Theory of Plasticity

Tutor: Dr. Owaisur Rahman Shah

 Course Outline
Introduction, Stress and Strain Applications of Plasticity Theory
1. Introduction to plasticity 1. Elasto-plastic analysis of frames
2. Stress tensor, general state of stress 2. Elasto-plastic bending of beams and residual
3. General state of strain, stresses,
4. Plane stress and plane strain 3. Beams under combined stresses,
5. Volumetric strain and finite strain 4. Plastic buckling of shells
6. Stress Transformation 5. Elasto-plastic torsion and residual stresses
7. Elasticity Equations 6. Elasto-plastic analysis of thin walled cylinders
8. Stress-strain curve 7. Elasto-plastic analysis of thick walled cylinders,
9. Engineering and true strain Autofrettage
10. Isotropy, anisotropy, 8. Metal forming processes
11. Small and large deformation Computational Plasticity Problems using FEM
Foundation and Physics of Plasticity 1. Elasto-plastic analysis of pressure vessel
1. Yield Criteria, 2. Elastoplastic analysis of tensile test specimen
2. Yield locus, Anisotropy in Yielding using isotropic and kinematic hardening
3. Plastic work 3. Elastoplastic analysis of tensile test specimen in
4. Rate independence, elasto-plastic range using multilinear plasticity
5. Strain hardening, Isotropic and kinematic hardening, model
6. Flow Rules, 4. Plastic Bending of a Clamped I-Beam
7. The principle of normality 5. Analysis of residual stresses in an autofrettaged
8. Plastic instability tube
9. Cyclic plasticity and shakedown,
10. Limit Analysis,
11. Crystal plasticity
12. Rate dependence, Creep and Viscoplasticity,
13. Theory of slip-line field for plane strain elastically rigid perfectly plastic solids,
 Introduction

• A metal may be regarded as macroscopically homogeneous and

isotropic when the small crystal grains forming the aggregate are
distributed with random orientations. As a result of plastic
deformation, the crystallographic directions gradually rotate toward
a common axis, producing a preferred orientation. An initially
isotropic material thereby becomes anisotropic, and its mechanical
properties vary with direction.
• The plastic deformation in a single crystal is generally produced by
slip, which is the sliding of adjacent blocks of the crystal along
definite crystallographic planes, called slip planes. The boundary
line separating the slipped region of a crystal from the neighboring
unslipped region is called a dislocation.
 Introduction

• The magnitude and direction of the relative movement in a slip is specified by

a vector known as the Burgers vector. A dislocation is said to be one of unit
strength when the magnitude of the Burgers vector is equal to one atomic
spacing. The terms edge dislocation and screw dislocation are used to
describe the situations where the Burgers vector is normal and parallel
respectively to the dislocation line. In general, a dislocation is partly edge and
partly screw in character, and the dislocation line forms a curve or a closed
loop.
 Introduction

• In a polycrystalline metal, the crystallographic orientation changes from one

grain to the next through a narrow transition zone, or grain boundary, which
acts as an effective barrier to slip. Dislocations pile up along the active slip
planes at the grain boundaries, the effect of which is to oppose the
generations of new dislocations. When the applied stress is increased to a
critical value, the shear stress developed at the head of the dislocation pile-up
becomes large enough to cause dislocation movement across the boundary.
 Introduction

• If the temperature of the strain-hardened metal is progressively increased,

the cold-worked state becomes more and more unstable, and the material
eventually reverts to the unstrained state. The overall process of heat
treatment that restores the ductility to the cold-worked metal is known as
annealing. The temperature at which there is a marked decrease in
hardness of the metal is known as the recrystallization temperature. The
dislocation density decreases considerably on recrystallization, and the
cold-worked structure is replaced by a set of new strain free grains. The
greater the degree of cold-work, the lower the temperature necessary for
recrystallization, and smaller the resulting grain size
 Stress Strain Behavior

• The true stress–strain curve

– The stress–strain curve of an annealed
material in simple tension is found to
coincide with that in simple compression
when the true stress σ is plotted against
the true or natural strain ε.
– The greater the degree of prestrain, the
higher the new yield point and the flatter
the train-hardening curve. For a heavily Let l denote the current length of a tensile specimen
and dl the increase in length produced by a small
prestrained metal, the rate of strain- increment of the stress. Then the true strain
hardening is so mall that the material may increases by the amount dε=dl/l. If the initial length
is l0, the total strain is ε=ln(l0/l), called the true or
be regarded as approximately natural strain. For a specimen uniformly compressed
nonhardening or ideally plastic. from an initial height h0 to a final height h, the
magnitude of the natural strain is ε=ln(h0/h). The
conventional or engineering strain e, on the other
hand, is the amount of extension or contraction per
unit original length or height. It follows that ε=ln(l+e)
in the case of tension, and ε=−ln(l−e) in the case of
compression. Thus ε becomes progressively lower
than e in tension, and higher than e in compression,
as the deformation is continued in the plastic range.
 Stress Strain Behavior

• The true stress–strain curve

– Suppose that a specimen that has been
completely unloaded from a tensile plastic
state, represented by the point C, is
reloaded in simple compression (Fig. 1.1).
The stress–strain curve will then follow the
path DF’, where the new yield point F’
corresponds to a stress that is appreciably
smaller in magnitude than that at C. This
phenomenon is known as the Bauschinger
effect
 Stress Strain Behavior

• The true stress–strain curve

– A generic point on the stress–strain curve in the
plastic range corresponds to a recoverable
elastic strain equal to σ/E, and an irrecoverable
plastic strain equal to ε−σ/E. If the stress is
plotted against the plastic strain only, and the
material is assumed to have a sharp yield point,
the resulting curve will begin at σ =Y. Let H be
the slope of the true stress–strain curve
excluding the elastic strain, and T the slope of
the curve including the elastic strain, for a given
value of the stress σ. The quantities H and T are
known as the plastic modulus and the tangent
modulus respectively. A stress increment dσ
produces an elastic strain increment dσ/E and a
plastic strain increment dσ/H, while the total
strain increment is dσ/T. Hence the relationship
between H and T is
 Stress Strain Behavior

• Some consequences of work-hardening

– A longitudinal extension in the tensile test is accompanied by a contraction in the
lateral direction. The ratio of the magnitude of the lateral strain increment to that of
the longitudinal strain increment is known as the contraction ratio, denoted by η.
– Since the ratio of the elastic parts of the lateral and longitudinal strain increments is
equal to −ν, the total lateral strain increment in uniaxial tension is

– where dεe is the elastic part of the longitudinal strain increment dε. In view of the
relationship dεe =dσ/E =(T/E)dε, the contraction ratio becomes
 Stress Strain Behavior

• Some consequences of work-

hardening
– Since dl/l is equal to dε, the
condition for the onset of necking
becomes

– When the true stress–strain curve

is given, the point on the curve that
corresponds to the tensile necking
can be located graphically from the
fact that the slope at this point is
equal to the current stress (Fig.
1.2a). A heavily prestrained metal
will obviously neck as soon as the
yield point is exceeded. Since
dε=de/(1+e), the condition for
necking can be expressed in the
alternative form
 Stress Strain Behavior

• Some consequences of work-hardening

– Using the constancy of volume, the load required during
the homogeneous compression may be written as

– where A0 is the original area of cross section of the

specimen. The graph for P against e shows an upward
inflection and rises continuously without limit (Fig. 1.2b).
Setting d2P/de2 =0, and using the fact that d/dε=(1−e)d/de,
the condition for inflection is found as
 Stress Strain Behavior

• Empirical stress–strain equations

– For theoretical computations, it is often necessary to represent an experimentally
determined stress–strain curve by an empirical equation of suitable form. When the
material is rigid/plastic, it is frequently convenient to employ the Ludwik power law

– where C is a constant stress, and n is a strain-hardening exponent usually lying between zero
and 0.5. The equation predicts a zero initial stress and an infinite initial slope, except for n=0
which represents a nonhardening rigid/plastic material. The higher the value of n, the more
pronounced is the strain-hardening characteristic of the material
– The simple power law (5) may be readily modified by including a constant term Y
representing the initial yield stress. The stress–strain equation then becomes

– where m and n are dimensionless constants. Although this formula represents the strict
rigid/plastic behavior of metals, it does not give a better fit for an actual stress– strain curve
over a wide range of strains. When n=1, the above equation represents a linear strain-
hardening, which is a reasonable approximation for heavily prestrained metals. A more
successful formula, due to Swift,‡ is the generalized power law
 Stress Strain Behavior

• Empirical stress–strain equations

– For theoretical computations, it is
often necessary to represent an
experimentally determined stress–
strain curve by an empirical equation of
suitable form. When the material is
rigid/plastic, it is frequently convenient
to employ the Ludwik power law

– Voce power law

– where e is the exponential constant

 Stress Strain Behavior

• Empirical stress–strain equations

– When the elastic and plastic strains are of comparable magnitudes, it is necessary
to replace ε in the preceding equations by the plastic strain εp. Considering the
power law (5), the plastic part of the strain may be assumed to vary as σm, where
m=1/n, Since the elastic part of the strain is equal to σ/E, the total strain may be
expressed
 Stress Strain Behavior

• Empirical stress–strain equations

– When the elastic and plastic strains are of comparable magnitudes, it is necessary
to replace ε in the preceding equations by the plastic strain εp. Considering the
power law (5), the plastic part of the strain may be assumed to vary as σm, where
m=1/n, Since the elastic part of the strain is equal to σ/E, the total strain may be
expressed by the Ramberg-Osgood equation.

– where σ0 is a nominal yield stress and α a dimensionless constant. The slope of the
stress–strain curve given by the above equation continuously decreases from the
value E at the origin. At the nominal yield point σ =σ0, the plastic strain is α times
the elastic strain, and the secant modulus is E/(1+α). The tangent modulus at any
point of the curve is given by
 Stress Strain Behavior

• Empirical stress–strain equations

– The second term on the right-hand side is equal to E/H in view of (1). The stress–
strain curve for a range of materials can be reasonably fitted by Equation (9) with
α=3/7. For a nonhardening material (m=∞), the equation degenerates into a pair
of straight lines meeting at the yield point σ =σ0.
– The material behavior can be simplified as

– While for an ideally plastic material a curve fitting can be of the form
 Stress Strain Behavior

• Analysis of Stress
– Stress tensor When a body is subjected to a set of external
forces, internal forces are produced in different parts of the
body so that each element of the body is in a state of statical
equilibrium. Through any point O within the body, consider a
small surface element δS whose orientation is specified by
the unit vector l along the normal drawn on one side of the
element (Fig. 1.8a). The material on this side of δS may be
regarded as exerting a force δP across the surface element
upon the material on the other side. The limit of the ratio
δP/δS as δS tends to zero is the stress vector T at O
associated with the direction I. For given external loading,
the stress acting across any plane passing through a given
point O depends on the orientation of the plane. The
resolved component of the stress vector along the unit
normal l is called the direct or normal stress denoted by σ,
while the component tangential to the plane is known as the
shear stress denoted by τ.
 Stress Strain Behavior

• Analysis of Stress
– The mean of the three normal stresses, equal
to (σx +σy +σz)/3, is known as the hydrostatic
stress denoted by σ0. A deviatoric or reduced
stress tensor sij is defined as that which is
obtained from σij by reducing the normal
stress components by σ0. This gives the
deviatoric stresses as
 Stress Strain Behavior

• Stresses on an oblique plane

– Consider the equilibrium of a small
tetrahedron OABC of which the edges OA,
OB, and OC are along the coordinate axes
(Fig. 1.9). Let (l,m, n) be the directions
cosines of a straight line drawn along the
exterior normal to the oblique plane ABC.
These are the components of the unit
normal 1 with respect to Ox, Oy, and Oz. If
the area of the face ABC is denoted by δS,
the faces OAB, OBC, and OCA have areas
n δS, l δS, and mδS respectively. The
stress vector T acting across the face ABC
has components Tx, Ty, and Tz along the
axes of reference. Resolving the forces in
the directions Ox, Oy, and Oz, we get
 Foundation and Physics of Plasticity

• The Criterion of Yielding

– Suppose that an element of material is subjected to a system of stresses of

gradually increasing magnitude. The initial deformation of the element is
entirely elastic and the original shape of the element is recovered on
complete unloading. For certain critical combinations of the applied
stresses, plastic deformation first appears in the element. A law defining
the limit of elastic behavior under any possible combination of stresses is
called yield criterion. The law applies not only to loading directly from the
annealed state, but also to reloading of an element unloaded from a
previous plastic state.
 Foundation and Physics of Plasticity

• A geometrical representation
– Consider a system of three mutually
perpendicular axes with the principal
stresses taken as rectangular coordinates
(Fig. 2.1). The state of stress at any point in
a body may be represented by a vector†
emanating from the origin O. Imagine a line
OH equally inclined to the three axes, so
that its direction cosines are (1/ √3, 1/√3,
1/√3). The stress vector OQ, whose
components are (σ1, σ2, σ3), may be
resolved into a vector OG along OH and a
vector OP perpendicular to OH. The vector
OG is of magnitude √ 3σ0 and represents
the hydrostatic stress with components (σ0,
σ0, σ0). The vector OP represents the
deviatoric stress with components (s1, s2,
s3) and its magnitude is √2J2 by Eq. (28),
Chap. 1. For any given state of stress, the
deviatoric stress vector will lie in the plane
passing through O and perpendicular to OH.
This plane is known as the deviatoric plane
and its equation is σ1 +σ2 +σ3 =0 in the
principal stress space. Since a uniform
hydrostatic stress has no effect on yielding,
it follows that yielding can depend only on
the magnitude and direction of the
deviatoric stress vector OP.
 Foundation and Physics of Plasticity

• Deviatoric yield loci of Tresca and von Mises.

 Foundation and Physics of Plasticity

• Deviatoric yield loci of Tresca and von Mises.

 Strain-Hardening Postulates

• Isotropic hardening
– We have seen that an element of material
yields when the magnitude of the deviatoric
stress vector is increased to a value such that
the stress point reaches the yield locus. Unless
the locus is a circle (as for the Mises criterion),
the magnitude of the stress vector causing
yielding depends on its final direction in the
deviatoric plane. If the material is
nonhardening, the plastic stress state can
change in such a way that the stress point
always lies on a constant yield locus. For a
strain hardening material, the size and shape of
the yield locus depend on the complete history
of plastic deformation since the previous
annealing. It is assumed that the material is
isotropic at the annealed state and that the
anisotropy and the Bauschinger effect
developed during the coldwork may be
neglected. The preceding discussion of the
yield criterion is then appropriate for any given
state of hardening of the material.
 Strain-Hardening Postulates

• Isotropic hardening
– A convenient mathematical formulation for strain-hardening is obtained by assuming further that the yield
surface uniformly expands without change in shape, as the state of stress changes along a certain path P0P
in the stress space (Fig. 2.8), the amount of hardening being given by the final plastic state. Since the yield
locus merely increases in size, any given state of hardening may be defined by the current yield stress in
uniaxial tension. It is, therefore, necessary to relate the current yield stress to the amount of plastic
deformation following a given initial state of yielding. To this end, we replace Y in the yield criterion by σ,
which is known as the equivalent stress, effective stress, or generalized stress. Referring to the von Mises
yield criterion, we write
 Strain-Hardening Postulates

• Isotropic hardening
– A convenient mathematical formulation for
strain-hardening is obtained by assuming
further that the yield surface uniformly expands
without change in shape, as the state of stress
changes along a certain path P0P in the stress
space (Fig. 2.8), the amount of hardening being
given by the final plastic state. Since the yield
locus merely increases in size, any given state
of hardening may be defined by the current
yield stress in uniaxial tension. It is, therefore,
necessary to relate the current yield stress to
the amount of plastic deformation following a
given initial state of yielding. To this end, we
replace Y in the yield criterion by σ, which is
known as the equivalent stress, effective stress,
or generalized stress. Referring to the von
Mises yield criterion, we write
 Strain-Hardening Postulates

• Isotropic hardening
– with the fact that no hardening is produced by purely elastic strains. If the plastic part of
the strain increment tensor is denoted by dεpij = ˙εpijdt, where ˙εij is the rate of
deformation and dt the time element, the increment of plastic work per unit volume is

– where δij is the Kronecker delta whose value is unity when i = j and zero when i ≠ j.

– where the last step follows from the condition dεpij = 0, implying that there is no plastic
volume change. Plastic incompressibility of metals is in close agreement with
experimental observations, and is also consistent with the fact that a uniform hydrostatic
stress produces no plastic strain. The work-hardening hypothesis may be stated
mathematically as
 Strain-Hardening Postulates

• Isotropic hardening
– where the integral is taken over the actual strain path starting from some initial state. The
function can be determined from the true stress–strain curve in uniaxial tension or
compression. If the true stress σ is plotted against the plastic part of the strain, then Wp is
equal to the area under the curve up to the ordinate σ. Since σ =σ in this case, the area
directly gives the argument of
 Strain-Hardening Postulates

• Isotropic hardening
– In an alternative hypothesis, more frequently in use, σ is regarded a function of a certain
measure of the total plastic strain. Considering the second invariant of the plastic strain
increment tensor, an equivalent (or generalized) plastic strain increment is defined as

– where only the positive root is implied. The numerical factor in the above expression is so
chosen that in uniaxial tension, dεp equals the longitudinal plastic strain increment. This
follows from the fact that the magnitude of the lateral compressive plastic strain in the
tensile test of an isotropic bar is half the longitudinal tensile plastic strain. The strain-
hardening hypothesis may now be expressed as
 Flow Rules

• The Rule of Plastic Flow

– When an element of material is unloaded from a certain plastic state, it recovers
elasticity and the stress point moves inside the yield locus. If anisotropy is disregarded,
the elastic behavior of the material is characterized by two independent elastic constants
which retain their initial values. When the element is reloaded along a certain strain-
path, yielding will again occur when the stress point reaches the current yield locus. For a
work-hardening material, a further plastic flow can be enforced only by increasing the
stress to a point outside the yield locus. If the stress increment is such that the stress
point remains on the same yield locus, no hardening is produced and the plastic strain
increments are zero. Such changes in stress are called neutral since they represent
neither loading nor unloading. The elastic part of the strain increment corresponding to
any plastic flow is directly related to the stress increment by means of Hooke’s law. It is
therefore necessary to relate the increment of plastic strain to the stress increment and
the current stress.
 Flow Rules

• The plastic potential

– The observed plastic behavior of polycrystalline metals clearly indicates that for isotropic
materials, the principal axes of the plastic strain increment coincide with those of the
stress. The plastic strain increment may therefore be regarded as a vector

– in the principal stress space, the factor 2G being introduced to obtain the dimension of
stress. Since

– the plastic strain-increment vector may be regarded as normal to a right cylinder

perpendicular to the deviatoric plane. The curve in which the cylinder is intersected by
the deviatoric plane is a level curve of a scalar function of the deviatoric principal stresses
(s1, s2, s3).
 Flow Rules

• The plastic potential

– The observed plastic behavior of polycrystalline metals clearly indicates that for isotropic
materials, the principal axes of the plastic strain increment coincide with those of the
stress. The plastic strain increment may therefore be regarded as a vector

– in the principal stress space, the factor 2G being introduced to obtain the dimension of
stress. Since

– the plastic strain-increment vector may be regarded as normal to a right cylinder

perpendicular to the deviatoric plane. The curve in which the cylinder is intersected by
the deviatoric plane is a level curve of a scalar function of the deviatoric principal stresses
(s1, s2, s3).
 Flow Rules

• The plastic potential

– It is reasonable to stipulate that the ratios of the components of the plastic strain
increment depend on the current stress and not on the stress increment. The magnitude
of the strain increment is, however, determined by the stress increment through the
strain-hardening characteristic of the material. The plastic strain-increment vector is
therefore parallel to the normal to the curve at the point where it is intersected by the
deviatoric stress vector. Since, in an isotropic material, the strain increments are
interchanged when the stresses are so, the curve must be symmetrical with respect to
the stress axes. Moreover, the reversal of the sign of the applied stresses in our idealized
material should merely change the sign of the strain increments. Hence the slope of the
curve at the opposite ends of a diameter must be the same. This is possible if the curve is
also symmetrical about diameters perpendicular to the three axes. Thus the curve, like
the yield locus , is identical in each of the 30◦ segments marked off by the directions
representing the states of uniaxial stress and pure shear. It follows that the equation of
the potential surface, whose size is immaterial, may be written in the form
 Flow Rules

• The plastic potential

– The function g, defining the ratios of the components of the plastic strain increment, is
known as the plastic potential.‡ In a nine-dimensional space, the plastic strain increment
may be expressed by the flow rule

– where dλ is a positive scalar that depends on the stress increment, and is generally a
function of the space variables as well as the time scale. Since g is independent of the
hydrostatic stress, the plastic incompressibility condition dεpii=0 is identically satisfied. If g
is assumed to be a homogeneous function of degree n, involving the stress components,
the increment of plastic work per unit volume may be written as
 Flow Rules

• Particular Stress–Strain Relations

– Lévy-Mises and Prandtl-Reuss equations
– The plastic flow rule corresponding to any particular choice of the plastic potential may
be readily obtained. The simplest form of the potential curve, having all the properties
described in the preceding section, is obviously a circle. In this case θ =ψ, or equivalently
μ=ν, whatever the form of the yield locus. Taking g = J2 = 1/2 sijsij, and employing,

– we obtain the corresponding flow rule

– The stress–strain relation in this form was suggested independently by Lévy and von
Mises, who used the total strain increments instead of the plastic strain increments
 Flow Rules

• The Total Strain Theory

• Hencky’s stress–strain relation
• Although the Prandtl-Reuss stress–strain relation provides the most
satisfactory basis for treating plasticity problems, the theory is incremental
and generally leads to mathematical complexities. Considerable
simplifications are often achieved by using a system of equations due to
Hencky,† who postulated a one-to-one correspondence between the stress
and the strain. Thus, the components of the total plastic strain are taken to
be proportional to the corresponding deviatoric stresses. Assuming small
strains, the plastic stress–strain relation proposed by Hencky may be
written as
 Flow Rules

• The Total Strain Theory

• where λ is positive during loading and zero during unloading. When the material work-
hardens, λ depends on the equivalent stress σ, which may be regarded as a function of an
equivalent total plastic strain εp defined as

• the relationship between σ and εp being given by the uniaxial stress–plastic–strain curve.
Substituting from

• and using

• for σ, which implies the von Mises yield criterion, we have

 Flow Rules

• The Total Strain Theory

• where S is the secant modulus of the uniaxial stress–strain curve at σ = σbar. For an
incompressible material, εeij= 3sij/2E by Hooke’s law, and (56) then furnishes εij =3sij/2S. The
incremental form of (56) is

• The Hencky equation is equivalent to the Prandtl-Reuss equation when the ratios of the
deviatoric stress components are held constant. This may be shown by writing sij =(σ/Y)s0ij,
where s0ij is the deviatoric stress at the initial yielding. Since dsij =(dσ/σ)sij

• the quantity dεp in this case being identical to dεp. When the stress–plastic–strain relationship
in uniaxial tension is represented by a power law, the Hencky equation is initially equivalent to
the Prandtl-Reuss equation.† Indeed, if σ varies as the nth power of εp, it is easily shown that
dσ/σ =n(dεp/εp), while εp/σ→0 as εp tends to zero. The complete equivalence of the two
equations is established by setting dεp = (1−n)dεp at the initial stage.
 Flow Rules

• The Total Strain Theory

• The Hencky theory can be extended to large strains by using a suitable definition of the strain
tensor εij, the most natural definition in this context being εij = dεij, where the integral is taken
along the path of the particle. When the principal axes of strain increment remain fixed in the
element, it leads to the logarithmic strains in the principal directions.
 Flow Rules

• The Total Strain Theory

• If the principal axes of the stress do not rotate with
respect to the element, a geometrical representation
of stresses and strains is again possible. The vectors
OP and PS, representing the deviatoric stress and the
plastic strain respectively (Fig.), are directed along
the same radial line whatever the stress path.
According to the Prandtl-Reuss theory, this happens
only when the stress path is a straight line through
the origin. The vector OP also represents the
deviatoric elastic strain, and the vector OS represents
the total deviatoric strain. The magnitudes of the
vectors OP and PS are √2/3σ and 2G√3/2εp
respectively. Hence
 Flow Rules

• The Total Strain Theory

• When the point S describes a certain curved strain
path, the stress vector rotates so that it is always
along OS. The stress increment PP’ corresponding to
a strain increment SS’ is obtained by locating P’ on
OS’ by the above relation. The vector SR, equal to PP’,
represents the elastic strain increment, and the
vector RS’ represents the plastic strain increment.
This is evidently incompatible with the Prandtl-Reuss
flow rule which requires RS’ to be parallel to OP. The
stress path corresponding to the Hencky equation
therefore differs from that given by the Prandtl-Reuss
equation. The projection of RS’ on a line through R
parallel to OS is RT =2G√3/2dεp, where dεp, is the
increment of the equivalent total plastic strain. The
scalar product of the vectors OP and RS’ gives the
increment of plastic work dWp =σdεp, which agrees
with that obtained from the scalar product of (57)
with the deviatoric stress sij .
 Applications of Plasticity Theory

• ELASTOPLASTIC BENDING AND TORSION

• In a deformable body subjected to external loads of gradually increasing magnitude, plastic
flow begins at a stage when the yield criterion is first satisfied in the most critically stressed
element. Further increase in loads causes spreading of the plastic zone which is separated
from the elastic material by an elastic/plastic boundary. The position of this boundary is an
unknown of the problem, and is generally so complicated in shape that the solution of the
boundary-value problem often involves numerical methods. The solution must be carried out
in a succession of small increments of strain even when the deformation is restricted to an
elastic order of magnitude. It is necessary to ensure at each stage that the calculated stresses
and displacements in the elastic and plastic regions satisfy the conditions of continuity across
the elastic/plastic boundary. In this chapter, we shall be concerned mainly with the problems
of bending and torsion in the elastic/plastic range, assuming the deformation to be sufficiently
small. The related problems of limit analysis will be discussed in the next chapter.
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

• As a simple application of the Prandtl-Reuss theory, consider the frictionless compression of a
rectangular block of metal between a pair of rigid overlapping platens (Fig. 3.1). The edges of
the block are parallel to the rectangular axes, with the x axis taken in the direction of
compression. A condition of plane strain is achieved by suppressing lateral expansion in

• the z direction with the help of rigid dies. It is therefore a case of homogeneous compression
in which σy =0 throughout the deformation, and σz =νσx while the block is still elastic. If
Tresca’s yield criterion is adopted, yielding begins when σx =−Y in each element of the block.
The relevant stress–strain equations in the plastic range are
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

• If the material is non-hardening, σx =−Y throughout the plastic compression. The elimination
of dλ from

• then gives

• At the initial yielding, σz=−νY and εx =−(1 − ν2)Y/E. Under these initial conditions, the above
equation integrates to
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

• giving the variation of σz with the amount of compression. As the deformation proceeds, the
first term becomes increasingly unimportant, while σz rapidly approaches the limiting value
−1/2Y. Taking ν=0.3, for instace, σz is found to have the value −0.49Y when εx is only 3.5 times
that at the initial yielding. When the material yields according to the von Mises yield criterion
σ2x −σxσz+ σ2z =Y2, the initial yielding of the block corresponds to σx =σ0x and σz =νσ0x , where

• During the subsequent compression, the yield criterion can be identically satisfied by writing
the stresses in terms of a parameter θ (which is the deviatoric angle) as

• The condition σz =νσx at the initial yielding furnishes the initial value of θ as
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

• When ν=0.3, we get σ0x = −1.127Y and θ0=13◦. Substitution from

• into the stress–strain relations

• gives
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

• Since dλ must be positive, the second equation indicates that θ decreases as the compression
proceeds. Eliminating dλ from the above equations, we get

• Using the initial condition ε =−(1 − ν2)σ0x/E when θ =θ0, the above equation is readily
integrated to obtain

• As the deformation continues, the first term on the right-hand side soon becomes negligible.
The angle θ rapidly approaches the limiting value zero, the corresponding values of σx and σz
being −2Y/√3 and −Y/√3 respectively. It is found that σz is within 1 percent of its limiting value
when εx is only four times that at the initial yielding, for ν=0.3. Owing to the rapid initial
change in stress, the elastic and plastic strain increments are comparable up to a total strain
which is three to four times that at the elastic limit. A graphical comparison of the solutions
based on the Tresca and Mises criteria is made in Fig
 Applications of Plasticity Theory

• (i) Plane strain compression of a block

 Applications of Plasticity Theory

• Plain Bending of a beam

• Torsion of a cylindrical shaft
 Applications of Plasticity Theory

• Thin-Walled Tubes Under Combined Loading

 Rate-Independent Plasticity

• Flow Rule and Work-Hardening

• Flow Rule
• In keeping with the formulation of rate-independent plasticity as the limit of classical
viscoplasticity for infinitely slow processes, we henceforth consider all processes to be
“infinitely” slow (compared with the material relaxation time τ), and correspondingly regard
the material as “inviscid plastic,” “rate-independent plastic,” or simply plastic. The inelastic
strain "i will from now on be called the plastic strain and denoted "p instead of "i. The flow
equations may be written as
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Assignment
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming
Metal Forming