Professional Documents
Culture Documents
Second edition
J . G . HICKS
Consultant in welded design,
fabrication and quality assurance
A B I N G T O N P U B L I S H I N G
Woodhead Publishing Limited in association with The Welding Institute
First published by Granada Publishing in Crosby Lockwood Staples 1979
Second edition published by BSP Professional Books 1987 (ISBN 978-0-632-01874-1)
Second edition reprinted by Abington Publishing, Woodhead Publishing Limited,
Abington Hall, Abington, Cambridge CB1 6AH, England, 1997
Conditions of sale
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise without the prior permission of the copyright owner.
ISBN 978-1-85573-337-4
1987 J. G. Hicks
Acknowledgements
Manual Metal A r c
Completed weld
tions which cleanse and refine the molten metal a flux added separately in the form of granules or
before it solidifies into the weld bead. The flux powder over the arc and weld pool.
coating also generates a local gaseous atmosphere The flux has four principal functions:
which contributes to the stability of the arc and (a) To protect the molten metal by forming a protec
prevents absorption of atmospheric gases by the tive slag layer.
weld metal. (b) To stabilise the arc.
Electrodes for manual metal arc welding are (c) To cleanse the molten metal, particularly when
made w i t h core wire diameters from 2 m m to 10 there is rust on the surface.
mm although the range generally used is from 2.5 (d) To control the composition of the weld metal.
to 6 mm in lengths of between 2 0 0 and 4 5 0 m m .
There are many types of electrodes. The main Fluxes can be classified by their method of
differences between them are in the flux coating. manufacture and their chemical characteristics.
There are three principal groups of electrode for Flux may be made by melting the constituents
carbon and carbon-manganese steels used in most together and then grinding the solidified mix w h e n
conventional fabrications. it has cooled, by bonding the constituents together,
Rutile electrodes have a high proportion of tita or by simply mixing the component minerals.
nium oxide in the flux coating. They are relatively The chemical characteristics range from the acid
easy to use and might be termed general-purpose types, containing manganese or calcium silicates
electrodes for applications which do not require together w i t h silica, to the basic types again con
strict control of mechanical properties. The steels taining calcium silicate, but w i t h a lower proportion
on which they are used should have good weldabil- of silica than the acid types.
ity (see Chapter 2). The acid fluxes are used for general-purpose
Basic electrodes have a coating w h i c h contains welding, whereas the basic fluxes are used for
calcium and other carbonates and fluorspar. They welds requiring control of fracture toughness and,
are more difficult to use than the rutile electrodes for steels of high hardenability, to prevent cracking.
but can produce welds w i t h better strength and The electrode wire is usually of a 0.1 per cent
notch toughness. If they are dried in a properly carbon steel w i t h a manganese content of between
controlled manner they produce welds w i t h low 0.5 and 2 per cent and w i t h a relatively low silicon
amounts of hydrogen and can be used to weld the content (0.2 per cent).
more hardenable steels without the risk of cracking The arc is completely enclosed, and high current
(see Chapter 2). can be used without the risk of air entrainment or
Cellulosic electrodes have a high proportion of severe spatter. A high current gives the weld pool a
combustible organic materials in their coating. The deep penetration into the parent metal and thicker
arc produced by this type of electrode is very sections can be welded without edge preparation
penetrating and is often used for the root runs in than w i t h manual metal arc welding.
pipeline welding and occasionally in structural The process is used mainly in a mechanised
work. The high quantities of hydrogen w h i c h are system feeding a continuous length of electrode
released from the coating require that precautions from a coil on a tractor unit, which carries the
be taken to prevent cracking in the steel after welding head along the joint. Alternatively, the
welding. welding head may be fixed and the job traversed or
Both the rutile and the basic electrodes can have rotated under it. A welding head may feed several
iron powder added to the coating. This improves wires, one behind the other, so that in effect a multi-
productivity by producing more weld metal for the run w e l d can be made in one pass. AC or DC can be
same size of core wire and hence the same welding used in submerged arc welding, and w i t h a multi-
current. The larger weld pool which is created head unit DC and AC may be used w i t h the different
means that the iron powder electrodes cannot be so wires - DC on the leading wire to give deep
readily used in all positions as the plain electrode. penetration and AC on the other wires to provide a
high w e l d metal deposition rate. Welding currents
of up to 1000A per wire can be used, but where
mechanical properties are to be controlled a weld
Submerged A r c can be made w i t h a number of small runs. Manually
This welding process uses a bare wire electrode and operated versions of the submerged arc welding
Arc Welding Processes 3
A 3 9 m m thick, 2.3 m diameter steel tube, part of a n oil production platform for BP's North Sea Forties Field, being welded by a
submerged arc w e l d i n g installation at t h e Graythorp, Hartlepool, w o r k s of Laing Pipelines Offshore.
Note that t h e operator needs only normal industrial clothing, t h e arc is screened by t h e flux.
Some 4 3 k m of w e l d i n g holds together t h e 2 5 , 0 0 0 t o n n e s of steel tube that makes up the platform a n d flotation raft. The platform
stands in 1 2 2 m of w a t e r in t h e North Sea.
Photograph by courtesy of GKN Lincoln Electric Ltd.
4 Welded Joint Design
process are used in which current levels are limited Gas Shielded Welding
to about 400A.
As a mechanical process, submerged arc welding In this process a bare wire electrode is used, and a
is capable of greater consistency and productivity shielding gas is fed around the arc and weld pool.
than manual welding, although to balance this the This gas prevents contamination of the electrode
process is not suitable for areas of difficult access and weld pool by air, and provides a local atmos
and multiposition work in situ. Because the wire phere giving a stable arc. The process is known as
and flux are separate, the welding engineer can MIG (metal inert gas) w h e n argon or helium gases
choose a combination w h i c h will give the required are used - generally for non-ferrous metals such as
weld metal properties and can exercise further aluminium, titanium, and nickel alloys. For carbon
control w i t h the heat input and welding speed. and carbon manganese steels, the gas commonly
W e l d i n g a vehicle body w i t h t h e gas shielded w e l d i n g process. The welder w e a r s leather gloves and a leather apron t o protect h i m
f r o m heat a n d spatter and looks at t h e w e l d t h r o u g h a mask w i t h a dark glass screen. The tube coming away f r o m t h e w e l d i n g g u n
above t h e w e l d e r ' s hand is extracting t h e f u m e s produced at t h e point of w e l d i n g .
Photograph by courtesy of GKN Lincoln Electric Ltd.
Arc Welding Processes 5
Welding in Practice
Flat position 1
Horizontal-vertical position 2
Vertical position 3
Overhead position 4
Arc Welding Processes 7
The limits of intermediate positions are given in Welding in the flat position allows a greater rate
the relevant Standard, e.g. B.S. 499: Part 1: 1983. of w e l d metal deposition than the other positions, in
A.W.S. D1.111 -86. The rate at which weld metal can w h i c h the maximum size of weld run is limited by
be placed and, therefore, the speed of welding the tendency of the molten weld metal to run out of
depends on the current drawn. The size of the weld the joint. In these positions the weld may have to be
pool, the arc forces, and the heat dissipated place a built up by depositing a number of smaller runs.
top limit on the current at w h i c h a welder can For example w i t h a 9 m m leg length weld made w i t h
produce a consistently sound weld; a practical manual metal arc rutile or basic coated electrodes:
maximum amperage for manual work in the flat
position is about 400A. The mechanical properties
of the welded joint are influenced by the heat input
and can place a limit on the current and therefore
the size of the weld bead.
This limitation on operating current also affects Full joint penetration is made possible by leaving
the depth of penetration into the parent plate. If t w o a gap between the abutting edges (open square butt
plates w i t h square edges are butted together w i t h joint). The gap should be about half the plate thick
out a gap (a close square butt joint) and one run is nesses. Above 6 m m thickness the gap becomes
deposited on each side, the centre of the joint will be impracticable and instead the edges are bevelled.
unfused in thicknesses greater than about 3 m m . The vee so formed is then filled by depositing a
number of runs. The shape and dimensions of t h e
prepared edges will depend on a number of factors
w h i c h are discussed in Chapter 5.
10 Welded Joint Design
Inaccessible Accessible
l/i N.
1st 2nd
\
\
\
\
\
-A. \ V
I — » — 1
CHAPTER 2
Chemical C o m p o s i t i o n
C Mn Si Ni Cr
%wt 0.16 1.6 0.3 0.2 0.1
V Cu S P
%wt 0.05 0.1 0.003 0.005
1. Limit the amount of Ensure cleanliness of steel before Requires clean, dry conditions, elec
hydrogen welding and use low hydrogen pro trode drying ovens.
cesses, e.g. M I G / M A G low hydrogen M I G / M A G may not be suitable or
M M A electrodes, basic submerged arc available.
fluxes. Submerged arc not necessarily
suitable.
3. Reduce cooling rate (a) employ high welding currents or Restricts positional welding. Low
slow welding speeds to increase rate speeds increase costs. Low productiv
of heat input to joint and so reduce ity. Unable to achieve high notch duc
cooling rate. tility in weld metal.
Requires time, energy consumption,
(b) Heat steel before welding, i.e.
manpower and equipment.
preheat.
(c) Optimise steel thickness in design.
Eliminate hydrogen Do not let job cool down after welding Delays completion of job, requires
after welding and keep hot for one or t w o hours energy, manpower and equipment.
after welding - post heat. This
releases some hydrogen.
Stee/s and their Weldability 15
Two other features influence the risk of hydrogen Through thickness direction ^ ^
cracking: the first is the rigidity of the joint w h i c h
can restrain the material from expanding and con
tracting, and increases local stresses; the second is
the fit up in fillet welded joints, there being a
tendency for poor fits to encourage hydrogen
cracking.
Cold
Hot
Cold
Yield stress
20 Welded Joint Design
to the stresses set up by welding, makes the A box girder is to be made from sections to be
distortion w o r s e - t h e diagram illustrates this. There welded together on site. Diaphragms are welded
Welds and J o i n t s
Butt joint
22 Welded Joint Design
Toes.
A w e l d face.
A root.
Penetration bead
Concave fillet
24 Welded Joint Design
necessarily strengthens a joint; a better term is full penetration butt welds, the general rule is:
overfill. design throat thickness, U ~ thickness of the thinner
part joined.
Size of Weld
Fillet Welds
Fillet weld sizes are calculated by reference to
allowable shear stress on the throat area, i.e. throat
area = design throat thickness x length of weld. The
size required is specified on drawings in terms of leg
length, /, or throat thickness, t.
These are the limits of a corner joint: There has been no really satisfactory all-purpose
definition of a butt weld. In some cases, for example
in plate members joined at their edges, the defini
tion is self-evident. In other cases - the corner joint
is one - the distinction between a fillet weld and butt
weld becomes blurred. It seems from common
usage that a butt weld is one which substantially
maintains the cross-section of the joined members
t > 10mm
joined. The fillet weld can then be made correctly
but its size should allow for the thickness of the
The A.W.S. Structural Welding Code, requires
packing. Mismatch between parts to be butt welded
the following.
can be more difficult to overcome. A skilled welder
For plug welds the hole diameter should not be less can deal w i t h a certain amount but beyond that a
than the plate thickness plus 8 m m and not greater defective weld is almost certain to result as can be
than 2.25 times the plate thickness. seen in the illustration on page 12.
28 Welded Joint Design
A t t e m p t t o make weld t o plate edge flange splice to permit a sound, continuous butt
weld to be made. Generally speaking, these cope
made to exactly this width. A n y attempt to do this holes are best left unfilled, but where they are to be
will result in the melting away of the corner and a watertight, plates can be fillet welded to each side.
subsequent loss of weld size. The weld leg length
should be at least 2 mm less than the available plate
width.
Weld Preparations
Included angle
11
Root gap
30 Welded Joint Design
several problems, notably by its propensity to melt Warning: Backing strips give a built-in crevice
away at the first weld run, so letting the weld metal susceptible to corrosion and give a lower fatigue life
drop through unless very low welding currents are than the simple butt weld.
used; in addition any wander in the line of the cut
edge will give a varying gap. These snags can be The single bevel preparation, as it is called,
avoided by leaving a root face. This provides a solid requires a volume of weld metal roughly propor-
nose which will fuse but at the same time hold up
the weld metal. It also acts as a buffer for cutting tional to the square of the plate thickness and its
tolerances by maintaining a constant root gap. lack of symmetry can give rise to distortion. Other
Variations in
Original plate profile
cutting line
Backing strip
The single U
The double U
Weld Preparations 31
i vertical
Single root
run
Over 6 45° 0 All Double root
20° 0 run
10 10
+2 Welded from
-0 one side with
steel backing
3 m m max. 2 m m max. strip
USf
R
Over 20 20° 5 5 5 All Welded from
t both sides
Detail at Y
Note: The angle of intersection 9 of t h e axes of t h e circular hollow sections should not be less t h a n 3 0 ° unless adequate efficiency of
the j u n c t i o n has been demonstrated.
36 Welded Joint Design
Note: The angle of intersection 0 o f the axes of the circular h o l l o w sections s h o u l d not be less t h a n 3 0 ° unless adequate efficiency of
the j u n c t i o n has been demonstrated.
Weld Preparations 37
1-2.5mm. / T
1-2.5mm
20°- 25°
L
H
H min. = T
2-3mm
1 i
2-3mm
p
Where d = D e< 60°
Details at Y Details at Z
38 Welded Joint Design
2mm 1
max. —L_
I 2mm max.
e <60° Where d
i—r
Where d = D
Details at Y
Note: The angle of intersection 0of t h e axes of t h e rectangular h o l l o w sections should not be less t h a n 3 0 ° , unless adequate efficiency
of t h e j u n c t i o n has been demonstrated.
CHAPTER 6
Weld Symbols on
Drawings
Detail Drawing
Fillet
\
Square butt
TT
Single-V butt
V
X
00
Double-V butt
Single-U butt
DOo
Double-U butt
42 Welded Joint Design
Single-bevel butt
7
Double-bevel butt
< 1
Single-J butt
p
Double-J butt
Weld Symbols on Drawings 43
Sealing run
o
Backing strip
V
Dressed flush
\/
- and a very useful symbol
2
Full penetration butt
weld by a welding procedure
to be agreed
44 Welded Joint Design
Static Strength of
Welded Joints
90 000
Stress = — — - = 1 5 0 N / m m 2
600
Leg length — 8 m m
Weld throat area = 5 0 x 2 x t = 1 0 0 t
w w
1 0 0 ^ x 1 1 5 = 90 000
fw = 7 . 8 2 m m The volume of weld metal is still 3 2 1 8 m m 3
100f,v
The maximum weld throat stress is then (in) acting in the throat parallel to the axis of the
weld and a shear stress (ti) acting in the throat
PL
transverse to the axis of the weld. Normal stress (an)
833.3f„
along the axis of the weld has no effect on the
The combined stress is then: strength of the weld.
A n equation then allows us to check the weld size
in terms of the allowable tensile stress in the steel.
%/_(l00t )
w
+
(s33-3rj _ =
^>/_(lOo) +
(aS*) . {Pa)
The equation is:
For P = 10 kN and L = 100 m m this becomes:
j3/[a/ + 3(T / 7
2
+ T / ) ] <Pa
10002^(0.0001 + 0.0144) = 1 ^ * For mild steel, e.g. B.S. 4 3 6 0 Grade 43. P = 0.7
'w t w For high yield steel, e.g. B.S. 4 3 6 0 Grade 50, p =
With an allowable stress of 115 N / m m 2 0.85
The example on this page can be checked using this
tw^lO mm equation.
52 Welded Joint Design
in is equal to the shear stress due to P and this is This is less than the allowable tensile stress of
155 N / m m and so w e can reduce the weld size in
2
PL 1 000 000
= 84-87 N / m m 2
throat thickness compared w i t h the previous size of
833-3f V w
2 8 3 3 3
x 10/2 10.0 m m .
The fillet weld design chart below gives a quick
So 0 - 7 / [ 8 4 - 8 7 + 3 (84-87 +
2 2
20 )]
2
method of calculating weld size from load per
= 121-26 N / m m 2
millimetre length of weld.
Weld size
Throat Leg
mm mm
1 2 3 4 5 6
kN/mm load
Two practical examples will clarify the way to W e could then call for 10 m m leg length fillet welds
calculate fillet welded joint strengths. w h i c h will be a saving in weld metal and so
represent a cost saving.
40 KN 2 . A connection between a hollow section and a
plate. What fillet weld size is required?
3 KNm
torque to the shaft. W e can take the weld length as A simple method is to assume that the shear is
n x 7 5 m m and this has a torque arm equal to the taken on the vertical (web) welds and the moment is
radius of the shaft. If the weld throat is t, since w e taken on the horizontal (flange) welds. For weld
have a t w i n fillet w e l d then w i t h an allowable throat metal matching grade 50 steel the allowable throat
stress of 115 N / m m 2
10 000
units on each side of the equation) and t = = 031 mm
32 000
16x 10 6
= 94.57 N / m m 2
54 Welded Joint Design
Area A mrrr Ay m m 3
Ay m m 2 4
/mm 4
12
= 166 667?
T u _ 416 667 0 0 0 0
Then Z = — — — = 8 3 3 3 f m m 3
3 x 10 6
8333f = •
160
3 * 10 6
due to bending
8333f
10x10 3
(360J + _ 242
361
then t = — = 2-25 mm
160
Fatigue Life
—
Yield .
1 lllj
stress !
Log stress range
:
...
Log cycles
The SN lines below are for the 50 and 97.5 per cent
survival levels.
2
Stress range, S, N/mm SN line for 50% survival SN line for 97-5% survival
2
Stress range, S, N/mm
10 ll I 111 I I I h m l I I I ln..l I I I I I
10 s
10 6
10 7
10 8
Endurance, N, cycles
Endurance, N, cycles
The table on pages 6 0 - 6 5 gives the classification Steel tubes are often used as structural members
of the more common types of welded joints to be in products as wide-ranging as building frames,
used w i t h the SN lines above. The table is based on cranes and offshore structures. The design of joints
the proposals by Dr T R Gurney of The Welding between tubular members requires special consid
Institute which have been used to formulate the eration because of the secondary stresses w h i c h
British Standard for Bridges (B.S. 5400) and the are caused by local bending of the tube walls.
United Kingdom Department of Energy Guidance Most of the joints in structures made from tubes
on the design of offshore installations. These docu consist of one or more brace members joined to a
ments should be consulted for fuller details in main, or chord, member:
respect of those types of structure, although this The secondary bending set up in the chord
design data in its basic form applies to any welded member means that the nominal stress in t h e
steel item. It will be seen that for offshore structures members cannot be used w i t h the SN curves above
a correction to the SN curves has to be made if the because the life would be seriously underesti
thickness of the material is more than 2 2 m m . This mated. Instead the greatest stress in the member at
is a result of research which shows that the thicker the toe of the weld has to be calculated; this is not an
the material, the lower the fatigue life. easy task because the stress varies around the joint.
Fatigue Life 59
Load applied in
the brace Local deformations in a tubular joint
Crown point
Having calculated the hot spot stress, the engi
neer is then in a position to undertake a fatigue
analysis. A special set of SN curves is used w i t h
these hot spot stresses. These curves are published
Chord in various codes, standard specifications and gui
dance. The example shown here is from Gurneyand
refers to circular section tubes. A s w i t h the SN
Nodal joint nomenclature
curves for plates and sections, there is an adjust
Increase in stress due
ment in the curves to be made for the effect of the
Nominal stress
thickness of the material on the fatigue life. The
to overall joint geometry-
curves shown here are for 9 7 . 5 per cent probability
Stress in brace
of survival w h i c h is the level commonly used in
Extrapolation of geometric most industries.
stress distribution to
weld toe This f o r m of presentation has been extended to
include rectangular section tubes and has been
Chord wall proposed for the Eurocode EC 3 Common unified
code of practice for steel structures. A detailed
discussion of the derivation of these curves has
Stress increase due
to weld geometry Brace hot spot stress= been prepared by Professor Wardenier of Delft
nominal stress in brace
X SCF Brace University.
Stress distribution in brace
Extrapolation of geometric
stress distribution to weld
toe
Stress in chord
Increase in stress
due to overall
joint geometry .
Nominal-
stress
Chord wall
10* 10°
Stress distribution in chord Endurance. cycles
Hot spot stresses in a nodal joint Diagram by permission of The Welding Institute
60 Welded Joint Design
No tack welds
^
Members w i t h attachments welded to
their surface.
Fatigue Life 63
Joints between tubular members can have very range S\ the constant amplitude fatigue life is N }
high stress concentrations w h i c h have to be cycles, then n cycles will use up a fraction n^/N of
} y
allowed for. Their fatigue life can be predicted from the fatigue life. In a stress history w i t h stress ranges
the foregoing classification provided that the joints S\* S2, S , etc. occurring n
3 n^ n , etc. times
u 3
So for one year's operation. These figures are for a steel plate w i t h a stiffener
attached w i t h a butt or fillet weld under a stress
n_ _ 6000 6000 1800 range, for w h i c h the life without treatment would be
N " 1 400000 +
10 80000
8 +
Brittle Fracture The ship below broke in t w o at sea off the coast of
Canada in 1979.
W h a t is seen in this picture is what is called a
'brittle fracture'. It is a fast-moving fracture which
gives the impression that the steel has lost its easy
plastic property and has changed into a glass-like
solid. This is not the case; w h a t has happened is that
Chapter 7 described h o w steel under tension a crack grew at high speed through the metal from
stretches elastically and then, w h e n the stress is some small irregularity. Welding can introduce
high enough, plastically. This sequence of proper irregularities such as slag inclusions, lack of fusion
ties is used to advantage in the rolling or forming of and cracks. W h e n the welded steel is stressed,
steel plates into various shapes, w h i c h would be these irregularities interrupt the stress field and,
impossible without this plastic property of steel. like all such features, have the effect of a notch
Consequently engineers have been surprised w h e n w h i c h locally increases the stress. In the case of the
steel structures and pressure vessels, items Kurdistan the origin of the fracture was a weld
designed to conventional working stresses and defect in the bilge keel.
Brittle fracture in a ship's hull - M. V. Kurdistan. Photograph by courtesy of The We/ding institute
68 Welded Joint Design
level. The temperatures over w h i c h this change The value of the stress intensity factor w h i c h
takes place are in the 'transition temperature causes brittle fracture is called the 'critical stress
range', a feature of these commonly-used structu intensity factor, Kc'. This concept is valid for low
ral steels. stresses, small cracks and brittle materials, but for
The required level of Charpy energy to confirm notch ductile steels an alternative concept has to be
resistance to brittle fracture at any service temper used. One such concept is based on the extent to
ature is also a function of the thickness of the steel. w h i c h a crack in the steel can be stretched open
As a result, product specifications will call for before brittle fracture or ductile failure occurs. The
Charpy tests at temperatures w h i c h will depend not measure of this opening, is made on a standard type
only on the service temperature but on the material of test specimen and is called 'crack tip opening
thickness. These requirements have been drawn up displacement', CTOD.
on the basis of research and experience and their The CTOD specimen looks like this and is loaded
origins are not explicitly stated in the specifications. in a conventional test machine:
For some specialised areas of engineering the
definition of notch toughness by the Charpy test is
not considered sufficient, and a more discriminating
approach is used, based on concepts called 'fracture
mechanics' w h i c h examine the stress in the metal
at the tips of cracks. In conventional engineering
calculations the stresses in the item under load are
derived, but the stress concentrating effect of such
details as bolt holes is often neglected, although
their cross sectional area may be deducted from the Load • 1 T h i s
sP
a v
millimetres 372
70 Welded Joint Design
The values of K or CTOD at which fracture occurs (a) Is the item under tensile stress greater than 25
can be used, w i t h a little mathematics, to define the per cent yield?
fracture toughness of the material which is required (b) Is it thicker than 5 mm?
to resist brittle fracture from a certain size of crack
(c) Is it to operate at temperatures below 5° C?
under a certain stress or strain level. Conversely
they can be used to calculate whether a certain (cf) Does the fabrication involve cold bending or
defect is likely to start a brittle fracture in a steel of forming?
known fracture toughness. This is the basis of the (e) Would a sudden fracture have serious
'fitness for purpose' approaches which are used to
consequences?
set quality standards.
If the answer to all of these questions is no then
attention to notch toughness is not essential.
If the answer to any of the questions is yes then
consideration should be given to controlling the risk
^ Fracture of brittle fracture. There are t w o principal means of
controlling this risk and a number of subsidiary
actions w h i c h can be taken. The primary approach
is to use steel and weld metals w h i c h have a notch
toughness sufficient to avoid brittle fracture under
the working conditions. This assumes that the
quality of fabrication is of a minimum standard
w h i c h must exclude any crack-like defects and the
larger volumetric defects. Depending on the sever
ity of the demands for controlling the risk of fracture,
Notch width Vmm the achievement of this level of quality can be
Record of CTOD test result accepted on the basis of a quality assurance
programme in which non-destructive testing
As w i t h the Charpy test, the CTOD varies w i t h methods appropriate to the risks to be accepted and
temperature and a transition curve can be plotted. the configuration of the welded joints may be
The CTOD specimen is of the same thickness as the included (see Chapter 12).
steel in the structure and so the test value can be Steel of less demanding notch toughness can be
applied directly to the actual job. employed if the second means of controlling the risk
Successful use of these concepts in material is employed in which the fabrication is post-weld
selection and in setting quality standards requires a - heat treated, known as 'stress relieving'. This redu
far deeper knowledge and understanding of the ces residual stresses and can improve the notch
subject than is given in this chapter. Nonetheless toughness of the steel around any weld defects.
the engineer needs to know when this question of There are service temperatures so low, such as
brittle fracture merits attention at the design stage. those at which liquefied gases are stored, that none
There is no direct answer to this except for those of the conventional structural steels can give a low
w h o are working to rigidly enforced specifications. risk of brittle fracture. For such purposes it is the
In other circumstances the engineer will find it practice to use aluminium alloys or some of the
helpful to make a checklist as follows for any welded stainless steels w h i c h do not have a change in
ferritic steel construction: notch toughness w i t h temperature.
Brittle Fracture 71
Welding Procedures
Address:
No
Contract
Consumable treatment
Pre-heat temp. *C
Method
Retention time hrs I
Gouge method
Weld preparation Gouge check method
Weld sequence & position
Inter pass cleaning
Max. weave width
Notes Electrode stick out mm
Welding progression
Weld finish
Signed Date
PWHT NDT
Prepared by (Contractor)
Heating Rate *C/h
Approved for test (Client)
Soaking Temp. •c
Time h
Cooling Rate •C/h Approved for construction
Withdrawal Temp. •c
To code/spec
For
1. Contractor
No Description Date By Chk 1 2 3 2. Client
Approved 3. Certifying authority
Revisions
Welding Procedure
Contractor:
Contract
No
Welders Name N o Date Welded
Location
Supervisor's Name
Power Source
Welding Machine/Gun
Measured preheat temp. 'C
These sketches required only if different from Weld Proc. Spec.
Max. measured interpass temp. *C
Weld preparation Weld sequence
Visual
Magnetic particle
Radiography
Ultrasonic
Cross weld tensile
Weld metal tensile
Root bend
Face bend
Side bend
Nick break
Macro
Hardness
Charpy
COD
Note 1 All NDT and mech anical test reports must be attached to this form Signed Date
and must carry w itnesses' signature and stamps. Prepared by
2 Names of witnesses in this part are for reference only. Checked by
CHAPTER 11
Quality Management
bureacracy which has no effect on product quality. (e) References: sources of information
Although the basic principles of quality assu (f) Responsibilities: who is responsible for execut
rance remain the same, their application to any ing the procedure.
particular industry or even any company w i t h i n an (g) Narrative: the actual procedure.
industry must suit the way in which that industry or The written procedure should say clearly and
company operates. A n attempt to force an unsuita concisely what is to be done, but stop short of those
ble system onto an organisation will result in activities which are part of a person's ordinary skills
resistance from personnel, possible damage to and tasks. If these are to appear anywhere they
quality and, without doubt, unjustifiable* costs. A should be in the company's training manuals.
properly designed system will enhance product
quality, motivate personnel, reduce costs and in the
end satisfy customers to the benefit of all.
Quality Assurance in Welded Fabrication
Welder Tests
The methods of inspection used for welded con Each inspection or non-destructive testing activity
struction are described in Chapter 12, but how is recorded in a report w h i c h should be listed in an
inspection fits into the overall pattern of quality inspection record and, depending on the practice of
assurance is explained here. the type of industry, such a record may or may not
Inspection activities, like other tasks, need to be be passed to the customer. In products for which
performed by personnel w i t h training and expe statutory approval is required, the inspection
rience. Inspection of welded joints tends to be records will form part of a package which has to be
broken down into visual inspection and non submitted to the authority before permission to use
destructive testing or examination. Visual examina the product is given. This applies to such products
tion requires an inspector w i t h a sound knowledge as offshore platforms, aircraft, and nuclear power
of welding and the associated activities. The inspec stations.
tor's task before welding will entail:
(a) Checking of material type.
(b) Edge preparations and joint fit-up.
(c) Pre-heat temperatures. Standard Specifications f o r Quality Assurance
(d) Other quantities which appear on the WPS.
On completion of the joint, the inspector will The British Standard most commonly used as the
examine the surface profile of the weld and the basis of quality systems in the United Kingdom is
compliance of the component w i t h the drawing in B.S. 5 7 5 0 'Quality sytems'. This lists the following
respect of dimensions. Depending on the particular requirement of a quality system:
organisation, the welding inspector may direct and (a) Quality system.
receive the results of any non-destructive testing. In (b) Organisation.
some situations the NDT operator will say whether (c) Review of the quality system.
the weld is acceptable or not, an activity known as (d) Planning.
'sentencing' in some quarters. In other situations (e) Work instructions.
the judgement on the acceptibility of the weld may if) Records.
be made by the welding inspector or other {g) Corrective action.
personnel. (h) Design control.
Non-destructive testing in whatever form must (i) Documentation and change control.
be undertaken by personnel w i t h the appropriate (/) Control of inspection, measuring and test
training and experience. Evidence of such training equipment.
and experience is given by qualifications awarded {k) Control of purchased material and services.
by a number of bodies on the basis of examinations {!) Manufacturing control.
requiring demonstration of both practical ability in, (m) Purchaser-supplied material.
and theoretical knowledge of, the particular tech (n) Completed item inspection and test.
nique. These bodies include CSWIP, the Certifica (o) Sampling procedures.
tion Scheme for Weldment Inspection Personnel, (p) Control of non-conforming material.
administered from the United Kingdom, and ASNT, (q) Indication of inspection status.
the American Society for Non-Destructive Exami (r) Protection and preservation of product quality.
nation, administered from the USA. Other schemes is) Training.
exist for specific industries but CSWIP and ASNT Another useful standard is the B.S. 4778 'Glos
are the most widely accepted. Although the techni sary of terms used in quality assurance'. For the
cal requirements are not substantially dissimilar, nuclear power industry there is a specialised stand
CSWIP and ASNT differ in that examinations for ard, B.S. 5882 'Specification for a total quality
ASNT can betaken in the inspector's o w n organisa assurance system for nuclear installations'. There
tion under ASNT qualified supervision from w i t h i n are standards published in other countries which
the organisation, whereas CSWIP conducts its o w n cover similar ground.
independent examinations at various locations
around the world.
CHAPTER 12
Chapter 1
P. T. Houldcroft
Welding Process Technology
Cambridge University Press, 1977
A. A. Smith
C0 Welding of Steel 3rd Edition
2
Chapter 2
R. W . K. Honeycombe
Steels - Microstructure and Properties
Edward Arnold, 1981
G. E. Linnert
Welding Metallurgy 3rd Edition Vols 1 and 2
American Welding Society, 1 9 6 5 / 6 7
K. G. Richards
We/dabi/ity of Steel
The Welding Institute, 1967
Anon.
Lamellar Tearing in Welded Steel Fabrication
The Welding Institute, 1972
F. R. Coe
We/ding Steels without Hydrogen Cracking
The Welding Institute, 1973
Chapter 3
Anon.
Control of Distortion in Welded Fabrications
The Welding Institute, 1968
F. M. Burdekin
Heat Treatment of Welded Structures 2nd Edition
The Welding Institute, 1969
86 Welded Joint Design
Chapter 8
T. R. Gurney
Fatigue of Welded Structures 2nd Edition
Cambridge University Press, 1979
J. Wardenier
Hollow Section Joints
Delft University Press, 1982
Chapter 9
G. M. Boyd (Ed)
Brittle Fracture in Steel Structures
Butterworths, 1970
access 10
aluminium 4, 71
arc 1, 5
argon 5
carbon 11
carbon equivalent 13
Charpy test 68
C0 5
2
cooling rate 12
corner joint 22, 26, 28
corrosion 19, 30
cruciform joint 22
CTOD 69
cumulative damage 65
electrode 1, 2, 5
face 23
fatigue 55
fatigue crack 55
fatigue life 56
filletweld21,22, 23
fit 27
fitness for purpose 70
flux 2
fracture toughness 70
hardenability 12
88 Index
hardness 12 radiography 82
heat affected zone 12, 23 reinforcement 24
hot spot stress 59 residual stresses 19
hydrogen cracking 13 restraint 14, 18
root 23
identification 79 root face 30
inclusions 14 root gap 29
inspection 80, 81
intermittent welds 48 sidewall 12, 31
size 24, 25, 47
J preparation 31 slot lap joint 27
slot weld 27
lamellar tearing 14 SN curve 56
lap joint 22, 26 splice 28
leg 25, 47 stainless steel 71
static strength 49
machining 19, 31 steel composition 11
MAG 5, 35 stick electrode 1, 31
magnetic particle 76 stiffeners 28
manganese 11 stress, allowable 50
manual metal arc 1 stress, shear 50
MIG 5, 31 stress concentration 57, 68
moment 53 stress intensity 69
mousehole 28 stress range 56
submerged arc 2
non destructive testing 80, 81 survival 57
notch toughness 67 symbols 39