5

SIMPLE WELDED
CONNECTION
INTRODUCTION
5.1
When two structural members are jointed by means of welds, the con
nection is called a welded comnection. A few decades ago designers had
a feeling that welded connections were less fatigue resistant and that
a good-quality welded connection could not be made. These negative feelings had
a great impact on the use of welding in structures. But the progress made in weld
ing equipment and electrodes, the advancing art and science of designing for weld
ing, and the increasing trust and acceptance of welding have combined to make it a
powerful implement for the expanding construction industry. Further, the economics
inherent in welding help to offset increase in the prices of material and cost of labou.
In addition, the shortened production cycle made possible by welding, have helped
effect aquickening in the paceof new construction. Welding will become increas
ingly important as more people acquire agreater depth of knowledge and experience
that goes with it. Today most of the regulatory agencies and government depart
ments accept, in fact, prefer welded joints. There are a number of reasons for using a
welded design, but a few basic ones are:
1. Welded designs offer the opportunity to achieve amore efficient use of mnate
rials. Welding is the only process that produces a one piece construction.
2 The speed of fabrication and erection helps compress production schedules.
3. Welding saves weight and consequently cuts costs. Connecting steel plates
are reduced or eliminated since they often are not required.
4. No deductions are there for holes; thus the gross section is effective in carry
ing loads.
5, Welded joints are better for fatigue loads, impact loads and vibrations.
Further, a properly welded joint is stronger than the jointed material. Fused joints
create a rigid structure in contrast to the non-rigid structures made using other types
of joints. The compactness and greater rigidity of welded joints permits design
assumptions to be realized more accurately.
 Simple Welded Connection 193

Welding offers the best method for achieving a rigid connection, resulting in
redued bcam depth and weight. Thus, it noticeably fowers the overall height of a
huilding. The weightof the structure and consequently the static loading is consider
ably reduced.This saves column steel and reduces foundation requirements. Saving
in transportation, handling time and erection is proportional to the weight savings.
Welding permits architects and structural engineers complete freedom of design.
The usageof outstanding design advancements such as open-web expanded (castel
lated) beams, tapered beams, vierendeel trusses, cellular floor construction, tubular
column and trusses are a few examples of welded constructions.
TYPES
5.2 The basic types of welded joints can be classified depending upon the
type of weld, e.g. fillet weld, groove weld (or butt weld), plug weld, slot
weld, spot weld, etc., position, e.g. flat weld, horizontal weld, vertical
veld and overhead weld, etc., and type of joint, e.g., butt welded, lap welded, tee
welded and corner welded. These welds are shown in Fig. 5.1.

Horizontal
Tee fillet Flat butt weld
butt weld
weld
Corner
butt weld
Vertical
fillet weld Slot weld
Overhead
fillet weld

Flat fillet weld

Flat
fillet weld Plug weld

Fig. 5.1 Iypes and positions of welds

Groove and fillet welds are most frequently used. Groove welds, are provided
when the members to be jointed are lined up. Various types of groove welds are
shown in Fig. 5.2. Groove welds require edge preparation and are thuscostly. Single
V, U, J, etc., are cheaper to form, but require double the weld metal than double
grooved joints. The choice between single and double grooves is usually a question
of whether the higher cost of preparation is offset by saving in weld
metal.
butt welds are used for plates up to 8 mm thickness only. Fillet welds are Square
when two members tobe jointed are in different planes. This situation is provided
met within structures. Therefore, fillet welds are more common than frequentty
groove welds.
Various types of fillet welds are shown in Fig. 5.3.
Fillet welds are normally easier to make, require less material preparation, and
are easier to fit than groove welds. On the other hand, for a given amount of weld
lalerial, they are not as strong and they cause greater concentration of stress. In
 84 Linu State Design of Steel Structures

Square Single Double Single Double

vee vee U

Single Double Single Double Single

tee tee bevel bevel J

Fig 52 Dpes ef groove welds

lightly stressed structures where stiffness rather than strength controls design and
fatigue or brittle fracture is not a problem, fillet welds are entirely adequate and gen
erally more economical. Reliable fillet welds may be, and frequently are, designed
for severe stress and service conditions, but, as a rule, groove welds are better in
highly stressed structures where smooth flow of stress is necessary. If the butt joint
has the same characteristics as the parent metal, is finished smooth with it on both
sides (by proper grinding or the like), and has complete penetration with no unweld
zones, it approaches the condition of no joint at all and, for most common types of
parent metal, may have impact and fatigue resistance equal or superior to that of the
base metal.
Slot and plug welds are used to supplement the fillet welds, when required length
of fillet weld cannot be provided. Since, the penetration of these welds cannot be
ascertained and since these are difficult to inspect, slot and plug are avoided.

Surface not
in tension
Surtace in
tension

Convex Concave Mitre

Fig S.3 ypes of fillet weld

SYMBOLS
5.3
A knowledge of welding symbols is essential for a site
able to read the drawings. Symbols save a lot of space engineer
as
to be
notes can be omitted. descriptive
Table 5.Idepicts the symbols and the method of their
ings. The drawings need to indicate the side of welding,representation on the draw
size, contour and finish,
Spacing and whether it is field or shop weld. Some examples of symbols and dimen
sions of welds are shown in Tables 5.2 and 5.3, respectively.
 Simple Welded Connection 95

tohle S.J Description f Dpe f Weld and its Representation

Type of weld
Fillet Groove |Seam Spot Plug Field
weld

Square V Bevel U J Vwith Bevel with Weld

broad broad root with
root face face raised
edges
vl PY J
Shape of weld surface Symbol MethodContour
offinish
representation
Shape OS Unweld
ofweld length
Flat (usually finished flush) Size
Other side
&
Convex
V
ldentification line
Reference line Arrow
6 50 (100)
line
Concave
Arrow side 6 mm fillet weld convex
finish 50 mm weld length
and 100 mm no weld length Joint

Table 52 Examples ef Use gf Main and Supplementary Symbols for Welds

S. Designation Illustration Symbolization
|No. syimbol

1. Single-V grove
weld

3 Double V
groove weld

4
Concave fillet
weld
 196 Limit State Design of Steel Structures

Table S.3 Dimensions of Welds

S. Designation
No.
Definition
Inscription
the
Groove S: minimnum distance from V

weld surface ofthe part tothe

bottom of the penetration, sll
which cannot be greater than
the thickness of the thinner
part

Continuous r: heightof the largest isosceles

fllet weld
S
triangle that can be inscribed in t
JKAABA, the section s: side of the largest
isosceles triangle that can be
inscribed in thesecction

3. Intermitent I: length of the weld (without

illet weld end crates).
(e): distance betwcen adjacent
weld elements,
n: number of weld
clements sn(e)
(same as in 2)

4 Staggered (Same as in 3)
intermittent
fillet weld T
(Same as in 2) nxI T(e)
nx| (e)

Notes
1. The position of the arrow line with respect to the weld is generally not
significant.
2. The reference line should be a straight line preferably drawn parallel to the
bottom edge of the drawing.
3. The synbol is placed on the continuous line side of the reference line if the
weld is on the arrow side of the joint.
4. The symbol is placed on the broken line side if the weld is on the other side
of the joint.

WELDING PROCESS
5.4
Welding consists of joining two steel sections by establishing a metal
lurgical bond between them through the application of pressure and/or
through fusion. The most commonly used process is arc welding-
fusion process. The bond bctween the metals is produced by reducing the surfaces t0
 Simple Welded Connectlon 197

Electrode
Welding machine ac or dc Extruded
power sOurce and controls cooling
Gaseous shield
Electrode holder
Molten pool
Electrode Stag Arc stream
Arc Base metal

Ground cable
Electrode cable

Fig. 5.4 Metal arc welding

be joined to a molten state and then allowing the molten metal to solidify. Whenthe
molten metal solidifies, union is completed.
In the arc welding process (Fig. 5.4), the intense heat required to reduce the metal
to liquid state is produced by an electric arc. The arc produces atemperature of about
3600"C between the sectionto be welded and' the electrode. The tremendous heat
at the tip of the electrode melts filler metal and base metal, thus liquifying them in
acommon pool called a crater. As the arcas solidify, the metals are joined into one
solid homogeneous piece. By moving the electrode along the joint to be welded, the
surfaces to be jointed are welded together along their length. In all modern arc weld
ing processes, the arc is shielded for the following reasons:
1. To protect the molten metal from air, either with gas vapour or slag
2. Toadd alloying or fluxing ingredients
3. To control the melting of electrode for more effective use of the arc energy.
Note Generally, the electrode used for welding is stronger than the parent metal. For
example, the electrode used for welding structural steel Fe 410 is E 41. This electrode
gives aweld deposit of yield strength of 330 MPa and ultimate tensile strength of
410 to 51O MPa.

5.5 WELD DEFECTS

Good welding techniques, standard electrodes and proper joint prepara
tion are the basic tools to achieve a sound weld. However, defects are
inevitable and a knowledge of these is essential to minimize them.Some
of the common defects inthe welds are discussed below and are shown in Fig. 5.5.
Incomplete Fusion is the failure of the base metal toget completely fused with the
weld metal. It is caused by rapid welding and also because of foreign materials on
the surfaces to be welded.
Mcomplete Penetration is the failure of the weld metal to penetrate the conplete
depth of joint. It is normally foundwith single vee and bevel jointsand also because
of large size electrodes.

Porosity Occurs due to voids or gas pockets entrapped in the weld while coolirg. It
results in stress concentration and reduced ductility of the metal. Normally porosity
 S3 LimitState Design of Steel Structures

Lack of fusion
Lack of fusion

(a)

Incomplete
penetration

Root-face
Incomplete
penetration
(b)

Gas pocket
Slag inclusion

(c) (d)

-Undercut Undercut

(e)
Fig. 5.5 Weld defects
is not a problem because each void is spherical and not a notch. Even with a
slight
loss in the section because of the voids, their spherical shape may be
to allow a smooth flow of stress around the void without considered
any measurable loss in
strength. Mainly these are caused because of careless use of back-up plates, presence
of moisture in the electrodes, hydrogen in the steel and
excessive current.
Slag Inclusions are metallic oxides and other solid compounds which are
times found as elongated or globular inclusions. Being lighter than the molten some
nmate
rial these float and rise to the weld surface from where these are
removed after cool
ing of the weld. However, excessive rapid cooling of the weld may cause them to e
trapped inside the weld. These present a problem in vertical and overhead welding.
Cracks are divided as hot and cold. Hot cracks occur due to the presence of sulpnu:
carbon, silicon and hydrogen in the weld metal. Phosphorus and hydrogen trappe
in the hollow spaces of the metal structure give rise to the formation of cold craCKs.
Preheating of the metal to be welded eliminates the formation of crackS.
Undercutting is the local decrease of the thickness of the Parent metal at the weld
toe. This is caused by the use of excessive current or a very long arc. An undelte
may result in loss of gross section and acts like a notch.
 Simple Welded Connevtion 199

INSPECTION OF WELDS
5.6
A poor weld leads to collapse: therefore proper inspection of the
weld is necessary. Some of the methods for inspecting welds are as
follows:

Magnetic Prticle Mechod Iron fillings are spread over the weld and it is then
subjected to an eleetrie curent. The fillings form patterns which are interpreted to
locate surtace cracks,

Dye Penetration Method The depth of acrack can be estinated by this method.
Adye is applied over the weld surface and then the surplus is removed. Adye
shsorber is placed over the weld which o0zes the dye giving an idea of the depth of
the crack.
Ulerasonic Method In this method, ultrasonic sound waves are sent through the
weld. Defects like flaws, blow holes, ete., affect the time interval of sound transmis
sion identitying the defect.
Radiography X-raysor gamma rays are used to locate defecs. This method is
Ised in groove welds only. lt cannot be used in fillet welds because the parent mate
rial will alsoform partof the projected picture.
ASSUMPTIONS IN THE ANALYSIS OF WELDED
5.7 jOINTS
The following assumptions are made in the analysis of welded joints.
1. The welds connecting the various parts are homogeneous, isotropic
and elastic elements.
2. The parts connected by the weld are rigid and their deformations are there
fore, neglected.
3. Only stresses due to external loads are considered. Efects of residual
stresses, stress concentrations and shape of the welds are neglected.
DESIGN OF GROOVE WELDS
5.8
Agroove weld is shown in Fig. 5.6. Asquare groove weld is provided
for sections less than 8 mm. Above this a single U, Vee or double U,
Vee, etc., groove welds are provided. Groove weld is usually designed
for direct tension or compression but if shear is also there then due provision should
he made. Since agroove weld involves no abrupt change in section at the joint, it is
the most suitable form of the weld for transmitting alternating stresses.
5.8.1 Specifications Fusion zone
Reinforcement Reinforcement is the Reinforcement
extra weld metal which makes the throat
Throat
dimension at least 10% greater than the
thickness of the welded material. The
Teason for providing reinforcement is to fig S.6 Gnve weld
 200 Limit State Design of SteelStrucres
increase the eficiency of the joint. The provision of reinforcement ensures that
depth through the weld is at least equalto the thickness of the plate and in cae
thick plates the metal within the thickness of the plate is in the anncaled
However, any reinforeement on the weld is ignored in caleulating the strength tcondition.
also difficult for a wclder to make the weld flush with the parent metal; so extra me
isdeposited.
Reinforcement makes the groove\weld stronger for static loads. But in case of
repcated and vibrating loads stress concentration develops in the reinforcement, lead.
ing to early failure. Hence, under such type of loads it is undesirable and the uat
surface is made flush. Where a flush surface is required, the groove weld is fir
reinforced and then dressed flush. Subsequent removal of reinforcement is not con.
sidered as reducing the strength of the joint. In any case, the reinforcement shoutd
not exceed 3 mm.
Size The size of the groove weld used is specificd by the throat dimension. This is
alsocalled efective throatthickness. The groove welds may bc of partial penetration
or of complete penetration (Fig. 5.7), the former being not preferred. It is difficult to
achieve complete penetration for single-U, Vee, J or bevel welds. However, this can
be achieved by using back-up strips as shown in Fig. 5.7 (a). These back-up strips
may be left in place or removed after the welding process is over. Double-U, Vee, J or
bevelwelds are examples of complete penetration groove welds. In case of complete
penetration of the groove weld the effective throat thickness is taken as the thickness
of the thinner member jointed. In case the full penetration of groove weld cannot be
achieved, an effective throat thickness of 7/8th of the thickness of the thinner mem
ber should be used. But for calculating the strength, the effective throat thickness is
assumed tobe 5/8th of thickness of the thinner member.

Backing bar
Tee Corner
(a) Complete penetration groove welds

(b) Partial penetration groove welds

Fig 5.7 Complete and partial penetration groove welds
Except for the squareedge type made in relatively thin metals, complete penetra
tion welds generally are welded from both the sides. Incomplete penetration but
welds, usually are of the single Vor single bevel type. In fact, incomplete penetration
is the failure of the base metal and the weld metal (to fuse at the root (Fig. 5.5 (b)).
This defect may be due to faulty design of the groove such as excessive root-1ace
dimension, root gap or groove angle, or it may b due to faulty technique such as
the use of excessively large size clectrodes, excessive speed or insufficient current.
Incomplete penetration isparlicularly undesirable sinceit causes stress concentration
under load and may cause shrinkage cracks.
 Simple Welded Connection 201
Ffective Area The effective area of the butt weld is the product of effective
thickness and the effective length of the butt weld, Effective length is the throat
the weld for which the required size of the weld is length of
done.
5.8.2 Design Strength
The design strength of the groove weld in
yield.
tension or compression is governed by

()
where f =smaller of yield stress of the weld (f) and
L = effective length of the weld in
the parent metal (f,) in MPa
mm
L.=effective throat thickness of the weld in
mm
Ymy = partial safety factor
= 1.25 for shop welding
=1.5 for site welding
The design strength of the butt weld in
shear is also governed by yield,
Vdw
Ymw (2)
where f =smaller of shear stress of weld
(fN3) and the parent metal (f,/N3)
Ymy = partial safety factor as defined above
f = yield stress of weld (MPa)
5.8.3 Design Procedure
1. Incase of complete penetration of the
not required as the weld strength at thegroove weld, design calculations are
joint is equal to the
member connected. strength
of the
2. In case of incomplete penetration of the
butt weld, the
ness is computed and the required effective length is effective throat thick
the strength equal to the strength of the determined to furnish
members connected.
3. The design strength of groove welds is
appropriate.
calculated by Eq. (1) or/ by Eq. (2) as

59 DESIGN OF FILLET WELDS

The fillet weld is done for members
which overlap each other. For
Jsuch joints the critical stress is shear stress. They are also subjected to
direct stresses but these are not of much
concave shaped fillet welds are provided dependinginmportance.
upon
Usually, convex
the situation. To all
external appearances, the concave fillet weld seems to be larger
weld. However, a check of the cross than the convex
penetration section may showthe concave weld to have less
and asmaller throat (Fig. 5.3) than first thought; therefore the convex
illet weld may actually be stronger eventhough it may have less deposited metal.
ZU2 Limit State Design of Steel Structures

Originally a concave fillet weld was favoured because it seemed to offer :a

path for the flow of stress. But expericnce has shown that single pass fillet
this shape have a greater tendency to crack upon cooling, which outweighs thewelds of
smoother
of improved stress distribution. cffect
When a concave fillet weld cools and shrinks, its outer face is stressed in
(Fig. 5.3), whereas a convex fillet weld on cooling shrinks and stresses the outer f. tension
in compression. Therefore in concave fillet welds cracks are developed on
However, when concave welds are desired they are made in two or more passes the
first slightly convex, and the other passes are buil-up to form a concave fillet weld.
cocoling,
Concave fillet welds are most suitable under alternating stresses.
5.9.1 Specifications
Size(s) The size of fillet weld is specified as the minimum leg length of the weld
The leg length is the distance from the root to the toe of the fillet weld (Fig. 5,8 (a
It is measured by the largest right angle triangle which can be inscribed within the
weld (Fig. 5.8 (a). This definition would allow an unequal-legged fillet weld (Fig
5.8 (b). Another definition stipulates the largest inscribed isosceles right angle tri.
angle and would limit this to an equal-legged fillet weld (Fig. 5.8 (c)). Equal-legged
fillet welds are preferred since they are easy to make. Unequal-legged fillet welds
are sometimes used to get additional throat area, hence strength, when the vertical
leg of the weld cannot be increased. An example of this is the attachment of a chan
nel (shear connector) over a beam flange, shown in Fig. 5.9. Here the vertical leg of
the fillet weld must be held to the thickness at the outer edge of the channelflange.
Additional strength must be obtained by increasing the horizontal leg of the fillet.
The weld size for unequal-legged fillet weldis specified by both leg
lengths.
Fusion Zone Increased effective Increased effective
Throat throat = 1.262t throat = 1.262t
Weld face Added weld
Added weld
()
metal metal
Toe

Leg - 20 k1.2620
(a) (b) Throat increased by 26% (c) Throat increased by 26%
Area increased by 100% Area increased by 59%
Fig. 58 Leg length of fllet weld
(a) Maximum Size of Weld The maximum size of a
fillet weld is obtained by
subtracting 1.5 mm from the thickness of the thinner member to be jointed. This
Specification limits the size of the fillet weld so that total strength may be developed
without overstressing the adjacent metal. In case of welds applied to the round toe of
steel sections the maximum size of the weld should not exceed 3 of the
thickness o!
the section at the toe.
(b) Minimum Size of Weld A small weld placed on a thick member is undesir
able. The heat generated in depositing the small size weld may not be appreciably
enough to expand the base metal.Consequently, as the weld cools, it contracts and iS
 Simple Welded Connection 203

Channel prevented from doing so by the stable base

shear metal. This results in initial stresses in the
attachment
weld metal. Also the welding process will
not heat the heavy plate much beyond the
axb immediate vicinity of the weld. The mass
Beam flange Unequal-legged
fillet weld
of the thick plates will cool the weld to0
rapidly and tendto make it brittle. To help
Fig 59 Welding of channel shear control this situation, IS: 800 specifies the
Connector minimum size of a fillet weld as given in
Table 5.4. These are the minimum sizes of
the first run, to avoid the risk of cracking in the absence of
size of the weldshould not exceed the thickness of the preheating. The minimum
weld size should not be less than 3 mm. thinner part jointed. Also, the

Table 54 Minimum Size of Weld

Thickness of thicker member
Minimum size (mm)
over (mn) up to and including (mn)
10 3
10 20 5
20 32 6
32 50 8 first run
10

Usually, a weld size closer to the minimum size is selected for the
reasons stated
as follows:
1. Large size welds require more than one run of welding
which means that
after the first run of welding chipping and cleaning will be required for
proper bond of successive weld runs. This increases the cost.
2. A smaller size weld will be cheaper than a large one for the
same strength
considering the volume of welding. For example, a 300 mm, 5 mm size fillet
weld will have the same strength (198.84 kN) as a 150 mm long and 10 mm
size fillet weld. However the volume of weld metal for a 10 mm weld will
150 = 7500 mmwhich is about twice that of 5mm size weld

xsx 300 =3750 mm:

Eifective Thro¡t Thickness is the shortest distance from the root of the fillet weld
to the face of the diagrammatic weld (line joining the toes) as shown in Fig. 5.10
la). The effective throat thickness should not be less than 3 mm. Also, it should not
CKCeed 0.7tor 1.0r under special circumstances, wheret is the thickness of thinner
plate of elements being welded.
Effective throat thickness = Kx size of weld =KS
 204 Limt State Design of Steel Structures

Throat KS
=0.707S Failure
plane
S
45
Root
K-S+t

(a) Throat of weld (b) Effective length of weld

Fig. S.10 Fillet weld dimensions

where S is the size of weld in mm and Kis a constant. The value of Kdepends upon
the angle between the fusion faces and is given in Table 5.5.
Table 5.5 Values of Constant Kfor Different Angles between Fusion Faces
Angle berween fusion faces 60°-90° 91°-100° 101°-106° 107°-113° 114°-1200
K 0.70 0.65 0.60 0.55 0.50

Note: Fillet weld is not recommended if the angle between fusion faces less than 60° or more than
120°

Efective Length is the length of the fillet weld for which the specified size and
throat thickness of weld exist (Fig. 5.10 (b)). It is taken equal to its overall length
minus twice the weld size. The deduction is made to allow for craters to be formed
at the ends of the welded length. End returns as End return
shown in Fig. 5.11are made equal to twice the size
of the weld to relieve the weld length from high
stress concentrations at their ends. This is particu 2s
larly important on the tension end of the parts car
rying bending loads. The actual -length of the weld
should not be less than four times the size of the
weld
Fig. S.11 End returns
Mote Inspite of all the precautions exercised by
the welder, there is always slight tapering off at the start and end of weld length.
Therefore, a minimum length of four times the size of weld is specified.
If the welds are parallel with the lines of stress, as in the case of
longitudinal fillet
welds, and are placed at the edges of the plates, there is a serious concentration of
stress at the edges of the plate. Therefore, the length of the longitudinal (side) filletS
on the flat should be not less than the width of the flat. The
unevenness of the stress
distribution is accentuated as the width of the plate increases. For these reasons, the
perpendicular distance between longitudinal fillet welds is limited to 16 times thick
ness of the thinner plate jointed. If the plate is wider than this limit, slot or plug welds
must be introduced. As a rule the provision of a slot tends to improve the distribution
of stressin the plates as shown in Fig. 5.12. Longitudinalfillet welds in slots have the
same strength as ordinary longitudinal fillet welds.
 Simple Welded Connection 205

Section A-A
Section B-B
A A
4 B

0
Id

(a)
(b)
Fig. 512 Distribution of stress in side-welded tension bar with and without slot
Slots if provided should be set back behind the
edges of the plate to ensure that the effective area ofbeginning of the welds on the
the deposition of slots does not lead to plate is not reduced and that
serious concentration of stress within the
plates.On first sight the joint as shown in Fig. 5.13 (b) might
strength to the one in Fig. 5.13 (a), but it fails at smaller appear to be superior in
(c) owing to the stress concentration in the plate in loads as shown in Fig.5.13
the re-entrant angles of notches.
If the length of the welded joint of a splice or
end
tension element exceeds 150 times the throat size ofconnections in a compression or
the weld, the reduction in weld
strength as per long joint is done. However, for flange to web
welds are loaded for the full length this connection since the
limitation does not apply. For long joints the
design capacity of weld is reduced by a factor,
0.2
B= 1.2. <0.1 (3)
150 1,
where = length of joint in the direction of force transfer
1, = throat size of the weld

(a) (b) (c)

hg 5.3 Unsatisfactory method of increasing length of fillet welds by means
f square slots
Effective Area The effective area of afillet weldlis equal to effective length of the
Weld multiplied by effective
throat thickness.
 20S Limit State Design of Steel Structures
be less
Overlap The overlap of plates to be fillet welded in a lap joint should not
whichever is
than four times the thickness of the thinner part joined or 40 mm,
more.

Transverse Spacing When the end of an element is connected only by parallel

longitudinal fillet welds, the length of the weld along either should not be less than
the transverse spacing between longitudinal welds.
Packing When a packing is welded between two members and is less than 6 mm
thick, it should be trimmed flush with the edges of the element. The size of the weld,
in such a case, along the edges is increased over the required size by an amount equal
tothe thickness of packing. Otherwise the packing is extended beyond the edges and
is fillet welded to the pieces between which it is fitted.
5.9.2 Design Strength
The design stress of afillet weld
(4)
where Si = nominal strength of fillet weld =f,/N3
The design strength of a fillet weld is based on its throat area and is given by
P =L (5)

or Pew = L KS

where L = effective length of the weld in mm

1,= throat thickness in mm
S= size of weld in mm
=smaller of ultimate strength of the weld and the parent material in
N/mm
P = design strength of weld in Newtons
Yonw = partial safety factor
=1.25 for shop welding
= 1.5for site welding
Mote The design strength in shear and tension for site welds made during erection
of structural members should be calculated by Eq. (5) but with a partial safety factor
Tmy of 1.5.

5.9.3 Design Procedure

Afillet weld may be subjected to direct, bending and shear stresses but shear controls
the design since the fillet weld always fails in shear.
1. The size S of the weld is assumed based on the thickness of the
members to
be jointed.
2. A fillet weld fails by shear at an angle of about 45° through the throat. The
strength of the fillet weld is calculated by Eq. (5).
 Simple Welded Connection 207
3. Pullor thrust to be transnitted by the
connections is estimated.
4. Effective length of the weld is calculated by dividing the
thrust by the strength of weld/mm. The length can either be factored pull or
adjusted as lon
gitudinal fillet welds (parallel to the load axis)or as transverse fillet welds
(perpendicular to the load axis) along with longitudinal fillet welds. It is a
common practice totreat both the welds as if they are stressed equally. If the
length of the weld exceeds 150 ,, the design capacity is reduced as described
in the specifications foreffective length.
5. If only a longitudinal fillet weld is made, a check is
provided to see that the
length of cach longitudinal fillet weld is more than the
tance between them. perpendicular dis
6. End returns of length twice the size of the weld are
provided at each end of
the longitudinal fillet weld.
Notes
1. It is assumed that the strength of the longitudinal and
transverse fillet welds,
as shown in Fig. 5.14, is same. Actually the strength of the
transverse fillet
weld is about 30% more than the longitudinal fillet weld, because a trans
verse fillet weld is stressed more uniformly for full length whereas a longitu
dinal fillet weld is stressed non-unifornmly due to varying deformations along
the weld length. Another reason for the greater strength of the transverse
fillet weld is given by tests which show that failure occurs at an angle other
than 45°, giving transverse fillet welds a larger effective throat area.
2. The main reasons for neglecting the greater strength of transverse fillet
welds are probably an interest in
simplifying design and the fact
that in most cases little would be
saved by differentiating between
D»»»»»)»»I))
transverse and longitudinal
welds. Many joints contain com
binations of longitudinal, trans
verse, and oblique welds. Any
Longitudinal fillet weld Transverse fillet weld attempt at athorough analysis
would become complicated and
Fig. 5.14 difficult to justify as a design
measure.

FILLET WELD APPLIED TO THE EDGE OF A

510 PLATE OR SECTION
For a fillet weld applied to the square edge of a part, the specified size
of the weld should generally be at least 1.5 mm less than the edge thick
ness in order to avoid washing down of the exposed arris (Fig. 5.15 (a)).
For the fillet weld appliedto the rounded toe of arolled section, the specified size
of the weld should generally not exceed 3/4 of the thickness of the section at the toe
208 Linit State Design of Steel Structures

Specified 1
1.5 mm size 4

(b)
(a) Toe of Rolled Section
Square Edge of Plate or Round
ig. S.5 Fillet Welds on

(b)Acceptable because (c) Not acceptable because of

(a) Desirable reduced throat thickness
of full throat thickness
Section
5.16 Full Size Fllet Weld Applied to the Edge of a Plate or
Fig.
fillet weld is such that the parent metal
(Fig. 5.15(b). Where the size specified for a
the outer cover or covers is allowed
willnot project beyond the weld, no melting of
reduce the
to occur to such an extent as to
205
Chamfer
throat thickness (Fig. 5.16).
When fillet welds are applied to the edges -t

of a plate or section in members subjected to

dynamic loading, the fillet weld should be of
full size, that is, with its leg length equal to h:b= 1:2 or flatter
the thickness of the plate or section, with the
limitations enumerated through Fig. 5.16.
The end fillet weld normal to the direc Force Force
tion of force should be of unequal size with
a throat thickness not less than 0.5t where t
is thickness of the part as shown in Fig. 5.17.
The difference in thickness of the welds in
such a case is negotiated at auniform slope. Fig. S.17 End,fllet Weld Normal to
Direction or Force

5.11 FILLET WELD FOR TRUSS MEMBERS

Truss members are composed of single-angle or
tions. The following points should be borne in minddouble-angle sec
the weld. while designing
1. The calculated weld length is
placed as longitudinal fillet welds either on the
two sides parallel to the axis of
shown in Fig. 5.18 (b)), i.e., the load (Fig.5. 18 (a), or on three sides a
A longitudinal fillet weld transverse welds along with longitudinalvwelds.
there will be possibility oflength should never be placed on one side only a
2. The centre of rotation.
gravity of the weld should coincide
section used as a truss member. If the with the centroid oI
will be placed member is
symnmetrical,
symmetrically but if the
channel) as is usually the case of a truss member is
the welO
unsymnmetrical
(ang
member, the lengths of longituane
 Simple Welded Connection 209

(a) (b)

Fig. S.18 Fillet Welds for Truss Member

fillet welds are kept different on the two sides, as shown in Fig. 5.18(a), to
achieve the above condition.

Let L, L, =length of longitudinal fillet welds on two sides

P, P, = factored design loads along lengths L, and Lz, respectively
P = factored load acting on the centroid of the section
Taking moment about the line passing through length L, (Fig. 5.18 (a),
Psh - Ph, =0
Ph, (6)
Or P, =
Similarly.
Ph, (7)
P, = h
lengths can be
Once the factored design force P, P, are known the fillet weld
designed as described in Section 5.9.
connecting the two side
The effectiveness of a fillet weld can be increased by
when the weld length to
welds by an end weld as shown in Fig. 5.18 (b). Further,
be provided cannot be accommodated over the
two parallel sides because of limited
reduces the size of gusset
Overlap, it may be the only choice. This arrangement also
plate and results in economy.
Let L, L = length of longitudinal fillet welds
h = length of end fillet weld
the section
P= factored load acting on the centroid of
fillet weld
P, = factored design force in the end
= hi, (from Eq. (4))
3 Ymw
Total length of weld required = L + Lyt+ h
laking moment about the line passing through L,
(8)
h
P,h +P37 - Ph, = 0
laking moment about the line passing through L:
20 Limit State Design of Steel Structures

P,h +P Ph, =0 (9)

and
Since P, from Eq. (5) is known: the above Eqs. (6) to (9) can be solved for P,
lengths /
P,. Once the factored design forces P, and P, are known the fillet weld
and L, can be designed.
WELDS
DESIGN OF INTERMITTENT FILLET
5.12
Intermittent fillet wvelds are provided to transfer calculated stress across
a joint when the strength required is less than that developed by a con
tinuous fillet weld of smallest practicalsize, e.g., in case of connections
of stiffeners to the web of plate girders. The fillet weld length required is computed
as a continuous fillet weld and a chain of intermittent fillet welds of total length equal
The types
to the computed length, with spacing as per IS specifications is provided. fillet
of intermittent fillet weld are as shown in Fig. 5.19. Staggered, intermittent
welds as shown in Fig. 5.19 (a) are structurally better than the chain pattern intermit
tent fillet welds shown in Fig. 5.19 (b).

(a) (b)
Fig. 5.19 Intermittent Fillet Weld

The question of whether and to what extent intermittent welds should be used
involves consideration of the following points:
1. Intermittent fillet welds are not economical unless the weld is of minimum
size. Asmaller size fillet weld of a longer length is usually more economical
for the same strength. This is because the strength of a fillet weld increases
directlywith size but the weight of the weld metal increases with the square
of the size.
2. If automatic welding is to be used the weld should be continuous.
3. If the structure is exposed, the use of continuous welds may be pretera
ble as they are conducive to greater ease of maintenance and longer
lite of
structure.
4. If severe dynamic loads, repetitive and
mittent welds must
alternating act on the structure, inter
not be used.
Note Intermittent butt welds may be used to resist only shear forces and in general
are not recommended.

5.12.1 Design Procedure

1. The size of the weld is
assumed and the total effective length of the tillel
weld required is computed.
2 Any intermittent fillet weld section
should have a minimum effective length
offour times the size of the weld wih aminimum of 40 mm,except for plate
girders.
 Simple Welded Connection 21l

3. The clear spacing between an intermittent fillet weld should not exceed 12r
for compression and 16r for tension and should in no case be more than 200
mm, where tis the thickness of thinner part jointed.
4. At the cnds, the longitudinal intermittent fillet weld should be of a length
not less than the width of the member or else transverse welds should alsO
be provided. If transverse welds are provided along with longitudinal inter
mittent fillet welds, the total weld length at thecends should not be less than
twice the width of the member.

PLUG AND SLOT WELDS

5.13
Plug and slot welds are used most often to tie two parts together and, in
particular toreduce the unsupported dimensions of cover plates in com
pression, which increases the critical stress. They may also be used for
shear transission. Their use is generally reserved for locations where it is impracti
cal to make afillet weld yet possible to provide a plug or slot. The critical section for
either a plug weld or a slot weld is the faying surface between the connected parts.
The unit shearing resistance on this section is essentially the same as that of a fil
let weld. Plug and slot welds should not be used to transmit tension that is, a force
normal to the faying surface. Tensile resistance depends largely upon the degree of
penetration of the weld, which is apt to be rather than uncertain in either a plug or
aslot weld. Some engineers distrust plug and slot welds because of the difficulty of
inspection. It is rather ease to make a plug or slot weld which appears excellent on
the surface yet contains voids at the critical section.
The following specifications should be adhered to while designing plug and slot
welds.
1. Width or diameter should be >3t and also 25 mm.
2. Corner radius in slotted hol: should be 1.5t and also 12 mm.
3. Clear distance between holes should be >2t and also 25 mm where t is the
thickness of plate having a hole or slot.
Rote A combination of plug weld and other types of welds is permissible and the
strength of the joint is the sum of the individual capacities of the welds.
STRESSES DUE TO INDIVIDUAL FORCES
514 When subjected to either compressive or tensile or shear force alone,
the stress in the weld is given by
P
S, or q = (10)

where
J,=calculated normal stress due to axial force in N/mm
4= shear stress in N/mm
I'=force transmitted (axial force Nor the shear force Q)
effective throat thickness of weld in mm
'= effective length of weld in mm
22 Limit State Design of Steel Structures

515 COMBINATION OF STRESSES

Fiilet Welds When subjected to acombination of normal and shear
stresses, the equivalent stress f, should satisfy the following
f. =V'+3q°s (l1)
V3 Ymw
where
bending moment
Ja= normal stress, compression ortension, due to axial force or
determined from Eq. (10)
determined from Eq. (10)
q= shear stress due to shear force or tension
not be done
For fillet welds check for the combination of stresses need
1. for side fillet welds joiningcover plates and flanges plates, and
2. for fillet welds where sum of normal and shear stresses does not exceed f
determined from Eq. (4).
done
Butt Welds For butt welds check for the combination of stresses need not be
if
1. butt welds are axially loaded, and
2. in single and double bevel welds the sum of normal and shear stresses does
not exceed the design normal stress, and the shear stress does not exceed 50
per cent of the design shear stress.
Combined Bearing, Bending and Shear Where bearing stress, fs, is combined
with bending (tensile or compressive) and shear stresses under the most unfavou
able conditions of loading, the equivalent stress f. is obtained from the following
formula:

f= +fr tffor + 3q (12)

where
f, = equivalent stress
f, = calculated stress due to bending in N/mm
Sor = calculated stress due to bearing in N/mm
q =shear stress in N/mm
The equivalent stress f,as calculated from Eq. (12) should not exceed the values
allowed for the parent metal.

FAILURE OF WELDS
5.16
Butt Welds When the butt weld is reinforced on both the sides of the
plate, the section through the weld is increased to such an extent thal
it is unlikely for failure to occur in the weld, and the fracture normally
)ccurs some distance away (Fig. 5.20 (a)). The reinforcement actsas a supporting rib
vhich inhibits necking in the immediate vicinity of the weld.
 simple Welded Connetlon 213

If the weld is ground flush with the

surface of the plate, the position of
the fracture depends upon the relative
(a) (b)
strength of the plate and weld metal. If
Fig 520 Failure of butt welds
thetensile strength or the yield point is
lower for the weld metal than for the
plate. failure takes place through the
centre of the weld (ig. 5.20(b)), But if the tensile strength and yield point of the
weld metal are higher, failure takes place in the plate away from the weld. Failure in
the weld junction is quite unusual.
Under conditions of free bending across areinforced butt weld, the
effect of the reintorcement inhibits failure at the joint. But since stress stiffening
occurs at the weld junction, due to change of the section, there is some concentration
for a crack to start at the weld junction itself with a flush butt weld, failuretendency
under
such a condition occurs in the middle of the weld, If there is a wide difference in the
tensile strength or yield point between the two metals, the failure may occur at the
junction.
End Fillet Weld The plane of the fracture in normal profile convex end fiület
weld(Fig. 5.21 (a)) is along the diagonal fromthe root of the fillet (Fig. 5.21(b).
When the fillet is subjected to shear in addition to the tensile stress, the positionof
the line of fracture departs from the diagonal according to the relative magnitude of
the two stresses (Fig. 5.21 (c)). If the fillet legsare unequal, afracture usually occurs
near the shorter leg (Fig. 5.21 (d)). If the tensile strength of the weld metal is consid
erably greater than that of the plate, the fillet may remain intact and be pulled right

(a)
Line of Line of
fracture fracture

(b) (c)

Line of Line of
,fracture fracture Line of
tracture

(d) (e)
Fig. 521 Position ffiracture in side fllet welds
 24 Limit State Design of Steel Structures
abruptly. afte.
out of the plate (Fig. 5.21(e). With all endfillet welds failure occurs
asmallamount of deformation.
shear stress along the
Side Fillet Weld In a side convex fillet weld subjected tocommences
weld, failure occurs downthe throat of the weld. The break at the toe f
fracture
the fillet at one or both the ends of the weld and as it progresses, the planeofofthe fllet
rotates (Fig. 5.22). The failure is gradual and considerable deformation
and usuallyalso of the plates takes place before the final fracture.

Fracture

Fg S22 Position of fracture in side fillet welds

517 DISTORTION OF WELDED PARTS

Fllet Weld Consider a fillet weld as shown in Fig. 5.23 (a). After the
weld is cooled, shrinkage is greater along the face of the weld where
large portion of fillet metal is deposited. If the vertical plate is free to
move (Fig. 5.23 (b),it is pulled over by the shrinkage of the weld without residual
stresses due to shrinkage. This is so because no resistance is offered to such distor
tion. Now a fillet weld on other Side of the vertical plate (Fig. 5.23 (c)). if
willtend to pull the plate back into the original position deposited.
stress in both the welds. This residual stress generated generating large residual
because of asymmetry of
welding sequence can be minimized by depositing the two welds
simultaneously.

(a) (b) (c)

Fig 523 Distortion and residual stresses of fillet welds
Short-Length Butt Weld Consider a butt weld
shown in Fig. 5.24. When molten connecting two plates Aand A, as
are heated rapidly to a very high metal is deposited part A and zone B of plate Ai
Zone C being heated less than the temperature allowing the steel to behave plasticaly.
The heatedportions tend to remaining two is assumed to behave elastically.
expand in
Cbeing less heated
tends to expand proportion
less and
tothe change in
temperature. Zols
Band part A. The distorted shape is shown in restrains some of expansion of zone
are restrained from expanding while in a Fig. 5.24 (b). Since part Aand zone B
plastic state, in cooling they tend to shrink
 Simple Welded Connection 215
A

B
A 4

(a) (b) (c)

Fig. 524 Distortion and residual stresses in
butt weld
oa length less than their original dimension. Zone C, which did not undergo plastic
deformation tends to shrink to itsoriginal dimensionand thus will restrain part Aand
20ne B from free shrinkage. When the plates are completely cooled, part A will be
shorter than its original length and will present a distorted shape shown in Fig. 5.24
(c). Since part A and zone B have been prevented from full shrinkage, they are sub
jected to tension, whereas zone C is subjected to compression and bending, result
ing in essentially compressive stresses at the boundary between zones B and C, and
tensile stresses at the free edge of zone C. If the two plates shown in Fig. 5.24 were
Symmetrical about the weld, the distortions and residual stresses would be reduced
considerably.
Long Butt Weld Consider a long butt weld as shown in Fig. 5.25. One part of the
plate zone D, is in a heated or expanding state similar to Fig. 5.24 (b) while the zone
Cis in a cooling or contracting state similar to
Fig. 5.24 (c). The relative areas of two zones,
final distortions, and the residual stresses
D
dependon speed of welding and on arrange
mentand sequence of welds designed to mini
mize distortions and residual stresses. There
area number of techniques to minimize resid
ual stresses and prevent distortions. Some of
which are: preheating the parts before weld
Fig 525 Hieating and cooling f a ing, hammering the weld to elongate it locally
long butt weld and to relieve the shrinkage forces (preening),
etc.

510 FILLET WELD VS BUTT WELD

Fillet welds are preferred in comparison to butt welds due to the fol
lowing reasons:
1. Afllet weld saves the operation of veeing and finishing the ends of
members.
2. In case of a butt weld, members are fabricated slightly long and cut exactly
to have a close fit inthe field. This process in uneconomical.
S. Butt welds have higher residual stresses.
25 Limit State Design ofSteel Structures Nature

isfor a
WELDED JOINTS VS BOLTED AND andhigh
5.19 RIVETED JOINTS boltsan
and bol.
1. Welded joints are economical. This is because splice plates strengt
reGuired ar
rivet materials are eliminated. Also, the gusset plateslength. Iabe
HSFGb

of a smaller size because of the reduced connection Availa

cost is also less as only one person is required to do the welding whereas at
bolts, h
and four for riveting.
least two persons are required for bolting continuity of the section at the
canbe
2. Welded structures are more rigid (due to the
bolted/riveted joints, cover
joint) as compared to bolted/riveted joints. In member during load
CoSt

plates,connecting angles, etc., deflect along with the

Connec

transfer and make the joint more flexible. maybe

3. Due to the fact that the strength of a welded joint is the
same as that of the be fea
otherwise is ascran
parent metals, even a smallest piece of the metal which be giv
can be used, bringing overall economy. fabrica
4. With welding it has become possible to connect tubular sections, which are Un
structurally very economical. small
5. Due to the fusion of two metal pieces jointed, a continuous structure is skillec
obtained, which gives a better architectural appearance than bolted/riveted for fat
joints. resisti
6. Alterations can be done with less expenses in case of welding as compared Rivets
tobolting/riveting. a sim
7. The process of welding is quicker in comparison to bolting/riveting. found
8. The process of welding is silent, whereas in the case of riveting a lot of noise nectic
isproduced.
9. In welding less safety precautions are required for the public in the vicinity,
whereas a hot rivet may toss and injure the persons working.
10. As splice plates, bolts/rivets, etc., are not used, the
welded structures are easier and less time details and drawings of Solh
11. The efficiency of welded joint is more consuming. Note
fact, a than that of a bolted/riveted joint. In
proper welded joint
12. Members to be jointed maymay have 100% efficiency.
cess, whereas there is no suchdistort due to the heat during the welding pro Exar
13. The possibility of a possibility in
brittle fracture is more bolted/riveted joints.
gro

compared to bolted/riveted joints. in the case of forc

14. The welded jonts be
inspection of
riveted joints can bewelded joints is difficult and expensive, whereas bolted
15. Amore skilled inspected simply by tapping the joint with ahammer.
person is required to make a
bolted/riveted joint. welded joìnt as compared to a A

5.20 SELECTION OF Sol

After FASTENERS
being conversant with
choose a
particular type basedvarious
on the
types of fasteners, a designer may
following considerations: