Engineering Mechanics 2

Mechanics of Materials
Chapter 1

Tension and Compression in Bars

Introduction

• Real physical bodies were approximated by rigid bodies.

• In many engineering problems the deformations also have to be

calculated

• The bodies must then be considered as being deformable.

1.1 Stress
• The external load (F) causes internal forces
• The internal forces can be visualized by an imaginary cut of
the bar
• They are distributed over the cross section (see Fig. 1.1b)
and are called stresses.
• they have the dimension force per area and are measured
• for example, as multiples of the unit MPa (1 MPa = 1
N/mm2)
• Stresses are denoted by σ. We assume that they act
perpendicularly to the exposed surface A of the cross
section and that they are uniformly distributed.
• Since they are normal to the cross section they are called
normal stresses
• Their resultant is the normal force N shown in Fig. 1.1c

N = σA
• positive normal force N (tension) the stress σ is then positive (tensile stress).

• Reversely, if the normal force is negative (compression) the stress is also

negative (compressive stress).
• imagine the bar being sectioned by a cut which is
not orthogonal to the axis of the bar so that its
direction is given by the angle φ (Fig. 1.1d)
• The internal forces now act on the exposed
surface
A∗ = A/cosφ.

• Again we assume that they are uniformly

distributed.
• We resolve the stresses into:
a) a component σ perpendicular to the surface Fig. 1.1
(the normal stress)
b) a component τ tangential to the surface (shear
stress) (Fig. 1.1e)
If we section a bar near an end which is
subjected to a concentrated force F (Fig. 1.2a,
section c − c)??
• we find that the normal stress is not
distributed uniformly over the cross-
sectional area.
• The concentrated force produces high
stresses near its point of application
(Fig. 1.2b)
• This phenomenon is known as stress
concentration
• the high stresses decay rapidly towards the
average value σ0 as we increase the distance
from the end of the bar.(Saint-Venant’s
principle)
• The uniform distribution of the stress is also disturbed by
holes, notches or any abrupt changes (discontinuities) of
the geometry.
• for example, a bar has notches the remaining cross-
sectional is subjected to a stress concentration.
A bar with only a slight taper:
• The cross-sectional area A and the stress σ
depend on the location along the axis

• If volume forces act in the direction of the

axis in addition to the concentrated forces,
then the normal force N also depends on the
location
• In engineering applications structures have to be designed in
such a way that a given maximum stressing is not exceeded.

• the stress σ must not exceed a given allowable stress (σallow) :

|σ| ≤ σallow. (Note that the allowable stresses for tension and
for compression are different for some materials.)

• The required cross section Areq of a bar for a given load and
thus a known normal force N can then be determined from
σ = N/A

This is referred to as dimensioning of the bar.

• the allowable load can be calculated from |N| ≤ σallowA in the case
of a given crosssectional area A.
• If the cross-sectional area is not constant or if the
bar is subjected to volume forces acting along its
axis, the strain may depend on the location.

we have to use a local strain as follows:

• We consider an element of the bar (Fig. 1.6) instead of the whole bar
• It has the length dx in the undeformed state. Its left end is located at x, the right end
at x + dx.
• If the bar is elongated, the cross sections undergo displacements in the x-direction
which are denoted by u.
• They depend on the location: u = u(x). Thus, the displacements are u at the left end of
the element and u + du at the right end.
• The length of the elongated element is dx+(u+du)−u = dx+du. Hence, the elongation
of the element is given by du.
• Now the local strain can be defined as the ratio between the elongation and the
undeformed length of the element:
• Now the local strain can be defined as the ratio between the
elongation and the undeformed length of the element:

➢ If the displacement u(x) is known, the strain ε(x) can be determined

through differentiation.
➢ if ε(x) is known, the displacement u(x) is obtained through
integration.

• The displacement u(x) and the strain ε(x) describe the geometry of the
deformation.
• Therefore they are called kinematic quantities.
• Equation (1.7) is referred to as a kinematic relation.
1.3 Constitutive Law

• The stresses and the strains are not independent!

➢ depends on the load which acts on the body.

Constitutive law: The physical relation that connects stress and

strain, and describes the behavior of the material of the body under a
load. It depends on the material and can be obtained only with the aid
of experiments.
Finding the relationship between stress and strain:
• One of the most important experiments to find the relationship
between stress and strain is the tension or compression test.

• Here, a small specimen of the material is placed into a testing machine

and elongated or shortened.

• The force F applied by the machine onto the specimen can be read on
the dial of the machine; it causes the normal stress σ = F/A.

• The change ∆l of the length l of the specimen can be measured and the
strain ε = ∆l/l can be calculated.
 stress-strain diagram
Proportional limit σP:
below this value, the stress is proportional to the
strain of the strain exhibited small values and the
relationship is linear (straight line).

Yield stress σY:

From this point of the stress-strain diagram the
strain increases at a practically constant stress: the
material begins to yield.
Note that many materials do not exhibit a
pronounced yield point.
 stress-strain diagram
Strain hardening:
At the end of the yielding, the slope of the curve
increases again which shows that the material can
sustain an additional load.

lateral contraction:
The elongation of the bar leads to a reduction of
the cross-sectional area A. The cross-sectional area
decreases uniformly over the entire length of the
bar in the case of small stresses.

Necking:
The cross-sectional area begins to decrease locally
at very high stresses
True stress or physical stress: σt = F/Aa
It represents the true stress in the region where necking takes place.
Since the actual cross section Aa may then be considerably smaller than
the original cross section A, the stress σ = F/A does not describe the real
stress any more. It called nominal or conventional or engineering stress
Consider a specimen being first loaded by a force which causes the stress σ.
• When σ < σY . Subsequently, the load is again removed. Then the specimen will return to
its original length: the strain returns to zero. In addition, the curves during the loading
and the unloading coincide. This behavior of the material is called elastic; the behavior
in the region σ ≤ σP is referred to as linearly elastic.

• When σ > σY is reached. Then the curve during the unloading is a straight line which is
parallel to the straight line in the linear-elastic region, If the load is completely removed
the strain does not return to zero: a plastic strain εpl remains after the unloading. This
material behavior is referred to as plastic.
(The constitutive law) Hooke’s law
In the region of linearly elastic which will be called from now on elastic, we
have the linear relationship between the stress and the strain

The proportionality factor E is called modulus of elasticity or Young’s modulus

The relation (1.8) is valid for tension and for compression: the modulus of
elasticity has the same value for tension and compression. However, the
stress must be less than the proportional limit σP which may be different for
tension or compression.
The modulus of elasticity depends on the composition of the
material and on the temperature.
• The strains are not only caused by forces but also by changes of
the temperature.
• Experiments show that the thermal strain εT is proportional to the
change ∆T of the temperature if the temperature of the bar is
changed uniformly across its section and along its length:

• Coefficient of thermal expansion (αT): It is a material constant and is

given in the unit 1/◦C.

• If the change of the temperature is not the same along the entire
length of the bar (if it depends on the location) then (1.10)
represents the local strain
εT (x) = αT ∆T (x)
If a bar is subjected to a stress σ as well as to a change ∆T of the
temperature, the total strain ε is obtained through a superposition of (1.9)
and (1.10):
 1.4 Single Bar under Tension or Compression
Derivation of the equilibrium condition
for an element:

• consider a bar which is subjected to two forces F1 and F2 at its ends and to a
line load n = n(x)
• The forces are assumed to be in equilibrium.
• Element of infinitesimal length dx
• The free-body diagram shows the normal forces N and N + dN, respectively,
at the ends of the element; the line load is replaced by its resultant ndx
Equilibrium of the forces in the direction of the axis of the bar
→: N + dN + n dx − N = 0
yields the equilibrium condition:

In the special case of a vanishing line load (n ≡ 0) the normal force in the bar is
constant:
If we insert the kinematic relation and σ = N/A into Hooke’s
law we obtain:

• This equation relates the displacements u(x) of the cross sections and the
normal force N(x). It may be called the constitutive law for the bar.

• The quantity EA is known as axial rigidity.

• Equations (1.13) and (1.14) are the basic equations for a bar under
tension or compression.
➢ In the special case of a bar (length l) with constant axial rigidity (EA = const) which is
subjected only to forces at its end (n ≡ 0,N = F) and to a uniform change of the
temperature (∆T = const), the elongation is given by:

• A change of the temperature causes only thermal strains (no stresses!) in a

statically determinate system.
• If we solve (1.14) for N and insert into (1.13) we obtain:

we obtain:

When EA = const and ∆T = const:

• If the functions EA(x), n(x) and ∆T (x) are given, the displacement u(x) of an arbitrary cross
section can be determined through integration of (1.20a).

• The constants of integration are calculated from the boundary conditions:

for example,
o one end of the bar is fixed then u = 0 at this end
o one end of the bar can move and is subjected to a force F0, then applying (1.14) and N =
F0 yields the boundary condition u′ = F0/EA + αT ∆T . This reduces to the boundary
condition u′ = 0 in the special case of a stress-free end (F0 = 0) of a bar whose
temperature is not changed (∆T = 0).

• Frequently, one or more of the quantities in (1.20) are given through different functions
of x in different portions of the bar (e.g., if there exists a jump of the cross section). Then
the bar must be divided into several regions and the integration has to be performed
separately in each of theses regions. In this case the constants of integration can be
calculated from boundary conditions and matching conditions
Example of a statically determinate system:

• We cut the bar at an arbitrary position x

• The normal force N is equal to the weight W∗ of the
portion of the bar below the imaginary cut.
N(x) = W∗(x) = W(l − x)/l.
• Equation (1.4) now yields the normal stress:

• Accordingly, the normal stress in the bar varies

linearly; it decreases from the value σ(0) = W/A at
the upper end to σ(l) = 0 at the free end.
The elongation ∆l of the bar due to its own weight is
obtained from (1.16):

• It is half the elongation of a bar with negligible weight

which is subjected to the force W at the free end.
We may also solve the problem by applying the differential equation (1.20b)
for the displacements u(x) of the cross sections of the bar. Integration with
the constant line load n = W/l yields:

The constants of integration C1 and C2 can be determined from the boundary conditions. The
displacement of the cross section at the upper end of the bar is equal to zero: u(0) = 0. Since
the stress σ vanishes at the free end, we have u′(l) = 0. This leads to C2 = 0 and C1 = W .
Thus, the displacement and the normal force are given by

Since u(0) = 0, the elongation is equal to the displacement of the free end:
Chapter 4

Bending of Beams

To derive the equations which enable us to determine the stresses

and the deformations during bending
4.1 Introduction
• Beams are among the most important elements in structural engineering.
• A beam has the geometrical shape of a straight bar, i.e., the dimensions of its cross-sectional
area are much smaller than its length.
• However, in contrast to the members of a truss it is loaded by forces which are perpendicular
to its axis
Due to the applied loads, the originally straight beam deforms (Fig. 4.1a). This
is referred to as the bending of the beam.

As a consequence, internal forces (= stresses) are generated in the beam,

the resultants of which are the shear force V and the bending moment M
Let us consider the beam in Fig. 4.1b where the z-axis is an axis of symmetry.
• the load is assumed to cause a bending moment M only (no shear force and no normal
force) (pure bending).

• in the region between the two forces F acting

at the beam in Fig. 4.1c.
• In pure bending there are only normal
stresses σ which act in the direction of the x-
axis in the cross-sectional areas (Figs. 4.1b,d)
• They are independent of the y-coordinate
and they are linearly distributed in the z-
direction
• the normal stresses may be written as:

where c is as yet an unknown factor of proportionality.

The bending moment M is statically equivalent to the sum of the moments of the distributed
normal stresses with respect to the y-axis

is called moment of inertia. With (4.3), the above equation can be written as c = M/I, and
(4.1) yields a relation between the stresses and the bending moment:
• the stress at a fixed value of z depends not only on the bending moment M but also on the
moment of inertia I
• The moment of inertia is a purely geometrical quantity which is related to the shape of the
area.
4.2 Second Moments of Area

Integrals which contain the square of the distances of the element dA or

the distances as a product are called second moments of area. They are
also referred to as moments of inertia of the area.
• The integrals Iy and Iz in (4.6a) are referred to as rectangular moments of inertia,
• Iyz is called the product of inertia and
• Ip is the polar moment of inertia.
• The second moments of area have the dimension length4; they are given, for example,
in the unit cm4 .
• The magnitude of a moment of inertia depends on the location of the origin and on
the orientation of the axes. Whereas Iy, Iz and Ip are always positive (the integrals
involve the squares of the distances)
• the product of inertia Iyz may be positive, negative or zero (the integrals contain the
product of y and z which need not be positive). In particular, Iyz = 0 if one of the axes is
an axis of symmetry of the area A.

Let, for example, the z-axis be an axis of symmetry (Fig. 4.3a). Then for every infinitesimal
area dA located at a positive distance y, there exists a corresponding element located at a
negative distance. Therefore, the integral (4.6b) is zero.
Frequently, an area A is composed of several parts Ai, the moments of inertia of which are
known (Fig. 4.3b). In this case, the moment of inertia about the y-axis, for example, is
obtained as the sum of the moments of inertia Iyi of the individual parts about the same
axis:
Example: we consider a rectangular area (width b, height h). The coordinate system with the
origin at the centroid C is given; the y-axis is parallel to the base (Fig. 4.4a). In order to
determine Iy, we select an infinitesimal area dA = bdz according to Fig. 4.4b. Then every point
of the element has the same distance z from the y-axis. Thus, we obtain
Example we calculate the moments of inertia of a circular area (radius R), the origin of the
coordinate system being at the centroid C (Fig. 4.5a). Due to the symmetry of the problem,
the moments of inertia about every axis through C are equal. Therefore, according to (4.6c),

• The product of inertia Iyz is zero (symmetry)

To determine the polar moment Ip, we choose an infinitesimal circular ring
with the area dA = 2π r dr (every point of this ring has the same distance r from
the center C):
The moments of inertia of a circular ring with outer radius Ra and inner radius Ri (Fig.
4.5c). Subtraction leads to

the moments of inertia of a thin circular ring (Fig. 4.5d)

• Rm = 1/2 (Ra + Ri) (= arithmetic mean value of the radii Ri and Ra)
• the thickness t = Ra − Ri of the ring
4.2.2 Parallel-Axis Theorem
• consider two different coordinate systems, namely, y,z and y ,̄ z ,̄ as
shown in Fig. 4.7.
• The axes of the two systems are assumed to be parallel and the origin of
the y, z-system is the centroid C of the area.
• In the following we shall investigate how the second moments of area
with respect to the different coordinate systems are related.

the parallel-axis theorem

• These are the relations between the moments of inertia with respect to the
axes through the centroid C and the moments of inertia with respect to axes
which are parallel to them.
• The Equations(4.13) are known as the parallel-axis theorem.
• Note that when applying the parallel-axis theorem one of the two axes must be
a centroidal axis.
Example how to determine the moments of inertia
with respect to the axes y ̄, z ̄ for the rectangle shown:
4.3 Basic Equations of Ordinary Bending Theory

The bending moment M and the shear force V are the resultants
of the normal stresses σ (acting in the x-direction) and the shear
stresses τ (acting in the z-direction), respectively
Every point of a cross section undergoes the same deflection in the z-direction.
This implies that the height of the beam does not change due to bending:
εz = ∂w/∂z = 0.
4.4 Normal Stresses
the normal stresses in the cross section of a beam
are obtained:

• This equation is called the bending formula. It shows that the normal stresses,
which are also referred to as the flexural or bending stresses, are linearly
distributed in z-direction as shown in Fig. 4.14.
• If the bending moment M is positive, the stresses are positive (tensile stresses)
for z > 0 and they are negative (compressive stresses) for z < 0. For z = 0 (i.e., in
the x, y-plane) we have σ = 0 and called neutral surface.
• The bending stresses (tensile or compressive) attain their maximum values at
the extreme fibers.
• the notation zmax for the maximum value of z:
• The quantity W (often also denoted by S) is called the section modulus.
• we obtain the maximum tensile or compressive stress, respectively:

If the state of stress in a beam is investigated, it often suffices to determine only the
normal stresses since the shear stresses are usually negligibly small (slender beams!).

There are several different types of problems arising in this context.

If, for example, the bending moment M, the section modulus W and the allowable
stress σallow are known, one has to verify that the maximum stress σmax satisfies the
requirement
On the other hand, if M and σallow are given, the required section modulus can be
calculated from:
4.5 Deflection Curve
4.5.1 Differential Equation of the Deflection Curve

These Equations represent four differential equations for

the stress resultants V , M and the kinematic quantities ψ,
w.

Provided the bending moment M and the flexural rigidity EI are known,
integration of (4.31) yields the slope w′(x) and the deflection w(x), also
called the elastic line.
• If the load q and the flexural rigidity EI are given, the deflection w can be determined
through integration of the above differential equation.

• The constants of integration are determined from boundary conditions

• Geometrical boundary conditions are statements concerning the geometrical

(kinematic) quantities w or w ′.

• Statical boundary conditions are statements referring to the stress resultants V or M .

• Two boundary conditions can be formulated at each end point of a beam

4.5.2 Beams with one Region of Integration
• the integration can be performed in one region, i.e., we assume that each of the
quantities q(x), V (x), M(x), w′(x) and w(x) is given by one function for the entire length of
the beam.
M = −F (l − x)
Introducing into (4.31) and integrating yields:
• Consider the three cases in the figure below:

• We introduce a coordinate
system and integrate

• These equations are independent of

the supports and therefore are valid
for all three cases
The elastic line:
The elastic line:
The elastic line: