A study of plastic strain and plastic strain rate in machining

of steel AISI 1045 using FEM analysis

J.P. Davim
*
, C. Maranho
Department of Mechanical Engineering, University of Aveiro, Campus Santiago, 3810-193 Aveiro, Portugal
a r t i c l e i n f o
Article history:
Received 18 January 2008
Accepted 15 April 2008
Available online 22 April 2008
Keywords:
FEM analysis machining
Plastic strain
Plastic strain rate
AISI 1045
High speed machining (HSM)
a b s t r a c t
The present paper employs the nite element method to determine the plastic strain and plastic strain
rate when machining an AISI 1045 steel. A validation of the process using literature experimental values
was also conducted in order to verify if the obtained results with the commercial nite element software
were close to those found within the literary research. The comparison shows that nite element mod-
ulation can be used to determine either plastic strain or plastic strain rate if special attention is taken into
consideration when using the analytical models. The effect of high speed machining (HSM) in the plastic
strain and plastic strain rate when cutting steel AISI 1045 was observed.
From the simulations, it can be concluded that both the plastic strain and plastic strain rate can be pre-
dicted with good precision when machining with the FEM model.
2008 Elsevier Ltd. All rights reserved.
1. Introduction
AISI 1045 steel is widely applied nowadays because of its
mechanical properties and is commonly used in mechanical con-
structions and also as part design. Its a material with good
machinability and has proven to be capable of providing engineers
a good and reliable solution when submitted to effort, corrosion,
etc. Due to its characteristics, its the steel with more applicability
among them all.
High speed machining (HSM) process is being used for manu-
facturing complex machine parts, mould and die, aerospace and
automotive components where maintaining structural integrity
of materials (while removing high volumes of material) is of ex-
treme importance [13]. It presents several advantages besides
high removal rates such as reduction in dead times and lower cut-
ting forces, leading to excellent dimensional accuracy and surface
quality as well [4]. Having this said, HSM processes are growing
industrial interest not only because they allow larger material re-
moval rates but also because of the positive inuence on the n-
ished workpiece properties. An interesting feature of HSM
processes is that specic cutting force for a great amount of mate-
rials decreases with the increase in cutting speeds and then reach-
ing a plateau. However, the reason for the cutting forces reduction
is still somewhat unclear. Thermal softening, a decrease in friction
or segmental chip formation (assuming that heat evacuation
through the chip is energetically favourable) at high cutting speeds
are possible causes [5]. Unlike conventional machining, where chip
formation is followed by plastic deformation, at HSM serrated chip
or segmentation processes occur [6].
2. Plastic strain and plastic strain rate effects during HSM
Machining of metals is a non-linear process and the complex
coupling between deformation and temperature is not completely
understood nowadays. The deformation process is concentrated
in a small zone and the temperatures that are generated in this zone
greatly affect both the tool and the workpiece. High cutting temper-
atures strongly inuence tool wear, tool life, workpiece surface
integrity, chip formation mechanismand also contribute to thermal
deformation of the cutting tool. An increase in temperature in the
primary deformation zone softens the material, allowing for lower
cutting forces and less energy in the shear process [2]. It is com-
monly known that during metal cutting, a great amount of defor-
mation energy is transformed into heat near the tool cutting edge.
A small percentage of this heat is transferred through conduction
to the uncut material ahead of the cutting tool, having an effect in
the integrity of the machined workpiece. The bigger percentage of
heat remains in the removed chip and this produces a local increase
of temperature in a narrow zone where high strain occurs and adi-
abatic shear bands are formed [4] (shear bands can be observed in a
metal if the plastic strain rate is higher than 10
2
s
1
).
The nature of the plastic strain and related phenomenon of
materials still remain considerably unclear nowadays. It has been
realized that strain rate and temperature affect material properties,
although strain rate has a greater effect on ow stress in hot-work-
0261-3069/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2008.04.029
* Corresponding author. Tel.: +351 234 370830; fax: +351 234 370953.
E-mail address: [email protected] (J.P. Davim).
Materials and Design 30 (2009) 160165
Contents lists available at ScienceDirect
Materials and Design
j our nal homepage: www. el sevi er . com/ l ocat e/ mat des
ing range and a smaller effect in the cold-working range, especially
when large strains are present [7,8]. Plastic deformation of steel is
temperature sensitive, the rate at which the deformation takes
place also has inuence on the plastic strain. This sensitivity is di-
rectly related with time and temperature dependency of the mech-
anisms that govern the deformation and the evolution of the
deformation in the material. The main mechanism by which plastic
strain takes place is thermally activated motion of dislocations past
obstacles that exist within the lattice over a wide range of strain
rates and cutting temperatures. The material response is signi-
cantly affected by the nature and density of the obstacles (which
may change as the deformation takes place). When dealing with
metals, experimental results show that the stress required for plas-
tic strain often reduces with the increase of temperature and with
the decrease of plastic strain rate. It can then be said that temper-
ature and plastic strain rate greatly inuence the material re-
sponse. In general, the stress decreases with the increasing of
temperature and decreasing the plastic strain rate [9]. Actually,
temperature and strain rate effects are coupled, since one inu-
ences the other. Temperature affects the rate of deformation,
which is controlled mainly by a thermally activated mechanism.
On the other hand, plastic strain at high rate generates signicant
heating and cause an increase in temperature which leads to
mechanical instability and the localization of deformation into nar-
row sheets of material (the adiabatic shear bands), which act as
precursor for eventual material failure [7].
Deformation at low strain rates or under quasi-static loading, is
relatively homogeneous because is governed by slip and twinning
mechanisms. On the contrary, deformation at high strain rates is a
much complex phenomenon that is characterized by extreme
strain localization along the adiabatic shear bands. Each material
has a different susceptibility to adiabatic shear because it depends
on properties like heat capacity, heat conductivity, strength level,
microstructure, geometry, defects and strain rates. It is also known
that adiabatic shear banding precedes material failures at high
strain rates. Adiabatic shear banding is usually accompanied by a
loss in stress capacity owing to intense thermal softening in the
shear bands and, in many cases, shear bands serve as sites for crack
initiation and growth during subsequent dynamic fracture [10].
Localized adiabatic shearing can be considered a unique conse-
quence of severe plastic deformation at high strain rates. As both
thermal and strain softening lead to rapid deformation localization,
a shear band forms via a nearly adiabatic process. Also of note is
that grain renement can occur within shear bands and severe
plastic strain (which can reach 520) can also appear within these
shear bands [11].
Plastic strain rate can be divided in three zones: the low strain
rate region (<1 s
1
), the medium rate region (comprehended be-
tween the low and high strain rate region values) and the high
strain rate region (above 10
3
or 10
4
s
1
). The inuence of these
zones on the ow stress is, respectively, weak, sensitive and great
[7].
Duan et al. [12] showed that the plastic strain rate increases
with the increase of the cutting speed. These authors also con-
cluded that the hardness of the workpiece is able to inuence
the chip formation and the deformation mechanism.
The objective was to study the behavior of AISI 1045 steel when
machined with HSM (in this case, 3000 m/min). Also of note is that
a conventional machining (of 300 m/min) was also applied to the
workpiece keeping the remaining machining parameters the same.
This allowed for a better comparison between both conventional
and high speed cutting regimes. The experimental values worth
noting for the calculation of the plastic strain and plastic strain rate
are presented in Table 1 (taken from Denkena et al. [13]).
From the developed work within this paper, it was possible to
predict with good precision both plastic strain and plastic strain
rate with the nite element analysis when compared to the litera-
ture data supplied by Denkena et al. [13].
3. Analytic model
Merchant theory [14] was followed to make all the calculations
(all the presented results in the analytic model are based on liter-
ature values [13]).
The chip compression ratio is obtained with the following
formula:
k
t
0
t
1
where t
0
is the measured chip thickness within the experimental
process and t is the theoretical chip thickness that can be obtained
by
t f sinv 2
where f is the feed rate and v is the tool position angle.
The shear angle can be obtained by
/ arctg
cos c
k sinc

3
where c is the tool rake angle. The plastic strain was calculated
using the following method:
e
1 k
2
2k sinc
k cos c
4
Finally, the plastic strain rate can be found with the following
formula:
_ e
V
c
cos c
cos/ c

1
Dx
5
where k is the chip compression ratio, V
c
is the cutting speed, / is
the cutting shear angle and Dx is the elemental chip thickness.
For a cutting speed of 300 m/min:
/ arctg
cos6
2:1 sin6

33

e
1 2:1
2
2 2:1 sin6
2:1 cos6
2:8
_ e
300
60
cos6
cos33 6

1
21 10
6
30:1 10
4
s
1
For a cutting speed of 3000 m/min:
/ arctg
cos6
1:5 sin6

43

e
1 1:5
2
2 1:5 sin6
1:5 cos6
2:39
_ e
3000
60
cos6
cos43 6

1
10 10
6
67 10
5
s
1
Table 1
Literature values (presented in Denkena et al. [13])
Experimental values
Cutting speed (V
c
) (m/min) 300 (conventional machining) 3000 (HSM)
Elemental chip thickness (Dx) (lm) 21 10
Chip compression ratio (k) 2.1 1.5
Feed (f) (mm/rev) 0.3
Depth of cut (ap) (mm) 3
Rake angle (c) () 6
Relief angle (a) () 6
Workpiece material AISI 1045
Tool material HC P30-P40
Tool coating Ti(CN)/Al
2
O
3
J.P. Davim, C. Maranho / Materials and Design 30 (2009) 160165 161
4. Finite element analysis
Predicted results may vary with software and with the input
data so the choice of the software is of extreme importance. A nite
element software, specic for machining operations, was chosen to
simulate the metal cutting process (in this case, a turning opera-
tion). Therefore, Advantedge
TM
, supplied by Third Wave Systems,
was used in this study. This commercial software package was
built from the start with metal cutting operations in mind, allow-
ing to simulate turning, drilling, milling, micro machining, etc in
both two or three dimensions. It uses adaptive meshing to help
improving the quality and the accuracy of the predicted results
and it also supports several workpiece material library. Unfortu-
nately, the solver cannot be controlled by the user (is controlled
by the software itself which sometimes can be problematic) but
fast setups for several simulations can be done easily because of
the easy to use software interface.
The input parameters (for the simulation) of the machining
operation can be found in Table 2.
Third Wave Advantedge
TM
allows improvement and optimiza-
tion of machining processes and with it, is possible to determine
optimum machining parameters and tool congurations allowing
lower cutting and feed forces and temperature, without spending
time and money with experimental processes.
To model the thermalvisco plastic behavior of the workpiece
materials, the software uses a constitutive equation, the
JohnsonCook law, which can be represented by the following
formula:
r
eq
A Be
n
1 C ln
_ e
_ e
0

1
T T
room
T
m
T
room

m

6
where e is the plastic strain, _ e is the plastic strain rate (s
1
), _ e
0
is the
reference plastic strain rate (s
1
), T is the temperature of the work-
piece material (C), T
m
is the melting temperature of the workpiece
material (C), and T
room
is the room temperature (C). Coefcient A
is the yield strength (MPa), B is the hardening modulus (MPa) and C
is the strain rate sensitivity coefcient, n is the hardening coef-
cient and m the thermal softening coefcient [15].
The friction coefcient was obtained using the Coulomb model
and was calculated through the following reasoning:
U
F
f
F
c
tany
F
c
F
f
tany
7
where F
f
represents the feed force, F
c
the cutting force and c is the
tool rake angle.
Table 2
Software input parameters
Workpiece
Workpiece length 4.0 mm
Workpiece height 3.0 mm
Workpiece material AISI 1045
Tool
Rake angle 6
Rake face length 2.0 mm
Relief angle 6
Relief face length 2.0 mm
Cutting edge radius 0.02 mm
Material Cemented carbide (P)
Coating (3 layers) Al
2
O
3
:0.03 mm
TiC:0.015 mm
TiN:0.015 mm
Process
Depth of cut 3 mm
Length of cut 4.0 mm
Feed 0.3 mm/rev
Cutting speed 300 and 3000 m/min
Friction coefcient
*
0.69 for 300 m/min and 0.31 for 3000 m/min
Coolant Not used
Simulation
Maximum number of nodes 12,000
Maximum element size 0.1 mm
Minimum element size 0.02 mm
*
Values obtained using the experimental cutting and feed forces [13].
Table 3
Errors between analytical and simulation values for both conventional machining and
HSM
Plastic strain (%) Plastic strain rate (%)
300 m/min 2.5 1.4
3000 m/min 1.6 6.5
Table 4
Comparison between analytical and simulation values for both plastic strain and
plastic strain rate at conventional machining and HSM (the presented values should
be considered an average)
Plastic strain
(analytical)
Plastic strain
(simulated)
Plastic strain
rate (analytical)
(s
1
)
Plastic strain
rate (simulated)
(s
1
)
300 m/min 2.77 2.84 2:74 10
5
2:78 10
5
3000 m/min 2.36 2.40 6:3 10
6
6:7 10
6
Fig. 1. Plastic strain distribution in the workpiece, chip and burr. (a) 300 m/min, and (b) 3000 m/min.
162 J.P. Davim, C. Maranho / Materials and Design 30 (2009) 160165
It is also important to refer that the workpiece height is more
than 5 times bigger than the feed. This ratio is of extreme impor-
tance to obtain simulations with good precision and the RTS (rela-
tive tool sharpness) which is, by denition
Fig. 2. Plastic strain rate distribution in the workpiece, chip and burr. (a) 300 m/min, and (b) 3000 m/min.
Fig. 3. Comparison between experimentation and simulation for the plastic strain
(a) and plastic strain rate (b) for the machining of AISI 1045. The cutting speed was
300 m/min with a feed of 0.3 mm/rev and a depth of cut of 3 mm.
Fig. 4. Comparison between experimentation and simulation for the plastic strain
(a) and plastic strain rate (b) for the machining of AISI 1045. The cutting speed was
3000 m/min with a feed of 0.3 mm/rev and a depth of cut of 3 mm.
J.P. Davim, C. Maranho / Materials and Design 30 (2009) 160165 163
RTS
t
r
r
n
8
where t
r
is the uncut chip thickness (mainly affected by the feed)
and r
n
is the tool edge radius [16,17] is bigger than 10 which is con-
sidered to be the minimum value to have reliable simulations.
5. Results and discussion
In this paper, the validation was made using the experimental
data given in Denkena et al. [13] so it was possible to calculate both
experimental plastic strain and plastic strain rate. Therefore, the
obtained results were compared with the predicted values found
in the simulations. The errors between experimentation and simu-
lation were calculated using the following formula and can be
found in Table 3:
Error
jExperimental value Simulated valuej
Experimental value
100% 9
In Table 4, a comparison between the analytical and simulated val-
ues for both the plastic strain and plastic strain rate is presented (all
values should be considered an average in the primary shear zone).
Iqbal et al. [18] stated that a realistic value for the friction coef-
cient is comprehended between 0.6 and 0.75 when machining
AISI 1045 with a cemented carbide cutting tool (with conventional
cutting speeds). The simulations in the present paper were con-
ducted under a friction value of 0.69 for 300 m/min (conventional
machining) and 0.31 for 3000 m/min (HSM). The friction values
were obtained using Coulomb model and the experimental forces.
A validation of the plastic strain and plastic strain rate was con-
ducted in order to verify the simulated values. However, this anal-
ysis was made using analytic methods (with experimental
measured data as an input for the formulation).
In Fig. 1, a comparison of the distribution of the plastic strain in
the workpiece and chip between both cutting regimes is presented.
Like expected, the plastic strain had felt a slight reduction in the
average value from the higher cutting speed to the lower cutting
speed (from 2.84 at 300 m/min to 2.40 at 3000 m/min in the sim-
ulated process and from 2.77 at 300 m/min to 2.36 at 3000 m/min
in the experimental process). This behavior can be explained by the
fact that the workpiece material suffered both a thermal softening
and a friction reduction due to the increase of the cutting speed
from 300 to 3000 m/min. When machined at 300 m/min, the sim-
ulated average temperature was about 1200 C and when ma-
chined at 3000 m/min, the simulated average temperature
reached about 1400 C. The friction coefcient, like already men-
tioned, has seen a decrease with the increase of the cutting speed.
The affected zone by the plastic strain is higher when machining at
high speed (comparing with the conventional speed machining).
The comparison between the plastic strain rates on both cutting
regimes can be observed in Fig. 2. The distribution of the strain rate
can be seen in the workpiece and chip. Once more, the results con-
verged to what was initially expected. As the cutting speed in-
creases, the plastic strain rate increases accordingly, showing a
rise from 2:78 10
5
at 300 m/min to 6:7 10
6
at 3000 m/min in
the simulated process and from 2:74 10
5
at 300 m/min to
6:3 10
6
at 3000 m/min in the experimental process. The affected
zone by the plastic strain rate is also higher when machining at
high speed (comparing with the conventional speed machining).
Also of note is that the elemental chip thickness is more evident
when machining at high cutting speeds and the chip has more curl-
ing when machined at higher cutting speeds (like can be observed
in both Figs. 1 and 2).
In Figs. 3 and 4, a comparison between the simulated and exper-
imental plastic strain and plastic strain rate (using analytical mod-
els) values are presented.
A reduction in the cutting force can be understood as an effect
of thermal softening (this causes a change in the effective stress
strain curve of the workpiece material, increasing the shear angle
and lowering this way the amount of plastic deformation required
to deform the chip). Although not observed in the simulated pro-
cess, segmented chips cause an additional decrease in the cutting
force due to being energetically favourable.
6. Conclusions
Friction coefcient is signicantly lower in high speed machin-
ing (HSM) when compared to traditional speed machining. This
leads to a reduction in the plastic strain, acting together with a
thermal softening due to an increase in the cutting speed. As defor-
mation proceeds, adiabatic heating occurs along the narrow bands
and thermal softening begins to dominate the deformation process.
This has an impact in the ow stress (starts to decrease with the
increase in strain). If a critical strain is reached, mechanical insta-
bility results in strain localization along narrow bands.
It was possible to determine with great precision both the plas-
tic strain and plastic strain rate. For conventional speed machining,
the plastic strain presented an error of only 2.5% and 1.4% for the
plastic strain rate. In high speed machining, the errors are also very
small presenting 1.6% of error in the plastic strain and 6.5% in the
plastic strain rate. Having that said, it is believed that nite ele-
ment analysis can be considered a good method to predict both
plastic strain and plastic strain rate in conventional and high speed
machining.
To sumit all up, the effect of HSMwas felt in multiple ways like:
a considerable reduction in the friction coefcient from conven-
tional to HSM,
in the chip morphology (more serrated shape in HSM),
an increase in the cutting temperature which in turn causes a
reduction of the plastic strain, and
a signicant increase of the plastic strain rate from conventional
to HSM.
Acknowledgments
The authors wish to thank the Foundation for Science and Tech-
nology (FCT), Portugal, Project POCTI/EME/61676/2004, for the use
of the software licence Advantedge

.
References
[1] Pawade R, Joshi S, Brahmankar K. Effect of machining parameters and cutting
edge geometry on surface integrity of high-speed turned Inconel 718. Int J
Machine Tools Manuf 2008;48:1528.
[2] Abukhshim N, Mativenga P, Sheikh M. Heat generation and temperature
prediction in metal cutting: a review and implications for high speed
machining. Int J Machine Tools Manuf 2006;46:782800.
[3] Campbell C, Bendersky L, Boettinger W, Ivester R. Microstructural
characterization of Al-7075-T651 chips and work pieces produced by high-
speed machining. Mater Sci Eng A 2006;430:1526.
[4] Velasquez J, Bolle B, Chevrier P, Geandier G, Tidu A. Metallurgical study on
chips obtained by high speed machining of a Ti6 wt.%Al4 wt.%V alloy. Mater
Sci Eng A 2007;452453:46974.
[5] Baker M. Finite element simulation of high-speed cutting forces. J Mater
Process Technol 2006;176:11726.
[6] Ekinovic S, Dolinsek S, Begovic E. Machinability of 90MnCrV8 steel during
high-speed machining. J Mater Process Technol 2005;162163:6038.
[7] Lee W, Yeh G. The plastic deformation behaviour of AISI 4340 alloy steel
subjected to high temperature and high strain rate loading conditions. J Mater
Process Technol 1997;71:22434.
[8] Gnevko A, Kazakov N, Lazarev D, Velichkovskii V. Effect of the loading rate on
the microhardness of metallic materials and the micromechanism of plastic
strain. Met Sci Heat Treat 2002;44(910).
[9] Gilat A, Wu X. Plastic deformation of 1020 steel over a wide range of strain
rates and temperatures. Int J Plast 1997;13(6-7):61132.
164 J.P. Davim, C. Maranho / Materials and Design 30 (2009) 160165
[10] Odeshi A, Al-ameeri S, Bassim M. Effect of high strain rate on plastic
deformation of a low alloy steel subjected to ballistic impact. J Mater
Process Technol 2005;162163:38591.
[11] Xue Q, Liao X, Zhu Y, Gray G. Formation mechanisms of nanostructures in
stainless steel during high-strain-rate severe plastic deformation. Mater Sci
Eng A 2005;410411:2526.
[12] Duan C, Wang M, Pang J, Li G. A calculational model of shear strain and strain
rate within shear band in a serrated chip formed during high speed machining.
J Mater Process Technol 2006;178:2747.
[13] Denkena B, Amor R, Len-Garca L, Dege J. Material specic denition of the
high speed cutting range. Int J Machining Machinability Mater 2007;2(2).
[14] Davim JP. Application of Merchant theory in machining particulate metal
matrix composites. Mater Des 2007;28:26847.
[15] Umbrello D, MSaoubib R, Outeiro J. The inuence of JohnsonCook material
constants on nite element simulation of machining of AISI 316L steel. Int J
Machine Tools Manuf 2007;47:46270.
[16] Outeiro J. Inuence of tool sharpness on the thermal and mechanical
phenomena generated during machining operations. Int J Machining
Machinability Mater 2007;2(3/4).
[17] zel T, Zeren E. Numerical modelling of meso-scale nish machining with
nite edge radius tools. Int J Machining Machinability Mater 2007;2(3/4).
[18] Iqbal S, Mativenga P, Sheikh M. Characterization of machining of AISI 1045
steel over a wide range of cutting speeds Part 1: investigation of contact
phenomena. J Eng Manuf 2007;221(5):90916.
J.P. Davim, C. Maranho / Materials and Design 30 (2009) 160165 165