ARTICLE IN PRESS

International Journal of Machine Tools & Manufacture 46 (2006) 17861794

www.elsevier.com/locate/ijmactool

Experimental and numerical modelling of the residual stresses

induced in orthogonal cutting of AISI 316L steel
J.C. Outeiroa,, D. Umbrellob, R. MSaoubic
a

Portuguese Catholic University, 3080-024 Figueira da Foz, Portugal

Department of Mechanical Engineering, University of Calabria, 87036 Rende (CS), Italy
c
Corrosion and Metals Research Institute (KIMAB), SE-114 28 Stockholm, Sweden

Received 4 October 2005; received in revised form 21 November 2005; accepted 24 November 2005
Available online 4 January 2006

Abstract
Residual stresses in the machined surface layers are affected by the cutting tool, work material, cutting regime parameters (cutting
speed, feed and depth of cut) and contact conditions at the tool/chip and tool/workpiece interfaces. In this paper, the effects of tool
geometry, tool coating and cutting regime parameters on residual stress distribution in the machined surface and subsurface of AISI
316L steel are experimentally and numerically investigated. In the former case, the X-ray diffraction technique is applied, while in the
latter an elasticviscoplastic FEM formulation is implemented. The results show that residual stresses increase with most of the cutting
parameters, including cutting speed, uncut chip thickness and tool cutting edge radius. However, from the range of cutting parameters
investigated, uncut chip thickness seems to be the parameter that has the strongest inuence on residual stresses. The results also show
that sequential cuts tend to increase supercial residual stresses.
r 2005 Elsevier Ltd. All rights reserved.
Keywords: Metal cutting; Machining; Residual stresses; Modelling; FEM

1. Introduction
The reliability of a mechanical component depends to a
large extent on the physical state of its surface layers. This
state includes the distribution of residual stresses induced
by machining. Depending on their nature (compressive or
tensile stresses) they could either enhance or impair the
ability of a component to withstand severe loading
conditions in service such as fatigue, creep, stress corrosion
cracking, etc. Furthermore, the residual stress distribution
on a component may also cause dimensional instability
(distortion) after machining [1]. This poses enormous
problems in engine/structural assembly and affects the
structural integrity of the whole part.
The direct inuence of residual stresses on the functional
behaviour (the static and dynamic strength, chemical and
electrical properties, fatigue, rust, etc.) of the component is
relatively well known. However, a number of questions still
Corresponding author. Tel.: +351 233 428445; fax: +351 233 428847.

E-mail address: jcouteiro@crb.ucp.pt (J.C. Outeiro).

0890-6955/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmachtools.2005.11.013

persist about the causes and the mechanisms of residual

stress generation in machining and how these residual
stresses could be controlled in order to achieve a desirable
distribution. Therefore, the understanding of residual
stresses and proper control of these in machining is a
prerequisite in order to enhance component performance
and minimize risks of failure.
The study of machining residual stresses is particularly
important when critical structural components are machined, especially if the objective is to reach high reliability
levels. This is the case of austenitic stainless steels, widely
used to produce critical structural components in chemical
industries and nuclear power stations because they provide
a unique combination of high mechanical properties and
corrosion resistance. However, austenitic stainless steels
are often regarded as difcult-to-machine materials because of their low thermal conductivity, high sensitivity to
strain and stress rate and severe work hardening. Moreover, their low thermal conductivity leads to heat concentration in the cutting zone, resulting in high localized
temperatures. As a result, machining of such steels induces

ARTICLE IN PRESS
J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

Nomenclature
E
n
r
T
h
k
Vc
t1
w
e
e_

Youngs modulus, MPa

Poisson ratio
mass density, kg/m3
temperature, 1C
convection coefcient, W/m2 K
thermal conductivity, W/m K
cutting speed, m/min
uncut chip thickness, mm
width of cut, mm
plastic strain
strain rate, s
1

relatively high residual stresses in the machined surface

layers, and therefore, greatly affects the properties of these
steels and their ability to withstand severe loading
conditions.
Several studies on residual stresses induced by machining
have been performed. Unfortunately, due to limitations in
nite element (FE) modelling of the metal cutting process
and the complex physical phenomenon involving the
formation of machining residual stresses, most of these
studies remain experimental in nature [24]. Although
many studies on FE modelling of the orthogonal cutting
process have been published until now, these were mainly
applied to predict with reasonable accuracy the strains,
stress and temperatures during cutting [5,6]. Only a few
studies on FE modelling involving the prediction of the
machining residual stresses with decent accuracy can be
found in the literature, with special attention to the residual
stresses in plain carbon steels and hardened steels [5,6].
Concerning modelling machining residual stresses in
stainless steels, the available studies are even more
restricted.
Wiesner [7] studied the residual stresses generated after
orthogonal cutting of AISI 304 steel using uncoated
cemented carbide tools. Using the X-ray diffraction
technique, Wiesner determined the inuence of the cutting
speed and cutting depth on in-depth distribution of the
residual stresses in the direction of primary motion (the
cutting speed direction). High tensile residual stresses (close
to +700 MPa) were found on the machined surface. In
order to explain these high tensile residual stresses, a nite
element method (FEM) was employed to analyse the
inuence of the thermal and mechanical effects on the
residual stress state separately, although in the paper he
presents the results for the thermal effect only. Wiesner
concluded that the thermal effect is not the only reason for
tensile residual stresses in machined components. The
mechanical effect does not always produce compressive
residual stresses, but can also contribute to tensile residual
stresses.
Liu and Guo [8] proposed an FE model to investigate the
effect of sequential cuts and tool-chip friction on residual

e_ 0
Tm
Troom
A
B
C
n
m
gn
an
kr
ls
rn

1787

reference plastic strain rate, s
1

melting temperature, 1C
room temperature, 1C
yield strength, MPa
hardening modulus, MPa
strain rate sensitivity coefcient
hardening coefcient
thermal softening coefcient
rake angle, deg
ank angle, deg
tool cutting edge angle, deg
tool cutting edge inclination angle, deg
tool cutting edge radius, mm

stresses in a machined layer of AISI 304 steel. They

reported a reduction in the supercial residual stresses
when the second cut is performed. Moreover, the residual
stresses can be compressive, depending on the uncut chip
thickness of the second cut. They also found that residual
stress on the machined surface is very sensitive to the
friction condition of the toolchip interface. Later, using
the same work material, Liu and Guo [9] presented a
similar study on the effect of sequential cuts on residual
stresses. They showed that decreasing the uncut chip
thickness below a critical value in the second cut may result
in favourable compressive residual stress distribution.
Thus, they conclude that it would be better to set an
appropriate nishing cut condition in consideration of the
effects of sequential cuts to control the residual stress
distribution. Unfortunately, Liu and Guo did not present
any experimental evidence for the work material under
investigation (AISI 304 steel) to validate their FE model.
Yang and Liu [10] performed a sensitivity study of the
friction condition on the toolchip contact, the cutting
forces and the residual stresses in machining-affected layers
of AISI 304 steel. In this study they proposed a new stressbased polynomial model for modelling the toolchip
contact, which represents a simple curve tting the
experimentally obtained shear and normal stresses acting
at the toolchip interface. When comparing this new
friction model with other friction models based on an
average friction coefcient deduced from cutting forces or
from stresses, they found signicant differences among the
predicted residual stresses. They concluded that the
conventional force-based friction model is inadequate to
predict the residual stresses induced by machining, and
they showed the potential for improving the quality in
predicting machining residual stress by adopting the stressbased polynomial model. Although it is widely accepted
that friction conditions will change along the toolchip
contact length, the authors did not present any experimental evidence to support their conclusions.
The prior investigations show that modelling residual
stress in machining stainless steel was studied for a very
limited range of cutting conditions and for specic analysis

ARTICLE IN PRESS
1788

J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

(inuence of the friction conditions and of the sequential

cuts on residual stresses). Therefore, there is a need for a
more fundamental model of the process, which takes into
consideration workpiece material behaviour, effects of
cutting tool geometry and material, and cutting regime
parameters concurrently to predict the residual stress
prole in machining.
In the present investigation, a numerical model of
orthogonal cutting of AISI 316L steel is employed with
the objective of predicting the in-depth residual stress
proles in the machined component for several cutting
parameters, including: cutting speed, uncut chip thickness,
tool geometry and the tool material. This allows the effects
of such cutting parameters and of the sequential cuts in the
surface and subsurface residual stress distributions to be
investigated. This numerical model is validated by comparing the predicted results (such as chip morphology, cutting
forces, temperatures and residual stresses induced by the
operation) to the experimental evidence conducted in the
laboratory.
2. Experimental and numerical procedures
2.1. Material properties, cutting tools and cutting regime
parameters
Orthogonal cutting tests were carried out on a lathe
using radial feed, as shown in Fig. 1. In order to keep
experiments to a manageable number, priority was given to
the parameters having greatest relevance to the present
investigation.

Fig. 1. Conguration of the orthogonal cutting tests and the directions of

the measurements of the cutting forces and the residual stresses.

Starting with the work material, round bars (150 mm in

diameter) of austenitic stainless steel AISI 316L were
selected for this study. Microstructure of the steel consisted
of an equiaxed grain structure characterized by an
approximate grain size of 50 mm and material hardness
was close to 170 HV. The chemical composition was
presented in earlier investigation [2] and the physical
properties are presented in Table 1. To model the elastic
behaviour of AISI 316L steel a Poisson ratio (n) of 0.30 and
a temperature (T) dependant Youngs modulus given by
Eq. (1) were used [11]:
ET 197845 140:76  T

0:5714  T 2 0:000T 3

MPa.

To model the thermo-viscoplastic behaviour of AISI

316L steel a JohnsonCooks constitutive equation [12] was
employed, which can be represented by the following
equation:




e_
n
seq A Be 1 C ln
e_0




 
T
T room m
 1

MPa,
2
T m
T room
where e is the plastic strain, e_ is the strain rate (s
1), e_0 is the
reference plastic strain rate (1 s
1), T (1C) is the workpiece
temperature, Tm(1399 1C) is the melting temperature of the
workpiece material and Troom (20 1C) is the room temperature. Coefcient A (MPa) is the yield strength, B (MPa) is
the hardening modulus, C is the strain rate sensitivity
coefcient, n is the hardening coefcient and m the thermal
softening coefcient. These material constants were obtained by Tounsi et al. [13] being A B 514 MPa,
C 0:508, n 0:0417 and m 0:533.
Furthermore, orthogonal cutting tests were conducted
using uncoated (ISO M10-M30) and coated (ISO P05-P25)
tungsten carbide tools. The chemical vapour deposition
(CVD) tool coatings employed were triple layer coatings
TiC/Al2O3/TiN, TiN being the external layer. The geometry of the cutting tools according to the ISO Standard
3002/1-1982 [14] is given in Table 3, concerning the normal
rake angle (gn), the normal ank angle (an) and the tool
cutting edge radius (rn). The tool cutting edge angle (kr)
was 901 and the tool cutting edge inclination angle (ls)
was 01.
Finally, the range of cutting regime parameters used in
the tests was selected using the data provided by tool
manufacturers for an optimal tool life. The cutting regime
parameters taken into consideration were the cutting speed

Table 1
Physical properties of AISI 316L steel [11]
Density (rw) (kg/m3)
Specic heat (cp,w) (J/kg K)
Thermal conductivity (kw) (W/m K)

rw T 7921
0:614  T 0:0002  T 2

cp;w T 440:79 0:5807  T
0:001  T 2 7  10
7  T 3
kw T 14:307 0:0181  T
6  10
6  T 2

ARTICLE IN PRESS
J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

(Vc), uncut chip thickness (t1) and width of cut (w). The
cutting speed was kept constant by the lathe controller,
which varied the rotation speed depending on the tool
position in radial direction. The range for the cutting
speed, uncut chip thickness and width of cut are given in
Table 3. No cutting uid was used in the tests.
2.2. Experimental set-up and parameters
Orthogonal cutting tests were performed on a 15 HP
numerically controlled lathe equipped with a specially
designed experimental set-up. In measuring the cutting
force components (tangential force and the thrust force) a
Kistler type 9255B three-component piezoelectric dynamometer was used. As shown in Fig. 1, the tangential force
component was measured in the circumferential direction
(the direction of the primary motion) and the thrust force
component in the axial direction (the direction perpendicular to the primary motion, along the tool cutting edge).
The output signals from the charge ampliers were
interfaced with a computer equipped with a data acquisition board. This experimental set-up also included thermal
imaging equipment developed to assess the temperature
distribution in the deformation zone. A detailed description of this equipment and its calibration can be found
in [15,16].
The residual stress state in the machined layers of the
workpiece was analysed by the X-ray diffraction technique
using the sin2 c method [17]. The parameters used in the
X-ray analysis are the same used in earlier investigation [2].
The residual stresses were determined for the stable
cutting zone (where the cutting force components are
stable) on the surface and in-depth, in the directions shown
in Fig. 1. These directions are selected because the analysis
of the residual stress tensors at the workpiece surface and
subsurface allowed us to conclude that these directions are
the directions of the principal stress components. To
determine the in-depth residual stresses, successive layers
of material were removed by electropolishing, to avoid the
reintroduction of residual stress. Further corrections to the
residual stress data were made due to the volume of
material removed. Due to circularity of the workpieces, a
circular mask with a diameter of 2.5 mm was applied to
limit the region of analysis. The error associated with the
method used was generally less than 50 MPa, but in some
cases where the depth exceeded 250 mm, it reached
100 MPa, due to the large grain size of the work material.

1789

elements, while the tool, modelled as rigid, was meshed and

subdivided into 1000 elements. A plane-strain coupled
thermo-mechanical analysis was performed using orthogonal assumption. The constitutive law used and implemented in the FEM model to take into account this complex
material behaviour is given by Eq. (2). The thermo-physical
properties of the workpiece are in Table 2 and that of the
uncoated and coated cutting tools in Table 3. As far as
friction modelling is concerned, a simple model based on
the constant shear hypothesis was implemented with the
shear factor equal to 0.8.
Another consideration concerns the interface heat
transfer coefcient, h, utilized in the preliminary thermomechanical numerical simulation. Usually a large h value is
adopted [18], assuming perfect interface heat transfer
conditions. However, the role of this parameter requires
careful investigation, since it controls the effectiveness of
the numerical simulation when a fully thermo-mechanical
analysis is run, as reported in a recent study conducted
by Filice et al. [19]. They found a value of h close to
1000 kW/m2 K, which permitted a satisfactory agreement
between their numerical data and the experimental
evidence in all the investigated cases.
Due to the known drawbacks of the present chip
separation criteria to simulate the material separation in
the form of a chip from the rest of the workpiece, a
remeshing technique was applied [20].
During the machining of AISI 316L steel, tensile stress
plays an important role, therefore its effect should also be
included in the material response such as the fracture
criterion. In this research, Cockroft and Lathams criterion
[21] is employed to predict the effect of tensile stress on
chip segmentation during orthogonal cutting. Cockroft
and Lathams criterion is expressed as
Z ef
s1 de C,
(3)
0

Table 2
Physical properties of the different materials of the cutting tools [2628]
Tool materials
WC-Co
Coating layer thickness (mm)

Density (rt) (kg/m3)

13,000
Specic heat (cp,t) (J/kg K)
234
Thermal conductivity (kt) (W/m K)
62.7

TiC
2.5
4940

Al2O3
1
3970

TiN
3
5430

Function of the temperature. See the above references for detailed

information.

2.3. Numerical model and parameters

The commercial FEA software DEFORM-2DTM, a
Lagrangian implicit code designed for metal forming
processes, was used to simulate the orthogonal cutting
process of AISI 316L (Table 1). An FEM model of the
orthogonal cutting process was developed and was
composed of the workpiece and the tool. The workpiece
was initially meshed with 3000 isoparametric quadrilateral

Table 3
Tool geometry and cutting regime parameters
Cutting tool g (deg) a (deg) rn (mm) Vc (m/min) t1 (mm)
Uncoated

Coated

5
0
5
0

11
11

0.030
0.055
0.100
0.030

w (mm)

100200

0.10.3

100200

0.10.3

ARTICLE IN PRESS
1790

J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

where ef is the effective strain, s1 is the principal stress, and

D is a material constant. Cockroft and Lathams criterion
states that when the integral of the largest tensile principal
stress component over the plastic strain path in Eq. (3)
reaches the value of D, usually called damage value,
fracture occurs or chip segmentation starts. The determination of D value was established by an iterative procedure
on chip geometry in this research and it was found to be
equal to 68 MPa.
Taking into account that, at the moment is not possible
to obtain directly the numerical residual stresses using the
DEFORM-2DTM FE code, the authors have proposed the
follow procedure:
1. An elasticviscoplastic model was implemented for the
investigated cases and it was executed for a total time
step long enough to reach the steady-state condition.
2. For several time steps the tool was released from the
machined surface (unloading phase);
3. The numerical residual stress proles were collected for
a few selected time steps and the average values were
taken into account.
More detailed information on the FEM model, material
model and simulation results of chip morphology, temperatures, etc. is provided in Refs. [22,23].
3. Comparison between experimental and numerical results
To validate the orthogonal cutting model the predicted
and experimentally measured chip morphology, cutting
forces, temperatures and residual stresses on the machined
affected layers were compared and their differences were
discussed.
Table 4 shows both predicted and measured chip
morphology, concerning the average values of the chip
geometry (valley, peak and pitch) and chip compression
ratio (CCR). As shown in Table 4, the predicted and
measured chip geometry parameters are very close.
Comparison of the measured and predicted cutting force
components and temperatures generated during the cutting
process are shown in Table 5 and Fig. 2, respectively. As
shown in Table 5, the average values of both predicted and
measured cutting force components are almost identical.
However, Fig. 3 shows that in general the measured
temperatures are higher (about 200 1C more) than the
predicted ones. The main reason for this difference could
be due to the fact that the temperatures are measured for
steady-state conditions, while by simulation the temperatures are obtained for a very short cutting process time,
which is not sufcient to reach steady-state conditions [19].
Finally, to fully validate the numerical model, the
predicted and measured residual stress distributions at
surface and subsurface were compared. In order to
properly compare the predicted residual stresses with those
measured, these should be extracted from the FEM model
in the same conditions as those applied experimentally. It is

Table 4
Comparison between experimentally and numerically obtained chip
geometry. Cutting parameters: H13A tool, V c 100 m/min, t1 0:2 mm
and w 6 mm (CCRChip compression ratio)

Table 5
Comparison between experimentally and numerically obtained cutting
forces. Cutting parameters: H13A tool, V c 100 m/min, t1 0:2 mm and
w 6 mm

Experimental
Numerical

Tangential force (N)

Trust force (N)

3494
3435

2838
2883

well known that residual stresses obtained by the X-ray

diffraction technique are averaged over a nite volume of
work material (a given supercial area and a depth of a few
micrometers), which depends on the nature of the X-ray
beam employed [17]. Therefore, the predicted residual
stress values presented in this paper were obtained by
calculating the average residual stresses over several points
within the same material volume as that affected by the
X-ray beam.
Fig. 3 shows both predicted and measured in-depth
residual stresses for a cutting speed of 200 m/min and an
uncut chip thickness of 0.1 mm. As shown in Fig. 3, the
predicted and measured in-depth residual stress proles are
very well correlated. For all investigated cutting conditions
the surface and subsurface residual stresses are almost
tensile in circumferential and axial directions, the residual
stresses being higher in the circumferential direction. As a
consequence, the circumferential residual stresses are the
most critical to component performance. They are higher
at the surface, sometimes reaching a value of around
1000 MPa in the circumferential direction. The level of
residual stresses in both directions decreases continuously
with depth into the machined surface, stabilising at a level
corresponding to that found in the work material before
machining (zero in axial direction and about 200 MPa in
circumferential direction). The actual depth at which the

ARTICLE IN PRESS
J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

1791

Fig. 2. Comparison between experimentally (a) and numerically (b) obtained temperature distributions.

effects of sequential cuts on residual stresses will also be

performed. In this study the width of cut was kept constant
and equal to 6 mm.
As seen in the previous section, the circumferential
residual stresses are higher than the axial residual stresses,
so they are the most critical for part performance.
Therefore, the following analysis will be performed, taking
into account only the circumferential residual stresses.

Experimental - Axial

1000

Residual Stresses (MPa)

FEM - Axial
800

Experimental - Circumferential
FEMl - Circumferential

600
400
200

4.1. Influence of cutting regime parameters

0
0.0

0.1

0.2

0.3

0.4

0.5

-200

Depth (mm)
Fig. 3. Comparison between experimentally and numerically obtained
residual stresses.

residual stresses reach the zero stress value can be thought

of as the thickness of the tensile layer due to machining.
Because the circumferential residual stresses did not reach
zero stress within the rst 500 mm in this study, this
thickness will only be evaluated in the axial direction.
As a result, the predicted results were nearly of the same
magnitude as those obtained experimentally, and therefore
the FEM model can be applied to study the inuence of
cutting parameters on residual stresses.
4. Inuence of cutting parameters on residual stresses
In this section the inuence of cutting parameters on
surface and subsurface distributions of residual stresses will
be studied. These parameters include the cutting speed,
uncut chip thickness, tool geometry (rake angle and cutting
edge radius) and tool material (coated and uncoated tool
with the same substrate). Furthermore, a study on the

The inuence of cutting regime parameters on residual

stresses was studied using both coated and uncoated tools
having 01 rake angle and varying the cutting speed and the
uncut chip thickness.
The inuence of the cutting speed on residual stresses
was analysed keeping the uncut chip thickness constant at
0.1 mm. It was observed that when using the uncoated tool,
the supercial circumferential residual stresses did not
change at all when the cutting speed increased from 100 to
200 m/min, whereas when using the coated tool, the
supercial circumferential residual stresses increased
(Fig. 4). Using both coated and uncoated tools, a reduction
of the thickness of the tensile layer was observed for high
cutting speeds. These results are in agreement with those
obtained experimentally by MSaoubi et al. [2].
The inuence of the uncut chip thickness on residual
stresses was studied, keeping an invariable cutting speed of
100 m/min. As seen in Fig. 5, the circumferential residual
stress at the surface increased by about 400 MPa when the
uncut chip thickness increased from 0.1 to 0.2 mm.
Furthermore, an increase of the thickness of the tensile
layer was observed when the uncut chip thickness
increased. Comparative to the earlier experimental results
obtained in [2], the predicted results are in agreement
concerning the thickness of the tensile layer, but in

ARTICLE IN PRESS
J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

1792

1400
Vc=200 m/min

1200

Vc=150 m/min

Circu. Residual Stresses (MPa)

Circu. Residual Stresses (MPa)

1400

Vc=100 m/min

1000
800
600
400
200
0
0.0

0.1

0.2

0.3

0.4

0.5

gama = 5

1200

gama = -5
gama = 0

1000
800
600
400
200
0
0.0

-200

0.1

0.2

0.3

0.4

0.5

-200

Depth (mm)

Depth (mm)
Fig. 4. Evolution of the circumferential residual stresses in depth for three
different cutting speeds, using the coated tool.

Fig. 6. Evolution of the circumferential residual stresses in depth for three

different tool rake angles.

1400

t1=0.1 mm

1200

t1=0.2 mm
1000
800
600
400
200
0
0.0

0.1

0.2

0.3

0.4

.
0.5

-200

Depth (mm)
Fig. 5. Evolution of the circumferential residual stresses in depth for two
different uncut chip thicknesses.

disagreement concerning the residual stress distribution at

the surface.
4.2. Influence of tool geometry
To study the inuence of tool geometry on residual
stresses, uncoated tools were used. Of all the cutting
geometry parameters the rake angle and the cutting edge
radius were selected for this study, due to their well known
strong inuence on residual stresses. The cutting speed and
the uncut chip thickness were kept constant and equal to
150 m/min and 0.1 mm, respectively.
The inuence of the tool rake angle on residual stresses
was analysed for three different rake angles:
51, 01 and 51.
As shown in Fig. 6, the circumferential residual stresses at

Circu. Residual Stresses (MPa)

Circu. Residual Stresses (MPa)

1400

Rn = 0.100 mm

1200

Rn = 0.055 mm
1000
Rn = 0.030 mm
800
600
400
200
0
0.0

0.1

0.2

0.3

0.4

0.5

-200

Depth (mm)
Fig. 7. Evolution of the circumferential residual stresses in depth for three
different tool cutting edge radius values.

the surface slightly decreased when the rake angle increased

from
51 to 51. This decrease on supercial residual
stresses with the rake angle was accomplished with a
reduction of the thickness of the tensile layer. The same
variations of the surface residual stresses and thickness of
the tensile layer with rake angle were obtained experimentally in earlier investigation [2].
The inuence of the tool cutting edge radius on residual
stresses was analysed for three different edge radii: 0.030,
0.055 and 0.100 mm. As seen in Fig. 7, the supercial
circumferential residual stresses increased when the cutting
edge radius increased from 0.030 to 0.100 mm. Moreover, a
reduction of the thickness of the tensile layer was observed
when the tool cutting edge radius increased. As far as the
authors know, no previous study on the inuence of the

ARTICLE IN PRESS
J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794
1400

Uncoated tool

Circu. Residual Stresses (MPa)

Circ. Residual Stresses (MPa)

1400
1200

1793

Coated tool

1000
800
600
400
200

3rd cut
2nd cut

1200

1st cut

1000
800
600
400
200
0
0.0

0
100

150

-200

200

0.1

0.2

0.3

0.4

0.5

Depth (mm)

Cutting Speed (m/min)

Fig. 8. Supercial circumferential residual stresses for coated and
uncoated tools.

tool cutting edge radius on residual stresses in austenitic

stainless steels has been reported. However, in this work
the variation of the supercial residual stresses with the
tool cutting edge radius is opposite to that found in the
literature when machining hardened AISI 52100 steel [24].
4.3. Influence of tool material
The inuence of the tool material on residual stresses
was studied using coated and uncoated tools having 01 rake
angle, keeping the uncut chip thickness constant at 0.1 mm.
This study was performed at three different cutting speed
values: 100, 150 and 200 m/min. The obtained results
demonstrated that when the uncoated tool was replaced by
the coated tool the most relevant change on residual
stresses was observed at the surface of the machined
workpiece. As shown in Fig. 8, the difference between the
residual stresses produced by the coated and uncoated
tools is remarkable only for the highest value of the cutting
speed. For such conditions, the uncoated tool produces
higher circumferential residual stresses when compared
with the coated tool. These results are once again very
similar to those found experimentally in earlier investigation [2].

Fig. 9. Effects of sequential cuts on in-depth distribution of the

circumferential residual stresses.

thickness of the tensile layer was also observed with the

number of cuts, this difference also being greater between
the rst and the second cuts. These results are opposite to
those found numerically by Liu and Guo [8], using AISI
304 stainless steel, and by Ee et al. [20], using 1045 steel.
Unfortunately, from one cut to another the tool wear
was not taken into account. This may have to be
considered in future work.
5. Conclusions
An experimental and numerical analysis of residual
stresses induced by orthogonal cutting of AISI 316L was
performed in the present investigation. The reasonable
agreement obtained between the experimental and numerical results indicate that the proposed FEM model appears
to be suitable for studying the inuence of cutting
parameters on residual stress arising from machining.
The following conclusions can be drawn from our
investigation:




4.4. Effect of sequential cuts

To study the effects of sequential cuts on residual stresses
an uncoated tool with 01 rake angle was used, keeping the
cutting speed constant at 200 m/min and the uncut chip
thickness at 0.1 mm. Fig. 9 shows the results obtained for
three successive cuts. As seen in this gure, the supercial
circumferential residual stresses increased with the number
of cuts. Additionally, a greater variation on supercial
circumferential residual stresses was observed between the
rst and the second cuts. Moreover, an increase in the




Residual stresses increase with most of the cutting

parameters, including cutting speed, uncut chip thickness and tool cutting edge radius.
When the uncoated tool was replaced by a coated one
the supercial residual stresses increase by 240 MPa,
when the highest cutting speed value was used.
An increase in tool rake angle seems to reduce the
supercial residual stresses by 140 MPa, when this angle
changes from
51 to 51.
Within the range of cutting parameters investigated, the
uncut chip thickness seems to be the parameter having
the largest inuence on residual stresses (about
390 MPa, when the uncut chip thickness increases from
0.1 to 0.2 mm), which is in agreement with most of the
experimental results found in the literature [25].

ARTICLE IN PRESS
1794

J.C. Outeiro et al. / International Journal of Machine Tools & Manufacture 46 (2006) 17861794

The sequential cuts tend to increase supercial residual

stresses by 280 MPa from the rst to the third cut. This
result is an opposite trend to the result found
numerically by Liu and Guo [8], using AISI 304 stainless
steel, and by Ee et al. [20], using 1045 steel.

Acknowledgements
The rst author (J.C. Outeiro) gratefully acknowledges
the Portuguese Foundation for Science and Technology
(FCT) for the nancial support. Authors also thank the
reviewers for their suggestions and comments.
References
[1] R. Schultz, M. Karabin, Characterization of machining distortion by
strain energy density and stress range, in: Proceedings of the Sixth
European Conference on Residual Stresses, Materials Science Forum,
2002, pp. 6167.
[2] R. MSaoubi, J.C. Outeiro, B. Changeux, J.L. Lebrun, A.M. Dias,
Residual stress analysis in orthogonal machining of standard and
resulfurized AISI 316L steels, Journal of Materials Processing
Technology 96 (13) (1999) 225233.
[3] J.C. Outeiro, A.M. Dias, J.L. Lebrun, V.P. Astakhov, Machining
residual stresses in AISI 316L steel and their correlation with the cutting
parameters, Machining Science and Technology 6 (2) (2002) 251270.
[4] R. MSaoubi, H. Chandrasekaran, Role of cutting parameters and
microstructure on chip formation and surface integrity during machining of AISI 316L, Transactions of NAMRI/SME 30 (2002) 199206.
[5] J. Mackerle, Finite-element analysis and simulation of machining: a
bibliography (19761996), Journal of Materials Processing Technology 86 (13) (1999) 1744.
[6] J. Mackerle, Finite element analysis and simulation of machining: an
addendum a bibliography (19962002), International Journal of
Machine Tools & Manufacture 43 (2003) 103114.
[7] C. Wiesner, Residual stresses after orthogonal machining of AISI 304:
numerical calculation of the thermal component and comparison with
experimental results, Metallurgical Transactions A 23A (1992) 989995.
[8] C.R. Liu, Y.B. Guo, Finite element analysis of the effect of sequential
cuts and toolchip friction on residual stresses in a machined layer,
International Journal of Mechanical Sciences 42 (2000) 10691086.
[9] Y.B. Guo, C.R. Liu, FEM analysis of mechanical state on
sequentially machined surfaces, Machining Science and Technology
6 (1) (2002) 2141.
[10] X. Yang, C.R. Liu, A new stress-based model of friction behavior in
machining and its signicant impact on residual stresses computed by
nite element method, International Journal of Mechanical Sciences
44 (4) (2002) 703723.

[11] P. Lacombe, B. Baroux, G. Beranger (Eds.), Les Aciers Inoxidables,

Les Ulis, Franc- a, Les Editions de Physique, 1990.
[12] G.R. Johnson, W.H. Cook, A constructive model and data for metals
subjected to large strains, high strain rates and high temperatures, in:
Proceedings of the Seventh International Symposium on Ballistics,
1983, pp. 541547.
[13] N. Tounsi, J. Vicenti, A. Otho, M.A. Elbestawi, From the basic
mechanisms of orthogonal metal cutting toward the identication of
the constitutive equation, International Journal of Machine Tools
and Manufacture 42 (2002) 13731383.
[14] ISO3002/1, Basic Quantities in Cutting and GrindingPart 1,
International Organization for Standardization (ISO), 1982.
[15] R. MSaoubi, C. Le Calvez, B. Changeux, J.L. Lebrun, Thermal and
microstructural analysis of orthogonal cutting of low alloyed carbon
steel using an infrared-charge-coupled device camera technique,
Proceedings of the Institution of Mechanical Engineers, Part B:
Journal of Engineering Manufacture, 216 (2002) 153165.
[16] J.C. Outeiro, A.M. Dias, J.L. Lebrun, Experimental assessment of
temperature distribution in three-dimensional cutting process,
Machining Science and Technology 8 (3) (2004) 357376.
[17] I.C. Noyan, J.B. Cohen, Residual StressMeasurement by Diffraction and Interpretation, Springer, New York, 1987.
[18] Y.C. Yen, J. Sohner, B. Lilly, T. Altan, Estimation of tool wear in
orthogonal cutting using the nite element analysis, Journal of
Materials Processing Technology 146 (2004) 8291.
[19] L. Filice, D. Umbrello, F. Micari, L. Settineri, On the nite element
simulation of thermal phenomena in machining processes, in: Proceedings of the Eighth ESAFORM Conference, 2005, pp. 729732.
[20] K.C. Ee, O.W. Dillon, I.S. Jawahir, Finite element modeling of
residual stresses in machining induced by cutting using a tool with
nite edge radius, International Journal of Mechanical Sciences 47
(10) (2005) 16111628.
[21] M.G. Cockroft, D.J. Latham, Ductility and workability of metals,
Journal of the Institute of Metals 96 (1968) 3339.
[22] S. Kobayashi, S.K. Oh, T. Altan, Metal forming and the niteelement method, Oxford University Press, Inc., Oxford, 1989.
[23] Deform2D, User Manual V 8.1, Scientic Forming Technologies
Corporation, Columbus, OH, USA, 2003.
[24] J.D. Thiele, S.N. Melkote, R.A. Peascoe, T. Watkins, Effect of
cutting-edge geometry and workpiece hardness on surface residual
stresses in nish hard turning of AISI 52100 steel, Journal of
Manufacturing Science and Engineering 122 (2000) 642649.
[25] E. Capello, Residual stresses in turning. Part I: inuence of process
parameters, Journal of Materials Processing Technology 160 (2005)
221228.
[26] ASM Handbook, Machining, vol. 16, ASM, 1989.
[27] I.S. Jawahir, C.A. Van Luttervelt, Recent developments in chip
control research and applications, Annals of the CIRP 42 (2) (1993)
659693.
[28] C. Le Calvez, Etude des Aspects Thermiques et Metallurgiques de la
Coupe Orthogonale dun Acier au Carbone, Ph.D. Thesis, Ecole
Nationale Superieure dArts et Me`tiers, Paris, 1995 (in French).