.
Malleability. Malleability is the property that relates to the ability of
metal to be permanently deformed by compression, usually by
rolling or hammering. Most ductile metals are also malleable.
Fatigue. This property refers to the ability of a metal to withstand
repeated or fluctuating loads. Fatigue failures always occur at a
stress level that is below the yield strength of the metal. Several
standard tests are used to measure this property. From these tests the
endurance limit of a metal can be determined, which is defined as
the stress below which the metal will withstand an indefinitely large
number of cycles of stresses without failure. Fractures due to fatigue
are often the result of sharp corners, scratches on the surface of the
metal, or tool marks.
174

Grain Structure
In the solid state, metals are in the form of crystals called grains.
Some of the grains can be quite large and can be seen by the naked
eye. However, most grains are very small and require powerful
magnification to be seen.
These crystals arc composed of a more or less orderly arrangement of atoms called a lattice structure. Each atom is in a fixed
position; that is, it oscillates about a fixed position.
Although the atoms cannot actually be seen, if it were possible to
magnify a crystal about 35 million times, it would be possible to see
the space lattice. Each lattice structure is composed of a number of
unit cells that are repeated over and over again to form the lattice.
Metals are composed of four basic unit cells which are shown in
Fig. 11-3. The lines between the atoms do not actually exist; they are
drawn to illustrate the geometrical arrangement of the atoms.
For example, chromium and tungsten have a body-centered cubic
structure. Iron, when below about 1666F, and steel, when below
1660 to 1333F, also have a body-centered cubic structure. Aluminum, copper, nickel, silver, and gold have a face-centered cubic
structure. When iron is above 1666F and plain carbon steel is above
1333 to 1666F, they have a face-centered cubic structure. Depend~
ing on its composition, steel may be partially face-centered cubic
and partially body~centered cubic between 1333 and I 666F. The
hard component of steel called martensite, that has been formed by
heating and quenching, has a body-centered tetragonal structure.
Actually martensite is iron which has carbon atoms trapped in its
structure to elongate the body-centered cubic structure into a
BODY CENTERED CUBIC
,,/"
FACE CENTERED CUBIC
./
./
BODY CENTERED TETRAGONAl.

CLOSE PACKED HEXAGONAL
Fig. 11-3. The four basic unit cells in metals.
175
Chapter 11
body-centered tetragonal structure. Indium and tin have a bodycentered tetragonal structure. Magnesium and zinc have a closepacked hexagonal structure.
It will be noted from the above that iron and steel have two
structures or phases, depending on their temperature. This important fact makes it possible to harden steel by heat treatment. When
steel is cooled slowly it will transpose from one structure to another
without difficulty, and is called "phase change." However. when this
is done rapidly, carbon atoms interfere with this change. and the
structure cannot assume its natural configuration at room temperature. As already mentioned, it is body-centered tetragonal (martensite) instead of body-centered cubic. This distorts the orderly
arrangement of atoms, producing internal stresses which cause it to
harden.
Anything that can be done to a metal that will disturb or distort
the lattice structure will cause it to harden. In the case of steel, the
lattice is distorted by a phase change brought about by heat treatment. Cold working a metal distorts the structure and thereby
hardens it. Note that cold working does not crystallize the metal; it is
always in the crystalline form. Inserting foreign atoms in the structure by alloy additions distorts the structure and hardens it. When
atoms are dissolved in a structure in the solid state and are then
precipitated out, the structure is distorted and hardened. This is the
mechanism used to harden aluminum. It is called age hardening or
precipitation hardening.
Metals are actually an aggregate of crystals, or grains, as shown in
Fig. 11-4. Each grain is surrounded by other grains, except at the
surface of the metal. When polished and etched with a suitable
reagent and then viewed under a microscope, the grains can clearly
be seen.
Fig. I 1-4. The grain structure in metals.
176

Depending on the heat treatment received by the metal, the grains
can have different sizes. Often the grains will be either large or small
throughout the entire structure of a piece of metaL In some cases,
where metal has been welded, for example, regions of large and
small grains will exist side-by-side. When a metal has been coldworked, the grains will appear distorted or flattened in the direction
of the COld-working (see Fig. 11-17).
The grains in an alloy mayor may not appear to be identical. In
many alloys, different kinds of grains exist side-by-side in a structure. The structure of a metal, when viewed under the microscope, is
called its "microstructure."
Structural Characteristics of Steel
The constituents in the microstructure of steel determine its properties, to a large extent. Some of the important constituents in steel
will now be examined.
Ferrite. (Fig. 11-5.) This is a solid solution of a very small percentage of carbon in iron. It appears at a temperature below 1333F and
the iron has a body-centered cubic structure. Ferrite grains appear
when the carbon content is less than about .8 percent in plain carbon
steel. The number of ferrite grains increases as the carbon content of
the steel decreases. It is the softest constituent in steel and as the
amount of ferrite increases, the steel becomes softer. In alloy steels,
some of the alloying elements may be dissolved in the ferrite as a
solid-state solution.
Cem~'!-t.~te. (Fig. 11-6.) Iron-carbide (Fe3C), a compound of iron
and carbon, is caHed cementite. It is very hard and wear-resistant. Its
composition will be varied when other carbide-forming alloys are
COUrTesy
oflhe Hoban 8ra/hers Co.
Fig. 115. Ferrite. The white grains are ferrite.
177
Chapter 11
present in the steel. Cementite may appear in several different ways
in steel. In Fig. I 1-6. it appears as a network surrounding the grains
in the region of grain boundaries and also within the grain, where it
is in the form of thin plates. Cementite may also appear as round,
roughly globular-shaped particles in steel that has been spheroidized. In their softened condition. tool steels and other high carbon
and alloy steels should be spheroidized because in this condition they
are easier to machine.
Courll'SY of rhe Hoborr Brarhers Co.
Fig. 116. Cementite. (Fe3CJ AIS(! called iron carbide. The white grains are
cementite.
(ourresy of the Hoban BrOlhers Co.
Fig. 11-7. Pearlite.
Pearlite. (Fig. 11-7.) This is a very tine platelike structure consisting

of thin plates of ferrite sandwiched together with thin plates of
cementite. The structure within the grains in Fig. 1]-6 is pearlite. In
Fig. 11-7. the white plates are ferrite and the darker plates are
cementite. When the steel is cooled slowly. the pearlite is coarse; it
becomes finer as the cooling rate is increased. Pearlite is a very
strong and tough structure that adds to the strength and toughness of
steel. The amount of pearlite in plain carbon steel increases until a
maximum is reached at approximately .8 percent carbon.
178
Martensite. (Fig. 11-8.) This structure results when steel with a

relatively high carbon content is cooled very rapidly. Small amounts
of martensite may be obtained by quenching a steel having a lower
CoUTtesy of the Hobart BTothers Co.
Fig. 11-8. Martensite.
carbon content. The addition of alloys to steel makes it possible to

obtain martensite at slower cooling rates. (Air-hardening tool steel
can be hardened by cooling in air.) Martensite has a body-centered
tetragonal structure in which carbon atoms are trapped. Martensite
is a very hard and brittle structure. Usually it is tempered, forming a
structure called "tempered martensite," which is body-centered
cubic. Tempered martensite becomes increasingly soft as the temper~
iog temperature is increased.
Widma.nstatten structure. (Fig. 11-9.) This structure OCCUrs in the
~.;ld zone of mild steels having a relatively low carbon content. It is
made up of white interlaced masses of ferrite distributed throughout
the grain, resulting from the formation of the ferrite in long, continuous plates. It is less ductile and has a lower impact strength than
ordinary ferrite.
CouTle"y of /he Hobart BTo/hers Co.
Fig. 11-9. Widmanstatten structure.
179
Chapter 11
Austenite. (Fig. 1110.) This is the face-centered cubic form of iron
occurring in plain carbon steels at temperatures above 1333F. It
normally is not stable at room temperatures; however, in highly
alloyed steels, such as tool steels and stainless steels, it can appear at
room temperatures. In steels, austenite can dissolve all of the carbon
that is present. This property is important in the heat treatment of
steels. At room temperature it has a good tensile strength and a
strong tendency to work-harden.
Spheroidite. (Fig. II-II.) This structure consists of many small
spheroidal-shaped particles of cementite dispersed in ferrite. It can
be obtained through several different heat treatments. When highly
alloyed steel and plain carbon steel are in the unhardened condition,
it is the preferred structure because of the marked improvement in
machinability obtained.
Bainite is another structure that can occur in steel. It may have
either a feathery or an acicular appearance. Bainite is produced by
quenching steel from an elevated temperature to some temperature
above about 400F (depending on the composition of the steel) and
"
-:,,,<~'t
Courlesy of the Hobart Brothers Co.
Fig. 11-10. Austenite,
Courrl'sy of the Hobart Bra/hers Co.
Fig. II-II. Spheroidite.
180

holding the steel at this temperature for a long time. It does not
normally occur as a result of welding.
The Iron-Carbon Diagram
A slightly modified form of the iron-carbon diagram is shown in

Fig. 11-12. The modification is in the naming of the constituents
where, in this case, the names used in this text are inserted in place
of the usual Greek names or symbols.
The iron-carbon diagram illustrates the temperatures at which the
different constituents in an iron-carbon alloy exist under equilibrium
(very slow heating and cooling) conditions. It is a most useful tool in
predicting the final structure of these alloys.
The principal iron-carbon alloys are steel and cast iron. The
region to the right of 2 percent carbon represents cast iron and that
to the left represents steel (although the limit of carbon content of
commercial steels is actually about 1.5 percent). Furthermore, this
diagram represents only plain carbon steels. When large amounts of
alloy additions are made to steel, the iron-carbon diagram changes
considerably.
As an example, take a .2 percent carbon steel (AISI 1020)~ as it is
cooled slowly from the liquid. the following changes occur on the
DELTA
LIQUIO
,,-
L.IQUID
DELTA--..../
/'
OELTA---"'I
AUSTENITE
,,- / '
LIQUID
IRON CARBIDE
I
AUSTENITE 8 IRON CARBIDE (CEMENTITE)
I
FERRITE
a IRON CARBIDE
I
(CEMENTlTE)
1
fERRITE
lloE-i~-'loI-- PEARLITE a CEMENTITE _.,...1--------;>ol

1.5 2.0
4.3
PER CENT CARBON (BY WEIGHT)
FERRITE
a PEARLITE
Fig. 1112. Iron-carbon diagram.
181
6.6
Chapter 11
iron-carbon diagram. First, some nuclei of delta iron form in the
melt. Delta iron is a body-centered cubic form of iron that occurs
only at very high temperatures and is shown in the small areas at the
top left-hand corner of the diagram. As the liquid cools further, the
nuclei form small crystals which continually grow larger.
When the temperature of 2720F has been reached, the melt
becomes mushy and consists of many solid crystals and some liquid.
At this temperature a change occurs in the solid crystals. The melt
briefly enters the austenite + 'liquid region at this temperature and
the solid crystals in the melt change from delta iron into austenite,
which, it will be recalled, is the face-centered cubic form of iron.
At a slightly lower temperature (approximately 2700F) all of the
liquid has solidified. The solidified metal now consists of a large
number of individual grains of austenite. All of the carbon atoms are
dissolved in the solid austenite crystals or grains.
No further change occurs until the austenite is cooled to approximately I 580F, at which time some of the austenite (particularly in
the region of the grain boundaries) changes into ferrite, forming a
number of small grains of ferrite. As the metal cools further, these
ferrite grains continue to grow and additional grains of ferrite form,
At about 1333F the microstructure consists of almost 75 percent
ferrite and 25 percent austenite. Then, at this temperature, the
remaining austenite transforms to pearlite. No further changes occur
as the metal cools to room temperature.
The final microstructure consists of grains of ferrite and pearlite.
It should be remembered that pearlite consists of plates of ferrite and
cementite (iron carbide).
Similarly, it is possible to predict the structure of other compositions of iron and carbon from this diagram when they are cooled
slowly.
Grain Size
A fine grain size promotes both increased strength and increased

ductility in a metal. The grain size in steel may be altered by heat
treatment. When it is heated above certain critical temperatures, a
phase change occurs and new grains wilJ nucleate and grow within
the old grains.
A part of the iron-carbon diagram is shown at the right in Fig.
II-D. The critical temperatures are labeled Act. AC3, and AC3J,
and A,m. It will be seen that these temperatures are at the boundaries
of the different phases in steel.
The microstructure of an AISI 1020 steel (.2 percent carbon) will
consist of grains of pearlite and ferrile. When it is heated no change
182

I
I
I
I
-+AUSTEN ITE
t;
~
1:\
\0<- AGS ->l------'~"'-k----AG-S - - -
01
~~I
11--+---+.----'% CARBON
AGS" AV(RABE
",0
",0
(jR,5d~
SIZt:
u"w
Fig. 11-13. &hematic drawing of grain-coarsening effect when heating and
cooling a 2 percent carbon steel.
will occur until it reaches the lower critical temperature (Ac!). At

this temperature the pearlite will be transformed into austenite and
in this process many new grains of austenite will form in the old
grains of pearlite. As the temperature is increased further, more
austenite grains are formed, this time within the remaining ferrite
grains. This continues until the upper critical temperature for the
AISI t 020 steel (Ac3) is reached, at which the microstructure will
consist entirely of austenite grains.
Because many new grains of austenite were formed within each of
the former grains, the steel now has more grains which are finer than
the previous grains. The fact that the average grain size has been
decreased upon heating it above the upper critical temperature is
shown schematically by the diagram at the left in Fig. 11-13.
If this steel is now heated to a temperature above the upper
critical (Ac3) temperature, little change in the average grain size will
occur until the grain~coarsening temperature is reached. When this
temperature is exceeded, the smaller grains will coalesce, or grow
together, forming larger grains. Above this temperature, the average
grain size will increase rapidly as shown in Fig. 11-13.
It is not possible to define a specific temperature that will be the
grain-coarsening temperature for a specific class of steel. Much
depends on how it is made; however, it is always above the upper
critical (AC3) temperature.
If this piece of steel is now cooled from above the graincoarsening temperature, there is apt to be a slight increase in the
average grain size until the grain-coarsening temperature is reached,
183
Chapter 11
as shown in Case 1, Fig. II-D. When the steel reaches the upper
critical temperature. new grains of ferrite form in the austenite
grains; this process continues until the lower critical temperature is
reached. At this temperature, the remaining austenite transforms to
pearlite. As in the case of heating the steel, the grain size was refined
when the steel passed through the temperature region between the
two critical temperatures. Cooling the steel below the lower critical
temperature, however, does not change the grain size.
In Case 1, Fig. 11-13, the AISI 1020 steel was heated above the
grain-coarsening temperature resulting in a very coarse austenitic
grain size. When this steel is cooled to room temperature; the grains
will be larger than they were prior to heating.
If this steel had been heated only slightly above the upper critical
temperature and then cooled, a significant grain refinement would
have occurred, as can be seen from Case 2, Fig. 11-13. In this case,
the steel is cooled from a fine austenitic grain size. Sometimes steels
are deliberately heaHreated in this manner to refine the grain.
When depositing a weld, the metal adjacent to the weld is heated
and cooled in the manner just described. The result is that there will
be regions of coarse-grain size and regions of fine~grain size adjacent
to the weld. In welding, the metal is heated and cooled more rapidly
than described, and for this reason, the grain coarsening and the
critical temperatures will be different than shown in Fig. II ~ 13.
However, the effects will be the same, with the changes described
occurring at slightly different temperatures.
Size Change
It is a common experience that metals expand when they are

heated. Like any metal, steel will also expand when heated; however. in the region between the critical temperatures, Ac}, and AC3, a
marked contraction OCcurs upon heating. This is shown schematically in Fig. 11-14 for a .2 percent carbon steel (AISI 1020).
When the steel cools slowly from an elevated temperature, like
other metals, it will contract. However, here again, something unexpected happens: the steel expands in the region between the critical
temperatures, as shown in Fig. 11-14.
The reason for this behavior is that a phase change occurs in the
region between the critical temperatures. Austenite is a denser
structure than ferrite and pearlite. so that upon heating to austenite
the metal must contract. Upon cooling from austenite to pearlite and
ferrite, the metal must expand.
When .8 percent carbon steel is rapidly cooled or quenched from
a temperature at which it is austenite, ferrite and pearlite do not
184

form. Instead, the hard martensite structure forms. Martensite is less
dense than pearlite, ferrite, or austenite. Thus, when the martensite
is formed, the metal must expand. As shown in Fig. 11-14, this
always occurs at a lower temperature.
R~PlD
HEAriNG
COOUNG
(QUENCHING)
SLOW
COOLING
AUSTENITE
---+
AUSTENITE 8
IRON CARBIDE
- ---,
I
I
!
'"
IX
IX
'"
Q.
~
w
....
MORTENSITE
LENGTH
.~C
LENGTH
.2C
LENGTH
.BC
..,
:>
'"
l!!'"
:l!
'"....
Ao
3,1
FERRITE a hRON CARBIDE
I
I
I
.B
% CARBOO
Fig. 11-]4. Schematic drawing showing the effect on the elongation of plain
carbon steel when heating and cooling.
Alloying Elements in Steel
Alloying elements are added to steel for the following purposes:

1.
2.
3.
4.
5.
6.
To increase hardenability
To increase strength at ordinary temperatures
To improve high-temperature properties
To improve toughness
To increase wear-resistance
To increase corrosion-resistance.
When alloys are added in the right amount and in the correct
combination, it can be said that the steel will be improved in one or
more of the above-mentioned characteristics. It is important, however, to have the right amount and the right combination of alloying
elements. Merely adding more alloy does not necessarily improve
the steel proportionately; it may even have a harmful effect. Two
alloys when added to a steel are more effective than the same
amount of a single alloy, if in combination, they enhance a property
of the steel. The effect of a single element depends upon the effects
of other elements, and this must be taken into account when evaluating specific compositions of steels.
185
Chapter 11
To simplify the discussion to follow, each major alloying element
will be discussed separately; however, the reader must not lose sight
of the fact that these elements usually act in combination with other
alloying elements.
Carbon. Carbon is the most important and effective alloying element in steel. Each small increase in the carbon content increases
the hardness and tensile strength of the steel in the "as rolled" or
"normalized" condition. When the carbon content exceeds .85 percent, the resulting increase in hardness and strength is less than in
the lower carbon ranges; however, the wear-resistance continues to
increase above this carbon percentage.
The presence of carbon accounts for the ability to harden steel by
heat treatment. For plain carbon steel the maximum hardness that
can be attained by heat treatment increases with increased carbon
content until the carbon content is about .60 percent. Above this
carbon content the rate of increase in hardness is very small. The
effect of alloying elements is to lower the carbon content at which
the maximum hardness occurs.
Only a very small percentage of carbon can be dissolved in ferrite;
however, all of the carbon present in steel will dissolve in the
austenite, which, it will be remembered, normally exists at a higher
temperature. At room temperature the carbon is combined with iron
to form the very hard and brittle iron carbide (Fe3C), called cementite. Cementite strengthens the steel, increases its hardness, and
increases its wear-resistance in the unheat-treated condition.
Manganese. Next to carbon, manganese is probably the most important alloy that is added to steel because it combines with the sulfur to
form manganese sulfide. Sulfur, when not combined with manganese, is very harmful, causing hot shortness in steeL When combined with manganese, the sulfur is harmless. Therefore, manganese
is an essential ingredient in all steels; for the purpose of combining
with sulfur, in quantities ranging from .40 to 1.00 percent. When
present above this amount, manganese is considered to be an alloying element; it also acts to deoxidize steel.
When present as an alloying dement, manganese contributes to
the strength and toughness of steel, and it greatly increa~es its
hardenability. In very large amounts (12 to 15 percent) the steel will
remain austenitic at room temperature. Austenitic manganese steel is
an extremely tough alloy that is used for applications requiring
severe impact and abrasion resistance, such as for power shovel
blades. High manganese content adversely affects the weld ability of
steel by increasing its crack sensitivity.
Silicon. Silicon is one of the principal deoxidizers used in steeL It is a
186

very important element in the metallurgy of gray cast iron, In steel it
dissolves in the ferrite increasing its strength and toughness. Steel
having less than .10 percent carbon and about 3 percent silicon has
excellent magnetic properties and is used in the cores and poles of
electrical machinery. A steel containing 1 to 2 percent silicon is used
for structural applications requiring a high yield point.
Nickel. Nickel is another very important alloying element in steel.
When present in appreciable amounts it improves the toughness and
impact resistance of the steel, particularly at lower temperatures. It
contributes to easier and more foolproof heat treatment of steel
thereby reducing costly heat-treating failures. It dissolves in the
ferrite and strengthens it. It causes more pearlite to form and the
pearlite formed is finer; therefore, it is a stronger and tougher
pearlite. Nickel steels are particularly suitable for case hardening.
Other than making steels more hardenable, the presence of nickel in
steel causes no difficulty in welding.
Chromium, Chromium increases hardenability and abrasion resistance. When present in quantities in excess of 4 percent, the corrosion resistance of steel is improved. It is one of the most effective
alloys in promoting hardenability. High-chromium steels are airhardening. Chromium forms a very stable carbide that has exceptional wear-resistance. It also promotes carburization of the steel.
The presence of chromium in steels presents problems in welding,
however, as the increased hardenability can cause cracking in and
adjacent to the weld joint. Steels containing 5 to 6 percent chromium
can be welded only by using special techniques.
Molybdenum. Molybdenum, manganese, and chromium have a
greater effect on hardenability than any other commonly used alloying element. Molybdenum has a powerful effect in increasing the
high-temperature strength of steel and it retards grain growth at
temperatures above the upper critical temperatures. Quenchhardened molybdenum steel is fine grained and very tough at all
hardness levels. It is used as an alloy in many grades of high-speed
tool steels.
Vanadium. Vanadium is used to inhibit grain growth in steel at
elevated temperatures, thereby causing the steel to be fine grained
at room temperature, adding to its strength and toughness. It
improves the hardenability of medium carbon steels when present in
amounts of .04 to .05 percent. Above this content, hardenability
decreases when the steel is heated to normal hardening temperature;
however, if a higher hardening temperature is used, the hardenability is increased. In high-speed steels used for cutting tools, vanadium is an essential alloying element to improve hardenability and to
obtain a fine-grain size.
187
Chapter 11
Tungsten. Tungsten is used in high-speed tool steels to promote the
retention of the hardness, obtained through heat treatment, at high
temperatures. It forms an extremely hard carbide that is very wearresistant.
Cobalt. Cobalt is an unusual element because it decreases the
hardenability of steel. It strengthens the ferrite and it is used in some
highspeed steels to increase resistance to abrasion at high temperatures.
Boron. Boron is used in steel for only one purpose, to increase the
hardenability of steels having less than .60 percent carbon content.
It is effective when used in quantities of only a few thousandths of a
percent. Perhaps, for this reason, the degree of effectiveness of
boron steels is sometimes rather unpredictable.
Titanium. Titanium has a very strong tendency to form carbides and
to reduce the ability of the steel to be hardened by heat treatment. It
is used as a deoxidizer and in deep-drawing steels to prevent age
hardening. It is also used for this purpose in stainless steels and in
heaHesisting steels, to increase their strength.
Aluminum is used principally as a deoxidizer in steel although it
also promotes a fine austenitic grain size. Copper is sometimes added
to steel to improve its resistance to atmospheric corrosion. Lead is
added to some steels to improve their machinability. While sulfur is
normally considered to be an impurity in steel, it is sometimes
intentionally added, along with the required amount of manganese
to form manganese sulfide. This is done to improve the machinability of the steel. Phosphorus is considered to be an impurity.
Stress Relieving
Cold-working imparts stresses in metals that remain after the

cold-working operation is finished. These stresses are called residual
stresses. When the cold-working is severe the grains are distorted as
a result of this operation (see Fig. 11-17). Machining operations also
impart residual stresses in metals as a result of the cold-working
effect, the severity depending upon the size of the cut, the sharpness
of the cutting tool, and the type of machining operation. There are
other causes of residual stresses in metals such as grinding stresses
caused by the grinding operation and residual stresses in castings
resulting from a solidification of the cast metal in the mold. Welding
will also impart residual stresses in a metal.
The grains of a bar of cold-drawn or cold-worked steel are
distorted in a direction perpendicular to the cold-working operation
and this bar will contain severe residual stresses. If this bar is heated
188

. to below the lower critical temperature, usually 1000 to 1200F, the
distorted grains will recrystallize and the residual stresses will be
relieved. A new, and finer, grain structure will result from this
operation, and the hardness of the bar will be lower.
This operation is called "stress relieving" or "stress-relief' annealing. When welding a piece of cold-drawn or cold-rolled Steel, the
metal adjacent to the weld that has been heated to a temperature
above 950 to lOOOF, but not exceeding 1333F, will also be stress
relieved as described and the grain structure will be refined.
Structure of the Weld

When making a butt weld (see Fig. 11-15) to join two mild steel
plates, the liquid in the puddle is mostly above the melting point, but
~
SECTION NN
Courtesy of George E. LinneTt, "Welding Metallurgy"

(New York: American Welding Society, 196.5)
Fig. 11-15. Three views of crystals in a single-pass weld. (Top) Single-pass butt
weld on a plate. (A) Three views of crystals or grains from top, side, and end, if!
the single-pass weld shown above. (B) A backing strip may exert a chilling
action causing the crystals to grow vertically upward at the root of the beau.
189
Chapter 11
at the boundary of the parent metal it is just at the melting point.
The parent metal adjacent to the weld is in a mushy condition.
As the source of heat (the arc) is moved on, the mushy region
solidifies and the molten metal in the puddle starts to freeze. The
molten meta] adjacent to the metal at the sides of the weld and, if
present, at the solid weld bead starts to solidify first. Grains start to
grow out from these surfaces. These grains have a columnar shape,
as shown in Fig. 11-15.
As the metal in the weld continues to cool it solidifies completely.
Upon cooling from below the solidification temperature to room
temperature it undergoes phase changes (see diagram in Fig. 11-12).
The temperatures at which these changes occur are lower than those
shown in the iron-carbon diagram because of the relatively fast
cooling rate of the weld.
The phase changes would occur exactly as described on pages
165 and 166 in a .2 percent carbon steel if the weld cooled slowly.
However, the actual weld puddle cools very rapidly. For this reason
a somewhat different microstructure will form. The microstructure
in the weld will consist partly of a Widmanstatten structure and
partly as minute plates of pearlite.
Structure of the Weld and the Weld-Affected Zone
During the welding process, the metal adjacent to the weld is also
heated. The temperatures existing in a mild steel plate during welding are given in Fig. I 1-16. Since the temperatures adjacent to the
weld exceed the critical temperatures, the grain size of the metal in
this zone is affected.
The effect of the welding heat on the metal adjacent to the weld is
shown in Fig. 11-17. In this case the metal is iron and two welds are
shown. The upper weld is made on iron that has not been coldworked, while the lower weld is on previously cold-worked iron.
Observe the difference in the shape of the original grains in these
two metals.
In both cases, the metal in the weld has columnar grains. Adjacent to the weld the temperature exceeded the grain-coarsening
temperature and, as a result, this region always is characterized by
its coarse grains. Further away from this region there is an area
where the maximum temperature exceeded the critical temperatures
but did not exceed the grain-coarsening temperature. This results in
a region of very fine grains that is always present in the case of iron
and steel.
Still further away there is a region where the maximum temperature was above 950F but did not exceed the lower critical tempera-
190
400" F
1100" F
1335 F
156ctf
1850" F
2550" F
Courtesy of George E. Lin",!r', "Welding Metallurgy"

(New York: American Welding Society, 1965)
Fig. 11-16. The distribution of the temperature in a mild steel plate at an instant
during welding. W-Uquid weld metal (puddle). The shaded area is the metal
that is in the mushy stage.
ture. The temperature range did not affect the grains in the upper
piece (Fig. 11-17), in which the original grains are not cold-worked.
However, in the lower piece, which was cold-worked, the grains
were refined and the residual stresses relieved. In the lower view the
second region of refined grains can be seen. It is interesting to note
the coarser-grained region between the two fine-grained regions.
This is the result of heating to just below the lower critical temperature. To avoid this coarse-grain structure, the maximum temperature
for stress-relief annealing should not exceed 1200F.
In summary, the region adjacent to the weld is always characterized by coarse grains which is followed by a region where the grains
are highly refined. Cold-worked steels have a second region where
the grain structure has been refined, which does not exist in a steel
that has not been cold-worked. The region in which the large grains
exist is less ductile than the tine-grain region and the other regions
191
Chapter 11
75
50
100
[25
BRINELL HARDNESS
Courtesy of George E. Linnerr, "Welding Melaliurgy"

(New York: A",nican Welding SoddY, 1965)
Fig. 11-17, Single pass weld in iron. nOp) Iron annealed before welding. Note
grain refinement in the vicinity of the zone that has reached 1670r-: during
welding. Chart at the right shows that welding d.id not affect hardness. (Bottom)
Iron cold-worked (cold-rolled) before welding. Note grain refinement in vicinity
of the zones that reached 950F (approx.imate temperature of recrystallization)
and 1670F. Chart at right shows that welding has softened the iron in the
zones that were heated above 950F.
where smaller grains exist, unless they are very severely coldworked.
An advantage of multiple-pass welding is that the following pass
refines the grain in the previous pass. The second pass of a two-pass
weld in mild steel will, for example, refine the grains in the first bead.
192
.,.8\.

The weld will consist of a finegrained lower bead and a coarsegrained upper bead; therefore, it will have a better ductility than a
singlepass weld.
In the case of mild steel with a .2 percent carbon content, the weld
metal will have a Widmanstatten structure at room temperature.
All of the metal adjacent to the weld that has been heated above
2000F will also have this structure. The metal that did not reach
2000F, but was heated above 1500F, has a structure consisting of
small grains of ferrite among which small grains of pearlite are
distributed. This structure is more ductile than the Widmanstlitten
structure.
The next region is that which has not been heated above 1560F
but above the lower critical (l333F) temperature. This region will
have some rather large ferrite grains and clusters of finer ferrite and
pearlite grains.
The region that did not reach the lower critical temperature
remains essentially unchanged although there is some tendency for
spheroidite to form. Usually the speed of welding is so high that
spheroidization of the cementite in the pearlite rarely occurs. This
region will consist of the original ferrite and pearlite grains.
Residual Stresses and Distortion In Welds
Earlier in this chapter stress was likened to an in,ternal load

existing inside of the metal. Stresses in metal are a result of external
loads and when the external load is removed, the stresses are
relieved.
Stresses inside metals can also result from other causes, such as
cold working, machining, grinding, heat treating, casting, and welding. Since there are no external loads to remove, these stresses
cannot readily be relieved and they remain locked up inside the
metal. These stresses are called residual stresses. Residual stresses
can be relieved by heat treatment, by physically removing a section
of the metal, or by yielding (distortion) of the metaL
Residual stresses always react against other residual stresses inside
a metal in order to achieve a balance. No attempt will be made here
to show exactly how these stresses are distributed inside the metal,
but rather to show their cause and effect.
An edge weld is shown in Fig. 1118A. Assume that this weld
bead could be separated from the base metal while it is still very hot,
perhaps just below the solidification temperature. Of course, the
metal bar adjacent to the weld bead would also be very hot
however, for simplicity, let it be assumed that it is at room tempera
193
Chapter 11
ture. This condition is shown in Fig. 11-18B, where the length of the
bead is equal to the length of the block (L 1).
Let the weld bead cool to room temperature. When this happens
the weld bead shrinks in length and, as shown in Fig. 11-18C, the
length of the weld bead, Lz, is less than the length of the block, Lt.
In order to attach the weld bead to the block so that the ends of
the bead are flush with the ends of the block, it is necessary to pull
on the bead in order to stretch it, and to push on the block to
compress it, Fig. ] 1-180. Pulling on the bead results in a tensile
stress in the metal inside of the bead. Similarly, a compressive stress
is set up inside the block in the region adjacent to the bead.
In Fig. 1118E, assume that the weld bead and the block are now
firmly joined together. The block, having been compressed, wants to
stretch out again because of the elasticity of the metal. In doing so it
~r...
)~-----)~
.F
Fig. I 1-18. Schematic drawing showing why residual stresses exist in welds and
their effect in causing weld distortion.
pulls on the weld bead thereby setting up a permanent residual

tensile stress in the metal inside the weld bead. The weld bead, in
turn, was stretched in order to join it to the block. The elasticity of
the weld metal causes it to want to pull together, thereby exerting a
permanent compressive force on the metal in the block that is
adjacent to the bead.
194

The end result is that a permanent residual tensile stress remains
in the weld bead and a permanent residual compressive stress
remains in the metal in the block that is adjacent to the bead. In
actual practice, the situation is much more complex; however, this
provides a model for visualization.
The reaction of the weld bead on the block in compressing the
metal adjacent to the bead tends to bend the block, as shown in Fig.
I 118F. The degree of actual bending that might occur will depend
upon many factors such as the size of the block, the size of the weld,
and the restraints upon movement placed on the block by other
members to which it may be attached.
Similarly, it can be shown (Fig. I 119) that a butt weld has
residual tensile stresses in the weld bead and residual compressive
stresses in the metal within the plates that are adjacent to the weld.
In this case, however, the two plates react against each other, if they
are of approximately the same width, and they will not bend as
showninFig.II-18E.
However, if the plates are unrestrained, as in Fig. II20A, the
weld will tend to offset them at an angle "X," as shown in Fig.
11-20B. The reason for this is the shrinkage of the metal in the weld
bead upon cooling to room temperature. As can be seen, there is
Fig. J 1-19. Generalization of the action of residual stresses in the lengthwise

direction in a butt weld.
more weld metal at the top of the bead than at the bottom; therefore,
the top will shrink a greater distance than the bottom. This action
results in the movement of the plates as shown,
If these plates are prevented from moving, as in Fig. 11-20C, the
metal in the weld is prevented from shrinking. While the temperature is such that the metal in the weld bead is soft and plastic, it will
deform; but when it has cooled down enough to obtain strength it
will be stretched elastically, thereby setting up a permanent residual
tensile stress in the weld bead.
The residual tensile stress in the weld bead tends to pull on the
plates to which it is attached, and in doing this sets up residual
195
Chapter 11
o
Fig. 11-20. Generalization of the action of resIdual stresses perpendicular to the
lengthwise direction in a bUll weld.
tensile stresses in the plates, Fig. 11-200. Again, the actual stress
pattern in the weld is more complicated; however, this example
helps to visualize why distortions occur and how residual stresses
within the weld occur. Without going into the reasons why at this
time, the welded plates in Fig. II-20C would tend to distort in the
general direction of the plates in Fig. II-20B.
In Fig. 11-21 A, the fillet weld causes the plates to move at an
angle, in a manner similar to the butt weld in Fig. 11-20. If the
plates in Fig. 11-21 are restrained from movement, residual stresses
will be set up in the weld bead and in the plates being joined.
When a fillet weld is made on relatively thin plates, where the
temperature rises to nearly a red heat at the bottom of the lower
plate, the metal may be upset, as shown in Fig. 11-21B and C.
The metal in the lower plate is heated in the region of the weld
and wants to expand. However, it is prevented from doing so by the
surrounding colder metal and the restraints. As a result, the relatively weak hot metal near the weld is upset by compression and
possibly bent. During cooling the plates were not sufficiently rigid to
iron out the bend.
196
A
r-
..
...
.......
B
"
/."
...-
c
Fig. I 1-21. Distortion occurring in fillel welds.
197
CHAPTER
12
Distortion in Pipe Welding

Distortion caused by welding can seriously affect the alignment
and the Iocational accuracy of a pipe installation unless preventive
measures are taken to avoid these problems. While in the last
chapter the basic causes of distortion in weld joints were treated, this
chapter will show how the distortion in the weld joint can affect the
fabrication of the pipe installation and explain the steps necessary to
overcome this.
On the job, the pipe welder must always be aware of the distortion
and must take those steps required to prevent this from affecting the
quality of his work. The amount of distortion cannot be calculated,
even by engineers. However, this should not deter the welder from
estimating the direction and possible magnitude of the distortion.
Having done this, he should plan his work in advance to minimize
the effect of the distortion. The following examples will describe
how this is done.
Example 1
When cross-country pipelines were discussed in Chapter 1, it was

pointed out that line-up clamps are used to hold the pipes in place
while welding the root bead. Their primary purpose in pipeline
welding is to prevent the root bead from cracking as it is being
deposited, but these clamps also serve to align the pipes.
Line-up clamps are also used on other types of pipe-fabricating
jobs. Here their main purpose is to align the pipes and to hold them
in position. A sufficient amount of the root bead is deposited to
assure alignment of the pipes after they are welded and to prevent
the root bead from cracking before removing the alignment clamps.
These clamps can be obtamed in several different styles and sizes
(see Fig. 8-5),
When welding longer lengths of pipeline, where several standard
lengths of pipe are welded together, the pipes will bend under their
own weight. It is important then to provide an adequate means of
support under the pipes to prevent them from sagging while they are
being welded together.
198

ExampJe2
A typical welded-pipe fabrication is shown in Fig. 12-1. Several

welds are required to fabricate the installation, which presents some
distortion problems that are frequently encountered in pipe welding.
The header, Fig. 122, has four branch pipes that are welded along
the top of the header pipe, as shown. Weld joints, such as made by
these branch lines, are sometimes called "weldolets."
Simply welding the branch lines to the header pipe would cause it
to bend, as shown in the lower view of Fig. 12-2. To prevent this
from occurring, a strongback is attached to the header on the side
opposite the branch pipes. The purpose of the slrongback is io add
to the stiffness of the header pipe, and to resist the tendency of this
pipe to bend as a result of heat input, in making the welds.
TYPICAL FABRICATION
Fig. 12-1. A typical welded-pipe fabrication.
1---- - - - - - - - - - - - - - d n~n
----------------MANIFOLD
'-
~A~
B-
STRONGBACK
~--
DISTORTED
Fig. 12-2. Crop) Use of a strongback to prevent distortion. (Bottom) Pipe

distorted (bent) as a result of welding.
199
Chapter 12
A strongback may be a channel section or an I-beam of adequate
size in relation to the header pipe. On mild steel pipe it may be
attached to the header by a series of tack welds. After the branch
pipes have been welded in place, the header is removed with an
oxyacetylene cutting torch. On some jobs this procedure for attaching the header is not permissible, especially when the header pipe is
made from a higher alloy steel. In this case the strongback is
attached by using heavy-duty clamps.
The branch lines should not be welded in the alphabetical order
shown in Fig. 12-2. To minimize distortion, they should be welded
in the following order: C, A, D, and B. By welding the branch pipes
in this sequence, the amount of heat put into the area surrounding a
header at one time will be reduced. While the temperature of the
weld must be high enough to obtain fusion, the heat will diffuse
more rapidly and only a small area will attain a temperature high
enough to cause serious distortion. It is the expansion and contraction of the metal surrounding the weld, as well as the weld metal
itself, that causes the distortion. The overall effect of using this
sequence, together with the strong back, is to reduce the amount of
distortion that occurs, or to eliminate it entirely.
To be able to more fully understand how distortion occurs when
the metal adjacent to the weld is heated to a high temperature, a
schematic example will be given. The bar of steel in Fig. 12-3 is to
have a small section heated to a high temperature where plastic flow
can readily occur. This can be likened to heating this part of the bar
to the forging temperature. Moreover, heating a small section of the
bar to this temperature can be compared to the condition of the bar
when welding a single bead.
Assume that the section of the bar to be heated can be cut out and
that this section will fit tightly into the resulting slot, as in Fig.
12-3A. If the section is removed from the bar and then heated, it will
expand and no longer fit into the slot, as in Fig. 12-3B.
In order that it again may fit tightly into the slot, the removed
section must be upset by applying a compression force, as shown in
Fig. l2~3c. This section is then placed back in the slot, Fig. 12-3D,
and allowed to cool to room temperature. When this has occurred,
the upset section will shrink and, as shown in Fig. 12-3E, it will fit
loosely in the slot.
Two things occur to make the loose section fit tightly in the slot
again; these are shown in Fig. 12-3F. The loose section is stretched
by pulling on it in tension and the sides of the slot are pulled inward.
However, pulling inward on the sides of the slot will tend to cause
the large bar of steel to bend. Figure 123G shows how the bar of
200

steel may be bent permanently when the loose section is back in the
slot, fitting tightly.
Heating a part of the bar of steel or a part of a pipe only will
result in a situation that is very similar to the schematic example
discussed in the preceding paragraphs. When the small part of the
bar or pipe is heated, it will be upset by the unyielding colder metal
that surrounds it. As the upset metal cools to room temperature it
contracts and, if left by itself, it would be even shorter than before
because it was upset while hot In contracting, the upset metal pulls
on the surrounding colder metal while, at the same time, it is being
pulled by the colder metal. By pulling on the surrounding colder
metal, the heated part of the bar tends to cause the entire bar to bend
as it cools.
Actually, the bar may bend very noticeably or very little bending
may take place. The amount of permanent bending that occurs will
depend on the amount of heat input, the size of the area heated, the
size of the bar of steel, and its shape.
NEr,lL TO 8E "lEATEtr
/" r----
COlOME1AL
---'U-"T-''''---I
) I
F
/HtATEO MET.-L _HER W'SETTI~
I11III--
COlP MUlC
~-----
PREViOUSlY HEATEO ....O<Jl>SU

METAL WHE'" COLO AGlLlhl
COLD YE'TAt
cE
E
Fig. 12-3. SChematic drawings showing how a bar of metal is bent as a result of
heating a small volume of metal on one surface. A. Original bar; B. Small
section of bar removed and heated to a forging temperature; C. Heated section
upset by compression; D. Upset section placed in notch with a tight fit; E. Upset
section after cooling to room temperature; F. Forces on the bar and the upset
section required to obtain a tight fit; G. Bar bent to close notch in order to
obtain a tight fit of the upset piece; H. Strongback prevents the bar from
bending. When upset piece is attached to ends of notch. it will be elongated or
"ironed out."
201
Chapter 12
If a strongback is added to the bar of steel to increase its stiffness,
Fig. 12-3H, the amount of permanent bending will be reduced Of,

perhaps, it can be prevented entirely. In this case the bar of steel is
stiff enough to resist bending and the tensile forces pulling on the
heated section will elongate or "iron out" this section as it cools.
It should be remembered that heating and cooling a small area of
the bar will set up residual stresses within the metal. These occur
whether or not a strongback is used. When the stfongback is removed, the residual stresses will usually cause a slight amount of
bending, but often this is a negligible quantity. The residual stresses
can be removed only by heating the entire bar of steel to a temperature that is just below the lower critical temperature.
Example 3
When welding branch lines onto a pipe, where the branch lines
are not in the same plane, the welder must plan to weld the branches
in the correct sequence in order to minimize the effect of distortion.
In Fig. 12-4, the branch line C is to be welded on the opposite side
of A and B.
Because they are located further from the center of the pipe, the
welds at A and B will cause less distortion and bending than the weld
at C. which is located near the center of the pipe. For this reason,
branches A and B should be welded first. If the pipe is bent as a
result of the welds at A and B, welding the branch at C will tend to
straighten the pipe out again because this weld has the greatest
bending effect on the pipe.
Fig. 12-4. Correct welding sequence for welding branch lines thai are in differ
ent planes along the header pipe.
Example 4
To maintain alignment when welding and after the completion of

a weld, a branch pipe must not be welded by depositing the ~oot
p'~ad completely around the circumference of the pipe at one tIme.
202

The central axis of the pipe is called the neutral axis. In order to
balance the distortion resulting from the shrinkage of the weld metal
when it cools (thereby maintaining the alignment of the branch
pipe), the welds should be balanced around this neutral axis, as
much as possible, as shown in Fig. 12-5.
The procedure to follow is to weld two short tack welds on
opposite sides of the branch pipe and two more 90 degrees from the
first two. Then weld a short section of the root bead on one side of
the pipe as in Fig. 12-5, and another short section of the root bead
on the other side. After this the root bead is welded on the remaining
sides.
While it is most important to weld the root beads in sequence, it
will also help to minimize distortion and to maintain the alignment
of the branch pipe(if the intermediate layers are welded in the same
sequence. The cover layer may be welded by going entirely around
the circumference of the pipe.
While the weld must be built up to have enough strength to
withstand any load to which it might be subjected, it is a mistake to
overweld the joint. This can cause distortion and will reduce the
elasticity of the joint. Particularly when the piping system is to
contain hot fluids, some elasticity in the piping system is required to
allow for the temperature changes that will occur. Elasticity is also
required for an outdoor piping system if it is subjected to large
seasonal temperature variations.
BRANCH PIPE
ROOT BEAD
ROOT BEAD
TOP VIEW
Hg. 12-5. Method of balancing the root beads around the neutral axis of a
branch pipe to minimize distortion and maintain alignment.
Example 5
Frequently, short sections of pipe, as in Fig. 12-6, must be welded

together so that they will align with two or more pipes. They must be
carefuny aligned when they are fit-up and they must be welded
together in the correct sequence; otherwise distortion can cause the
vertical pipe to be misaligned with respect to the pipe to which it is
attached.
203
Chapter 12
VERTiCAL PIP
Fig. 126. Correct welding sequence for welding the root beads of the pipe
assembly.
Hot metal deposited in the weld joint will shrink as it cools to

room temperature. For a given metal, such as steel, the amount of
shrinkage will depend upon the size of the deposit of weld metal and
the freedom of the pipes to move. After the tack welds have been
made to hold the pipes in place, the root bead should be welded in
place on all of the joints in Fig. 12-6 before the remaining beads are
deposited. The root beads are relatively narrow and cause Jess
distortion than the wider intermediate and cover beads. Moreover,
they stiffen the entire assembly, thereby reducing the distortion
resulting from the shrinkage of the wider beads,
It is equally important to weld the root beads in the correct order.
The last root bead should always be welded in the joint at Eand the
other root beads should be welded in a balanced sequence on either
side of E. In this way, a misalignment of the vertical pipe can be
corrected by heating the tack welds at E and drawing the misaligned
pipe into position before the root bead is deposited in this joint.
The first step in fabricating the pipe assembly is to fit-up the pipes.
Since this subject is treated in detail in Chapter 14, it will only be
necessary to mention here that the pipes must be aligned in the
specified position and tack welded in place. Then the root beads are
welded in all pipe joints and in the correct sequence. The correct
sequence is the alphabetical order given in Fig. 12-6, i.e., joints A.
B, C. D, and E. By following this procedure, the root beads are
alternately welded in joints on one side and then on the other side of
E, and joint E is welded last. The remaining beads should also be
welded by welding the joints in this sequence.
Example 6
Right-angle pipe joints, Fig. 12-7, are frequently encountered in

pipe welding. The 90degree turn is made by welding a pipe elbow in
place as shown. As always, the pipes must be ht-up correctly;

however, a good fit-up job can be spoiled by using an incorrect
welding procedure. As shown in Fig. 12-7B, distortion caused by
welding can cause the pipt:s to move so that they no longer are at
right angles after they are welded together.
To prevent this from happening, the elbow should first be tack
welded to the pipe that is in the fixed location, which, in this case, is
the horizontal pipe. If the vertical pipe is in a fixed location, the
,,
I
TACK
TACK
Fig. 12-7. A. Using a support to maintain the correct alignment of a right-angle

pipe joint; B. Distortion of a right-angle pipe joint caused by welding.
elbow should first be tacked to this pipe. After the elbow has been
tack welded to the horizontal pipe, the angularity of the pipe should
be checked by placing a level across the unwelded face of the elbow.
If necessary, the elbow can be aligned by heating the tack welds and
bending it slightly-The vertical pipe is then placed in position and aligned with the
elbow. The correct spacing of the root opening can be obtained with
a bent piece of wire. Four tack welds are then deposited around this
joint in the usual manner and the alignment of the assembly is
checked.
Before welding the remainder of the root bead, a support is added
to hold the pipes in position. This can be in the form of an "angle
iron" (correctly called a "steel angle") which is tack welded to the
two pipes to which the elbow is attached. The angle iron should be
welded at approximately 45 degrees, with respect to the elbow, to
form a right triangle (disregarding the curvature of the elbow). This
205
Chapter 12
will provide the maximum stiffness to the joint for holding the pipes
in place while welding.
With the pipes held in the correct angular relationship to each
other, the root beads are deposited in both weld joints, after which
the intermediate and cover beads are welded. When the welds in
both joints are finished, the angle iron is removed with an oxyacetylene cutting torch. It is easy to visualize how joints can be welded at
other angles, as in Fig. 12~8, by using this procedure.
In summary, while pipe welds tend to distort, errors caused by
distortion can be avoided by planning ahead and by using the
Fig. J 2-8. Using a support to maintain the correct angular alignment when
welding a pipe joint.
correct welding procedures. The real craftsman not only can make
good welds, he can fabricate pipe installations that meet all of the
dimensional requirements.
Following are some suggestions that will help to prevent distor~
tion:
1. Plan how each job is to be done before starting. Determine the
best sequence in which the pipe joints are to be welded.
2. Check the position and the alignment of each joint before the
filler layers are deposited. Usually, this should be done after tack
welding.
3. Balance the welds, especially the tack welds, about the neutral
axis of the pipe. Weld short root beads opposite each other.
4. Keep to a minimum the heat input in any section along the
pipe. Do not allow the heat to build up but distribute it as evenly as
possible.
5. Clamp or weld temporary supports to the pipes in order to
maintain the alignment while welding.
6. Make allowances for the contraction of the weld when lengths
of pipe are welded together.
7. Never weld the pipe joint to completion unless the pipes are
known to be in correct alignment and in the correct position. Before
starting to weld, make sure that the job will be right when it is
finished.
206
CHAPTER
13
Pipe Welding Defects

Many of the defects that occur in pipe welding have, of necessity,
already been discussed in previous chapters. In this chapter additional information will be given on the causes of these defects and
preventive measures required to avoid their occurrence. The subject
will be covered from the point of view of the welder, rather than that
of the engineer.
Arc Strike Cracking
There are many reasons why cracking occurs in the weld and in
the adjacent weld-affected zone. One cause of cracking is carelessly
striking the arc outside of the weld groove, or in the area in which
the weld deposit is to be made, Fig. 13-1.
When the arc is struck, the metal with which it is in contact is very
rapidly heated. Then when the arc is moved on it is very rapidly
cooled again; in effect, it is quenched. This affects only a relatively
small volume of metal, the remainder not being heated appreciably.
As explained in Chapter I], when the carbon content in steel is
high enough, heating it to above the upper critical temperature and
quenching it results in the formation of martensite. Low-carbon
martensite can even form in low-carbon steel if the quench is drastic
enough. With the addition of alloying elements, martensite will form
at a lower carbon content and at a slower cooling rate. The martensite forms while the metal is cooling at some temperature below
700F and metal that has formed into martensite expands. This
expansion of the metal results in internal stresses that may cause
sman micro-cracks to occur on, or just below, the surface.
If the arc is struck in the area over which the bead is to be
deposited, the metal that is stressed or possibly cracked is melted a
short time later as the bead is deposited, which eliminates the
problem. On the other hand, if the arc is struck outside of the weld
zone, the stresses and the cracks remain.
If cracks resulting from the arc strike appear on the surface, a
weld bead must be deposited over them. The bead should be made
reasonably long and wide in order to increase the amount of heat
input to the defective area, thereby reducing the cooling rate. Inter-
207
Chapter 13
Fig. 13-1. Deposits caused by striking the arc outside of the weld joint.
nal stresses can be corrected by postheating; i.e., by heating the

affected area with an oxyacetylene torch allowing the metal to cool
slowly. A most effective way to prevent the occurrence of these
stresses and cracks is to preheat the weld-affected area so that the
cooling rate is slowed down. Of course, the cardinal rule to follow is
never to strike the arc outside of the weld joint.
Underbead Cracking
This form of cracking occurs within the base metal at a very short
distance away from the fusion line. It occurs in low-alloy and
high-alloy steels. The factors responsible for this form of cracking
are not, as yet, completely understood. It is known that dissolved
hydrogen gas must be present; otherwise underbead cracking does
not occur. It is believed to be caused by stresses within the metal that
are the result of:
1. The unequal contraction of the base metal and the weld metal
2. The restraint of the cooler base metal
3. Stresses caused by the expansion of the metal when martensite
is formed
4. Stresses set up by the precipitation of hydrogen out of the
metal to form molecular hydrogen.
208

The presence of hydrogen seems to act as a trigger to start the
formation of these cracks.

The cracks occur below the surface of the metal and when a pipe
is subjected to a load, they spread very rapidly, causing the joint to
fail.
Since the presence of hydrogen is the primary cause of underbead
cracking, everything possible should be done to prevent hydrogen
from entering the molten metal. It can enter this molten metal from
the atmosphere, from ingredients in electrode coatings, and from
moisture.
To prevent hydrogen from entering from the atmosphere, the
molten metal should always be blanketed with the gaseous shield
formed by the electrode coating or, in the case of GTAW welding,
by the inert gas. Low-hydrogen electrodes are so designed that the
coating contains only the smallest trace of hydrogen. While these
electrodes eliminate underbead cracking, they are difficult to use in
some pipe-welding applications and, in the case of downhill welding,
they cannot be used at all.
A primary source of hydrogen in welding is moisture. Some
welders like to soak their electrodes in water in order to obtain a
better arc characteristic. This practice should be discouraged. The
electrode coating should be protected from any form of dampness.
Before attempting to weld, the weld joint should be dry and welding
should never be attempted in rain or snow.
If the weld metal cools more slowly, some of the hydrogen gas
that has been dissolved will have an opportunity to escape from the
weld by precipitating out. For this reason, preheating the metal
before welding helps to prevent underbead cracking. Preheating will
also prevent the formation of martensite, as explained before,
thereby eliminating one of the factors contributing to underbead
cracking. While postheating will help to relieve locked-up stresses in
the metal, it is seldom effective in preventing underbead cracking.
Restraint Cracking
Restraint cracking is usually associated with small welds made on

thick metal sections. This form of cracking occurs in the weld bead
and it is caused in part by the solidification and cooling pattern of
the weld. Because the parent, or base, metal is much cooler than the
weld metal, solidification progresses from the fusion line inward. For
this reason the center of the weld is at a higher temperature than the
remainder of the weld while it cools.
209
Chapter 13
The mechanism of restraint cracking can be illustrated by again
resorting to an imaginary situation, as seen in Fig. 13-2. Cross
sections through an imaginary weld are indicated. For simplicity, a
square butt joint is shown. Above each joint a small graph indicates
the distribution of the temperature across the joint.
Shown on the graph are the solidification temperature and the
strengthening temperature of the weld metal. Actually, there is no
fixed strengthening temperature, although each metal does have a
temperature range below which it attains a reasonable amount of
strength and above which it is very weak when slowly cooled.
In Fig. 13-2A, the weld metal has just solidified and, as can be
seen in the graph above the weld, the temperature drops off very
rapidly at the fusion line because of the chilling effect of the heavy
plate. In Fig. 13-2B, C, D and E, the temperature is assumed to have
Fig. 13 -2. Mechanism of restrai nt cracking. A. Temperature distribution across

butt joint just after all of the metal has solidified; B. Assumed shrinkage of weld
metal ifseparated from wall after the weld has cooled to temperature shown; C.
Weld metal must be pulled to rejoin base metal. thereby setting up tensi Ie
stresses; D. Weaker metal in center of weld will crack as a result of lhe ten~i Ie
stresses; E. Thicker weld joint may yield instead of cracking.
dropped an equal amount; i.e., they represent the same instant

during the cooling cycle.
For a moment, let it be assumed again that the weld metal could
be separated from the sides of the weld as it cools from the solidification temperature shown in Fig. 13-2A to the temperature shown in
the other diagrams. In this case the weld metal would separate
completely from the sides because it would have to shrink as it cools,
as shown in Fig. 13-2B.
In order to attach the weld metal to the sides while at this
temperature, it would be necessary to pull on the weld in order to
stretch it as shown in Fig. 13-2C. When this is done, the weld metal
is in tension and tensile stresses exist within it.
Most of the weld metal is strong enough at this temperature to
resist the tensile stresses; however, as shown in Fig. 1320, the metal
in the center of the weld, which is at the highest temperature, is still
very weak. The weak metal in the center will crack when subjected
210

to internal stresses. Of course, the weld metal does not aclUalJy
separate from the wall, but in shrinking as it cools, stresses arc set up
within the metal that can cause the weaker metal in the center of the
weld to crack.
As the weld shrinks upon cooling it must deform if it remains
attached to the sides of the weld. The amount of deformation that
can take place before cracking occurs depends upon the amount of
metal that is available. A thick weld, having a larger volume of
metal, will accommodate more deformation without cracking than
will a thin weld. Therefore, as shown in Fig. 13-2E, a heavy weld is
less likely to crack than a small weld, when they are on thick metal
sections.
Welding Faults and Their Prevention
Undercutting. When the base metal along the edge of the weld is
reduced from its original thickness, as shown in Fig. 13-3, the weld
is said to be undercut. There are several reasons why undercutting
occurs.
Excessive current can cause the edge of the joint to melt and the
molten metal will wash into the weld, Jeaving a drain-like impression
at the edges of the weld. This can occur in varying degrees; even
with a normal arc length, undercutting can occur if the electrode is
not manipulated in such a manner as to supply an adequate amount
of filler metal to the molten edge of the weld. Because the edge of the
weld cools morc rapidly than the center of the weld, the arc should
pause at the edges, when weaving, in order to deposit filler metal and
to supply additional heat to the area.
Pausing at the edge also has the effect of churning the metal in
this region, thereby obtaining a better mixture of the base metal and
Fig. 13-3. Undereul.
211
Chapter 13
the filler metal. If the composition of the weld metal varies greatly
from the edge to the center, it wili solidify at a different temperature,
and for this reason, the correct composition of filler metal should
always be used. Using incorrect filler metal can also contribute to
undercutting.
If the electrode angle is too small, the arc force will tend to wash
away the molten metal at the edges of the joint to cause undercutting. To avoid this cause of undercutting, it is important to maintain
the correct electrode angle while welding. Other causes of undercutting are dampened ~Iectrodes, using an excessive arc length, and a
welding speed that is too slow.
In summary, undercutting is caused by anyone or a combination
of the following factors;
1. Excessive welding current
2. Incorrect electrode manipulation
3. Using an incorrect welding rod which provides filler metal of
the wrong composition
4. Electrode angle too small
5. Arc length too long
6. Welding speed too slow
7. Using dampened electrode.
Slag Inclusions. Slag inclusions (nonmetallic particles of slag embedded in the weld) can have a serious, adverse effect on the quality of
a weld. Usually the slag is from the electrode coating, although, in
some instances, particles of slag (slag inclusions) appear in the base
metal and they can be retained in the weld. Figure 13-4 illustrates a
weld in which a large amount of slag inclusions appear.
These slag inclusions frequently appear at the edges of a weld if
the correct welding procedure is not used. The molten slag, being
lighter than the weld metal, rises to the surface of the liquid pool,
rapidly forming a blanket that covers the metal. Furthermore, the
slag solidifies at a lower temperature than the metal. Thus, when an
edge is not completely filled with metal and has a drooping contour,
the slag blanket slips around this contour and settles in the corner.
When the metal solidifies, more slag will settle in the edge to form a
tightly adhering inclusion.
During weaving, the arc should pause at the edge of the joint to
provide sufficient heat in this area so that it will not cool too rapidly
and will deposit additional filler metal to join the surfaces of the
solid and liquid metal without an undercut appearing. By pausing,
any previously trapped slag in this area will also be remelted and will
have time to rise to the surface of the weld metaL
212
Fig. 13-4. Inclusions in weld metal.
The arc force churns the metal and forces some of the slag into
the body of molten metal. Under certain conditions, some of this
slag can become entrapped below the surface of the metal by other
metal that is solidifying from the walls of the joint and up from the
bottom of the weld as well. This can result from erratic electrode
manipulation. Most frequently, however, it is caused by maintaining
a pool of molten metal that is too large in relation to the electrode
diameter.
If the body of liquid metal is too large it will have a tendency to
roll, especially if it is also excessively hot and very fluid. As shown in
Fig. 13-5, the layers of liquid metal above the fusion Hnc will move
LIQUID LAYERS ABOVE

FUSION LtNE
LIQUlD LAYER AT
FUSlON LINE
Fig. 13-5. Liquid layers above the fusion line will move out further than liquid
layer at fusion line.
213
Chapter 13
out further than those close to the fusion line. The liquid slag which
blankets the molten metal will be rolled over toward the fusion line
where it will be trapped as the metal solidifies. To prevent slag
inclusions from this cause, the puddle must not be allowed to
become too large in relation to the electrode size and an excessively
high current setting should not be used.
Another frequent cause of slag inclusions is carelessness in deslagging a previous layer of weld metal. Proper cleaning and deslagging
is essential prior to welding any additional beads. If this has not been
done, some of the particles of the slag coating may not have enough
time to rise to the surface when the weld is restarted, thereby
becoming entrapped. At all times, the solidified slag coating of a
previously deposited bead must be chipped off and this should be
followed by a vigorous application with a wire brush to remove any
remaining particles. When restarting the weld to continue a bead,
this should only be done for a distance of one or two inches behind
the crater and in the crater, in order to retain as much heat as
possible in the bead. However, before another bead is deposited over
this bead, all of the slag coating must be removed.
Inclusions can also be caused by heavy oxides such as rust and
surface scale. These oxides remain undissolved in the molten metal
and do not readily rise to the surface. When the weld metal freezes,
the oxide inclusions remain entrapped in the weld.
Fig.
3-6. Porosity below the surface of the weld bead. Part ground to expose
porosity.
214
Porosity. Porosity is caused by oil, grease, or moisture entering the

weld metal to form gas bubbles which become trapped in the
freezing metal. Porosity can appear in the form of large blow holes
or in the form of small pinholes. In Fig. 13-6, the weld bead has
been ground down slightly below the surface to show porosity in the
form of small blow holes. It is possible that there are additional blow
holes or pinholes below those shown.
Moisture is, perhaps, the principal cause of porosity. When moisture enters the molten metal it forms a vapor, and at the temperature
of the melt, the water vapor (H 2 0) quickly disassociates into hydrogen and oxygen gas. A part of these gases rises to the surface and
escapes to the atmosphere. However, a part of these gases is dissolved by the molten metaL
The amount of gas that a metal can hold in solution is dependent
upon the temperature of the melt. Less gas can be held in solution as
the temperature decreases. When the temperature of the molten
metal drops, it must precipitate out some of the gas that was in
solution. The precipitated gas forms a bubble which grows in size as
it rises to the surface of the puddle.
When the temperature of the molten metal drops, the liquid metal
also becomes less fluid. Ultimately, as the temperature continues to
drop, the metal becomes mushy. When the mushy stage has been
reached, gas bubbles rise very slowly and it is easy for them to be
trapped in the weld metal upon further cooling and solidification of
the melt.
Not all of the porosity is caused by the gases that were precipitated from the molten metal. If the speed of welding is very fast, the
molten metal will freeze so rapidly that bubbles of undissolved gas
are trapped. Also, excessively long whipping or weaving strokes can
cause the molten metal to periodically freeze very rapidly thereby
entrapping the bubbles. Moreover, by exposing the surface of the
puddle to the atmosphere, additional gases will be dissolved by the
molten metal when weaving or whipping excessively.
To prevent porosity, excessive whipping and weaving must be
avoided and the speed of welding must not' be too fast. All oil and
grease must be removed from the weld joint prior to welding.
Handling the electrodes with oily, greasy, or damp gloves is an
all-too-common cause of porosity, something of which many welders
are not aware. Oil, grease, and moisture can be soaked up by the
electrode coating and passed on into the weld metal. Electrodes
should not be soaked in water and whenever rain or snow can reach
the weld joint, welding should be stopped. The weld joint must be
free of any moisture; if necessary, the joint should be dried with an
215
Chapter 13
oxyacetylene torch. When they can be used, low-hydrogen electrodes reduce porosity to a significant extent.
Lack of Fusion and Incomplete Penetration. Lack of fusion and

incomplete penetration are unacceptable in many industrial piping
systems, particularly in systems subjected to high pressures. When
these defects are present in root beads they are in the form of small
channels and crevices in which corrosive compounds can settle, as
shown in Fig. 13-7. These small defects can enlarge into more
serious defects. especially when the pipes contain corrosive substances such as acids, liquefied gases, and sulfur compounds.
Primarily, lack of fusion and incomplete penetration occur when
the welding procedures described in the previous chapters have not
been followed or when the welder is careless in depositing the weld.
In any particular case, the reader must refer back to the pertinent
chapters when seeking measures to prevent a particular kind of
defect. All that can be said in a general way is that all facets of the
welding procedures must be given the most careful attention. This
would include the current setting, arc length, electrode angle, electrode manipulation (weave or whipping), the keyhole in the case of
root beads, and always the pool of molten metaL
Fig. 13-7. Incomplete penetration at the bottom of a root bead.
216
WELD REPAIR
This section discusses some of the proper methods for evaluating and
repairing defective welds. The defects covered here are those created
during and after welding, such as those caused by improper manipulation, insufficient protection of the weld metal pool, oxidation and
impurities, and willful neglect to procedures such as proper cleaning,
edge preparation and spacing of the joint to be welded. Within the
industry, some supervisors themselves ignore these procedures for the
sake of productivity. Although avoiding defects has its cost, repairs
can be even costlier, even more so when the repairs are not handled in
a systematic way.
The repairs to be covered here are based on the following defects,
which have been discovered by nondestructive testing: I) porosity, 2)
lack of fusion, 3) lack of penetration, 4) Wagon track, and 5) root bead
cracking, excessive penetration, and cold lap.
First Step in Repair

The first step in any report is to locate the defect and establish its
size. Even those defects that are visible to the eye must be evaluated
for size in order to avoid excessive removal of the base and weld metal.
Second Step in Repair

If a welding procedure already exists for the weld being made,
then it should be reviewed in terms of the following:
Material composition
Electrode classification
Joint preparation
Interpass temperature
Preheating temperature.
This procedural review helps establish those items that will be used
as part of the welding repair. It may not be necessary to undertake an
elaborate study if other welds in the piping system are satisfactory._
The failure of the weld may be due to anyone of the conditions mentioned above. However, it is important to realize that the weld has gone
through a cycle of intense heating and cooling; the internal or shrinkage stress can be quite high. If the pipe material is of medium alloy, or
heat resistant steel, then a repair procedure should be prepared to verify the steps needed for the repair. This procedure will include the fol-
217
Chapter 13
lowing steps;
Method of exploring the defect
Method of removing metal in and around the crack
Explanation of the applications of preheating and inter
pass temperature, both before and during welding
Postweld heat treatment
Electrode classification to be used in making the repair
Whether the weld will be supervised by anyone experi
enced in repairs
Use of qualified welders
Sketches showing the final shape and detail of the pre
pared area to be repaired
On most repair jobs, metal must be removed so that there is no evidence of discontinuity, creating, a sound base upon which to make
weld metal deposit. The following factors must be established:
a.
b.
c.
d.
e.
the type of alloy

the location, size, and depth
wall thickness
the propensity to harden
probing method
The question at this point might well be: Why is there so much fuss
over the repair ora simple weld? If the defect is discovered in a medium carbon steel pipe system, the repairs will not need the precaution
used with those materials alloyed for services such as high temperature and high pressure services. At the same time, preparation of a
defect, regardless of the material involved, is similar in many respects.
The repair procedure document has important reference information
about similar situations in which repairs were successful.
When repairing a defective root bead, one with cracking, the procedure will be to remove metal. It is important to establish whether the
crack was propagated into the second and third passes. If it did, then
the defect is more complicated. In this case, it is possible that that
crack was caused by shrinkage stress, complemented by part of the
root bead being too weak to withstand or accommodate the stress during cooling. However, there are still other reasons which could be
associated with such a defect.
For instance, the preheating temperature may have dropped lower
than that specified white initially welding the joint (fthe alloy mate~
rial is of the high hardenable type, the weld may crack. If there were
218

just a few passes, the cracks may not be visible to the naked eye
because the area is still at a high temperature and has not experienced
full contraction. Furthennore, cracks do not always develop instantly.
However, when the temperature falls, the microstructure of the hardened type develops and can lead to cracking.
When developing a welding procedure for weld repairs, the following factors must be considered:
The type of micro structure
Heat treatment prior to attempting removal of weld
metal for repair
Exploratory method for removing defective metal, with
the hope of not encountering transverse cracking
A crack that is restricted to the root bead can be looked at differently. Here, the shrinkage stress is the cause, perhaps complemented
by other factors such as stress risers, inadequate root opening that led
to lack of penetration, or poor restart when depositing the root bead.
In other instances, where the material on which the repair is to be
made is of a medium-alloy type, the procedure becomes more challenging. The welding procedure in the first instance (making the weld
initially) requires preheating and interpass temperature as well as a
normalizing or stress relieving postweld heat treatment. After welding,
if the postweld heat treatment was not already performed, then it will
be required before the repair exercise begins, and to be normalized
when the weld (repair) is completed.
In some cases involving a weld on a pipe with a heavy-wall thickness, even if preheating and interpass temperature were part of the initial welding procedure, both would be applied in repairing the weld. In
that case, the sequence will be changed because of the variables
involved.
Removing the metal by the carbon process (gouging) requires that
preheating should be conducted the same way it was during the initial
welding, due to terminal shock of the base metal. Gouging on cold
metal that has the potential to harden can create microcracks. This
effect is not different from that of welding an alloy that will harden
when cooling exceeds the critical rate,
If the mass is great and the joint is complicated, then there is a possibility of multiaxial stress. In this case, surface heating is not an
accepted part of the procedure. Heating should be slow so that the full
thickness of the section as well as a short distance surrounding the area
is uniformly brought to the required temperature.
219
Chapter 13
Repairs
Welding has been ignored in many ways and has often had to overcome a bad reputation. Too often a welder is requested to run a pass
over a surface crack which is not properly prepared or badly prepared,
yet is still expected to produce a sound weld. If the weld turns out to
be sound, it is because of sheer luck and the welder's ingenuity. In general, either the process or the electrode are to blame for such unsatisfactory workmanship. Often, it is not sufficiently considered that the
defective area was not prepared properly, that all the scale and oxide
were not properly removed, or that the welder was not instructed to use
preheating and interpass temperature.
To produce a good repair weld, one must follow a set of instructions
that are stipulated in the procedure, as follows
I.
2.
3.
4.
cleaning
preparation of the joint to be repaired
selection of the welding process and electrode
execution of proper preheating and interpass temperatures
Carbon arc gouging is actually part of the method used for removing
metal when repairing a weld. Therefore, care should be taken to ensure
that the temperature of the weld is held at the required level in order to
prevent cracking. This type of cracking is known as terminal shock
cracking. It is caused when the surface is heated at a rapid rate and then
allowed to cool at a rapid rate. In fact, pipes with greater mass (wall
thickness) actually quench the surface, which is expanded on heating.
The cooling at a rate of quenching can be so fast that the surface actually shrinks beyond the greater mass of the weld. The surface will
develop stress.
Removal of Defects
A defect should be clearly marked by the inspector or supervisor so
that the metal can be excised with a minimum of material loss.
Normally surface defects and those of reasonable depth can be removed
by grinding or gouging. Defects that go into the weld are removed by
the gouging method; the weld must be preheated to the required temperature and maintained as such. The preheating temperature is the
same as that used for making the weld.
There are two stages of repair: (a) removal of defects and (b) preparation ofthe defective area for welding. The cavity, which has been prepared by gouging, must now be dressed by a grinder or a burring tool
that can effectively remove the rough surface in the cavity. The repair
cavity must be wide enough so that electrode can fully access the defective area.
220
A weld that is fully prepared should be as follows:

I.
2.
3.
The prepared cavity surface must be smooth without

peaks and valleys.
All traces of oxide must be removed,
All spatter and globular deposits in the groove during
each welding pass should be removed repeatedly.
As shown in the figure below, the root opening is reestablished. This

part of the root bead must first be deposited, because the base metal is
that of a medium alloy composition. If the root pass was originally
deposited by the gas tungsten arc welding process, it should again be
applied here. However, if the machine to be used for welding has an
attachment for pulse arc, then it should be used instead to control bum
through. The second pass should be slightly wider than the first, using
the pulse arc. From there on, the shielded metal arc process can be used
to deposit stringer beads of a limited size. All slag particles and dust
should be removed from the cavity after each deposit. As shown in Fig
the figure below, the starting point should be approximately 3/16 of an
inch on the base metal. It should then be either ended or discontinued
the same distance on the base metal. Stringer beads should be used
until it is filled, and then two weave passes or a single, wider weave
can complete the weld.
.
Starting Point
Preparation of weld repair
221
CHAPTER
14
Fitting-Up Pipe
Fitting-up pipe is one of the basic skills of pipe welding. In simph
terms, fitting-up means to position the pipes in the correct location
as specified by the blueprint. The general procedure used to fit-up
pipes is basically a's follows:
1. Align the pipe or pipe fitting as closely as possible and hold it
in this position
2. Weld a single tack weld in place
3. Measure the location of the pipe or pipe fitting
4. If necessary, adjust the position of the pipe or pipe fitting until
it is in the specified location
5, Weld a second tack weld opposite the first tack
6. Check the location of the pipe or the fitting again and, if
necessary, adjust the position
7. Weld the two remaining tack welds in place.
Usually all or a number of joints are fitted-Up as described above,
before the complete root bead is welded, in order to keep the joints
flexible so that adjustments can be made in the positions of the pipes.
Each joint is held together by the four tack welds, unless it is
necessary to weld the entire root bead to enable the joint to carry the
weight to which it is subjected. Braces may also be used to support
the load.
While the general procedure for fitting-up is the same, regardless
of where the job is done, variations in which the details are performed are to be expected. For example, very heavy pipes and
fittings have to be lifted by cranes or by chain hoists, using chains or
steel cables to hold the pipes or the fittings. On the other hand,
smaller pipes can usually be held more conveniently by a helper
while the welder welds the first tack weld.
The pipe welder should know the basic methods of fitting-up and
should be skilled in their application. He should plan each job in
advance. By Joing this he can avoid costly errors. Fitting-up is
interesting work because it requires thought, planning, and often a
considerable degree of ingenuity, since it is seldom that two jobs are
the same.
222
Fitting-Up Pipe
Tools Required
For very heavy pipes and fittings, tackle is required for lifting and
holding them. Chains, steel cable, heavy-duty C-clamps, chain
hoists. and cranes are used, if available. Hammers of various sizes,
crowbars, and pinch bars are needed occasionally to bend the pipe
joints in order to align them. These tools should always be used
carefully so that they will not damage the pipe or the pipe joint.
Steel rules, steel tapes, spirit levels, protractors, and rafter squares
are indispensable tools for fitting-up pipe. Careless usage of these
tools must be avoided as it may impair their accuracy. A level that
does not read correctly or a square that is "out-of-square" can cause
serious errors. For this reason, levels and squares should be checked
occasionally.
A level is checked by placing it on a plane surface that is
horizontal, or nearly so. In Fig. 14-1: this is called POSITION 1 and
the encircled numbers identify the ends of the level. When in
POSITION I, carefully read the position of the bubble in the glass.
Then turn the level around to POSITION 2, Fig. 14-1. If the level is
POSITION 2
POSITION I
~;Ig. 14-1. fl.1ethod ot checkmg a level. The bubble should read the same when
the level is in either position shown.
reading correctly, the bubble will be in the same place but on the
opposite side of the glass in relation to the ends of the level. The
vertical and 45-degree glasses can only be checked by placing the
level on surfaces that are known to be vertical or 45 degrees. Most
levels have provision for adjusting the glass so that it will read
correctly.
The best method of checking a square is to hold it against a
square that is known to be true. Both squares should be placed on a
perfectly flat surface, as shown in Fig. 14-2A. Another method,
shown in Fig. 14-2B, is to set the square on a flat surface and against
another, smooth surface on which a line can be scribed. Scribe a line
uf;.ing the square and then turn it around and scribe another line in
tl1.e same place. If. the two scribed lines coincide. the square is true.
A square that 1S not true can only be corrected in a precision
machine shop. Usually it is less expensive to purchase a new square.
223
Chapter 14
Fig. 14-2. Methods of checking a square. A. Comparing a square 10 another

square that is known to be true; B. Scribing two lines with square blades resting
on table in opposite directions.
Pipe Fittings. A wide variety of standard pipe fittings is shown in

Fig. 14-3. These fittings are produced especially for welded pipe and
they are purchased with the ends beveled correctly in preparation for
welding.
Pipe-lo-Pipe Fit-Up
When there are two lengths of pipe that must be fitted-up together, they can be aligned by the methods shown in Fig. 14-4A
and B. A very effective method is illustrated in Fig. 14-4A, where an
ordinary piece of straight angle (or steel angle) is used as a gage to
check the alignment. The angle iron must be straight and the burrs
on the ends, caused by the saw cuts, must be removed.
The angle iron is first placed on one side of the two pipes,
bridoing the joint, and then it is placed against the opposite side of
224
Fitting-Up Pipe
PICTORIAL INDEX
rUBE.TU." W.JJi.g Fimffgs /I,,; FJIIlIg.S
I'
~(r'
naows
l""'ll_
'900'
rII
o..lod
RIDIJ(lIjD
mows
IMoa I.it<
n RlOWS
W1ONl911 o
911' IlBOWS
!.hortJl.... 'li
II
U" lltOWS
1....'-'"
#'It;;
ItO' IITlJRKS
lootR....
180 RUDIN!
SllortRocIi
illS
Slnig!lt ...
'~Oodl.t
Ii
W"IlMlWS
I... R...t
{lOSS~
Dot"',
$"qItt_ ""
(OIKUnRI(
IlDUCIllS
!C(INT.1e
.IDUCIlS
~ W!~IN(; NICK
(AM
~ $l1.ON IIANDIS
IIANDIS
JI,
-,
ell
LAP IllINT
STlJIIHDS
..&
.. I I
f8;
LAP JOINT fLANDIS
IATIRAlS
S..........
.-.,.;.., Ooi""
. . . otIIIU!D lLANGIS
SNAPIO KIPI'L!5
lUND !UNGIS
SIIlVII
~
$OCKIT TYPE
'
WttDING FUNGIS
SAODU1
. . 'IPUClNG lLANGIS
flllllN<IR<lEMfNT
SADDm
"'''1\
V ." )
IONO WILDING N!CKS
"MIDING RINGS
HIIlGIO (IOSU.1S
T8011
allSu.1S
0.111(1 flANGES
\..../
j"
'.
lARGI DIAMtIR
!LAHGIS
ANCNOR fOlIOING!
Cmrrlt'.I)' of til<' Tub'! Tum.! Dil'., Chl'IHt'rron Corp.
Fig. 14-3. Standard pipe fittings used for welded pipe.
the pipe. In each position an attempt is made to rock the angle iron.
When it will not rock in either position. the pipes are aligned.
Sometimes, the angle iron is held over the pipes while the tack weld
is deposited on the other side.
The second method, shown in Fig. l4-4B, is simply to bridge the
joint with a straightedge or with a blade of a rafter square. At least
two positions on the pipe, 90 degrees apart, should be checked in
this manner. While aligning the pipes in preparation for the first tack
weld, the correct root opening must be maintained. This can be done
by placing a piece of bent wire between the pipes as shown in Fig.
14-4C. Of course, the diameter of the wire must be equal to the root
opening.
The procedure for fitting-up two pipes is illustrated in a stepby-step manner in Fig. 14-5. This general procedure is also used to
225
Chapter 14
=-=-AfF=
~lt:=
Fig. 14.4. Method of aligning two pipes. A. Using an.gle iron to ali~n the pipe.s ;
B. Straightedge used to align the pipes; C. USing a piece of bent wire to obtain
the correct root open mg.
align the fittings, with appropriate variations, as will be explained

further on.
Welding Pipe Parts to Pipe
Flanges. Frequently, a flange must be welded to the end of a pipe, as

shown in Fig. 14-6. It must be centered with respect to the axis of
the pipe and the face of the flange must be perpendicular to the axis.
The boh holes in the flange are usually in multiples of four and it is
Gammon practice to weld the flange so that two bolt holes share the
uppermost position on the flange.
226
Fitting-Up Pipe
6. REPEAT STEPS 1THROUGH 5 TO

MAKE THE 3RD AND 4TH TACK
WELD,9()'l FROM THE POSITION SHOWN
Fig. 145. Step-by-step procedures for fitting-up
lWO
pipes.
The first step in fitting-up the pipe is to hold it in place as shown

in Fig. 14-6A Two bolts are inserted in the bolt holes and a spirit
level is placed so that it rests on the bolts. A helper holds the flange
in the center of the pipe so that the bubble of the spirit level is
centered. While the helper holds the flange in this position, the
welder deposits the first tack weld in place. If bolts are not available,
the level can be held as shown in Fig. 14-6B, with the top of the level
tangent to the two bolt holes.
227
Chapter 14
After the first tack weld has been deposited, the flange is carefully
aligned. It is aligned centrally with respect to the axis of the pipe by
checking the internal and external surfaces of the flange and the
pipe. If the pipe is known to be in a horizontal position, the flange
may be positioned perpendicularly to the pipe axis, in one plane, by
placing a spirit level against the face, as in Fig. 14-7. The flange is
aligned by tapping it with a hammer until the bubble in the vertical
glass on the level reads zero; however, it is necessary to protect the
flange against dents by placing a piece of soft metal or wood where
the hammer blow is struck. When the flange is aligned, the second
tack weld can be deposited.
Another method of aligning the face of the flange is illustrated in
Fig. 14-8. One blade of a square is placed against the face while the
B
Fig. 14-6. A. Aligning bolt holes on flange by placing the level on two bolts
inserted in bolt holes; B. Aligning bolt holes on flange with a level.
Fig. 14-7. Aligning a flange using a level. when the pipe is known to be in a
horizontal position.
228
-~
-""'-'
-----
Fitting-Up Pipe
Fig.. 14-8. Method of aligning a flange perpendicular to the axis of the pipe to
which it is to be welded.
other blade rests on the side of the flange. With the square in the
center of the pipe, two measurements are made between the surface
of the pipe and the blade of the square. When both measurements
are equal, the second tack weld can be made.
Before the two final tack welds are deposited the flange must be
aligned again, this time 90 degrees from where it was first aligned. If
the pipe is horizontal, a level cannot be used to check this alignment.
A square and a rule must be used, as shown in Fig. 1.4-8. The last
two tack welds can be deposited when both measurements are equal.
If the flange is to be attached to a pipe that extends out from a
vessel, as in Fig. 14-9, it must be welded in place a certain distance
from the vessel. To obtain this distance is largely a matter of welding
the pipe to the vessel correctly; however, the location of the flange
should be checked. This can be done by drawing a horizontal chalk
line along the side of the vessel that is the same height as the center
of the pipe, as shown in Fig. 14-9. The measurement is made by
placing a straightedge or the blade of a rafter square against the face
of the pipe and measuring between it and the chalk line.
The flange is held in the correct position by a helper while the
welder deposits the first tack weld. When this has been done the
flange must be checked and aligned before the second tack is
deposited by placing a level against the face of the flange, as was
shown in Fig. 14-7. After the second tack weld has been deposited
the straightedge, or square, is again placed against the face of the
flange and two measurements that are a distance apart are made
229
Chapter 14
Fig, J 4-9, Aligning the flange and measuring the distance from the face of the
flange to the vessel.
between the straightedge and the chalk line. When the flange has
been adjusted so that both measurements are equal, the remaining
tack welds can be deposited.
90-Degree Elbows. If the pipe to which the elbow is attached is

horizontal, the most convenient method of alignment is by using a
spirit level, as shown in Fig. 14-10. By placing the spirit level on the
face of the elbow, as shown in Fig. 14-lOA, the elbow is leveled
before the first tack is welded. After this tack weld has been made, it
is again checked with the elbow in this position and adjusted until
the bubble is centered. The spirit level then is placed in the position
Fig. 14-10. A. Aligning a 90-degrec elbow. using a spirit level. prior to deposit.
ing the first tack weld. B. Aligning a 90.degree elbow prior to depositing the
second tack weld.
230
Fitting-Up Pipe
shown in Fig. 14~ 1OB and the elbow is aligned until the bubble on
the level is centered. The bottom tack is then deposited. The elbow
should be checked in both directions again and realigned, if neces~
sary, before the two remaining tacks are deposited.
If the pipe is not exactly horizontal, the elbow can be checked for
crosswise alignment as shown in Fig. 14-IOA. For lengthwise alignment, a straightedge or the blade of a rafter square is used, as in Fig.
14-11. When two measurements between the straightedge and the
center of the pipe, taken a distance apart, are equal, the elbow is
aligned. The checking, aligning, and welding are done as before.
Fig. 14-11. Method of aligning a 90-degree elbOW using a straightedge or rafter

square.
Fig. 14-12. Checking. the alignment of a 45-degree elbow with a spirit level
when the pipe to which it is to be attached is horizontal.
45-Degree Elbows. When the pipe to which the 45-degree elbow is

attached is horizontal, the most convenient method of aligning the
elbow in the direction parallel to the pipe is to use a spirit level, as
shown in Fig. 14-12. The elbow is aligned when the bubble in the
45-degree glass on the level is centered.
If the pipe is not horizontal, as in Fig. 14-13, a protractor level
can be used. The bubble in the glass is centered when the protractor
is resting on top of the pipe and the reading, in degrees, is obtained.
231
Chapter 14
Forty-five degrees are added to this reading and the protractor is
adjusted to this position. For example, if the reading on the pipe is
12 degrees, the protractor is adjusted to read 57 degrees (12 0 +
45 0 = 57). Adjusted to this setting, the protractor level is placed
on the face of the elbow, as shown, and the elbow is aligned when
the bubble is centered. Sometimes a flat piece of steel must be placed
across the elbow, on which the protractor can rest.
5.1" REAl>ING
Fig. 14-13. Method of checking the alignment of a 45degree elbow with a

protractor level when the pipe is not perfectly level.
[SCRIBE REFERENCE LINE!
Fig. 14- 14. Method of checking the crosswise alignment of a 45-degree elbow.
A. Laying out the reference line; 8. Method of supporting the level to align the
elbow.
232
Fitting-Up Pipe
Frequently, the 45~degree elbow is aligned crosswise, with respect
to the axis of the pipes, by visual estimation. A method that can be
used is shown in Fig. 14-14. The pipe is shown lying on a flat surface
in Fig. 14-14A, with a square positioned against the face. Chalk is
rubbed on the face and a reference line then is scribed across the
face, using the blade of the square as a guide. The chalk makes the
scribed line more visible. Two shallow center-punch marks are often
punched at each end of the reference line and the angle iron is
aligned with respect to these marks.
When the elbow is to be aligned against the pipe, the spirit level is
placed on an equal-legged angle iron (3 x 3 x lf4 in. or 1'/2 x Ilh Xlf8
in.) and the angle iron is positioned parallel to the reference line or
the center-punch marks on the face of the elbow, as shown in Fig.
14-14B. With the level in this position, the elbow is aligned when the
bubble is centered.
Fig. 14-15. A. Aligning a tee prior to depositing the first tack weld; B. Aligning
the tee prior to depositing the second tack weld; and C. Alternate method of
aligning the tee.
Chapter 14
Tees. A spirit level can be used to align a tee when the pipe is
horizon taL As shown in Fig. 14-15, the level is placed crosswise and
lengthwise on the upper face of the tee to align it in these two
directions.
When the pipe is not exactly horizontal, the crosswise alignment is
obtained by using the level as shown in Fig. 14-15A. Lengthwise
alignment is obtained by using a rafter square, as in Fig. l4-1SC. A
blade of the square may be held firmly to the end face or to the top
face of the tee, and two measurements are made between the top of
the pipe and the square. When the two measurements are equal, the
tee is aligned in the lengthwise direction.
When the branch pipe to which the tee is to be connected is
horizontal, the tee must be positioned as shown in Fig. 14-16. In this
case, the tee can be aligned by placing the spirit level against one
face and then the other face of the tee, as shown.
Fig. 14-16. A. Aligning a tee with a spirit level prior to depositing the tirsltack
weld; B. Aligning the tee prior to depositing the second tack weld.
Vertical Pipe Alignment. The procedures for aligning vertical pipes

are shown in Fig. 14-17. Short runs of vertical pipe can be aligned
by using a level, as shown in Fig. 14-17A. Two positions on the pipe
that are 90 degrees apart should be checked in this manner to make
certain that the pipe is truly vertical. The pipe should be checked in
this manner before depositing each tack weld. A piece of bent wire
can be used to obtain the root opening when the pipe is in this
position.
Longer lengths of vertical pipe can also be checked with a level.
The pipe should be checked in two places around the pipe as before,
but this must be done in several locations along the length of the
pipe. A procedure that is frequently used to check long runs of
vertical pipe is shown in Fig. 14-178. A plumb bob is attached to
the end of a line hanging from a convenient place near the top of the
pipe. The distance between the wall of the pipe and the line is
234
Fitting-Up Pipe
Fig. 14-17. A-I and A-2. Aligning a verucaT pipe with a spirit level; B. Aligning
a vertical pipe with a plumb bob.
measured in several places. When these measurements are all equal,

the pipe is vertical in the direction shown.
Perpendicular Pipe Alignment. Frequently, a pipe must be welded
perpendicular to another pipe. The elbow alone cannot be depended
upon to provide the necessary alignment. A typical example of the
method of aligning the pipe after the first tack weld has been made is
shown in Fig. 1418. The rafter square is held firmly against the
Fig. 14-18. Method of aligning two pipes perpendicular to each other.
235
Chapter 14
Fig. 14-19. Another method of aligning two pipes perpendicular to each other.
horizontal pipe and several measurements are made between the

blade of the square and the surface of the pipe. By lifting the end of
the horizontal pipe, it is adjusted until the measurements are all
equal, after which the second tack is welded in place.
As shown in Fig. 14-19, the measuring procedure could have
been reversed, with the square held against the vertical pipe and the
measurements taken between the blade of the square and the horizontal pipe. However, in this illustration the other end of the
horizontal pipe is tack welded in place before the second tack weld is
deposited on the elbow. Before the other end of the pipe is tack
welded in place, the square is used, as shown, to make certain that
the pipes attached to the elbow are 90 degrees apart. By this
procedure, the horizontal pipe is held firmly in the correct position
before the remaining Hick welds are deposited on the joint at the
elbow.
236
CHAPTER
15
Qualification of the Welding

Procedure and the Welder
On many welding jobs, the fabricator is required to separately
qualify the welding procedure to be used and the welder who is
actually to weld the pipe. How this is done is specified by a code.
One code frequently encountered in pipe welding is the ASME
Boiler and Pressure Vessel Code, Section IX. While this code is not
reproduced here in detail, the discussion in this chapter is based
largely on the code.
Qualifying a Procedure
To qualify a procedure, the fabricator must prove the adequacy

of his procedure with consideration given to the service conditions,
particularly to the consequences should the piping system fail. A
pipe weld is made according to the procedure to be qualified.
Samples are cut from the weld; the positions from which they are cut
are shown in Fig. 15-1. Figure 15-1 also specifies the dimensions of
the test samples and the type of test to which they are to be
subjected. After the weld specimens have proven to be acceptable,
the procedure is considered to be valid.
Qualifying the Welder
In order to qualify the welder, the fabricator must have him weld
a pipe joint in both the horizontal (2G), and the vertical (5G),
positions, using the procedure that has been qualified for doing the
job. Four samples, Fig, 15~2, are cut from each of the welds and
bend test samples are prepared. As shown in Fig. 152, two face~
bend tests and two root-bend tests are made from each weld.
The fabricator must maintain a record of the tests made by each
welder, who is assigned an identification number for use as a
reference. No welder is allowed to perform on the job unless he has
been successfully qualified by passing the bend tests.
Passing one particular procedure will not qualify the welder for
welding all different pipe sizes and wall thicknesses, but will restrict
the operator to welding within a specific range of wall thickness. If
237
Chapter 15
MACHINE THE MINIMUM AMOIJNT
NG:OfD TO DBT""N PLANE
PAAAlLEL FACES OIll'R THE
3i4" WIDE ~DUCED SECTION.
2" R
1/4
'
REDUCED SECT ION
THESE mGES MA' eE flAME CUT
I'
I
L
6" MIN.
IJ
}J
]~
_
FACE BEND SPfC IMEN
TENSILE
I-
6" MIN.
li
_J~l'lt t----- --u---------J 4T

u
ll-liS'" MAX.
ALL CORNERS
ROOT BfND SPECIMEN
R-1fS" MAX.
All CORNEllS
Fig. 15-1. Specification of test specimens for qualifying the pipe welding procedure.
tests are conducted on pipe with %-inch wall thickness, the operator
is then qualified to weld wall thicknesses one-half the size of th~
sample, or twice the sample size. Therefore, before attempting to
weld heavier wall pipe, the welder, and procedure, must have new
qualification.
A welder shall be requalified when there are one or more changes
made, which differ from that of the already established procedure.
The following changes are responsible for new qualifications:
A. A change from a base metal listed under one letter subgroup
to one listed under another subgroup.
B. A change from one diameter wall-thickness range to another.
C. The addition of any welding position in which a large number
of welds is made.
D. If the direction of welding is changed from bottom to top, or
top to bottom.
There are other changes which are not quite related
process:
to
this
E. An operator who is qualified with an E-60 lOis not necessarily

permitted to use a low-hydrogen electrode. If he is qualified with a
low-hydrogen electrode, it is possible for him to use an E6010 if the
job permits.
238
Qualification of the Welding Procedure and the Welder

ROOT BEND
ROOT BEND
FACEBENO
ROOT BEND
VERTICAL (5Gl WELD
FACE BEND
SPECIMEN
uu
ROOT BEND
FEMALE DIE
FACE BEND
Fig. 15-2. Bend test specimens for qualifying the pipe welder. The weld must
bend 180 degrees without breaking.
Qualifying welders according to a procedure is the responsibility

of a qualified inspector. An inspector not only inspects the completed weld samples, but also conducts this test in a systematic
manner. There are three basic steps in testing:
1. V isual inspection
2. Acceptance quality
3. Final result.
Perfect Weld Requirements
A. The pipe nipples and the edges to be welded should be free

from oil, grease, etc.; after they are tacked together, the root
openings are inspected. During the course of welding, each layer of
welded metal is checked for porosity, undercut, slag inclusion, and
surface roughness. An inspector has the responsibility of terminating
any particular test at any time if he finds that the operator does not
have the skill required to produce satisfactory welds,
B. When a test weld is completed, further visual inspection is

conducted to see that the weld is free from cracks, that there are no
239
Chapter 15
arc strikes on the pipe surface, that the melHhrough does not exceed
% inch, that concavity of the root bead does not exceed IA6 inch,
that incomplete fusion is not present, that restarts and tie-ins do not
exceed 1/16 to lf64 inch in cavity, and that the cover pass is uniform
and smooth.
C. Having passed the visual inspection, four specimens are

extracted from the welded pipe as shown on page 227, comprising
two face bends and two root bends. Those bends must be inspected
for the following:
A Slag inclusion
B. Cracks
C. Incomplete fusion between the welded metal and the base
metal, and between any of the weld passes.
Bend Tests
If the welder is to pass the guided bend test, the specimen shall
have no cracks or other defects exceeding Vg inch in any direction,

with the exception of corner cracks. However, if corner cracks are
caused by lack of fusion, slag inclusion, or other defects, the welded
specimen can be rejected.
A complete record is kept on each welder if the specimens have
passed the described test. Procedure and qualification tests for each
specimen, including tensile and bend tests, are recorded in detail.
In addition, a written document is kept, showing:
A Specification of material being welded
B. Specification of filler metal

C.
D.
E.
F.
G.
H.
Edge preparation
Preheat and postheat, if being used
Volts and amps
Spacing or root opening
Nurn ber of passes
Size of electrode and classification
I. Wall thickness and diameter of pipe being tested
J. The test conditions, open air or closed shop.
All this and other important information is documented and made

available to supervisors and inspectors.
It is important to note that when tests are conducted on pipe with
wall thickness exceeding % inch, the test specimen is sectioned out
differently from those in Fig. 15-2. Bend specimens are prepared
240
Qualification of the Welding Procedure and the Welder
SIDE SEND SPECIMEN

(FOR PIPE OVER 318" THICKl
St.andard Dimensions, Inc.

(a)
0.505
Specimen
lb)
0.353
Specimen
See Note 4
A-Length of reduced See Note 4
section
O.500O.OIO 0.3500.OO7
DDiameler
318. min
R-Radius oftillet
lf4. min
13/s, approx
Ills. approx
s- Length of end
section
3/.
C- Diameter of end
'I.
section
Ie)
,d)
0.252
Specimen
0.188
See Note 4
Specimen
See Note 4
0.250 0.005 O.1880.003

J/s. min
:VIe. min
%. approx
'I. approx
3fs
'I.
NOTES:
(I) Use maximum diameter specimen (a). (b), (c), or (d) that can be cut from
the section.
(2) Weld should be in center of reduced section.
(3) Where only a single specimen is required the center of the specimen should
be midway between the surfaces.
(4) Reduced Section "A" should not be less than width of weld plus two "D."
(5) The ends may be of any shape 10 fit the holders of the Testing Machine in
such a way that the Load is Applied Axially.
Fig. 15-3. Test specimens for thick-wall pipe.
in a way which allows them to be bent sideways. The tensile specimen is also prepared differently for thick-wall pipe (Fig. 15-3).
It is also important to note that when welded specimens are being
prepared for bend tests, the surface is smoothed by grinding which
should be done in the direction shown in Fig. 15-4.
241
Chapter 15
For information concerning The American Petroleum Institute Butt Weld Procedure Qualification Test (API) 1104
see the latest edition of the API Standard 1104, for sale by
WELD SPECIMEN - - . . ,
UPANO DOWN
MOVEMENT
fCig. 15-4. Method of grinding the weld of the lest specimen.
the American Petroleum Institute, 1271 Avenue of the Americas,

New York, N.Y. 10020.
The points covered in this limited discussion on Inspection and
Qualification Procedures will vary over a period of time. Inspections
and qualifications are ~hangeabfe and kept up-to-date. Therefore, it
is always necessary to consult the,. approoriate code (ASME
BOILER AND PRESSURE VESSEL CODE, SECTION IX).
242
CHAPTER 16
General Welding Safety

After a half century, commercial welding has proved itself not to
be injurious to health. However, as in most trades, if a welder is
careless, some features of the welding process can cause discomfort
and actual danger.
Essentially, welding is not a hazardous occupation if proper
precautionary measures are always observed. This requires continuous awareness of the possibilities of danger and habitual safety
precautions taken by the welder. In addition, it requires that the
supervisor be alert, responsible, and stringent in enforcing safety
regulations.
SAFETY MEASURES
1. Always wear dry, protective, fire-resistant clothing, cuffless

trousers covering the shoe tops, leather gloves, jacket, apron, and
proper dark lenses.
2. Always keep a safe. clean work area.
3. Make sure there arc no flammable materials nearby.
4. Do not weld in the vicinity of explosive materials or carbon
tetrachloride.
5. Always make sure there is enough ventilation to give three or
four complete changes of air per hour.
6. Use air exhaust at the weld whenever welding lead, cadmium,
chromium, manganese, brass, bronze, zinc, or galvanized metals.
7. Never weld or cut in a confined area without ventilation.
8. Keep all welding equipment in good condition.
9. If it is necessary to couple lengths of cable, make sure joints
are insulated and all electrical connections are tight. Use no cable
with frayed, cracked, or bared spots.
10. When electrode holder is not in use, hang it on welding
machine or special holder. Never let it touch a gas cylinder.
11. Always have welding machine properly grounded, usually to
cold-water pipe.
12. Make sure pedal controls are guarded to prevent accidental
starts.
243
Chapter 16
13. If need arises to weld in damp or wet conditions. wear rubber
boots and/or stand on dry cardboard or wood.
14. Stand only on solid items, floor, or ground.
15. When welding in high places without railing, use safety belt
or lifeline.
16. Always wear proper eye protection, especially when grinding
or cutting.
17. Keep booth curtains closed to protect the eyes of others.
18. Never weld or cut directly on a concrete floor.
19. When using a water-cooled torch, check for water leakage.
SAFE HANDLING OF GAS CYLINDERS
I. Be very careful when moving any gas cylinder never
handle roughly. Always have cylinders capped when moving. Never
roll cylinders horizontally.
2. Never use welding gas as compressed air for blowing away
dirt or debris.
3. Before attaching a regulator to a cylinder, open and close the
valve quickly. This is commonly called "cracking" the cylinder.
4. Open valve on cylinder slowly after regulator is attached.
5. Be sure all connections are clean and gas-tight. Check with
saliva or soapy water.
6. When the regulator is not in use, the adjusting screw should
be screwed out until diaphragm is free.
7. Always protect the hose from rupture or mechanical damage.
8. Always close the cylinder and rekase the pressure from the
regulator and hose when work is done.
9. Always leave safety plugs alone.
10. Always keep the cylinders in an upright position.
11. Always mark the cylinders that have been used "Empty,"
or uMT."
12. Never open tank valves until you are certain that regulator
valves are closed finger-tight.
13. Never open the valves on the cylinders with a hammer.
14. Never screw the regulator screw in tightly against the regulator.
15. Never use a cylinder, even when empty, as a roHer.
16. Do not store cylinders in a room where the temperature is
higher than 80 degrees F.
244
Gereral Welding Safety

SAFETY REGULATIONS FOR OXYACETYLENE WELDING
1. Check all connections before lighting the torch.

2. When the torch flashes back, or is burning on the inside, turn
both cylinders off immediately - first the oxygen and then the
acetylene.
3, Extra tips should be put away when not in use. Never put
them in a drawer with other tools; this may damage the seat of the
tip.
4. Always use the right-size tip for cutting and welding.
5. Always turn the oxygen cylinder valve all the way open.
Open the acetylene cylinder valve no more than one turn. One-half
turn is preferred.
6. Always place the welding tip so that it points to the side of
the torch to which the acetylene hose is attached.
7. Always weld at least 5 feet away from the cylinders.
8. Do not use any oil or grease on any oxygen or acetylene
connections.
9. Never hammer on oxygen or acetylene regulators.
10. Do not light a torch with a match or an open flame. Use a
striker.
11. Before lighting torch, be positive that hose, tanks, or any
inflammable material will not be exposed to heat, flame, or sparks.
12. Beware of high acetylene pressure. Never use acetylene gas
when the pressure is greater than 15 pounds per sq in. (acetylene gas,
when compressed to more than 15 pounds per sq in., becomes a very
high explosive).
13. Do not hold welding or cutting tip too close to the work.
14. Never use a tip that gets too hot.
IS. Never use a torch that leaks.
16. Never leave the torch burning and go away from it.
17. Never leave torch valve open.
18. Do not adjust, alter, change, build, or do any experimental
work on cylinders, regulators, torches, or any other gas equipment.
t 9. Never weld a closed or jacketed vessel without an air vent.
Also, never weld a vessel which has contained any explosive or
flammable material until you are positive that it has been thoroughly
emptied and purged, then use extreme care.
20. Always write the word "HOT" on workpieces that are very
hot, especially when they are located in work areas frequented by
other workers.
21. Always coordinate movement, when shifting large sections of
fabricated pipe, if two or three people are working together.
245
Chapter 16
22. Never congest work area with completed jobs, so that evacuation can be quick in case of extreme danger.
23. Always secure pipe from rolling during fabrication and storing.
24. When fabricated sections are in their working position, they
should be secured by tacking or bracing properly before the load is
released.
25. When removing a single link of pipe from a pile or pipe rack,
always start from the very top to avoid the cluster of pipe link from
rolling off
26. Avoid resting tools (hammer, chisel, small pieces of metal,
etc.) on pipe, and when working above ground level, to avoid injury
to those working on a lower level by falling metal or tools.
246
Index
Alignment, perpendicular pipe 235. 236
vertical pipe, 234, 235
Alloying elemenls in steel, 185- J88
boron, 188
carbon, 186
chromium, 187
cobalt, 188
manganese. 186
molybdenum, 187
nickel, 187
silicon, 186, 187
titanium, 188
tungsten, 188
vanadium, 187
Aluminum as a deoxidizer in steel, 188
"Amperage," 10
Arc length, 12, 13
Arc strike cracking, 207, 208
Austenite, 180
Bainite, 180, 181
Base metal - Hiler metal combinations, 75-77
Bend tests, 240-242
Boron as an alloying element in steel. 188
Brittleness, (def.) 174
Carbon as an alloying element in steel, 186
Cementite, [77,178
Chromium as an alloying element in steel, 187
Cleaning the joint surfaces, 23
Coating, electrode. 12
Cobalt as an alloying element in steel. 188
Codes and standards, 8, 9
Complicated pipe joints, full lateral joint in,162-165
intennediate and cover passes in, 160, 161
rool beads in, 158-160
special precautions ill welding, 168
T-joint with intersection at side, 166, 167
T-joint with intersection on bottom, 166
T-joint with intersection On top, 165
welding, 158169
V-joint, 167
Copper added to steel, 188
Cover passes, in horizontal pipe welding, 148, 149
in mUltilayer welding, 91-93
in Otinwall pipe welding, 122-23
in welding complicated pipe joints, 160, 161
Cracking, arc strike, 207, 208
restraint, 52, 209211
underbead 208, 209
Cryogenic services, 156, 157
Current flow. 10
Current setting in multilayer welding, 84-86
Defects in pipe welding, 207-221
Distortion in pipe welding, 198206

preventing, 206
Distortion in welds, 193- J97
Downhill pipe welding, 3-7
of thinwall pipe, 102106
heavy wall and large diameter, 126128
DUClility, (def.) 174
Electric arc welding, basic concepts of, 10, II
Electrode coating, 12
Electrodes for GTAW process, 62,63
Electrodes for multilayer welding, low-hydrogen, 93, 94
Elements in steel, alloying, 185188
Fatigue, 174
Faults and their prevention in welding, 211-216
Ferrite, 177
Fifth, sixth, and seventh passes
pipe welding. 148, 149
Fittmgs, pipe, 224
Fitting-up pipe, 23, 24, 222-236
Fit-up, pipe-to-pipe. 224-226
poor, 45-49
Flanges to pipe, welding, 226-230
Flat welding the root bead, 4041
45-degree elbows to pipe, welding, 231-233
Full lateral joint in welding complicated
pipe joints, 162-165
Gas cylinders, safe handling of, 233, 234
Gas tungsten arc welding (GTAW), 57-58
welding the root bead by, 5780
Grain size, 182184
Grain structure, 175-177
GTAW, equipment for, 59-63
shielding gas for, 59, 61
torch for, 62
welding machine for. 59
GTAW process,
f;aution, 74, 80
cover passes, 132-135
depositing root bead, 128-129
downhill pipe welding, 126-128
electrodes for, 62, 63
features of, 57
filler passes, 129-132
Heavy-wall pipe welding. 126-135
making a tie-in in, 74, 75
preparation of the weld joint in, 65
procedure for welding the root bead in, 65-70
pulse current, 72-74
rool bead, 7280
second pass in, 72
shielding the weld metal in, 6365
stop and restart in, 70, 71
247
Index
Handling of gas cylinders, safe, 244
Hardenability, 173
Hardness. (def.) 114
Heat distribution, 14, 1520
Heat input, 14, 15
Heat withdrawal from weld zone, I5
High temperature services, alloy steel in, I54-156
Horizontal pipe welding, 8, 136-157
coverp~esin, 148. 149
fi fih, sixth, and seventh passes in, 148, 149
making a tie-in in, 14 I
of the root bead, 137-14Q
of thinwall pipe, 104
poor fitup in, 141143
second pass in, 144146
stop and restart in, 140
third and fourth passes in, 146-148
using lowhydrogen electrodes, 143
Hot pass in thin-wall pipe welding, 115-119
Nickel as an alloying element in steel, 187

9O-degree elbows to pipe, welding, 230, 231
OXYllC\:tyJene welding, safety regulations for,
245,246
1nelusions of slag, 212214
Incomplete penetration, 216

inspecting the weld in uphill welding the rootbead,
44-52
intermediate and cover passes, in welding

complicated pipe joints, 160, 161
on heavy-wall pipe, 8199
intermediate passes, in multilayer welding, 86-89
in thin-wall pipe welding, 119122
iron-carbon diagram, 181,182
Keyhole, 27, 28
in uphill welding the root bead, 33, 34
Lack of fusion, 216
Large diameter pipe welding, 126-135
depositing root bead, 128-129
downhill pipe welding, 126-128
Lead added to steel, 188
Low temperature serv ice, 156, I57
Malleability, 174
Manganese as an alloying element in steel, 186
Martensite, 179
Metallurgy, welding, 170-197
Metals, properties of. I 70-174
size changes in, 184, 185
Metal transfer, 11, 12
Molybdenum as an allOying element steel, 187
Multilayer welding, 8182
current setting in, 84-85
intermediate passes in, 86-89
low-hydrogen eleclfodes for, 93, 94
stop and restart in, 89-91
the cover pass in, 91-93
tie-in in, 91
Multiple layers, 81-82
Pearlite, 178
Perfect weld requirements, 239, 240
Phosphorus in steel, 188
Pipe fittings, 224
Pipeline welding, outdoor, 104, 105
Pipe parts to pipe, welding, 226-236
Pipe-IO-pipe fit-up, 224226
Poor fitup, in horizontal pipe welding, 141143
narrow root opening, 48, 49
root face 100 thin, 49
root face too wide, 49
wide and narrow rool openings, 48, 49
wide root opening I 4548
Porosity, 215, 216
Positions for pipe welding, basic, 1,2
Precautions in welding compl icated pipe joints, 168
Preparation of the pipe joint, 21-30
Preparation of the rool bead in multilayer
welding, 82-84
Preparation oflhe weldjoinl in GTAW pro-cess, 65
Preparing pipe edges, 2123
Preventing distorticm in pipe welding, 206
Prevention of welding faults, 211216
Procedures, bask pipe welding, 3-8
for fitting-up pipe, 222
qualifYing, 237
Properties of metals, 170-1 74
Pulse current, root bead, 72-74
Qualification of the welding procedure and
the welder, 237242
Relieving of stresses, 188, 189
Requirements for a perfect weld, 239, 240
Residual stresses and distortion in welds, 193-197
Restart in uphill welding the rool bead, 37, 38
Restraint cracking, 52, 209-211
Rool bead, flat welding the, 40-41
in complicated pipe joints, 158160
in uphill welding on heavy-wall pipe, 31-34
vertical uphill welding of the, 3839
Root bead, GTAW process, 7280
pulse current, 72-74
Root bead in multilayer welding, prepara-tion of
the, 8284
Root bead in uphill welding, starting the. 34-36
Root bead welding, summary of, 52-56
thin-wall pipe, 103108
with low-hydrogen electrodes, 49-52
248
Index
Safe handling of gas cylinders, 244
Safety, general welding, 243-246
Safely regulations for oxyacetylene welding, 245,246
Second half of the pipe joint, welding the, 41
Second pass in horizontal pipe welding, 144-146
Shear stress, 172
Shielded metalarc welding. 10-13
electrode, 94-99
Shielding gas for GTAW process, 59, 61
Shielding the weld metal in GTAW process, 63-65
Silicon as an alloying element in steel, 186, 187
Size changes in metals, 184, 185
Slag inclusions, 212-214
Spheroidite, 180
Standards and codes, 8, 9
Standard symbols designating welding posi-lions, 2
Starting the root bead in uphill welding,3436
Steel, structural characteristics ot: 177-181
Stop and restart, in GTAW process, 70, 71
in hori7,Ontai pipe welding, 128, 129
in multilayer welding, 89-91
in uphill welding the root bead, 37, 38
Stopping the weld, 28-30
Strain, (def.) 172
Strength, 172, 173
ultimate tensile, 173
yield, (def.) 173
Stress, (def) 17I, 172
shear, 172
torsional, 172
Stresses and distortions in welds, residual, 193-197
Stress relieving, 188, 189
Structural characteristics of steel, 177-181
Structure of the weld, 189-193
Structure sleel 78, 79
Sulfur added to steel, 188
Symbols designating welding positions, 2
Tack welding, 25-27
thin-wall pipe, 106-110
Tees to pipe, welding, 234
Tensile strength, ultimate, 173
Tests, bend, 229-231
Thin-wall pipe welding, 100-135
cover pass in, 122123
downhill, 102106
horizontal, 104
pass in, 115, II 6 joint tor, 105,106
Thin-wall pipe welding, intermediate passes in,
uphill, 100, !O I
welding the root bead in, 110114
with pipe axis at 45-degree POSilion,124-125
Third and fourth passes in horizontal pipewelding,
146-148
Tie-in, in GTAW process, 72
in horizontal pipe welding, 141
in multilayer welding, 91
procedure, 41-44
TIG (tungsten inert gas) welding, 57-58
Titanium as an alloying elemen! in sleel, 188
T-joint, in welding complicated pipe join Is, 164-167
with intersection at side in welding com-plicated
pipejoinls, 166,167
wilh intersection on top in welding com-plicated
pipejoints, 165
Tools required for fitting-up pipe, 209-2 II
Torch for GTAW process, 62
Torsional stress, 172
Toughness, (def.) 174
Tungsten as an alloying element in steel, 188
Ultimate tensile strength, 173
Underbead cracking, 208, 209
Undercutting, 211,212
Uphill pipe welding, 7, 8
wilh pipes in an angular (6G) position, 161,162
Uphill welding the root bead, 36-38
keyhole in, 34
on heavy-wall pipe (50 position), 31-48
vertical, 38-39
UphHl welding thin-wall pipe, 100, 101
Vanadium as an alloying element in steel, 187
Venical uphill Welding of the root bead, 3839
Voltage, 10
"Weaving," 20
Weld-affected zone, 190-193
Weld repair, 217-221
"Whipping." 20, 28
"Whipping and pllllSe," 39
Widmanstatten structure, 179
Withdrawing heat from weld zOne, 15
Yield strength, (def.) 173
V-joint in welding complicated pipe joints, 167
119-122
poor fitup in, 124
preparation for the hot pass in, 115, 116
preparation of the pipe joint for, 105, 106
stop and restart in, 1114, 115
tack welding, 106-110
tie-ins in, 115
249

Pipe Welding Procedures

Uploaded by

Copyright:

Available Formats

Pipe Welding Procedures

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pipe Welding Procedures

Uploaded by

Copyright:

Available Formats

PIPE WELDING

Library of Congress Cataloging-in-Publication Data

Industrial Press Inc.

Pipe Welding Procedures

Copyright 2003. Printed in the United States of America.

2. Essentials of Shielded Metal-Arc Welding Technology

3. Heat Input and Distribution

4. Reperation of the Pipe Joint

5. Uphill Welding the Root Bead on Heavy-WallPipe

6. Welding the Rood Bead by the Gas Tungsten Arc

7. The Intermediate and Cover Passes

8. Welding Thin-Wall Pipe

9. Horizontal Pipe Welding (20)

10. Welding Complicated Pipe Joints

11. Introduction to Welding Metallurgy

12. Distortion in Pipe Welding

13. Pipe Welding Defects

14. Fitting-UP Pipe

15. Qualification of the Welding Procedure & the Welder 237

Foreward from First Edition

This book is intended to help meet this need by assisting in the

Finally, the author would like to gratefully acknowledge the

Fig. I-I. Example of a high quality pipe weld.

learned by reading a book alone; however, if incorrect techniques

FLAT POSIT ION

liE RTlCAL posmON

PLATES VERTICAL AND

PLATES A'lO A)<IS

PIPE SHALL NOT BE

Count'syo!th.. Nobart Broth"" Cu.

Fig. ] -2. Standard symbols designating the welding positions.

The type of metal to be welded

Basic Pipe Welding Procedures

When the pipe is in the SG position, with its axis horizontal as in

Downhill Pipe Welding. Regardless of the method used. the pipes

desirable because a softer and more ductile grain structure is then

Fast moving welding crew.

Cross country pipeline laying.

Horizontal Pipe Welding. When the pipe is in the 2G position, with

by cross~country pipelines. In the construction industry pipe welded

Essentials of Shielded Metal-Arc

Essentials of Shielded Metal-Arc Welding Technology

Metal Transfer. Of particular importance is that the pipe welder

Electrode Coating, Almost aU modern electrodes are coated. The

MOLTEN WELD METAL

Essentials of Shielded Metal-Arc Welding Technology

PROPER ARC LENGTH

Heat Input and Distribution

The heat input is controlled primarily by the current setting. To

Heat Input and Distribution

The distribution of the heat and the rate at which it is withdrawn

Conductivity of the work material

When compared to most other materials, ali of the metals are

WELD QUALITY INSPECTION

SMOOTH, WELl FOIlMEO BEAD

COMMON WELDINO MISTAKES

TOO MUCH SPATTBl

ARC TOO LONG

BEAD TOO SMAlL,

Fig. 3-1. Weld quality inspection.