CHAPTER IV CONNECTION

1. INTRODUCTION
Connections or joints are used to transfer the forces supported by a structural member to other parts of the
structure or to the supports. They are also used to connect braces and other members which provide
restraints to the structural member.
The arrangement should be chosen to avoid excessive costs, since the design, detailing, manufacture, and
assembly of a joint is usually time consuming; in particular the joint type has a significant influence on
costs.
A joint is designed by first identifying the force transfers from the member through the components of the
joint to the other parts of the structure. Each component is then proportioned so that it has sufficient strength
to resist the force that it is required to transmit. Connections can be classified according to:
▪ The type of connecting medium used: bolted connections, welded connections, bolted-welded
connections riveted connections.
▪ The type of internal forces the connections are expected to transmit: shear (semi-rigid, simple)
connections, moment (rigid) connections.
▪ The type of structural elements that made up the connections: single-plate angle connections,
double web angle connections, top and seated angle connections, seated beam connections, etc.
▪ The type of members the connections are joining: beam-to-beam connections (beam splices and
hanger), column-to column connections (column splices).
Attention should be paid to:
➢ use of standardized details
➢ the clearances necessary for safe erection
➢ the clearances needed for tightening fasteners
➢ the need for access for welding
➢ the requirements of welding procedures
➢ the effects of angular and length tolerances on fit-up
➢ subsequent inspection and maintenance.
Members meeting at a joint should normally be arranged with their centroidal axes intersecting at a point.
Where there is eccentricity at intersections, the members and connections should be designed to
accommodate the resulting moments. For bolted angles, the setting out lines of the bolts may be adopted
instead of the centroidal axes.
Joint components
A. Bolts
Bolts may transfer loads by shear and bearing, by friction between plates clamped together or by
tension.
1. Design of bolts
Bearing bolts in shear (Page 89 table 8.4 ES EN1993-1-3-2015)
The resistance Fv of a bolt in shear depends on the shear strength of the bolt (of tensile strength fub) and
the area A of the bolt in a particular shear plane (either the gross area, or the tensile stress area through the
threads As, as appropriate). It can be expressed in the form

1
 CHAPTER IV CONNECTION

Eq6.1

where γM2 = 1.25 is the partial factor for the connector resistance (1.50 for Grade 4.6 bolts), n is the number
of shear planes, An and αvn are the appropriate values for the nth shear plane. It is common and conservative
to determine the area An at a shear plane by substituting the tensile stress area As for the shank area A of
the bolt.

Bolts in tension (Page 89 table 8.4 ES EN1993-1-3-2015)

The resistance of a bolt in tension (Figure 9.1c) depends on the tensile strength fub of the bolt and the
minimum cross-sectional area of the threaded length of the bolt. The design force Ft,Ed is limited by EC3-
1-8 [1] to
Eq6.2

where γM2 = 1.25 (1.50 for Grade 4.6 bolts) and As is the tensile stress area of the bolt. The use of 0.9fub
for the limit state of bolt fracture in addition to the partial factor γ M2 for the connector resistance reflects
the reduced ductility at tensile fracture compared with shear failure.

Bearing bolts in shear and tension (Page 24 table 3.4 ES EN1993-1-8-2015)

Test results for bearing bolts in shear and tension suggest a circular interaction relationship for the strength
limit state. EC3-1-8 [1] uses a more conservative interaction between shear and tension, which can be
expressed in the form of equation 6.3, and
where Fv,Rd is the shear resistance when there is no tension, and Ft,Rd is the tension resistance when there
is no shear.
Eq6.3

1.1. Bolts in bearing (Page 24 table 3.4 ES EN1993-1-8-2015)

It is now commonly the case that bolt materials are of much higher strengths than those of the steel plates
or elements through which the bolts pass. As a result of this, bearing failure usually takes place in the plate
material rather than in the bolt. The design of plates against bearing failure is discussed.
For the situation in which bearing failure occurs in the bolt rather than the plate, EC3-1-8 [1] requires the
design bearing force Fb,Ed on a bolt whose ultimate tensile strength fub is less than that of the plate material
fu to be limited to
Eq6.4

in which γM2 = 1.25 (1.50 for Grade 4.6 bolts), t is the thickness of the connected element, and d is the
diameter of the bolt. Equation 6.4 ignores the enhancement of the bearing strength caused by the triaxial
stress state that exists in the bearing area of the bolt, but it seldom governs except for very high-strength
plates.
1.2. Design against slip (Page 27 ES EN1993-1-8-2015)
For design against slip in a friction-grip joint for either serviceability or at ultimate loading, the shear force
on a preloaded bolt Fv,Ed,ser or Fv,Ed determined from the appropriate loads must satisfy

2
 CHAPTER IV CONNECTION

Eq6.5

in which γM3 = 1.25 for ultimate loading (for serviceability γM3, service = 1.1), n is the number of friction
surfaces, ks is a coefficient given in Table 3.6 of EC3-1-8 [1] which allows for the shape and size of the
hole, µ is a slip factor given in Table 3.7 of EC3-1-8.

When the joint loading induces bolt tensions, these tend to reduce the friction clamping forces. EC3-1-8
requires the shear serviceability or ultimate design load Fv,Ed,ser or Fv,Ed to satisfy
Eq6.6
in which Ft,Ed,ser or Ft,Ed is the total applied tension at service loading or ultimate loading, including any
prying forces, and the slip resistance partial safety factor is γM3 = 1.1 for service loading and 1.25 for
ultimate loading.
2. Design of bolted plates
A connection plate used in a joint is required to transfer actions which may act in the plane of the plate, or
out of it. These actions include axial tension and compression forces, shear forces, and bending moments.
The presence of bolt holes often weakens the plate, and failure may occur very locally by the bearing of a
bolt on the surface of the bolt hole through the plate (punching shear), or in an overall mode along a path
whose position is determined by the positions of several holes and the actions transferred by the plate, such
as that considered in chapter 2 for staggered connectors in tension members.
2.1. Bearing and tension (Page 24 table 3.4 ES EN1993-1-3-2015)
Bearing failure of a plate may occur where a bolt bears against part of the surface of the bolt hole through
the plate, as shown in Figure 6.1a. After local yielding, the plate material flows plastically, increasing the
circumference and thickness of the bearing area, and redistributing the contact force exerted by the bolt.
EC3-1-8 [1] requires the plate-bearing force Fb,Ed due to the design loads to be limited by
Eq6.7

in which fu is the ultimate tensile strength of the plate material, d is the bolt diameter, t is the plate thickness
and γM2 = 1.25 is the partial factor for connector resistance (1.50 for Grade 4.6 bolts).

The coefficient αd in equation6.8 is associated with plate tear-out by shearing in the direction of load
transfer, commonly when the bolt is close to the end of a plate, as shown in Figure 6.1 b. In this case, EC3-
1-8 requires αd to be determined from
Eq6.8

for internal bolts if these bolts are closely spaced, in which d0 is the bolt diameter and the dimensions e1
and p1 are shown in Figure 2.1 of chapter-2. The coefficient k1 in equation 6.9 is associated with tension
fracture perpendicular to the direction of load transfer, commonly when the bolt is close to an
edge, as shown in Figure 6.1 e. Hence, EC3-1-8 uses
Eq6.9
for internal bolts, where p2 is the distance between rows of holes

3
 CHAPTER IV CONNECTION

Figure 6.1

2.2. Shear and tension(Page 29 ES EN1993-1-8-2015)

Plate sections may be subjected to simultaneous normal and shear stress, as in the case of the splice plates
shown in Figure 6.2a and b. These may be designed conservatively against general yield by using the shear
and bending stresses determined by elastic analyses of the gross cross-section in the combined yield
criterion, and against fracture by using the stresses determined by elastic analyses of the net section in an
ultimate strength.
Block failure may occur in some connection plates as shown in Figure 6.2c and d. In these failures, it may
be assumed that the total resistance is provided partly by the tensile resistance across one section of the
failure path, and partly by the shear resistance along another section of the failure path. This assumption
implies considerable redistribution from the elastic stress distribution, which is likely to be very non-
uniform. Hence EC3-1-8 [1] limits the block-tearing resistance for situations of concentric loading such as
in Figure 6.2c to
Eq6.10

in which Ant and Anv are the net areas subjected to tension and shear, respectively, fy/√3 is the yield stress
of the plate in shear, γM0 = 1.0 is the partial factor for plate resistance to yield and γM2 = 1.1 is the partial
factor for fracture. Equation 6.10 corresponds to an elastic analysis of the net section subjected to uniform
shear and tensile stresses. For situations of eccentric loading such as in Figure 6.3d, the block-tearing
resistance is limited to Eq6.11

4
 CHAPTER IV CONNECTION

Figure 6.2

A. Pins Pin joints used to be provided in some triangulated frames where it was thought to be important
to try to realize the common design assumption that these frames are pin-jointed
B. Rivets In the past, hot-driven rivets were extensively used in structural joints.
C. Welds Structural joints between steel members are often made by arc-welding techniques,
in which molten weld metal is fused with the parent metal of the members or component plates being
connected at a joint. Welding is often cheaper than bolting because of the great reduction in the
preparation required, while greater strength can be achieved, the members or plates no longer being
weakened by bolt holes, and the strength of the weld metal being superior to that of the material
connected. In addition, welds are more rigid than other types of load-transferring fasteners. On the
other hand, welding often produces distortion and high local residual stresses, and may result in
reduced ductility, while site welding may be difficult and costly.

a. BUTT WELDS, such as that shown in Figure 6.1a, may be used to splice tension members. A full
penetration weld enables the full strength of the member to be developed, while the butting
together of the members avoids any joint eccentricity. Butt welds often require some machining
of the elements to be joined. Special welding procedures are usually needed for full strength welds
between thick members to control the weld quality and ductility, while special inspection
procedures may be required for critical welds to ensure their integrity. These butt welding
limitations often lead to the selection of joints which use fillet welds.

b. FILLET WELDS (see Figure 6.3b) may be used to connect lapped plates, as in the tension
member splice (Figure 6.3c), or to connect intersecting plates (Figure 6.3d). The member force is
transmitted by shear through the weld, either longitudinally or transversely. Fillet welds, although
not as efficient as butt welds, require little if any preparation, which accounts for their extensive
use. They also require less testing to demonstrate their integrity

5
 CHAPTER IV CONNECTION

Figure 6.3

3. Design of welds
3.1. Full penetration butt welds
Full penetration butt welds are so made that their thicknesses and widths are not less than the corresponding
lesser values for the elements joined. When the weld metal is of higher strength than those of the elements
joined (and this is usually the case), the static capacity of the weld is greater than those of the elements
joined. Hence, the design is controlled by these elements, and there are no design procedures required for
the weld. EC3-1-8 [1] requires butt welds to have equal or superior properties to those of the elements
joined, so that the design resistance is taken as the weaker of the parts connected.

Partial penetration butt welds have effective (throat) thicknesses which are less than those of the elements
joined. EC3-1-8 requires partial penetration butt welds to be designed as deep-penetration fillet welds.

3.2. Fillet welds

In the simple design method of EC3-1-8 [1] the design weld forces FTy,Ed, FTz,Ed, and FL,Ed per unit
length due to the factored loads are limited by
Fw,Ed ≤ Fw,Rd
where F w,Ed is the resultant of all of the forces transmitted by the weld per unit length given by
Eq6.12

and
Fw,Rd = fvw,da
is the design weld resistance per unit length, in which

Eq6.13

is the design shear strength of the weld, fu is the ultimate tensile strength of the plate which is less than that of the
weld, βw is a correlation factor for the steel type between 0.8 and 1.0 given in Table 4.1 of EC3-1-8, and γM2 = 1.25
is the connector resistance partial safety factor.

6
 CHAPTER IV CONNECTION

EC3-1-8 also provides a less conservative directional method which makes some allowance for the dependence of the
weld strength on the direction of loading by assuming that the normal stress parallel to the axis of the weld throat does
not influence the design resistance. For this method, the normal stress perpendicular to the throat σ⊥ is determined as

Eq6.14

Figure6.4

D. Plates and Cleats Intermediate plates (or gussets), fin (or side) plates, end plates, and angle or tee
cleats are frequently used in structural joints to transfer the forces from one member to another. Examples
of flange cleats and plates are shown in the beam-to-column joints of Figure 6.1, and an example of a gusset
plate is shown in the truss joint of Figure 6.3i. Stiffening plates, such as the seat stiffener (Figure 6.3 f) and
the column web stiffeners (Figure 6.3g), are often used to help transfer the forces in a joint.