173 

 
Chapter 8
Welded Connections

Fig.8.1 Welded plate girder with stiffeners

Nowadays, welding is used to connect steel plates to construct members such as plate
girders and box sections, as well as to connect structural members together. The welding
processes usually employed in the steel construction industry are

1) Shielded metal arc welding (SMAW)

2) Gas metal arc welding (GMAW)
3) Flux cored arc welding (FCAW)
4) Submerged arc welding (SAW)

To fuse two pieces of steel requires heat and/or pressure to bring the metallic atoms into
close proximity with one another to form metallic bonds, and protection of the metal atoms from
combining with oxygen atoms in the air. Arc welding processes use only heat from electric arc to
melt the metals being joined, and use fluxes and filler metals to protect and deoxidize the weld
pool. Both the base metals and the filler metals are melt in the welding process, as differentiated
from brazing and soldering, where only the filler metals are melt. The typical welding by SMAW is
shown in Fig.8.1.

Fig.8.2 Typical SMAW welding circuit

 
 174 
 

Fig.8.3 SMAW electrodes

Electrodes for SMAW (Fig.8.3) consist of a solid-core wire, the filler metal, surrounded by a
coating called flux, which decomposes as the filler metal is consumed. The purpose of the flux is
to shield the molten weld metal from the atmosphere. Once the decomposed flux contacts the
weld pool, they float to the surface and shield the pool, keeping oxygen and nitrogen (causing
embrittlement) from contaminating the weld deposit. After welding, the decomposed flux forms a
slag coating the covers the final weld (Fig.8.4). The slag is removed after welding, revealing the
shiny weld beads underneath.

Fig.8.4 SMAW process

As opposed to SMAW, which is also called

"stick" or manual welding and are used in both
shop and field welding, electrodes for the other
welding processes are supplied in coils or reels
(Fig.8.5), facilitating automatic or semi-automatic
shop welding.

Fig.8.5 FCAW electrodes

 
 175 
 
GMAW uses inert gases, such
as argon and helium, or CO2 to
shield the molten metals (Fig.8.6
top). FCAW uses tubular electrodes
that contain flux powders inside and
might also use gas shield (Fig.8.6
middle), making some FCAW and
GMAW equipment similar enough to
be used interchangeably. SAW
(Fig.8.6 bottom) uses granular flux to
shield the moten metal. It is called
"submerged" because the arc is
completely covered by the flux and
almost invisible. Since the granular
flux must stay in place to shield the
weld pool, SAW is restricted to the
flat and horizontal welding positions.

Fig.8.6
GMAW process (top)
FCAW process (middle)
SAW process (bottom)

 
 176 
 
Fig.8.7 shows the four basic welding positions, which are differentiated according to the
position of the electrode relative to the joint and can affect welding economy and quality.

1) Flat with the face of the weld nearly horizontal. The electrode is nearly vertical, and welding is
performed from above the joint.

2) Horizontal with the axis of the weld horizontal. For groove welds, the face of the weld is nearly
vertical. For fillet welds, the face of the weld usually is about 45° relative to horizontal and vertical
surfaces.

3) Vertical with the axis of the weld nearly vertical. (Welds are made upward.)

4) Overhead with the face of the weld nearly horizontal. The electrode is nearly vertical, and
welding is performed from below the joint.

Where possible, welds should be made in

the flat position. Weld metal can be deposited
faster and more easily, and generally the best and
most economical welds are obtained. In a shop,
the work is usually positioned to allow flat or
Flat Vertical horizontal welding. In the field, vertical and
overhead welding sometimes may be necessary.
The best assurance of good welds in these
positions is use of proper electrodes by
experienced welders.

Fig.8.7 Welding positions

Horizontal Overhead

The strength of a weld depends on the weld

metal used, i.e. the welding electrodes. For the
commonly used steel grades, only the US's AWS
E70XX electrodes or just E70 is considered here.
The E70 electrode has an ultimate tensile strength
Fuw  70 ksi  483 MPa and is used with steel having
Fy  60 ksi  414 MPa . The '1' in the designation
E7018 of an SMAW electrode indicates that the
electrode can be used in all positons.

 
 177 
 
8.1 Types of Joint

Pieces of steel are joined

in one of the five configurations
shown in Fig.8.8. Butt, T-, and
lap joints are common in steel
construction.

Fig.8.8 Joint types

8.2 Weld Types

Welds may be grouped into 3 categories: groove welds, fillet welds, and plug/slot welds,
Fig. 8.9. Groove and fillet welds are most widely used in steel construction. Plug/slot welds are
sometimes used when more weld is needed than length of edge is available.

Fig.8.9 Weld types

8.3 Groove Welds

Groove welds are the welds deposited in a gap, or groove, between two parts to be
connected. They are mainly used for butt, tee, and corner joints. In most cases, one or both of
the connected parts will have beveled edges, called prepared edges, as shown in Fig.8.10. A
variety of edge preparations are possible, Fig.8.11, depending on the thicknesses of the parts,
accessibility for welding, available equipment, as well as the knowledge and experience of the
contractor. From the view point of weld penetration, groove welds are divided into complete joint
penetration (CJP) groove welds and partial joint penetration (PJP) groove welds, Fig. 8.10.
Terminology associated with groove welds is shown in Fig.8.12. The most important is the
"throat", which is theoretically the weakest plane in the weld and therefore controls the design of
many welds. For CJP shown in Fig.8.12, the throat is equal to the thickness of the plates being
joined. For statically loaded structures, CJP groove welds develop the full strength of the
connected parts, therefore do not require a design calculation. On the other hand, the strength of a
PJP groove weld depends on the amount of penetration; once determined, the design will follow
that of a fillet weld, described in the next section.

 
 178 
 

Fig.8.10 Types of groove weld

Groove Types Single Double

Square

Vee

Bevel

Fig.8.11 Different groove weld preparations

Fig.8.12 Weld terminology: Groove weld

 
 179 
 
8.4 Fillet Welds

Fillet welds are the welds placed in a corner formed by two parts in contact, Fig.8.13. In
design, only the straight dotted-line dimensions are used, and for equal-legged fillet welds applied
to surfaces that are oriented 90° apart (Fig.8.13, 8.14), the throat dimension is found by multiplying
the leg size by 0.707 (i.e., sin 45°). The size of a fillet weld is specified in terms of the leg size, but
the strength of the weld is theoretically controlled by the throat dimension. No matter what is the
direction of the force applied to the joint, the shear failure is assumed on the throat area (Fig.8.14).
The strength of the fillet weld of leg size 'a' with effective length 'Le' and stress on the failure plane
are given by:

Shear strength of weld: Fnw  0.60 Fuw (8-1)

Fnw
Allowable shear stress: Fvw  ,   2.0 (8-2)

or Fvw  0.30 Fuw (8-3)
If P is the force applied either perpendicular to or along the weld axis shown in Fig.8.14,
Shear stress on weld:
P
f vw   Fvw (8-4)
te Lw

where te  0.707 a

Fig.8.13 Weld terminology: Fillet weld

Fig.8.14 Assumed failure plane of a fillet weld

 
 180 
 
AISC places the following limitations on using a fillet weld.

8.4.1 Minimum Fillet Welds

To have enough fusion and avoid cracking due to rapid cooling of small welds and restraint
to weld shrinkage provided by the surrounding metal, the minimum size shall not be less that the
size shown in Table J2.4.

8.4.2 Maximum Fillet Welds along Edge

To avoid melting the upper corner of a thick plate, which results in having actual throat less
than being designed and difficult leg-size inspection by eyes (Fig.8.15a), the maximum fillet size 'a'
shall be 2 mm less than the plate thickness for plate thickness of 6 mm and thicker. For thinner
plates, the maximum size is the plate thickness.

a) Incorrect welding b) Correct welding

Fig.8.15 Welding along edge of plate 6 mm thick or thicker

 
 181 
 
8.4.3 Length of Fillet Welds

If L is the length of fillet welds, the effective length used in the computation Lw is
for L  4a, Lw  L

L
for L  4a, Lw 
4

8.4.4 Flat Bar Tension Members with only Longitudinal Welds

The length of weld cannot be less than the width as shown in Fig.8.16.

Fig.8.16 Limitation of longitudinal welds of a flat bar under tension

8.4.5 Minimum Lap Length

To avoid large joint rotation in lap joints, the minimum overlap L shall be as shown on
Fig.8.17.

L  Max[ Min[5t1 ,5t2 ], 25 mm]

L

t2

t1

Fig.8.17 Minimum lap length

 
 182 
 
8.4.6 Fillet weld terminations

Fillet weld terminations are permitted to be stopped short or extend to the ends or sides of
parts or be boxed except:

1) For lap joints where one part extends beyond the edge of another part that is in tension
(Fig.8.18), fillet welds shall terminate not less than the size of the weld from that
edge.

Fig.8.18 Weld termination on overlapped tension member

2) Fillet welds that occur on opposite sides of a common plane shall be interrupted
at the corner common to both welds (Fig.8.19).

Fig.8.19 Fillet weld on opposite sides of a common plane

 
 183 
 
8.4.7 Fillet Welds under In-Plane Shear

When a weld group is subjected to in-plane shear as shown in Fig.8.20, the welds must be
designed to resist the combined effect of direct shear due to P and torsional shear due to the
eccentricity (e) of the line of force from the center of gravity (C.G.) of the welds. The analysis can
be done in much the same way as bolted connections as shown in Fig.8.20. The welds are
assumed to resist an equal share of the direct shear P , and a share of the torsion Pe proportional
to the distance from the C.G. of the weld. The resultant vectorial sum of the stresses is the
required strength f vw of the welds.
Lwx

y fvt
1 fvs
2 P
e P
Lwy

= + Pꞏe
CG x
3
cgx 4
Direct shear Torsional shear
Fig.8.20 Fillet welds under in-plane shear

The shear stress due to direct shear, f vs , is given by Eq.8-4 as:

P
f vs  (8-5)
te Lw

Consider a point on the welds, the stress due to torsional

fvt fvtv shear Pe is given by

fvth f vt 
 Pe  d (8-6)
v Jw
or in terms of the resolved components:
d
f vth 
 Pe  v (8-7)
 Jw

C.G. h f vtv 
 Pe  h (8-8)
Jw

Fig.8.21 Stress due to

torsional shear

where J w  polar moment of inertia of the welds computed assuming that each weld segment is a
line segment, and neglecting the moment of inertia of the line segment about the axis coinciding
with the line.

 
 184 
 
8.4.8 Weld Eccentricity and Weld Balancing

AISC J1.7 requires the CG. of welds to coincide with the CG. of the axially loaded member
unless provision is made for the eccentricity, with the exception of the end connections of angles
under static loading. As shown in Fig.8.22a, if welds are placed symmetrically about the center
line of the angle, a small eccentricity always exist since the line of force through the centroid of the
angle is close to the unconnected leg. To reduce or even eliminate the eccentricity, the welds have
to be balanced about the CG of the angle, resulting in having longer weld on the side of the
unconnected leg in Fig.8.22b.

Fig.8.22 Weld balancing

8.4.9 Fillet Welds under Out-Of-Plane Force

When a fillet weld is subjected to out-of-plane force as shown in Fig.8.23, the weld is
subjected to the combined action of in-plane force and bending moment. For the weld in Fig.8.23,
the stress at the critical point is computed from the vector sum of the direct shear stress f vs from
the load P and the flexural tensile stress f tm due to bending moment Pe , as shown in Fig.8.24.

e P

Fillet weld Fillet weld

Fillet weld
both sides

Fig.8.23 Fillet weld under out-of-plane shear

 
 185 
  te
fvs= P/teLw) ftm= Pe/Sw
1

L
Pꞏe
P

te
Direct shear Bending moment

Fig.8.24 Stresses on fillet weld under P and Pe

8.5 Welding Symbols

Welds are specified on design drawings using standard symbols. Some of the more
frequently used symbols are shown in Fig.8.25. If the fillet welds in Fig.8.23 are specified using the
welding symbols, the drawing will be as shown in Fig.8.26.
Fillet Weld

6 6 mm

6 mm
Field Weld

Groove Weld

45 Weld-all-around

6
6

Fig.8.25 Frequently used welding symbols

e P

Fig.8.26 Fig.8.23 redrawn using welding symbols

 
 186 
 
Ex8.1 Fig.8.27 shows two SS400 brackets connected to a steel column using fillet welds with leg
size equal to 16 mm. The welding electrode to be used is E70. Check if the fillet welds can safely
transmit P  400 kN to the column.

Fig.6.9 Ex6.1

SS400, for t  16 mm, Fy  245 MPa , Fu  400 MPa , E  200, 000 Mpa
P  400 kN , Fvw  483 MPa 130 40

a  16 mm, te  0.707 a  11.3 mm P (kN)

16
Lwx Lwx  130 mm , Lwy  300 mm

y L2wx
2

300
cgx  2  30.2 mm
2 Lwx  Lwy
CG
Lwy

x 3 2
L  t L 16
lwx  2te Lwx  wy   e wy  9,162, 7200 mm 4 16
cgx  2  12

 t L3 L  
2

lwy  te Lwy cgx 2  2  e wx  te Lwx  wx  cgx    10, 798, 987 mm 4

 12  2  
J w  I wx  I wy  102, 426,187 mm 4
Lwx D
Direct shear Torsional shear
y e fvt
fvs
1
2
P
P
Lwy

CG x = + Pꞏe
3
4
cgx

From the above decomposition, the most critical points are the points no. 1, 4.
D  40 mm, e  Lwx  D  cgs  139.8 mm

Lwy fvs fvtv

h1  Lwx  cgx  99.8 mm, v1   150 mm
2
Lw  2 Lwx  Lwy  560 mm #1 +
P
fvth
f vs   63.1 MPa
te Lw

f vth 
 Pe  v1  81.9 MPa
Jw

f vtv 
 Pe  h1  54.5 MPa
Jw

 f vs  f vtv   f vth2  143.4 MPa  0.3 f uw  144.9 MPa OK !

2
f vw 

 
 187 
 
Ex8.2 Fig.8.28 shows a steel plate with length L  260 mm welded to a steel column using fillet
welds with leg size equal to 6 mm. The welding electrode to be used is E70. Check if the fillet
welds can safely transmit P  140 kN with e  60 mm to the column.

e P

6
Fig.8.28 Ex8.2 6
6
6

SS400, for t  16 mm, Fy  245 MPa , Fu  400 MPa , E  200, 000 Mpa

P  400 kN , Fvw  483 MPa

te
fvs= P/teLw) ftm= Pe/Sw
a  6 mm, te  0.707 a  4.2 mm 1

L
Pꞏe
P

te
Direct shear Bending moment

From the above figure, the most critical point is the point no. 1.
te L2
Lw  2 L  520 mm, S wx  2  95, 586 mm 3
6
P
f vs   63.5 MPa
te Lw

Pe
f tm   87.9 MPa
S wx

f vw  f vs2  f tm2  108.4 MPa  0.3 f uw  144.9 MPa OK !

 
 188 
 
Ex8.3 An SS400 angle section is under tension and has its end welded to an SS400 gusset plate
which is also welded to a steel support as shown in Fig.8.29. Welding is to be carried out using
E70 electrodes. Check if the member with the given connection can safely transmit P  400 kN to
the steel support.
P-16
L 10

125
P

75
e
L-150x100x12

75
125
13
13
200

Fig.8.29 Ex8.3

SS400, for t  16 mm, Fy  245 MPa , Fu  400 MPa , E  200, 000 Mpa

  P  400 kN , E 70 : Fuw  483 MPa

Angle properties: B  150 mm, t  12 mm, Ag  28.56 mm 2 , cx  4.88 cm, c y  2.41 cm

B
e  cx  26.2 mm
2
1. Welds at angle's end: a  10 mm, te  0.707 a  7.1 mm

Lwx  200 mm , Lwy  B  150 mm

Lwx
L2wx
y 2
cgx  2  72.7 mm
2 Lwx  Lwy
2
CG L  t L
3
Lwy

lwx  2te Lwx  wy   e wy  17,895, 938 mm 4

x  2  12

cgx  t L3 L  
2

lwy  te Lwy cgx 2  2  e wx  te Lwx  wx  cgx    17,139, 394 mm 4

 12  2  
J w  I wx  I wy  35, 035,331 mm 4
Lwx
Direct shear Torsional shear
y fvt
1 fvs
2
P
e P
Lwy

CG x = + Pꞏe
3
4
cgx
 
 189 
 
From the decomposition figure, the most critical points are the points no. 1.
Lwy
h1  Lwx  cgx  127.3 mm, v1   75 mm
2
fvtv
Lw  2 Lwx  Lwy  550 mm fvs
f vs 
P
 102.9 MPa
#1 +
te Lw fvth

f vth 
 Pe  v1  22.4 MPa
Jw

f vtv 
 Pe  h1  38.1 MPa
Jw

 f vs  f vth   f vtv2  131 MPa  0.3 f uw  144.9 MPa OK !

2
f vw 

2. Welds at plate's end: a  13 mm, te  0.707 a  9.2 mm, L  250 mm

te
fvs= P/teLw) ftm= Pe/Sw
1

P
L
Pꞏe

te
Direct shear Bending moment
From the above figure, the most critical points are the points no. 1.
te L2
Lw  2 L  500 mm, S wx  2  191, 479 mm 3
6
P
f vs   87 MPa
te Lw

Pe
f tm   54.7 MPa
S wx

f vw  f vs  f tm  141.8 MPa  0.3 f uw  144.9 MPa OK !

3. Angle's tensile strength:

Gross yielding:
R1  0.6 Fy Ag  419.8 kN  P  400 kN OK !

 
 190 
 
Tensile fracture: An  Ag  2856 mm 2

From Table D3.1 case 2, from the figure below and steel table, LC  Lwx  B  200 mm

cy
U  1  0.88
LC U  1  X / LC
LC

Centroid

X
LC

Therefore, Ae  UAn  2512 mm 2

R2  0.5 Fu Ae  502.4 kN  P  400 kN OK !

4. Plate strength: t p  16 mm, B p  L  250 mm

Block shear of gusset plate:

Agv  2 Lwx t p  6400 mm 2

Anv  Agv  6400 mm 2

Ant  t p B  2400 mm 2

From Fig.C-J4.2, U bs  1.0 .

R3  min  0.3 Fu Anv , 0.3 Fy Agv   0.5U bs Fu Ant  950.4 kN  P  400 kN OK !

Combined tensile and flexural strength:

1
+
125

P
P-16
L
e
Bp

+ +
125

-
The critical point is point no.1.
tp P/A Pꞏe/S
A = tpBp
0.6 Fy
R4   361 kN  P  400 kN OK ! S = tpBp2/6
1 6e

t p B p t p B p2

Therefore the member and given connections cannot support the given tension P  400 kN .