STUDY NOTE: STRESS – STRAIN ANALYSIS OF SIMPLE STRUCTURES

Please note that this study note is a combination of Prof Tabakov’s notes and sections
taken from the prescribed book: Mechanics of Engineering Materials by PP Benham &
RJ Crawford.
Remember that the test questions will be very similar to those in this Study Note so
go through them carefully!

The analysis of stress

Definition of stress
Consider the solid, three-dimensional body of arbitrary shape which is subjected to
a set of external forces P 1 t o P6. The body is in equilibrium.

Figure 1: Forces within a body

Because points of application of the forces are separated physically, they must
communicate with one another through the body itself. We say that the body is stressed.

In continuum mechanics, stress is a physical quantity that expresses the internal

forces that neighbouring particles of a continuous material exert on each other.

Now considering only ∆P acting on ∆A, the stress at this point is defined by stress f
where
and f acts in the same direction as ∆P.
The force ∆P acting on ∆A can be resolved into two components, one component acting
normal to the surface ab and the other acting tangential to the surface. Let these
components be: ∆PN and ∆PT respectively, then the normal (direct) stress at the point is:

The tangential (or shear) stress at the point is:

The general state of uniform stress on a three-dimensional element

We now want to examine the full range of stresses which may act at a point within a
stressed body. Because the element is of differential size dx times dy times dz, we will
neglect any variation of stresses across its surface and assume that they are equal to the
stresses at the point itself. The differential element is therefore said to be in a state of
uniform stress.

Figure 2: General state of stress on a 3-dimensional element.

Because this picture is difficult to understand, we will build it step by step. Thus,
although the element has six faces, with three stresses on each face, there are only nine
different stresses involved. These nine stresses fully define the uniform state of stress of
the element, and they can be conveniently written in the form of a stress array:

This form of presentation is also called the stress tensor.

Figure 3: The general state of stress on a 3-dimensional element demonstrated in 2-D

Notations and sign convention

Normal stresses: The indices x, y, and z in σx, σy, and Plane strain describe the direction
of the stress, ie. they are parallel to the coordinate axes X, Y and Z, respectively. Normal
stress is positive when it is tensile, otherwise, when the stress is compressive, it is
negative.
Shear stresses: Shear stresses have two indices, eg. τxy, τyz etc. The first index refers to
the surface perpendicular to the corresponding coordinate. That is, the stress τxy is
parallel to the surface which is perpendicular to the axis X. The second index shows
the direction of the stress, i.e. parallel to the coordinate axis Y.

Complementary shear stress. There are only three independent values of shear stress
which, together with the three normal stresses, fully define the stress state of the element.
From Figure 3a, taking moment about d gives:

From Figure 3b, taking moment about g gives:

From Figure 3c, taking moment about e gives:

These equal-magnitude pairs of shear stress, each pair acting in the same plane along
adjacent faces of the element, are called complementary shear stresses. It follows
that the stress tensor is symmetrical about the leading diagonal.

Deformation
Deformations may occur in a material for a number of reasons, such as external
applied loads, change in temperature, tightening of bolts, irradiation effects, etc.
Bending, twisting, compression, torsion and shear or combinations of these are
common modes of deformation. In some materials like rubber or plastic, the
deformations are quite large for relatively small loads, and observable by eye. In
metals, however, the same loads would produce very small deformations requiring
the use of sensitive instruments for measurement.

Stress values do not always provide the limiting factor in design, for although a
component may be safe with regard to stress, the deformations accompanying that
stress might be dangerous or inconvenient.

The analysis of strain

Strain is the measure of the deformation of the material.

When considering stress, it was helpful to think of the total stress as consisting of a
stress σ acting normal to the plane and shear stress τ acting along the plane.

Similarly, with strain we will analyse the total strain condition in terms of a direct
strain ε and shear strain γ.

Before defining ε and γ there is a distinction which must be made between the
displacement of an element and the distortion or deformation of the element.
 An element in a structure will usually be both displaced and deformed.

NB: The strain of an element is associated only with its deformation and not with
its bodily displacement.

The strain is a measure of deformation or change of shape (due to the stresses

acting on the element) and not of bodily displacement.

Notations and sign convention. The use of indices in normal and shear strains
and the sign convention is the same as in the stresses.

Figure 4: Illustrating the displacement and deformation of an element in the web of an

I-beam.

Figure 5: Direct and shear strains of an element demonstrated in separate diagrams.

Stress-strain relations (Constitutive relations)

In structural analysis and design, it is necessary to know how structural materials
respond to the application of stress. In structures which are statically indeterminate, it is
impossible to analyse the structure without a knowledge of the stress-strain relations.

For statically determinate structures, although the stresses can be determined without a
knowledge of the material stress-strain relationship, the calculation of deflections
requires this knowledge.
The relationship between stress and strain for any material is an empirical one; it is
determined only by experiment and observation. It cannot be deduced theoretically.

For the large majority of structural analysis, the particular stress-strain relationship used
is the simplest uniaxial stress-strain characteristic, which is found by applying either a
tensile force or a compressive force to a specimen of the material and measuring the
corresponding strains in the direction of the force.

Poisson's ratio
At this point., we confine our attention to only isotropic material. We also assume that
the material is homogeneous, ie. that the strain response to stress is the same thorough
the mass of the material.

Poisson's ratio is an empirically-determined constant which relates the strain in all

directions at right. angles to the direction of the applied stress to the strain in the same
direction as the applied stress.

Figure 6: Poisson's ratio.

Thus, Poisson's ratio is defined as v, where

Because the numerator and denominator of this ratio will always be of opposite sign, it
follows that v is always a positive quantity.

Poisson's ratio is a fundamental quantity used in structural analysis.

Plane stress and plane strain
The condition of plane stress (in the xy plane) is defined by the stress array

Here σz = 0 → εz ≠ 0 because of Poisson's ratio. Hence, the corresponding strain array is

The condition of plane strain (in the xy plane) is defined by the strain array:

In this case, εz = 0 → σz ≠ 0. This stress will be necessary in order to prevent any

strain in z-direction. The corresponding stress array will therefore be

Stress-strain relations for direct stress and strain

Figure 7: Deformation of an element of the bar.

 Let us now consider what happens to an element of material within the bar shown
in Figure 7.

The element of original dimensions, dx, dy, dz will change shape as a result of the
application of σz, to the deformed element of dimensions (dx + εxdx), (dy + εydy)
and (dz+εzdz), where, because of the Poisson's ratio effect:

where E is Young's modulus. The stress array for the element in Fig 7 is F1, where

From eqn. (1) it follows that the stress array F1 causes the strain array S1, where

Now suppose that, instead of σx the stress σy acted alone giving a stress array F2, where

F2 would cause a strain array S2, where

Similarly, a stress array F3 (σz acting alone) would cause a strain array S3, where
and

If all the three direct stresses act simultaneously, thus (F = F1 + F2 + F3) causes a state
of strain described by (S = S1 + S2 + S3). That is,

causes the state of strain

Equation (5) can be written as

or, in matrix form:

Equations (6) and (7) are statements of Hooke's law.

Example 1.
A mild steel plate (E = 200kN/mm2 and v = 0.3) lies in the plane xy. It has dimensions
250mm x 250mm x 10mm. A compressive stress of 210N/mm2 is applied to the plate in
the x-direction. Find the changes in dimensions of the plate.
Figure 8: Example 1.

Solution:
We must find the resulting strains (using Hooke’s Law) in the z, y and z directions and
use them to calculate the new dimensions

We are given/know the following:

Example 2.
The mild steel plate in the previous example is again subjected to σx = 210N/mm2,
but this time the top and bottom edges of the plate are prevented from moving in
the y direction by two smooth walls which allow movement to take place in both
the x and z directions. What are changes in the plate dimensions?
Figure 9: Example 2.

Solution:
We must again find the resulting strains (using Hooke’s Law) in the z, y and z directions
and use them to calculate the new dimensions.

We are given/know the following:

Stress-strain relations for shear

The element in Figure 10, when subjected to pure shear stress τxy (= τyx) in the
xy-plane, will deform as shown. The total shear strain is γxy·
Figure 10: Deformation of a 3-dimensional element subjected to shear stress only in the
x and y directions.

The elastic relationship between τxy and γxy is given by

where G is the shear modulus for the material (sometimes referred to as the modulus of
rigidity).

Similarly, independent, relationships exist in the other two planes, namely

The general statement of Hooke's law

Combining eqns. (6), (8) and (9), we have the general statement of Hooke's law for an
isotropic, elastic element, namely

Inversion of eqn. (10) gives stresses in terms of strains as follows:

Stress-strain relations for plane stress
A state of plane stress in the xy plane is defined by σz = τxz = τyz = 0 whilst σx , σy and
τxy may have non-zero values. In this case, eqn. (10) reduces to:

Solving for σx, σy and σy from eqns. (12) gives:

In matrix notation, these relationships become

Note that equations (12), (13) and (14) are true for a state of plane stress. They are not true
for a state of plane strain.

Stress-strain relations for plane strain

A state of plane strain in the xy plane is defined by εz = γxz = γyz = 0 whilst εx, εy and
γxy may all have non-zero values. In this case, therefore eqns (10) become:
From the third line of eqn. (15), it follows that

Inversion of eqns. (15) gives stresses in terms of strains, for a state of plane strain.

Volumetric stress-strain relations

Figure 11 shows a cube subjected to stresses σx, σy and σz. As a result, the x, y and z
dimensions of the cube, each of which had an original length equal to unity, suffer strains
εx, εy and εz respectively. The new lengths of the sides therefore become (1 + εx), (1 +
εy) and (1 + εz). From eqns. (10):

Figure 11: A cube subjected to direct stresses only in the x, y and z directions.
The volume of the cube has changed from unity to

The change in volume is therefore:

Ignoring all products of strains we arrive at

Therefore, the volumetric strain εv is given by:

ie.

From eqn. (10), it follows that:

ie.

The relationships between the elastic constants E, G, and v

The elastic constants are all empirical but they are not independent of one another.
Consider the element abcd of Figure12a, lying in the xy plane and subjected to principal
stresses σx = σ1, σy = -σ1 and σz = 0, ie. a state of plane stress. The corresponding
strains are given by eqns. (10) as
NB: This is a special case in which a state of plane stress (σz = 0) creates also a state
of plane strain (εz =0) and so from eqns. (20) and (22):

The above relationship is always true for an isotropic, elastic material.

Example 3.
The state of stress at a point in a material with E = l00kN /mm2 and v = 0.25 is given by
F where

Find the corresponding strain array.

Solution:
We must again find the resulting strains (using Hooke’s Law) and arrange them in an
array that looks like this:
1
𝜀𝜀x γ𝑥𝑥𝑥𝑥 0
2
S = 1
γ𝑥𝑥𝑥𝑥 𝜀𝜀y 0
2
0 0 𝜀𝜀z

This is the strain matrix for plane stress (remember εz ≠ 0!)

We are given/know the following:

Example 4.
Determine the increase in the volume of the cylindrical steel boiler under internal
pressure, neglecting the deformation of the ends and taking σx = 30MPa and σy =
60MPa. Also E = 2.1 x 105 M Pa and v = 0.3.

Solution:
We must again find the resulting strains (using Hooke’s Law) and then find the new
cylinder length and radius (you must work with the radius and not diameter when finding
the change in dimension)
Example 5.
The spherical pressure vessel has an inner diameter of 2 m and a thickness of 10 mm. A
strain gauge having a length of 20 mm is attached to it, and it is observed to increase in length
by 0.012 mm when the vessel is pressurized. Determine the pressure causing this deformation
as well as the normal stress in the pressure vessel. The material is steel, for which E =
200GPa and v = 0.3.

Note that σ = Pd/2t

Solution:
The strain gauge strain can be calculated as ε=∆L/L. Also, the pressure vessel is thin-walled
(as t << d)

Example 6.
A cube of concrete is compressed in two perpendicular directions by force P = 120kN.

Determine the change per unit volume of the cube if it is 10cm on a side. The
compression stress is uniformly distributed over the face. Take E = 1.8 x 104MPa;
ν = 0.12. (Answer: -1.0133 x 10-3)
Solution method:
As usual, use Hooke’s law to calculate the strains, then find the change in the dimensions and
use equation 18 for the volumetric strain to calculate the answer.

We know σz = 0 (no applied load)

σx = P/A = 120 x 103/0.12 = -12 MPa (compressive)

σx = σy (the loading is symmetric), thus εx = εy = 1/E (σx – νσx) = -5.87 x 10-4

εz = 1/E (- ν(σx+ σy)) = 0.16 x 10-3

Since the volumetric strain is

Thus εv = 2(-5.87 x 10-4) + 0.16 x 10-3 = -1.013 x 10-3

Example 7.
A large spherical tank contains gas at a pressure of 3.75 MPa. The tank is 19 m in
diameter and is constructed of high-strength steel having a yield stress in tension of 570
MPa. Determine the required thickness to the nearest millimetre of the wall of the tank if
a factor of safety ( SF) of 3.5 with respect to yielding is required.

Solution method:
For thin spherical tanks, σ=Pr/2t. Also, the diameter d = 19m is the distance to the middle of the
tank wall centreline.

Let σ = P/SF = 570/3.5 and σ = 570 x 9.5/2t

Thus t = 91mm

Example 8.
A rubber ballis inflated to a pressure of 60 kPa. At that pressure the diameter of the ball is
230 mm and the wall thickness is 1.2 mm. The rubber has modulus of elasticity E = 3.5 MPa
and Poisson's ratio v = 0.45. Determine the maximum stress and strain in the ball.

Solution:
The ball is a small spherical pressure vessel and so σ=Pr/2t.

σx = σy = Pr/2t = 0.06 x 115 / (2 x 1.2) = 2.88MPa

ε = 1/E (σ- νσx) = 1/3.5 x (2.88 – 0.45 x 2.88) = 0.452

Example 9.
A closed cylindrical tank containing a pressurized fluid has an inside diameter of 250 mm and
a wall thickness of 10 mm. The stresses in the wall of the tank acting on a rotated element
have the values shown in the figure. Determine the fluid pressure.
Solution:

Solving a different way, 45 = p (125/10 + 125/20) → p = 2.4 MPa.