MODULE 21

Welding Metallurgy
of Steels

8260 Parkhill Drive, Milton ON L9T 5V7

Tel: 1-800-844-6790 Fax: 905-542-1318
www.cwbgroup.org

® Copyright 2010 The CWB Group industry Services

Revised July 2010 | All rights reserved

The works of authorship contained in this document, including but not limiled to all design, text and images, are the property,

except as otherwise expressly stated, of the CWB Group-Industry Services or one of its affiliates, the CWB Canadian Welding

Bureau and QUASAR, This document, whether in current form or electronic, may not be copied, reproduced, transmitted,

displayed, performed, distributed, rented, sublicensed, altered, stored for subsequent use, or otherwise used in whole or in

part in any manner without the prior written consent of the CWB Group industry Services, except to the extent that such use

constitutes "fair use" under the Copyright Act of 1976 as amended, and except for a temporary copy in a single computer's

memory and one unaltered permanent copy to be used by the viewer or student for persona! and non-commercial use only.
m
Table of Contents

Welding Metallurgy of Steels

1. Introduction
1.1 Objectives

2. Steel Metallurgy
2.1 Crystal Structure
2.2 Solubility of Carbon
2.3 Iron-Iron Carbide Phase Diagram
2.4 Transformation Diagrams
2.5 Heat Treatment

3. Properties of Steels
3.1 Effect of Carbon
3.2 Effect of Thickness
3.3 Normalizing
3.4 Alternate Methods of Strengthening

4. The Heat-Affected Zone

4.1 Structure
4.2 HAZ Grain Size
4.3 Hardening Curves
4.4 HAZ Toughness
4.5 Hydrogen Induced Cold Cracking HICC In the HAZ
4.6 Stress Relief Cracking Reheat Cracking

5. Weld Metal
5.1 Structure
5.2 Weld Metal Composition
5.3 GMAW Electrodes
5.4 Fluxes
5.5 Properties of Welds
5.6 Hydrogen Induced Cold Cracking HICC in the Weld Metal

6. Summary

7. Additional Resources

Guide and Exercises

m
 MODULE 21 ! Welding Metallurgy of Steels

1. Introduction
The increasing sophistication of steel property requirements—especially that of fracture toughness—has
resulted in the development of a new generation of steels made through complex processing routes.
These new requirements and materials present Important challenges to the welding engineer, who must
select welding consumables and develop procedures that will give specific properties and at the same
time avoid cracking problems. The task is made all the more difficult by the growing sophistication of steel
metallurgical design and demands a thorough knowledge of underlying metallurgical principles. This is the
subject of this module.

The structure of steels and how they obtain their strength and toughness will be presented. Various
diagrams will be used to determine the effects of heat treatment on the structure of steel. The factors
that determine the properties of we Id ments—both the weld metal and the heat-affected zone—will be
presented and finally help will be given to determine a weiding procedure that will avoid cracking problems.
Steelmaking and ways of classifying and specifying steel will not be presented in this module. These topics
are covered in Module 8 and it is recommended that It be studied in addition to this module.

1.1 Objectives
After successfully completing this module you will be able to:

• describe the ways steels derive their strength and toughness

• describe the effects of heat on steel properties

• explain how the welding procedure affects weld metal and heat-affected zone properties

• determine procedures necessary to avoid various cracking problems

2. Steel Metallurgy
Steels form the largest group of commercially important alloys. There are several reasons for this:

• the great abundance of Iron In the earth's crust

• the relative ease of extraction and low cost

• the wide range of properties that can be achieved as a result of solid state phase transformation

• other unique properties such as magnetism

Steels are alloys of iron with less than 2% carbon plus a wide range of other elements. Some of these
are added deliberately to produce special properties and others are impurities not completely removed
during the steel-making process. Elements may be present in solid solution or combined as intermetallic
compounds with iron, carbon or other elements. Some elements, namely carbon, nitrogen, boron and
hydrogen, form interstitial solutions with iron, whereas others such as manganese and silicon form
substitutional solutions. Beyond the limit of solubility these elements may also form intermetallic compounds
with iron or other elements.

2.1 Crystal Structure

Iron has the special property of existing in different crystallographic forms In the solid state. Below
910°C, the structure is body-centred cubic BCC but between 910°C and 1390°C it assumes a
face-centred cubic structure FCC. Above 1390°C and up to the melting point at 1534°C, the

© Copyright CWB Group - Industry Services Page 1

 MODULE 21 I Welding Metallurgy of Steels

structure reverts back to the body-centred cubic form. These are known as allotropic forms of Iron.
The face-centred cubic form is a close-packed structure, being more dense than the body-centred
cubic form. Consequently, iron will actually contract as it is heated above 910°C, when the structure
transformation takes place. Figure 2.1 illustrates this.

Expansion

Temperature •'C
Figure 2.1: Heating a bar of iron causes it to expand, but wfien transformation occurs from tlie BCC
Structure to tfie FCC structure at 910°C a small contraction takes place. FCC crystal structure is close^
packed and occupies less volume ttian the BCC structure.

2.2 Solubility of Carbon

The solubility of carbon in the BCC form of iron is very small, with the maximum solubility being only
about 0.02 weight percent at 723°C. As shown in Figure 2.2, there is negligible solubility of carbon in
iron at room temperature {less than 0.0001 wt%. Since steels nearly always have more carbon than
this, the excess carbon is not in solution but rather present as the intermetallic compound iron-carbide
fefi. known as cementite.

_J f I I
0,005 0,01 0.015 0-02
Weight % carbon
Figure 2.2: Solubility of carbon in a BCC iron as a function of temperature.

Page 2 © Copyright CWB Group - Industry Services

 MODULE 21 I Welding Metallurgy of Steels

In contrast, the face-centred cubic form of iron dissolves up to 2.0% carbon, well In excess of the
usual carbon content of steels. Stee! can therefore be heated to a temperature at which the structure
changes to FCC and all the carbon goes into solution. The way in which the carbon is obliged to
redistribute itself upon cooling back below the transformation is the origin of the wide range of
properties achievable in steels.

2.3 Iron-Iron Carbide Phase Diagram

Fundamental to a study of steel is an understanding of the Iron-carbon phase diagram. The diagram
is actually the metastable Iron-iron carbide system. The true stable form of carbon in iron is graphite,
but except for cast irons this only occurs after prolonged heating. Since in steels carbon is normally
present as iron carbide, it is this system that will be considered. Figure 2.3 shows the iron-iron
carbide system up to 6 wt% carbon. Several important features of this diagram will be discussed.

delta ferrite
1500

1400

1300

1200

1100

1000

o
o 900
9!

800
Q.
E
700

600

500

400

300

200

100

1.0 2,0 3.0 4,0 5,0 6,0

Weigtit % carbon

Figure 2.3: Tfie iron-iron carbide ptiase diagram.

© Copyright CWB Group - Industry Services Page 3

 MODULE 21 I Weiding Metallurgy of Steels

A eutectic Is formed at 4.3 wt% carbon. Liquid of this composition on solidification at 1147°C
transforms to a mixture of two solid phases austenite + cementite. This region is important when
discussing cast irons, but is not relevant to steels.

2.3.1 Austenite 7
The region in which iron is face-centred cubic, identified in Figure 2.4, dissolves up to 2%
carbon. This phase is termed austenite, or gamma phase. With no carbon present it begins at
giCC on heating but with 0.8 wt% carbon it starts at 723°C. When a steel is heated into the
austenite region, all carbon and most other compounds dissolve to form a single phase.

_l I I I I I I I I I
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Weight % carbon

Figure 2.4: The austenite region of the iron-iron carbide diagram. Austenite dissolves up to about 2%
carbon.

2.3.2 Ferrite a
The region shown in Figure 2.5 where carbon Is dissolved in BCC Iron is very narrow,
extending to only 0.02 wt% carbon at 723°C. This phase is termed ferrite, or alpha phase.
Although the carbon content of ferrite is very low, other elements may dissolve appreciably in it
SO ferrite cannot be considered "pure Iron".

© Copyright CWB Group - Industry Services

Page 4
 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Weight % carbon

Figure 2.5: Tlie ferrite region of tfie iron-iron carbide diagram.

2.3.3 Pearlite
At 0.8% carbon and 723°C a eutectoid is formed. This is similar to a eutectic but occurs
completely in the solid state. As Figure 2.6 illustrates, a steel of eutectoid composition will,
when cooled from the austenite region, transform at 723°C to a mixture of two solid phases,
ferrite and cementite Fe3C. This eutectoid mixture is called pearlite.

The morphology of pearlite is shown in Figure 2.7. Note that pearlite is not a single phase
itself, but rather a mixture made of two phases. Pearlite is but one of many constituents given
different names but made of the same two phases of ferrite and iron carbide. Cementite itself
is extremely hard—about 1150 HV—but when mixed with the soft ferrite layers in pearlite the
average hardness of pearlite is considerably less.

© Copyright CWB Group - Industry Services Page 5

 MODULE 21 I Welding Metallurgy of Steels

0.8% C stee! cooled

from the austenite region

O
o cementite
g
=3 FeoC
TO
Q.
E a eutectoid mixture
of ferrite and cementite

r JL JI
0,4 O.B 1.2

Weight % carbon

Figure 2.6: The eutectoid part of the iron-iron carbide piiase diagram.

Figure 2.7: A typical example of pearlite showing the two phases ferrite and cementite iron carbide in
layers. Light areas are ferrite and dark areas are cementite.

2.3.4 Peritectic
The region at the top left portion of the diagram, enlarged In Figure 2.8, Is where iron reverts
bacl< to the BCC structure. Here again the solubility for carbon is low, at only 0.1 wt% at
1493°C. This phase is known as delta ferrite. The part of the diagram at 0.16% carbon having
the appearance of an inverted eutectoid is called a peritectic. At this point a two-phase mixture
of delta ferrite and liquid transforms on cooling to a single-phase solid of austenite. This portion
of the diagram is used to describe the solidification structure of low-carbon-steel weld metal.
The genera! shape of the phase diagram in this region will not be discussed in detail but,
should be recognized, as it is used to explain various hot cracking phenomena in welding.

Page 6 © Copyright CWB Group - industry Services

 MODULE 21 I Welding Metailurgy of Steels

The part of the diagram that is of most interest to the study of steels is the region containing the
eutectoid Figure 2.6 in the bottom left corner of the iron-iron carbide diagram.

2.3.5 Peariite Growth

A Steel with 0.8 wt% carbon transforms on cooling through 723''C to the two-phase eutectoid
constituent pearlite. In peariite, the two phases, ferrite and cementite, are mixed closely
together in fine layers. The transformation begins by the formation of a cementite particle on
the austenite grain boundaries. Since the ferrite contains very little carbon while the cementite
has 6.7%, carbon atoms must diffuse to the growing cementite plates, as Figure 2.9 indicates.
The distance they can diffuse—and hence the spacing of the plates—depends on how fast the
pearlite is growing. A fast growth rate means less time for diffusion and a finer pearlite. The
morphology depends on the growth rate, which in turn depends on the cooling rate of the steel.

Direction of peariite grovi/th ^ Fast growth rate

'A*^ " '^^^^^

Ferrite plate

Austenite '"" r Austenite

Carbon atoms must With a fast growth rate

diffuse to cementite there is less time for
plates awa/ from carbon diffusion and
the ferrite atoms cannot travel
far. Piates must therefore
be closer together giving
fine pearlite

Figure 2.9: Tfie growth of pearlite is diffusion controlled.

© Copyright CWB Group - Industry Services Page 7

 MODULE 21 j Welding Metallurgy of Steels

2.3.6 Proeutectoid Ferrite

if ttie steel tias less than 0.8% carbon termed hypo-eutectoid, upon cooling from the austenite
region ferrite will be formed first from the austenite. The example in Figure 2.10 shows a steel
of 0.4% carbon and the ferrite formed is called proeutectoid ferrite. As transformation continues
and the temperature drops, the remaining austenite becomes richer In carbon. At 723''C
the steel Is comprised of ferrite and remaining austenite with 0.8% carbon. The latter then
transforms to pearlite, so the final structure of the steel is a mixture of proeutectoid ferrite and
pearlite. The amounts of proeutectoid ferrite and pearlite can be estimated by application of the
lever rule. For a 0.4% carbon steel, about 50% will be proeutectoid ferrite and 50% pearlite.

Similarly a steel of more than 0.8% carbon up to 1,8%, termed hyper-eutectoid first transforms
to proeutectoid cementite, with the remaining austenite forming pearlite Figure 2.11.

steel of 0.4% carbon cools

from austenite region

On reaching about Sie^C

some ferrite starts to form

Remaining austenite becomes

© richer in carbon

On reaching 723°C remaining

© austenite transforms to peadite

0,2 0.4 0,6 0.8 1.0 1.2 1.4

Weight % carbon

Figure 2.10: Diagram stiowing We sequence of transformation of a 0.4% carbon steel.

Steel of 1.2% cools

0 from austenite region

Remaining austenite becomes

© leaner in carbon

On reaching 723°C remaining

0.2 0.4 0.6 0.8 1.0 1.2 1.4 © austenite transforms to pearlite

Weight % carbon

Figure 2.11: Sequence of phase transformation in a steel with greater than 0.8% carbon.

Page 8 © Copyright CWB Group - Industry Services

 MODULE 21 I Weiding Metallurgy of Steels

2.3.7 Morphology of Proeutectoid Ferrite

The appearance of polished and etched sections viewed under the microscope is shown In
Figure 2.12. As indicated, the proeutectoid ferrite forms initially in the grain boundaries of the
austenite, with nucleation taking place at several points around each austenite grain. Since
each region of ferrite becomes an individual grain, its grain size will be considerably smaller
than that of the parent austenite. Ferrite continues to be formed and to grow until the final
transformation of any remaining austenite to pearlite. The ferrite does not always appear as
neat, equiaxed grains as shown In the figure, but can occur as long spikes from the grain
boundaries or even nucleate within the austenite grain, especially in welds. The form of the
proeutectoid ferrite is determined by the cooling rate.

On reheating the steel, the process reverses and the pearlite and proeutectoid ferrite grains
transform back into single-phase austenite, with completely new grains being formed. The
temperature required to get complete transformation depends on the carbon level, as seen
from the phase diagram, and ranges from 910°C for zero carbon to 723°C for 0.8% carbon.

Figure 2.12: Ptiotomicrograpfi of a steel with less than 0.8% carbon sfiowing proeutectoid ferrite at the
grain boundaries.

2.3.8 Grain Growth

Heating to higher temperatures than those necessary to get complete transformation to
austenite causes the austenite grains to grow. The grains grow to lower the surface energy
at the grain boundaries. The final size of the austenite grains depends not only on the
temperature reached, but also on the type of steel. Some steels contain small precipitates,
such as aluminum and vanadium nitride, which pin the austenite grain boundaries and retain
small grain size up to high temperatures. These are termed fine grain steels. Steels may be
deliberately developed as coarse grain or fine grain. Fine grain steels are specified when
toughness is a consideration, but coarse grain steels may be desirable when high temperature
creep resistance is important. For most structural applications, fine grain steels are specified.

© Copyright CWB Group - Industry Services Page 9

 MODULE 21 I Welding Metailurgy of Steeis

2.3.8.1 Fine Grain Steels

Ttie effect of austenitizing temperature on grain size is stiown in Figure
2.13. it stiows that, although grain growth is restricted in a fine grain steel
at a sufficiently high temperature, the precipitates dissolve and the steel
behaves as a coarse grain steel. This effect is important in the heat-affected
zone HAZ of a weld. Figure 2.14 illustrates the difference in HAZ grain size
between a coarse grain and a fine grain steel. Note that the grain size is the
same in the coarse and fine grain steels close to,the fusion boundary. The
fusion boundary is often called the weld interface.

700 800 900 1000 1100

Austenitizing temperature °C

Figure 2.13: Ttte effect of temperature on austenitic grain size in a coarse and a fine grain steel.

Course grain
steel

boundary

Fine grain
steel

boundary

Figure 2.14: Austenitic grain size in the heat-affected zone of a weld.

© Copyright CWB Group - Industry Services

Page 10
 MODULE 21 \ Weiding Metallurgy of Steels

2.3.9 HAZ Grain Size

in the heat-affected zone HAZ of coarse grain steel, there is a wide region where grain
growth has occurred, but in fine grain steel growth is restricted except in the narrow region
immediately adjacent to the weld fusion boundary weld interface where temperatures are very
high. Notice that in this narrow region the grain size is the same for both the coarse and fine
grain steels.

Steels containing titanium are exceptions, In that fine HAZ grain size may be achieved right up
to the fusion boundary weld interface. Titanium nitride is very stable and may not completely
dissolve in the HAZ, even at the temperature immediately adjacent to the fusion boundary.
Modern steels, designed specifically to retain small grain size in the HAZ of high heat input
welds sometimes contain titanium for this purpose.

2.3.10 Non-equilibrium Cooling

The phase and microstructures predicted by the iron-iron carbide diagram in steels occur only
if the cooling is very slow. Moreover, the diagram assumes that carbon is the only alloying
element present in the steel. If the concentration of other alloy elements is small, the basic
diagram can stili be used, except that it will be distorted and the lines may move to slightly
different locations. For example, the presence of alloy elements changes the eutectoid
temperature, as shown in Figure 2.15. In structural steels, the concentration of alloys is quite
small and the basic iron-iron carbide diagram is not distorted very much from equilibrium
conditions,

1300

500
0 2 4 6 8 10 12 14 16 18

Weighit % of alloying element

Figure 2.15: Diagram sfiowing tlie effect of alloying elements on the eutectoid temperature in steel.

© Copyright CWB Group - Industry Services Page 11

 MODULE 21 \ Welding Metailurgy of Steels

2.3.11 Effect of Rapid Cooling

The rate of cooling has a major effect on the types of microstructures formed. Unless the stee!
cools slowly, the iron-iron carbide diagram cannot be used because the transformation of
austenite to pearlite requires the diffusion of carbon to the sites of growing carbide, a process
that takes time. A faster cooling rate produces finer pearlite. With even faster cooling rates,
less time is available for diffusion and pearlite cannot form at all. Alternate microstructures
form, with their exact morphology dependent on just how quickly the steel cools. In a water
quench, for example, the cooiing rate is so rapid there is no time for any diffusion, and the
carbon remains trapped in the same place as it was in the austenite.

A rapid quench cannot suppress the crystal structure change from FCC to BCC, but the
presence of trapped carbon in the BCC phase distorts it to a tetragona! shape BCT, as
shown in Figure 2.16, rather than a true cubic structure. This structure is actually formed by a
shearing action, thus dislocation density is very high.

300 r

Iron
atoms CD 290
"53
E
ro
Carbon ro
atoms Q.
M
CD 280

o
i=i
ro

270
0.2 0.4 0,6 0.8 1.0 1.2

Weight % carbon

Figure 2.16: Carbon trapped in the martensite lattice causes a distortion to a tetragonal structure. The
shape depends on the carbon level.

2.3.12 Martensite
This constituent, formed by rapid cooling, is known as martensite. Under the microscope,
martensite has the appearance Figure 2.17 of a mass of needles, Martensite can be a very
hard and brittle constituent when it contains appreciable carbon. The hardness depends almost
exclusively on the carbon content, with other elements having little effect. Figure 2.18 shows
the relationship between martensite hardness and the carbon content.

Page 12 © Copyright CWB Group - Induslry Services

 IHustrated in Figure 2.19 are a number of welds In a carbon steel that have cooled at various
rates. In each weld the heat-affected zone transforms to a microstructure dependent on the
cooling rate of that weld. For the small, rapidly-cooled welds, martensite is formed. For the
large, slowly-cooled welds, the HAZ structure is pearlite. The hardness of the HAZ is much
higher in those welds In which martensite is present.

© Copyright CWB Group - Industry Services Page 13

 MODULE 21 1 Weiding Metallurgy of Steels

HAZ hardness Hv 475 450 300 275

Steel composition
C = 0.36%
Mn = 0.77%
Si = 0.24%

Figure 2.19: Examples of four welds of different heat input. The small welds cool rapidly enough to form
martensite in the HAZ with a hardness exceeding 450 Hv.

2.3.13 Bainite
Intermediate between a rapid quencti that produces martensite and a slow cool producing
pearlite, other constituents may form, particularly in alloy steels. The most important of these is
bainite. Bainite is still a two-phase mixture of ferrite and iron carbide, but unlike the cementite
plates in pearlite, the carbide in bainite forms as stringers {see Figure 2.20. Bainite formed
above SOO'C contains relatively coarse particles of the Fe^C form of iron carbide cementite
and is termed upper bainite. When formed below 300'C, bainite has a much finer structure
due to less diffusion, with the carbides tending to form striations across the ferrite laths. This is
called lower bainite. The carbides in lower bainite are Fe^^C e carbide.

Figure 2.20: Photomicrograph of bainite. The growth rate is too rapid for carbides to form as plates of
cementite as in pearlite.

2.4 Transformation Diagrams

Since the iron-iron carbide diagram is only valid for very slow cooling rates, alternate diagrams for
determining the constituents present in a more rapidly cooled steel have been developed. There
are two types, covering the case of continuous cooling CT diagrams and the case of isothermal
transformation, where the steel sample Is held at a constant temperature until transformation is
complete IT diagrams.

© Copyright CWB Group - Industry Services

Page 14
 MODULE 21 I Welding Metallurgy of Steels

2.4.1 Isothermal Transformation IT Diagrams

Imagine a sample of steel heated until fully austenitic, then quenched to some temperature
below the equilibrium transformation temperature see Figure 2.21. If the steel Is held at this
temperature, there is a delay before transformation begins and a further elapse of time while
transformation takes place. The delay depends on the temperature at which the steel is held.
This information can be plotted on a diagram of temperature against time for a given steel
composition.

time

% carbon

Figure 2.21: Procedure for determining time-temperature-transformation IT diagram behaviour. Steel is

heated to the austenitic region, quenched to a specific temperature, then held at that temperature until
transformation is complete.

An example of an isothermal transformation IT diagram for carbon steel is shown in Figure

2.22. Note that at high temperatures see Figure 2.23 the steel transforms to proeutectoid
ferrite, followed by pearlite. At lower temperatures, less proeutectoid ferrite is formed and
the pearlite is finer. At about 550°C the pearlite forms in the shortest time and there is no
proeutectoid ferrite see Figure 2.24. Below this range Figure 2.25 transformation to
bainite occurs, taking a longer time at lower temperatures. At a sufficiently low temperature
Figure 2.26 martensite begins to form. Note that it forms almost instantaneously and does
not grow as a function of time. For each steel, there is a fixed temperature M J at which
martensite starts to form and a fixed temperature M, at which transformation is complete. The
percentage of martensite formed therefore depends only on the temperature to which the steel
is cooled and not on how long it Is held there.

© Copyright CWB Group - Industry Services Page 15

 MODULE 21 I Welding Metallurgy of Steeis

800

_ Peadite start
700 •—• " Peadite finish

o 600
o
S!
500 BainiteN
a.
E Ms Martensite start start^/

400

Mf Martensite finish
300

11111
1 10 10^ 10^ 10-* 10^

Time seconds

Figure 2.22: Typical IT, or isothermal transformation, diagram.

? Specimen quickly cooled from austentite

region then held at about 650°C

2 Proeutectoid ferrite begins to form

3 Pearlite starts to form

4 Transformation to peadite complete

Time

Figure 2.23: Steel undergoing isothermal transformation at about 650°C.

T Specimen quickly cooled from austentite

region then held at about 550°C

2 Pearlite starts to form

3 Transformation to pearlite complete

o.
£

Time
Figure 2.24: Steel transformed at about SSO^C. No proeutectoid ferrite is produced, only pearlite.

© Copyright CWB Group - Industry Services

Page 16
 MODULE 21 I Welding Metallurgy of Steeis

Time

Figure 2.25: Transformation at about 450°C to bainite.

T Specimen quickly cooled from austentite

region to beiow Mg. Martensite forms
immediately.

2 If held at constant temperature the

proportion of martensite does not change.

3 Specimen cooled to room temperature.

Complete transformation to martensite
occurs on crossing Mf temperature.

Time

Figure 2.26: Specimen quenched to a temperature below the martensite start. Martensite forms
essentially instantaneously, but transformation is not complete until the specimen is cooled below the
martensite finish temperature.

2.4.2 Continuous Cooling Transformation CT Diagrams

Consider the case of continuous cooling. A cooling curve can be superimposed on the IT
diagram to get an idea of what microstructures form, but it is more accurate to use a diagram
established under continuous cooling conditions. The continuous cooling transformation CT
diagram is slightly different from the IT diagram. Continuous cooling transformation diagrams
can be used with cooling curves representing weld cooling rates to determine microstructures
produced in welds and heat-affected zones. The example shown for the steel in Figure
2.27 reveals that there is a critical cooling rate at which the nose of the curve is missed and
martensite forms. This is an important concept since the cooling rate at which martensite
can form strongly influences the risk of cracking during welding and gives an indication of the
"weldability".

© Copyright CWB Group - Industry Services Page 17

 MODULE 21 I Welding Metallurgy of Steels

Figure 2,27: Transformation diagram with cooling curves superimposed.

2.4.2.1 Critical Cooling Rates

Because of the shape of the CT diagram for plain carbon steels, bainite
cannot form during continuous cooling. The bainite region is tucked under
the peariite area, so a cooling curve either hits the pearlite curve or misses
it completely. At cooling rates fast enough to miss the nose of the curve,
martensite is formed. At slov/er rates, the cooling curve intercepts the ferrite
start and a ferrite/pearlite structure results. This is also apparent in the
example shown in Figure 2.19, where the HAZ is martensitic for small welds
and pearlitic for large, slowly cooled welds.

2.4.2.2 Determining CT Diagrams

The exact shape of a CT curve depends on the details of the heating and
cooling cycles. CT diagrams can be established specifically for the conditions
pertaining to welds, but are different from the conventional curves in that
they must be determined with high austenitizing temperatures to simulate the
conditions in the HAZ, This can be done using a dilatometer. In which the
start of transformation is detected by sensing the changes in length of a small
sample as it cools. An alternate method for establishing a CT diagram for
the HAZ of a low-alloy steel for six standard cooling rates is shown in Figure
2.28.

Page 18 © Copyright CWB Group - Industry Services

 MODULE 21 I Weiding Metallurgy of Steels

Time to cool from 1000°C sec

Figure 2.28: Typical CT diagram for a C-Mn-Nb steel showing six cooling rates and the corresponding
hardnesses. Welds cooling at these rates would be expected to show these hardness levels In the HAZ.

2.4.3 Effectof Alloy Elements

Alloy elements have a significant effect on the shape of the CT and IT diagrams, since the
alloys aWow different microstructures to be produced in alloy steels. Molybdenum and boron, for
example, shift the top pearlite part of the curve to the right, to longer times, thus exposing the
bainite region. Steels containing these elements may produce bainite on continuous cooling.
The entire curve may also move to greater times, allowing martensite to form at much slower
cooling rates. This increases the hardenability ease with which a hardened structure can be
produced from austenite of the steel.

The other notable effect of ailoy elements is to change the martensite start and finish
temperatures. Increasing carbon, for example, depresses the martensite start to lower
temperatures as shown in Figure 2.29. Other elements affect martensite formation. The
combined effect can be approximated by the following equation:

M/C = 550 - 350x%C - 40x%Mn - 35x%V - 20x%Cr - 17x%NI - 10x%Cu ~ 10x%Mo -

5x%W + 15x%Co + 30x%AI

If the M, is below room temperature, martensite transformation Is not completed and the steel
contains some retained austenite.

© Copyright CWB Group - industry Services Page 19

 MODULE 21 ! Welding Metallurgy of Steels

2.5 Heat Treatment

If a bar of steel is heated Into the austenite region then quenched in water or brine salt water the
cooling rate at the surface of the bar will be much higher than at the centre of the bar see Figure
2.30. Thus, the microstructure formed In the surface region could be quite different from that formed
In the interior.

Maximum cooling Minimum cooling rate

at surface of bar at centre of bar

Figure 2.30: Wtien a tiot bar is quenctied the cooling rate is a maximum at the surface but is slower in
the centre of the bar.

© Copyrigtit CWB Group - Industry Services

Page 20
 MODULE 21 I Weldmg Metallurgy of Steels

2.5.1 Hardenability
In a large bar, the centre could cool so slowly that no martensite Is formed and the steel
would not be hardened at all. The distance below the surface that is hardened is of great
practical importance. It determines the diameter of bar that can be hardened right through. The
largest diameter of bar that will just produce 50% martensite in Its centre under ideal, perfect
quenching conditions is known as the ideal critical diameter D., as shown in Figure 2.31. it is
a measure of the steel's hardenability. If alloy elements are added that allow a larger diameter
bar to produce 50% martensite at its centre, then the hardenability has been increased.

50%
martensite

Figure 2.31: Tfie ideal critical diameter is tfiat which will just produce 50% martensite at the centre of the
bar under an ideal quench.

2.5.2 Measuring Hardenability

The hardenability of a steel can be determined by performing a Jominy end-quench test, as
illustrated in Figure 2.32. The test specimen is a cylinder that is one inch in diameter and four
inches long, which is heated to the hardening temperature and then placed in a fixture, where it
is quenched by water impinging on one end. The faster quench is therefore on the sprayed end
of the specimen, with progressively slower cooler rates being experienced away from that end.
The specimen is examined by grinding two flats along the length of the cylinder on opposite
sides and making hardness measurements at intervals along the length from the quenched
end. The distance from the end at which 50% martensite is produced can be correlated directly
with the hardenability D..

© Copyright CWB Group - Industry Services Page 21

 MODULE 21 I Welding Metailurgy of Steeis

round bar
4 in long by
1 in diameter

Figure 2.32: Arrangement for carrying out tfie Jominy end-quencfi test for measuring liardenabHity.

2.5.3 Calculating {Hardenability

Many researchers have attempted to establish quantitative relationships between the
composition of the steel and the hardenability. The classic work of Grossman led to the
concept of multiplying factors for each element. For a given carbon level and austenite grain
size, a base diameter D, carbon is determined, These are listed in Table 2.1. This is multiplied
by a factor for each of the alloying elements depending on their concentration. Factors are
shown graphically in Figure 2.33. For greater accuracy refer to the values published in
tabular form. The example that follows illustrates how the calculation works. This method
has been superseded by more sophisticated computer-based models, but Is still useful for
approximations. Notice that since the hardenability depends on the product of various factors,
high hardenability can be obtained by using small quantities of several alloy elements rather
than a large quantity of just one element.

Tabie 2.1: Base diameters in incfies for carbon steels of various austenitic grain sizes.

Carbon % Base diameter D,^,,^„

ASTM Grain Size

#6 #7 #8

0,05 0.0814 0.0750 0.0697

0.10 0.1153 0.1065 0.0995
0.15 0.1413 0,1315 0.1212
0.20 0.1623 0,1509 0.1400
0,25 0.1820 0.1678 0.1560
0.30 0.1991 0.1849 0.1700
0.35 0.2154 0.2000 0.1842
0.40 0.2300 0.2130 0.1976
0.45 0,2440 0.2259 0.2090
0.50 0.2580 0.2380 0.2200
0.55 0.273 0.251 0.231
0.60 0.284 0.262 0.241
0.65 0.295 0.273 0.251
0.70 0.306 0.283 0.260
0.75 0.316 0.293 0.270
0.80 0.326 0.303 0.278
0.85 0.336 0.312 0.287
0.90 0.346 0.321 0.296

Page 22 © Copyright CWB Group - Industry Services

 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Weight % of element

Figure 2.33: Values of multiplying factors for various alloy elements as a function of tfie alloy content.

2.5.4 Example of a Hardenability Calculation

In this example the hardenability of AISI 4340 steel is calculated. Hardenability is measured by
the ideal critical diameter. This is the diameter of a round bar in which 50% martensite will be
produced at the centre when given an ideal quench.

Step 1: Determine the base diameter from the carbon content using Table 2.1. Assume the
grain size of the austenite before quenching is ASTM 8.

Assume average composition AISI 4340 wt%:

0.40 C, 0.25 Si, 0.70 Mn, 0.8 Cr, 0.25 Mo, 1.8 Ni
Base diameter = 0.1976 in

Step 2: Determine the multiplying factors for each of the alloying elements. Use Figure 2.33 or
published tables for greater accuracy.

Alloy Element wt% Multiplying Factor

Silicon 0.25 1.2

Manganese 0.70 3,3

Chromium 0.80 2.7

Molybdenum 0.25 1,75

Nicl<el 1.8 1,7

Step 3: Calculate the ideal critical diameter

D. = D. carbon x multiplying factors

D, = 0.1976 X 1.2 X 3.3 x 2.7 x 1.75 x 1.7 = 6.29 in

Note: A quench in practice is never ideal. Real quenching media always cool the specimen
slower than the ideal. The actual critical diameter is therefore less than the ideal. In this
example, a water quench without any agitation will reduce the ideal critical diameter to a value
of about 5.5 in.

© Copyright CWB Group - Industry Services Page 23

 MODULE 21 I Weiding Metallurgy of Steels

2.5.5 Hardenability in the HAZ

Similar attempts tiave been made to develop models for hardenability in welding. Depth of
hardening and critical diameters are not relevant concepts In a welding situation, but of interest
is the hardness produced at a given cooling rate or the critical cooling rate to produce a given
hardness in the heat-affected zone. The simplest model is one in which the effects of individual
alloying elements are added together a linear model in a carbon equivalent, which in turn
relates to the critical cooling rate to produce a given hardness. Figure 2.34 shows the carbon
equivalent plotted against the critical cooling rate at 540°C to give a hardness of 350 HV, and a
reasonable correlation exists.

As might be expected from a consideration of the multiplying effect apparent in the Grossman
approach, the linear carbon equivalent method tends to overestimate hardenability for
plain carbon steeis and underestimate it for multi-alloyed steels. It is useful, nevertheless,
in estimating welding conditions for the normal composition ranges of common weldable
structural and pressure vessel steels. Carbon equivalents are used widely in industry as
measures of weldability and it is important to understand their origin. Several different formulae
have been proposed and some are incorporated Into national standards and codes.

0.80

in
0.70 h

0.60 h

0.50
CD

+ 0.40
O
II
UJ Cooling rate at 540°C
o JIL I III
0.30
100 50 30 20 10 32

Critical cooling rate °C/s for HAZ hardness of 350 Hv

Figure 2.34: Relationship between the carbon equivalent and the critical cooling rate In a weld to
produce a heat-affected zone hardness of 350 Hv.

2.5.5.1 HAZ Hardness

Several models have been proposed for predicting the hardness of the heat-
affected zone. One example is presented below for structural steels:

Hv = 2019[1 - 0.51ogt3jC + O.SICE^ - C] + 661-0.8logtg,5

Hv is the Vickers diamond pyramid hardness of the HAZ.

© Copyright CWB Group - Industry Services

Page 24
MODULE 21 I Welding Metallurgy of Steels

In this equation the term CEg represents the steel composition as a linear
carbon equivalent and represented by:
CEg{% = C + Si/11 + Mn/8 + Cr/5 + Mo/6 + V/3 + Ni/17 + Cu/9
and t is the time, in seconds, for the HAZ to cool from 800*'C to 500°C.

2.5.6 Tempering
Martensite produced in a quenched steel is hard and brittle and in most cases the steel is
unusable in that form. The toughness may be improved by a process called tempering. This
involves reheating the steel to a temperature below that necessary for transformation, holding
it at that temperature for a period of time, then cooling to room temperature. During tempering
the carbon trapped as an interstitial in the martensite tetragonal structure is released.
Carbon atoms diffuse and precipitate as small carbides. With enough time and at sufficiently
high temperatures, cementite will form, not as plates as in pearlite, but as round particles.
Improvement In toughness is accompanied by a loss of hardness, which is a function of both
temperature and time. Over limited ranges these two factors may be combined into a single
parameter, as proposed by Holloman and Jaffe:

H = r 20 + log 0

where T is the temperature in kelvin K = °C + 273 and t is the holding time in hours. Figure
2.35 shows the hardness of a carbon stee! that has been quenched and tempered at various
temperatures and times. The hardness loss is far more sensitive to temperature than to
time and an Increase of 50°C at eOO^C is equivalent to increasing the time ten times. The
temperatures typically selected for thermally stress relieving weided steel structures are high
enough to cause rapid tempering. Many of the benefits derived from a thermal stress relief are
actually due to tempering

I I I1IIIIi I
9 10 11 12 13 14 15 16 17

H = T { 20 + log t X 10-3 T = K, t = hrs

Figure 2.35: The effect of tempering time and temperature on the hardness of a quenched steel.
Temperature and time are combined in the Holloman-Jaffe parameter, H.

© Copyright CWB Group - Industry Services Page 25

 MODULE 21 I Welding Metallurgy of Steels

2.5.6.1 Secondary Hardening

In some steels containing specific alloy elements, tempering may actually
cause an increase in hardness as the tempering temperature is raised, as
Figure 2.36 shows. Strong carbide-form ing elements such as molybdenum
Mo, chromium Cr, and tungsten W, cause alloy carbides to be
precipitated in certain temperature ranges. This behaviour of secondary
hardening is put to good use in the design of tool steels high speed steels
that need to retain their hardness at the high temperatures experienced in
machining.

60 r
5 % molybdenum
t/i
to 50 h
o
c
•a
k. 40
0
JC
b 0 % molybdenum
30
0
I 20
u 0.35 % carbon
o
DC 0.30 % chromium
10 II X ± ± J
200 300 400 500 600 700

Tempering temperature °C

Figure 2.36: Hardness as a function of tempering temperature for a piain carbon and a molybdenum
alloyed steel. The molybdenum alloyed steel shows secondary hardening.

2.5.7 Welding Quenched and Tempered Q&T Steels

Welding of quenched and tempered steels may pose several problems. A careful study of the
steel and the intended application is necessary before specifying the welding procedure, if the
welding is carried out after the heat treatment it may be difficult to achieve uniform properties
across the joint. In some cases, therefore, it is preferable to perform the quenching and
tempering after completing the weld. This assumes that the welded component is small enough
to allow this. The weld metal In this case must be selected to have a matching composition so
that uniform properties are achieved after quenching and tempering.

2.5.8 Effects of Preheating

A second method of achieving uniform properties without heat treating after welding is to
apply sufficient preheat to the component to hold it above the martensite start temperature
until welding is complete. All deposited metal and the HAZ remain austenitic during welding

© Copyright CWB Group - Industry Services

Page 26
 MODULE 21 I Weldmg Metallurgy of Steels

and transform together on cooling, producing a uniform structure. This avoids any tempering
of adjacent we\6 beads as each bead is deposited. In applying this approach, the IT diagram
for the stee! can be studied to determine the preheat temperature and the maximum allowable
time for the completion of welding. The holding time after weld completion to ensure complete
transformation of the austenite should be adjusted because of the effect of grain size on
diffusion-controlled transformations. The adjustment is usually to double the holding time
indicated on the IT diagram. This method Is applied for buildup and surfacing of alloy steel
shafts, but is not used in structural steels. The example shown illustrates how the IT diagram
can be used to develop a procedure for this type of application {see Figure 2.37.

Preheat for Roll Build-up

A
Y
A roll is to be built up by weiding. The roil is
turned and the weld metal deposited by SAW in
a continuous spiral bead. A uniform hardness is
required in the weld metal and HAZ to avoid
uneven wear in service.
\1 WJM

C Mn Or Sr Ni Mo
Roll 0.5 1.0 1.5 0.5 1.0 0.5
Deposit 0.15 1.9 12.0 1.0 -

To avoid each bead tempering the one

before the weld metal and HAZ must be
held above the Mg so that transformation
is delayed long enough for several beads
to be deposited. Temp

Deposit must held at a temperature in

the "bay" of the IT diagram. This gives Preheat
the longest delay before transformation.
Preheat is above the Mg.

From the formula given in 2.4.3, the preheat

can be estimated.
Mg. for the roH and deposit metal.
Mg. roll = 283°C
Mg. deposit ~ ISSX

Preheat specified 300°C

Figure 2.37: An example of tiow an IT diagram can be used to develop a preheat procedure.

© Copyright CWB Group - Industry Services Page 27

 MODULE 21 [ Weiding Metallurgy of Steels

If the weld is not to be heat treated, or the isothermal method cannot be used, then the
problem will be welding the steel without cracking while obtaining adequate weld properties.
The high hardenability of Q & T steel means the HAZ tends to transform to martensite on
cooling. Cracking must be prevented by applying adequate preheat. A high preheat may cause
auto-tempering as the martensite forms but stays hot. A weld metal must be selected that will
provide adequate strength without the benefit of a subsequent quench and temper

2.5.9 Stress Relief of Q & T Steels

It is common practice to apply a post-weld temper stress relief heat treatment, which softens
the HAZ, removes hydrogen and lowers residual stress. The temperature for this treatment
must not be higher than the original tempering temperature of the steel or the steel strength
could drop below specification. To reduce the risk of cracking, the post-weld heat treatment
may be carried out immediately after welding without letting the weld cool down.

3. Properties of Steels

3.1 Effect of Carbon

Plain carbon steels represent the simplest family of steels in which various strength levels are
achieved through the appropriate carbon content. In a steel slowly cooled from the austenite phase,
the structure is pearlitic and the strength is proportional to the amount of peariite formed. The general
relation between carbon and strength is shown in Figure 3.1. increasing the percentage of pearlite
has a stronger effect on the ultimate tensile strength UTS than on the yield strength. Carbon steels,
therefore, tend to have a lower yield to ultimate strength ratio than low-alloy steels. This is one
reason carbon steels are popular for pressure vessels where the UTS usuaiiy determines design
requirements. In structural applications e.g., bridges where the yield stress controls design, low-alloy
steels with higher yield to ultimate ratios are favoured.

L_l I I I I I I
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Weight % carbon

Figure 3.1: The effect of carbon content on yield and ultimate tensile strength in a plain carbon steel.

© Copyright CWB Group - Industry Services

Page 28
 MODULE 21 I Welding Metallurgy of Steels

At 0.8% the steel carbon is 100% pearlite and the maximum strength is obtained. Any increases
of carbon above 0.8% serve only to produce pro-eutectoid cementite, with no further increase in
strength. Plain carbon steels of about 0.8% carbon are often specified where maximum strength is
desired, for example in rails and piano wire.

3.1.1 Effect on Ductility

Increasing carbon content, however, reduces the ductility as shown in Figure 3.2 and this
often limits the applications for plain carbon steels. The impact toughness is also lowered as
shown by the effect of carbon content on the Charpy transition curve is illustrated in Figure
3.3.

1000 I-

800
£
to
.E
c 600
.2
O m
3
•D 400
i
200
^^^^

± A. J. A J.
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Weight % carbon

Figure 3.2: Tiie effect of carbon content on ductility in plain carbon steels.

200 r 0.11% C

0.20% C

1%

0.80% C

-100 -50 0 50 100 150 200

Temperature °C

Figure 3.3: Effect of carbon content on the Charpy V-notch impact energy transition curves of plain
carbon steels.

© Copyright CWB Group - Industry Sen/ices Page 29

 MODULE 21 I Welding Metallurgy of Steels

3.2 Effectof Thickness

Under continuous cooling conditions, the coarseness of pearlite depends on cooling rate. A thick plate
cooling slowly produces a coarse structure with lower strength. This is reflected in the specifications
for carbon steels that may have lower specified strength requirements for thick sections, or may allow
higher carbon levels to compensate. An example for ASTM A516 is shown in Table 3.1. This has
important implications for weiding, since not only does a weld on a thick plate cool faster than one
on a thin plate, but the thick plate Is likely to have a higher carbon content. These two factors greatly
increase the risk of cracking in thick plates.

Table 3.1: Maximum specified carbon levels for various thicknesses of ASTM A516 Grade 70 steel.

Thickness range in inches mm Carbon content % max

1 in 25 mm and under 0.31
Over 1 to 2 in 25 to 50 mm, ind. 0.33
Over 2 lo 4 in 50 to 100 mm, incl. 0.35
Over 4 to 8 in 100 to 200 mm, incl. 0.35
Over 8 in 200 mm 0.35

The coarsest pearlite and ferrite structure is achieved in a carbon steel after a full anneal—that is,
heating until fully austenitized at 900°C to 950°C, followed by a very slow cool in a furnace anneal.
A full anneal is used when maximum softening is required, but the coarse microstructure has very low
toughness. Although the ductility in a tensile test is improved, the impact toughness is poor. The fully
annealed condition is rarely desirable in a steel for structural use. A full anneal might be specified prior
to machining, particularly when this is followed by a subsequent heat treatment.

3.3 Normalizing
The toughness of carbon structural and pressure vessel steels can be improved by normalizing to
reduce the grain size see Figure 3.4. This treatment involves austenitizing at about 900°C, followed
by air cooling. The temperature Is the lowest practical temperature at which the steel becomes fully
austenitized, and the holding time need only be sufficient to ensure complete austenitizatlon has
occurred. Prolonged heating may, in fact, cause grain growth, which is undesirable. A holding time
of half an hour at temperature for carbon steels up to 75 mm is typicai, although various codes and
standards may specify differently.

IIIII
-40 -20 0 20
Nil DuctilityTemperature''CNDT

Figure 3.4: The toughness of carbon steels is improved by normalizing. This diagram shows a typical
distribution of NDT temperatures for as-rolled and normalized steels.

© Copyright CWB Group - Industry Services

Page 30
 MODULE 21 I Welding Metallurgy of Steels

Some of the older steel specifications allow a hot forming operation carried out by the fabricator to
count as a normalize. If the fabncator chooses this option there are two important things to note: First,
the steel may be supplied in the as-rolled un-normalized state, but the mill test certificates will report
tests on coupons normalized in the laboratory. Second, the temperature for hot forming must not
significantly exceed the normalizing temperature, otherwise grain growth and toughness deterioration
could occur. In this regard, tine grain steels are more tolerant to higher temperatures than those with
coarse grain.

3.3.1 Cooling After Normalizing

To prevent coarse pearlite from forming, normalizing is followed by air cooling rather than
furnace cooling. An inadequate cooling rate could result In toughness deterioration. If
several components are normalized together they must be properly stacked to allow good air
circulation between them. With thick sections, forced air cooling using fans may be specified,
but care must be taken to avoid uneven cooling that results in distortion.

3.4 Alternate Methods of Strengthening

The negative effects of carbon on toughness and weldability have led steel makers to employ other
strengthening mechanisms. The most recent and important for structural steels are precipitation
hardening and grain refinement.

The first generation of high-strength low-alloy structural steels used the carbo-nitride forming
elements vanadium and niobium columbium as precipitation hardening agents. These elements had
the advantage that they could be added to semi-killed steels because of their limited tendency to form
oxides. The increased strength due to precipitation hardening allows a reduction in carbon content,
improving weldability and toughness. Typical examples of these steels are ASTM A441 and A572.

3.4.1 Controlled Rolling

More recent steeis have employed grain refinements in addition to precipitation hardening
to produce high strength coupled with excellent toughness. Fine grain steels are produced
through thermo-mechanical processing, in which the reduction of plate thickness by rolling
Is integrated with temperature. During hot rolling the austenite grains breakup, but at high
temperatures they re-crystallize and grow again see Figure 3.5. If rolling is performed at
lower temperatures controlled rolling there is less of a tendency for the austenite grains to
re-grow. The re-crystallization of the austenite grains is further retarded by the presence of
micro-alloy elements—particularly niobium—and the austenite range is extended to lower
temperatures by alloys such as manganese.

© Copyriglit CWB Group - industry Services Page 31

 MODULE 21 I Welding Metaliurgy of Steels

during liot rolling

Original austenite grains If rolling Is finished at high temperature
grains recrystallize and grow again

If rolling is finished at lower temperature

tendency to regrow is reduced

If rolling is flnislied at low temperature

and recrystaflization Is retarded by
niobium small grain size will be retained

Figure 3.5: Using lower finistiing temperatures during hot rolling and micro-alloy elements to retard re-
crystallization can result in steels of very fine grain size.

A steelmaker may use controlled rolling for the lighter sections of plate and normalizing
for thick sections when excellent toughness is required. Several other devices are being
developed to further Improve properties, among them being on-line accelerated cooling and
direct quenching.

3.4.2 Effectof Work Hardening

Since heating a steel above 400°C causes recovery and re-crystallization to begin, any
process Involving temperatures above this removes the effects of work hardening. Thus,
castings, hot-forged, hot-rolled, or any heat-treated products will not gain any strength from
work hardening. However, any cold finished product or one subject to subsequent cold working
such as bending or forming will have a strength component that results from work hardening.
Figure 3.6 shows the increase in strength due to cold deformation, while Figure 3.7 reveals
the accompanying reduction in ductility.

Work hardening as a strengthening method is particularly valuable in the production of wires.

The wire drawing process itself provides a greater degree of reduction than is achieved with
any other method. High strength achieved in low carbon steel wires is particularly valuable in
making gas metal arc welding electrodes. GMAW wires are usually pushed by a wire feeder
along a conduit to the welding gun and a high column strength is desirable to prevent kinking.

© Copyright CWB Group - industry Services

Page 32
 MODULE 21 I Welding Metailurgy of Steels

0 20 40 60 80 100
% reduction in area in rolling

Figure 3.6: Tfie effect of cold rolling on tensile strengtfi of steel.

30 -

25 I

20 I
c

0 20 40 60 80 100

% reduction in area in rolling

Figure 3.7: Tfie effect of cold rolling on ductility as measured by elongation in a tensile test.

3.4.3 Restoring Ductility

Notice in Figure 3.7 that the loss of ductility is substantial for the first ten percent of the
deformation. In bending steel plates for structural applications it may be necessary to give the
plate a thermal stress relief to restore ductility when deformation levels are substantial. ASME
Section VIM Division 1 requires a stress relief heat treatment for most vessel shell sections,
heads and other pressure boundary parts of carbon and low-alloy steels if the extreme fibre
elongation exceeds 5%.

© Copyright CWB Group - industry Services Page 33

 MODULE 21 I Weiding Metallurgy of Steels

When a coid-worked material is welded, the heat from welding may be sufficient to cause
a marked loss of strength in the HAZ where re-crystallization takes place. This may be
important, for instance, when resistance welding sheet that has a hard temper The loss of
strength or the width of the softened zone can be reduced by modifying the welding procedure
to reduce the time.

4. The Heat-Affected Zone

4.1 Structure
Adjacent to the weld, the meta! undergoes vanous changes depending on the peak temperature
and cooling rate experienced at each location. Close to the fusion boundary, the peak temperature
will be high enough to cause complete transformation to austenite and some grain growth. At some
distance from the weld interface {fusion boundary, the peak temperature is insufficient to cause any
microstructural changes, although other effects, such as strain aging may occur. In between, a range
of mixed structures may be observed, Figure 4.1 illustrates the various zones adjacent to the weld
metal.

Figure 4.1: Diagram stiowing ttte various parts of the heat-affected zone HAZ and their relation to the
iron-iron carbide phase diagram.

4.2 HAZ Grain Size

The austenite grain size in the HAZ is controlled mainly by the weld heat input, but is also influenced
by the shape of the fusion boundary. The effect of heat input on austenite grain size In the HAZ is
shown in Figure 4.2. In very high heat input processes, such as electroslag, the HAZ is very wide and
the grains can grow to their equivalent isothermal size, that is, the same size grains as in a specimen
uniformly heated in a furnace at the same temperature. The heat input affects the time at peak

Page 34 © Copyrigtit CWB Group - Industry Services

 MODULE 21 ! Welding Metallurgy of Steels

temperature and also ttie temperature gradient in the HAZ. A steep gradient from a low-heat-lnput
weld restricts the size to which grains can grow. This effect Is seen in welds that have non-circular
fusion boundary shapes. The concave region of the submerged arc weld shown in Figure 4.3 has a
large austenite grain size compared with the convex region despite the same energy Input. The region
in the HAZ is usually in the concave region because the larger grain size increases hardenability.

0.5 1.0 5 10 50 100

Emergy input kJ/mm

Figure 4.2: Austenite grain size close to ttte fusion boundary as a function of weld heat Input.

Figure 4.3: Section of submerged arc weld showing how the shape of the fusion boundary affects the
grain size in the HAZ.

4.2.1 Width of Coarse Grain HAZ

The austenite grain size is affected by the composition of the steel as well as the time at
temperature. A coarse grain steel has a wide grain-coarsened austenite region in the HAZ
and a fine grain steel has a narrow one. Except for the recent titanium steels, the grain size
immediately adjacent to the fusion boundary will be the same for all types of steel. In titanium
steels the presence of very stable titanium nitrides prevents grain coarsening, with fine grains
being retained right up to the fusion boundary.

© Copyright CWB Group - Industry Services Page 35

 MODULE 21 I Welding Metallurgy of Steels

4.2.2 Fine Grain Region

The fine grain region is outside the coarse grain region and exhibits good mechanical
properties, This region has fine grains because of the double-phase transformation that
occurs: transformation to low temperature austenite during heating, then transformation to
ferrite and carbides during cooling.

4.2.3 Partial Transformation

Adjacent to the fully austenitic zone is a region where partial transformation to austenite can
occur Sometimes, particularly in small welds in higher carbon steels that cool rapidly, small
patches of high carbon white etching martensite can be seen in this zone, These often have
a hardness higher than would be expected from the average composition of the steel. They
occur in the region that just exceeds 723°C, where partial transformation to austenite higher in
carbon occurs and rapid cooling results in the formation of martensite. The martensite regions
are localized and it Is not certain whether the microstructure as a whole becomes any more
susceptible to cracking.

4.3 Hardening Curves

On cooling, the austenite region of the HAZ transforms to a structure dependent on the cooling
rate and the hardenability of the steel. Continuous cooling transformation diagrams are helpful in
determining the structures and hardness that occur. In cases where only the hardness is of interest,
the standard hardening curve determined directly from weld samples at various heat inputs can be
used. The hardening curve Is helpful in establishing welding conditions to prevent HAZs forming with
hardnesses higher than a given value. Two examples of hardening curves are shown in Figure 4.4.
One, for a carbon steel, has a steep transition from the high hardness associated with martensite to a
low hardness where the microstructure is pearlite. The other example is for a quenched and tempered
alloy steel. In this example the hardenability is high enough for a hard HAZ to be formed at any
cooling rate experienced in normal welding operations.

100 50 10 5 1
Cooling rate at 540°C "C/s

Figure 4.4: Plot of HAZ mean maximum hardness versus cooling rate hardening curves for two steels.
Note the HAZ hardness of the alloy steel does not change much with cooling rate. For the carbon steel
there is a large change of hardness with cooling rate.

Page 36 © Copyrigtit CWB Group - Industry Services

 MODULE 21 I Welding ivletaiiurgy of Steels

4.4 HAZ Toughness

Studies of brittle fractures in welded structures show that the site of initiation of the fracture is often
the HAZ. The HAZ is usually at a point of stress concentration, is subjected to high residual stresses,
and may contain defects. If, in addition, the HAZ toughness Is low, it becomes a prime candidate for
starting a brittle fracture. HAZ toughness is quite difficult to measure because the HAZ is so smaii and
the notches of fracture and impact specimens are hard to locate precisely. Nevertheless, testing for
toughness in the HAZ is increasingly being specified and it is important to understand the factors that
control toughness.

4.4.1 Martensite
Examining the hardening curve of a typical steel indicates that at fast cooling rates the HAZ
transforms to a hard martensitic structure, with the hardness dependent on the carbon content.
This microstructure, unless tempered by a stress relieving heat treatment, is very brittle—
particularly when the carbon content is high. For steels with greater than about 0.2% carbon,
hardness exceeding 450 Hv is obtained in the HAZ of rapidly cooled welds, and the toughness
of the HAZ will be unacceptably low for many applications.

4.4.2 Arc Strikes

Very fast cooling rates are encountered in arc strikes, tack welds and small "cosmetic" weld
beads, and can induce brittle heat-affected zones in steels with sufficient carbon. Many of
the older structural steels had carbon levels over 0.2% and brittle fractures occurring in these
steels could often be traced to these small welds. As a result, welding standards such as CSA
W59 and AWS Dl .1 state that arc strikes are to be avoided and if they occur on cyclically
loaded structures they must be removed by grinding. Pipeline standards require arc strikes to
be repaired or removed.

To avoid the presence of hard, brittle martensitic structures in the HAZ, welding energy input
and preheat processes are selected to ensure an adequately slow cooling rate. The hardening
curves provide guidance on how slowly the weld must cool and the critical cooling rate can be
equated to the minimum heat input or preheat using published cooling rate diagrams. In most
of the newer structural steels e.g., those used for pipelines and offshore oil and gas structures
the carbon content is sufficiently low that martensite formed in the HAZ at very fast cooling
rates is not excessively hard and is, in fact, quite tough. In fact, with most modern HSLA steels
it is the microstructures formed at slow cooling rates that are sometimes of low toughness.
The general trend, therefore, is for HAZ toughness to decrease with increasing heat input as
illustrated In Figure 4.5.

© Copyright CWB Group - Industry Services Page 37

 MODULE 21 [ Welding Metallurgy of Steeis

Heat Affected Zone Toughness decreases when

Toughness hard martensite is produced
at very low heat inputs

Conventional
carbon steel
Increasing heat input

Heat Affected Zone

Toughness

Low carbon
HSLA steel \ ^
Increasing heat Input

Figure 4.5: Diagram illustrating the general effects of heat input on HAZ toughness.

4.4.3 Effect of Grain Size

Research Is not yet sufficiently advanced to provide rules for predicting the toughness of the
HAZ, but certain factors are known to influence it. Grain size is one of the most important and
an increasing austenitic grain size in the HAZ is likely to result in deteriorating toughness. The
grain size is determined largely by the heat input and it may be necessary to restrict heat input
to achieve adequate toughness. In practical terms, restrictions on heat input may mean some
welding processes, such as electroslag or tandem submerged arc, cannot be used. There may
be a maximum interpass temperature and the width of weave of a vertical SMAW weld may be
limited, Some results of fracture toughness tests on an experimental steel are shown in Figure
4.6 and illustrate dramatically the effect of heat input on HAZ toughness.

© Copyright CWB Group - Industry Services

Page 38
 MODULE 21 j Welding Metallurgy of Steels

.40 I 111i
2468
Energy input kJ/mm

Figure 4.6; Effect of fieat input on HAZ tougiiness of a niobium steel. Toughness measured by the
instrumented Charpy test

AAA Precipitation l-fardening

Precipitation hardening from the presence of micro-alloy elements can also lower the HAZ
toughness. Again, precipitation is encouraged by high heat Input because of the longer times
at high temperature and the slower cooling rates. A post-weld heat treatment {stress relief
can cause a considerable amount of precipitation with a substantial decrease in toughness in
the HAZ, as indicated in Figure 4.7. Unlike plain carbon and carbon manganese steels, post-
weld heat treatment of some micro-alloyed steels can cause an increase in the HAZ hardness
due to precipitation of fine carbides. Figure 4.8 shows the effect of post-weld heat treatment
temperature on the HAZ hardness of a niobium-vanadium steel.

2 ol 1 1 1
0 0.05 0.1 0.15

Niobium content wt %

Figure 4.7: Effect of niobium content on HAZ toughness in the as-welded and stress-relieved condition.
Precipitation during stress relief results in deterioration of toughness at high niobium levels.

© Copyright CWB Group - industry Services Page 39

 MODULE 21 I Welding Metaliurgy of Steels

500 600 700

Stress relief temoerature °C
Figure 4.8: Ttie effect of stress relief temperature on fiardness in a micro-alloyed steel.

4.4.5 Other Causes of HAZ Embrittlement

The microstructure itself may have an effect on toughness in the HAZ. Some researchers have
observed low toughness associated with certain bainltic structures, but there is insufficient
data to predict precisely when such structures might occur A further source of embrittlement
in the HAZ is strain aging. The contraction of a cooiing weld may cause substantial plastic
deformation, particularly around defects residing in the HAZ. In certain parts of the HAZ
this may result in dynamic strain aging with a consequent loss of toughness. Since the
phenomenon Is associated with the presence of free nitrogen it is most common in steels that
are not killed, aluminum treated grades.

4.5 Hydrogen Induced Cold Cracking HICC in the HAZ

Martensite formed in the HAZ is highly susceptible to HICC, such as that shown in Figure 4.9. In
carbon or carbon-manganese steels {i.e., those with steep hardening curves welding conditions
can be selected to avoid the cooling rates at which martensite Is produced. But in low-alloy steels
or those where a hard HAZ cannot be avoided, other steps must be taken to prevent cracks. These
often involve applying preheat to accelerate the diffusion of hydrogen out of the weld. When preheat is
used, little hydrogen remains by the time the weld cools to room temperature. Figure 4.10 shows that
quite moderate temperatures are highly effective in removing hydrogen.

Figure 4.9: An example of a liydrogen crack in the HAZ of a weld.

Page 40 © Copyright CWB Group - Industry Services

 _J I I I t
400 300 200 100 0
Temperature of weld during cooling °C

Figure 4.10: Diagram showing the fraction of hydrogen remaining in an example weld as It cools.

4.5.1 Predictive Formula

The preheat temperature required depends on the susceptibility of the HAZ to hydrogen
cracking. Much work has been done to find composition formulae to predict this. One widely
used formula developed by Ito and Bessyo is:

P^,, = C = Si/30 + Mn/20 + Cu/20 + Ni/60 + Cr/20 + Mo/15 + V/10 + 58

CM
This applies to cracking susceptibility in steels of less than about 0.2% carbon and should be
distinguished from the hardenability carbon equivalent given earlier.

The level of preheat also depends on the original hydrogen level in the weld metal. These
two factors—the plate composition and the hydrogen—can be combined in an Index of
susceptibility equal to:

i2PoM + iogH

where H is the hydrogen determined by the standard IIW method. This index is used in an
Appendix to both CSA W59 and AWS D1.1 as part of an alternate guide for estimating preheat
levels.

4.6 Stress Relief Cracking Reheat Cracking

During thermal stress relief the residual elastic strains are converted to plastic strains by a drop In
the material yield point and creep. If these strains are not spread uniformly through the metal, but are
concentrated in local regions, they may be sufficient to cause cracking. This phenomenon, known as
stress reliefer reheat cracking, has been observed in the heat-affected zones of welds in particular
steels. Cracking is more likely in highly restrained joints and very heavy sections and where there is a
stress concentration such as an unfused root or the toe of a large fillet weld.

Precipitation of carbides of secondary hardening elements during stress relief is the primary cause of
cracking. Precipitation occurs preferentially within the grain, causing the grain interior to strengthen
relative to the grain boundary. This shifts the strain onto the grain boundary where cracks may form.

Studies of stress relief cracking reveal that composition has a dominant effect and empirical relations
have been established to predict whether a steel is sensitive to this type of defect.

© Copyright CWB Group - industry Services Page 41

 MODULE 21 i Welding Metallurgy of Steels

One formula suggested in the research literature is:

P^^ = Cr + Cu + 2Mo + 10V + 7Nb + Ti - 2

When P^^ is greater than zero, stress relief cracks may occur. This formuia does not work for low
carbon < 0.1% or high chromium >1.5% and these steels are resistant to stress relief cracking.

5. Weld Metal

5.1 Structure
Since weld metai starts off as a small casting, cools rapidly and does not undergo heavy deformation,
it is not surprising that the properties of weld metal can be quite different from those of the parent plate
and the heat-affected zone. Consequently, the optimum compositions of weld metals are rarely exactly
the same as the steels they are joining. For example, a thick plate of pressure vessel steel of 0.25%
carbon may be joined with a weld metal containing only 0,1% carbon. On the other hand, when alloy
steels are selected for specific properties such as corrosion resistance or high temperature strength,
the weld metai will likely have a similar composition to the base metal.

5.1.1 Columnar Grains

The microstructure of a weld metal is influenced first by the solidification structure that exists
after the metal freezes. The weld metal solidification begins by the liquid transforming to solid
6 ferrite by the addition of atoms to the grains at the weld interface following a crystallographic
pattern straight lines. Subsequently, long austenite grains extend from the weld interface
towards the centre of the weld see Figure 5.1 growing approximately perpendicular to the
isotherms. They are epitaxial with the grains in the HAZ, meaning they grow directly from them.
This is an important observation, since the grain size of the weld metal can be influenced
by the grain size of the HAZ. In most steels the HAZ grain size at the weld interface fusion
boundary depends only on the heat input and not on the steel composition, so the weld metal
grain size is also independent of the steel composition. As demonstrated earlier, however,
some modern titanium steels have very small HAZ grain size and this is reflected in a smaller
weld metal grain size.

Figure 5.1: Weld metal structure showing austenite grains extending from the fusion boundary where
they are epitaxial with the HAZ grains.

© Copyright CWB Group - industry Services

Page 42
 MODULE 21 I Welding MetaNurgy of Steels

5.1.2 Ferrite Morphology

In a typical carbon manganese weld metal, austenite grains transform to a mixture of ferrite
and carbide. Ferrite may form first at the grain boundaries of the austenite and give rise to the
familiar "veins" commonly observed in weld metals and visible in Figure 5.2. Ferrite may also
form within the grain if nucleation sites are available. Small oxide particles provide excellent
sites for nucleation and a weld metal having more than about 250 ppm of oxygen present has
enough oxide particles to nucleate many tiny grains of ferrite within the austenite grain. This
intragranular ferrite, shown in Figure 5.3, is termed acicular ferrite and is associated with
weld metals having good toughness. Ferrite may also be present in the form of spikes or laths
growing into the grain from the edge. This form, sometimes termed side-plate ferrite, has been
associated with poor toughness.

After ferrite is formed, the remaining austenite becomes richer in carbon and can transform
into a number of different structures. The carbon may therefore be present as a carbide in
small islands of martensite or in retained austenite.

Figure 5.2: Proeutectoid ferrite "veins" formed at ttie austenite grain boundaries in a C-l\/ln-Si weld
metal.

Figure 5.3: Acicular ferrite formed within the austenite grain of weld metal. The very small grain size of
this intragranular ferrite contributes to good toughness.

5.2 Weld Metal Composition

The final composition of the weld metal is determined by three factors:

the filler metal composition

• dilution from the parent metai

chemical reactions

© Copyriglit CWB Group - Industry Services Page 43

 MODULE 21 I Welding Metallurgy of Steels

Note that the filler metal composition includes elements coming from the flux as well as other
ingredients present on the electrode coating or in a cored wire. It is not easy to determine the
composition if fluxes are present, and for this reason standards may specify composition in terms
of deposit chemistry which is not necessarily the same as the electrode. This approach is useful
for specification purposes if the method of making the deposit Is standardized. For example, CSA
W48 specifies flux cored electrodes in terms of their deposit chemistry. In the case of solid wires
for shielded or submerged arc welding, the composition can be specified in terms of the actual wire
chemistry. In this case it is important to recognize that the weld deposit may not be the same as the
wire chemistry,

5.2.1 Dilution Diagram

The effects of dilution can be predicted quite accurately for some elements by taking the
proportion of the element contributed by the parent metal and adding it to that contributed
by the electrode and assuming complete mixing. This only works for those elements that
do not enter into chemical reactions, such as nickel and molybdenum. It does not work for
active elements like carbon or manganese. The simple diagram shown in Figure 5.4 assists
in determining the final weld composition. The example shows that an electrode of 8% Ni
on a plate of 3% Ni with a 60% dilution weld would result in a weld metal containing 5% Ni.
The major difficulty in using this method is that the percentage dilution Is not known with any
degree of certainty; however, it is a useful method for setting ranges.

% Element in parent metal

1 23456789
% Element in electrode

Figure 5.4: Diagram to estimate weld metai composition from dilution.

5.2.2 Chemical Reactions

Chemical reactions that take place between the weld metal and the slag or the surrounding gas
can result in significant changes to the weld metal composition. We consider first the reactions
between shielding gases and the weld metal in gas metal arc welding.

The main purpose of the shielding gas is to exclude oxygen and nitrogen from the weld pool,
but this is not always achieved. Some welding-grade gases may deliberately contain oxygen

Page 44 © Copyright CWB Group - Industry Services

 MODULE 21 I Welding Metallurgy of Steels

or active gases and these will react with the weld metal, if iron Is in contact with pure oxygen
gas the oxygen continues to dissolve up to the solubility limit, then forms FeO as a slag on the
surface of the liquid iron. At 1550°C the solubility is about 0.2% [O] but at higher temperatures
it increases to more than 2% {see Figure 5.5.

1500 2000 2500 3000

Temperature °C

Figure 5.5: Solubility of oxygen in liquid steel in equilibrium with pure iron oxide.

If a pool of liquid iron is surrounded by a carbon dioxide CO^ atmosphere, the CO^ reacts to
produce carbon and oxygen, which both dissolve in the liquid iron:

C03 = [C],^ + 2[0],^

Since there is a steady supply of CO^ gas, the reaction above tends to result in equilibrium
concentrations of carbon and oxygen in the iron, regardless of the original levels of carbon and
oxygen present. In the case of a weld in which the weld pool is moving, fresh metal is being
melted ahead of the arc, while solidification takes place behind it and this influences the weld
metal composition. But the constant supply of CO^ shielding gas to the weld pool results in a
similar reaction In which the oxygen and carbon content of the poo! tend toward an equilibrium
value.

At the back of the pool where solidification takes place, the falling temperature causes oxygen
and carbon to react together producing carbon monoxide:

[OU + [C],, = CO

5.2.2.1 Deoxidation
The carbon monoxide tries to escape but, getting trapped, creates porosity.
Porosity can be prevented by lowering the oxygen content in the weld pool
before solidification takes place. This is achieved by adding deoxidants, such
as silicon, to the electrode wire. Silicon reacts with oxygen to form a slag of
silica, which floats out on the surface of the weld metal:

2[0], + [SiL = SiO,

© Copyright CWB Group - Industry Services Page 45

 MODULE 21 [ Welding Metallurgy of Steels

The effects of adding silicon to the pool are that it:

lowers the oxygen content

raises the carbon level at which carbon monoxide could form

• raises the carbon level that is in equilibrium with CO^ gas

5.3 GMAW Electrodes

A typical weld using COj shielding, where silicon Is added through the wire, resuits in a weld metal
carbon content of about 0.1%C, regardless of the original wire or plate carbon level. In practice,
wires can be purchased with various levels of deoxidants to accommodate the range of oxygen that
may be met. Welding rimming steel or rusty plate requires a higher level of deoxidation than welding
a clean, fully killed steel. The effectiveness of silicon as a deoxidant is increased by the addition
of manganese. Further deoxidation is obtained by adding titanium, zirconium or aluminum triple
deoxidized wires. The deoxidizing power of a solid wire electrode depends on the combination of
silicon, manganese and other deoxidants. The list in Table 5.1 shows a number of commercially
available wires in ascending order of deoxidizing power and a guide to their selection.

Table 5.1: GMAW electrodes and their relative deoxidizing power

Classification Typical Electrode Analysis wt% Deoxidation

Power
CSAW48 AWSA5.18 0 Mn Si Other

1
B-G 49A2CG3 ER70S-3 0,10 1.15 0.60

B-G 49A3CG4 ER70S-4 0.10 1.25 0.75

B-G 49A 3 C G6 ER70S-6 0.10 1.60 0,95

Low alloy, high strength 0.10 1.85 0,65 Mo 0.50

AWS A5.28 ER80S-D2
B-G 49A 3 C G2 0,05 1.15 0.50 A! 0.05
Ti 0,07
Zr 0.10

5.4 Fluxes
Fluxes are used in several welding processes, such as submerged arc weiding and flux cored arc
welding. The prime function of the flux in submerged arc welding is to form a slag that protects the
molten weld metal from the air, as Figure 5.6 shows, but there are secondary functions such as arc
stabilization and weld pool support. Naturally, the liquid slag formed should be immiscible with the
molten steel so that two layers form. The melting point of the slag and its density should be lower than
that of the steel so that the slag floats out.

Page 46 © Copyright CWB Group - Industry Services

 MODULE 21 I Weiding Metailurgy of Steels

no t Direction of welding

Liquid slag in contact Solidined slag

'iit^MMTSHS^ftttm-JiiCmiH '"^'aa^ifSiii's detachfis easily

Flux feed

Arc cavity
Molten weld metal

Figure 5.6: Tfie main features of the submerged arc process. Weid metai composition is affected
marf<edly by chemical reactions between the weld pool and the molten slag.

SAA Submerged Arc Fluxes

Since some mechanical support to the weld metal is also desirable, the melting temperature
of the slag should not be too far below that of the steel. A solidification temperature range, as
opposed to a sharp freezing point, is preferable, since a suitable viscosity will assist in giving
a good bead shape. In addition, the thermal contraction of the slag relative to the steel must
allow the slag to detach easily.

Fluxes of metal oxides melting in the range of 1200°C to MOCC are satisfactory and are
chemically quite stable. There are, however, some important chemical reactions that can occur
between the slag and the weld metal. Because most single metal oxides melt well above the
melting temperature of steel, two or more oxides are used in combination, often as naturally
occurring minerals. Figure 5.7 shows the phase diagram for the MnO-SIO^ system, which has
been used as the basis of many commercial fluxes. It is clear from the diagram that appropriate
melting temperatures can be found by carefully selecting the composition.

1900

Two liquids

Si02 + liquid

MnSiOs
+
SiOa

MnO 30 40 50 60 70 80 90 SiO'
Weight % SiO.

Figure 5.7: MnO-SiO^ phase diagram showing how certain compositions may produce a flux with a
melting temperature suitable for welding steel.

© Copyright CWB Group - Industry Services Page 47

 MODULE 21 [ Welding Metallurgy of Steeis

5.4.2 Slag-Metal Reactions

Although silica SiO^ has many desirable properties for a weiding flux, it does become
chemically active at high temperatures. Consider for a moment a pool of liquid iron on which
floats a slag of pure silica. Oxygen and silicon Si in the metal will be in equilibrium with the
silica SiOj such that the concentrations satisfy the equilibrium equation:

K = [SiO,]/[Si][0]^

where K is the equilibrium constant and [ ] represents the activity of the component.

5.4.3 Oxygen and Silicon in Weld Metal

For pure silica the activity is such that the product [Si][0]^ is a constant at any given
temperature. If the silicon content of the iron is raised by adding silicon, the oxygen level must
go down to restore equilibrium. If no silicon or oxygen were initially present in the iron, the
equilibrium levels would be the stoichiometric values where one atom of silicon from the silica
enters the liquid metai for every two oxygen atoms.

If, instead of being pure, the silica is mixed with other components, the activity of the silica is
reduced, depending on the concentration. In the case of submerged arc welding, equilibrium
conditions cannot be assumed, but is has been shown experimentally see Figure 5.8 that the
above equation is approximately true in the sense that the term [Si][0]^ is proportional to the
Si02 content of the flux. The data in Figure 5.8 may be redrawn in terms of weld metal oxygen
and silicon levels for various proportions of silica In the flux. Figure 5.9 shows this.

100

[%Si][%0]2 xlO-*

50

50 60 70

% SiOj in flux

Figure 5.8: Experimental data showing the relationship between the flux silica content and the oxygen
and silicon levels In the weld metal.

Page 48 © Copyrigtit CWB Group - Industry Services

 % Silicon [Si]

Figure 5.9: Weid metal oxygen and silicon levels for various silica contents in the flux. Drawn for the
data shown in Figure 5.8.

The implications of this are important. If the flux contains a high percentage of silica, then the
value of [Si][0]^ is high. In effect, both silicon and oxygen are transferred to the weld metal
from the flux. A lower oxygen level could be achieved by adding more silicon [Si] from the wire,
giving a very high silicon level but keeping the value of [Si][0]^ the same, but to achieve both
lower silicon and lower oxygen the silica content of the flux must be reduced. Silicon in small
quantities may have an apparent beneficial effect on weld metal toughness because It lowers
the oxygen content. But, as iliustrated in Figure 5.9, high levels of silicon have a diminishing
effect on oxygen, and high levels of silicon increase the yield strength and lower the toughness.

5.4.4 Oxide Inclusions

Oxygen does not remain in solution when the metal solidifies, but forms small, round
inclusions. These inclusions have a marked effect on the energy to cause ductile fracture in the
steel, a measure of which is provided by the upper shelf of the Charpy transition curve. Thus,
the oxygen content of the weld metal directly influences the Charpy impact properties, This
becomes an important factor in selecting a flux when impact properties are a consideration.
High-silica fluxes have many desirable properties, such as high current capacity and
excellent bead shape control, but their use should be restricted to those applications where
high toughness is not a prime consideration. Some commercial high silica fluxes are oniy
recommended for single pass welds.

5.4.5 Flux Basicity

The concept of oxygen transfer from fluxes can be extended to other flux components and
has given rise to the basicity index as a measure of potential oxygen transfer. Although not
originally based on oxygen reactions, experiments have shown that the basicity index relates
quite well to the weld metal oxygen, as Figure 5.10 demonstrates.

© Copyright CWB Group - industry Sen/ices Page 49

 MODULE 21 I Weiding Metallurgy of Steels

Ttie basicity index is given by:

CaO + MgO + Na^O + K^O + LI2 + CaF^ + 1/2MnO +FeO

SiO^ + 1/2Al203 + TiOj + ZrO^

0.10 r

0 1 2 3
Flux basicity index

Figure 5.10: Effect of flux basicity on the oxygen content of weld metal.

Commercial basic fluxes w'Ah high basicity are often specified when excellent toughness at low
temperatures is required.

5.4.6 Active Fluxes

Up to this point the discussion has been concerned with chemical reactions with the oxide
components of the flux. Not all the flux components, however, are oxides. In some fluxes made
by the bonding or agglomeration process, "active" deoxidants may be present. Silicon in the
form of ferro-silicon may be added to improve the deoxidizing capacity of the flux, allowing it
to handle rusty plate. The amount of silicon transferred to the weld pool this way depends on
the quantity of flux melted, which in turn depends on the arc length and voltage. A high voltage
causes a long arc within the flux and results in more flux being melted. Figure 5.11 shows
typical results illustrating the effect of voltage on flux consumption. Since voltage has iittle
effect on the electrode melting rate, an increase in voltage increases the concentration of any
active element transferred from the flux.

© Copyright CWB Group - Industry Services

Page 50
 MODULE 21 [ Weiding Metallurgy of Steels

40 r

30 h

20 h
u
< 600 amp DC
after Jackson and Shrubsall

10
0.1 0.2 0.3 0.4 0.5 0.6

Rate of flux melting Ib/min

Figure 5.11: Tfie amount of flux melted is directly proportional to the arc voltage.

5.4.6.1 Effect of Voltage

Figure 6.12 shows the weld metal silicon contents for two types of flux as a
function of voltage. Voltage sensitivity is important when making multipass
welds, since it is a common practice to increase the voltage on the final
layers to spread the bead and get a good weld profile. With voltage-sensitive
fluxes this could result In a silicon or manganese or other elements buildup
leading to undesirable properties. Other alloy elements such as chromium
are present in some alloy fluxes used for welding weathering steels, stainless
steels and other alloy steels. If fluxes of this type are employed, very careful
attention to voltage is required to ensure proper control of weld metai
composition.

0.3 I 1 1 1
20 30 40 50
Arc Voltage

Figure 5.12: The amount of silicon transferred from active fluxes depends on the arc voltage.

© Copyright CWB Group - Industry Services Page 51

 MODULE 21 I Welding Metallurgy of Steels

5.5 Properties of Welds

In most cases the strength of the weld metal Is determined by the composition and whether or not
the weld is subject to post-weld heat treatment. Welding heat input may also have an effect, since
it controls the cooling rate of the weld, Although an accurate prediction of weld metal strength from
composition is not possible, some general trends have been established for carbon-manganese-
silicon weld metals. In most cases the ultimate tensile strength UTS is of greatest importance, since
weld metal yield strength is usually substantially higher than that of the plate when the ultimate tensile
strength matches.

5.5.1 Tensile Strength

Figure 5.13 shows the relationship between composition and the UTS of gas metal arc
welds made under CO^ shielding. Wires of different deoxidizing capability produce a range of
manganese and silicon levels in the weld metal. High levels of these elements lead to weld
metal of high strength and hardness, which may have unacceptably low ductility. Figure 5.14
shows a similar relation for submerged arc welds. The wide scatter of results indicates that
several factors other than composition affect the strength.

UTS MPa

Figure 5.13: Relationstiip between the tensile strength and the weld metal composition for CO^ welds.

© Copyright CWB Group - Industry Services

Page 52
 MODULE 21 I Welding Metallurgy of Steeis

UTS MPa

Figure 5.14: Reiationsiiip between the tensiie strength and the weld metal composition for submerged
arc welds.

5.5.2 Toughness
Toughness cannot be simply related to composition, and a considerable amount of research
has been done to determine the factors that control toughness. An understanding of these
factors is necessary when faced with the problem of selecting welding consumables that must
meet certain toughness requirements.

5.5.2.1 Ductile Fracture

Ductile fracture in steel results from the growth and subsequent joining of
microvoids that form around inclusions within the metal. How the presence
of inclusions from deoxidation influence ductile fracture as represented by
Charpy energy have been presented previously. Reduction in the numberof
inclusions particularly large inclusions by lowering the weld meta! oxygen
can lead to substantia! improvements in toughness. Low oxygen is achieved
through the use of basic fluxes or processes with low oxygen potential such
as inert gas shielded welding. In the case of flux cored arc welding, the effect
of the flux can override that of the gas, so that using an inert gas alone is not
sufflcient to ensure high toughness. The progress made in developing flux
cored wires with very high toughness was recognized in the CSA Standard
W48, which provided for classification of wires with impacts tested down to
-40''C.

5.5.2.2 Effect of Microstructure

Microstructure has a pronounced effect on toughness, although it is not
always possible to predict the microstructure from the composition. CCT

© Copyright CWB Group - Industry Services Page 53

 MODULE 21 I Welding Metallurgy of Steels

or structure diagrams for weld metal are useful In helping to determine the
microstructure that may be expected. Two examples of weld metal structure
diagrams are shown in Figure 5,15 and Figure 5.16. They show the
proportion of each microstructural constituent expected at various cooling
rates. Research has shown that structures containing a large percentage of
intragranular, acicular ferrite are associated with good toughness.

•a 100 r

Grain boundary
proeutectoid
ferrite

'Side plate"
ferrite

Intragranular
acicular ferrite

30 40 50 6070 100

Weld cooling time 800-500 s

Figure 5.15: Weld metal structure diagram showing the percentage of each constituent present. This
weld metal exhibited a low percentage of acicular or intragranular ferrite.

0
tn
s Other constituents
5 801- Grain boundary
proeutectoid
ferrite
Ic 60 k
o
o
5b. 40 h
= "Side piate"
•J
u ferrite
3
*j
V 20 h Intragranular
o acicular ferrite
o
g J I I I IIII J I I I IIIM
3 4 5 6 7 10 20 30 40 50 6070 100

Weld cooling time 800-500 X s

Figure 5.16: Weld metal structure diagram. This weld metal had a medium oxygen level and showed a
high proportion of intragranular, acicular ferrite.

Page 54 © Copyrigtit CWS Group - Industry Services

MODULE 21 I Weiding Metailurgy of Steels

5.5.2.3 Grain Size

As it does in the heat-affect zone, the grain size appears to play a prominent
role In determining weld metal toughness, and low heat input and fast cooling
rates generally give better toughness.

5.5.2.4 Multipass Welds

In multipass welds, the heat from subsequent passes refines the grain
structure and may improve the properties. A welding procedure giving a high
proportion of refined structure can result in better toughness, even though
the heat input per pass may be higher Lower impact values are usually
recorded in the root region of groove welds in thick sections. Factors believed
to contribute to the loss of toughness are dynamic strain aging, high dilution
and rapid cooling rates. Many specifications call for impact specimens to be
tested from this region see Figure 5.17.

-J I I 1 I I
-60 .40 -20 0 20 40
Charpy test temperature "C

Figure 5.17: Tfie location of tlie impact specimens has a marked effect on the impact toughness
measured. The root region usually shows the lowest toughness.

5.5.2.5 Effect of Post-weld Heat Treatment

Both strength and toughness can be affected by post-weld heat treatment.
There is usually a reduction in strength, but toughness may either increase
or decrease. The extent of strength loss depends on both the temperature
and the time of post-weld heat treatment. It is possible to combine these two
factors into the single Holloman-Jaffe parameter introduced earlier:

H = r20+logO

This applies only when the temperature is steady. If the temperature is

changing, for example during the heating and cooling stage of a stress relief
treatment, then a correction factor can be added to the time. The correction
is:

f^ = T/2.3/f[20-log/]

where k is the heating or cooling rate In degrees K per hour.

© Copyright CWB Group - Industry Services Page 55

 MODULE 21 I Welding Metailurgy of Steels

In Figure 5.18 the UTS of typical carbon-manganese and molybdenum weld

metals is plotted against the parameter H. Over the practical range of interest
it shows a linear relationship, If the initial strength of a weld Is known, this
graph can be used, for example, to determine the maximum stress relief time
at a given temperature before the weld metal strength drops below a specified
value.

600 r-

Figure 5.18: Drop in the tensile strength of weld metai resulting from stress relief. The drop depends on
time and temperature and the weld metal composition.

5.5.3 Effect of Toughness

Unfortunately it is not possible to predict the change in toughness of a weld metal subject to
a post-weld heat treat PWHT stress relief. In many cases the toughness can improve by
a substantial amount, as shown in the example in Figure 5.19. Conversely, deterioration of
toughness could occur, for example due to precipitation hardening. As with the heat-affected
zone, the presence of certain micro-alloy elements, such as niobium, in sufficient quantities
causes precipitation during stress relief resulting in increased hardness and loss of toughness.
Figure 5.20 illustrates this effect. CSA W59 places a limit of 0.05% on weid metal vanadium
content when welds are to be stress relieved.

© Copyriglit CWB Group - Industry Services

Page 56
 MODULE 21 j Welding Metaliurgy of Steels

-60 -40 -20 0 20 40 60

Charpy test temperature °C

Figure 5.19: A stress reiief PWHT generally improves the toughness ofC-Mn weld metal, as shown in
these typical results.

20
Stress relieved

3 0
1
a
E -20
0
c As welded
o
tn -40 h
c
CO

>» -60
a
I.
11 Two pass submerged arc welds
JZ
O
0 0.01 0.02 0.03 0.04 0.05
Nb + 1/2 V in weld metal Weight %

Figure 5.20: Stress relief of weld metal containing significant levels of niobium columbium
usually causes a deterioration of toughness, as shown here by the increase in toughness transition
temperature.

5.6 Hydrogen Induced Cold Cracking HICC in the Weld Metal

Hydrogen induced cold cracking can occur in either the heat-affected zone or the weld metal, and
can be either transverse or longitudinal to the weld axis. The level of preheat or other precautions
necessary to avoid cracking will depend on which region, weld metal or HAZ, is the more sensitive.
In the older carbon-manganese medium-strength structural steels, the HAZ was usually the more
critical region and weld metal rarely caused a problem. Modern steels with low carbon are now very

© Copyright CWB Group - Industry Services Page 57

 MODULE 21 I Welding Metallurgy of Steels

resistant to hydrogen cracking. This resistance is attributed to the low carbon. Even if a hardened
structure is produced, the hardness will not be very high. High strength consumables may have higher
hardenability than the base metal, thus procedures need to be designed to avoid hydrogen induced
cold cracking in the weld metal.

HICC in the weld metal depends on the same fundamental factors as in the HAZ, that is, hydrogen
content, microstructure hardness and stress on the weldment. The controlling variables in practice
are usually: strength level, hydrogen content, restraint and stress concentrations. In single-pass welds
and the root bead of multipass welds, the gap provides a stress concentration to the transverse stress
and this can lead to longitudinal cracks in the weld metal.

The higher dilution of the root bead often results in a harder weld bead that is more likely to crack.
This is the predominant form of cracking in applications such as pipelines. In cases where transverse
notches exist, such as an unspliced backing bar, the longitudinal stress becomes the controlling
factor and transverse cracks may occur in heavy multipass welds, cracking will generally be in the
transverse direction, either normal to the surface or at an angle of about 45°. The latter type has
been called "chevron cracking". The solution to these cracking problems rests with proper control
of hydrogen through proper handling of consumables, drying of fluxes and electrodes, and the
application of the correct preheat

Weld metal hydrogen induced cold cracking can also be a problem when welding modern high
strength steels with high toughness requirements. To match the base metal properties the
consumables usually need to contain Ni, which means the weld metal has higher hardenability than
the base metal. This Is particularly true when multipass weiding is used weld beads that are not
diluted by base metal will have a higher Ni content. It is important to remember that hydrogen induced
cold cracking can be delayed by as much as a few days due to the diffusion of hydrogen and the
buildup of stresses during cooling to ambient temperature. Therefore, inspection of weldments should
not be performed until sufficient time is allowed for hydrogen diffusion and weldment cooling.

5.6.1 Summary of Methods to Avoid Hydrogen induced Cold Cracking

The major factors that affect HICC are the microstructure hardness, the stress on the weld
joint including concentrations and the presence of hydrogen. Without hydrogen present,
hydrogen cracking cannot occur.

A list of possible methods used to avoid this kind of cracking is presented below:

use low hydrogen consumables lower hydrogen content in weldment

preheat/heat input lower hydrogen content in weldment, softer microstructure for steels
with steep hardening curve

• maintain preheat after welding lower hydrogen content—may have high hardness

isothermal transformation/preheat above temperature produces bainite structure; use

where PWHT Is not possible

temper bead technique temper hardened structure by appropriate bead placement; use
where PWHT is not possible

© Copyrigtit CWB Group - industry Services

Page 58
 MODULE 21 I Welding Metallurgy of Steels

• use austenitic weld metal when preheat is restricted weld metal not sensitive to hydrogen
but has hard HAZ; use where preheat is not possible

post-weld heat treatment temper hardened structure; applied immediately after welding
and above hydrogen cracking temperature apporox. 150°C, where hard HAZ is produced
by welding

6. Summary
Steels may employ a wide variety of methods to achieve specific combinations of desirable properties.
Modern steels with their demands for strength, toughness and good welding behaviour have evolved to
depend less on carbon content as a strengthening agent and more on fine grain size and precipitation
hardening. This has meant a shift in the types of problems faced by the welding engineer, who may now
have to determine a welding procedure and select consumables to meet stringent property requirements as
well as to avoid cracking and other defects during welding.

This module described how steel may be heat treated to alter its properties and how such changes may
take place in the heat-affected zone of a weld. The effects of welding on the properties of the heat-affected
zone have been discussed, as well as those aspects of the welding procedure that must be addressed to
achieve adequate properties. The effects of various fabrication heat treatments, such as post-weld stress
relief, have been discussed, together with their effects on properties.

Several major types of cracking problems have been presented, along with guides to the procedural
controls necessary to avoid them. This module presented an overview of how welding procedure variables
affect the properties of welds in steel, and how to estimate the procedures necessary to avoid metallurgical
problems.

7. Additional Resources
Below is a brief list of additional sources of information that you may find valuable in your study of the
welding metallurgy of steels.

1. A comprehensive work on all aspects of steel making but not covering welding or fabrication is:

The Making, Shaping and Treating of Steei, United States Steel, Pittsburg, Pa., USA.

2. The basic principles of heat treatment of steel are covered in:

Reed-Hill, R.E., Physical Metallurgy Prindpies, D. Van Nostrand Company, Inc.

3. Recommended sources for information on welding of steel are:

Welding i^andbook. Vol. 4 American Welding Society, 550 N.W. LeJeune Road, Miami, FL
33126, USA.

Metals Handbook, Vol 6 American Society for Metals, Metals Park, Ohio 44073, USA.

4. If you are interested in specific welding problems such as hydrogen cracking, solidification
cracking and so forth, you should consult the research literature published in major journals
such as the AWS Welding Journal or the CWA Weiding Journal

© Copynght CWB Group - industry Services Page 59

 MODULE 21 \ Welding Metallurgy of Steels

Guides and Exam Exercises

To obtain maximum benefit from this moduie we suggest that you follow this guide and complete the
exercises as indicated. It is important that you work through the text methodically, studying each section
thoroughly before moving on. The exercises are designed to give you an indication of whether you have
learned the material and can move on or whether you need to go back and study the section again.

Do the exercises honestly, They will not help you unless you take them seriously. If you get a question
wrong, go back through the text until you understand where you have gone wrong and know the correct
answer.

The length of time required to complete the module will vary from student to student. Find you own pace.
Do not rush. Remember you are trying to teach yourself something, not win a race.

Some people like to underline sections when they read a text. We suggest that you use caution if you
do this. What you think is important first time you read it may be different after reading it three times. We
suggest you read a section three times thoroughly before highlighting anything.

The exam exercise is designed to give you an indication of whether you are ready to take the closed-book
exams. The exercise questions are of a similar standard to the official exams. Do not take the exam until
you feel you are ready. You may wish to study several modules before taking the exams on each. If you
have any difficulties with this module do not hesitate to ask for help.

You may contact us at 1-800-844-6790, fax us at 905-542-1318 or email us at info@cwbgroup.org.

Remember, if you fail the examination you can always try again later.

© Copyriglit CWB Group - Industry Services

Page 60
 MODULE 21 I Welding Metailurgy of Steels

Guide 1

Carefully read the Sections 2.1 to 2.3.10 and answer the following questions. Check your answers
for accuracy. If any of your answers are wrong, re-study the guide subject matter in the text until you
understand it.

The maximum solubility of carbon In a iron Is:

a 2%.

b 0.002%.

c 0.02%.

d There is no solubility.

True or false?

When a steel Is heated into the austenite region, all carbon and most other compounds dissolve
to form a single phase.

True or false?

Peariite is a single phase.

4. What is the resulting structure of a 0.6% carbon steel slowly cooled from 900°C to room
temperature?

Most fine grain steels retain their fine grains:

a up to the melting temperature.

b up to the grain coarsening temperature, but above that behave as coarse grain steel.

c oniy up to the transformation temperature.

© Copyright CWB Group • Industry Services Page 61

 MODULE 21 I Welding Metallurgy of Steels

Guide 2

Carefully read Sections 2.3.11 to 2.5.4 and answer the following questions. Check your answers for
accuracy. If any of your answers are wrong, re-study the guide subject matter in the text until you
understand it.

1. Why is martensite produced rather than pearlite when a steel is rapidly quenched from 900°C?

2. The hardness of martensite depends mainly on the:

a alloy content.

b rate of cooling.

c carbon content.

d size of the part.

3. Can bainite normally be produced In continuous cooling of a plain carbon steel?

4. Increasing the carbon content:

a lowers the martensite start temperature M^.

b raises the M^.

c has no effect on M^.

5. Calculate the ideal critical diameter hardenability for a steel with 0.2%C, 0.25%Si, 1,0%Mn,
0.5%Cr, 0.5%Mo, 1.0%Ni assume an initial grain size of ASTM 8. Use section 2.5,3 and 2.5.4
as a reference.

© Copyright CWB Group - Industry Services

Page 62
 MODULE 21 I Welding Metaliurgy of Steels

Guide 3

Carefully read Sections 2.5.5 to 2.5.9 and answer ttie following questions. Ctieck your answers for
accuracy. If any of your answers are wrong, re-study the guide subject matter in the text until you
understand it.

1. Using Figure 2.34, estimate the fastest rate a weld could cool without the HAZ becoming
harder than 350 HV for a steel of 0.2%C, 0.18%Si, 1.2%Mn, 0.35%Cr.

2. What happens to the carbon trapped in the tetragonal structure when martensite is tempered?

3. Secondary hardening during tempering results from:

a retained austenite.

b a slow heating rate.

c the presence of strong carbide formers.

4. If a thermal stress relief is applied to a welded quenched and tempered steel, the temperature
should not exceed the:

a transformation temperature.

b original tempering temperature.

c grain coarsening temperature.

© Copyrigtit CWB Group - industry Services Page 63

 MODULE 21 I Weiding Metailurgy of Steels

Guide 4

Carefuliy read Section 3 and answer the following questions, Check your answers for accuracy. If any of
your answers are wrong, re-study the guide subject matter in the text until you understand it.

1. The effect of increasing the carbon content of a steel on the Charpy transition curve is to:

a lower the transition temperature.

b raise the upper shelf

c raise the transition temperature and lower the upper shelf.

2. What simple heat treatment is widely used to improve the toughness of structural steeis?

3. The main purpose of using a lower finishing temperature during controlled rolling of steei is to:

a limit re-growth of austenite grains.

b allow time for precipitates to dissolve.

c reduce the force necessary for rolling.

4. If plates are deformed more than 5% during fabrication, the simple heat treatment often
specified to restore ductility is:

a a full furnace anneal at 900°C,

b normalization.

c stress relief, typically at 635°C.

© Copyrigtit CWB Group - industry Services

Page 64
 MODULE 21 I Welding Metallurgy of Steels

Guide 5

Carefully read Section 4 and answer the following questions. Check your answers for accuracy, if any of
your answers are wrong, re-study the guide subject matter In the text until you understand it.

1. The austenite grain size close to the weld interface fusion boundary is determined mainly by
the:

a composition of the steel.

b cooling rate.

c weld heat input.

2. One of the main reasons for limiting the maximum heat input when welding HSLA steels is to:

a keep the HAZ hardness below 350 HV.

b preserve the HAZ toughness.

c promote a bainite structure.

3. True or false?

A post-weld heat treatment stress relief always improves the fracture toughness of the HAZ.

4. Name two ways in which preheating may reduce the risk of hydrogen cracking in structural
steels.

5. Would you expect a 1%Cr, 1%Mo, 0,35%V steel to be sensitive to stress relief cracking?

© Copyright CWB Group - industry Services Page 65

 MODULE 21 I Welding Metallurgy of Steels

Guide 6

Carefully read Sections 5 to 5.4 and answer the following questions. Check your answers for accuracy, if
any of your answers are wrong, re-study the guide subject matter in the text until you understand it.

1. Acicular ferrite in weld metal is:

a proeutectoid ferrite formed at the grain boundaries,

b a type of martensite.

c ferrite nucleated within the austenite grains.

2. Wires for CO^ welding usually contain silicon to:

a increase the strength of the weld.

b allow easy feeding of the wire.

c deoxidize the weid pool.

3. Which wire has the greatest deoxidizing power?

a B-G 49A 3 C G2

b B-G 49A 2 C G3

c B-G 49A 3 C G6

4. If you wished to obtain a deposit with very low silicon using SAW you would use the
combination of a:

a low silicon wire with an acidic non-basic fiux.

b high silicon wire with a basic flux.

c low silicon wire with a basic flux.

5. When might it be appropriate to specify basic fluxes for SAW?

6. Which welding parameter has the largest effect on the transfer of elements from an active flux?

© Copyright CWB Group - Industry Services

Page 66
 MODULE 21 I Welding Metallurgy of Steeis

Guide 7

Carefuiiy read Sections 5.5 to the end and answer the following questions. Check your answers for
accuracy. If any of your answers are wrong, re-study the guide subject matter in the text until you
understand it.

1. Will the use of argon shielding with flux cored arc welding always ensure a low oxygen deposit?
Explain.

2. The region of a weld where the lowest Charpy impact values are usually recorded is:

a close to the weld interface.

b the root region.

c the surface.

3. A weld metal had a strength of 500 MPa after a post-weld stress relief of two hours. If the
time were extended to four hours is it likely that the weld metal would still exceed a specified
minimum strength of 480 MPa?

4. Does a post-weld stress relief always improve the toughness of a weid metal? Yes or No?

© Copyrigiit CWB Group - industry Services Page 67

 MODULE 21 I Welding Metallurgy of Steels

Guide Answers

Guide 1

1. b

2. True

3. False

4. About 25% proeutectoid ferrite and 75% pearlite

5. b

Guide 2

1. insufficient time for carbon to diffuse to form carbides.

2. c

3. No

4. a

5. 5.0

Guide 3

1. 30°C/s at 540°C.

2. The carbon is released

3. c

4. b

Guide 4

1. c

2. Normalizing

3. a

4. c

© Copyright CWB Group - Industry Sen/ices

Page 68
MODULE 21 J Wekiing Metallurgy of Steels

Guide 5

1. c

2. b

3. False

4. Slow cooling to produce softer structure and facilitate removal of tiydrogen

5. Yes

Guide 6

1- c

2. c

3. a

4. c

5. When excellent toughness at low temperature is required

6. Voltage.

Guide 7

1. No, it depends on the flux basicity.

2. b

3. Yes

4. No

© Copyright CWB Group - Industry Services Page 69

 MODULE 21 I Welding Metallurgy of Steels

Exam Exercise
This test is designed to determine whether you are ready to attempt the forma! examination.

Complete the ANSWER SHEET and compare the results with the ANSWER KEY. If you score less than
70%, we suggest you restudy the material.

1. Steels form the largest group of commercially important alloys because of the:

a great abundance of iron in the earth's crust.

b relative ease of extraction and low cost.

c wide range of properties that can be achieved as a result of the solid state phase
transformations.

d All of the above.

2. Under equilibrium conditions, for a plain carbon steel the average composition of pearlite is:

a 0.02% carbon.

b 0.2% carbon.

c 0.8% carbon.

d 1.2 % carbon.

3. Martensite is a phase produced in hardenable steels that are quenched rapidly because the:

a reduced diffusion causes the carbon atoms to distort the usual BCC crystal into a BCT
crystal by a shear mechanism.

b high thermai gradient produces high strains.

c steel is restricted from expanding by the cool outer surface during quenching.

d carbon atoms accumulate on the austenite grain boundaries.

4. Hardenability of steel is affected by:

a carbon content and austenite grain size.

b carbon content and small mass.

c alioy content and austenite grain size.

d the mass of the part and the cooling rate.

© Copyright CWB Group - Industry Services

Page 70
 MODULE 21 \ Weiding Metaliurgy of Steels

5. Secondary tiardening Is caused by:

a ttie formation of cementite during stress relief heat treatment.

b the formation of alloy carbides during stress relief heat treatment.

c the formation of cementite on the prior austenite grain boundaries during stress relief
heat treatment.

d a slip mechanism due to the high thermal gradients.

6. Bainite can only be produced in a hardenable plain carbon steel through:

a rapid quench.

b normalizing.

c isothermal transformation.

d high austenite temperature followed by rapid quench.

7. The only strengthening mechanism that will also increase toughness is:

a grain size refinement.

b precipitation hardening.

c secondary hardening.

d work hardening.

8. The region in the heat-affected zone that has the highest hardenability is the:

a fine grain region.

b partial transformation region.

c coarse grained region.

d base metal region.

9. The general trend for HAZ toughness in steel Is:

a higher heat input gives increased toughness.

b higher heat input gives lower toughness.

c heat input has no effect on HAZ toughness.

d the addition of oxygen increases HAZ toughness.

© Copyriglit CWB Group - Industry Services Page 71

 MODULE 21 I Weiding Metallurgy of Steels

10. The major factors affecting hydrogen cracking are:

a microstructure hardness, local stress and weld zone hydrogen.

b hydrogen, hardenability and solidilication range.

c hydrogen, cooling rate and welding process.

d hydrogen, microstructure and temperatures above 200°C.

© Copyrigtit CWB Group - Industry Services

Page 72
 MODULE 21 I Welding Metaliurgy of Steeis

Answer Sheet—Module 21
Complete ttie "Answer Stieet" and compare the results with the "Answer Key". If you have a pass mark less
than 70%, you are advised to re-study the material.

Please circle only ONE letter corresponding to the answer you think is most correct.

QUESTIONS ANSWERS

1 a b G d

2 a b 0 d

3 a b 0 d

4 a b c d

5 a b c d

6 a b c d

7 a b 0 d

8 a b 0 d

9 a b c d

10 a b c d

The answer key below Is provided for your use in the event that you wish to retest yourself.

QUESTIONS ANSWERS

1 a b c d

2 a b 0 d

3 a b c d

4 a b c d

5 a b c d

6 a b c d

7 a b c d

8 a b c d

9 a b c d

10 a b c d

© Copyrigtit CWB Group - Industry Services Page 73

 MODULE 21 i Welding Metallurgy of Steels

Answer Key—Moduie 21
Compare your answer sheet to this key.

QUESTIONS ANSWERS

1 a b c

2 a b c d

3 a b c d

4 a b c. d

5 a c d

6 a b d

7 b c d

8 a b d

9 a b c d

10 a b c d

Page 74 © Copyrigtit CWB Group - Industry Services