Journal of Manufacturing Processes 85 (2023) 216–226

Contents lists available at ScienceDirect

Journal of Manufacturing Processes

journal homepage: www.elsevier.com/locate/manpro

Evolution of residual stress when turning a fillet radius in stainless steel

Yang Liu a, *, Jian Weng a, b, Rachid M'saoubi a, c, Aylin Ahadi a, Jinming Zhou a
a
Department of Mechanical Engineering Sciences, Lund University, Box 118, 22100 Lund, Sweden
b
Hubei Digital Manufacturing Key Laboratory, School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
c
Seco Tools AB, Björnbacksvägen 10, 737 82 Fagersta, Sweden

A R T I C L E I N F O A B S T R A C T

Keywords: Most studies have been carried out to investigate the surface integrity induced by metal cutting process. How­
Finite element (FE) modelling ever, the previous studies are limited to a longitudinal turning or orthogonal cutting operations and the residual
CEL stresses generated in a fillet radius have been ignored. This study uses a combination of experiments and nu­
Fillet radius turning
merical simulations to study the evolution of cutting forces, temperature, chip morphology, and residual stress
Thermal-mechanical load
Residual stress
distributions while turning a fillet radius in AISI 304. Finite Element (FE) models were developed with a Coupled
Eulerian and Lagrangian (CEL) method, where the geometric model of the workpiece was established taking into
account the previous machined surface profile at the four specific cutting faces. The model was validated by
experimental cutting forces, chip morphology, and residual stress profiles. The changing trend of shape and area
of uncut chip cross-section during fillet turning were analyzed to explain the evolution of cutting forces and
temperatures. The results show that the cutting force components in cutting speed and tangential directions
increase during the early stage of the fillet turning process and decrease after that, while the force in the radial
direction shows an increasing trend during this process. The maximum temperature at the machined surface is
increased along the tool path. In addition, magnitude and depth of residual stress are slightly changed during the
fillet radius turning process, but a reduction of the residual stress profile can still be noticed.

1. Introduction crack nucleation and propagation [4], demonstrating the importance to

get deeper insights into the thermal-mechanical load and the turning-
Residual stress induced by the metal cutting processes is of consid­ induced surface integrity during a fillet radius turning process.
erable importance because it plays a crucial role in the performance and Some preliminary cutting tests were conducted to reveal the modi­
service life of machined components. Tensile residual stresses in fication of surface integrity along the fillet radius. Dumas et al. [5]
machining are unwanted since they can lead to premature fatigue and characterized the residual stresses, microstructure and mechanical
corrosion failures [1]. On the other hand, compressive residual stresses properties in a martensitic stainless steel while performing such turning
are often beneficial since they can increase corrosion resistance and operation. It was found that the microstructure remained stable along
fatigue strength [2,3]. Therefore, the investigation of residual stress is the fillet radius while residual stress shows a significant difference be­
particularly important for the design and manufacture of components. tween longitudinal turning and end face turning. This was explained by
Many research have been conducted to reveal residual stresses distri­ the small grain size of the 15-5PH martensitic structure and low strain
butions influenced by workpiece material and a wide range of cutting hardening modulus that prevents microstructure modifications to be
parameters (cutting speed, feed rate, tool geometries, etc.) [3]. Never­ observed. A further investigation was conducted to explore the surface
theless, most of these studies are limited to the restrictive assumptions of integrity of a 316 L austenitic steel when turning a fillet radius [4]. It
orthogonal cutting or longitudinal cutting, and machining-induced was revealed that the affected depth in terms of microstructure and
surface integrity when turning a fillet radius was seldomly reported in mechanical property modifications is twice larger for longitudinal
the previous literature. In fact, in the industrial applications components turning than for end face turning. In addition, the compressive residual
are commonly consist of many surfaces, and the joints of these surfaces stress penetrates deeper into the workpiece with a much higher
are usually machined into fillet radius. Due to the geometrical-induced magnitude during the longitudinal turning than end face turning. On the
stress concentration, the existence of fillet radius may promote fatigue other hand, Weng et al. [6] found a contrary trend of residual stress

* Corresponding author.
E-mail address: yangliux@mit.edu (Y. Liu).

https://doi.org/10.1016/j.jmapro.2022.11.038
Received 19 May 2022; Received in revised form 13 November 2022; Accepted 15 November 2022
Available online 29 November 2022
1526-6125/© 2022 The Authors. Published by Elsevier Ltd on behalf of The Society of Manufacturing Engineers. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

evolution during fillet radius turning of AISI 304, where a more Table 1
compressive residual stress was found in end face turning. The reason for Summaries of fillet radius turning and 3D metal cutting simulations.
this could be the different workpiece materials and cutting parameters Reference Workpiece Tool Cutting Remarks
employed in these studies. material material parameters
As it was reviewed, experimental measurement is the main approach Dumas 316 L Carbide Fillet radius Residual stress
to reveal the surface integrity of the components generated in fillet et al. austenitic tools turning becomes more
radius turning operations. However, the evaluation of surface integrity, [4,5] stainless Vc = 100 m/ compressive during
such as, surface roughness, microhardness, microstructure, and residual steel min the longitudinal
f = 0.3 mm/ turning than end
stress using experimental methods is a very time-consuming work, and rev face turning. The
because of the destructive nature of most testing techniques used, it is d1 = d2 = 0.3 study was carried
also often an expensive and a non-sustainable production approach mm out with
[2,7]. In addition, there are some interesting aspects which are difficult R = 5 mm experimental tests.
Weng et al. AISI 304 CVD- Fillet radius Residual stresses
to measure by experiment. For example, it is very hard to obtain the
[6] stainless coated turning have a slight
temperature distribution in the cutting zone due to influence of gener­ steel carbide Vc = 180 m/ difference between
ated chips [8]. Also, the in-depth distribution of residual stress in fillet tools min outer face and end
radius zone cannot be measured using X-ray diffraction with electro­ f = 0.15 mm/ face, but a
polishing due to the difficulty in etching the arc faces [9]. The limita­ rev reduction of the
d1 = d2 = 0.3 profile is still
tions of the experiment method restrict the knowledge of residual stress mm noticeable. The
distribution on the fillet radius and a quantitative understanding of the R = 5 mm study was carried
thermo-mechanical effects during the fillet radius turning process. out with
Residual stresses are generated in machined components due to the experimental tests.
Özel et al. AISI 4340 PCBN tools Turning The presence of
combination effect of thermal load, mechanical load, and metallurgical
[20] steel Vc = 125, variable design
structural alterations [3]. It is therefore imperative to obtain the rele­ 175, 300 m/ cutting tool inserts
vant information (temperature, forces, stress, plastic strain, etc.) during min was considered in
machining process as detailed as possible to reveal the basic mechanism f = 0.1, 0.15 FE modelling of 3D
of residual stress generation while turning a fillet radius. Over the past mm/rev turning process.
ap = 0.5, 1
decades, FE methods have been widely used in the simulation of metal
mm
cutting processes [10]. With the use of FE methods, temperature dis­ Agmell AISI 316 L PCBN tools Orthogonal 3D FE model with
tributions [11], cutting forces [12], chip morphology [13], stress fields et al. stainless cutting CEL formulation
within the cutting tools [14], and surface integrity characteristics (re­ [21] steel Vc = 250 m/ was proved to be a
min feasible and
sidual stresses [15,16], plastic deformation [17], microstructural alter­
h1 = 0.2 mm effective approach
ations [18], white layer thickness [19]) on machined surface have been b = 1 mm to study both a
investigated comprehensively by many scholars. Other than predicting segmented chip
the orthogonal cutting process, some scholars have achieved the three- formation and the
dimensional metal cutting simulations to offer a more realistic periodic loads
acting on the tool.
machining process modelling. Through the use of 3D FE simulations
Valiorgue AISI304L TiN coated Turning A 3D hybrid model
[20], the temperature and stress distributions were predicted to reveal et al. stainless carbide Vc = 50–200 was proposed to
the advantages of variable edge design tools. Mathias et al. [21] [22] steel tools m/min predict residual
developed a 3D model for a serrated chip formation when machining f = 0.2–0.4 stresses induced by
mm/rev turning process,
AISI 316L. Valiorgue et al. [22] proposed a 3D hybrid model to predict
ap = 0.3 mm which significantly
residual stresses induced by turning process. In this model, the chip reduced
formation process was removed and replaced by acting equivalent computation time.
thermo-mechanical loadings that was obtained from experimental Arrazola Inconel 718 Uncoated Turning 3D FE simulations
measurements, which significantly reduced computation time. Arrazola et al. carbide Vc = 30, 70 were established to
[23] tools m/min explore the
et al. [23] investigated the residual stress distributions influenced by the
f = 0.15, influence of
material model and frictional model applied in 3D FE simulations. With 0.25 mm/rev material flow stress
a 3D machining model, Ozel et al. [24] studied the effect of modified ap = 0.15 and friction
material model on the cutting forces, temperatures, strain and stress mm parameters
employed in FE
distributions when turning Inconel 718. Yadav et al. [25] established a
simulations on the
3D model using DEFORM 3D to investigate the effect of cutting pa­ machining induced
rameters on material removal rate and flank wear. Previous literature stress predictions.
indicates that most of the numerical work took orthogonal cutting or Ozel et al. Inconel 718 Carbide Face turning This work presents
longitudinal turning into account and thus they are not adequate to [24] tools Vc = 30, 70 a comparison of 3D
m/min machining models
understand the mechanisms and surface integrity in turning fillet radius.
f = 0.25 mm/ developed using
Table 1 lists the summaries of the previous research about the fillet rev ABAQUS/Explicit
radius turning and 3D metal cutting simulations. ap = 0.15 and DEFORM™ 3D
In this paper, a Coupled Eulerian and Lagrangian method is mm Machining.
employed to predict residual stresses distributions while turning a fillet Yadav et al. Inconel 718 CVD Turning 3D machining
[25] coated N= simulations were
radius in AISI 304 using coated carbide inserts. First numerical insights carbide 421–1020 carried out to
about the thermal-mechanical load and surface integrity modifications tools RPM predict the flank
when turning a fillet radius are provided. The workpiece in four f = 0.08–0.2 wear and material
different cutting faces is modelled with a pre-generated surface profile. mm/rev removal rate.
ap = 0.4–1
The 3D FE model is validated by comparing cutting forces, chip
mm
morphology and residual stress distributions between the predicted re­
sults and experimental measurements. In addition, the evolution of force
components, temperature distributions, chip morphology and residual

217
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

stresses profiles along the fillet radius are discussed with the analysis of Table 2
the varying cutting parameters during this process. Tool geometry and cutting conditions.
Parameters Description Value
2. Experimental procedures
Tool geometry Kr ( )
◦
Major cutting-edge angle 72.5
λs (◦ ) Tool cutting edge inclination angle − 6
Fig. 1a shows the experimental settings of the present study. The γ n (◦ ) Normal rake angle − 6
cutting experiments were performed on an SMT500 CNC turning center. εr (◦ ) Tool included angle 55
The workpiece material employed in this work is AISI 304 stainless steel. r (mm) Tool nose radius 0.8
rβ (μm) Tool edge radius 60
The employed tool is CVD coated carbide tools (ISO DNMA150608). Cutting conditions R (mm) Fillet radius 5
Several revolutions were conducted prior to the experiment to generate d1 (mm) Depth of cut in outer face turning 0.3
a fillet radius. After that, the experiment was performed with a new tool. d2 (mm) Depth of cut in end face turning 0.3
Cutting conditions and tool geometries are listed in Table 2. The edge Vc (m/min) Cutting speed 180
f (mm/r) Feed 0.15
radius of the tool was pre-measured by Edge Master Module of Alicona
to ensure its accuracy (Fig. 1b). Several measurements were carried out
along the tool nose and the uncertainty of the edge radius was found Abaqus/Explicit v6.14-2. CEL method is used to completely avoid the
within ±3 μm. During the experiment, Kistler 9129AA dynamometer problem with severe element distortion and ensure proper chip
was used to measure the real-time cutting force exerted on the cutting morphology [16]. In CEL method, the cutting tool is described by the
tool. Original signals of cutting forces were acquired at an acquisition Lagrangian formulation and the AISI 304 stainless steel workpiece with
rate of 100 Hz in the cutting force measurement. Since the acquisition a Eulerian one. The relative position of the tool and workpiece is
frequency is low and the transition of the force components from outer adjusted based on the parameters used in the experiment. As an
face to end face are smoothly during the fillet radius turning process, example, schematic views of the FE model and the applied boundary
there is no filter utilized in the cutting force measurement. The chips conditions when turning the outer face are shown in Fig. 3a. The
generated during cutting experiments was collected and measured dimension and mesh employed for the cutting tool and the workpiece
under microscope. After the experiment, machining-induced residual are illustrated in Fig. 3b and c. The tool is described as a rigid body and
stresses in cutting direction were measured using X-ray diffraction and fixed in all displacement degrees of freedom. The Euler-part is dis­
electropolishing techniques. For the outer face and end face, residual cretized by 8-node thermally coupled linear Eulerian brick elements
stresses distributions along the depth beneath the machined surface (EC3D8RT), and 4-node thermally coupled tetrahedron elements
were measured by electropolishing to remove successive layers of ma­ (C3D4T) are used for the cutting tool. The investigation on mesh
terial. Since it is difficult to etch the arc faces, only the magnitude of the sensitivity of chip morphology, temperature distributions and cutting
residual stress at the machined surface was measured at the cutting faces forces when using CEL method was already performed in previous
of Φ = 30◦ and 60◦ (indicated in Fig. 1a). For each test, five-time investigation [26]. It was recommended to use square elements to avoid
measurements were repeated to guarantee the accuracy of the results. any influence of the results due to the orientation of the elements. Also,
it showed that when decreasing the element size from 10 μm to 5 μm,
3. Finite element modelling fewer differences of chip morphology, temperatures and forces is noted.
10 μm square elements could be used to achieve fast results computing
3.1. FE modelling of workpiece and cutting tool with a lower accuracy but still satisfactory. Regarding the residual stress
distribution, it was found in our previous study [12] that the CEL model
To fully explore the residual stress evolution along the fillet radius, with 10 μm square elements is accurate enough to simulate the residual
four cutting faces are selected to perform the simulations, that is Outer stress profile. Therefore, in this study, the smallest element size of 8 μm
face, Deg30 face, Deg60 face, and End face respectively (Fig. 2). In each × 8 μm × 8 μm is used in Euler-part in the area near the machined
simulation, the workpiece is created with the shape of previous surface and chip to describe the residual stress gradient and chip
machined surface profile with the help of the “Merge/Cut Instances” morphology precisely. A coarse mesh with element size of 15 μm × 15
feature in Abaqus. The reason for selecting Deg30 face and Deg60 face μm × 15 μm are utilized in the rest of the Euler-part to improve
on the fillet radius is due to their special position, making it excep­ computing efficiency. The total number of elements are more than 2.5e6
tionally convenient to calculate the tool position in these two cutting for the Euler-part including the initial shape of the workpiece and the
faces (point C and D) and in the previous cuts (point C′ and D′ ) based on area where the chip will form. A computational time of two weeks is
the available angles and dimensions. needed for one cutting face simulation on a computer equipped with 8
These four 3D CEL Finite Element models were established in cores. Further refining the mesh is unacceptable for this cutting simu­
lation due to the huge computational source consumption. The simula­
tion was carried out in two steps. In the first step, the cutting process was
performed and in the second step the cutting tool was removed from the
workpiece to relieve the mechanical load. The machined workpiece was
not cooled down in the CEL simulation for two reasons. First, the
cooling-down step is extremely time consuming when it is carried out in
explicit solver ABAQUS/Explicit [27]. To address this issue, Nasr et al.
[28] developed an efficient model where an explicit solver ABAQUS/
Explicit was used for the cutting process, and an implicit solver ABA­
QUS/Standard for the unloading and cooling-down steps. This method
was proved to cut down the computational time from the order of days
(obtained with explicit solver) to just few minutes. However, Coupled
Eulerian-Lagrangian method was adopted in this study and thus the
cooling-down step can be performed only in Dynamic, Explicit steps
[29]. In our cases, it will cost long computational time (the order of
months) to obtain residual stress profiles after cooling down the
Fig. 1. Experimental setup: (a) cutting experiments and force measurements; machined workpiece with such a large amount of elements (2.5e6) and
(b) edge geometry.

218
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

Fig. 2. Selected cutting faces and the determination of tool positions for numerical simulation.

Fig. 3. The simulation settings when turning the outer face: (a) configuration and boundary conditions of the CEL model; (b) the mesh for the cutting tool; (c) the
mesh and dimension for the Euler-part.

small element length (8 μm). Second, the significance of thermal stress shown in Fig. 10a that the maximum temperature at the machined
in residual stress generation depends largely on the magnitude of tem­ surface is approximately 360 ◦ C in end face turning and can be lower in
perature in machined workpiece. The material employed in this paper is outer face turning. Matsumoto et al. [30] has concluded that the tem­
AISI 304 that usually does not cause extreme high temperature. It is perature increases in workpiece is so low when turning AISI 304 and the

219
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

effect of the temperature increase on the formation of residual stresses is

secondary. Lin et al. [31] also found that thermal load makes limited
difference of residual stress when the temperature is not very high.
Consequently, the cooling-down step was ignored in the CEL model due
to the excessive computing time and the limited influence of the tem­
perature gradient. Despite this, measures have been taken to minimize
the influence of temperature gradient as much as possible. For example,
the residual stress profiles are extracted in the location at some distance
(1.5 mm) away from the cutting edge, where the machined surface
temperature has been transferred from maximum value to around 88 ◦ C,
as shown in Fig. 4.

3.2. Material modelling Fig. 5. Illustration of stress-strain response of a metal specimen with damage
behaviour [29].
The Johnson-Cook model [32] is adopted in this research to model
the flow stress characteristics of AISI 304 stainless steel (curve ab in
Fig. 5). This model is capable of modelling the plastic material behav­ Table 3
AISI 304 material constants for J-C constitutive model [33].
iour at high strains, at high strain rates and at high temperatures. The set
of Johnson-Cook parameters are given in Table 3. This model is A (MPa) B (MPa) n C m Tmelt (◦ C) Troom (◦ C)
expressed as follows: 277 556 0.794 0.0096 0.944 1400 20
[ ( ) ][ ]m
ε̇ T − Troom
σ = [A + B(ε)n ] 1 + Cln 1− (1)
ε˙0 Tmelt − Troom
Table 4
Thermo-physical properties of AISI 304 steel and carbide cutting tool [16,34].
where σ is the material flow stress; ε is the plastic strain; ε̇ is the strain
rate; ε˙0 is the reference strain rate; T is the temperature of the material; Thermo-physical parameters Workpiece (AISI 301) Tool (Carbide)

Tmelt is the melting point of the material and Troom is the room temper­ 3
Density (kg/m ) 7930 14,860
ature. The material constants are as follows: A is the yield stress; B is the Thermal conductivity (W/m◦ C) 16.2 82
Thermal expansion (◦ C− 1) 16e-6 5.2
hardening modulus; C is the strain rate sensitivity coefficient; n is the
Young's modulus E (GPa) 193 600
hardening coefficient and m is the thermal softening coefficient. The Poisson's ratio v 0.28 0.2
physical properties of the workpiece and the cutting tool are given in Specific Heat (J/Kg◦ C) 500 249.8
Table 4.
The criterion for damage initiation of an element is met when the
scalar damage parameter ω exceeds 1, which is given by Eq.(2),

Fig. 4. The way extracting approximate residual stress distribution: (a) the selection of Path a; (b) temperature distribution along the Path a; (c) residual stress
distribution in the cutting speed direction (RS “SZZ”) along the Path a.

220
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

∑n ( )
Δε
ω= (2)
i=1 ε0 i

where Δε is the increment of equivalent plastic strain during an incre­

ment i; ε0 is the equivalent plastic strain at the onset of damage, which is
defined by the Johnson-Cook damage model [35], cf. Eq.(3),
[ ( ) ][ ( ) ][ ( )]
P ε̇ T − Troom
ε0 = D1 + D2 exp D3 1 + D4 ln 1 + D5 (3)
σ ˙
ε0 Tmelt − Troom

where D1-D5 are material constants; P/σ is the stress triaxiality where P
is the hydrostatic stress. The damage initiation is illustrated as point b in
Fig. 5.
After damage is initiated, the material stiffness is degraded pro­
gressively, thus the stress-strain behaviour of the damaged materials
cannot be accurately described by Eq. (1). Continuing to use the stress-
strain relation introduces a strong mesh dependency based on strain
Fig. 6. Determination of the equivalent plastic strain at the onset of damage
localization, such that the energy dissipated decreases as the mesh is (ε0 ) and the maximum stiffness degradation (Dmax).
refined. A different approach is required to follow the strain-softening
branch of the stress-strain response curve. Hillerborg's fracture energy
zone into two zones: sticking zone and slding zone. Sticking zone is
model [36] is used to reduce mesh dependency by creating a stress-
formed near the cutting edge along the tool-workpiece interface, where
displacement response after damage is initiated, cf. Eq. (4).
the frictional stress is considered as the shear stress of AISI 304 stainless
∫ ∫
steel. Coulomb friction model is adopted to capture the sliding area. The
Gf = ε0 εf Lσ dε = uf σ du (4)
0 model is defined as Eq. (7):
( )
where, L is the characteristic length associated with an integration point; τf = min μσn , τy (7)
ε0 is the equivalent plastic strain at damage initiation; εf is the equiva­
where τf is the frictional stress; τy is the yield shear stress of AISI 304
lent plastic strain when material is fully degraded; uf and σ are the
stainless steel; μ is the friction coefficient; and σ n is the normal stress
equivalent plastic displacement and flow stress.
between the tool and workpiece. It is known from the literature that the
In this study, Gf is provided as an input parameter and is calculated
friction coefficients between different workpiece materials and the
by Gf=K2c (1 − v2)/E where Kc is the fracture toughness determined by
matching cutting tools obtained by researchers vary with the cutting
[37].
conditions [41]. Among them, stainless steel have friction coefficient in
The degradation of the material upon damage is specified in expo­
the range of 0.2–0.8 [17,21,42,43] in the FE modelling of machining
nential form, which is given as:
process. In this study, the value of friction coefficient is select as 0.3
( ∫ )
u σ according to the previous study [43].
D = 1 − exp − du (5)
0 Gf

During the softening phenomenon (curve bc in Fig. 5), the stress is, 3.4. Thermal modelling
therefore, given by Eq. (6):
The heat generated in machining comes from the friction work
σ = (1 − D)̃
σ (6) occurring in the tool-workpiece interface and the plastic deformation
happening in the workpiece material. The fraction of the plastic work
where ̃ σ is the undamaged stress computed in the current increment. It
converted into heat is set to be 0.9, and that of the friction work is set to
represents stresses that would exist in the material without damage,
be 1.0 [44]. In the tool-workpiece interface, the weight factor of fric­
shown as curve bb’ in Fig. 5.
tional heat distribution between the tool and workpiece is set to be 0.5
In this study, an upper bound of damage variable, Dmax = 0.5, is
[27].
specified, which means that no further damage is accumulated at an
Heat conduction between cutting tool and workpiece is considered in
integration point once D reaches 0.5, and the plastic equivalent stress
the simulation and defined by gap conductance as a function of the
σ evolution starts to follow the curve cc′ .
pressure. The heat conduction coefficient is adopted from the literature
It is noticed that the values of ε0 and Dmax are determined according
[44]. Regarding the convection, this type of heat transfer is not
to the experimental results. A set of values were chosen and to perform
considered in the step of cutting process due to a very limited cutting
the test simulations until the discrepancy between the numerically and
time. Since the amount of heat radiation can be negligible compared to
experimentally obtained average chip thickness (T), average chip width
that of the thermal conduction, this type of heat transfer is not taken into
(W), and average value of the force component in the direction of cutting
consideration throughout the simulation.
speed (Fc) are below 20 %. The flow chart for determining these two
values is illustrated in Fig. 6. Eventually, ε0 and Dmax are set to 2.6 and
4. Results and discussion
0.5 respectively.

4.1. Cutting forces

3.3. Friction modelling
The evolutions of force components in x, y, and z directions (illus­
The frictional model used in this work is the Zorev's sticking-sliding trated in Fig. 2) are recorded by Kistler dynamometer during a fillet
friction model [38], which is the most commonly used frictional model radius turning. Fig. 7 shows the measured forces evolution during fillet
in metal cutting simulation to describe the frictional behaviour between turning process and indicates the value of Fx, Fy, and Fz which corre­
the tool and workpiece. It was demonstrated that the Zorev model is an sponds to the specified four cutting faces. After coordinate trans­
effective model for machining using the split-tool method [39] and formation, three cutting force components Ft, Fr, and Fc are obtained,
Partially Restricted Contact Length Tools [40]. It divides the frictional representing the cutting force components in the direction of tangential,

221
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

350 changed significantly during the process of fillet radius turning. The
Fx Fy Fz
length of the uncut chip cross section increases by around 64.3 % from
the outer face turning to end face turning. This may be the reason for the
175 increasing trend for the force component Fr during this process. Fig. 9b
presents the evolution of uncut chip cross-sectional area during fillet
Forces (N)

radius machining process. The area at each cutting position is divided

0 into several increments. For each increment, the length and width are
calculated based on geometric analysis that was already reported in our
previous study [9]. The total area is therefore obtained through inte­
-175 grating the area of each increment along the engaged edge. The uncut
chip area increases firstly and then decreases with the angle of the
cutting face position (Φ) increasing from 0◦ to 90◦ . This could be the
-350 reason for the evolution of Ft and Fc, both showing the similar changing
Outer Deg30 Deg60 End trend as the uncut chip cross-sectional area.
Cutting faces position

Fig. 7. Experimental cutting force components evolution during fillet 4.2. Temperatures
radius turning.
Cutting temperature is a critical parameter worthy of attention
radial, and cutting speed respectively. Fig. 8 shows the comparison of because the temperature has a significant influence on the magnitude of
the force components obtained from simulation and experiment at these the tensile stresses on the surface of the machined components. How­
four cutting faces. It can be seen in this figure that the proposed CEL ever, it is difficult to measure cutting temperature during turning pro­
model shows a good correlation with Fc, but it underestimates Ft and Fr. cess because of the complex geometry of the contact area and the
The discrepancy of the average value for the Fc are less than 20 %, while influence of chip. With the help of the FE simulation, this work can
the prediction of Ft and Fr are 35.5–62.5 % underestimated. A reason for provide the predicted temperature distribution during fillet radius
the underestimation of the cutting force components is likely attributed turning. Fig. 10a indicates the predicted temperature field in chip and
to the lower value selection of the friction coefficient. It is shown in [40] workpiece at each cutting position at the same cutting time of 0.5 ms.
that in a sliding region the local friction coefficient higher than 1 can be Fig. 10b shows the evolution of the temperature over time at point F
observed using Partially Restricted Contact Length Tools. Increasing the (indicated in Fig. 10a) in outer face turning. It is worth to mention that
friction coefficient can cause larger tool–chip contact area, such that the short simulation time (0.8 ms) was carried out for the four cutting
more work is required to form the chip and thus result in higher cutting position simulation due to the large amount of computational time.
forces [45]. In addition, the fact that the real tool edge microgeometry Therefore, the steady-state thermal distribution were most probably not
will deform rapidly in the experiment because of the creep behaviour of reached for the four models [47]. However, this doesn't prevent
tool material would contribute to the underestimation of the predicted comparing the temperatures between these four cutting positions
tangential and radial forces. It has been reported that [46] the initial because the same cutting time was selected for temperature comparison.
wear or deformation of the cutting tool (WC-10%Co) occurs in the first It can be observed from Fig. 10a that the high concentration of heat
seconds while machining AISI 304 stainless steel. Through the simula­ appears at the root of the chip and around the edge radius of the cutting
tion of the tool creep behaviour, it was proven that the tool geometries tool for all the cutting faces. In addition, the temperature is higher at the
variation is a significant source of feed force error in metal cutting location away from the machined surface due to the increasing depth of
simulations. This can also be proved by the investigation [16] that the cut where more plastic deformation occurs.
deviations of the feed force are significantly reduced with the increasing Fig. 10c shows the maximum temperature at the machined surface at
of flank wear and edge radius. each cutting position, which increases from 295 ◦ C to 361 ◦ C with the
Fig. 8 shows that in fillet radius turning of stainless steel the variation angle of the cutting face position (Φ) increasing from 0◦ to 90◦ . This
trend of the cutting force components obtained from simulation and could be explained by the phenomenon that the engaged edge rounding
experiment are similar. Along the tool path, both force components Ft (blue-shaded region in Fig. 9a) increases during this process. Edge
and Fc increase up to a maximum value and then decrease until the tool rounding acts as an effectively negative rake angle in machining [48],
exits from the fillet radius, while Fr shows an increasing trend during this which will lead to a more severely plastic deformation of the workpiece
process. The evolution of force components is significantly influenced by material. The increased area of edge rounding means that the volume of
the uncut chip cross-sectional geometry and area along the tool path. It materials compressed by this region becomes larger along the tool path,
can be seen in Fig. 9a that the uncut chip cross-sectional geometry so that more heat is generated as a result. In addition, the change of the
local normal rake angle (γ l) also contributes to the augmented

Experiment Simulation
300 300 300

200 200 200

Fc (N)

Fr (N)
Ft (N)

100 100 100

0 0 0
Outer Deg30 Deg60 End Outer Deg30 Deg60 End Outer Deg30 Deg60 End
Cutting face position Cutting face position Cutting face position

Fig. 8. Comparison between simulated and experimental forces components.

222
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

Fig. 9. (a) Geometry of uncut chip in outer face turning and end face turning; (b) evolution of uncut chip area with the angle of the cutting face position.

400 400
Maximum temperature (°C)
Temperature (°C)

300 300

200 200

100 100

0 0
0 0.2 0.4 0.6 0.8 Outer Deg 30 Deg 60 End
Cutting time (ms) Cutting face position
(b) (c)
Fig. 10. Predicted cutting temperature at the four cutting positions: (a) temperature fields in the workpiece; (b) the evolution of temperature over time at point F in
outer face turning; (c) the evolution of the maximum machined surface temperature during the fillet radius turning.

temperature. Fig. 11a shows the extraction of the local normal rake fillet radius turning process, although there is a slight recover in the final
angle that is located in Cutting plane A, and Fig. 11b presents the evo­ stage of this process. Therefore, in addition to the influence of the
lution of this angle during the fillet radius turning process. The overall engaged edge rounding, the decreased trend of local normal rake angle
trend of local normal rake angle tends to be more negative during the also results in the increased amount of heat generated along the tool

223
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

Fig. 11. Local normal rake angle at each cutting position: (a) method extracting local normal rake angle; (b) evolution of this angle.

path. thickness and width predictions show 3.9–10.8 % and 3.9–18.8 % pre­
diction error, which proves the correctness and accuracy of the proposed
numerical model.
4.3. Chip morphology

The chip thickness in fillet radius turning tests is non-uniform due to 4.4. Residual stresses
the variation of uncut chip thickness with the presence of the tool nose.
Accordingly, only the chip width and chip thickness are discussed in this Fig. 13 presents the experimental and numerical residual stress
section because it is difficult to acquire the section view of the chip. The profiles at different cutting faces during fillet radius turning. The
chip generated in outer face turning and end face turning were collected experimental results (Fig. 13a) show that there is not evident change
and measured with the help of optical microscope. Fig. 12 compares the with the residual stress profile when the cutting tool advances from the
chip morphology obtained from simulation and experiment, in which T outer face to end face. However, a slight decrease of maximum tensile
is the average chip thickness that was measured from the view of the residual stress and an increase of the magnitude and depth of
chip on the wider side, and W is the average chip width. The distribu­ compressive residual stress can still be found from outer face turning to
tions of equivalent plastic strain (PEEQ) are shown in predicted chip end face turning. The similar trend of residual stress distribution during
morphology. The way extracting the average width and average thick­ this process are also noticed in the simulated results (Fig. 13b). This is
ness of the predicted chip from the FE simulation model is also illus­ likely due to the variation of mechanical load acting on the workpiece
trated. It can be noticed from the experimental results that a significant during the fillet radius turning process. As already indicated in Fig. 8
smaller chip thickness is formed at the end face compared that at the that both the force components in the radial and cutting speed direction
outer face due to the shape variation of uncut chip region as shown in increase when the cutting face position (Φ) changes from 0◦ to 90◦ . The
Fig. 9a. At the same time, the width of chip increases because the workpiece material is therefore stretched more severely with a larger
engaged edge especially the straight cutting edge increases during this cutting force components at end face turning, leading to more
process. The simulation results follow the same trend of the chip compressive residual stress generated in the machined workpiece at this
morphology as the measured ones. In addition, the average chip cutting position. Fig. 13c compares the magnitude of residual stress at

Fig. 12. Comparison of the measured and predicted chip morphology.

224
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

1000 1000
Outer face

Residual stresses (MPa)

Residual stresses (MPa)

Outer face
800 800 Deg 30
End face
Deg 60
600 600 End face

400 400

200 200

0 0

-200 -200

0 50 100 150 200 250 0 50 100 150 200 250

Depth from machined surface (Pm) Depth from machined surface (Pm)
(a) (b)
1000
Residual stresses (MPa)

800

600
Exp.
Sim.
400

200
Outer face Deg 30 Deg 60 End face
Cutting face position
(c)
Fig. 13. Residual stresses in cutting speed direction: (a) experimentally obtained residual stress distributions along the depth direction; (b) numerically obtained
residual stress distributions along the depth; and (c) surface residual stress comparison between the experiment and simulation.

the surface of the four cutting positions which are obtained from introduced as a supplementary method to contribute to the under­
simulation and experiment. The error of the average surface residual standing of underlying mechanism of residual stress induced by fillet
stress between simulation and experiment is less than 20 %. radius turning because it can provide more information that is
There exist various error resources in numerically and experimen­ difficult to get from experiment, such as temperature distribution,
tally obtained residual stress distribution which would explain the residual stress distribution on the fillet face, local normal rake angle.
mismatch between them. In simulation, the sources of errors can be 2) Both the force components in the direction of cutting speed and
generalized as follows: not cooling down the machined workpiece tangential increase firstly and then decrease, while the force
because of the limited computing resources; the deviation of the friction component in radial direction increases during the fillet radius
coefficient selection; treating the workpiece material as pure homoge­ turning process.
neous when establishing the FE model, etc. In addition, simulations were 3) In fillet radius turning, cutting temperature in the machined work­
carried out with a constant tool edge, while in practice the tool edge may piece increases when the cutting tool advances from outer face to end
deform in less time than the first second of turning process, which could face. This is caused by the combination influence of the engaged edge
change the value of the cutting forces and thus the residual stresses. On rounding region and the local normal rake angle, which become
the other hand, it was reported that there are testing errors in the larger and more negative respectively during this process.
measurement of residual stress [49], which could be one of the error 4) In fillet radius turning, tensile residual stresses were generated
sources occurring in experiment. In addition, before carrying out the within the depth of 25 μm from the surface throughout the area of
experiment the workpiece was pre-machined as a fillet radius. The fillet radius. Magnitude and depth of residual stress are slightly
previous revolutions could produce remaining residual stresses with changed during the fillet radius turning process, but a reduction of
various magnitudes and distributions prior to the experiment [50], the residual stress profile can still be noticed.
which will more or less affect the residual stress distributions obtained
from the experiment. Declaration of competing interest

5. Conclusions The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
This study reveals the evolution of process variables induced by fillet the work reported in this paper.
radius turning of AISI 304 regarding the cutting forces, temperature,
chip morphology, and residual stress. A 3D CEL model was developed at Acknowledgement
the four specific cutting face positions and experimentally verified. The
following conclusions can be drawn: Authors would like to thank Seco Tools for providing the necessary
tools. One of the authors (Y. Liu) appreciates the financial support from
1) A good agreement between the simulated and experimental results China Scholarship Council (No. 201706130144).
has been achieved in terms of the evolution of cutting forces, chip
morphology and residual stress. Numerical investigation can be

225
Y. Liu et al. Journal of Manufacturing Processes 85 (2023) 216–226

References predictions. Mach Sci Technol 2011;15:21–46. https://doi.org/10.1080/

10910344.2011.557950.
[25] Yadav RK, Abhishek K, Mahapatra SS. A simulation approach for estimating flank
[1] Capello E. Residual stresses in turning: part I: influence of process parameters.
wear and material removal rate in turning of Inconel 718. Simul Model Pract
J Mater Process Technol 2005;160:221–8. https://doi.org/10.1016/j.
Theory 2015;52:1–14. https://doi.org/10.1016/j.simpat.2014.12.004.
jmatprotec.2004.06.012.
[26] Ducobu F, Rivière-Lorphèvre E, Filippi E. Mesh influence in orthogonal cutting
[2] Jawahir IS, Brinksmeier E, M’Saoubi R, Aspinwall DK, Outeiro JC, Meyer D, et al.
modelling with the Coupled Eulerian-Lagrangian (CEL) method. Eur J Mech A/
Surface integrity in material removal processes: recent advances. CIRP Ann - Manuf
Solids 2017;65:324–35. https://doi.org/10.1016/j.euromechsol.2017.05.007.
Technol 2011;60:603–26. https://doi.org/10.1016/j.cirp.2011.05.002.
[27] Muñoz-Sánchez A, Canteli JA, Cantero JL, Miguélez MH. Numerical analysis of the
[3] Ulutan D, Ozel T. Machining induced surface integrity in titanium and nickel
tool wear effect in the machining induced residual stresses. Simul Model Pract
alloys: a review. Int J Mach Tool Manuf 2011;51:250–80. https://doi.org/
Theory 2011;19:872–86. https://doi.org/10.1016/j.simpat.2010.11.011.
10.1016/j.ijmachtools.2010.11.003.
[28] Nasr MNA, Ng EG, Elbestawi MA. A modified time-efficient FE approach for
[4] Dumas M, Kermouche G, Valiorgue F, Van Robaeys A, Lefebvre F, Brosse A, et al.
predicting machining-induced residual stresses. Finite Elem Anal Des 2008;44:
Turning-induced surface integrity for a fillet radius in a 316L austenitic stainless
149–61. https://doi.org/10.1016/j.finel.2007.11.005.
steel. J Manuf Process 2021;68:222–30. https://doi.org/10.1016/j.
[29] Abaqus Analysis User’s Guide, Version 6.14 n.d.
jmapro.2021.05.031.
[30] Matsumoto Y, Da-Chun H. Workpiece temperature rise during the cutting of AISI
[5] Dumas M, Valiorgue F, Rech J, France F, Stief P, Dantan J, et al. Evolution of the
4340 steel. Wear 1987;116:309–17. https://doi.org/10.1016/0043-1648(87)
surface integrity while turning a fillet radius in a martensitic stainless steel
90179-7.
15–5PH. Procedia CIRP 2020;87:101–6. https://doi.org/10.1016/j.
[31] Zone-ching Lin BL. An investigation of the residual stress of a machined workpiece
procir.2020.02.038.
considering tool flank wear. J Mater Process Technol 1995;51:1–24.
[6] Weng J, Zhuang K, Xu D, Saoubi RM, Zhou J. A comprehensive study on cutting
[32] Johnson RG. A constitutive model and data for materials subjected to large strains,
mechanisms and surface integrity of AISI 304 when turning a curved surface. Mater
high strain rates, and high temperatures. In: Proc 7th Inf Sympo Ballist; 1983.
Manuf Process 2021;00:1–14. https://doi.org/10.1080/10426914.2021.1906893.
p. 541–7.
[7] Withers PJ, Bhadeshia HKDH. Residual stress. Part 1 – measurement techniques.
[33] Nie D, Lu Z, Zhang K. Hot bending behavior of SUS 304 stainless steel sheet assisted
MaterSci Technol 2013:0836. https://doi.org/10.1179/026708301101509980.
by resistance heating: multi-field coupling numerical simulation and experimental
[8] Li T, Shi T, Tang Z, Liao G, Han J, Duan J. Temperature monitoring of the tool-chip
investigation. Int J Adv Manuf Technol 2016:2763–74. https://doi.org/10.1007/
interface for PCBN tools using built-in thin-film thermocouples in turning of
s00170-016-8653-x.
titanium alloy. J Mater Process Technol 2020;275:116376. https://doi.org/
[34] Zhang W, Wang X, Hu Y, Wang S. Predictive modelling of microstructure changes,
10.1016/j.jmatprotec.2019.116376.
micro-hardness and residual stress in machining of 304 austenitic stainless steel.
[9] Weng J, Liu Y, Zhuang K, Xu D, M’Saoubi R, Hrechuk A, et al. An analytical method
Int J Mach Tool Manuf 2018;130–131:36–48. https://doi.org/10.1016/j.
for continuously predicting mechanics and residual stress in fillet surface turning.
ijmachtools.2018.03.008.
J Manuf Process 2021;68:1860–79. https://doi.org/10.1016/j.
[35] Johnson G, Cook W. Fracture characteristics of three metals subjected to various
jmapro.2021.07.004.
strains, strain rates and temperatures and pressures. Eng Fract Mech 1985;21(1):
[10] Sadeghifar M, Sedaghati R, Jomaa W, Songmene V. A comprehensive review of
31–48.
finite element modeling of orthogonal machining process: chip formation and
[36] Hillerborg A, Modéer M, Petersson P-E. Analysis of crack formation and crack
surface integrity predictions. Int J Adv Manuf Technol 2018;96:3747–91. https://
growth in concrete by means of fracture mechanics and finite elements. Cem Concr
doi.org/10.1007/s00170-018-1759-6.
Res 1976;6:773–81. https://doi.org/10.1016/0008-8846(76)90007-7.
[11] Calamaz M, Coupard D, Girot F. A new material model for 2D numerical simulation
[37] Farahmand B. Virtual testing and predictive modeling: for fatigue and fracture
of serrated chip formation when machining titanium alloy Ti–6Al–4V. Int J Mach
mechanics allowables. 2009. https://doi.org/10.1007/978-0-387-95924-5.
Tool Manuf 2008;48:275–88. https://doi.org/10.1016/j.ijmachtools.2007.10.014.
[38] Zorev NN. Inter-relationship between shear processes occurring along tool face and
[12] Liu Y, Agmell M, Xu D, Ahadi A, Stahl JE, Zhou J. Numerical contribution to
shear plane in metal cutting. Int Res Prod Eng 1963;49:143–52.
segmented chip effect on residual stress distribution in orthogonal cutting of
[39] Kato S, Yamaguchi K, Yamada M. Stress distribution at the interface between tool
Inconel718. Int J Adv Manuf Technol 2020:993–1005. https://doi.org/10.1007/
and chip in machining. J Manuf Sci Eng Trans ASME 1972;94:683–9. https://doi.
s00170-020-05702-2.
org/10.1115/1.3428229.
[13] Zhang YC, Mabrouki T, Nelias D, Gong YD. Chip formation in orthogonal cutting
[40] Ortiz-de-Zarate G, Madariaga A, Arrazola PJ, Childs THC. A novel methodology to
considering interface limiting shear stress and damage evolution based on fracture
characterize tool-chip contact in metal cutting using partially restricted contact
energy approach. Finite Elem Anal Des 2011;47:850–63. https://doi.org/10.1016/
length tools. CIRP Ann 2021;70:61–4. https://doi.org/10.1016/j.
j.finel.2011.02.016.
cirp.2021.03.002.
[14] Agmell M, Ahadi A, Gutnichenko O, Ståhl JE. The influence of tool micro-geometry
[41] Du M, Cheng Z, Wang S. Finite element modeling of friction at the tool-chip-
on stress distribution in turning operations of AISI 4140 by FE analysis. Int J Adv
workpiece interface in high speed machining of Ti6Al4V. Int J Mech Sci 2019:163.
Manuf Technol 2017;89:3109–22. https://doi.org/10.1007/s00170-016-9296-7.
https://doi.org/10.1016/j.ijmecsci.2019.105100.
[15] Liu Y, Xu D, Agmell M, Ahadi A, Stahl J. Investigation on residual stress evolution
[42] Maranhão C, Paulo Davim J. Finite element modelling of machining of AISI 316
in nickel-based alloy affected by multiple cutting operations. J Manuf Process
steel: numerical simulation and experimental validation. Simul Model Pract Theory
2021;68:818–33. https://doi.org/10.1016/j.jmapro.2021.06.015.
2010;18:139–56. https://doi.org/10.1016/j.simpat.2009.10.001.
[16] Liu Y, Xu D, Agmell M, Saoubi RM, Ahadi A. Numerical and experimental
[43] Laakso SVA, Agmell M, Ståhl J. The mystery of missing feed force — the effect of
investigation of tool geometry effect on residual stresses in orthogonal machining
friction models, flank wear and ploughing on feed force in metal cutting
of Inconel 718. Simul Model Pract Theory 2021;106:102187. https://doi.org/
simulations. J Manuf Process 2018;33:268–77. https://doi.org/10.1016/j.
10.1016/j.simpat.2020.102187.
jmapro.2018.05.024.
[17] Nasr MNA, Ng EG, Elbestawi MA. Modelling the effects of tool-edge radius on
[44] Agmell M, Ahadi A, Zhou JM, Peng RL, Bushlya V, Ståhl JE. Modeling subsurface
residual stresses when orthogonal cutting AISI 316L. Int J Mach Tool Manuf 2007;
deformation induced by machining of Inconel 718. Mach Sci Technol 2017;21:
47:401–11. https://doi.org/10.1016/j.ijmachtools.2006.03.004.
103–20. https://doi.org/10.1080/10910344.2016.1260432.
[18] Jafarian F, Imaz Ciaran M, Umbrello D, Arrazola PJ, Filice L, Amirabadi H. Finite
[45] Sartkulvanich P, Altan T, Göcmen A. Effects of flow stress and friction models in
element simulation of machining Inconel 718 alloy including microstructure
finite element simulation of orthogonal cutting – a sensitivity analysis. Mach Sci
changes. Int J Mech Sci 2014;88:110–21. https://doi.org/10.1016/j.
Technol 2007:0344. https://doi.org/10.1081/MST-200051211.
ijmecsci.2014.08.007.
[46] Laakso SVA, Zhao T, Agmell M, Hrechuk A, Ståhl JE. Too sharp for its own good –
[19] Umbrello D, Filice L. Improving surface integrity in orthogonal machining of
tool edge deformation mechanisms in the initial stages of metal cutting. Procedia
hardened AISI 52100 steel by modeling white and dark layers formation. CIRP Ann
Manuf 2017;11:449–56. https://doi.org/10.1016/j.promfg.2017.07.135.
- Manuf Technol 2009;58:73–6. https://doi.org/10.1016/j.cirp.2009.03.106.
[47] Courbon C, Mabrouki T, Rech J, Mazuyer D, Perrard F, D’Eramo E. Further insight
[20] Özel T. Computational modelling of 3D turning: influence of edge micro-geometry
into the chip formation of ferritic-pearlitic steels: microstructural evolutions and
on forces, stresses, friction and tool wear in PcBN tooling. J Mater Process Technol
associated thermo-mechanical loadings. Int J Mach Tool Manuf 2014;77:34–46.
2009;209:5167–77. https://doi.org/10.1016/j.jmatprotec.2009.03.002.
https://doi.org/10.1016/j.ijmachtools.2013.10.010.
[21] Agmell M, Bushlya V, Laakso SVA, Ahadi A. Development of a simulation model to
[48] Liu Y, Hrechuk A, Agmell M, Ahadi A, Stahl JE, Zhou J. FE analysis on the
study tool loads in pcBN when machining AISI 316L. Int J Adv Manuf Technol
association between tool edge radius and thermal-mechanical load in machining
2018:2853–65. https://doi.org/10.1007/s00170-018-1673-y.
Inconel 718. Procedia CIRP 2021;102:91–6. https://doi.org/10.1016/j.
[22] Valiorgue F, Rech J, Hamdi H, Gilles P, Bergheau JM. 3D modeling of residual
procir.2021.09.016.
stresses induced in finish turning of an AISI304L stainless steel. Int J Mach Tool
[49] Jafarian F, Amirabadi H, Sadri J. Experimental measurement and optimization of
Manuf 2012;53:77–90. https://doi.org/10.1016/j.ijmachtools.2011.09.011.
tensile residual stress in turning process of Inconel718 superalloy. Meas J Int Meas
[23] Arrazola PJ, Kortabarria A, Madariaga A, Esnaola JA, Fernandez E, Cappellini C,
Confed 2015;63:1–10. https://doi.org/10.1016/j.measurement.2014.11.021.
et al. On the machining induced residual stresses in IN718 nickel-based alloy:
[50] Dumas M, Valiorgue F, Van Robaeys A, Rech J. Interaction between a roughing and
experiments and predictions with finite element simulation. Simul Model Pract
a finishing operation on the final surface integrity in turning. Procedia CIRP 2018;
Theory 2014;41:87–103. https://doi.org/10.1016/j.simpat.2013.11.009.
71:396–400. https://doi.org/10.1016/j.procir.2018.05.050. Elsevier B.V.
[24] Ozel T, Llanos I, Soriano J, Arrazola PJ. 3D finite element modelling of chip
formation process for machining inconel 718: comparison of FE software

226