Chapter Two: Tension Members

Chapter Two: Tension Members

2.1 Introduction
Tension members are structural members that carry pure tension loads. They are efficient
carriers of load and are used encountered in most steel structures. The bottom chords of
roof and bridge trusses are c1assic examples of tension members. Steel cables in suspension
and cab1e-stayed bridges, cab1es-supported roofs, guyed microwave and radio
communication towers and power transmission towers, elevator cables and those cables in
parts of hoisting equipment are all examples of tension members.
Certain web members of a truss system may be in tension for certain loading condition and
in compression for other loading conditions. Wind bracing in an X configuration is frequently
used where the members are so flexible that "buckling" takes place under compression
stresses developed by wind in one direction but functions as a tension member for the
reversed wind.
Tension members frequently appear as secondary members, being used as tie rods to stiffen
a trussed floor system or to provide intermediate support for a wa1l girt system.
The selection of their cross section is one of the simplest and most straightforward
procedures encountered in the design of steel components. Since stability is of minor
concern with tension members, the process of designing such structural members is reduced
to:

selecting a section with sufficient cross-sectional area to carry the design load
without exceeding the design tensile stress as stipulated in relevant codes of
practice
proportioning connections so that all relevant design specifications are met with
regard to arrangement as well as stress limitations.

In all these, the tensile strength of steel is used. In this stress configuration, member
buckling or warping is not a matter of concern. However, specifications normally require a
minimum amount of member stiffness or rigidity with the view of preventing undue sagging,
deflection and vibration and, accordingly, slenderness ratio is 1imited by design
specifications in order to account for this requirement.
Tension members are frequently subjected to bending stresses in addition to the principal
tensile forces. These conditions occur when the cross section is acted upon by eccentric
forces. This calls for additional investigation of the member for proper design and members
subjected to such a condition of combined bending and tensile stresses will be discussed
later.
2.2 Types of Tension Members
Tension members may consist of a single structural shape or they may be built up from a
member of structural shapes as shown in Fig. 2. 2
The cross sectional arrangement of axially stressed tension members is structurally
unimportant so long as the net cross sectional are is sufficient to carry the design loads and

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

Fig. 2.1 Tension Members in Buildings and Bridges

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

the shape can be conveniently connected to other members in the structure. In view of this,
their form is governed largely by the type of structure of which they for parts and by the
method of joining them to the connecting portions of the structure. The only other
structural requirement is that they be sufficiently stiff to prevent harmful vibration,
unsightly sagging, or, where a member may resist a chance of reversal stress to compression
of small but indeterminate magnitude, to prevent buckling.

Fig. 2.2 Cross-sections of Typical Tension Members

Accordingly, if a member at the end is to be connected by bolts or rivets, the angle,

channel, or I section, single or built-up, will be better suited. The use a particular rolled or
built-up shape will be dictated, in addition to its capacity, by the remainder of the
structure; i.e., by the availability of sufficient space on the joint where the member will be
framed into. On the contrary, plates and angles are mostly used in welded structures. For
light trusses and for bracing systems, single angle sections are commonly used. The use of
double angles is generally preferred since the joint will be more symmetrical both in and out
of plane, as opposed to using a single angle, which will always have an out-of-plane
eccentricity.

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

Tension rods may be used as suspenders for suspension bridges and for smalls-pan roof
trusses. For heavy building trusses and long-span bridges, the eye bar is economical to use.
For latticed girders, the chord members are generally built-up sections
.

For carrying greater tension the members have to provide larger net area and therefore
built up sections might be the only effective choice. Such members are also required when a
single or a pair of angles, or anyone of the standard rolled shapes does not have sufficient
rigidity (measured by L/r), or the joint will be impractical to fabricate. For long-span light
structures, tubular sections are ideally suited.
In general, therefore, the use of single structural shapes is more economical than built up
sections. However, the latter may be required under any of the following situations:

the tensile capacity of a single rolled section is not sufficient.

the L/r ratio (the ratio of the unbraced length to the minimum radius of gyration)
does not provide sufficient rigidity.
the effect of bending combined with the tensile behavior requires a large lateral
stiffness
usual connection details require a particular cross section
esthetic control

1.3 Design Consideration

Although the design of tension members is the simplest and most straight-forward one
compared to those for various other member types such as compression or bending. The
process nevertheless requires consideration of several factors. A member subject to axial
tension is supposed to develop a uniform tensile stress across the entire cross-sectional
area. The preconditions for such assumption are as follow:
Axial force is acting along the centroid of the cross section
No bending moment exists on the section
Inter-connections of members or joints are such that the center of gravity of the
member is collinear; that is, it has no eccentricity with the joint.
In order to fulfill these assumptions, due consideration need be given, among others, to
connection types and details, types of shapes available or required for the intended system,
and the effects of shear flow in the section.
Strength as a Design Criterion
The problem of designing a tension member is basically one of providing a member with
sufficient cross-sectional area to resist the applied loads with an adequate margin of safety
against tensile failure. The controlling strength limit state for tension member will be either:

a) yielding of gross cross-sectional area of the member away from the joints, or
b) Fracture of the effective net sectional area through the holes at the joints.
Net Area:
For tension members having holes for rivets and bolts, the reduced cross section is
referred to as the net area. The determination of the net section involves the geometric
spacing of the holes made to accommodate the connecting bolts and rivets.

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

The net area of a cross-section or element section shall be taken as its gross area less
appropriate deductions for all holes and other openings. When the fastener holes are not
staggered the total area to be deducted should be the maximum sum of the sectional areas
of the holes in any cross-section perpendicular to the member axis.
Accordingly, the net area Aeff for the determination of section capacity will be given by:
no

Aeff Ag

(2.1)

o ,i i

i 1

Where: Ag = gross cross sectional area

do,i = hole diameter at section i

tj = thickness of the section at i
If the holes are not disposed symmetrically about the centerline of the section, an
effective net area, obtained by multiplying the net area by a reduction factor kA, should be
used. For a single hole, the reduction factor is given by:

K A 1

do
b

2e
1 b

(2.2)

where: do is the hole diameter,

e is edge distance ( from hole center to edge )
b is width of the section.
When the holes are staggered, the stress distribution is more complicated and an
approximation is allowed (Fig. 2.3a).

Fig. 2.3 Determination of Net Section

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

The almost universally adopted procedure is as follows:

Take any reasonable and possible path across a chain of holes and deduct one hole width for
each bolt hole encountered.

For each change in direction from hole to the next hole, add back the quantity

s2
4p

where s is

pitch or longitudinal distance between adjacent holes and g is gauge distance between
adjacent holes across the width.
In general, the net sectional area Aeff can be determined from:
np

n0

A eff Ag d 0.i t i
i 1

s2
t j

i 1 4 g j

(2.3)

where
Ag is gross cross-sectional area
Do is nominal diameter of the ho1e (bolt cutout)
t is thickness of the component element (note that elements within cross section may have
different thickness, such as the webs and flanges in rolled sections)
S is the staggered pitch, the spacing of the centers of two consecutive holes in the chain measured
parallel to the member axis.
g is the gauge, the spacing of the centers of the same two holes measured perpendicularly to the
member axis.
In an angle, or other member, with holes in more than one plane, the gauge shall be measured along the
center of thickness of the material (Fig.2.3b).
Effective Net area
Then net area computed in the previous section may not correctly reflect the strength specially:

when the tension member has a profile consisting elements not in a common plane.

where the tensile load is transmitted at the end of the member by to some but not all of the
elements. Angle section having connection to one leg only is an example of such a situation.

due to shear lag effect (non uniformity of stresses in wide plates i.e. the shear transfer lags
or inefficient)
For such cases the tensile force is not uniformly distributed over the net area. To account for this,
LRFD provides for an effective area Aeff to be
Sym
computed as:
Aeff = U An
where:

(2.4)

U is a reduction coefficient

Gusset plates

gusset plate
C.G. angle

An is net area
x-

x-

x-

Fig. 2.4 Eccentricity in Joints

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter Two: Tension Members

The reduction coefficient is given by:

x
l

0 .9

(2.4a)

where: l is the length of the connection

x is the connection eccentricity (distance from centroid of element being

U

connected eccentrically to plane of load transfer as shown in Fig; 2.4.)

should be calculated using the maximum value of x.

1.4 Limit State Design of Tension Members

Limit state design of tension members calls for verification of the member to withstand
various kinds of failures related to tensile strength both in gross cross section and in
effective net section as well as block shear with respect to tension fracture and shear
fracture.
Ethiopian Building Code Standard EBCS 3, 1995
According to the EBCS 3 Specification, axially loaded tension members designed to resist a
factored axial force of Nt,Sd, calculated using appropriate load combinations, must satisfy
the condition:

N t .sd N

L , RD

where:
Nt, Rd = design tension resistance capacity of the cross-section, taken as a smaller of
either the design plastic resistance Npl,RD of the gross section or the design
ultimate resistance Nu,Rd of the net section at the bolt hole where, again,
Npl,Rd and Nu.Rd are determined as in the fol1awing expressions:

Pi , RD

NU ,

RD

Ag x f y

MO

0.9 x Aeff x fU

(2.5a)

(2.5b)

M2

The partial safety factor MO = 1.1 and while M2 = 1.25 represents resistance of the net
section at bolt holes.
AISC-LRFD Specification
According to the AISC-LRFD Specification, tension members designed to resist a factored
axial force of Pu, calculated using the appropriate load combinations, and must satisfy the
condition:
t Pn Pu

Design of Steel and Timber Structures CE-519

(2.6)

Yibeltal T.

Chapter Two: Tension Members

Where:
Pn = the design tensile strength of the cross section and it is evaluated based on
three limit states: yielding in gross section, fracture in effective net section,
and block shear.
t = 0.9 is the appropriate resistance factor in tension.

Yielding in the cross section away from the joint should be avoided to prevent excessive
deformation that results when steel yields. The design strength for this limit state is
evaluated from the equation:
t Pn = t x fy x Ag

(2.6a)

where
t = 0.9 = resistance factor for tension
fy = specified minimum yield stress of steel
Pn = nominal axial strength

Fracture in effective net section or fracture of the net section the joint should be avoided,
to prevent the loss of load-carrying capacity of the member. The design strength for this
limit state is evaluated from the equation:
t Pn t x fu x Ae

(2.6b)

where:
t = 0.75 = resistance factor for fracture in tension
Fu = specified minimum tensile strength of the material
Ae= effective net cross-sectional area of the member
Pn = nominal axial strength
For members without holes, fu11y connected by welds, both Aeff in EBCS 3 and Ae in AISCLRDF specifications are the smal1er of the gross area of the member and the effective
area of the welds.
As it can be seen from both EBCS 3 and AISC-LRDF specifications, the concept of net
section forms one of the criteria for the determination of limiting strength of the cross
section.
Block Shear
Block shear failure or rupture along a block shear failure path occurs when a segment of the
connecting member is torn out as a result of the combined effects of tension and shear.
Block shear must be checked if the load is transmitted by some but not all of the
component elements of the cross section.

Ethiopian Building Code Standard EBCS 3, 1995

The design value of the effective resistance Veff,Rd for rupture along a block shear failure
path shall be determined from:

Design of Steel and Timber Structures CE-519

Yibeltal T.

Chapter One: Introduction, Material and Design Concepts

Veff . RD

0.6 x f y x Av. ef f

(2.7)

MO

Where MO = 1.1 = partial safety factor

f = specified minimum yield stress of steel
Av,eff = effective shear area subject to block shear
The effective shear area Av,eff for block shear, Fig. 2.5, is
determined from:

Av,eff = t (Lv + L1 + L2 ndo)

(2.7 a)

in which L1 and L2 are given by:

L1 = 5.0d0 a1
L1 = 2.5d0 a2
and n = the number of fastener holes in the block shear
failure path

do = hole diameter
T = thickness of the web or bracket
Fig. 2.5 Net Shear Area for Block Shear.

AISC-LRFD Specification
According to the AISC-LRFD Specification, the design strength for block shear is
determined from the following two conditions:
Tension Fracture Shear Yield: t Pn = 0.75 x (0.60 x fy Agv + fuAnt)
(2.8a)
Shear Fracture Tension Yield: t Pn = 0.75 x (0.60 x fu Anv + fyAgv)
(2.8b)
Where
0.75 = resistance factor for block shear
fy, fu = specified minimum yield stress and tensile strengths, respectively
Agv = gross area of the torn-out segment under shear
Ant = net area of the torn-out segment under tension
Anv = net area of the torn-out segment under shear
Agt = gross area of the torn-out segment under tension
Normally, it is necessary to investigate both the tension fracture - shear yield and the
shear fracture-tension yield criteria. The larger of the two values calculated is to be used
for tPn.
Slenderness Ratio
In al1 tension members, minimum amount of member stiffness or rigidity is required with
the view of preventing undue sagging, deflection and vibration. This is accomplished by
limiting the slenderness ratio given by L/r where L is the length of the member and r is the
list radius of gyration.
____________________________________________________________________________________
Design of Steel and Timber Structures CE-519
Yibeltal T.

Chapter One: Introduction, Material and Design Concepts

AISC specifies an upper limit of 300 on

10

L
r

Adequacy of the Connection

Connections must also be carefully designed and detailed. This topic will be discussed in
detail in Structural Connections and Design of Joints.
Longitudinal Spacing of Connectors
The spacing of connectors in built-up tension members consist of elements in continuous
contact shall conform to the spacing requirements for fasteners. Details of this will be
presented in Structural Connections and Design of Joints.

_______________________________
ADDITIONAL READING

E.H. Gayloard and J.E. Stalmeyer

Chapter 3
Charles G. Salmon and Johne E. Johnson
Chapters 3
Robert Englekirk
Chapter 1
EBSC 3 and EC 3

____________________________________________________________________________________
Design of Steel and Timber Structures CE-519
Yibeltal T.