METAL FORMING

1. Classification of Forming Processes

The importance of metals in modern technology is due, in large
part, to the ease with which they may be formed, into useful
shapes such as tubes, rods, and sheets.
Useful shapes may be generated in two basic ways:
1. By plastic deformation processes in which the volume and mass of
metal are conserved and the metal is displaced from one location to
another.
2. By metal removal or machining processes in which material is
removed in order to give it the required shape.
Equal in importance to the creation of useful shapes by plastic
forming processes is the control of mechanical properties by the
metalworking process.
 The deformation processing situation is best viewed
as a total system (Fig. 2).
The deformation zone is concerned with the
distribution of stress, strain, and particle velocities,
and with the overall pressure required to perform the
operation.
Obviously, the applied forces must develop yielding
in the material but the stresses must not locally
create fracture.
Figure 2. Deformation system
Such metallurgical phenomena as strain hardening, recrystallization and fracture are
important, but often under specialized conditions of high strain rates and/or high temperature.

The flow stress of the material will be a strong function of strain, strain rate, and temperature.

The workpiece will be in contact with nondeforming (elastic) tools or dies.

The friction along this interface and the heat transfer from the workpiece to the die are
important considerations, as are such practical concerns as tool wear and surface finish of the
product.
 Mechanics of Metalworking
The analysis of the stresses in metalworking processes has been an active area of
applied plasticity. Since the forces and deformations generally are quite complex,
it is usually useful to use simplifying assumptions to obtain a tractable solution.
Because the strains involved in deformation processes are large, it is usually
possible to neglect elastic strains and consider only the plastic strains (plastic-
rigid solid).
The principal use of analytical studies of metalworking processes is for
determining the forces required to produce a given deformation for a certain
geometry prescribed by the process. A mechanics analysis of a process may also
be used to develop information on the frictional conditions in the process. An
important problem area is predicting the limiting deformation at which fracture
will occur.
Metal working occurs due to plastic deformation which is associated with
analysis of complex stress distribution- require simplification.
Only (large) plastic strain is considered while elastic strain is very small and can be
neglected.
Strain hardening is often neglected. Metal is considered to be isotropic and
homogeneous.
Normally plastic deformation is not uniform and also has frictions, but we need to
simplify the stress analysis in order to determine the force required to produce a given
amount of deformation to obtain a product in a required geometry.
Relating stress and strain in elastic deformation
1. Expressing stress and strain at a point
2. Relating stress and strain in a solid that obeys
Hooke’s law

Description of stress at a point

σx, σY, σZ are normal stresses acting on a plane

perpendicular to X, Y, Z directions respectively
Normal stress values greater than zero represents
tensile and less than zero represents compression.

Two subscripts are required to describe shear stresses

=> τXY, τYX, τXZ, τZX, τYZ, τZY τXY => shear stress on the
plane perpendicular to X-axis in the direction of Y
axis +ve if it points in the positive direction on the
positive face of unit cube; -ve if it points in the
negative direction of a positive face of a unit cube
 Nine quantities viz., σx, σY, σZ, τXY, τYX, τXZ, τZX, τYZ, τZY must be defined
to describe state of stress at a point

Simplification can be done by assuming the areas of the faces of the unit cube are
small enough so that the change in stress over the face is negligible.

Therefore τXY =τYX; τXZ =τZX; τYZ=τZY (take summation of moments about each axis)

State of stress at a point is completely described by six components

–three normal stresses and three shear stresses => σx, σY, σZ, τXY, τXZ, τYZ
Principal stresses and planes
Principal planes: Planes on which only normal stresses act and not shear
stresses. The stresses normal to these planes are called principal stresses. In 2-
D => σ1, σ2; In 3-D => σ1, σ2 & σ3 and σ1>σ2 >σ3 algebraically.

Equilibrium equations
Assume non-homogeneous condition for the force system shown and by
considering equilibrium conditions along each axes, we can get equilibrium
equations as shown below.
State of stress in three dimensions
The system shown is in equilibrium and forces acting on each of its
faces must balance.
Along X-axis,
σAl = σx Al + τyx Am + τzx An ⇒(σ–σx) l - τyxm –τzx n = 0
Similarly,
-τxy l + (σ- σy) m – τzy n = 0 (along Y axis) ;

-τxz l –τyzm + (σ-σz) n = 0 (along Z axis)

Here l, m, n are directional cosines of ‘σ’ i.e., cosines of angles between

‘σ’ and X, Y, Z axes.
JKL => Principal plane cutting through unit cube of area A
σ is the principal stress acting normal to the plane JKL.
These three equns. are linear, homogeneous equns. in l, m, n.
The nontrivial solution can be obtained by,
Solving the above determinant results in a cubic equation in ‘σ’
 The three roots of the equation, σ1, σ 2, σ 3 are called
principal stresses. The direction of the principal
stresses with x, y, z axes can be obtained by
substituting each of σ1, σ2, σ3 into the determinant
and solving them with the help of l2+m2+n2 = 1
relationship. Here, in general, σ1>σ2>σ3 .
The coefficients in the above equation do not vary
with changes in coordinate axes. They are called as
invariant coefficients, I1, I2, I3.
In terms of principal stress,
I1 = σx + σy + σz ; I2 = σxσy + σyσz + σxσz; I3 =
σxσyσz
We derived equations for stress on an oblique plane
(principal plane) where no shear stress is acting
Similarly equations can be developed for normal and
shear stresses acting on an oblique plane whose From τ2 = S2 – σ2
normal has DC’s l, m, n with x, y, z axes. In this
case, the total stress (S) on the plane will not be
coaxial with the normal stress (σ).
One can apply equilibrium principle on each of the
axis and evaluate equations for ‘σ’ and ‘τ’.
Here S2 = σ2 + τ2
Here l, m, n are DC’s between the normal to the
oblique plane and the principal axes. We need to find
the principle shear stresses and planes on which they
act. The principal shear stresses occur for the
following combinations of DC’s that bisect any of
the two principal axes.
We know that σ1 > σ3 algebraically, τ2 is the
maximum principal shear stress. So, τ2 = τmax = σ1-
σ3/2. Note that for each pair of principal stress, there
are two planes on which principal shear stresses act,
which bisect principal stresses direction.
Volume change:
It is referred by Volume strain, change in volume
per unit volume. In the case of rectangular
parallelepiped with dx, dy, dz as edges,
Volume strain, Δ = (1+εx) (1+εy) (1+εz) – 1
Since strain values are small, we can neglect the
strain products.
Hence,
Δ = εx + εy + εz = First invariant of strain tensor
Hydrostatic component of strain
(or Mean strain) = εm = (εx + εy + εz)/3 = Δ/3
The part of strain tensor that is responsible for shape
change rather than volume change is called strain
deviator tensor (εij).

This can be obtained by subtracting mean strain from

normal strain.The splitting of total strain tensor into
volume/shape strain and strain deviator tensor can be
represented by,
εij = εij’+εm = Strain deviator tensor + Mean strain
Mean and deviator stress tensor
Similar to strain tensor components, stress tensor
can be divided into mean stress tensor (σm) and
deviator stress tensor (σij’). Mean stress component
produce only elastic volume change and does not
cause plastic deformation. So, yielding is
independent of Hydrostatic or Mean stress tensor.
While deviator stress tensor involves shear stresses
and is important in causing plastic deformation.The
decomposition of stress tensor and further
simplification gives
Consider principal axes,
σ1’ = 2σ1-σ2-σ3/3 = (σ1-σ2) + (σ1-σ3)/3 = 2/3 [(σ1-
σ2)/2+ (σ1-σ3)/2] = 2/3(τ3+τ2)
It is clear that the stress deviator tensor involves
shear stresses.
The principal values of stress tensor are the roots of
the cubic equation
Where J1, J2, J3 are invariants of the stress deviator
tensor and given by,
J3 is the determinant of stress deviator tensor
 Elastic stress-strain relationship
How can we relate stress tensor and strain tensor?
This can be done by introducing material
properties in the elastic region. Elastic stress can
be related to elastic strain by Young’s modulus
(Hooke’s law)
σx = Eεx; where E is the Young’s modulus
3-D elastic stress-strain relations:
A tensile force along X-direction can produce
elongation in that direction and produces contraction
along Y and Z directions. The strains in three
different directions can be related by Possion’s ratio,
γ. means by applying σx to the solid, strains εy,
This
εz, and εx can be created and are related as above.
This holds good for σy and σz also.
The following table details this effect.
We have assumed that, (1) material is isotropic, and
(2) normal stresses does not produce shear strain on
x, y, z planes and shear stresses does not produce
normal strains on x, y, z planes.
Finally by superposition of strain components from
the table, we get elastic stress-strain relations in 3-D.
 Hooke′s law in 3-D

Shear stresses will create shear strains i.e., τxy =G

γxy; τyz =G γyz; τxz =G γxz;
where G is modulus of rigidity.

Bulk modulus, K = Hydrostatic pressure/volume

strain = σm / Δ;
Four elastic constants viz., E, γ, G, K can be related
as,

Also, we have assumed isotropic nature in the elastic deformation. Anisotropy of elastic
deformation is possible. Since elastic part is small during metal deformation.
Plastic deformation
Important points to remember in plastic deformation
- Hooke’s law is not valid in plastic deformation
- Irreversible process – material will not come back to its original
dimension
- Plastic strain depends on the loading path by which final state is
achieved (elastic deformation depends on initial and final states). There is
no easily measured constant relating stress and strain in plastic
deformation region, unlike elastic deformation region
- Important phenomenon ‘STRAIN HARDENING’ has to be addressed in
this regime - Plastic anisotropy, Bauschinger effects are important- Criteria for
yielding has to be developed.

Flow curve and flow equation

Unloading at A, total strain will decrease from ε1 to ε2. This strain decrease is
called recoverable elastic strain. Some plastic strain will disappear with time
(ε2 –ε3). This is known as anelastic behavior.

Unloading at A, stress decreases with strain parallel to elastic regime & upon
reloading curve bend over and reaches A, after which it takes the original
shape of stress-strain curve.
The true stress-strain curve is called as Flow curve and the eqn. that describes
the curve is called Flow equation. Generally the flow equations are empirical,
fit equations.
For eg., Hollomon’s equation, σ=Kεn where ‘σ’ is the true stress for
particular true strain ‘ε’, ‘K’ is the strength coefficient, ‘n’ is the strain
hardening exponent is a flow equation.
It should be noted that this eqn. is valid from onset of plastic
deformation to maximum load at which necking starts.
Relating engineering and true quantities
Dimensions of the specimen change with deformation; hence true quantities are better indicator of
forming than engineering quantities. Engg. quantities depend on the original dimensions of the
sample unlike true quantities. True strain, ε = ln (L/L0) Engg. strain, e = Δ L/L0 = (L-L0)/L0 = (L/L0)
– 1 => 1+e = L/L0
Hence, ε= ln (1+e) (This equation is valid up to maximum load and invalid after that point. This is because
of localization of neck after which the gage strain cannot be referred for measurement); ε = ln (L/L0) is
always useful.
=> Remember volume of solid remains constant during plastic deformation.
Hence εx + εy + εz = ε1 + ε2 + ε3 = 0 during plastic deformation. => This is not true in elastic deformation
regime i.e., Δ = ex + ey + ez = (1-2γ)/E [σx + σy + σz]; Δ will be zero only if γ = 1/2.
True stress, σ = Load / Instantaneous cross-section area = P/A
Engineering stress, S = Load / Initial cross-section area = P/A0
σ = P/A = (P/A0) (A0/A) = S (1+e); where A0/A = L/L0 = volume constancy principle
Strain hardening After yield point, further plastic deformation requires an increasing load but at
decreasing rate. Work hardening strengthens the material, but at the same time area of cross-section is
decreasing. The combined effect of these two phenomenon results in typical load-progression curve.
From yield point to ultimate load, work hardening is dominant. At ultimate load, condition of tensile
instability occurs. Till the ultimate load, deformation of sample gage length is uniform.
Necking–a local constriction begins along the gage section. From this point, incompatibility between
strain hardening and area decrease arises. As a result, the load required for further progression
decreases. This means that load carrying capacity of the sample decreases. After this, practically all
plastic deformation is concentrated in the small necked region. Finally, failure occurs in the necked
region.

Step 1: Initial shape and size of the specimen with no load.

Step 2: Specimen undergoing uniform elongation.
Step 3: Point of maximum load and ultimate tensile strength.
Step 4: The onset of necking (plastic/tensile instability).
Step 5: Specimen fractures. Step 6: Final specimen length.
Fitting stress-strain curve
Hollomon’s equation, σ = K εn where ‘σ’ is the true stress for particular true strain ‘ε’, ‘K’
is the strength coefficient, ‘n’ is the strain hardening exponent. It should be noted that this
eqn. is valid from onset of plastic deformation to maximum load at which necking starts
ln (σ) = lnK + n(lnε) => Y = C + mX Plot graph between ln (σ) and ln (ε)

If the true stress-strain plot is non-linear, it shows that material does not truly obey the Hollomon’s
eqn. and ‘n’ is not a constant. In this case, generally, ‘n’ will be defined w.r.t. strain.
Strain hardening exponent, n = d (lnσ)/d (lnε)
Physical restrictions of the hardening law:
Eqn. is valid between strain of 0.04 and strain at which necking begins. Predicting yield
strength using this eqn. should be avoided. Offset method has to be followed => For eg.,
Yield Strength = K (0.002)n is incorrect. Exclusion of elastic and transition regions leads to
little error.
Yield function
Yield function defines yield surface. Assume that yield surface is closed, smooth surface. At
any instant of time, yield surface is defined as f (σij) = f (σ11, σ22, σ33, σ23, σ13, σ12) = k => This is
6-D surface with each dimension represent one of the stress components. Assume isotropic
material – same properties in all directions; In this case, we can write in terms of principal
stresses only σ1, σ2, σ3 and surface is reduced to 3-D. We need cubic eqn. to relate these stresses
to principal stresses, σ3 – I1σ2 – I2σ – I3 = 0.
The stress invariants are related to principal stresses, I 1 = σ1 + σ2 + σ3;
I2 = - (σ1σ2 + σ2σ3 + σ1σ3); I3 = σ1σ2σ3
With isotropic assumption, we can write k = f (I1, I2, I3) or k = f (σ1, σ2, σ3); First assumption k = f (I 1, I2, I3). Plastic
deformation is pressure independent => Solids under hydrostatic pressure do not deform
plastically. Consider principal stress terms that differ by a pressure term
σ1 => σ1’= σ1 + p/3; σ2 => σ2’ = σ2 + p/3; σ3 => σ3’ = σ3 + p/3
Here hydrostatic pressure = - (σ1+σ2+σ3)/3
Now, σ1’=σ1 –(σ1+σ2+σ3)/3; σ2’=σ2 –(σ1+σ2+σ3)/3; σ3’=σ3 –(σ1+σ2+σ3)/3
It is clear that σi’ is nothing but stress deviatoric components as defined earlier. We can write
yield function in terms of deviatoric components to avoid pressure dependence, i.e., Isotropic,
pressure independent: f (I2’, I3’). Here I1’ = 0 because σ1’+ σ2’+ σ3’ = 0. Now we have reduced
yield surface to a function of two variables, I 2’ & I3’
Effect of temperature in metal forming:
Forming process requires stress above flow stress of the material being deformed. The effect of
external work done on work piece during forming is converted into heat. About 5 to 10% of the
work is stored within as internal energy. Friction can also result in heating and increase in
internal energy of workpiece. Assuming frictionless deformation, the temperature increases
during metal forming operation can be written as:
ΔT = uplastic/ρCp, where uplastic is plastic work done per unit volume of work piece. C pis specific heat and
ρ is density.
With friction,
ΔT = λutotal / ρCp, where λ is fraction of deformation work converted to heat. Normally, λ = 0.95 to
0.98.
Temperature rise is calculated using stress-strain curve, as the plastic work is calculated as the area
under stress-strain curve for plastic flow. For slow deformations, the temperature rise of the work piece
may be small as the heat generated gets dissipated through the die, surrounding air, etc. However,
adiabatic condition may prevail under large deformation speeds, resulting in large rise in temperature
of the work piece. This may cause incipient melting. Therefore, strain rate also influences the
temperature rise during working. For low carbon steel, the temperature rises for a true strain of 1 has
been estimated to be 553 K. This is without heat lost from the billet.
Cold and hot forming:
Cold forming is carried-out at a temperature lower than recrystallization temperature of the work piece material.
Hot working is a process carried out at temperatures above recrystallization temperature, namely, 0.6 T m. High
strain rates – 0.5 to 500 s-1are involved in hot working. No strain hardening takes place in hot working. Processes of
recrystallization, recovery and grain growth dominate in hot working. Energy required for hot working is low, as
flow stress decreases with increase in temperature. Large strains (ε = 2 to 4) are possible in hot forming because of
recovery process. Due to oxidation on surface during hot working, poor surface finish and poor dimensional
tolerances are inherent defects. Die wall chilling may result in non-uniform material flow. Upper limit for hot
working is hot shortness, in which the metal becomes brittle above a certain temperature due to grain boundary
melting or melting of low melting phase such as sulfur in steel.
Metals with high thermal conductivity will require higher working temperatures or rapid working.
During hot working, material softening happens due to two mechanisms – dynamic recovery or dynamic
recrystallization. In dynamic recovery, dislocation cross-slipping, climbing occurs. This mechanism is predominant
in high stacking fault energy metals, with low activation energy for creep. On the other hand, for metals with low
stacking fault energy, like copper, nickel, the dynamic recrystallization is predominant mechanism of softening.
During hot working static recovery can happen in between the working phases, thereby softening the metal. Rapid
cooling after hot working may bypass this static recovery, thereby retaining the high strength of the metal. Strain
induced precipitation or phase transformation can increase the flow stress, reduce ductility. Age hardenable (Al)
alloys are subjected to freezing temperatures before forming, to suppress precipitation during forming. Thermo
mechanical treatments can be used for achieving optimum levels of strength and hardness. Cold working leads to
work hardening. The formed material may have to be annealed to relieve internal stresses and reduce hardness and
strength after cold forming. However, if high strength and hardness are desirable, normally cold worked structure is
retained. Cold working has high dimensional accuracy. Working on a metal at temperatures above room
temperature but below recrystallization temperature is called warm working. Warm working may have the
advantages of reduced working pressures, reduced levels of residual stresses and oxidation, improved surface finish
and dimensional accuracy.
Analysis of forming - Slab Method
Forming of materials is a complex process, involving either biaxial or triaxial state of stress on the material
being formed. Analysis of the forming process therefore is highly involved. Prediction of forming load in a
particular process is rather empirical. However, fairly accurate methods have been developed in order to
predict the forming process and process parameters. Some of the early methods of forming analysis include
slab analysis; slip line field analysis, upper bound analysis etc. With the availability of high-speed computers,
we can depend on finite element method for accurate predictions of forming loads. Numerous metals forming
software have been developed based on finite element procedures for complex shapes with more realistic
boundary conditions.
Slab method is a simple analytical procedure based on principles of mechanics. We can assume a simple
relation between forming load and material flow stress in the form: F = kA, where k is an empirically
determined constant which takes into account friction, redundant deformation etc.
The general methodology involved in slab method can be stated as follows:
First the material under deformation is sliced into infinitesimally small portions. Then force balance is made
on the small element. From force balance a differential equation in terms of the forming stress, geometric
parameters of the billet and friction coefficient is formulated. This differential equation is solved with suitable
boundary conditions. The solution gives us the required forming stress. This method may involve some
simplifying assumptions. Hence this method may be considered approximate. Moreover, it may not be easy to
apply this method for more complex forming processes, such as impression die forging. Slab method is
developed with the assumption that the material flow is homogeneous during forming.

Slab method - Upsetting of a ring

Let us try to understand the slab method of forming analysis with the help of a simple example. Sliding or
Columbic friction often occurs at the material tool interface. As a result of friction the forming load is
enhanced. The flow of material is also non-uniform due to friction. Another type of friction condition, namely,
shear friction or sticking friction could be convenient to consider in the analysis. In shear friction model, we
assume the frictional shear stress to be proportional to shear yield strength of the material. Thus, we have: τ =
mk, where m is friction factor and k is shear yield strength. The following assumptions are the basis of the slab
analysis:
 1. The reference axes are in the directions of the applied stresses
2. Friction does not cause non-uniform deformation. Therefore, material is assumed to deform
homogeneously – a plane remains a plane after deformation.
Consider the homogeneous deformation of a ring-shaped specimen subjected to upsetting force.
Let us assume shear friction at tool-material interface. The ring compression process is widely
used for finding the coefficient of friction for given condition of friction. Consider an elemental
portion of the ring specimen and the various stresses on this element. The following diagram
shows the stresses acting on the elemental part of the ring.

Fig.: Stresses acting on elemental ring subjected to upsetting

Consider a small sector of an elemental ring of radius r, radial thickness dr, height h and the angle of
the sector as dθ. The ring is subjected to upset force F, which is to be determined.
There exists a neutral radius in the ring, such that the material deformation happens towards the axis for radii
less than the neutral radius. There is a decrease in diameter of the ring. For radii greater than the neutral
radius, the material flow is away from the axis-axially outward. This condition exists because of friction.
Therefore, the friction force is observed to act axially outward within the neutral section. It acts radially
inward in sections beyond the neutral section. The force balance along the radial direction gives:
In the above equation the bracketed term represents the factor which accounts for friction
effect during the forming.
 PROJECT WORK /
ASSIGNMENT
• 1. TITLE ----Multi position study table
• 2. SKETCH / DRAWING
• 3. MATERIAL /MACHINE
• 4. HOW TO MAKE IT (Procedure)
• 5. COST
 PROJECT WORK /
ASSIGNMENT
• Develop a production plan for making a
product like soap, door, bread.........
• Develop a strategy for marketting it
• How to do control the quantity & quality
of the product
• Obtain the production plan of any product
in the market
• Suggest an alternate product or process
for the available product