Chapter 1

Introduction

Solid and structural mechanics is a branch of applied mechanics that deals with the study
of the structures and the machine members under the action of external forces, taking
into account the internal forces created and the resulting deformation.
The principal objective of Solid and Structural mechanics is to determine the stresses,
strains and displacements in structures and machine members due to the loads acting on
them, and subsequently determine the limiting load which the member can stand before
failure (excessive deformation) of the material occurs.

1
Chapter 2

Stress and Strain

2.1 Tensile and compressive stresses

When a load is applied to a member of a machine or structure, the material distorts.

The stress intensity (the load transmitted per unit area of cross section) is the intensity
of the internal forces produced on the material, and the strain is the measure of the
resulting distortion.
Consider a prismatic bar shown in figure 2.1.

Figure 2.1: Stress-strain concepts

Let;

F the axial force applied at the ends of the bar.

δ the increase in length.

Lo the original length of the bar.

d the diameter of the bar.

The weight of the bar is neglected and it is assumed that the only active forces are the
axial forces at the ends of the bar.

2
The intensity of the force (stress, denoted as σ) is the force per unit area,

F
σ= (2.1)
A
Equation 2.1 is valid only if the force is uniformly distributed over the cross section.
This condition is realized if the axial force acts through the centroid of the cross section.
The units of stress are N/m2 (P a), but since the values of stress are large N/mm2 (M P a)
are usually used. If the force F stretches the member, then the stress is called tensile
stress and is positive. If the force F compresses the member, then the stress is called
compressive stress and is negative. Since these stresses act in a direction perpendicular
to the cross section, they are called normal stresses.
The bar in figure 2.1 will change in length when loaded axially, becoming longer when
stretched and shorter when compressed. Strain () is defined as the change in length per
unit length.
δ
= (2.2)
Lo
Similarly, when F stretches the bar we have tensile strain which is positive, but when F
compresses the bar we have compressive strain which is negative. Strain is dimensionless
quantity although for clarity, units like m/m, mm/m or µm/m are used.
Examples 2.1
Suppose the bar in figure 2.1 has a diameter d = 50mm and original length Lo = 2.0m.
When the load F = 27 × 103 N is applied axially, the bar elongates by 1.4mm. Calculate
the normal stress and strain in the bar.
Solution: Using equation 2.1, stress in the bar is;

F 27 × 103
σ= = = 13.8 × 106 N/m2 = 13.8M P a
A π/4(0.05)2

Using equation 2.2, strain in the bar is;

δ 1.4 × 10−3
= = = 0.0007m/m = 700 × 10−6 m/m = 700µm/m
Lo 2.0

Examples 2.2
A circular steel rod of length L and diameter d hangs in a mine shaft and holds an ore
bucket of weight W at its lower end as shown in figure below.

(a) Obtain a formula for the maximum stress σmax in the rod taking into account the
weight of the rod.

(b) Hence or otherwise, calculate the maximum stress if L = 40m, d = 8mm, W =

1.5kN and the mass density of the steel is γ = 7.85 × 103 Kg/m3 .

3
Solution: (a) The maximum axial force Fmax in the rod occurs at the upper end and is
equal to;
Fmax = W + Wr
where Wr is the weight of the rod itself and is given by;
Wr = γgVr = γgAL
In which A is the cross sectional area of the rod. Therefore using equation 2.1 the
maximum stress becomes,
Fmax W + γgAL W
σmax = = = + γgL
A A A

(b) To calculate the maximum stress, we substitute numerical values to the equation
above, thus
1.5 × 103
σmax = + 7.85 × 103 (9.81)(40) = 32.92 × 106 P a = 32.92M P a
π/4(0.008)2

2.2 Engineering stress and true stress

Consider a prismatic bar which is stretched until it breaks. In this case the stretching
load F will increase until necking begins, after which it will decrease until the specimen
fails.
The engineering stress (σeng ) is defined as the instantaneous load divided by the original
area of the cross section, that is,
Fi
σeng = (2.3)
Ao
Therefore the σeng decreases after the formation of the neck.
On contrary the actual σ −  curve shows a continuous increase in stress even after
necking. This is because the true stress (σt ) at any instant is obtained by dividing the
instantaneous load by the actual area of cross section at that point, that is,
Fi
σt = (2.4)
Ai

4
2.3 Engineering strain and true strain

Engineering strain is defined as;

δ Lf − Lo
= = (2.5)
Lo Lo
Equation above assumes that the deformation is uniform over specimen length and
throughout the loading process. For ductile materials, the deformation is non-uniform
over the specimen length beyond the elastic limit due to the necking phenomena. There-
fore the definition in equation 2.5 losses meaning in this case. In its place, true strain
(t ) is defined for incremental extension over the original length as,
Z Lf 
dL  L 
f
t = = Ln (2.6)
Lo L Lo
Ao Lf
Due to volume consistency Ao Lo = Af Lf which gives Af
= Lo
, hence equation 2.6
becomes,
A 
o
t = Ln (2.7)
Af

Practice: For a circular cross section show that the true strain can be given as,
d 
o
t = 2Ln
df

eng and t can be related as follows,

L 
f
t = Ln
Lo
L
+ δ  δ 
o
= Ln = Ln 1 +
Lo Lo
= Ln(1 + eng ) (2.8)

2.4 Shear stress and strain

If a piece of material is subjected to equal and opposite forces F which produces a state
of shear as shown in figure 2.2, then the shear stress is defined as the shearing force
divided by the cross sectional area of the plane parallel to the force.

5
 Figure 2.2: Shear stress and strain

Therefore, shear stress (τ ) is

F F
τ= = (2.9)
A ab

If the deformation in the direction of F is x, then the shear strain can be obtained as,
x
tan φ = (2.10)
c
x
where c is the perpendicular distance between the applied forces. Since c
is very small
then equation 2.10 can be written as,
x
φ= (2.11)
c
Therefore shear strain (φ) is shown to be the change of right angle in a body when a
shearing force is applied.

2.4.1 Equality of shear stresses on perpendicular planes

Lets consider a small element of material in shear in a form of a cube of unit thickness
as shown in figure 2.3.

Figure 2.3: Small element of material subjected to shear stresses and strains

6
The prefix convention of the shear stresses is τab where a is the axis perpendicular to the
stress and b is the axis in the direction of the stress.
There are three pairs of complimentary shear stresses: (τxy , τyx ), (τxz , τzx ) and (τyz , τzy ).
If we consider (τxy and τyx ), then summing the moments about the origin, we have;
(τxy × b × 1)a = (τyx × a × 1)b)
τxy = τxy (2.12)
Similarly,
τxz = τzx (2.13)
τyz = τzy (2.14)

2.4.2 Sign conventions for shear stresses and strains

• A shear stress acting on a positive face of an element is positive if it acts in the
positive direction of an axis and negative if it acts in a negative direction of an
axis.
• A shear stress acting on a negative face of an element is positive if it acts in the
negative direction of an axis and negative if it acts in a positive direction of an
axis.
• A shear strain in an element is positive when the angle between two positive faces
(or two negative faces) is reduced.
• A shear strain in an element is negative when the angle between two positive faces
(or two negative faces) is increased.

2.5 Poisson’s ratio

When a material is exposed to a longitudinal load, it experiences both longitudinal

and transverse strains. For instance when a prismatic bar is loaded in tension, the
axial elongation is accompanied by lateral contraction (that is contraction normal to the
direction of applied load). This change in shape is depicted in the figure 2.4. The ratio

Figure 2.4: Axial elongation and lateral contraction of a bar in tension

of the lateral strain (l ) to the axial strain (a ) is known as Poisson’s ratio (ν), thus
l
ν=− (2.15)
a

7
2.6 Volumetric strain

Since the dimensions of a bar in tension or compression changes when the load is applied,
the volume of the bar changes too. The change in volume divided by the initial volume
is called the volumetric strain (v ).
Consider a small element of an isotropic material cut from a bar in tension,

x
z
Figure 2.5: Volumetric strain

Original volume of the cuboid is,

Vo = xyz
If the changes in the axial dimensions are δx, δy and δz for x, y and z dimensions, then
the final volume of the cuboid is,

Vf = (x + δx)(y + δy)(z + δz)

Therefore the volumetric strain is, Similarly,

Vf − Vo
v =
Vo
(x + δx)(y + δy)(z + δz) − xyz
= (2.16)
xyz
Expanding equation 2.16 and ignoring products of small quantities, we have,
δz δy δx
v = + +
z y x
= z + y + x (2.17)

2.7 Tensile test

This is a very common laboratory experiment of determining mechanical properties of

materials.
The test specimen is installed between the two large grips of the testing machine and
then loaded in tension until failure. The machine automatically tabulates and graphs
the results in form of load-extension curve.

8
2.7.1 Stress-Strain curve

Tensile test results depend upon the size and material of the specimen being tested.
It is unlikely that we will design a structural or machine member with same size with
the test specimens. Therefore we need to express the test results in a form that can be
applied to members of any size. This is achieved simply by converting the results into
stress and strains at various magnitudes of the load then plotting a diagram of stress
versus strain.
Such a stress-strain curve is a property of the particular material being tested and conveys
important information about the mechanical properties of the material.
The resulting stress-strain curve depends on the following factors; brittleness or ductility
of the material, whether true stresses/strains or engineering stresses/strains are used.
A ductile material such as mild steel in tension have stress-strain curves as shown in the
figure 2.6.

σ
σ eng − ε eng

σ t − εt D

σ ULT
C

σF D
A
σ UYP
σ LYP B

Y .R S.H .R( P.R) N.R

E.R

O
ε
Figure 2.6: Stress-strain curve for a ductile material

Region OA: The curve begins with a straight line from the origin O to point A, meaning
that the relationship between stress and strain in this region is linear and propor-
tional. This is the linear elastic region and deformations are recoverable upon load
removal. Hooke’s law is obeyed and the slope os the curve at this region is the
Modulus of Elasticity of the material.

Region AB: This is the yielding region. Point A marks the elastic limit also called
Upper Yield Point and there is sudden drop in stress with further strain, and can
only be realized under circumstances where proper centralization of specimen has

9
 been done, tests conducted in ambient conditions, high strain rates are applied and
strain raisers are avoided. Point B is the Lower Yield Point and there is a marked
extension at almost constant load.

Region BC: Once the yield region has been passed, an increasing stress is required to
cause a continued straining, and this behavior is called work-hardening or strain
hardening and the metal in fact becomes harder. The rate of increase of stress is
large than the rate of increase of strain. This region is called plastic region and
the resulting deformation is irrecoverable. The strain energy stored in a specimen
over the plastic region is the toughness of the material.

Point C: This is called the Ultimate Tensile Strength and represents the maximum
stress the specimen can withstand before necking starts.

Region CD: After necking kicks off, further stretching of the specimen is accompanied
by a reduction in the load, and fracture finally occurs at point D.

After point C the reduction in area of the specimen becomes clearly visible and a pro-
nounce necking of the specimen occurs. If the actual cross section area is used to calculate
stress, the true stress-true strain curve (dashed curve) is obtained.

2.7.2 Proof stress

When a material does not show a well defined stress in tension or compression, the proof
stress is required. Its obtained by drawing a line parallel to the linear elastic region of
the curve at the appropriate strain of usually 0.1%.
Therefore proof stress is defined as the stress required to produced a permanent strain
of 0.001 on the removal of stress.
σ

σY

0.001
ε

Figure 2.7: Proof stress at 0.1% strain

10
2.7.3 Tangent modulus

When there is no well defined linear region in a stress-strain curve, a tangent modulus
is used. A tangent modulus is defined as the slope of a straight line that is tangent to
the stress-strain curve at the origin.

Figure 2.8: Tangent modulus

2.7.4 Secant modulus

In practice, obtaining a tangent modulus involves considerable use of judgement, so this

is not a well defined property.
Secant modulus is defined by drawing a line from the point of 0.3% permanent strain,
and where the line cuts the curve is joined to the origin and the slope of the resulting
line gives the secant modulus.

ε
0.003

Figure 2.9: Secant modulus

11
2.7.5 Working stresses and factor of safety

The working stress in a material is the stress the material carries under normal working
condition.
In design, the maximum allowable working stress on a material is usually less than
the stress at yield point (where the material suddenly becomes plastic) to allow for
overloading, non-uniformity of stress distribution, shock loadings, faults in material and
workmanship, etc.
The ratio of the yield stress (σY ) to the maximum working stress (σwmax ) is called the
factor of safety (nsf ), i.e
σY
nsf = (2.18)
σwmax

2.8 Elastic constants

Many structural materials behave both elastically and linearly when first loaded. This
means that their stress-strain curves begin with a straight line passing through the origin.
When a material behaves elastically and also have a linear relationship between stress and
strain, it is said to be linearly elastic. This type of behavior is important in engineering,
since by designing structural and machine members to operate in this region we avoid
permanent deformations due to yielding.
In the elastic region, three constants called elastic constants are,

2.8.1 Young’s Modulus

The linear relationship between the stress and strain for a bar in simple tension or
compression is given by,
σ = E (2.19)
Where σ is the axial stress,  is the axial strain, and E is the constant of proportionality
called the Young’s Modulus or Modulus of Elasticity of the material.
The modulus of elasticity is the slope of stress-strain diagram in the linearly elastic
region. From equation 2.19 we have,
σ
E= (2.20)

Since the strain is dimensionless, then the units of the E are N/m2 .

2.8.2 Modulus of Rigidity

A relationship between shear stress and shear strain may be obtained from a torsion test
on a cylindrical bar in which applied torques and angular twists are measured.

12
Shear stress is proportional to the shear strain within the elastic limit, that is,

τ = Gφ (2.21)

Where τ is the shear stress, φ is the shear strain, and G is the constant of proportionality
called the Modulus of Rigidity of the material.
From equation 2.21 we have,
τ
G= (2.22)
φ
Since the strain is dimensionless, then the units of the G are N/m2 .

2.8.3 Bulk Modulus

When a material is under the action of three equal normal stresses in three mutually
perpendicular directions, then this is termed as hydrostatic stress.
σy

σz
σx

Figure 2.10: Hydrostatic stress

For hydrostatic stress,

σx = σy = σz = σh (2.23)

The hydrostatic stress is directly proportional to volumetric strain within the elastic
limit, thus the constant of proportionality is called Bulk Modulus (K),
σh
K= (2.24)
v

2.9 Strain energy

When a material is loaded it deforms and work is done on it. This energy is stored in
the material and is recoverable when the material is unloaded provided that the material
remains elastic. This energy is called strain energy.
The ability of a material to absorb energy when it is deformed elastically and release it
when it is unloaded is called resilience.

13
2.9.1 Strain energy in tension and compression

Consider an axial load P applied gradually to a tensile specimen thereby producing an

extension δ. If a load-extension curve is plotted for the elastic region, then the work
done on deforming the material or strain energy is represented by the area under the
curve, as shown in figure 2.11

Figure 2.11: Strain energy in tension or compression

Strain energy,
1
U = Pδ (2.25)
2
σLo
From equation 2.1 P = σAo . Also from equation 2.2 and 2.19 we have δ = E
. Therefore
equation 2.25 becomes,
1 σLo
U = σAo
2 E
σ2
= Ao Lo
2E
σ2
= Vo (2.26)
2E

2.9.2 Strain energy in shear

If we assume the load-extension curve in figure 2.11 above to be for a specimen un-
der shear loading as shown in figure 2.2, then the expression for shear strain energy is
obtained as,
1 τ Lo
U = τ Ao
2 G
τ2
= Ao Lo
2G
τ2
= Vo (2.27)
2G

Examples 2.3

14
A tension specimen of circular cross section tapers uniformly from 20mm to 16mm
diameter over a gauge length of 200mm. When an axial force of 40kN is applied, the
extension measured on this gauge length is 0.4mm.

(a) Find the modulus of elasticity of the material.

(b) Compare the strain energy in this specimen with that in a specimen of the same
material of uniform diameter 18mm carrying the same load.

Solution: (a) The specimen can be represented as shown in figure below,

L0 = 200mm

C
L
X
A d 2 = 20mm

O d1 = 16mm d

B
Y
x
dx D

Using similarity in triangles OAB and OCD,

L L + 0.2
=
0.016 0.02
0.0032
L = = 0.8m
0.02 − 0.016

Using similarity in triangles OAB and OXY ,

L x
=
0.016 d
0.016x
d = = 0.02x
0.8

Stress on the element,

P 40 × 103 127.33 × 106

σ= = 2
= 2
N/m2
Ax π/4(0.02x) x

Extension on the element,

σdx 127.33 × 106 dx

dδx = =
E x2 E

15
Total extension on the bar is,
Z 1.0 
127.33 × 106 dx 
δ =
0.8 x2 E
6 Z 1.0 
127.33 × 10 1
= dx
E 0.8 x2

Note that,
Z
xn+1
xn dx =
n+1
Therefore,

127.33 × 106 h 1 i1.0

δ = −
E x 0.8
6
127.33 × 10 31.83 × 106
= (0.25) =
E E
31.83 × 106
0.0004 =
E
E = 79585 × 106 N/m2 = 79.585GN/m2 (GP a)

(b) From equation 2.25

1
U1 = P δ1
2
1
U2 = P δ2
2
Ratio of strain energy is,
1
2
P δ1 δ1
r= 1 =
2
P δ2 δ2

δ1 = 0.0004m
P Lo 40 × 103 × 0.2
δ2 = = = 0.000395m
A2 E π/4(0.018)2 × 79.585 × 109
0.0004
r = = 1.012
0.000395
=⇒ U1 = 1.012U2

Assignment 1

1. A round bar of length L tapers uniformly from radius r1 at one end to radius r2 at
the other. Show that the extension produced by a tensile load P is,
PL
e=
πEr1 r2
where E is the Young’s modulus of the material. If r2 = 2r1 compare this extension
with that of a uniform cylindrical bar having a radius equal to the mean radius of
the tapered bar.

16
2. Show that the volumetric strain of a bar of square cross sectional area undergoing
axial loading is given by,
v = x (1 − 2ν)
where x is the longitudinal strain and ν is the poisson’s ratio.

3. A steel ball of radius r, has equal and parallel flats machined on opposite sides
so that the thickness across the faces is 1.6r. Show that the decrease in thickness
when an axial load W is applied to these faces is given by,
2.197W
e=
πrE
Hint;

W
y
dx

x r

17
2.10 General stress-strain relations

Consider an element of a material shown inσ figure

y
2.12 subjected to an uniaxial stress
σx . The corresponding strain system is shown in the figure alongside.
σx
−ν
E

σx
y σx E
σx

σx
x −ν
E
z

Figure 2.12: Uniaxial stress and resulting strains

The uniaxial stress causes an axial strain in the direction of stress, and lateral strains in
the perpendicular directions. The lateral strains are obtained using the Poisson’s ration
as shown in equation 2.15.
A triaxial state of stress σx , σy and σz is shown in the figure 2.13. The resultant strains
in the xyz-directions are shown in the figure alongside.

σy σx σz
−ν −ν
E E E
σy σz
y

x
σx
σx σx σy σz
z −ν −ν
E E E

σz σy σz σx σy
−ν −ν
E E E

Figure 2.13: Triaxial stress and resulting strains

The total strain in the x-direction is composed of the axial strain due to σx , lateral strain
due to σy and a further lateral strain due to σz . Similarly for total strains in the other
directions. Therefore,
1 
x = σx − ν(σy + σz ) (2.28)
E
18
 1 
y = σy − ν(σx + σz ) (2.29)
E
1 
z = σx − ν(σx + σy ) (2.30)
E
There are no lateral strain associated with shear strains. Hence the shear stresses and
strains relate in the same way for both uniaxial and triaxial strain systems, that is
γxy = τGxy , γyz = τGyz and γxz = τGxz .

2.10.1 Relations of the elastic constants

The elastic constants (E, G and K)described in section 2.8 are not independent of each
other, and are related by the following equations,

E = 2G(1 + ν) (2.31)
E
K = (2.32)
3(1 − 2ν)
9GK
E = (2.33)
3K + G

Proofs

Examples 2.4
Show that E = 2G(1 + ν) for an elastic material.
Solution: Consider an element of material subjected to a tensile stress σ in one direction
together with an equal compressive stress σ in a mutually perpendicular direction. This
is shown in figure below

σ σ

σ
If the material is elastic, then x and y caused by the direct stress σ are given as,

1 σ
x = (σ + νσ) = (1 + ν)
E E
1 σ
y = (−σ − νσ) = − (1 + ν)
E E

19
If the sides of the element are of unit length, the work done in distorting the element
per unit volume is,
1 1 σ
W1 = σx − σy = (x − y )
2 2 2
σh σ σ i
= (1 + ν) + (1 + ν)
2 E E
σ h 2σ i σ2
= (1 + ν) = (1 + ν)
2 E E

In the state of pure shearing under stresses τ = σ, the shearing strain is give by,
σ
γxy =
G
and the work done per unit volume is,
1 σ2
W2 = σγxy =
2 2G

Since the one state of stress is equivalent to the other, the values of work done are equal,

W1 = W2
2
σ σ2
(1 + ν) =
E 2G
E = 2G(1 + ν)

Practice: During your revision times of this unit, show that, for an elastic material the
other two relations given by equations 2.32 and 2.33 holds.
Examples 2.5
A bar 30cm long is 5cm square cross section for 12cm length, 2.5cm diameter for 8cm
length, and 4cm diameter for the remaining length. If a tensile force of 10kN is applied
to the bar, calculate

(i) Maximum and minimum stresses produced in the bar assuming uniform distribution
of load over the area.

(ii) The total elongation of the bar given E = 210GN/m2 .

Solution: The bar is as shown below,

10kN

3
2

1
10kN

Part 1 has A1 = 0.052 = 2.5 × 10−3 m2 and L1 = 12 × 10−2 m.

20
Part 2 has A2 = π4 (0.025)2 = 0.49087 × 10−3 m2 and L2 = 8 × 10−2 m.
Part 3 has A3 = π4 (0.04)2 = 1.25 × 10−3 m2 and L3 = 10 × 10−2 m.
(i) The maximum and minimum stresses are,

P P 10 × 103
σmax = = = = 20.37M P a
Amin A2 0.49087 × 10−3

P P 10 × 103
σmin = = = = 4.0M P a
Amax A1 2.5 × 10−3

(ii) Total elongation,

δt = W2 δ1 + δ2 + δ3
P L1 P L2 P L3
= + +
EA1 EA2 EA3
P h L1 L2 L3 i
= + +
E A1 A2 A3
10 × 103 h 0.12 0.08 0.10 i
= + +
210 × 109 2.5 × 10−3 0.49087 × 10−3 1.25 × 10−3
= 6.152 × 10−6 mm

Examples 2.6
A rectangular plate of uniform thickness has a strain gauge rosette bonded to one surface
at the center as shown in the figure below. It is placed in a test rig which can apply
a biaxial force system along the plate edges. If the measured strains are +0.0005 and
0.0007 in the x and y directions, (E = 208GN/m2 and ν = 0.3)

(a) The corresponding stresses set up in the plate.

(b) The strain through the plate thickness.

σy

σx

σx

σy

21
Solution: (a) For biaxial stress system where σz = 0 equations 2.28, and 2.29 become,
1
x = (σx − νσy )
E
1
y = (σy − νσx )
E
Solving the equations above simultaneously for σx and σy gives,

E(x + νy )
σx =
1 − ν2
(σx − Ex )
σy =
ν
Which gives the stresses as,

208 × 109 (5 × 10−4 + 0.3 × 7 × 10−4 )

σx = = 162.3M N/m2
1 − 0.32
162.3 × 106 − 208 × 109 (5 × 10−4 )
σy = = 194.3M N/m2
0.3

(b) For biaxial stress system where σz = 0 equations 2.30 becomes,

1 
z = − ν(σx + σy )
E
which gives strain through the plate thickness as,
1 
6 6

z = − 0.3(162.3 × 10 + 194.3 × 10 ) = −5.143 × 104
208 × 109

Assignment 2

1. A tensile test has been carried out on a mild steel specimen 10mm thick and 50mm
wide rectangular cross section. An extensometer was attached over a 100mm gauge
length and load extension readings were obtained as follows:

Load 16 32 64 96 128 136 144 152 158

(kN)
Extension 0.016 0.032 0.064 0.096 0.128 0.137 0.147 0.173 0.605
(mm)
Load 154 168 208 222 226 216 192 185.4
(kN)
Extension 1.181 2.42 7.25 12.0 16.8 22.0 24.0 fracture
(mm)

Plot load-extension diagram for specimen and determine: (a) Young’s Modulus
(b) Proportional limit stress (c) Yield point stress (d) Ultimate tensile strength (e)
Percentage elongation.

22
 2. Figure below shows a round steel rod supported in a recess and surrounded by a
coaxial brass tube. The upper end of the rod is 0.1mm below that of the tube and
an axial load is applied to a rigid plate resting on the top of the tube.
(i) Determine the magnitude of the maximum permissible load if the compressive
stress in the rod is not to exceed 110M N/m2 and that in the tube is not to
exceed 80M N/m2 .
(ii) Find the amount by which the tube will be shortened if the compressive stress
in the tube is the same as that in the rod.
Esteel = 200GN/m2 and Ebrass = 100GN/m2

P
0.1mm

30mm

45mm
300mm

Steel 50mm Brass

100mm

2.11 Tutorial 1
Q1 A square steel bar of 50mm side and 1m long is subjected to an axial force of 250kN .
Determine the decrease in lateral dimensions due to this load given E = 200GP a
and ν = 0.3. (AN S : 7.5 × 10−3 mm)
Q2 Show that the volumetric strain in an element subjected to a triaxial state of stress
is give by,
1 − 2ν  
v = σx + σy + σz
E
Q3 Determine the shear strain energy stored in a torsion tube subjected to a state
of constant shear stress of 80M N/m2 throughout the wall thickness. Use G =
85GN/m2 . (AN S : 37.6KN m2 /m3 )
Q4 A piston of hydraulic ram is 40cm diameter and the piston rod has a diameter of
6cm. If the water pressure is 1M N/m2 determine, (E = 200GN/m2 )
(a) The stress in the piston rod. (AN S : 43.5M N/m2 )
(b) The elongation of a 1m length of the piston rod. (AN S : 2.18 × 10−4 m)
when the ram is under pressure from the piston rod side.

23
Q5 A straight steel bar 1m long and of uniform cross section rotates at 2500rev/min
about an axis at mid-length perpendicular to the length of the bar. Find the
maximum stress and total extension if E = 200GN/m2 and steel density is 7.8 ×
103 Kg/m3 .

Q6 The maximum safe compressive stress in a hardened steel punch is limited to

1GN/m2 and the punch is used to pierce circular holes in mild steel 20mm thick.

(a) If the ultimate shearing stress of the plate is 300M N/m2 , calculate the small-
est diameter of the hole that can be pierced. (AN S : 24mm)
(b) If the effective length of the punch is 75mm, calculate the maximum strain
energy stored in the punch during the piercing operation. Assume the modulus
of elasticity for the material of punch to be 200GN/m2 . (AN S : 84.8J)

Q7 A flat steel bar 10m long and 10mm thick tapers from 60mm at one end to 20mm
at the other. Determine the change in length of the bar when a tensile force
P = 12kN is acting along its axis if E = 200GN/m2 . (AN S : 1.648mm)

Q8 Calculate the overall change in length of the tapered rod shown in figure below. It
carries a tensile load of 10kN at the free end, and at the step change in section a
compressive load os 2M N/m evenly distributed around a circle of 30mm diameter.
Assume E = 208GP a. (AN S : −0.035mm).

70mm

35mm

30mm
600mm
24mm

2MN / m
600mm

12mm

10kN

24
Chapter 3

Analysis of Structures

3.1 Introduction

Structural mechanics is generally concerned with analyzing the effects of applied loads
on a structure. The applied loads are external to the material of a structure and result
in internal reacting forces, together with deformations and displacements.
Principles of statics which involves the concept of equilibrium of forces is essential.

3.1.1 Overview of statics

A particle is in state of equilibrium if the resultant force and moment acting on it are
zero, and therefore according to Newton’s law of motion, it will have no acceleration and
will be at rest.
Therefore a structure is in equilibrium if the resultant of external forces and moments are
zero. This can be expressed in the following equations which are necessary and sufficient
conditions for equilibrium of a body,
X
Fx = 0
X
Fy = 0
X
Fz = 0
X
Mx = 0
X
My = 0
X
Mz = 0

where Fx , Fy and Fz represent the components of force vectors in the x, y and z directions
respectively. Also Mx , My and Mz are the components of moment vectors caused by the
external forces acting in the x, y and z directions respectively.
Forces and moments are vector quantities and may be resolved into components. Con-
sider firstly a 2-D case,

25
 F
Fy

θ
Fx

Figure 3.1:

The force F can be resolved into two components along the x and y directions as,

Fx = F cos θ
Fy = F sin θ

For a three dimensional case, the force and moment can be resolved as,

y y

Fy My
F M
α α Mx
θ Fx
Mz θ
Fz x x
β β

z z

Figure 3.2:

Where,

Fx = F cos θ
Fy = F sin α
Fx = F cos β
Mx = M cos θ
My = M sin α
Mx = M cos β

Any system of forces can be replaced by a resultant through an a couple about any
chosen point.
Where
X
F̄x = Fx

26
 y

x
z

FR
−
Fx
−

A M
−
Fy

Figure 3.3:

X
F̄y = Fy
q
FR = F̄x2 + F̄y2
X
M̄ = Mz

3.1.2 Types of structural and solid body components

1. Tie: A slender member which prevents two parts of a structure from moving apart
and it is under tensile force.

2. Strut: A slender member which prevents two parts of a structure from moving
towards each other and is under compressive force.

3. Column: A vertical member which supports some of the mass of the structure.

4. Cable: Flexible string under tension which connects two bodies. It can not supply
resistance to bending action.

5. Beam: A member supported horizontally and carries transverse loading. A com-

mon special case is termed as a cantilever where one end is fixed and provides all
the necessary support.

6. Beam-column: Combines the separate functions of a beam and a column.

7. Shaft: A member which transmits torque and twist.

The members described above can have a variety of cross-sectional shapes, some of which
are illustrated in figure 3.5,

3.1.3 Types of support for structural components

The applied loading on a framework, beam or column is transmitted to the supports will
provide the required reacting forces to maintain overall equilibrium. Common supports
are shown in figure 3.6,

27
 Figure 3.4: Structural components

Figure 3.5: Common cross-sectional shapes for structural members

3.2 Pin-jointed frames

3.2.1 Statical determinacy

Frames can be categorized as;

1. Statically determinate (Just-stiff ): If the number of equations available from state

of equilibrium is equal to the number of unknown forces including support reactions.

2. Statically indeterminate (Over-stiff ): If the number of equations available from

state of equilibrium is less than the number of unknown forces including support

28
 A
A M
Rx

Ry

Rx

Ry

Ry

Figure 3.6: Common supports for structural components

reactions. In this case, there exist one or more members which should be removed
to make the frame just-stiff.

3. Under-stiff: If there are more equilibrium equations than the number of unknown
forces including support reactions. The system is not a structure but a mechanism,
and is not of practical interest in the preceding analysis. In this case, there exist
one or more members which should be added to make the frame just-stiff.

This can be expressed mathematically as follows; Let

j be the number of joints including support points in a frame.

m be the number of members on the frame.

r be the number of support reactions on the frame.

Then for,

Under-stiff frame: m + r < 2j

Just-stiff frame: m + r = 2j

Over-stiff frame: m + r > 2j

29
Examples 3.1
Determine the class of each of the frames below and state which members are required
to be added or removed to make the structure just-stiff.
Solution: (a) Number of members; m = 16 and number of support reactions; r = 3.
Therefore, m + r = 16 + 3 = 19. Number of joints; j = 10, therefore, 2j = 20. The
frame is under-stiff since m + r < 2j. It can be made statically determinate by adding
one member.
(b) Number of members; m = 18 and number of support reactions; r = 3. Therefore,
m + r = 18 + 3 = 21. Number of joints; j = 11, therefore, 2j = 22. The frame
is under-stiff since m + r < 2j. It can be made statically determinate by adding one
member.
(c) Number of members; m = 18 and number of support reactions; r = 3. Therefore,
m+r = 18+3 = 21. Number of joints; j = 10, therefore, 2j = 20. The frame is over-stiff
since m + r < 2j. It can be made statically determinate by removing one member from
the central hexagon.

3.2.2 Determination of axial forces in a frame

The following assumptions are made when solving for axial forces in frame members,

• Members are pin-jointed, hence they carry only axial forces and these forces may
be determined by considering the equilibrium of various parts of the structure as
free bodies.

• The frame is loaded only at the joints.

• The frame is perfect, that is, the frame has got just enough members to ensure
that it remains in equilibrium and doesn’t distort when loaded.

30
 • The weight of the frame members is negligible unless otherwise stated.

• The frame is assumed to be plane, that is, it lies in a xy plane,

There are three methods commonly used to determine the axial forces in frame members,

• Method of resolving the forces at joint, also called the method of joints.

• Graphical method

• Section method

We shall use the method of force resolution in the preceding analysis.

Method of joints

This method is used to determine the internal forces in members of statically determinate
frames only.
It involves isolating each joint of the frame in the form of free body diagram and then by
considering equilibrium equations in each of these joints, the axial forces in the members
of the frame are determined.
When applying this method, it is assumed initially that all the members are in tension.
The sign convention for the frame members is as shown below,

Figure 3.7: Sign convention

Examples 3.2
Using the method of resolution of forces at the joints, determine the axial forces of the
members in a plane pin-jointed frames shown below,
Solution
(a) Drawing a free body diagram of the whole frame,

31
 (a ) P
(b)
B B D

2m 2m
2m 2m

C F
A C A
2m 2m 2m E 2m

5kN

B B D

2m 2m
2m

A C
R Ax C A
2m 2m 2m E
R Ay RCy R Ay 5kN

Applying the principle of static equilibrium on the structure,

X
Fx = 0
=⇒ RAx = 0
X
Fy = 0
=⇒ RAy + RCy = P (3.1)

Taking moments about point A, and assuming moment in clockwise direction to be

positive, we have
X
MA = 0
−2RCy + P = 0
P
=⇒ RCy = (3.2)
2
Substituting equation 3.2 in equation 3.1, we get,
RAy = P − RCy
P
= P−
2
P
= (3.3)
2
32
Drawing a free body diagram at joint A,

Applying the principle of static equilibrium,

X
Fy = 0
FAB sin 600 + RAy = 0
−RAy −2RAy
=⇒ FAB = 0
= √
sin 60 3
 
P
−2 2 −P
= √ = √
3 3

X
Fx = 0
0
FAB cos 60 + FAC = 0
=⇒ FAC = −FAB cos 600
h −P  1 i P
= − √ = √
3 2 2 3

Drawing a free body diagram at joint B,

P
0
FAB cos 60 FBC cos 60 0

B 60 0
60 0 FBC sin 60 0
FAB sin 600

FBC
FAB

X
Fx = 0
FAB cos 60 − FBC cos 600 = 0
0

=⇒ FAB cos 600 = FBC cos 600

FBC = FAB
−P
= √
3

33
Therefore the axial force in member,

P
AB is √
3
(Compressive)
P
BC is √
3
(Compressive)
P
AC is √
2 3
(Tensile)

The values can be proved by applying static equilibrium equations at joint C, that is,

FBC cos 60 0

FBC
FBC sin 60 0
60 0

FAC
RCy

X
Fy = 0
FBC sin 600 + RCy = 0
√
−P  3  P
√ + = 0
3 2 2
=⇒ 0 = 0

X
Fx = 0
FBC cos 600 + FAC = 0
−P  1  P
√ + √ = 0
3 2 2 3
=⇒ 0 = 0

(b) Drawing a free body diagram of the whole frame,

Applying the principle of static equilibrium on the whole frame,
X
Fx = 0
=⇒ RF x = 0
X
Fy = 0
=⇒ RAy + RF y = 5000 (3.4)

34
 P

B B D

2m 2m
2m 2m

C F
C A RFx
2m 2m 2m E 2m
RCy R Ay 5kN RFy

Taking moments about point F , and assuming moment in clockwise direction to be

positive, we have
X
MF = 0
6RAy − 5000 × 4 = 0
5000 × 4
=⇒ RAy = = 3333.3N (3.5)
6
From equation 3.4, we get,
RF y = 5000 − RAy
= 5000 − 3333.3 = 1666.7N

The only joints at which there are two unknowns are A and F . Lets start at A, and
draw a free body diagram at the joint, Joint
Joint A
FAB cos 450

FAB
0
FAB sin 45
45 0
A FAC

R Ay

For static equilibrium

X
Fy = 0
FAB sin 450 + RAy = 0
−RAy −3333.3
=⇒ FAB = 0
=
sin 45 sin 450
= −4714N

35
 X
Fx = 0
0
FAB cos 45 + FAC = 0  
=⇒ FAC = −FAB cos 450 = − − 4714 cos 450
= 3333.3N

Joint B

FAB sin 450 B

FBD
FAB cos 450
45 0

FAB
FBC

For static equilibrium

X
Fy = 0
0
FAB cos 45 + FBC = 0  
=⇒ FBC = −FAB cos 450 = − − 4714 cos 450
= 3333.3N

X
Fx = 0
FAB sin 450 − FBD = 0  
=⇒ FBD = FAB sin 450 = − 4714 sin 450
= −3333.3N

Joint C
FCD cos 450 FCD
FBC

FCD sin 450

45 0

FAC C FCE

5000N

For static equilibrium

X
Fy = 0

36
 FCD sin 450 + FBC − 5000 = 0
5000 − FBC 5000 − 3333.3
=⇒ FCD = 0
=
sin 45 sin 450
= 2357.1N

X
Fx = 0 CD
FCE + FCD cos 450 − FAC = 0
=⇒ FCE = FAC − FCD cos 450 = 3333.3 − 2357.1 cos 450
= 1666.6N

Joint D
FBD D
FDF sin 450

FDF cos 450

45 0 45 0

FDE

F CD FDF
0
FCD cos 45
0
FCD sin 45

For static equilibrium

X
Fx = 0
FCD sin 450 − FDF sin 450 + FBD = 0
FBD + FCD sin 450 −3333.3 + 2357.1 sin 450
=⇒ FDF = =
sin 450 sin 450
= −2357.1N

X
Fy = 0
FDE + FCD cos 450 + FDF cos 450 = 0
=⇒ FDE = −FCD cos 450 − FDF cos 450
= −2357.1 cos 450 − (−2357.1 cos 450 )
= 0N

Joint E
For static equilibrium
X
Fx = 0
FCE − FEF = 0
=⇒ FEF = FCE
= 1666.6N

37
 FDE

FCE E FEF

Therefore the axial force in member,

AB is 4714N (Compressive)
BC is 3333.3N (Tensile)
AC is 3333.3N (Tensile)
BD is 3333.3N (Compressive)
CD is 2357.1N (Tensile)
CE is 1666.6N (Tensile)
DE is 0
DF is 2357.1N (Compressive)
EF is 1666.6N (Tensile)

3.3 Tutorial 2
Q1 For the plane pin-jointed frames shown below, determine using the method of
joint the support reactions and internal forces in all the frame members indicating
whether the forces are tensile or compressive.

5kN

(a ) (b) D 3kN
B 2m C 2m E
6kN
B
2m E
20kN

A D A F
1m C
10kN

Q2 The framework shown in the figure below is used to support a steel car body weigh-
ing 200Kg. When the car body is suspended in (a) air and (b) totally immersed
in a plating bath containing a liquid of density 1000Kg/m3 , calculate the support
reactions and the internal forces in all the members using the method of force
resolution at the joints. Density of steel is 7.8 × 103 Kg/m2 .

38
 A 3m B 3m C 3m D

6m E

39
Chapter 4

Analysis of thin-walled pressure

vessels

Thin-walled pressure vessels are those vessels whose wall thickness is less than or equal
to one tenth of the vessel radius. The distribution of stress across the vessel thickness in
this case is assumed to be uniform.
They are made of metal and used generally in engineering as storage containers for liquid,
powder, gas etc. Stresses will arise due to, say uniform internal liquid or gas pressure,
eg. in steam boilers, or pressure due to weight of substances contained.

4.1 Stresses in a thin-walled cylinders

When a thin cylinder is subjected subjected to an internal pressure, stresses are induced,

• on the longitudinal section XX due to the force tending to separate the top and
bottom halves. This stress is called the circumferential or hoop stress

• on the circumferential section Y Y due to the force tending to separate the right
and left ends of the cylinder. This stress is called the longitudinal stress.

Let the internal diameter and length of the cylinder be d and l respectively, the thickness
be t and the internal pressure be p.

4.1.1 Circumferential (Hoop) stress

Consider figure below,

The force tending to separate the top and bottom halves is the pressure multiplied by the
projected area in a direction perpendicular to the diametral plane, that is PC = p × dl.

40
 Y

σL

Y
X X σL σθ σL
σθ

σθ

Figure 4.1:

p l

X
X σθ
t d

Figure 4.2:

This force is resisted by the stress acting in the longitudinal section XX (σθ ). Therefore,

p × dl = (σθ × tl) × 2
pd pr
σθ = = (4.1)
2t t

4.1.2 Longitudinal stress

Consider figure below,

The force tending to separate the right and left hand halves is the pressure times the
2
area of one end, that is PL = p × πd4 . This force is resisted by the stress acting on the
circumferential section Y Y (σl ). Therefore,

πd2
p× = σl × πdt
4

41
 Y

PL
d p

t
Y

Figure 4.3:

pd pr
σl = = (4.2)
4t 2t

4.1.3 Circumferential and Longitudinal strains in a thin walled

cylinder

The hoop and longitudinal stresses can be considered as bi-axial stress system. Therefore,
Hoop strain (θ );
1
θ = (σθ − νσl )
E
1  pr pr 
= −ν
E t 2t
pr ν
= 1− (4.3)
Et 2

Longitudinal strain (l );

1
l = (σl − νσθ )
E
1  pr pr 
= −ν
E 2t t
pr  1 
= −ν (4.4)
Et 2

4.1.4 Volumetric strain in a thin-walled cylinder

When a thin-walled cylinder is exposed to internal pressure, it will experience a change

in dimension θ r and l l in the circumferential and longitudinal directions respectively.
Let the initial volume of the cylinder be Vo = πr2 l, and the final volume of the cylinder
be Vf = π(r + θ r)2 (l + l l). Therefore the volumetric strain (v ) is,

Vf − Vo
v =
Vo

42
 π(r + θ r)2 (l + l l) − πr2 l
=
πr2 l
πr l(1 + θ )2 (1 + l ) − πr2 l
2
=
πr2 l
2
= (1 + θ ) (1 + l ) − 1
= (1 + 2θ + 2θ )(1 + l ) − 1
= 1 + l + 2θ + 2θ l + 2θ + 2θ l − 1
= l + 2θ
pr  1  2pr  ν
= −ν + 1−
Et h 2 Et i 2
pr
= 0.5 − ν + 2 − ν
Et h
pr i
= 2.5 − 2ν (4.5)
Et

4.2 Stresses in a thin-walled sphere

Let the internal diameter be d, the thickness of metal be t and the internal pressure be p.
The force tending to separate the two halves on a section XX is the pressure multiplied
2
by the projected area in the direction perpendicular to XX, that is P = p × πd4

σθ

σθ
σθ
X d t X
σθ σθ
σθ
p

Figure 4.4:

Therefore,

πd2
p× = σθ × πdt
4
pd pr
σθ = = (4.6)
4t 2t

43
From the symmetry of the spherical shape, we obtain the same equation for the tensile
stresses when we cut though the center of the sphere in any direction whatsoever. Thus
we reach the following conclusion: A pressurized sphere is subjected to uniform tensile
stresses σθ in all directions. This is represented in figure 4.4(b) by the small stress
element with stresses σθ acting in mutually perpendicular directions.
The normal strain in a thin walled sphere is,
1
θ = (σθ − νσθ )
E
σθ
= (1 − ν)
E
pr
= (1 − ν) (4.7)
2Et

4.2.1 Volumetric strain in a thin-walled sphere

On application of internal pressure, the sphere radius will change by θ r.

Let the initial volume of the sphere be Vo = 43 πr3 , and the final volume of the sphere be
Vf = 34 π(r + θ r)3 . Therefore the volumetric strain (v ) is,
Vf − Vo
v =
Vo
4
π(r + θ r)3 − 34 πr3
= 3 4
3
πr3
= (1 + θ )3 − 1
= 1 + 3θ + 32θ + 3θ
= 3θ
3pr
= (1 − ν) (4.8)
2tE

Examples 4.1
A compressed air tank having an inner diameter of 450mm and a wall thickness of 7mm
is formed by welding two steel hemispheres.

(a) If the allowable tensile stress in the steel is 115M P a, what is the maximum permis-
sible air pressure pall in the tank?
(b) If the normal strain at the outer surface of the tank is not to exceed 0.0003 what is
the maximum permissible air pressure pall in the tank? Assume that Hooke’s law
is valid and that E = 210GP a and Poisson’s ratio for steel is ν = 0.28.

Solution: (a) From equation 4.6, the allowable pressure is,

2tσall
pall =
r
2 × 0.007 × 115 × 106
= = 7.16 × 106 P a
0.225
44
(b) From equation 4.7, the allowable pressure is,
2tEall
pall =
r(1 − ν)
2 × 0.007 × 210 × 109 × 0.0003
= = 5.44 × 106 P a
0.225(1 − 0.28)

Examples 4.2
A strain gauge is installed in the longitudinal direction on the surface of a cylindrical
aluminium beer can. The radius-to-thickness ratio of the can is 200. When the lid of
the can is popped open, the strain changes by  = 170µm/m. What was the internal
pressure p in the can? Assume E = 70GP a and ν = 0.33.
Solution: The longitudinal strain for a thin walled cylinder is given by equation 4.4,
therefore,
EtL
p =
r(1 − ν)
70 × 109 × 170 × 10−6
= = 0.35 × 106 P a
200(1 − 0.33)

Examples 4.3
A thin spherical steel vessel is made up of two hemispherical portions bolted together
at flanges. The inner diameter of the sphere is 300mm and the wall thickness is 6mm.
Assuming that the vessel is a homogeneous sphere, calculate the maximum working
pressure for an allowable tensile stress in the shell of 150M P a. If twenty bolts of 16mm
diameter are used to hold the flanges together, what is the tensile stress in each bolt
when the sphere is under full pressure.
Solution: From equation 4.6, the allowable working pressure is,
2tσall
pall =
r
2 × 0.006 × 150 × 106
= = 12.0 × 106 P a
0.15
The force tending to separate the two halves is,
πd2
F = p×
4
12 × 106 × π × 0.32
= = 848.23kN
4
Force per bolt,
848.23
Fb = = 42.41kN
20
Tensile stress per bolt,
42.41 × 103
σb = π 2
= 210.9 × 106 P a
4
× 0.016

45
Chapter 5

Torsion

One of the common modes of deformation in engineering is that of torsion in which a

member is exposed to a torque about its longitudinal axis resulting in twisting deforma-
tion.
A design analysis in twisting deformation is required in order to estimate the shear stress
distribution and angular twist deformations.
When a shaft, either solid or hollow is subjected to pure torque (not accompanied by
bending or axial thrust) loading at its end(s), it experiences pure twisting and thus pure
shear stress.
The following assumptions are made when developing stress and strain solutions for a
shaft under pure torque loading,

• The shaft is straight.

• The cross sectional area is constant over the shaft length.

• Plane sections remain plane even after twisting.

• Radii remain straight during twisting.

• Torque is constant over the shaft length.

• Deformation is by relative rotation of adjoining planes.

• The material is homogeneous, elastic and obeys Hooke’s law, ie. the shear stress
at any point is proportional to the shear strain at that point

5.1 Shear stress due to twisting

The shearing stress induced in the shaft produces a moment of resistance, equal and
opposite to the applied torque.

46
 r dA
x

Figure 5.1:

Since radii are assumed to remain straight, it follows that the strain (and hence the
stress) is directly proportional to the radius.
Thus if the stress at the surface of the shaft of radius r is τ , then shear stress on an
element of area dA at a distance x from the axis is xr τ . The shear force on the element
is,
x
dF = τ dA
r

Moment of the shear force on the element about the center of shaft is,
x
dM = τ dA × x
r

Therefore total moment of resistance is,

τZ 2
M = x dA
r
This moment is equal to the applied torque, T ,
τZ 2
T = x dA
r
x2 dA = J where J is the second polar moment of area of the cross section. Hence
R
But
τ
T = J
r
or
T τ
= (5.1)
J r

5.2 Angle of twist

Consider figure below,

Due to shear strain in the shaft, the longitudinal element ABCD is distorted into par-
allelogram AB 0 CD0 . The angle BAB 0 is the shear strain (φ), and the angle BOB 0 is the
angle of twist (θ).

47
 C
A
φ

D'
D
B'

O
θ B
L
r
T

Figure 5.2:

Length of arc BB 0 is,

BB 0 = rθ = φL

Therefore,
φL
θ =
r
τ
But φ = G
, therefore,

τL
θ =
Gr
Gθ τ
= (5.2)
L r
Which when combined with equation 5.1, we get the general twisting formula as,
Gθ τ T
= = (5.3)
L r J

5.3 Torsion of a solid shaft

In case of a solid shaft, shearing stresses are variable over the cross section of the shaft,
with τmax occurring at rmax = R. The second polar moment of area of a circular cross
section about an axis through the center is,

πR4
J =
2

48
 Figure 5.3:

Therefore equation 5.3 becomes,

Gθ τmax T
= = πR4
L R 2
πR4 τmax πR3
T = = τmax
2 R 2
πD3
= τmax (5.4)
16

5.4 Torsion of a hollow shaft

Ro

Ri
L

Figure 5.4:

In case of a hollow shaft τmax occurs at rmax = Ro . The second polar moment of area of

49
a circular cross section for a hollow shaft about an axis through the center is,

π(Ro4 − Ri4 )
J =
2
Therefore equation 5.3 becomes,
Gθ τmax T
= = π(Ro4 −Ri4 )
L Ro
2
π(Ro4 − Ri4 ) τmax
T =
2 Ro
4h 4i
πRo Ri τmax
= 1−
2 Ro4 Ro
πRo3 h Ri4 i
= 1− τmax (5.5)
2 Ro4

Examples 5.1
A solid shaft is required to transmit 50kW at 4rev/s. If the maximum allowable shear
stress is 70M P a, and the maximum allowable shaft twist is 0.50 /m length, calculate the
size of the shaft. Assume G = 100GP a.
Solution: Torque in the shaft,

P 50 × 103
T = = = 1989.4N m
ω 2π × 4

Shaft size based on maximum shear stress allowable,

πD3
T = τmax
16
16T 16 × 1989.4
D3 = =
πτmax π × 70 × 106
D = 0.0525m = 52.5mm

Shaft size based on maximum allowable angle of twist,

T Gθ
=
J L
θ
T = GJ
L
But
θ 0.5
= πrad/m
L 180
and
πD4
J =
32

50
Therefore,
θ πD4
T = G
L 32
32T 32 × 1989.4
D4 = θ = 0.5 2
GLπ 100 × 109 × 180 π
D = 0.0694m = 69.4mm

Since the shaft size under the twist criterion is large, then this is taken as the appropriate
size, ie. D = 69.4mm.

5.5 Torsion in composite shafts

A composite shaft is one which consists of two or more materials rigidly fixed together
so that the share the applied torque. If T1 and T2 are the parts of the applied toque T

Shaft 2

Shaft 1

Figure 5.5:

carried by the two materials, then T1 + T2 = T .

The angle of twist of the two parts is the same, that is θ1 = θ2 . Recall that,
Gθ T
=
L J
θ T
=
L GJ
θ1 θ2
And since the length of both parts are equal, then L1
= L2
, therefore,
T1 T2
= (5.6)
G1 J1 G2 J2

5.6 Torsion in non-uniform shafts

In this case the total torque is transmitted by each portion of the shaft, that is,
T1 = T2 = T

51
 T1
T2

T
D1 D2
T

L1 L2

Figure 5.6:

Therefore,
τ1 πr14 τ2 πr24
=
2r1 2r2
τ1  r 3
2
= (5.7)
τ2 r1

τ1 L1 τ2 L2
Angle of twists, θ1 = Gr1
, and θ2 = Gr2
.

θ1
L1 τ1 Gr2
θ2 = ×
L2
Gr1 τ2
τ1 r2  r2 3 r2
= =
τ2 r1 r1 r1
 r 4
2
= (5.8)
r1

The total deformation θ is due to θ1 over length L1 plus θ2 over length L2 . Therefore,
θ = θ1 + θ2
τ 1 L1 τ 2 L2
= +
Gr1 Gr2
1  τ1 L1 τ2 L2 
= + (5.9)
G r1 r2
Or in terms of torque, ( Gθ
L
= TJ ),
θ = θ1 + θ2
T L1 T L2
= +
GJ1 GJ2
T L1 L2 

= + (5.10)
G J1 J2

5.7 Torsion of a tapered shaft

Suppose a twisting moment T is applied to a tapered shaft of length L

52
 L
C
L1
X
A

T d1 d d2

O T

B
Y
x
dx D

Figure 5.7:

Let

τ1 = τmax at the end of radius r1 .

τ2 = τmax at the end of radius r2 .
τ = τmax at a cross section of radius r.

The resisting moment in all cross sections of the shaft must be equal to the applied
torque T . Then,

T1 = T2 = Tr = T
τ1 πr13 τ2 πr23 τ πr3
= = =T
2 2 2
τ1 r13 = τ2 r23 = τ r3 (5.11)

Let dx be the length of the small element at distance x from point O, and having a
radius r, and angle of twist dθ. We know that,
Gθ T 2T
= = 4
L J πr
x(r2 −r1 )
For the element θ = dθ, L = dx, and using similarity on the triangles r = L
,
therefore
dθ 2T L4
=
dx Gπx4 (r2 − r1 )4
2T L4
dθ = dx
Gπx4 (r2 − r1 )4
The total angle of twist for the entire length L is given by,
Z L+L1
θ = dθ
L1

53
 Z L+L1 h
2T L4 dx i
=
L1 Gπ(r2 − r1 )4 x4
2T L4 h 1 iL+L1
= −
Gπ(r2 − r1 )4 3x3 L1
2T L4 h 1 1i
= − +
3Gπ(r2 − r1 )4 (L + L1 )3 L3
Lr1 Lr2
But L1 = r2 −r1
and L + L1 = r2 −r1
which when substituted in equation just above, we
get,
2T L h r3 − r3 i
2 1
θ =
3Gπ(r2 − r1 ) r13 r23
2T L h r12 + r1 r2 + r22 i
= (5.12)
Gπ 3r1 r2

Examples 5.2
A composite shaft of circular cross section O.5m long is rigidly fixed at each end as shown
in the figure below. A 0.3m length is 50mm diameter and is made of bronze to which is
joined the remaining 0.2m length a 25mm diameter made of steel. If the limiting shear
stress in the steel is 55M N/m2 , determine the maximum torque that can be applied
at the joint. What is then the maximum shear stress in the bronze. (Gsteel = 82GP a,
Gbronze = 41GP a).

Bronze
Steel
T

0.3m 0.2m

Figure 5.8:

Solution: Since the torque is applied at the joint, then it is shared between the two parts
of the shaft, ie. T = Ts + Tb . Maximum torque that can be carried by the steel part is,
Ts τs
=
Js rs
τs πrs3 55 × 106 × π × 0.01253
Ts = = = 168.7N m
2 2

The angle of the twists for the two shaft parts are equal at the joint, ie. θs = θb . The
angle of twist for the steel part is,
Gs θs τs
=
Ls rs

54
 τs L s 55 × 106 × 0.2
θs = = = 0.0107rad
Gs rs 82 × 109 × 0.0125

Therefore θb = 0.0107rad, hence the torque carried by the bronze part is,
Gb θb Tb
=
Lb Jb
Gb θb Jb Gb θb πrb4
Tb = =
Lb 2Lb
41 × 109 × 0.0107 × π × 0.0254
= = 897.3N m
2 × 0.3
Therefore the total torque that can be applied at the joint is
T = Ts + Tb
= 168.7 + 897.3 = 1066N m

Maximum shear stress in the bronze part is,

τb Tb
=
rb Jb
Tb rb 2Tb
τb = = 3
Jb πrb
2 × 897.3
= = 36.6M P a
π × 0.0253

5.8 Strain energy in torsion

In pure torsion strain energy is given by;

1
U = Tθ
2
Gθ T
But from the general torsion formula L
= J
= τr , we have;
τJ
T =
r
τL
θ =
Gr
Therefore;
1 τJ τL τ2 JL
U = × × = × 2
2 r Gr 2G r

For a solid circular shaft maximum shear stress occurs at the surface of the shaft, that
4
is r = d2 and J = πd
32
. Therefore;
2
τmax πd4 L 2
τmax πd2 L 2
τmax πd2 L
U = × 2 = × = ×
2G 32 d4 2G 8 4G 4
τ2
= max × V olume (5.13)
4G

55
5.9 Tutorial 3
Q1 A shaft tapers from a radius r + a at one end toa radius r − a at the other end.
If the shaft is under an axial torque T and a = 0.1r, determine the percentage
error in the angle of twist for a given length, when the twist is calculated on the
assumption of a constant radius r. (Ans : 3.30%)

Q2 A hollow shaft is 50mm outside diameter and 30mm inside diameter. An applied
torque of 1.6kN m is found to produce an angular twist of 0.40 measured on a
length of 0.2m of the shaft. Calculate the value of modulus of rigidity. Calculate
also the maximum power which could be transmitted by the shaft at 200rev/min
if the maximum allowable stress is 65M N/m2 .(Ans : 65M N/m2 and 292kW )

Q3 A hollow shaft is to have an internal diameter di × outside diameter, and is required

to transmit 1500kW at a speed of 2500rev/min. It is to be fitted with a flanged
coupling having 8 bolts on a circle of diameter twice that of the shaft. Assuming
allowable shearing stresses in shaft and bolts of 70M P a and 55M P a respectively,
determine the necessary diameters. (Ans : 46.9mm, 78.2 and 14.54mm)

Q4 A steel shaft ABCD has a total length of 1.3m, made up as follows: AB = 0.3m,
BC = 0.4m and CD = 0.6m. AB is hollow, its outside diameter being 100mm and
its inside diameter dmm. BC and CD are solid, having diameters of 100mm and
88mm respectively. If equal opposite torques are applied to the ends of the shaft,
find to the nearest 1mm, the maximum permissible value of d for the maximum
shearing stress in AB not to exceed that in CD. If the torque applied to the shaft
is 9kN m, what is the total angle of twist? G = 82GP a (Ans : 75mm, 1.1480 )

Q5 A steel shaft ABCD having a total length of 1.5m is made of three lengths AB,
BCa nd CD each 0.5m long. AB and BC are solid, having diameters of 60mm
and 70mm respectively, and CD is hollow having outside and inside diameters
of 70mm and 45mm respectively. When an axial torque of 2kN m is transmitted
from one end of the shaft to the other, the total angle of twist from A to D is 1.20 .
Determine,

(a) the maximum shearing stress in the shaft and state where it occurs. (Ans :
49.4M N/m2 in AB)
(b) the angle of twist for each of the three lengths AB, BC and CD. (Ans : 0.5480 ,
0.2960 and 0.3560 )
(c) the modulus of rigidity of the material. (Ans : 82.2GP a)

Q6 The maximum safe compressive stress in a hardened steel punch is limited to

1GN/m2 and the punch is used to pierce circular holes in mild steel 20mm thick.

(a) If the ultimate shearing stress of the plate is 300M N/m2 , calculate the small-
est diameter of the hole that can be pierced. (AN S : 24mm)
(b) If the effective length of the punch is 75mm, calculate the maximum strain
energy stored in the punch during the piercing operation. Assume the modulus
of elasticity for the material of punch to be 200GN/m2 . (AN S : 84.8J)

56
Q7 A composite shaft consists of a steel rod 80mm diameter surrounded by a closely
fitting tube of brass firmly fixed to it. Find the outside diameter of the tube so
that when a torque is applied to the composite shaft, it will be equally shared by
the two materials. If the torque is 16kN m, calculate the maximum shearing stress
in each material and the angle of twist on a length of 3.5m. Gbrass = 40GP a and
Gsteel = 80GP a.(Ans : 105.3mm, 79.6M P a and 52.3M P a)

Q8 A round steel rod is surrounded by a close fitting tube of duralumin, the two being
secured fastened together to form a composite shaft. Find the diameter of the steel
rod and the outside diameter of the duralumin tube so that the maximum shearing
stresses in the two materials do not exceed 90M P a and 60M P a respectively when
the composite shaft is subjected to a torque of 0.7kN m. Also calculate the angle
of twist on a length of 1.2m. Gduralumin = 26GP a and Gsteel = 78GP a. (Ans :
18.76mm, 37.52mm and 8.460 )

Q9 A hollow steel shaft with an external diameter of 150mm is required to transmit

1M W at 300rev/min. Calculate a suitable internal diameter for the shaft if its
shear stress is not to exceed 70M P a. Compare the torque-carrying capacity of this
shaft with a solid steel shaft having the same weight per unit length and limiting
shear stress. Take G = 80GP a. (Ans : 112.3mm, and 2.35)

57
Chapter 6

Analysis of Close-coiled Springs

6.1 Introduction

Springs are directly concerned with theories of torsion, bending and strain energy.
Springs are incorporated in almost all machines to assist in their operation.
The principle function of a spring is to absorb energy , store it for a long or short period
of time, and then release it to the surrounding material. For example, springs in a watch
store energy for a long time, while springs in engine valves store energy for a very short
time.
The force required to produce a unit deformation is called the stiffness of the spring.
Helical springs are widely used in engineering and are manufactured in two categories;

• Close-coiled springs - The helix angle is very small and the coils almost touch each.
These are tension helical springs and are designed to resist stretching.

• Open-coiled springs - The helix angle is large and coils are spaced further apart.
These are compression helical springs and are designed to resist being compressed.
A typical use for compression coil springs is in car suspension systems.

NOTE: For more information on springs, a supplement reading material will be provided.

6.2 Axial load on a close coiled spring

Since the obliquity of the wire of a close coiled spring is small, then such a spring under
axial load may be assumed to be subjected to torsion only, the effects of bending and
direct shear being negligible.
Figure ?? shows a close-coiled spring under an axial load W .
Let;

58
 D be the mean diameter of the spring coil

d be the spring wire diameter

N be the number of coils making the spring

L be the length of the wire and is given by L = πDN

δ be the axial deflection of the spring.

The torque on the wire at all sections is;

WD
T =
2

Maximum Shear Stress

The general torsion formular is given as;

T τ
=
J r

The maximum shear stress on the wire material occurs at the surface of the wire, that
4
is r = d2 . Since the wire is of circular cross section then, J = πd
32
. The maximum shear
stress on the wire material is thus;
WD d
2 2
. 8W D
τmax = πd4
= (6.1)
32
πd3

Axial deflection of the spring

Equating the work done by the load to the shear strain energy, we have;
1 1
Wδ = Tθ
2 2
But from general torsion formula, θ can be expressed as;
TL
θ =
GJ
Hence;
1 T 2L
Wδ =
2 2GJ
1  W D 2 πDN 4W 2 D3 N
= 4 =
2 2 G. πd
32
Gd4

Hence the spring axial deflection can be expressed as;

8W D3 N
δ = (6.2)
Gd4

59
Strain energy

The strain energy is given by;

1 4W 2 D3 N
U = Tθ =
2 Gd4
This can be written as;
 8W D 2 π 2 DN d2
U = ×
πd3 16G
2
τmax πd2
= × πDN ×
4G 4
2 2
τ πd L τ2
= max × = max × V olume (6.3)
4G 4 4G
This corresponds with the strain energy in pure torsion of a solid shaft as represented in
equation 5.13

Example 6.1

A close coiled spring is to have a stiffness of 1kN/m in tension, a maximum load of 50N
and a maximum shearing stress of 120M P a. The solid length of the wire (that is, when
the coils are touching) is to be 45mm. Find the diameter of the wire, the mean diameter
of the coils and the number of the coils required. Take G = 50GP a.
W
Solution: Spring stiffness is defined as the load per unit deflection, that is, k = δ
. From
the deflection equation 6.2, we have,
8W D3 N
δ =
Gd4
W Gd4
k = =
δ 8D3 N
50 × 109 d4
1000 =
8D3 N
4
d
3
= 1.6 × 10−7 (6.4)
D N

The maximum shear stress is given by equation 6.1 as,

8W D
τmax =
πd3
8 × 50D
120 × 106 =
πd3
D
= 942477.8 (6.5)
d3

The solid length according to the definition is;

Ls = nd
N d = 0.045 (6.6)

60
Equations 6.4, 6.5 and 6.6 can be solved simultaneously for d, N and D.
0.045
From equation 6.6 we have; d = N
which when substituted in equations 6.4 and 6.5
leads to;

D3 N 5 = 25.63 (6.7)

and,

DN 3 = 85.88
85.88
D = (6.8)
N3
Substituting equation 6.8 in equation 6.7 we can solve for N .
 85.88 3
N 5 = 25.63
N3
N5 25.63
9
=
N 85.883
N = 12.54

Hence the number of coils are N = 13.

Therefore, the mean coil diameter D and the wire diameter are;
85.88 85.88
D = 3
= = 0.0436m = 43.6mm
N 12.543
0.045 0.045
d = = = 0.0036m = 3.6mm
N 12.54

6.3 Tutorial 4
Q1 A composite spring has two close-coiled helical springs in series. Each spring has
a mean coil diameter of 8 times the diameter of its wire. One spring has 20 coils
and a wire diameter of 2.5mm.

(a) Find the diameter of the wire of the other spring if it has 15 coils and the
stiffness of the composite spring is 1.25kN/m. (Ans : 1.97mm)
(b) Find the greatest axial load that can be applied to the spring and the cor-
responding extension for a maximum shearing stress of 300M P a. (Ans :
57.2N, 0.0458m)

Q2 The spring load against which a valve is opened is provided by an inner helical
spring arranged within and concentric with an outer helical spring. Both springs
are of steel, close-coiled, and the free length of the inner spring is 6mm longer than
that of the outer spring. The outer spring has 12 coils of mean diameter 25mm,
diameter of the wire 3mm and an initial compression of 5mm when the valve is
closed.

(a) Find the stiffness of the inner spring if the greatest force required to open the
valve 10mm is 150N .(Ans : 4018.6N/m)

61
(b) If the diameter of the inner spring is 16mm and the wire diameter is 2mm,
determine the number of coils it has. (Ans : 10Coils)

Take G = 80GP a.

62
Chapter 7

Thermal Stresses and Strains

7.1 Introduction

Consider a bar of length Lo at a temperature To . Let the coefficient of linear expansion

of the material making the bar be α. If the bar is subjected to a temperature increase
of say ∆T and the bar is completely free to expand, then its length will increase by
α∆T × Lo . Hence the new length of the bar after thermal expansion becomes,

Lf = Lo + α∆T × Lo = Lo (1 + α∆T )

If this expansion is prevented then the situation is as if a bar of length Lo (1 + α∆T )

is compressed to a length Lo . Hence compressive strain and consequently, compressive
stresses are induced on the bar. The compressive strain is given as;
α∆T × Lo α∆T
 = =
Lo (1 + α∆T 1 + α∆T

Since α∆T is very small compared to one, then it can be neglected and equation above
becomes;

 = α∆T

and therefore the compressive stress induced on the bar can be approximated as;

σ = E = Eα∆T

7.2 Thermal stresses in composite bars

Consider a composite bar consisting of a solid circular bar denoted as 1 contained inside
a circular tube 2, as shown in figure ??. The materials of the bar and the tube have
different coefficients of linear expansion, denoted as α1 and α2 respectively, and assumed
that α1 > α2 .

63
Let XX be the initial level of the composite bar and let YY and ZZ be the final levels of
the tube and bar respectively after a temperature rise of ∆T .
If both parts are free to expand, the extension of material 1 (the bar) would be;

δ1 = Lo α1 ∆T

and the extension of material 2 (the tube) would be;

δ2 = Lo α2 ∆T

If the ends of the bar and the tube are rigidly attache to each other, the the tube is
forced to extend a distance say e2 and a bar is forced to compress a distance say e1 as
shown in the figure ??.
It is these forced changes in lengths e1 and e2 which produce stresses in the materials.
From figure ??, we have;

δ1 − δ2 = e1 + e2
σ1 Lo σ2 Lo
Lo α1 ∆T − Lo α2 ∆T = +
E1 E2
σ σ2 
1
Lo ∆T (α1 − α2 ) = Lo + (7.1)
E1 E2

Since there is no external force applied to the composite bar, then the tensile force in
the tube must be equal to the compressive force in the bar. If we let this force be P ,
then,

σ2 A2 = σ1 A1 = P
P
⇒ σ2 = (7.2)
A2
P
σ1 = (7.3)
A1
Substituting equations 7.2 and 7.3 in equation 7.1 leads to;
P P 
Lo ∆T (α1 − α2 ) = Lo +
A1 E1 A2 E2
 1 1 
= P Lo +
A1 E1 A2 E2

Therefore the force P which is induced by the temperature change ∆T if the ends of the
two members are rigidly attached to each other is give as;

∆T (α1 − α2 )
P = 1 (7.4)
A1 E1
+ A21E2

This force is the one which induces thermal direct stresses on the materials.

64
Example 7.1

A bar of brass 25mm diameter is enclosed in a steel tube 50mm external diameter and
25mm internal diameter. The bar and tube are initially 1m long and are rigidly fastened
together at both ends. find the stresses in the two materials when the temperature rises
from 150 C to 950 C. Take the Young Modulus for steel and brass to be 200GP a and
100GP a respectively, and the coefficients of linear expansion for steel and brass to be
11.6 × 10−6 /0 C and 18.7 × 10−6 /0 C respectively.
Solution: Let subscript s and b denote the steel and brass material respectively. There-
fore, Es = 200GP a, Eb = 100GP a, αs = 11.6 × 10−6 /0 C and αb = 18.7 × 10−6 /0 C.
The temperature change is; ∆T = 95 − 15 = 800 C. Since the coefficient of linear expan-
sion of brass if higher than that of steel, then the force induce due to the temperature
change is,

∆T (αb − αs )
P = 1
Ab Eb
+ As1Es

But,

π(0.0252 )
Ab Eb = × 100 × 109 = 49.09 × 106 N
4
π(0.052 − 0.0252 )
As Es = × 200 × 109 = 294.52 × 106 N
4
Therefore;

80 × 10−6 (18.7 − 11.6)

P = 1 1 = 23899.5N
49.09×106
+ 294.52×106

Stress on brass bar is;

P 4P 4 × 23899.5
σb = = 2 = = 48.7M P a
Ab πd π × 0.0252

Stress on steel tube is;

P 4P 4 × 23899.5
σs = = 2 2
= = 16.23M P a
As π(do − di ) π(0.052 − 0.0252

65