Concept of Tensors Deformation of A Body Stress Equilibrium Equations Constitutive Equations Principles From Work and Energy Homework References
Concept of Tensors Deformation of A Body Stress Equilibrium Equations Constitutive Equations Principles From Work and Energy Homework References
Homework
References
Tensors are physical entities whose components are the coefficients of a linear relationship between
vectors.
The list of some of the tensors used in this course is given in Table 2.1.
It is often needed to transform a tensorial quantity from one coordinate system to another coordinate
system. This transformation of a tensor is done using direction cosines of the angle measured from
initial coordinate system to final coordinate system. Let us use axes as the initial coordinate axes
and as the final coordinate axes (denoted here by symbol prime – ). Now, we need to find the
direction cosines (denoted here by a ij) for this transformation relation. Let us use the convention for
direction cosines that the first subscript (that is, i) of a ij corresponds to the initial axes and the
second subscript (that is, j) corresponds to final axes. The direction cosine correspondence with this
convention in 3D Cartesian coordinate system is given in Table 2.2. The corresponding Cartesian
coordinate systems are shown in Figure 2.1.
Table 2.2 Direction cosines for 3D Cartesian coordinate system
From/To
Now, we will find all the direction cosines. The list is given below.
(2.1)
The matrix of direction cosines given above in Eq. (2.1) is also written using short forms for
. Then Equation (2.1) becomes
(2.2)
Note: The above direction cosine matrix can be obtained from the relation between unrotated and
rotated coordinates. For the transformation shown in Figure 2.2 (a) one can write this relation using
Now we will use the direction cosines to transform a vector, a second order tensor and a fourth order
tensor from initial coordinate (unprimed) system to a vector, a second order tensor and a fourth order
tensor in final coordinate (primed) system.
First, let us do it for a vector. Let denote the components of a vector P in unprimed and
primed coordinate axes. Then the components of this vector in rotated coordinate system are given in
terms of components in unrotated coordinate system and corresponding direction cosines as
(2.3)
Now, putting the direction cosines in terms of angles and summing over the repeated index j (=1, 2, 3)
in Equation (2.3) we get
(2.4)
(2.5)
(2.6)
Now, we will extend the concept to transform a second order tensor. Let us transform the stress
tensor as follows
(2.7)
(2.8)
The readers are suggested to write the final form of Equation (2.8) using similar procedure used to
get the last of Equation (2.7).
(2.9)
Thus, deformation map is a vector valued function. Similarly, for deformation of a point Q to , we
can write
(2.10)
(2.11)
where is called Deformation Gradient. In component form, one can write
(2.12)
Now, let us give the deformation map for the displacement of a point. Let us consider the point P in
reference configuration again. It undergoes a deformation and occupies a new position .
Thus, we can write this deformation as follows
(2.13)
(2.14)
or in component form
(2.15)
(2.17)
(2.18)
where is given as . Thus, the strain components are nonlinear in . Here,
(2.19)
Similarly,
(2.20)
The readers should observe that from the definition of strain tensor in Equation (2.18), the strain tensor is
symmetric (that is, ). If the gradients of the displacements are very small the product terms in
Equation (2.18) can be neglected. Then, the resulting strain tensor (called Infinitesimal Strain Tensor) is
given as
(2.21)
(2.22)
The readers are very well versed with these definitions. This strain tensor can be written in matrix form as
(2.23)
Note: The shear strain components mentioned above are tensorial components. In actual practice,
engineering shear strains (which are measured from laboratory tests) are used. These are denoted by .
The relation between tensorial and engineering shear strain components is
(2.24)
Stress
Now, we will introduce the concept of stress. The components of stress at a point (also called State
of Stress) are (in the limit) the forces per unit area which are acting on three mutually perpendicular
planes passing through this point. This is represented in Figure 2.4. Stress tensor is a second order
tensor and denoted as . In this notation, the first subscript corresponds to the direction of the
normal to the plane and the second subscript corresponds to the direction of the stress. For
example, denotes the stress component acting on a plane which is perpendicular to direction 2
and stress is acting in direction 3. The tensile normal stress components are
positive. The shear stress components are defined to be positive when the normal to the
plane and the direction of the stress component are either both positive or both negative.
The readers should note that the state of stress shown in Figure 2.4 represents all stress
components in positive sense. In this figure, the stress components are shown on positive faces
only.
(2.27)
In general, instead of using global 1-2-3 coordinate system, x-y-z global coordinate system is used.
Further, the shear stress components are shown using notation . Thus, the stress tensor in this
case can be written as
(2.28)
Note: The stress tensor will be symmetric, that is only when there are no distributed
moments in the body. The readers are suggested to read more on this from any standard solid
mechanics book. In this entire course, we will deal with symmetric stress-tensor.
The equilibrium equations, without body forces are written using xyz coordinates as follows:
(2.31)
The boundary conditions are very essential to solve any problem in solid mechanics. The boundary
conditions are specified on the surface of the body in terms of components of displacement or
traction. However, the combination of displacement and traction components is also specified.
Figure 2.5 shows a body, where the displacement as well as traction components are used to specify
the boundary conditions.
We define traction vector for any arbitrary point (for example, point P in Figure 2.5) on surface as
a vector consisting of three stress components acting on the surface at same point. Here, the three
stress components are normal stress and shear stress and . The traction vector at this
point is written as
(2.32)
where is the ith component of the unit normal to the surface at point P. For example, if this
surface is perpendicular to axis 2, then and the components of traction acting at a
point on this surface are given as follows
(2.33)
(2.35)
Plane stress problem corresponds to a situation where out of plane stress components are negligibly
small. Thus, we can say that the state of stress is planar. The planar state of stress in x-y plane is
shown in Figure 2.6. For the case shown in this figure, the normal and shear stress components in z
directions, that is are zero. Please note that the state of stress shown in this figure
assumes the stress symmetry.
For plane stress problem the equilibrium equations take the following form
(2.36)
Plane strain problem corresponds to a condition where all the out of plane strain components are
negligibly small. Here, we denote as out of plane strain components. The readers
are again cautioned to note that the out of plane stress components need not be zero. These
depend upon the constitutive equation. Further, the equilibrium equation is same as Equation (2.36)
and .
(2.38)
The readers should note that strain energy density is a scalar quantity. Further, it is a positive
definite quantity.
The principle of minimum of total potential energy states that of all possible kinematically admissible
displacement fields, the actual solution to the problem is one which minimizes the total potential
energy .
The total potential energy (for linearly elastic material) is defined as
(2.39)
Note: The kinematically admissible displacement field is a single valued and continuous
displacement field that satisfies the displacement boundary condition.
The principle of minimum of total complementary potential energy states that of all possible statically
admissible stress fields, the actual solution to the problem is one which minimizes the total
complementary potential energy .
The total complementary potential energy (for linearly elastic material) is defined as
(2.40)
Note: The statically admissible stress field is one that satisfies both equilibrium equations and
traction boundary condition.
4. Derive the principles of minimum of total potential and total complementary potential energy.