Journal of Materials Processing Technology 190 (2007) 375381

Die design and experiments for shaped extrusion

under cold and hot condition
S. Kumar , P. Vijay
Department of Mechanical Engineering, Institute of Technology, Banaras Hindu University Varanasi 221 005, India
Received 14 September 2005; received in revised form 9 December 2005; accepted 29 January 2007

Abstract
The paper presents designing, fabrication and experiments for shaped extrusion of aluminum alloy (Al 2024) and lead alloy (70Pb30Sn). Flat
and conical dies of H, T, L, elliptical and two-hole sections have been designed on the basis of upper bound technique by the authors for cold and
hot extrusion. Hot extrusion has been performed in the temperature range 300500 C. Experimental investigation have been conducted for average
extrusion pressure in cold case for lead alloy (70Pb30Sn) billets and in hot case for commercial grade aluminum (Al 2024) billets respectively.
Extrusion pressure in the both cases has been compared with the theoretical work by the author for cold extrusion and for hot extrusion of aluminum
alloy, result have been compared with a finite element based commercial package HyperXtrude. Theoretical results obtained by the upper bound
technique and the HyperXtrude compares well with the experiments.
2007 Elsevier B.V. All rights reserved.
Keywords: Extrusion; Upper bound technique; Finite element method; Die design

1. Introduction
An extrusion die through a proper die arrangement plays an
important role to extrude complicated shapes. Different dies
have been used for cold and hot extrusion of solid as well as
hollow sections. The aim of die design has been to withstand
a long die life with better mechanical and metallurgical properties for extruded products without defect. A good die design
involves: design of optimal die length with proper bearing length
and surface finish, selection of die profile, die arrangement, run
speed and proper heat treatments, etc.
An extensive literature exists on optimal die profile and
design based on power minimization criteria using the slip line
field technique, the upper bound technique and the finite element method. Gunasekera and Hoshino1 [1], Yang et al. [2],
Balaji et al. [3], Joun and Hwang [4]; Reddy [5] and Kumar
et al. [6] are some of them. Several attempts have been made
to analyse cold as well as hot extrusion problems. Some of the
cold cases are due to Gatto and Giarda [7], Boer and Webster
[8], Lee et al. [9], Yang et al. [1012], Gunasekera and Hoshino
[1], Shim et al. [12], Joun and Hwang [13,4], Onuh [14] and

Corresponding author. Tel.: +91 542 2368157; fax: +91 542 2368428.
E-mail address: santoshk itbhuv@yahoo.co.in (S. Kumar).

0924-0136/$ see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmatprotec.2007.01.033

most recently by Kumar et al. [15,16]. A variety of sections

both simple and complicated including many with re-entrant
sections are cold extruded economically with high-dimensional
accuracy. There has been a considerable interest in the investigation by Fragomeni [17] on the effects of die profile geometry
and extrusion parameters for structure of flow patterns, extrusion
pressure and mechanical properties of the product.
Many authors have shown that streamlined dies are superior
to straight converging dies for reducing extrusion pressure in
cold as well as hot conditions. Kumar et al. [15,16] proposed a
feature-based upper-bound analysis with strain hardening suitable for die design of non re-entry and re-entry cold extrudable
shapes. In case of hot extrusion, some of the recent attempts
are due to Altan et al. [18], Sheu and Lee [19], Hao and Li
[20], Zienkiewicz et al. [21], Smelsor [22] and Reddy et al. [23].
Recently commercial software HyperXtrude [24], DEFORM3D and FLUENT are being used for hot extrusion die design
extensively. HyperXtrude [24] is hp-adaptive finite element program to analyze fluid flow and heat transfer problems commonly
encountered in plastics and metal processing industry where
materials undergo plastic deformation and flow is described
by incompressible NavierStokes equations. HyperXtrude saves
design and analysis engineers time from performing several
mesh generation and analysis cycles that are common with conventional numerical tools.

376

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

In the present work an attempt has been made to design and

fabricate solid extrusion dies along with backers for H, L, T,
two-hole and elliptical sections and to analyze extrusion under
cold and hot conditions for the commercial grade aluminum (Al
2024) and the lead alloy (70Pb30Sn), respectively. Load versus
ram displacement obtained from the experiment has been compared to verify with the theoretical work Kumar et al. [15,16]
for cold extrusion of lead alloy. Commercially available FE software HyperXtrude [24] has been used to verify the hot extrusion
case of the aluminum alloy.
2. Design and manufacturing of extrusion die
Geometry and profile of an extrusion die constitute very
important aspect of die design and determines the extent of
redundant work done during the deformation. It cannot be totally
avoided in extrusion process rather it can be minimized by choosing a proper die profile that minimize the extrusion pressure.
Optimal die profile has been expressed as a function of several parameters as a polynomial with multiplying constants. The
present work is mainly related to design of dies for solid sections
to full-fill the following requirements as: (i) to design accurate
dimensions and product shape, so as to avoid need for any corrective work, (ii) to have the maximum possible working life,
(iii) to have the maximum length of the extruded section, (iv)
with a good-quality surface finish maintaining over many extrusions (i.e. infrequent die cleaning), (v) high extrusion speeds,
and (vi) to have a low manufacturing cost.
Following six factors have been considered in design and construction of a die design according to Laue and Stenger [25], i.e.:
(a) the flow pattern, (b) maximum specific pressure, (c) geometrical shape of the section, (d) wall thickness and tongue sizes,
(e) shape of the bearing surfaces, and (f) the tolerances of the
section. Following five common basic rules are also followed
for initial layout of a die for solid shape and to determine the
position of die aperture as: (i) the center of gravity of the section, where most material flows should be placed as close to the
edge of a die because the material tends to flow faster in the
center than at the edge, (ii) large cross sections run relatively
more quickly than smaller and thinner sections and therefore,
allows for balancing to be achieved without different bearing
lengths, (iii) symmetrical section is always placed with the axis
of symmetry considering with centre of a die, (iv) in case of
multi-hole die, the aperture should be arranged as symmetrically as possible, (v) layout of a multi-hole die is arranged to
prevent extrusions from rubbing together or running on top of
each other as they leave the press, and (vi) length of the die lands
should be at least twice the wall thickness for aluminum section
with minimum value of 2.53 mm.
The dimensioning of dies has been made according to stress
criteria proposed by Laue and Stenger [25] and proper tongue
length and die deflection rules have been considered in the die
design. Pressure on the die, bending of the die, shearing stresses
and equivalent stresses due to bending and shearing have been
calculated according to Schmoeckel et al. [26] and Vijay [27].
For conical and flat dies, rules proposed by Laue and Stenger [25]
are followed for the proper die bearing and corner radius design.

Fig. 1. Designed dies the die backers.

Thickness of the die depends on extrusion load and therefore a

proper die backer is required. Following thicknesses 13, 15, 20,
25, 30, 40, 50, 70 and 100 mm are commonly used for backer
design as standard practice.
Three factors have been considered in selecting a proper
material for a die and the die backer and the pressure pad. These
include (a) press capacity, (b) pressure required for extrusion,
and (c) maximum temperature developed. The extrusion set-up
toolings as fabricated by Bishwajit [28] has been used to conduct all the experiments mentioned. The set-up includes die and
die holder, ram, dummy block, die backer, pressure pad, bolster,
and horseshoe clamp and die holder carrier, etc.
After careful design of an extrusion die and the backer, it is
manufactured by machining. Dies and pressure pad are hardened
by heating up to 800 C at the rate of 400 C per hour and holding
at 800 C for 2 h inside a heating chamber and then quench to
servo-cutting oil. The fabricated dies along with die backers have
been shown in Fig. 1.
A small billet heating arrangement (induction type) has also
been fabricated. The temperature of each billet is measured and
controlled in the said induction furnace using thermocouples to
control the temperature ranging from 25 to 460 C. After heating
to a desired temperature, the billet is removed form the induction
coil and fed to the container for extrusion.
3. Experiments and validation
Experiments on the fabricated set-up by Bishwajit [28]
have been performed using H, L, T, two-hole and elliptical
dies manufactured in the Section 2. Average extrusion pressures for all the section is determined experimentally and
load versus ram displacement curve have been drawn. A comparison of the extrusion pressure for cold extrusion of lead
alloy is made with theoretical work obtained by the upper
bound analysis Kumar et al. [15,16]. HyperXtrude [24] has
been used to analyze the hot extrusion of aluminum alloy (Al
2024) using flat and conical dies. Following are the process

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

377

Fig. 2. Extruded lead alloy products.

conditions and parameters for all the cases (cold/hot):Billet

dimension: length = 45 mm, diameter = 25.09 mm.Billet materials properties: for hot extrusion: commercial grade aluminum
(Al 2024), specific heat (j/kg/K) = 880, density (kg/m3 ) = 2830,
conductivity (W/m/K) = 152, coefficient of thermal expansion (1/K) = 1 e005 , stress exponent = 2, activation energy
(j/mol) = 1.22 e5 , strain rate offset (1/s) = 1.0 e4 , liquidus temperature (K) = 916, solidus temperature (K) = 814 and reciprocal
of strain factor = 1.18 e8 . For cold extrusion: lead (70Pb30Sn)
at room temperature with flow stress without strain hardening = 12 MPa.Lubricant: for aluminum and lead both: mobil oil
mixed with graphite powder (20%) is used.Friction condition:
the friction factor (obtained from Ring test) in container and die
is based on billet condition are 0.078 for lead alloy (70Pb30Sn)
and 0.15 for Al 2024, respectively.
From the experimental results, the loads versus ram displacement curve have been plotted for all the cases. Pressure
versus ram displacement curves is drawn based on the load
displacement curves.
Figs. 2 and 3 show the defect free extruded products for cold
and hot condition. Fig. 4 shows the variation of load and pressure with ram displacement for lead alloy (70Pb30Sn) using flat
die of L shape. It is observed that initially, a rapid rise in the
extrusion load from O to A is due to initial compression of the
billet to fill the container and some extrusion of relatively undeformed material also takes place. After point A, load increases
due to shearing of the material at the die entry and then the load
decreases due to filling of material at sharp corner of the die and
subsequently further flow of material through die bearing takes
place; the load starts increasing from B to C due to friction at the
die bearing. At point C, material begins to exit from the die under

Fig. 3. Extruded aluminum alloy products.

Fig. 4. Load and pressure vs. ram displacement curve (lead alloy, flat dieL
shape, %r = 54.88, V0 = 0.120 mm/s).

steady state condition and therefore, load remains constant. The

average velocity of the ram is set as V0 = 0.120 mm/s. Nature of
the pressure curve is similar because it has been obtained from
the same loaddisplacement curve.
Fig. 5 shows the variation of load and pressure with ram
displacement for the extrusion of lead alloy (70Pb30Sn) using
a flat die of T shape. Nature of the curve obtained is similar to
Fig. 2 and 3. The velocity of the ram is set as V0 = 0.128 mm/s.
Figs. 6 and 7 show the variation of load and pressure with ram
displacement for the lead alloy (70Pb30Sn) using flat dies for H
and two-hole shape, respectively. Nature of the curve obtained
is similar to Figs. 2 and 3. The velocity of the ram is set as
V0 = 0.129 mm/s for H shape and V0 = 0.116 mm/s for two-hole
shape, respectively.
Fig. 8 shows the variation of load and pressure with ram
displacement for the lead alloy (70Pb30Sn) using conical die of
elliptical section with reduction as 40.97%. It is observed that
initially load increases from point O to A due to upsetting of billet

Fig. 5. Load and pressure vs. ram displacement curve (lead alloy, flat dieT
shape, %r = 51, V0 = 0.128 mm/s).

378

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

Fig. 6. Load and pressure vs. ram dispacement curve (lead alloy, flat dieH
shape, %r = 56.31, V0 = 0.129 mm/s).

Fig. 7. Load and pressure vs. ram displacement curve (lead alloy, flat dietwohole, %r = 57.93, V0 = 0.116 mm/s).

Fig. 9. Load and pressure vs. ram displacement curve (Al, flat dieL shape,
%r = 54.88, V0 = 0.114 mm/s).

Fig. 10. Load and pressure vs. ram displacement curve (Al, flat dieT shape,
%r = 51, V0 = 0.118 mm/s).

in the container. For point A to B, the rate of increase of load is

less due to shearing of material at die section. From point B to C
load remains constant to fill the edges of the ellipse. From point
C to D load increases due to flow of material through die land.
At point D, material begins to exit from die under steady state
condition and therefore, load remains constant. The velocity of
the ram is set as V0 = 0.069 mm/s.
Figs. 911 show variation of load and pressure with ram
displacement for extrusion of aluminium (Al 2024) under hot
working condition through flat and conical die of H, L, T, twohole and elliptical section. Figs. 9 and 10 show the variation of
load and pressure with ram displacement using flat die of L and

T shape, respectively. The initial temperature of the billet for L

shape is kept as 410 C and that of the container as 400 C. The
initial temperature of billet for the T shape is kept as 400 C and
that of the container as 390 C. It is observed that in both the
cases, initially loads increases gradually, remains constant for
a certain value of ram displacement, further increases and then
remains constant. From point 0 to A upsetting of billet takes
place in the container, from point A to C shearing at die section and filling of sharp corner of die takes place and flow of
the material through the die bearing takes place and therefore,
load starts increasing from C to D due to friction. The average velocity of the ram is set as V0 = 0.114 mm/s for L shape
and V0 = 0.118 mm/s for T shape, respectively. Nature of the
pressure curve is similar to load curve.
Figs. 11 and 12 show the variation of load and pressure with
ram displacement for the aluminium (Al 2024) billet using flat

Fig. 8. Load and pressure vs. ram displacement curve (lead alloy, conical
dieelliptical shape, %r = 40.97, V0 = 0.069 mm/s).

Fig. 11. Load and pressure vs. ram displacement curve (Al, flat dieH shape,
%r = 56.31, V0 = 0.129 mm/s).

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

379

Fig. 13. Load and pressure vs. ram displacement curve (Al, conical die -elliptical
shape, %r = 44.08, V0 = 0.116 mm/s).
Fig. 12. Load and pressure vs. ram displacement curve (Al, flat dietwo-hole,
%r = 57.93, V0 = 0.144 mm/s).

die for H shape and flat die for two-hole shape, respectively. The
initial temperature of the billet for H shape is kept as 410 C and
that of container as 400 C and initial temperature of the billet
for Two-hole shape is kept as 320 C and that of container as
310 C. It is observed that for H-shape initially the extrusion
load increases slowly and then increases rapidly and remains
constant for a certain value of ram displacement and further
increases and then remains constant The velocity of the ram is
set as V0 = 0.129 mm/s. For two-hole shape nature of curve is

similar to H shape. The velocity of the ram is however set as

V0 = 0.144 mm/s.
Fig. 13 shows the variation of load and pressure with ram
displacement for extrusion of aluminum (Al 2024) billet using
conical die of elliptical shape. The initial temperature of billet for elliptical shape is kept as 440 C and that of container
as 430 C. It is observed that, initially extrusion load increases
rapidly, remains constant for a certain value of ram displacement, further increases and then remains constant. The velocity
of the ram is set as V0 = 0.116 mm/s. The nature of pressure
curve is similar to the load curve.

Fig. 14. Variation of temperature as obtained from the HyperXtrude [24].

380

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

Table 1
Comparisons of power obtained from the experiment and theoretical work
(Kumar. et. al. [15,16]) using lead alloy

sion dies and processes in CAE environment to improve upon

the product quality and productivity by avoiding trail runs.

Die

Experimental
power (W)

Theoretical
power (W)

Acknowledgements

Flat dieL shape

Flat dieT shape
Flat dieH shape
Flat dietwo-hole
Conical dieelliptical shape
(% reduction = 40.97)

0.972
1.049
2.117
1.603
0.704

1.049
1.183
2.188
1.775
0.740

The financial support from Department of Science and

Technology (INDIA) [Grant No. III.5(109}/2001-SERC(Engg)
Dated 19.02.2002] is gratefully acknowledged. Authors wish to
express there thanks to Dr. A.K. Jha of Mechanical Engineering Department I.T.- B.H.U. Varanasi (INDIA) for his help and
support.

Table 2
Comparisons of power obtained from the experiment and theoretical work
(HyperXtrude [24]) using aluminum alloy
Die

Experimental
power (W)

Theoretical
power (W)

Flat dieL shape

Flat dieT shape
Flat dieH shape
Flat dietwo-hole
Conical dieelliptical shape
(%reduction = 44.08)

1.37
2.01
2.95
8.25
0.70

1.50
2.16
3.13
8.43
0.88

In all the above three cases (Figs. 1113) it is observed that,

initially the load increases rapidly because of the clearance
between billet-container interface being first filled during the
deformation. Also initially the contact area of friction between
billet and container wall is more therefore, the frictional force
requirement is proportionally followed. After that, load becomes
constant because of the frictional force decreases due to decrease
in contact area and the deformation load increases due to strain
hardening while passing through the die bearing. Once the material is on the exit side of die, load becomes constant. The variation
of temperature obtained from HyperXtrude [24] is shown in
Fig. 14.
Table 1 shows comparison of the power obtained using the
experimental and theoretical work ([15,16]) for extrusion of different sections of lead alloy. In Table 2, comparison of the power
obtained by experimental and theoretical work by HyperXtrude
[24] has been made for extrusion of different sections of aluminum alloy. It is observed that theoretical power obtained by
the upper bound analysis and FEM is satisfactory but it is slightly
more than the experimental power and therefore compares well
with the experiments.
4. Conclusion
The experimental result for complicated shape extrusion confirms the proposed validity of the die design to be safe and
satisfactory. Products have been found with better surface finish without any defect such as surface crack, central burst etc.
It is also observed that as the complexity of section increases,
extrusion power also increases and it compares well with the theoretical work for cold extrusion of lead alloy (70Pb30Sn) and
with aluminum (Al2024). The study indicates that upper bound
technique and FEM can be used confidently for designing extru-

References
[1] J.S. Gunasekera, S. Hoshino, Analysis of extrusion of polygonal sections
through streamlined dies, Trans. ASME J. Eng. Ind. 107 (1985) 229232.
[2] D.Y. Yang, H.S. Kim, C.M. Lee, C.H. Han, Analysis of three dimensional
extrusion of arbitrary shaped tubes, Int. J. Mech. Sci. 32 (1990) 101114.
[3] P.A. Balaji, T. Sundararajan, G.K. Lal, Viscoplastic deformation analysis
and extrusion die design by FEM, Trans. ASME J. Appl. Mech. 58 (1991)
644652.
[4] M.S. Joun, S.M. Hwang, Optimal process design in steady state metal
forming by finite element method. II. Application to die profile design in
extrusion, Int. J. Mach. Tools Manuf. 33 (1993) 6370.
[5] N.V. Reddy, P.M. Dixit, G.K. Lal, Die design for axi-symmetric extrusion,
J. Mat. Proc. Tech. 55 (1995) 331339.
[6] S. Kumar, K. Shanker, G.K. Lal, A feature recognition methodology for
extrudable product shapes, Int. J. Prod. Res. 37 (1999) 25192544.
[7] F. Gatto, A. Giarda, The characteristics of the three dimensional analysis
of plastic deformation according to the SEER method for converging dies,
Int. J. Mech. Sci. 23 (1981) 129148.
[8] C.R. Boer, W.D. Webster Jr., Direct upper bound solution and finite element
approach to round to square drawing, Tran. ASME. J. Eng. Ind. 92 (1985)
158168.
[9] C.M. Lee, D.Y. Yang, M.U. Kim, Numerical analysis of three dimensional
extrusion of arbitrarily shaped sections by the method of weighted residuals,
Int. J. Mech. Sci. 32 (1990) 6582.
[10] D.Y. Yang, C.H. Han, M. Kiuchi, A new formulation for three dimensional
extrusions and its application to extrusion of clover sections, Int. J. Mech.
Sci. 28 (1986) 201218.
[11] D.Y. Yang, C.H. Han, B.C. Lee, The use of generalized extrusion boundaries for the analysis of axi-symmetric extrusion through curved dies, Int.
J. Mech. Sci. 27 (1985) 653663.
[12] H. Shim, D. Kim, N. Kim, A simplified 3D finite element analysis of the
non axi-symmetric extrusion process, J. Mat. Tech. 37 (1993) 567576.
[13] M.S. Joun, S.M. Hwang, Optimal process design in steady state metal
forming by finite element method. I. Theoretical considerations, Int. J.
Mach. Tools Manuf. 33 (1993) 5161.
[14] S.O. Onuh, M. Ekoja, M.B. Adeyemi, Effects of die geometry and extrusion
speed on the cold extrusion of aluminum and lead alloys, J. Mat. Proc. Tech.
132 (2003) 274285.
[15] S. Kumar, K. Shanker, G.K. Lal, Feature based extrusion die design for
non re-entry cold extrudable shapes, Trans. ASME J. Manuf. Sci. Eng. 124
(2002) 7179.
[16] S. Kumar, S.K. Prasad, Feature based design of extrusion process using
upper-bound and finite element techniques for extrudable shapes, J. Mat.
Proc. Tech. 155156 (2004) 13651372.
[17] J.M. Fragomeni, An organizational design of experiments statistical
approach to investigate the effect of extrusion process manufacturing on the
mechanical behavior of an aluminium alloy, in: Proceedings of the Conference on Pioneering New Technologies: Management Issues and Challenges
in the Third Millennium, 1998, pp. 1114.
[18] B. Altan, M. Gevrek, S. Onurlu, A numerical method for predicting the temperature distribution in axisymmetric extrusion through flat dies, J. Mech.
Working Tech. 13 (1986) 151159.

S. Kumar, P. Vijay / Journal of Materials Processing Technology 190 (2007) 375381

[19] J.J. Sheu, R.S. Lee, A numerical model for simulating temperature and
speed effects in hot extrusion of rod, Int. J. Mech. Sci. 33 (1991)
985998.
[20] N. Hao, Li Kezhi, Numerical design of the die land for shape extrusion, J.
Mat. Proc. Tech. 101 (2000) 8184.
[21] C. Zienkiewicz, E. Onate, J.C. Heinrich, A general formulation for coupled
thermal flow of metals using finite elements, Int. J. Num. Meth. Eng. 17
(1981) 14971507.
[22] R.E. Smelsor, A thermo mechanically coupled analysis of streamlined
die extrusion including hardness predictions, in: Proceedings of the
Conference on Aluminum Technology, Institute of Metals, London,
1986.
[23] N.V. Reddy, P.M. Dixit, G.K. Lal, Die design for axi-symmetric hot extrusion, Int. J. Mach. Tools Manuf. 37 (1996) 16351650.

381

[24] HyperXtrude (Version 3), Simulation Software for Extrusion Process

Altair Engineering Inc. USA (2003).
[25] K. Laue, H. Stenger, Extrusion Processes, Machinery, Tooling, American
Society for Metals Pub, 1976, 20345 and 420550.
[26] D. Schmoeckel, G.F. Vonschlotheim, A. Wansel, Evaluation of the cold
extrusion production process, in: Proceedings of First International Symposium on Environmentally Conscious Design and Inverse Manufacturing,
1999, pp. 414419.
[27] Vijay, P., Design, fabrication, Instrumentation and Experiments for Shaped
Extrusion under Cold/ Hot Working Conditions, Unpublished M. Tech
Thesis IT BHU Varanasi (INDIA) (2003).
[28] Biswajit N., Design and Fabrication for a Shaped Extrusion Tooling
and Experimental Study, Unpublished M. Tech Thesis IT BHU Varanasi
(INDIA) (2003).

ID
794070

Title
Die design and experiments for shaped extrusion under cold and hot condition

http://fulltext.study/journal/758

http://FullText.Study

Pages
7