CHAPTER-2

FINITE ELEMENT MODELLING AND SIMULATION OF METAL FORMING FOR

PROCESS AND PRODUCT OPTIMIZATION (HOT UPSETTING, HOT

ExTRUSION, HOT CLOSED DIE FORGING)

2.1 Introduction
The goal in manufacturing research and process & product development is to
determine the optimum and cost effective means of producing sound formed products.
Modelling techniques for metal forming have continuously evolved throughout the
years as discussed in chapter-I , section-l.5. Under the pressure of industry, for
process design and optimization, a considerable effort was devoted to this area and the
progress can be assessed by reviewing the literature [Kob84, 8Sa, 8Sb , 00187, Che92,
Deme93 , Den94, Doe94, Esp94, Owe9S , Fou96c, Fou97, Che98b, Berg98, Chung98,
Fou98 , Ge198 , Gra98 , BarOO, ChouOO, DoeOO, Chun03 , BaWl , BraO!, Bor02, Ant02 ,
Ari03 , Do04, BonOSa, BonOSb, FliOS , Ach06, Fos,]. Today the numerical codes are
becoming more:
• realistic as regards material constitutive equations;
• complete by taking into account various coupling effects, e.g.: thermal coupling,
coupling with tools or coupling in multi-stand rolling;
• user friendly with automatic meshing and re-meshing capabilities;
• reliable by introduction of error estimation and adaptive re-meshing.

Numerical simulation is an important step of the design process, particularly in

forging where its cost is very low compared to industrial experiments. Moreover,
numerical simulation provides extensive information like material flow, tool stresses
and life span or elastic spring back, information that are difficult to obtain otherwise.
It is of great interest to use FEM to anal yze various metallurgical phenomena besides
mechanical and thermal ones. Figure 2.1 presents all the interactions between
phenomena.
 CHAPTER-2

1. Phase transformation

Stress and strain

Fig. 2.1: I.nteractions between stress, strain, temperature and metallic structure.

It also allows process optimization in a very simple and friendly way, for instance to
improve the tool geometries in order to decrease the required energy and increase the
tooling life span.

During the last two decades, FEM simulations of metal forming processes have
become important tools for designing feasible production processes. Because of the
unique capability of FEMs to describe the complex geometry and boundary condition
of the forming process, the FE method has proven itself by numerous success stories
experienced by the industry or reported by research organizations [Hib70, Zie74,
Suz79, Oh82, Axe84, Tek84, Sur87, Kob89b, Cou91, Hans92, Kus94, Pri94, Pas9S,
Rod9S, Lio97, War98, Pat99, Im99, LeeOO, BarOO, DoeOO, KimOO, WagOI, BalDI,
WarOI, LuOI, SanOI, HusOI , PicOI, Lin02, BarOO, ChouOO, DoeOO, KimHOO, SriOO,
Chun03, BalDI, WarOI, BraOI, Bor02, Ant02, Ari03, Do04, Gup04a, Gup04b,
BonOSa, BonOSb, FliOS, GylOS, HeiOS, RepOSa, RepOSb, Ach06, Mad07, Fos], Finite
element simulations in metal forming are currently employed as a virtual
experimentation tool that replaces the physical and costly trial-and-error experiments.

Objective of FE based simulation of the metal forming process on the computer have
been used by the forming industry primarily for providing the comprehensive details
of following:

2-2
 CHAPTER - 2

1. Establish the kinematic relationships (shape, velocities, strain rates and strains)
between the un-deformed part (billet or preform) and the deformed part (product),
i.e. predict metal flow during the forming operation.
2. Establish the limits of formability and producibility, i.e. determine whether it is
possible to perform the forming operation without causing any surface or internal
failures (under fill , lap formation , cracks and folds) in the deforming material.
3. Predict the stresses, forces and the energy necessary to carry out the forming
operation. This information is necessary for tool design and for selecting the
appropriate equipment, with adequate force and energy capabilities, to perform the
forming operation.
4. To analyze die stress for improving the die life.
5. To model process window control (e.g. press speed, forging temperature,
lubricant, and heat treatment procedure) for ensuring that the resulting
microstructure properties (such as grain size) meet the requirement.
6. Non-steady-state deformation can be treated by FE simu lations.
7. To predict temperature increase in the workpiece due to deformation and further
predict the heat transfer between the workpiece and the dies. This can help to
predict tool wear.
8. FE simulations can be used to conduct heat treatment simulations such as
quenching, carburising etc, which help to change the microstructure according to

the requirements of the finished product.

9. FE simulations are more efficient and cheaper than expensive trial and error
experiments to study the effects of various process parameters on the forging
process. It is easier to evaluate alternative designs on the computer rather than
doing the same experimentally.
10. The FE simu lation is used to study metal flow to ensure that the die cavity is
completely filled.

FEM simulations of metal forming processes are classified into to two categories i.e.
elastic-plastic and rigid-plastic analysis. In the elastic-plast ic simulation, material is
modeled as deforming elastic-plastically. Thus, the results are given for not only the
plastic deformation but also for elastic deformation like residual stresses and
springback. Rigid-plastic simulation assumes material to deform only plastically and
in comparison to elastic-plastic simulations, it results in shorter computing time.

2-3
 CHAPTER- 2

Minimizing the time and cost of production trials and design iterations continues to be
the goal of FEM development. Toward this goal, ongoing research is focused in the
areas of computational efficiency, robustness, ease of use and accuracy.

Die / Billet
Geometry
Forging
Initial Billet
Process
Temperature

Boundary
Conditions

Die fill .and Under fill

Lap formation
Strain distribution
Stresses
Forging Load and Energy
Final part Geometry
Temperature Distribution

Fig 2.2: Input and output parameters for FE Simulation of the metal forming
process

These FEM and modern visualization techniques assist the designer in improving the
process design to obtain optimal results. Some of the parameters that the designer can
alter include local die geometry, process parameters i.e. die velocity, friction at the
workpiece and die interface, temperature of billet and die etc. Effective application of
simulation technologies reduces the number of expensive die trials required, thus
leading to a significant reduction in lead time and cost for process design. The FEM
has been successful in the computer modelling of metal forming operations like
forging, extrusion, etc.

In metal forming, plastic strains usually outweigh elastic strains and the idealization
of rigid-plastic or rigid-viscoplastic material behavior is typically acceptable. The
resulting analysis based on this assumption is known as the flow formulation given by

2-4
 CHAPTER -2

Kobayashi et al. [Kob89b]. This formulation is quite useful for metal forming
analysis. In this formulation, variational method is used for the derivation of the basic
equations for finite element analysis. The solution to the boundary value problem is
obtained when the first order variation, bTr of the functional , Tr, vanishes where:

Tr = fO' t dV - f F,u, dS .... 2.1

V S,.-

where,
0' = effective stress,

t= effective strain rate,

F, = surface tractions and
U; =kinematically admissible velocity fie ld.

Using the penalty method to account for the incompressibility constraint and taking
the first order variation results in:

bTr = fO' cSt dV + K f £v b£v dV - f F,u, dS = 0 .... 2.2

v V S,_

where,
K = penalty constant
£v = £/1 = volumetric strain rate
The term K is introduced to keep the volumetric strain nearer to zero. The limiting
strain rate is also introduced to incorporate the incompressibility.

When the velocity solution is obtained, the geometry of the deformed workpiece is
obtained by updating the coordinates of the nodes (Lagrangian mesh system). This
procedure, know as the updated Lagrangian method, is typicall y employed to model
the deformation mechanics, Oh [Oh82] and Surdon [Sur87]. This implies that the
workpiece mesh model is allowed to evolve to represent the material flow. However,
due to the imposition of large plastic strains, typical to bulk forming processes, the
workpiece mesh undergoes severe distortion resulting in a situation that does not
allow the analysis to be continued due to zero or a negative Jacobian in one or more
elements. To continue the analysis, a new, valid mesh representing the deformed
workpiece must be generated using a "remeshing process" .

2-5
 CHAPTER - 2

The main mechanical and thermal equations for metal forming process simulation
analysis are presented here with the two well known approximations i.e. the flow
formulations and the elastic viscoplastic approach. The special case of a rigid but free
tool is considered and a convenient formulation is presented. The general finite
element discretization and time integration schemes are presented. A master and slave
discrete formulation for approximate coupling of part and tool with non co-incident
meshes is analysed.

2.2 Flow formulation and deformation mechanics for the simulation and analysis
of metal forming process
The deformation process governing laws are presented here to recapitulate the
concepts that are being used in the design process. In the analysis of metal forming,
plastic strains usually outweigh elastic strains, and the idealization of rigid-plastic or
rigid-visco-plastic material behavior is acceptable. The resulting analysis based on
this assumption is known as the flow formulation.

2.2.1 Visco plastic flow model

Figure 2.3 shows true stress-true strain curves for isothermal conditions generated
using a viscoplastic flow model. Stress is sensitive to strain rate and elastic
deformation is neglected. Rate sensitivity is applicable to hot metals and the analysis
neglects elastic strain because in hot forging elastic strains are small compared to
plastic strains.

IJight strain rate

Strain
Fig. 2.3: Rigid viscophlstic fl ow model

2-6
 CHAPTER -2

For hot forming the elastic contribution IS often negl ected and the Norton-Hoff
[Nor29, Hof54] equation is used:
r:;. )111-1 .
cr'=2K ( -y3E" E" .. .. 2.3

where:
(J' = deviatoric stress tensor
K = material consistency
m = strain rate sensitivity index

f: = effective strain rate

.... 2.4

Dense materials are considered as incompressible:

div(v)=O .... 2.5
Isotropic work hardening is described by putting:

.... 2.6

where the equivalent strain, E" , is computed by time integration of:

dE" -'-
-=c .... 2.7
dt

Figure 2.4 shows the influence of the rate sensitivity index , m, on the viscoplastic
behaviour defined by the Norton-Hofflaw. The following features are of note.

• With m = 1 flow is equivalent for Newtonian fluid with viscosity, 11 = 3K.

• With m = 0 flow is equivalent to plastic deformation for a material obeying the

Von- Mises criterion with flow stress "0 = K( J3) .

• The general form is with 0 < m < 1: For hot metals (ie when the metal is above
its recrystalisation temperature) m will generally lie between 0.1 and 0.2. The
consistency, K, is a general property used to quantify the resistance of the
metal to permanent deformation.
I11 2
• The dimensions of K depend on the value of m and are M.LT - where L
represents a length dimension, T a time dimension and M a mass dimension.

• The viscoplastic properties of a metal changes with temperature and

deformation.

2-7
 CHAPTER - 2

Il/ = 1

0.7
~--~---+----~----r--~--~4-~~

t----t---t---+-----1-r-~--+_-__.",.j 0.5

1/ / 3

Fig. 2.4:1-d viscopiastic behaviour defined by the Norton-Hoff Law

The FE model used in work uses these variations by defining consistency to be an
explicit function of temperature and strain. All the laws available and their definitions
are written in table 2.1.

It is also possible to define more generalised flow behaviour by coding any general
rule using FORTRAN in a user defined subroutine which can then be incorporated
into the main program by recompiling the finite element solver with the appropriate
modifications.

2-8
 CHAPTER-2

Constant consistency I<(I.£:)K,

Exponential Tcrna Law K( 1.1;) - 1\. ( i·;'- r:: )' C 1\ I

Strain Hardening Power Law 1I1

K(Li;-)=I\. ,/f: If:!) )"c '

Linear Strain Ihlnlening La\\ 1<(I.f )- I(, (I -. ;ll:)"C-I'i

Limited Linear Strain Hardeni ng La" I<.(I '-;;)- K,( 1- :I!:)c 1'1 jl'T; ST: 111:\\

- - -Iii
K( I'.T:) K \,( 1 ~t l'III,I\' )1.2 iI'r-~ --- T
::::... .... 111<1\

Solid Polymer La,,\' ;\1

K(T.~:) = K (i c r (I +e-";' )e h(

Interpohltion Funct ion K ( LT:)= K.); I (J)( I _c"" )e ll:

Linear Power Strain Hardening Law I'

I
I\.( I J:)-I\.u( 1TnT:" )c

Tahle 2.1: Material Laws

Where,
'£=effective strain, Ko=constant term, T =Temperature [K],

~ = temperature term, n =strain hardening index, w = strain hardening term,

a l = strain hardening term, h = strain hardening term, 9 = temperature [DC] ,

EO= strain hardening regulation term, k9(9) = interpolation function,

a =exponential temperature term

2.2.2 Elastoplastic and elastic visco plastic behaviour

For a more accurate description of the material behavior, the elastic effects must be
introduced. The additive decomposition of the strain rate tensor can be considered as a
satisfactory approximation for usual metallic materials:
.. .. 2.8

Where, £e is elastic part of the strain rate tensor, £Pis the plastic (or viscoplastic) part
of the strain rate tensor.

The material behaviour is independent of strain rate and elastic deformation is

included. This model, shown in figure 2.5, applies to most metals at room
temperature,

2-9
 CHAPTER-2

rj

Strain
Fig. 2.5: Elastic-plastic flow model
The elastic contribution obeys a rate form of the Hooke Law:
da
_J_ = "A e trace (£e )I + 2/-l e£e = De:£e . ... 2.9
dt

where A,e and j.. / are the usual Lame coefficients, I is the unit (second rank) tensor, De

is the resulting fourth rank elastic tensor. The Juman derivative of the stress tensor in
Eqn.2.9 is defined by:
da
_J_ = 0- - CDa + aCD .. .. 2.10
dt
where CD is the spin tensor. When the process involves moderate rotations, the simple
material derivative can be used in Eqn.2.9 as a satisfactory approximation.

For an elastoplastic material, the irreversible plastic deformation is computed with the
help of the Von Mises plastic yield function, f, defined by:

f(a) = 3I a;r2a~ (E)

ij
.... 2.11

Two regimes must be distinguished:

2.2.1.1 Deformation is purely elastic

£P = 0 iff(a) < 0,

or if (f(a) = 0 and af : 0- < 0) .... 2.12

8a
2.2.1.2 Deformation is elastoplastic

£P = "A P af if (f(a) = 0 and af :o- ~ 0) .... 2.13

8a 8a

2-10
 CHAPTER-2

where, the plastic multiplier A, Pis positive.

When the stress field depends on the strain rate, the irreversible strain rate is
expressed with viscoplastic law:

£P = _1 (0 -R )+,-I
_ eq_ 0' .... 2.14
K K
where the bracket function is defined by:
(x) = 0 ifx=O and (x) = x if x ~ 0 .... 2.15

the equivalent stress by:

.. ..2.16

and R is a function of "£ .

For materials exhibiting kinematic hardening, Eqn.2.14 and Eqn.2.16 must be

generalized, they can be replaced respectively by:

.... 2.17

and

o eq = .% ~::c
lJ
0;j2 - ~ i/ .... 2.18

Where, ~ is the back stress tensor, which is governed by a differential law of

hardening in term of the strain rate tensor.

2.2.3 Unilateral contact and friction behavior

We define the velocity difference at the interface oQe between the part and tools by:
Vs = v - v taal .... 2.19

The contact condition is written:

vs.n = 0 if (0.n).n = 0 .... 2.20

while for a loss of contact:

vs.n = 0 if(0.n) .n = 0 .... 2.21

where, ' n' is the normal to the contact surface, which is outside the work piece.

2-11
 CHAPTER - 2

The viscoplastic friction behavior is described by a non-linear relation between the

shear stress vector, r and the sliding velocity, Vs

.... 2.22

where,

Uf = the viscoplastic friction coefficient, can be a function of the normal stress Gn

q = the sensitivity parameter to the sliding velocity (which is often taken equal to m) .
It is easy to convince oneself that this value includes the Coulomb special case.

2.2.3.1 Coulomb Friction.

Friction is proportional to the stress normal to the surface of the workpiece. The
proportionality is defined by a coefficient of friction which is ass umed to be constant
throughout the metal forming operation.
. ... 2.23

where,

'tf = friction stress tangential to the surface

Il= coefficient of friction
an = compressive stress normal to the surface (contact pressure)

This law is not suitable to bulk metal forming because high contact pressures are
involved. At high contact pressures the Coulomb law predicts friction stresses greater
than the shear strength of the metal and sticking rather than sliding is modelled at the
interface. Coulomb friction is more applicable to sheet metal forming where surface

pressures are lower.

2.2.3.2 Tresca Friction.

Friction stress at the contact surface is equal to a fraction of the shear yield stress of
the workpiece material. It is calculated using a constant friction factor.

't f = rnk .... 2.24

where,
m' = friction factor
k = shear yie ld stress of material

2-12
 CHAPTER-2

This law is suitable to bulk metal forming because, unlike Coulomb friction, the
amount of friction is independent of the normal stress at the surface.

2.2.4 Thermal Analysis

The classical heat equation for deformable bodies is written simply:

pc dT =div (kgrad(T))+qv . .. .2.25

dt
Where, the heat dissipation represents a fraction fw (with 0 ~ fw ~ 1) of the irreversible
mechanical rate of work:

.... 2.26

and for viscoplastic materials it is:

.... 2.27

The constitutive law depends on temperature, for example we put in Eqn.2.3:

. . .. 2.28

For viscoplastic materials, the incompressibility Eqn.2.5 is replaced by the mass

conservation condition, taking into account linear thermal dilation with coefficient

trace(£) = div(v) = 3a d t .. ..2.29

The radiation condition on the free surface aos is written:

.. .. 2.30

where,

Cr = emissivity parameter,

CJ r = Stephan constant and

To = outside temperature.
On the surface of contact aOe , due to conduction with the tool s with temperature

Tloot. and friction energy dissipation governed by the effusivity of the part, b, and of
the tool, bloot. the condition is:

-k -aT = h ed (T - Ttoo, ) + b a rK I6.v Iq+1 .... 2.31

an b+b too'

In fact the parameter hCd, governing the exchanges between part and tools, IS a

function of the normal stress, CJ n , which must be determined experimentally.

2-13
 CHAPTER - 2

2.2.5 Integral formulations

The viscoplastic equations can be transformed into an integral mixed formulation; for
any virtual velocity field v*:

rV(x,£",v,p) = f2K(J3t t· '£: £* dV + f ucKlvsIP" vsv* dS - fpdiv(v*) dV = 0

n an, n

.... 2.32
and for any virtual pressure field p*:

rP(x, v, p) = fp*div(v) dV = 0 .... 2.33

n

Corresponding integral formulations, in term of the velocity field , can be derived for
elastoplastic or elastic viscoplastic behavior as it is shown in Chenot [Che98a, b]. But
most often a displacement formulation is preferred for simplicity.

2.2.6 Free tool Problem

We suppose that, in addition to the tools with a prescribed velocity VtooJ, there is one
(or several) free tool with an unknown velocity, Vfree, which is determined by the
contact condition with the work-piece on the boundary, an llee . We must impose the

non penetration condition on an'i-eo :

(v-v Cree ).n = 0 .... 2.34

when there is contact. When a penalty formulation, with penalty factor, PP' is chosen

for imposing the contact conditions given by Eqn.2.20 and Eqn.2.21, the integral
equation becomes:

rV(x,£", x Cree ' v, p, v free ) = f2K(J3tt·'£: £* dV + f ucKlvJ" vsv* dS - fp div(v*) dV

n an, n

+Pp f ((v-vrree )·n)(v*.n)dS = O .... 2.35

anfree
and for any virtual velocity V*free of the free tool :

rvfree eX, XCree , v, v Cree ) = Pp f(eVlrec- v).n) ev\ee· n)dS = 0 .... 2.36
ancree
where , X Cree I' S the coordinate vector of the centroid of the free tool.

2-14
 CHAPTER - 2

2.2.7 Coupling between work piece and tools

We consider here the case of a contact between two deformable bodies ~·i , with

velocity field VI, and Q 2 with velocity field i, through the contact surface r c can be

expressed as:

(Vi -v 2 ).n ~O if0 n =0

(Vi - v 2).n = 0 if on ~ 0 .... 2.37

which can be imposed practically by a penalty formulation . The mixed formulation is

written on the whole domain Q = QI U Q 2, with a(discontinuous) ve locity field v:

rV = f2K(fit )",-1£: £* dV - fp div(v*) dV + f((J"/+~t D: £/) £* dV + farK Ivl- v* dS I

Vs
n' n n' r,
+ p f((V I - v 2)·n)v*.n dS = 0 .... 2.38
r,

In which the third integral corresponds to the elastic behavior of the tool , and E' is the

strain rate deviatoric tensor and

rV(x, v, p) = fp* div(v) dV + fp * (p-~t K div(v) dV = 0 .. .. 2.39

n' n'

2.2.8 Finite Element Formulation

2.2.8.1 Space discretization
Utilizing usual isoparametric elements, the velocity field is discretised in term of the
nodal velocity vectors Vn and the shape function N n:
.... 2.40

~ is the local co-ordinate vector; the mapping with the physical space is defined by :

x = LXllNn(~) .. .. 2.41

and the strain rate tensor is computed with the help of the B linear operator;

£= LVnBn .... 2.42

The pressure field is discretized in term of nodal pressure Pill' with compatible shape

function Mill:
.... 2.43
m

2-15
 CHAPTER - 2

from Eqn.2.32 and Eqn.2.33 , the discretized mixed formulation for viscoplastic
material gives the following set of non linear equations:

R ~= f2K(.J3tyn-'£: BndY - f u rKlvl-' vsNndS - fptr(B Il

) dY = 0 . ... 2.44
n ~ n

R ~, = fM\1ldiv(v) dY =O .... 2.45

n

which takes the symbolic form:

R(X, £, V, P) = 0 ., .. 2.46

The time evolution strain is governed by Eqn.2.7 , and co-ordinate vectors by:

dX =Y ... .2.47
dt
For an elastoplastic or elastic viscoplastic material , the general form of the discretized
integral equation is:
R(X, £, V, P) = 0 ... .2.48

the equations of evolution being Eqn.2.7, Eqn.2.45 and the stress evolution equation:

do-
= 0- .... 2.49
dt

2.2.8.2 Time integration

For a viscoplastic materials, the velocity and pressure vectors are obtained after
solving Eqn.2.44, using the well known iterative Newton-Raphson linearization
method. Nodal and equivalent strain updates are performed according to Eqn.2.7 and
Eqn.2.47.
The popular one step Euler scheme gives the following procedure:
• at time t, Xl is known,
• compute yl from Eqn.2.46 written at time t, i.e.:
.... 2.50

• update co-ordinates vector X and equivalent strain according to:

.... 2.51

and:
\+61 I • I

I:> = £ + tlt£ .. .. 2.52

2-16
 CHAPTER-2

More accurate schemes can be easily applied, for example a two step Runge and
Kutta, Chenot [Che98a, c]. For elastoplastic of elastic viscoplastic materials, more
implicit schemes must be used to avoid violation of the yield criterion.

2.2.9 Free tools coupling

The discrete analog ofEqn.2.35 and Eqn.2.36 re-easily obtained:

R~ (X, £, x rree' V, P, v free )= f 2K(J3tyn-'£: BlldV - f urK!vs!P. ' vsNlldS - fptr(BIl)dV

Q an, Q

+p f ((v - vrree )·n ) .ndS = 0 .... 2.53

a!'lrrcc

and

RVrrcc (X, xrree, V, P, vfree)=p f (Cv rree -v).n)ndS .... 2.54

OOfn.'C

We obtain a non-linear system of equations, the symbolic form of which will be

written:
R(X, £, x frce' V, P, v rree ) = 0 ... .2.55

The time evolution equations are:

dx ... .2.56
~=v
dt free

Remark: if the free tool also possesses rotational degrees of freedom , additional
equations and differential equations must be introduced.

2.2.9 Master-slave coupling with the tools

We define first the master surface r c ' which is here the discretized boundary of ~i
and 7r , the projection operator on r c . A gap function g is also introduced to define
the distance of a material point from the master surface:
g(XI , t) = (X l -n(xl )).n l .. ..2.57

where n I is the normal vector to rc

The incremental contact condition is expressed by, Pichelin et. al. [PicO 1] as :

(vl(xl)-v 2(n(xl)).nl _ g(XI, t):sO ... .2.58

~t

so that the incremental penalty formulation will be obtained by replacing in Eqn 2.33
the penalty term by the following integral:

2-17
 CHAPTER-2

. . .. 2.59

In fact a nodal approximation of equation-54 is preferable in the form :

.... 2.60

where Sj is a surface area surrounding node number I, such that if Sc is the area of rc
we have the equality:

.. .. 2.61

The main mechanical and thermal equations (Eqns.) governmg metal flow during
forming processes, and the finite element discretization and time integration
procedures are summarized briefly.

2.3 Implementation of FEM analysis and simulation of metal forming processes

Finite Element Method (FEM) technique can be applied in analysis and simulation of
metal forming processes for a large range of boundary value problems with little
restriction on workpiece geometry . The ultimate advantage of computer aided design
and analysis in metal forming processes is achieved when reasonably accurate and
inexpensive computer software is available for simulating metal flow throughout a
particular metal forming process. Thus, forming experiments can be simulated on a
computer by simulating the finish formed parts that would result from an assumed set
of process parameters. A graphical display helps to better understand the metal
forming process being simulated.

In practical metal forming processes, a number of operations (preform steps) are

required to transform the initial simple billet geometry into a complex final geometry
while achieving desired tolerances and properties. The desired workpiece shapes are
obtained by using dies of various shapes. Therefore, a method of analysis that can
treat the boundary conditions of dies is necessary in order to full y utilize the
advantage of the finite element method in metal forming analysi s.

Numerous commercial FEM codes that are available which are used extensively in
simulating metal forming process analysis are specified in table 2.2 metal forming
processes. The user is provided with a library of data containing material properties

2-18
 CHAPTER -2

algebraic equations, which are to be solved , from the governing equations of the
boundary value problem. The governing equations include conservation principles,
kinematic relations, and constitutive relations. The FE preprocessor then solves the set
of linear algebraic equations to obtain the state variables at the nodes. It also evaluates
the flux quantities inside each element.

The following steps are pursued in this operation:

i) A suitable interpolation function is assumed for each of the dependent variables in
tenns of the nodal values.
ii) Kinematic and constitutive relations are satisfied within each element.
iii) Using work or energy principles, stiffness matrices and equivalent nodal loads are
established.
iv) Equations are solved for nodal values of the dependent variables.

2.3.3 Postprocessor
This operation prints and plots the values of state variables and fluxes in the meshed
domain. Reactions may be evaluated. Output may be in the form of data tables or as
contour plots.

2-20
 CHAPTER-2

~ )
CrT
,"I L.. Analysis and
Start
'1
Problem
I' d~sign decisions
,..... Stop
"Is "
I~

Pre-processor
" Processor Post-processor

• Reads process control

parameters.
• Generates element shape
functions.
• Prints and plots contours for
state variables.

• Reads or generates nodal • Calculates master element

equations.
• Returns element domain and
calculates flux and other
co-ordinates.
,.... I~
variables .
• Reads or generates element • Calculates transformation
• Prints and plots contours for
data. matrices.
flux and other physical
conditions.
• Reads material constants. • Maps element equations into
global system. • Evaluates and prints error
• Reads boundary conditions.
• Assembles element equations.
bounds.

• Introduces boundary conditions.

• Perfonns solution procedures.

Fig. 2.6: Implementation of FEA for metal forming analysis.

2-2 1
 CHAPTER-2

2.3.4 The FE Method

• Geometry of the workpiece (and dies in some cases) is meshed into a number of
elements. Dies are meshed only if die stress/wear analysis is required.
• Elements store data such as amount of deformation (related to work hardening) .
• Nodes are points, which connect elements. Nodes store variables such as
temperature and contact between the workpiece and tools. More elements: longer
simulation time.
• Boundary conditions (e.g. velocity, force, temperature) are applied to the surface
nodes.
• Material models are applied to elements.
• Forming equipment models are applied to dies.
2.3.4.1 Inputs in FEM
• Workpiece material properties
• Processing conditions (preform & tool geometries, die material, temperature)
• Friction (lubrication)

2.3.4.2
Outputs in FEM
Workpiece shape, die fill, forming load and energy, temperature, sprmg back,
mechanical properties, die stresses, fracture potential

2.4 Issues in the application of process simulation

2.4.1 Geometry
The first step in simulating any metal forming process is analyzing the geometry of
the workpiece. if the simulation is 2-D then one has to decide whether to simulate it as
axisymmetric (such as circular sections) or plane strain (long sections). The geometry
can either be drawn in the preprocessor of the FE code or imported from a solid
modelling software.

2.4.2 Mesh and remesh

The geometries to be studied have to be meshed with a sufficient number of regular
elements. There is a direct correlation between the number of elements and the
accuracy of the simulation, however a large number of elements will also increase
computing time. Hence one needs to reach a compromise between the two. The

2-22
 CHAPTER -2

number of elements in areas of interest should be high such as areas of high stress
concentration, corners etc. It is a general rule that there should be at least three
elements at any radius of the geometry. Remeshing is required when large strains are
encountered and the elements get highly distorted such as during shearing operations.

2.4.3 Workpiece and tool material properties

Another preliminary step in setting up a simulation is selecting the material models
for the objects involved in the process. The various options are rigid, elastic-plastic,
rigid-plastic, viscoplastic etc. The choice of the material model depends on the role of
the object to be studied in the actual process e.g. in die stress analysis the dies would
be specified as elastic to study the elastic deformation and stresses induced during
processing. The flow stress data can either be selected from the material data library
or can be user defined. It should be selected depending upon the operating
temperature since for hot forming operations the flow stress is dependent on
temperature and strain rate. This is a very crucial step since the flow stress defines the
material behavior during the forming process.

2.4.4 Boundary conditions

These include the relative positioning of the objects, movement, loading conditions,
interface friction etc. Proper constraints need to be specified depending upon the
process requirement. The various interface frictions affect the metal flow and should
be appropriately specified. Usually in bulk forming operations we use the shear
coefficient based on the constant shear stress law.

2.4.S Characteristics of the codeFE codes can be either implicit or explicit. Implicit
codes are slower but more accurate than the faster explicit type .

2.S FORGE2
FORGE2 is a domain specific two dimensional Finite Element (FE) software
designed to simulate axi-symmetric and plane-strain metal forging [Atk81]. It can
perform rigid-viscoplastic analysis which neglect the elastic deformation of the
workpiece and elastic-plastic analysis which include elastic effects. A rigid-
viscoplastic analysis is applicable to hot forging where elastic effects are
overshadowed by thermal effects and the large plastic deformations involved. Elastic-

2-23
 CHAPTER -2

plastic analysis are applicable to cold forging operations. Elasticity is included

because elastic effects such as residual stress and springback can become significant
in cold forging.

For rigid-viscoplastic simulations, FORGE2 uses a viscoplastic formulation based on

the Norton-Hoff [Nor29, Hof54] law to model the friction between die and workpiece.
Its elastic-plastic solver provides three alternative models of friction, Coulomb,
Tresca and the same viscoplastic law used in rigid-viscoplastic simulations.

In a FORGE2 model of a forging sequence the workpiece is represented by a

deformable mesh of 20 elements and the dies by lines which define the rigid die
surfaces. Mechanical and thermal properties are ascribed to the mesh and constants
which define thermal and mechanical (friction) effects at the surface of the workpiece
are given values. Once initial temperature conditions are prescribed the dies are
moved in small incremental steps and a solution calculated for each step. The die and
ambient temperatures are assumed to remain constant throughout the simulation.

FORGE2 is used extensively for research and can be applied to industrial forging
operations [Han92, Ab088, Sek93 , Mi194, Oen94, Kus94c, Han96, Mah94, Han99,
HanOOa, b, c, Alv02, Gup02, Tie02, Abd03, Oua04, Gup04a, Sri04a, Pen04, Mi104,
Gup05 , Gup04b, Fli05 , Gy105 , Hei05 , Gup06, Kis06, Mad07].

In a FORGE2 simulation of a forging sequence the workpiece is represented by a 20

finite element mesh. The dies, which are assumed to be rigid, are defined by lines
which represent the die surfaces. Mechanical and thermal properties are ascribed to
the mesh and constants which define thermal and mechanical (i.e. friction) effects at
the surface of the workpiece are given values. Once initial temperature conditions are
prescribed the dies are moved in small incremental steps and thermal and mechanical
solutions are calculated at each step. Since the mechanical properties of the metal are
dependent on its temperature and the thermal solution depends on the geometry and
heat generated by friction and plastic working of the material the two solutions are
interdependent. The way FORGE2 maintains thi s linkage is as follows: At any given
increment the temperature field is known. The mechanical solution is obtained using
the mechanical properties for this temperature field. The workpiece is then deformed

2-24
 CHAPTER-2

incrementally according to the strain rates calculated at each node and the thermal
analysis to produce a new temperature field is performed simultaneously.

Unlike most commercial finite element packages FORGE2 is not a general purpose
code. It is designed specifically for the analysis of hot and cold metal forming. This
means that key phenomena such as plasticity, viscoplasticity, friction and thermo-
mechanical linkage are incorporated in the code. Some sophisticated general purpose
packages will model all of these effects but because they need to be applicable to a
much wider range of structural and thermal analysis than FORGE2 they are not as
easy to apply to the analysis of metal forming. FORGE2 also incorporates an
automatic remesher. Remeshing becomes essential to avoid catastrophic element
distortion when analysing large scale deformations which occur in industrial closed
die hot forging operations.

Billet! die Die / die

Friction factor
temperatures shape

• Metal flow
• Forming load
• Energy
• Product quality
• Die life

Fig. 2.7: Interaction among state variables and uncertainty sources

2.6 Finite element modelling and simulation of hot upsetting process

Upsetting is defined as ' free forming, by which a billet or a portion of a workpiece is
reduced in height between usually plane, parallel platens" [ASM88]. Upsetting is a
basic deformation process which can be varied in many ways. A large segment of
industry primarily depends on the upsetting process for producing parts ranging in
complexity from simple bolts, screws, nuts, rivets or flanged shafts to wrench sockets
that require simultaneous upsetting and piercing. Hot upsetting is also occasionally

2-25
 CHAPTER-2

used as a finishing operation following hammer or press forg ing, such as in making
crankshafts. A sketch of the upsetting process is shown in figure 2.8.

Top Die (Moving)

~-----Initia l Billet

+-----Rillet after Upset

Fig. 2.8: Upsetting process.

In upsetting a rigid tool is pushed onto a block of material (billet). The material is free
to move at the right hand side.
Successful upsetting mainly depends on two process limitations:
1. Upset strain, s: that affects the forming limit or forgeability of the workpiece
material,

... .2.62

2. Upset Ratio (Ru): that affects the buckling of the workpiece

R, =( ;: J .. .. 2.63

.
Die Velocity

Fig. 2.9: Metal flow in non-steady state upset forging processes

In upsetting a ratio of Ru ~ 2.3 can be achieved in one hit if the deformation occurs

over a portion of the workpiece. Larger values of Ru require several deformation

stages. In upsetting the parameters that are significant are dimensions of the

2-26
 CHAPTER - 2

workpiece} its strength, its formability, the required upset ratio, the desired accuracy
and the surface quality. When forming in several stages, the design of the heading
preforms affects the fiber structure of the final shape. Heading preforms are to be
shaped such that the workpiece is guided correctly to avoid buckling and folding
[Lan85].

In metal forming, the flow of metal is caused by the pressure transmitted from the dies
to the deforming workpiece. Therefore, ,the frictional conditions at the die/workpiece
interface greatly influence metal flow, formation of surface and internal defects,
stresses acting on the dies, and load and energy requirements [Alt83 ].

A number of FEM simulations are performed for hot upsetting of a ck-45 steel

cylindrical preform at temperatures varying from 350 °C to 1300°C under different

punch velocities (1 to 5mm/s) and friction conditions (where ).!, the friction co-
efficient is 0.1 to 0.8) using the model described in section 2.2. The dies are kept at

350°C.

The geometry is axi-symmetric in nature so only one half of the section is simulated.
The billet height is initially kept at 10 mm and the final height is 5 mm. The diameter
of the billet is 20 mm. The forging force required for 50% reduction at different punch

velocities, at temperatures varying from 350°C to 1300 °c under varying friction

conditions is obtained (table 2.3). The variation of forging force with change in
temperature [Figure 2.10], the effect of friction co-efficient on forging force [Figure
2.11] and the influence of variation in die velocity on forging force [Figure 2.l2] are
plotted.

2-27
 CHAPTER-2

9
Velocity of die "" 2 mm!scc
Co-efficient of friction = 0,2
8

7
'"
<!.l
E 6
.8
-0"
ro
..s: 5
OJ)
t::
'5iJ
...
C
I.L.

1200 "C:

o~------~--------~------~--------~------~
5 6
..
7 8

Displacement of dic, rum

9 10

Fig.2.10: Effect of billet temperature on the evolution of forging force during hot
upsetting

45 ~------~-------r-------r------~~----~
Velocit y of die ~ I nUll isec
Tempemture oCbilkt = 350 "C

<Jl
<!.l
E
.8
-0
ro 25
..s:
OJ)
t::
'M
I-; 20
C
I.L.

IS

o ~ ______-L________ ~ ______ ~ ________ ~ ______ ~

5 6 7 8 9 10
..
Displacement of die, mm

Fig.2.11: Effect of friction co-efficient on the evolution of forging force during

hot upsetting

2-28
 CHAPTER - 2

10
Temperature of bille t = 1200 <Ie
9 co-efficient of tiction = 0.2

7
..,v:
r::
r:: 6
B
-0"
c;:
.!2 5
eJ)
r::
'0'
....0
0 4
t.L.

2 V = 2mm!scc

_. mm . V =3 mm!scc
·c

0
5 6 7 8 9 10

Displacement of die, mm

Fig. 2.12: Effect offriction co-efficient on the evolution offorging force during
hot upsetting

The radial compressive stress contours and equivalent stress contours indicate that
radial stress is maximum at the axis of symmetry and equivalent stress is maximum at
the comers indicating folding tendency of upset. The hydrostatic pressure contours
also depict that the maximum hydrostatic pressure occurs at centre of the specimen
along its axis of symmetry and at edges and minimum along the bulge of the upset.
The temperature contours indicate that the maximum temperature is at the central
region of the work piece away from the dies and free surface. It is minimum near the
dies due to the heat loses occurring at the die interface. The results of this analysis
also confirm that at high temperature the effect of heat conduction is largely more
important than those related to heat generation due to conversion of plastic work. The
temperature and equivalent strain contours are very significant in bulk forming as they
can be used to predict the phase changes inside the billet. The processing maps
indicate the problem regions in the billet as the defects may develop at unstable
regions and hence, the need for corrective measures in choosing process parameters.
Principal stress contours depict that maximum principal stress is compressive in
nature as is expected in upsetting and occurs at the bulge.

2-29
 C HAPTER- 2

Die Forging load (To nn es) for following initial temperatures (uC) of the ck-45 steel
Coeff. of billet at 50% deformation
Ve locity
Friction
mm/sec 350 400 500 600 700 800 900 1000 1100 1200 1300
I 32. 14 29.74 25.79 22.3 1 19.63 16.85 14.87 12.49 10.88 9.72 8.38
2 8.54 7.75 6.37 5.23 4.3 3.53 2.89 2.37 1.95 1.6 1.32
0.2 3 4 .65 4.18 3.37 2.72 2.19 1.77 1.42 1.15 0.93 0.75 0.61
4 3. 16 2.83 2.27 1.81 1.45 1.1 6 0.93 0.75 0.6 0.48 0.38
5 2 .25 2.0 1 1.59 1.26 1 0.8 0.63 0.5 0.4 0.32 0.25
1 33 .57 3 1.44 29.93 24 .17 2 1.08 17.84 15.82 13.98 12.08 10.38 8.84
2 8.87 8.05 6.61 5.43 4.46 3.65 ...,) 2.46 2.02 1.6 1.36
0.3 3 4 .82 4.33 3.49 2.82 2.27 1.83 1.48 1.19 0.96 0.78 0.63
4 3.27 2.93 2.34 1.88 1.5 1.2 0.96 0.77 0.62 0.5 0.4
5 2.33 2.08 1. 65 1.31 1.04 0.83 0.66 0.52 0.4 1 0.33 0.26
1 36.53 34.33 29.6 1 25. 18 21.79 19.54 16.63 14 .7 12.4 10 .7 1 9.4
2 9. 12 8.27 6.8 5.58 4.57 3.75 3.07 2.52 2.07 1.7 1.39
0.4 3 4.96 4.45 3.59 2.9 2.34 1.88 1.52 1.22 0.99 0.8 0.65
4 3.36 3.0 1 2.4 1 1.93 1.57 1.23 0.99 0.79 0.64 0.51 0.41
5 2.39 2.13 1.69 1.34 1.06 0.85 0.67 0.54 0.44 0.34 0.27
I 37.92 35.06 3 1.08 26.67 23.76 20 .5 17 .74 15.2 13.02 11 .56 9.48
2 9.42 8.54 7.01 5.75 4.7 3.85 3. 15 2.57 2. 11 1. 73 1.41
0.5 3 5.07 4.56 3.67 2.96 2.39 1.92 1.53 1.25 1.0 I 0.82 0.66
4 3.44 3.08 2.46 1.97 1.58 1.26 1.0 I 0.81 0.65 0.52 0.42
5 2.44 2. 18 1.73 1.38 1.09 0.87 0.69 0.55 0.43 0.35 0.27
I 39.08 37.49 33 .56 27.91 24.93 21 .22 18 .19 16.02 13.75 I 1. 9 1 10 .34
2 9.63 8.74 7. 18 5.89 4.83 3.96 3.25 2.57 2.09 1.7 1.39
0.6 3 5.25 4.63 3.73 3.01 2.42 1.95 1.57 1.27 1.02 0.83 0.67
4 3.5 3. 13 2.5 1 2.0 1 1.6 1.28 1. 02 0.82 0.66 0.53 0.42
5 2.49 2.22 1.76 1.4 I. I 1 0.89 0.78 0.56 0.44 0.3 5 0.28
1 40.86 39.23 34.33 30.29 25.2 1 22.63 19.63 17.33 14.57 12.55 10.86
2 9.69 8.78 7. 19 6.09 4.99 4.08 3.34 2.7 1 2.23 1.83 1.5 1
0.7 3 5.2 4.67 3.76 3.03 2.44 1.92 1. 58 1.27 1.03 0.83 0.67
4 3.55 3. 18 2.54 2.03 1.62 1.3 1.04 0.83 0.67 0.53 0.43
5 2.53 2.25 1.79 1.42 1.12 0.89 0.71 0.56 0.45 0.36 0.28
I 44.22 41.45 36.32 30.93 26.84 23.85 20.4 17.6 15. 18 12.89 I 1.1
2 9.89 8.96 7.34 6. 13 5.06 4. 14 3.33 2.73 2.24 1.84 1.52
0.8 3 5.15 4.63 3.73 3.01 2.42 1.88 1.57 1.27 1.02 0.82 0.67
4 3.57 3.2 2.56 2.04 1.63 1.3 1.04 0.84 0.67 0.53 0.43
5 2.7 1 2.42 1.93 1.5 3 1.22 0.97 0.78 0.62 0.49 0.39 0.3 1
Table 2.3: Forging load (tonnes) required in hot upsetting for 50% deformation
at different input conditions

2-30
 CHAPTER-2

DAYALBAGH_INSTITUTE

INCR. COMPUT No: 1 h = 10.00(inm) t = .0000(s) FILE : upset.don

< ....;....,--:----r-:-,,---,'-,--.......,..-..,....------...............- - -.....:.;;;;:-. -'CC:;'''

1"--

i'·:~·/
(,
1.+.:..

l~Y\« ..
.. ", T'-

I .. .:,. ./.. </ •

.... .
t"· • .. j'-:
I ~' ~ '
-.• j. -) "

I,.
C. !JO I' ,
L ........••:........................ L. ........ ,.........,
FORGE2 V2.7.4

DAYALBAGH_INSllTUTE

INCR. COMPUT No: 4 h = 5.000(mm) t = 5.000(s) FILE: upset.don

I,
I
I,
I,
! '" , ";"". . . .:, .. -

"j

.- .. ,- .. ~

", /

" ......J ..........~.. J.... j

FORGE2 V2.7.4
Fig. 2.13: Sample simulation of initial die/billet setup and tinal stage for
simulation hot upsetting for 50% deformation depicting automatic meshing and
rcmcshing

2-31
 CHAPTER-2

DAYALBAGH_INSTITUTE
HYDROSTATIC PRESSURE [Mpa) [P>O <=>compression)
INCR. COMPUT No : 4 h = 5.000[mm) 1= 5.000(s) FILE : upset.don

P = -20.96

P = -83.57

{> Maxi= 417.4

o Mini= -146.2

5.00

FORGE2 V2 .7.4
Fig. 2.14: Isocontours of hydrostatic pressure during hot upsetting process.

DAYALBAGH_INSTITUTE
TEMPERATURES [Celcius)
INCR. COMPUT No: 4 h = 5.000[mm) t = 5.000(5) FILE: upset.don

Tem = 466.9

Tem = 465.5

{> Maxi= 476 .6

o Mini= 464.1

(j 5.00

FORGE2 V2.7.4
Fig. 2.15: Isocontours of temperature distribution during hot upsetting process.

2-32
 CHAPTER-2

DAYALBAGH_INSTITUTE
EQUIVALENT STRESS SIGMAequivalent (Mpa)
INCR. COMPUT No : 4 h = 5.000(mm) t = 5. 000(5) FILE : up set.don

Seq= 335.4

Seq= 317.8

L> Maxi= 458.0

o Mini= 300.3

5.0n

FORGE2 V2.7 .4
Fig. 2.16:Isocontours of equivalent stress during hot upsetting process.

DAYALBAGH_INSTITUTE
EQUIVALENT STRAIN RATE [sec -1)
INCR. COMPUT No : 4 h = 5.000(mm) t = 5.000(5) FILE: upset.don

P= .4053

P= .3660

EbP= .1698

EbP= .1306

L> Max i= .4445

o Mini= 9.1349E-02

5.00

V2.7
Fig. 2.17: Isocontours of equivalent strain rate during hot upsetting process.

2-33
 CHAPTER-2

OAYALBAGH_INSTITUTE
EOUIVALENT STRAIN (no unit]
INCR. COMPUT No: 4 h = 5.000(mm) t = 5.000(s) FILE : upset.don

= 1.427

= 1.295
= 1.163

= 1.031
= .8997
= .7679

A Maxi= 1.558
o Mini= .3727

5.00

2 V2.7
Fig. 2.18: Isocontours of equivalent strain during hot upsetting process.

2.7 Study of effect of various hid ratios in hot upsetting process using finite
element modelling and simulation
A number of FEM simulations are performed for hot upsetting of a ck-45 steel

cylindrical preform at temperatures varying from 400 0C to 1200 0C under different

punch velocities (I mmls to 8 mm/s) using the FE model described in section 2.2.
This range of operating parameters is often used in industry for hot upsetting as per

the ASM specifications, ASM hand book [Sem96]. The dies are kept at 350 oc. The
material parameters chosen are m = 0.02, n = 0.1, a = 0.2. The geometry is axi-
symmetric in nature so only one half of the part is simulated. The forging force
required for 50% reduction at different hid ratios, punch velocities and preform

temperatures varying from 400 0C to 1200 0C under varying friction conditions is

obtained and given in table-2.4.

The results shown in table 2.4 indicate the following salient conclusions:

(i) To study the effects of various hid ratios on forging load, we further performed a
number of finite element simulations for hot upsetting of a ck-45 steel preform

2-34
 CHAPTER - 2

at temperatures varying from 400 °C to 1200 °c under different punch velocities

(I mml s to 8 mm/s) for hid ratios varying from 1 to 2.5 using the model
described above. The dies kept at 350 °C.

(ii) There is a wide variation of forging force with respect to the process parameters.
With increase in the any of the parameters such as billet temperature, die
velocity, and hid ratio (keeping the other parameters constant) the forging force
reduces considerably. However, the decrease in the forging force is more
significant with increase in billet temperature and die velocity.

(iii) The friction coefficient does not have a considerable effect on force required for
hot upsetting.

(iv) As is to be expected the minimum forging force be obtained for the combination
having maximum temperature, velocity and hid ratio, and minimum friction
coefficient. From a practical point of view the high temperature can be obtained
by appropriate heating, the hid ratio is fixed for a given billet, maximum
velocity is a property of the forging machine, and the friction coefficient for a
billet - die material pair can be altered with suitable lubricant.

2-35
 CHAPTER - 2

Forging load (Tonnes) for following initial

Die
hid Coefficient temperatures (OC) of the ck-45 steel billet at
Velocity
Ratio of Friction 50% deformation.
(mm/sec)
400 600 800 1000 1200
2 6.67 4.47 2.99 2.01 1.35
4 2.46 1.57 1 0.64 0.41
0.2
6 1.38 0.84 0.53 0.33 0.2
8 0.92 0.57 0.35 0.22 0.13
2 6.79 4.53 3.02 2.02 1.36
4 2.52 1.6 1.02 0.65 0.42
0.4
6 l.38 0.86 0.54 0.33 0.08
8 0.95 0.58 0.36 0.22 0.14
1
2 6.83 4.57 2.91 2 1.35
4 2.54 1.62 1.03 0.66 0.39
0.6
6 1.4 0.87 0.55 0.34 0.2 1
8 0.96 0.59 0.36 0.22 0.14
2 6.7 4.45 2.99 1.95 1.34
4 2.56 1.62 1.03 0.66 0.39
0.8
6 1.41 0.88 0.55 0.34 0.2 1
8 0.97 0.59 0.37 0.23 0.14
2 6.18 4.11 2.74 1.83 1.23
4 2.29 1.45 0.93 0.59 0.3 8
0.2
6 1.36 0.85 0.53 0.33 0.2 1
8 0.86 0.53 0.33 0.2 0.13
2 6.22 4.12 2.73 1.82 1.23
4 2.32 1.47 0.94 0.6 0.38
0.4
6 1.38 0.86 0.54 0.34 0.19
8 0.87 0.54 0.33 0.2 0.13
1.5
2 6.21 4.11 2.72 1.82 1.22
4 2.33 1.48 0.94 0.6 0.38
0.6
6 1.38 0.86 0.54 0.31 0.19
8 0.88 0.54 0.33 0.21 0.13
2 6.17 4.09 2.71 1.81 1.22
4 2.34 1.48 0.94 0.6 0.3 8
0.8
6 1.39 0.87 0.54 0.31 0.19
8 0.88 0.54 0.33 0.21 0.13
2 2 5.92 3.94 2.63 1.77 1.21
4 2.19 1.39 0.89 0.57 0.36
0.2
6 1.29 0.81 0.46 0.29 0.18
8 0.82 0.5 0.31 0.19 0.12
2 5.95 3.96 2.64 1.77 1.21
4 2.21 1.4 0.9 0.57 0.37
0.4
6 1.31 0.75 0.47 0.29 0.18
8 0.83 0.51 0.32 0.12 0.12
0.6 2 5.95 3.95 2.63 1.77 1.2
4 2.22 1.41 0.89 0.57 0.3 7
6 1.22 0.76 0.47 0.29 0.18

2-36
 CH APTER - 2

8 0.84 0.51 0.32 0.2 0.12

2 5.93 3.94 2.63 1.77 1.2
4 2.23 1.41 0.89 0.57 0.37
0.8
6 1.23 0.76 0.47 0.29 0.18
8 0.83 0.51 0.31 0.19 0.12
2 5.71 3.8 2.53 1. 71 1.17
4 2.11 1.34 0.86 0.55 0.35
0.2
6 1.25 0.78 0.49 0.31 0.19
8 0.87 0.54 0.33 0.21 0.13
2 5.74 3.81 2.54 1. 71 1.17
4 2. 13 1.35 0.86 0.55 0.35
0.4
6 1.26 0.78 0.49 0.31 0.19
8 0.87 0.54 0.3 0.18 0.11
2.5
2 5.7 3.8 2.53 1.71 1.17
4 2.14 1.35 0.86 0.55 0.35
0.6
6 1.26 0.79 0.49 0.31 0.19
8 0.88 0.49 0.3 0.18 0.11
2 5.73 3.8 2.53 1.71 1.17
4 2.14 1.35 0.86 0.55 0.36
0.8
6 1.26 0.79 0.5 0.3 1 0.2
8 0.88 0.5 0.3 1 0.1 9 0.12

Table 2.4: FEM evaluation of forging load (tonnes) for hot upsetting at 50%
deformation with different input conditions.

2.8 FE Simulations of hot extrusion process with finite element model

Hot extrusion is a metal forming process of forcin g a heated billet to be reduced in its
cross section by forcing it to flow through a shaped die opening under a high pressure
[ASM88] . During extrusion metal billet is under compression stress state in all three
direction and shear forces , figure 2.19. No tensile force is produced, which makes
high deformation possible without tearing the metal. The hot extrusion is the widely
used due to its relative low deformation resistance of the metal for production of long
straight metal products of constant cross section (such as bars, so lid and holl ow
sections, tubes, wires and strips) from materi als that can not be formed by co ld
extrusion. Hot extrusion is an attractive process in industry due to its ability to
achieve energy and material savings, quality improvement and development of
homogeneous properties throughout the component. In spite of these advantages the
process is rather complicate as it requires careful control. In the industrial applicati on
of the extrusion process, die design and process control are mainly based on empirical
knowledge. This empirical knowledge is not well documented and is to a large extent
only accessible through the experience of die des igners, die correctors and press

2-37
 CHAPTER - 2

operators. Because of this the performance of the extrusion process IS mainly

determined by the subjective influences of these people.
Ram

D ie velocity
Die

Fig. 2.19: Metal flow in steady state extrusion processes

In hot extrusion process a preheated billet is loaded into the container and ram presses
this billet through a die, producing a profile with a cross-section determined by the
shape of the die orifice. Solid profiles are generally produced with a die that consists
of only one part (at die). Hollow profiles are produced with a porthole die, which
consists of two parts, a mandrel to define the inner geometry of the profile and a die
plate which defines the outer geometry. Principal parameters for the hot extrusion are:
the extrusion ratio, the working temperature, the speed of deformation, the frictional
conditions and lubrication. The extrusion ratio is the ratio of the initial cross-sectional
area (Ao) of the billet to the final cross-sectional area (AI) after extrusion, R = Ao/A I.
This is actually another name for elongation . For temperature selection, the two
factors that's need to be considered are the temperature at which hot shortness occurs,
or, for pure metals, the melting point, temperature rise due to heat generation during
metal deformation and friction. This temperature rise is affected by the extrusion
ratio, speed, etc., Heat transfer at billet-tooling interface and heat conduction within
the billet and the tooling. Increasing the ram speed produces a tendency to increase
the extrusion pressure; on another hand, however, low extrusion speeds leads to
greater cooling of the billet and thus a tendency to increase the extrusion pressure.
The higher the temperature of the billet, the greater the effect of low extrusion speed
on the cooling of the billet it has. Therefore, high extrusion speeds are required with
materials (e.g., high strength alloys) that need high extrusion temperatures.

2-38
 CHAPTER-2

FOIward extrusion

Peak load Friction load

Displacement of die (stroke)

Fig. 2.20: Punch load versus punch displacement curves in forward rod
extrusion.

In extrusion, optimization of process conditions is focused on achieving the maximum

extrusion speed, for which the quality requirements imposed on the product are still
satisfied.

The extrusion process is limited by two factors, the maximum extrusion load and the
maximum exit temperature, figures 2.20 and 2.21. The maximum extrusion load is
either imposed by the strength of the die or by the maximum capacity of the extrusion
press. The extrusion load can be lowered by increasing the initial temperature of the
billet. However, this is limited by the maximum exit temperature of the material.
When this temperature gets too high, surface defects or even melting of the material
can occur. So basically optimization of the process comes down to choosing an
optimum initial temperature. In the design of extrusion dies, a major challenge is to
obtain a uniform exit velocity over the entire cross-section of the profile. The
objective of FE simulations here is to reduce the problems encountered in extrusion
practice. In simulation of extrusion viscoplastic model is used and elasticity of the
material is neglected. The reason for this is that the elastic deformations are small
compared to the very large plastic deformations that occur during the process. A
number of extrusion defects need to be avoided. During extrusion, the center of the
billet moves faster than the periphery. After about two-thirds of the billet is extruded,

2-39
 CHAPTER - 2

the outer surface of the billet, which was formed as dead zone in the early extrusion
stage, moves towards the centre and extrudes through the die near the axis of the rod .

Optimum
process conditions

Initial Extrusion Process Temperature

~' ig. 2.21: Extrusion r)rocess is limited by two factors, the maximum extrusion
load and the maximum exit temperature

Since the surface of the billet often contains an oxidized skin, this type of flow results
in internal oxide stringers or internal pipes. With higher friction, and the higher
temperature difference between the billet and extrusion container, the tendency for
formation of the extrusion defect grows higher. The reason is that. they promote faster
metal flow in the center part of the billet than in the surface skin. Other defects
include axial hole (or called funnel), surface cracking, center burst etc. Axial hole is
caused by radial metal flow into the die when extrusion is carried to the point at which
the length of billet remaining in the container is about one-quarter its diameter.
Surface cracking can be produced by longitudinal tensile stresses generated as the
extrusion passes through the die. Center burst, or chevron cracking, can occur at low
extrusion ratios. One common problem exists in the variation in structure and
properties from front to back end of the extrusion in both the longitudinal and
transverse directions. Extrusion die geometry, frictional conditions at the die billet
interface and thermal gradients within the greatly influence metal flow in extrusion.
The only recourse for modelling this process is to consider FE simulations in
obtaining such knowledge, providing insight into the process that cannot easily be
obtained in any other way. The influence of the various process parameters at the die-

2-40
 CHAPTER-2

billet interface on the geometrical accuracy of the extruded part has been investigated
for the extrusion process using a finite element analysis. High values of frictional
coefficient produced greater from errors in the extruded component due to the greater
compressive stresses at the contact surfaces of the die. Geometrical characteristics of
the extrusion die influence both the extrusion process and mechanical property of the
extruded material.

A number of finite element simulations are performed for forward hot extrusion of a
preform for transmission shaft with various die angles (ISo, 30°, 4So, 60° and 7So) at
temperatures varying from 1000 °c to 1260 °c under different punch velocities (1 SO
mm/sec to 203 mm/sec) using the finite element model described in section 2.2. This
range of operating parameters is often used in industry for hot extrusion as per the
ASM specifications, ASM hand book [Cou91]. The dies are kept at constant
temperature (3S0 Dq.

o-
I
j

I, ,
'j •

0, '

, '

-.1 ~30
Fig. 2.22: Schematic of die shape used for simulation of extrusion process.
The geometry is axi-symmetric in nature so only one half of the part is simulated,
figure 2.22. The forging force at SO% reduction in diameter for few simulations at 203
mm/s die velocity and 1260 °c with 0.4 friction is depicted as forging load graph
w.r.t. displacement of upper die, figures 2.23 and 2.24. The figure 2.23 clearly
indicates the evolution of extrusion load with respect to punch travel and effect of die
angle on the magnitude of extrusion force.

2-41
 CHAPTER-2

600~~--~--~--~--~~--~---

~
CJ:l
g 500
t::
o
C 400
I" 140
~r 300
at its ex it end

-
CJ:l
;:: 200
x
LLl
100

OL-~--~--~------~--~--~~
196 197 198 199 200
(

-+1 j.-30 Displacement of die (mm)

600~--~----~----~--~----~
1+-60-+1
I

at its exit end

200
100

OL---~----~----~----~--~
195 196 197 198 199 200
(

j.-30 Displacement of die (mm)

600 .-----y----,--------.------,------,,------,
1+-60-+1
I Extrus ion die without curvature
at its exit end

. 140

1 200
100

OL---~--~----L---~------~
194 195 196 197 198 199 200
(

j.-30 Displacement of die (mm)

2-42
 CHAPTER-2

600.---~----~----~----~--~
1+-60--+1
I ".-,
en
C) 500
§
~ Extrusion die without curvature
400 at its exit end

200
i1
-; 300
,2
rn
g>< 200
IJ;:l
100

O~----~----~----~----L---~
190 192 194 196 198 200
(

-+1 ~30 Displacement of die (rrun)

500

~400
(!)

it
§

140 00
Extrusion die without curvature

1
§ 00
'00
at its exit end
200 2
><
IJ;:l 100

OL-__~__~____~__~____~__~
194 195 196 197 198 ]99 200
(

Displacement of die (mm)

140 Extrusion die without curvature

at its exit end

1 200
40
30
20 ~--~----~----~----~----~
190 192 194 196 198 200
(

~30 Displacement of die (mm)

Fig. 2.23: Dies with different inclination angle w.r.t. horizontal with their
corresponding extrusion load curve w.r.t. displacement of the upper die.

2-43
 CHAPTER - 2

Extrusion load (Tonnes) for following Initial

Angle Co-efficient Velocity
Temperatures (0C) of the ck-45 steel billet.
(Degrees) of Friction (mm/sec)
1000 1090 1180 1260
168 1432.34 944 .54 764 .94 643 .7 1
186 1285.03 876.07 756 .64 541.01
0.4
203 1045.60 825.30 702 .76 526.02
168 1618.09 1234.67 1162.59 1036.22
15 0.6 186 1417.00 1181.96 881.90 761.93
203 1247.15 1069.71 867 .83 718.76
168 1772.00 1405.80 1289.07 1040.64
0.8 186 1614.32 1367.19 1106.47 980 .00
203 1571.16 1355.42 1092.67 960 .25
168 1166.74 942 .68 781.50 663. 08
0.4 186 1058 .06 850 .00 681.55 576.00
203 970.65 797 .28 652.48 542 .64
168 1649.59 1211.00 1038.85 831 .59
30 0.6 186 1397.00 1073.00 919.00 747 .00
203 1277.35 848.47 806.26 668.87
168 1924.37 1597 .89 1254.63 1160.48
0.8 186 1766.00 1535 .00 1127.00 962.00
203 1511.56 1329.94 994.45 820.95
168 1228.00 922.47 760 .79 615 .23
0.4 186 1093.92 801.00 680.00 542 .00
203 947.42 750 . 13 598 .50 500 .17
168 1859.00 1651.66 1215.80 978 .00
45 0.6 186 1664.38 1487.56 939.00 756 .77
203 1477.48 1310 .01 854 .02 754 .54
168 2274 .00 1919.66 1278.00 1052.00
0.8 186 1961.32 1706.4 7 1165.00 979.00
203 1638.08 1294.35 1080.00 978.44
168 1088.53 943.13 758 .99 618 .27
0.4 186 981.00 874.00 648 .00 529 .00
203 907 .77 741.45 602 .55 516 .23
168 1679.73 1538.44 1112 .88 940 . 10
60 0.6 186 1625.00 1337.39 895.00 800 .60
203 1612.46 1116.43 842.13 744 .57
168 2049.62 1789.98 1456.60 1369.25
0.8 186 1944.00 1672.49 1368 .00 1289.00
203 1789.98 1516.00 1154 .56 966.57
168 1141.62 932.18 752.43 635.39
0.4 186 976.00 769.09 553.00 527 .00
203 950.33 553 .08 634.64 430 .39
168 1723 .34 1400.00 1168.67 1088.00
75 0.6 186 1507 .00 1224 .00 972.41 888 .00
203 1375.11 1161.54 964 .00 715 .30
168 2287 .04 1886.82 1569.55 1264.98
0.8 186 1685.00 1656.84 1400.00 1133.00
203 1537.26 1534.00 1311.64 1088.00

Table 2.5 : FEM evaluation of extrusion loads for preform of a shaft at different
input conditions.

2-44
 CHAPTER - 2

The variation of extrusion force with different die angles, punch velocities, coefficient
of friction and temperature of billet are obtained, table 2.5.

The salient conclusions observed from the table 2.5 are as given below:
(i) Extrusion force reduces as the temperature of the billet, die angle, and velocity of
the die increase.
(ii) Extrusion force increases with increase in friction coefficient.

As is to be expected the minimum extrusion force is for a combination of parameters

with maximum temperature, velocity and die angle and minimum friction coefficient.
From a practical point of view the high temperature can be obtained by appropriate
heating, the die angle can be selected as large as possible, the maximum velocity
possible is a property of the forging machine, and the friction coefficient for a billet -
die material pair can be altered with suitable lubricant to some extent. However,
improper selection of the parameters may cause a folding defect in the billet and
inferior or defective parts.

203 mm/s Ram Velocity

?{hll,
r-r-
Die Angle
- I- 30° 45° I I 75°
r'\ I-
15°
•-\ ~ •~ 60°

~ 1400 ~
~l ' "
., ,r- "
~r\. "\ ~,
~
~~
~
~~
,
"<>
-:. 1200

~ \ ....
~ ~ . ~~ ~ ~ .~ ,;: ~\.
" ""
."

j'"
• ::.
IUtll)

"<> ~ "- t- ~ ~ "- ~ ~~

11
~ ~ .~ ~~~
~"
.;; guO

~ ..... ~ ~ I\. -...; ~

"~ ~
;I.
t r'I
~ 6no

4UO
.... r"i ~ ..... ~
~
200
i
A B C D ABC D ABC D ABC D A B C D

T~llIpc r. tul'c. ·c
Co-efficient of friction
- + - 0.4 - 0.5 - " - 0.6 ........ 0.7 - 0.8 A-l000 °c , 8-1090 °c, C-1180 °C, 0-1260 °C

Fig. 2.24: Finite element simulation results depicting extrusion load required for
50'Yt, reduction with various conical dies under different process conditions.

2-45
 CHAPTER-2

2.9 Study of effect of curvature at die exit end in hot extrusion process
The conical dies of different die exit exit angle with straight edges used in section 2.8
are modified to have a curvature at the exit ends and the finite element simulations are
performed, to study the effect of die profile on extrusion force.
1+-60--+1
I

140

1 200

---.1 I+- 30
Fig. 2.25: Schematic of die shape with curvature (Radius 4) at exit end used for
simulation of extrusion process.
Using the finite element model described in section 2.2 . a number of finite element
simulations are performed for forward hot extrusion with various conical dies with
end curvature (ISo, 30°, 4So, 60° and 7So) at temperatures varying from 10000c to
I 260°C under punch velocities of 203 mm/sec and the corresponding forging force is
evaluated, table 2.6. This range of operating parameters is often used in industry for
hot extrusion as per the ASM specification [Sem86]. The effects of end curvature in
conical dies on forging forces are predicted. The dies are kept at constant temperature
(3S0°C). The geometry is axi-symmetric in nature so only one half of the part is
simulated. The effects of giving end curvature in conical dies are clearly illustrated in
figure 2.26. Evolution of forging force is graphically depicted for both the cases i.e.
dies without end curvature and dies with curvature at its exit end. Also the forging
force for these two cases are computed at SO% reduction in diameter for simulations
at 203 mm/s die velocity and 1260 °c with 0.4 friction co-efficient, table 2.6.

2-46
 CHAPTER - 2

Friction Extrusion load (Tonncs) for following Initial

Die Angle
Coemcit~nt 'remperatures (0C) of the ck-45 steel billet.
(degree)
(/-1) 1000 1090 1180 1260
0.4 190.29 147 120.42 114.22
15 0.6 238 .37 193.42 156.19 127.90
0.8 293 .22 228.20 191.44 153.76
0.4 179.54 146. 14 11 9.42 104.87
30 0.6 271.38 207.10 169.27 141.52
0.8 318 260.94 213.94 173.47
0.4 190.07 152.72 121.5 I 101.06
45 0.6 258 .87 213.51 170.69 151.50
0.8 339.56 270.86 221.32 189.59
0.4 168.64 125.08 123.15 98.26
60 0.6 290.45 216.13 179. I 5 149.29
0.8 348.84 294.53 245 200.17
0.4 183.43 152. I 8 124.01 100.23
75 0.6 382.43 315.06 261 .29 220.47
0.8 279.79 230.00 191.31 159.74
Table 2.6: Finite element simulation results for extrusion load required to
extrude a shaft by 50% reduction in its diameter with various conical dies having
curvature at its exit end under different process conditions at a ram velocity of
203 mm/sec.

2-47
 CHAPTER- 2

/-60-1
500

Extrusion die without curvatur

at its exit end

Extrusion die with curviltllre

at its exit end

oL,..------,---;--:-=::::::;:~d
196 197 198 199 200
(
--I [--30 Displacement of die (mm)
/-60-1 600~--~----~----~--~----,

ij
I - r - --x-

500

I~ 400
Extrusi on die without curvature

111 300 ilt its exit end

l,~
..::;
200
Extrusion die with curvaturc
100 ___r--____ iat its exit end

oL-,---~::::==::::::::.::::::d
195 196 197 198 199 200
(

--l [--30 --l [--30 Displacement of die (mm)

/-60-1 /-60-1 600
I -r---r- I
I ~
Extrusion die without curvature
c;(S 500
§

~r400
.9
c 300
o
0;:;;

2 200
~
LLl Extrusion die with curvature
at its exi.t end

P94195 196 197 198 199 200

[--30 30 Displacement of die (111m)

2-48
 CHAPTER-2

600 .---~----~----~--~~--~

,-.. 500
Vl
~
Extru sion die wi thout curvature

~r::
at its ex it end

] 1200
~
Uj 100
OL---~ ____ ~ ____ - L_ _ _ _~_ _~

190 192 194 196 198 200

(
1--30 1--30 Displaccment of die (mm)

500

rI 400
§ Extru sion die w hout curvature

~1300
C<l
nel

] 200
'" Extrusion die with curvature
E _ _ _ _ at its ex it end
Uj 100
OL-__ _ L_ _~~_ __ L_ _~_ _ _ _~_ _~

194 195 196 197 198 199 200

(

1--30 1--30 Displacement of die (mm)

1--60-1 1--60-1

'11
< 140
Extrus ioll die with l:ttrvature
at its exit end

40
30
20 L-----~----~----~----~----~
190 192 194 196 198 200
-I 1--30 -I 1--30 Displacement of die (mm)
Fig. 2.26: Dies with different inclination angle w.r.t. horizontal (with and without
curvature at its exit end) with their corresponding forging load curve w.r.t.
displacement of the upper die.

2-49
 CHAPTER-2

DAYALBAGH_INSTITUTE
EQUIVALENT STRAIN RATE I.ec -1)
INCA. COMPUT No : 25 h= 198.~lmm) t = 1 . 9~85E ·031') FILE : exlru30.don

363.0
311.1
212.3
221.0
181.6
136.3
EbP= 90.92
EbP= 45.51

" Moxl= 408.4

o Mlnl= .2173

100
. ....1.. .• :.. _ .••

DAYALBAGHJNSTITUTE
STRAIN RATE Isec ,1)
INCR. COMPUT No : 150 h = 194.9Imm) t = 2.5214E·02(0) FILE: exlru30.don

332.5
290.9
249.4
201 .0
166.3

124.1
EbP= 83.13
EbP= 41.56

"Moxl= 314.1
o Mlni= 1.8672E-oO

100

EQUIVALENT STRAIN RATE (.ec -1)

INCR. COMPUT No : '200 h = 193.5Imm) t = 3.2128E-02(6) FlLE: exlru30.don

P'!1
EbP= 156.6 ~.
EbP= 131.0 1 : -
I ' ; ~

EbP= 117.~
l'
EbP= 91.86
Iil.i EbP= 70:29
j.
• EbP= 58.12 !'. "i
ii~.
EbP= 39.15
,~, EbP= 19.51
I ' \' ~
"Moxi= 116.2
o Mini= 2.4360E-09 1

100 I:
L - - - - - - - - - L - - - - -- I
- - - - - . . L . - - - - - - - - F O R G E 2 V2 .1.~-

Fig. 2.27: lsocontours of equivalent strain rate at various increments of extrusion

process simulation.

2-50
 CHAPTER- 2

DAYALBAGH_INSTITUTE
EQUIVAlENT STRAIN Ino unit)
INCA. COMPlJT No : 25 h: 196.4Imm) t : 1.9465E-031_) FILE: extru30.don

Eb: .5159
Eb: .4518

.Eb: Eb : .3811
.3236
" Eb ; .2596
• Eb: .1955
Eb ; .13H

Eb : 6.1364E-02

" Maxi: .5799

o Mini: 3.2938E-03

100

FORGE2 Vl .7.4
DAYAlBAGH _INSTITUTE ~
EQUIVALENT STRAIN Ino unit)
INCA. COMPUT No : 150 h: 1 !i4.9Imm) t: 2.5214E-02(o) FILE : extru30.don

: 2. 471
: 2.163
: 1.656
: 1.548
: 1.240
: .9328
Eb: .625.1
Eb: .3115

" Maxi: 2.119

o Mini: 9.6132E-03

100

DAYAlBAGH_ INSTITUTE
EQUIVAlENT STRAIN Ino unll)
INCA. COMPlJT No : 200 h : 193.5(mm) t : 3.2128E-02(5) ALE: extfu30.don

: 2.490

: 2.180

: 1.870
: 1.561

: 1 .251
: .9409
Eb : .6311

Eb: .3213

.. Maxi: 2.800
o Mini: 1.1530E-02

100

Fig. 2.28: Isocontours of equivalent strain at various increments of extrusion

process simulation.

2-5\
 CHAPTER-2

DAYAlBAGH_ INSTITUTE
EQUIVALENT STRESS SIGMAequiv.lent I Mp~J
INCR. COMPUT No: 25 h · 198Almm) t = 7.9485E-03{s) FILE : extru30.don

141.4
126.9

r, ,'~
112.4
97;96
1;'.. -;-;,
;~
83.50
69.04

.. Maxl = 155.8
o Mi ni= 25.64

80.00
i..... .!........i. ••••• i

DAYAlB.A.GHJNSTITUTE
EQU IVAlENT (l.4p.)
INCR. COMPUT No : 50 h = 197.81mmJ t =1.1 052E--021') f iLE: ext.u30.don

163.0
146.0
1.29.1
112.2
95.21
18_27
'Seq= 61.34

.Seq= 44.40

.. M.xl= 179.9
b Mlnl= 27.46

80.00

DAYAlBAGH_INSTITUTE
E.QUIVAlENT STRESS SIG!.4Aequlv.lent IMp.)
INCR. COMPUT No: 200 h = 193.5[mm) t = 3.2128E-021') FILE: ext,u30.don

142.8

124.9

101'.1
89.22

11.38
53.53

17.84

" M.xl= 160.6

o Minl= 5.4174E-04

l UO

V2.7.
Fig. 2.29: Isocontours of equivalent stress at various increments of extrusion
process simulation.

2-52
 CHAPTER-2

The salient conclusions observed from the finite element simulations are as given
below:
(i) Forging force reduces as the temperature of the billet and velocity of the die
mcrease.
(ii) Forging force increases with increase in friction coefficient.
(iii) The forging force initially decreases with increase in die angle up to some
optimal value and later increases.
(iv) The combination of process parameters i.e. temperature, velocity and die angle
and friction coefficient for minimum forging force needs an optimization
methodology to be applied in tandem with a new process model which can give
fast estimates of forging force unlike finite element model which requires
considerable time and effort.
(v) By comparing the simulations results of table 2.5 and 2.6 it has been observed
that there is a drastic reduction in forging force required to extrude a
transmission shaft by 50% reduction in its diameter when using the conical dies
with curvature at its exit end as compared to those without curvature for the
same set of process parameters, figure 2.26.
A sample simulation illustrating isocontours of equivalent strain, equivalent stress and
. temperature during hot extrusion of transmission shaft using 65 ° die with end
curvature is shown in figures 2.27, 2.28 and 2.29.

2.10 Die shape design for hot extrusion process using curvilinear profiles.
Finite element modelling of hot extrusion using conical dies with and without
curvature is already attempted in section 2.8 and section 2.9. In the design of
extrusion dies, a major challenge is to obtain a uniform exit velocity over the entire
cross-section of the profile. At the moment, the design of extrusion dies is primarily
based on the experience of the designer. When the die is manufactured, it is first
tested in a trial pressing. Often a distorted or curved profile is produced during this
trial pressing, caused by an unbalance in the exit velocity of the profile. If this is the
case, the die is corrected in order to obtain a uniform exit velocity. After the
correction, a new trial pressing is performed until the profile satisfies the
requirements. At that point the die is used in production. During the production phase
of a die, wear and plastic defom1ation occur. This requires additional corrections to

2-53
 CHAPTER-2

the die and eventually causes the end of its lifespan. The goal of the simulation tools
developed here is to reduce the problems encountered in extrusion practice. In
simulations of extrusion it is common to use a viscoplastic model and neglect the
elasticity of the material. The reason for this is that the elastic deformations are small
compared to the very large plastic deformations that occur during the process.

In this work a novel approach is adopted for obtaining the optimal die shape that
minimizes extrusion force for extruding a transmission shaft of 30 mm diameter from
a billet of 60 mm diameter by a hot extrusion process. From the earlier study (section
2.8 and section 2.9) it has been concluded that extrusion force depends number of
process parameters viz., coefficient of friction, temperature of billet, velocity of die
and die profile. This study is now enriched by incorporating die shape as a variable .
The methodology adopted here is: initially nine different extrusion die profiles are
created. Each die profile is generated with the help of different curves, figures 2.30
and 2.31.

This die profile

region is
generated with
the help of
curve of certain
radius. Various

(
profiles are
generated by
variation of
radius.
~'ig. 2.30: Set up for design of die profile for hot extrusion process.

2-54
 LHAPT ER-~

( ( r I
I

I
I

R=70
II

I
I ~
R=-40
I
I

I
I
I
~
R=-50 R=-60 R=-70
R=18 R=40 R=50 R=60

Fig.2.31: Various extrusion dies used for analysis

2-55
 CHAPTER-2

After creating these die profiles, their finite element (FE) modelling using the finite
element model described in section 2.2 . is done with following process parameters,
i.e., co-efficient of friction is 0.4, temperature of billet is 1260 oC, velocity of die is
203 mm/sec. The results obtained are depicted in table 2.7.

Profile Radius Forging Load Max. Eq. strain rate

18 175 .14 103 .5
40 140 218.2
50 144 149.3
60 141.38 159.7
70 152.39 118.5
-40 209.61 127.1
-50 198.19 85.85
-60 204.08 124.3
-70 204 .20 125 .7
Table 2.7: Forging load obtained with various die profiles with co-efficient of
friction = 0.4, velocity of die = 203 mm/sec, and temperature of billet = 1260 °c

2-56
 -
CHAPTER-2

FORGING LOAO IN TONNES

180.

i'r
"\11

.~
147.
'\ "\' '.;i';'w\'!.t 1\
\
\
115. \

82.

50.

17.
100. 112. 12~. 136. 14B. 160.
Hpilote (mm)

Extru Profile. R ~ 18

150. rFc.:0:.:.R.:.:G"'IN.:.:G=-L::c0:.:.A=O-'IN:.:..c.TO.:.:N:.:.N.:.:E.:.:S'---_ _ _ _ _ _ _ _ _ _ _ _ _ _----,

I
123.
fl' !!1 V'I
\'.

97.

70.

44.

17. L -_ _ _- _ _ _ _ _ _ _ _ _ _- -___ - +_ _ ~_-+ _ _+-_-"

100. 112. 124. 136. l~B . 160.

Hpilote (mm)

Extru Proflle. R~ 40

150. r=-FO=-R=G=IN:.:.G~LO=AD~I=-Nc.:T=O=N=N=-ES=___ _ _ _ _ _ _ _ _ _ _ _ _ _ _,

it \
123.
T,
\
97.

70.

44.

17. L--_-~- _ _- _ - _ -_ _- ' _ _----I

100. 112. 124. 136. 148. 160.
Hpllote (mm)

Extru Profile. R~50

2-57
 CHAPTER-2

91.

66.

42.

11. L-__ ~ __ ~ __ ~ __ ~~ __ ~ __ ~ __ ~ __ ~ ____ ~~

100. 112. 124. 136. 148. 160.

Hpilol. (mmJ

Extru Profil • • R=60

160. rF~O~R~G~IN~G~LO~AD~I~N~T~O~N~N~ES~______________________________,

131.

'! ,~\ k'·

103.

14.

46.

11. L-__+_--__- -__- -________- -______- -__----__ ~

100. 112. 124. 136. 148. 160.

Hpilol. (mmJ

Extr. Prom • • R=10

21 O. rF~O~R~G~IN~G~LO~AD~I~N~T~O~N~NE~S~___,,___--------------------------_,

/i \/,i!~V~~
~
111 .

\\'. \

133.

94.

56.

11. L---+_--__- -__--~~--+_--__- -__- -__----+_--~

100. 112. 124. 136. 148. 160.
Hpllol. (mmJ

Extru Prom • • R=-40

2-58
 CHAPTER-2

FORGING LOAD IN TONNES

200.
. <t\ 1,
,\';' . \~ \:1\
~\Uj l
'j'

163. 'l\
I\ ..
.Y1v.

127.

90.

54.

17.
100. 112. 124. 136. 146. 160.
Hpllote [mm)

Extru Profile. R=-50

FORGING LOAD IN TONNES

210 . ,.:.c:.:.;.:.:.:.:.:::.=.:.=..::.:..cc::.:.::.==----------------~

171.

133.

94.

56.

17. L -_ _ _- __- __~-__- __-~--__- __-~_~

100. 112. 124. 136. 146. 160.

Hpllote [mm)

Extru Profile. R=-60

127.

90.

54.

17 . L------------------------------~
100. 112. 124. 136. 148. 160.
Hpllote [mm)

Extru Profile. R=-70

Fig.: 2.32: Evolution of extrusion force with different curvilinear dies

2-59
 CHAPTER-2

DAYALBAGH_INSTITUTE
EQUIVAlENT STRAIN RATE (sec -I)
INCR. COt.4PUT No : 20 h = 118.4(mm) t = .1558(s) FILE: expo.don

68.97
57.48
45.98
34.49
EbP= 22.99
EbP= 11.50

I:. t.4.xl= 103.5

o t.4lnl= 5.9898E-06

60.00

DAYAlBAGH_INSTITUTE
EQUIVALENT STRAIN RATE (sec -1)
INCR.COMPUTNo : 40 h= 114.7(mm) t= .1741(s) FILE : expo.don

132.8
116.2

99.56
82 .97
66.38
49.78

" M.xl= 149.3

o Mini= 4.9924E-06

60.00 o
........ t ,.,A ... '...... L .... :
L - - - - - - - - - - ' - - - - - - - - - - - - ' - - -- ------"nD'O"?V2.7
DAYALBAGH_INSTITUTE ~
EQUIVALENT STRAIN RATE (sec -1)
INCR. COMPUT No: 45 h = 113.5(mm) t = .1799(s) FILE : expo.don

EbP= 48.49

EbP= 24 .24

A M.xl= 218 .2
o Mlnl= 1.2197E-D5

!) 60.00

L - - - - - - - - - - - ' - - - - - - - - - - - - - - L - - - - - - - - - - F O R C , F ? V?7 .•

2-60
 CHAPTER- 2

DAYALBAGH_IN STITUTE
EQUIVALENT STRAIN RATE (8ec -1)
INCR. COMPUT No : 5~ h = 121.1 (mm) t = .1423(5) ALE : expo.don

.:- ......j
I, ~" 1
~
1
]
EbP= 70.98 1
~~
-EbP= 53.23 1
EbP= 35.49
1

(
EbP= 17.H
1
A 159.7
Maxi=
o Mlnl= 1.0849E-06 1

I
tiO .OO 1
.. 1. .. ~.

FOR r. F? V? 7.4
DAYALBAGH_INSTITUTE ~
EQUIVALENT STRAIN RATE (sec -1)
INCR. COMPUT No : tiD h = 122.0(mm) t = .1380(8) ALE : ex po.don

52.67
39.50
EbP= 2ti.34
EbP= 13.17

l> Maxi= 11 8.5

o Mlnl= 2.1798E-Oti

60.00

' - - - - - - - - - - - ' - -- - - - - - - - - - ' - - - - - - - - - - F O RGE2 V2.7.4

DAYAlBAGH _INSTITUTE ~
EQUIVALENT STRAIN RATE (sec -1)
INCR. COMPUT No : 20 h = 125.0(mm) t = .1233(5) FILE : expo .don

_
EbP= 38.16
. EbP= 28.ti2
EbP= 19.08
_ • • EbP= 9.539

b Maxi= 85.85
o Mini = 4.7151 E-05

tiO.OO

' - - - - - - - - - - - ' - - - - - - -- -- - -- ' - - - - - - - -- -FORr. F ? V?.7.4

2-61
 CHAPTER-2

EQUIVAlENT STRAIN RATE (sec -l)

INCR. COMPUT No : 20 h: 123.0(mm) t: .1330(s) FILE: expo.don

112.9 ..
98.82
84.70
I
70.59 l-
56.47
42.35
I
I

", Maxi: 127.1

o Mini: 1.0596E-07

60.00
i(
t
DAYAlBAGH_INSTITUTE
EQUIVAlENT STRAIN RATE (sec -1)
INCR. COMPUT No : 25 h : 123.7(mm) t: .1295(sl FILE : expo.don

110.5
96.69
82.88
69.06
I
55.25 I
41.44 I·,·
Ebp: 27.63
Ebp: 13.81
I
I
'" Maxi: 124.3
o Mini: 6.5162E-06

60.00
I
I
i
'1
0

DAYAlBAGH_INSTITUTE
EQUIVALENT STRAIN RATE (sec · 11
INCR. COMPUT No: 25 h: 122.8(mm) t: .1339(s) FILE : expo.don

111 .8
97.80
83.83
69.85
55.88
41.91
EbP= 27.94
EbP: 13.97
.J
'" Maxi: 125.7
o Mini: 6.2002E·06

60.00

V?.7

"Fig. 2.33: Isocontors of equivalent strain rate obtained during hot extrusion process
using curvilinear dies

2-62
 CHAPTER- 2

DAYALBAGHJNSTITUTE
EQUIVALENT STRESS SIGMAequlvalenl (Mpa)
INCA. COMPUT No : 20 h= 118 . ~(mm) I = .1558(s) FILE: expo.don

Seq= 125.5
Seq= 109.9

Seq= 9~ .37

78.82
63.26

47.71

l> Maxl= 141.0

o Mini= 1.042

60.00

DAYALBAGHJNSillUTE
EQUIVALENT STRESS SIGMAequlvalenl (Mpa)
INCA. COMPUT No : 45 h = 113.5(mm) 1 = .1799(s) FILE : expo.don

146.2

128.1

110.1
92.06

74.02

55.99
Seq= 37.96

Seq= 19.93

A M8Xi= 164 .2
o Mlnl= 1.896

60.00

DAYALBAGH_INSillUTE
EQUIVALENT STRESS SIGMAequivalenl(Mpa)
INCA.COMPUTNo : 40 h= 114.7(mm) 1= .1741(s) ALE : expo.don

Seq= 34.36

Seq= 17.59

"Maxl= 151.8
o Mlnl= .8148

o :(
60.00

FORGE2 V2.7. 4

2-63
 CHAPTER-2

DAYAlBAGH_INSTIT\JTE
EQUIVALENT STRESS SIGMAequlvalent (Mpa)
IN CR. COMPUT No: 55 h: 121.1 (mm) t: .1423(5) FILE: expo.don

. Seq: 138 . ~

. seq: 121.1

I s e q : 103.8
Seq: 86 . 5~
Seq: 69.26
. s e q : 51.97
Seq: 34.69
Seq: 17.41

6 .... axl::: 155.7

o Mini::: .1241

60.00
o

FORGE? V? 7 4
DAYALBAGH _INSTITUTE
EQUIVAlENT STRESS SIGMAequlvalenl (Mpa)
INCR. COMPUT No: 60 h: 122.0(mm) I : .1380(5) FILE: expo.don

131.1
114.7
98.39
82.06
Seq: 65".72
Seq: 49.38
Seq: 33.05
Seq: 16.71

" Maxi: 147.4

o Mini::: .3715

60.00

FORGE2 V2.l.1
DAYALBAGH_INSTITUTE
EQUIVAlENT STRESS SIGMAequivalenl (Mpa)
INCR. COMPUT No: 20 h : 125.0(mm) I : .1233(5) FILE : ex po.don

125.7
110.9
96.02
81.17
66.32
51.46
36.61
21.16

0. h/laxJ= 140.6
o Mini::: 6.904

60.00

L-------------------L-----------------------L------------------~ 'n c'r_~?V? 7

2-64
 CHAPTER-2

DAYALBAGH_INSTITUTE
EQUIVALENT STRESS SIGMAequivalent (t.4pa)
INCA. COMPUT No : 20 h = 123.0(mm) t = .133 0(s) FILE: expo.don

. s e q: 66.79
Seq: 50.10
Seq: 33 . ~0

Seq: 16.71

.. Maxi: 150.3
o Mini: 1.6956E-02

60.00

DAYALBAGH_INSmUTE
EQUIVALENT STRESS SIGMAequivalent (t.4p8)
IN CA. COMPUT No : 25 h: 123 .7(mm) t : .1295(s) FILE : expo.don

131.7
115.3
99.01
82.68
66.36
50.03

.. Maxi: 148.0
o Mini: 1.050

60.00
... .1 ..... ~ ... :....... 1... .... ,.. .J

V?.7
DAYALBAGH_INSTITUTE ~
EQUIVALENT STRESS SIGMAequivalent (Mpa)
INCA. COMPUT No: 25 h: 122.8(mm) t : .1339(s) FILE : expo.don

Seq: 13~ . ~

Seq: 117.8
seq: 101.1
• Seq: 8~.~1

LJ seq: 67.73
. s e q: 51.06
Seq: 34.39
Seq: 17.71

A ~axi:;: 151 .1
o Mini: 1.0~1

60.00

L..- -- - - - - - - ' - - - - - - - - - - -- - ' ' - - - - - - - - - - F ORGE2 V2.U

Fig. 2.34: Isocontors of equivalent stress obtained during hot extrusion process
using curvilinear dies

2-65
 CHAPTER- 2

2.11 Finite element modelling and analysis of hot closed die forging process for
an automotive piston

. Geometric Parameters Process Parameters Material Parameters

Selection of Optimum Process Parameters

Fig. 2.35: Modelling of closed-die forging.

Finite element modelling of automotive piston forging has been attempted. This
example is taken from industrial practice [Wer97]. The material chosen for piston
forging is aluminium alloy: AI-Cu4.5-Mg1.5-MnO.6. The realistic computational data
for piston dimensions, preform dimensions, punch velocity, friction coefficient and
process temperature have been taken into consideration while modelling the piston
forging. The geometry of piston forging is illustrated in figure 2.36. Although the
shape of piston forgings are similar to an axially symmetric body, yet their
mechanical manufacture is difficult because of the high mechanical quality required.
The thermal and mechanical properties of this alloy are collected from ASM
handbook [Sem96]

The forming of piston starts with the process of free upsetting of the forging . As the
corner gets filled, the backward extrusion starts and the material flow freely in the
piston side-wall and the forging load further increases as a function of the punch

2-66
 CHAPTER-2

stroke. Thereafter the material touches the upper die wall and the upper corner gets
filled with the material. The material further flows out to form the piston flange and
an exponential increase in the forging load takes place. A number of finite element
simulations are performed, using the finite element model described in section 2.2, for
piston forging with various punch velocities (1.2 mm/sec to 2.4 mm/sec) at
temperatures varying from 150 DC to 600 DC under different friction conditions (a
ranging from 0.1 to 0.5) using the finite element model described above The
simulations are performed in isothermal conditions i.e. the billet and dies are kept at
the same temperature. The geometry is axi-symmetric in nature so only one half ofthe
part is simulated. The variations of final forging load in the final product with
different punch velocities, friction coefficients and temperature of billet are obtained
as shown in table 2.8.

I"
Fig. 2.36: Shape geometry of the automotive piston

2-67
 CH APTER-2

a v Final forging load, O! (tonnes) and max. eg. strain rate,

O 2 (/see) for the following temperatures (DC)
150 200 300 400 500 600
1.2 FL(O!) 0.6876 0.5562 0.3602 0.2399 0.15 36 0.1048
ESR(02) 1.3480 0.8959 0.8374 0.9982 0.8810 0.8747
1.6 FL(O!) 0.7671 0.6038 0.3905 0.2577 0.1686 0.1163
ESR(02) 2.3670 1.7600 1.9710 1.6400 2.385 1.9130
0.1 0.4044
2.0 FL(O!) 0.7908 0.6213 0.2663 0.1733 0.1147
ESR(02) 2.7260 1.8970 2.6660 3.000 2.5160 1.977
2.4 FL(O!) 0.7750 0.6367 0.4295 0.2775 0.1763 0.1154
ESR(02) 3.123 3.059 3.0470 4.3510 2.3932 3.33 7
1.2 FL(OJ) 0.8675 0.6820 0.4684 0.2930 0.191 3 0.124
ESR(02) 0.8580 0.9380 1.4250 1.223 1.417 0.9100
1.6 FL(O!) 0.9205 0.7489 0.4828 0.3 128 0.2098 0.1372
ESR(02) 1.764 2.9820 1.9750 3.2400 2.700 1.474
0.2
2.0 FL(O!) 0.9576 0.7839 0.5014 0.3 197 0.2 177 0.1376
ES R(02) 3.890 3.7630 2.4990 2.4000 2.5890 2.9180
2.4 FL(O!) 0.9671 0.8045 0.5054 0.3302 0. 2 199 0.1438
ESR(02) 2.3000 5.9180 3.1630 3.8760 2.8380 2.7950
1.2 FL(O!) 1.0120 0.8170 0.5370 0.3540 0.2230 0.1520
ESR(02) 0.6991 0.7585 0.8705 1.4020 1.8570 0.8220
1.6 FL(O!) 1.1350 0.8810 0.6060 0.3885 0.2461 0.1674
ESR(02) 1.5380 1.9170 3.2550 2.114 2.1150 2.3180
OJ
2.0 FL(O!) 1.1530 0.9200 0.5861 0.3883 0.2501 0.1669
ESR(02) 3.3230 2.6020 2.2480 2J610 8.8140 4.928
2.4 FL(O!) 1.1680 0.9380 0.6300 0.3960 0.2600 0.1736
ESR(02) 2.9920 3.7410 2.8550 3.0210 2.0310 3.0210
1.2 FL(O!) 1.1560 0.9340 0.5997 0.3935 0.2509 0.1695
ESR(02) 1.0640 0.8762 1.6710 0.6988 0.9064 0.7675
1.6 FL(O!) 1.2960 1.0320 0.6539 0.4360 0.2756 0.1790
ESR(02) 1.9250 3.2740 1.7990 1.6630 2.2220 1.9320
0.4 2.0 1.2889 1.0470 0.6859 0.4369 0.2846 0.1891
FL(O!)
ESR(02) 6.2630 3.0320 3.0460 3.4450 2.3390 3.0720
2.4 FL(O!) 1.3120 1.0490 0.6854 0.4425 0.2870 0.1881
ESR(02) 4.8020 2.0410 4.2690 2.8400 2.2660 3.8490
1.2 FL(O!) 1.2670 1.0580 0.6720 0.4332 0.2592 0.1844
ESR(02) 0.6670 0.7946 3.3300 0.7857 0.7355 1.6000
1.6 FL(O!) 1.3900 1.1250 0.7537 0.4874 0.3066 0.2076
ESR(02) 6.8090 3.4190 2.2080 1.5300 4.7070 1.5760
0.5 2.0 FL(O!) 1.4650 1.1750 0.7730 0.4850 0.3126 0.2064
ESR(02) 2.1640 2.8120 3.3550 2.2250 6.8330 5.5200
2.4 FL(O!) 1.4780 1.1880 0.7548 0.4947 0.3314 0.2184
ESR(02) 2.6070 2.6190 5.6550 9.1250 2.9870 2.8270
Table 2.8: Finite Element Simulation forging load results for hot closed die
forging of an automotive piston forging. a-Coefficient offriction, v-punch veLocity
(mmlsec).

2-68
 CHAPTER - 2

0.9

0.8

0.7
V)
ill
t::
.... 0.6
...,
0
"0"
ro
..2 0.5
00
t::
'eo... 0.4
0

'"'"'
0.3

0.2

0.1

0
20 30 40 50 60 70 80

•
Displacement of die, mm
Fig. 2.37: Forging load curve w.r.t. displacement of upper die with dies with
straight edges
0.9

0.8

0.7

<n
Q) 0.6
E
2
"0"
0:: 0.5
..2
00
c
'00
..... 0.4
0
I.l..

0.3

0.2

0.1

0
20 30 40 50 60 70 80
...
Displacement of die, mm

Fig. 2.38: Forging load curve w.r.t. displacement of upper die with dies with
curved edges

2-69
 CHAPTER - 2

1.6

1.4

1.2
Ralll Velocity 2.4 nun/;;;
...
:;'" 1

...S 0.2 _ 0.32 - - 0.4 - 0.51

oe Friction coeffi cient

-
eo; 0.8
Q

bJ)
·a ,0.6
S
~
0.4

0.2

0
150 200 300 '100 500 600

Fig. 2.39: FE simulation results depicting forging load variation w.r.t. process
parameters required for hot closed die forging of an automotive piston at ram
velocity 2.4 mm/s

2-70
 CHAPTER-2

DAYALBAGH_INSTITUTE

INCR. COIo4PUT No : 25 h = 6·3.70(mm) t = 8.582(s) FILE : pl.Ion.don

INCR. COIo4PUT No : 75 h = 60.39(mm) 1= 11 .34(s) FILE : pislon .don

1.:;.dJ

FOR r. F? V?7 .4
DAYALBAGH )NSmUTE ~
INCR. COt.4PUT No : 446 h = 51 .53(mm) t = 18.72(s) FILE : plston.don

',;. i·V

r"
' - - - - - - - -- -L-- - - -- - -.L...-- -- -- - - - - FOR r.F?V? .7.4-
Fig. 2.40: Sample simulation of simulation hot closed die forging of an
automotive piston forging depicting automatic meshing and remeshing

2-71
 CHAPTER-2

DAYALBAGH_INSmUTE
EQUIVALENT STRAIN (no unit)
INCR. COIlAPUT No: 25 h : 63.10Imm) t : 8.582Is) FILE : pi ston. don

: 1.011 >1' .....

.9138

.8164
: .1189

: .6215

:: .52 41
Eb : .4267
Eb: .3292

'" lIAaxl: 1.1 09

o IIAlni: .2318

30 .00

DAYALBAGH_INSTITUTE
EQUIVALENT STRAIN Ino unil)
INCA. COMPUT No : 300 h = 53.99(mm] I : 16.67Is) FILE: pislon.don

: 2.379
: 2.118

: 1.857
: 1.597
; 1.336

Eb; .8144

Eb; .5537

'" lIAaxl: 2.639

o lIAini: .2930

V?.7

DAYALBAGH INSTITUTE
-
EQUIVALENT STRAIN Ino uni~
INCR. COIlAPUT No : 446 h = 51.53Imm) I : 18.72l s) FILE : piston.don

1 ...
.Eb:
II jEb: 5.182

• Eb:
4.661

4.140
I:.'; ,. JJ
I Eb-3.6 19 . >'

.Eb:
1 ' Eb: 3.098
2.578
Eb: 2.057
"

Eb: 1.536
Ii
~ Maxl= 5.702
o Mini= 1.015
i ~

30.0!
" 1
~.
.- ...;. ..: .- .

FORr.F? V?1 .4
.-
Fig. 2.41: Isocontours of equivalent strain at various increments of hot closed die
forging process for automotive piston

2-72
 CHAPTER-2

DAYAlBAGH_INSTITUTE
EQUIVAlENT STRAIN RATE (sec ·11
INCR. COMPUT No: 25 h = 63.10(mml 1= 8.5821sJ FILE : piston .don

!
. !IEbl': .2595
I <t
I
j
. Ebl': .2288
EbP= .1981
! EbP= .1675
.EbP= .1368
EbP= .1 062
EbP= 7.5505E·02
EbP= 4.4847E' 02
I. $'
.2901
Il. ~axi ;::
o Mlnl= 1.4190E'02

!' 30.00 I
I FORr.F? .-
V?1 .4
DAYAlBAGH_INSTITUTE ~
EQUIVALENT STRAIN RATE Isec ·lJ
INCR. COMPUT No : 300 h = 53.99(mml 1= 16.611sl FILE : pislon.don

Ebl': .3955
EbP= .1978

'" M.xI= 1.780

o Ml nl= 2.4569E-09

30.0 0

FORr.F ? V?7 .•

DAYAlBAGH_INSTITUTE
EQUIVALENT STRAIN RATE (s ec ' 11
INCR. COMPUT No : 420 h = 51 .841mml 1 = 18.47(s) FILE: pislon .don

IIEbP= 3.710 I
- Ebl': 3.246
- EbP= 2.783
IJltEbP= 2.319
o·
1~
_IEbl': 1.855
EbP= 1.391
EbP= .9275
EbP= .4638
I ~;t t
'" M~xl = 4.114
o Minl= 2.2414E-07
I 1
! j
30.00 Ir ~' ~.". ....". . ".- . . . ,...,. U
L-- - - -- - - - - ' - - - - - - - - ' - - - - - - - - - - - -- -FORGE2 V2.1.4 -

Fig. 2.42: Isocontours of equivalent strain rate at various increments of hot

closed die forging process for automotive piston

2-73
 CHAPTER- 2

OAYALBAGHJNSTITUTE
EOUIVALENT STRESS SIGMAequiv.lent IMp.)
INCR. CO MPUT No : 300 h ; 53 .99(mm) t ; 16.67(5) FILE : plston .don

4.9160E-02
4.2140E-02
3.5121E-02
2.8102E-02

· Seq ; 1.4063E-02
Seq; 7.0434E-03

A M.xi; 6.3199E-02
o Mini' 2.4025E-05

DAYALBAGH_INSTiTUTE
EQUIVALENT STRESS SiGMAequi",i,nt (Mpa)
INCR. COMPUT No : 75 h ; 60.39(m01) t ' 11.34(5) FiLE : piston. don

5.7732E-02
5.330 4E-02
4.8817E-02
4.4449E-02
4.0 022E-02
3.5594E-02

A Maxi; 6.2159E-02
o Mini' 2.2311 E·02

30.00

V?7
DAYALBAGH INSTITUTE
- ~
EOUIVALENT STRESS SIGMAequi".ient IMpa)
INCR. CO MPUT No : 446 h ' 51.53(mm) t ; 18.72(5) FILE: piston. don

I se q ; .1461 !
. se q , .1285
Seq' .11 09
I
I m ~eq, 9.3250E·02 I
Seq ' 7.5641 E-02
. s e q , 5.8032E-02
I
Seq' 4.0423E·02 I '~ ~
.!
Seq' 2.2814E-02 I 0 ,j
I' J
6.Maxi: .1637
o Mlni = 5.2056E-03
I ~

.~
,!

I ; ,j

I ..j
i" -
h 50.00
... _......
= ,=
--
FORGE2 V2.7.4
.-
Fig. 2.43:1socontours of equivalent stress at various increments of hot closed die
forging process for automotive piston

2-74
 CHAPTER-2

DAYALBAGH_INSTITUTE
TEMPERATURES (Celeius)
INCR. COMPUT No: 25 h: 63.70(mm) I : 8.582(5) FILE: pislon.don

595.5
595.1
594.6
594.1
593.7
593.2
Tern: 592.8
Tern: 592.3

"Max!: 596.0
" Mini: 591.9

50.00

DAYALBAGH_INSTITUTE
TEMPERATURES (Celeius )
INCR. COMPUT No: 75 h: 60.39(mm) I : 11 .34(5) FILE : pi s lon.don

594.5
593.9
593.3
592.7
592.1
591.5

" Maxi: 595.1

" Mini: 589.8

50.00

DAYALBAGH_ INSTITUTE
TEMPERATURES (Celclus)
INCR. COMPUT No : 420 h: 51.84(mml I : 18.47(5) ALE: pislon.don

594.6
593.2
591.7
590.3
588.8
587.4
Tern: 506.0
Tem = 584.5

" Maxi: 596 .0

" Mini: 583.1

50.00

V?7

Fig. 2.44: Isocontours of temperature distribution at various increments of hot

closed die forging process for automotive piston

2-75
 CHAPTER - 2

The salient conclusions observed from the finite element simulations are as given
below:
(i) There is a wide variation of forging force with respect to process parameters.
Forging force reduces as the temperature of the billet increases however; it
increases with increase in friction coefficient.
(ii) Though the forging force increases with an increase in ram velocity, it is not
very significant.
(iii) The change in equivalent strain rate with respect to chosen process parameters is
significant and does not follow a fixed pattern.
(iv) The combination of process parameters i.e. temperature, velocity and friction
coefficient for minimum forging force as well as minimum equivalent strain rate
needs an optimization methodology based on multi-objective approach to be
applied in tandem with a new process model which can give fast estimates of
forging force and equivalent strain rate unlike finite element model which
requires considerable time and effort.
(v) The change in equivalent strain with respect to chosen process parameters is
also observed. This change is found insignificant i.e. it varies in a very narrow
range with a mean value of about 3.000.
A sample simulation illustrating initial and final setup for simulation depicting
meshing and remeshing, evolution of isocontours of equivalent strain, equivalent
strain rate, equivalent stress and temperature during hot closed die forging of an
automotive piston is shown in figures 2.40, 2.41 , 2.42,2.43,2.44.

2-76