3D FEA Simulations in Machining

SpringerBriefs in Applied Sciences and Technology
Manufacturing and Surface Engineering

Panagiotis Kyratsis · Anastasios Tzotzis ·
J. Paulo Davim
3D FEA Simulations in
Machining
Series Editor
Joao Paulo Davim , Department of Mechanical Engineering, University of
Aveiro, Aveiro, Portugal
This series fosters information exchange and discussion on all aspects of manu-
facturing and surface engineering for modern industry. This series focuses on
manufacturing with emphasis in machining and forming technologies, including
traditional machining (turning, milling, drilling, etc.), non-traditional machining
(EDM, USM, LAM, etc.), abrasive machining, hard part machining, high speed
machining, high efficiency machining, micromachining, internet-based machining,
metal casting, joining, powder metallurgy, extrusion, forging, rolling, drawing, sheet
metal forming, microforming, hydroforming, thermoforming, incremental forming,
plastics/composites processing, ceramic processing, hybrid processes (thermal,
plasma, chemical and electrical energy assisted methods), etc. The manufacturability
of all materials will be considered, including metals, polymers, ceramics, compos-
ites, biomaterials, nanomaterials, etc. The series covers the full range of surface
engineering aspects such as surface metrology, surface integrity, contact mechanics,
friction and wear, lubrication and lubricants, coatings an surface treatments, multi-
scale tribology including biomedical systems and manufacturing processes. More-
over, the series covers the computational methods and optimization techniques
applied in manufacturing and surface engineering. Contributions to this book series
are welcome on all subjects of manufacturing and surface engineering. Especially
welcome are books that pioneer new research directions, raise new questions and
new possibilities, or examine old problems from a new angle. To submit a proposal
or request further information, please contact Dr. Mayra Castro, Publishing Editor
Applied Sciences, via [email protected] or Professor J. Paulo Davim,
Book Series Editor, via [email protected]
Panagiotis Kyratsis · Anastasios Tzotzis ·
J. Paulo Davim
3D FEA Simulations
in Machining
Panagiotis Kyratsis Anastasios Tzotzis
Product and Systems Design Engineering Product and Systems Design Engineering
University of Western Macedonia University of Western Macedonia
Kila Kozani, Greece Kila Kozani, Greece
J. Paulo Davim
Mechanical Engineering
University of Aveiro
Aveiro, Portugal
ISSN 2191-530X ISSN 2191-5318 (electronic)

ISSN 2365-8223 ISSN 2365-8231 (electronic)
ISBN 978-3-031-24037-9 ISBN 978-3-031-24038-6 (eBook)
https://doi.org/10.1007/978-3-031-24038-6
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023

This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
1 A Comparative Study Between 2D and 3D Finite Element

Methods in Machining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Comparison Between 2D and 3D FEM Analysis Techniques . . . . . . . 3
1.2.1 Evaluation of Cutting Forces and Torque . . . . . . . . . . . . . . . . . 3
1.2.2 Chip Formation and Dimensions Assessment . . . . . . . . . . . . . 4
1.2.3 Examination of Residual Stresses . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.4 Temperature and Tool-Wear Assessment . . . . . . . . . . . . . . . . . . 6
1.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Fundamentals of 3D Finite Element Modeling in Conventional
Machining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.1 FEM-Based Studies in Machining . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.2 Advantages and Limitations of FEM . . . . . . . . . . . . . . . . . . . . . 17
2.2 Finite Element Modeling in Machining . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.1 Meshing and Element Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.2 Mesh Adaptivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.3 Tool-Workpiece Representation . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2.4 Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.5 Material Flow Stress Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2.6 Friction Modeling and Contact Description . . . . . . . . . . . . . . . 30
2.2.7 Material Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.2.8 Tool Wear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.3 Typical FEM-Based Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3.1 Cuttings Forces, Torque, and Residual Stresses . . . . . . . . . . . . 37
2.3.2 Chip Morphology, Temperature Distribution, and Wear . . . . . 38
2.4 Conclusions and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
v
vi Contents
3 FEM-Based Study of AISI52100 Steel Machining: A Combined

2D and 3D Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.2.1 Machining Process Framework . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.2.2 Preliminary FE Model Assessment . . . . . . . . . . . . . . . . . . . . . . 50
3.2.3 Numerical Modeling of the Turning Process in Three
Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.3 Results and Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.3.1 Machining Forces Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.3.2 Chip Geometry Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4 Experimental and 3D Numerical Study of AA7075-T6 Drilling
Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.2.1 Layout of Experimental Testing . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2.2 Finite Element Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.3 Results and Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3.1 Cutting Forces and Torque Analysis . . . . . . . . . . . . . . . . . . . . . 70
4.3.2 Chip Morphology Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.3.3 Temperature Distribution Analysis . . . . . . . . . . . . . . . . . . . . . . . 72
4.3.4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5 3D Finite Element Simulation of CK45 Steel Face-Milling: Chip
Morphology and Tool Wear Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.1 Experimental Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.2.2 Face-Milling CAD-Based Setup . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.2.3 Numerical Modeling of the Face-Milling Process . . . . . . . . . . 81
5.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3.1 Chip Formation Analysis and Temperature Distribution
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3.2 Tool Wear Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Chapter 1
A Comparative Study Between 2D
and 3D Finite Element Methods
in Machining
Abstract The Finite Element Method (FEM) is arguably one of the most valu-
able tools when studying manufacturing processes. Therefore, it is widely used for
the development of models that are able to predict the behavior of a manufacturing
process under a variety of conditions with acceptable accuracy. Especially as compu-
tational resources develop, so does the FEM, making it an integral part of modern
studies. Even though FEM can nowadays be implemented in 3D, 2D modeling still
holds an important role. In the present work, a comparison between the application
of 2D and 3D FEMs in machining was made, with both the advantages and disad-
vantages in mind. Specifically, an effort was made to capture the effectiveness of
each method when studying standard machining results such as the cutting forces
and torque, the temperatures, the residual stresses, and the tool wear.
Keywords 2D FEM · 3D FEM · Machining · Cutting forces · Residual stresses ·

Tool wear · Cutting temperature
1.1 Introduction
The Finite Element Method (FEM) is a well-known tool, implemented often in

machining studies. The fact that it can reduce the experimental work and that it allows
the investigation of variables that cannot be examined in other manners, makes it an
integral part of modern machining studies. Both 2D and 3D FEM studies display a
number of advantages and disadvantages. Even though at the moment it seems that
the one supplements the other, it is considered that the 3D FEM will replace the 2D
due to the restrictions that exist in the two dimensions.
Cepero Mejias et al. [1] used a 2D FEM technique to study the orthogonal
machining of thick unidirectional laminates. To obtain realistic results, the authors
implemented a novel linear stiffness degradation coupled with a continuum damage
mechanic model. Similar strategies, especially for the damage propagation [2] were
applied to investigate the machining of unidirectional composites, focusing on the
influence of the rake and relief angle of the tool, as well as the cutter edge radius.
Zhou et al. [3] simulated the machining of alloy ZL109 in two dimensions with
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

P. Kyratsis et al., 3D FEA Simulations in Machining,
Manufacturing and Surface Engineering,
https://doi.org/10.1007/978-3-031-24038-6_1
2 1 A Comparative Study Between 2D and 3D Finite Element Methods …
a polycrystalline diamond (PCD) tool to acquire the optimal finishing parameters

and to study the effects of the cutting parameters on the residual stresses, cutting
forces, and temperature. The authors verified the simulation results with experi-
mental work. Peng et al. [4] developed a 2D FEM chip formation model for the cutting
forces prediction during machining of aged Inconel 718, by using a Coupled Eule-
rian–Lagrangian (CEL) method. Carbide-cutting tools of different cobalt percentages
have been utilized, to generate varying tool wear. Mishra et al. [5] studied the turning
process with uncoated-textured and coated-textured tools on Ti6Al4V by using 2D
Finite Element (FE) simulations. The authors made a comparison with equivalent
plain tools and evaluated the performance of the tool with regard to the cutting forces,
friction coefficient, contact length, chip behavior, and shear stress. Tagiuri et al. [6]
established a 2D numerical model for the prediction and study of the effects of tool
nose morphology combined with a number of cutting parameters when machining
AISI1045 steel. The performance parameters studied, were the cutting tempera-
ture and developed forces, the effective stress, as well as the tool wear. Numerous
studies such as [7–13] still use 2D FEM for the examination of several characteristics
that are involved in machining. Typical materials studied are AISI1045, AISI630,
and AISI4140 steels, Inconel 718, titanium alloys such as Ti6Al4V and Ti-5553,
composites, as well as aluminium alloys. The aforementioned studies focus mostly
on the investigation of the generated cutting forces and temperatures, the developed
stresses, the chip morphology, and tool wear under dry conditions with both coated
and uncoated tools. In some cases, cryogenic conditions were implemented, and the
microstructure of the material was examined. The Arbitrary Lagrangian–Eulerian
(ALE) and the CEL were the two common methods for the development of the FE
models, with modifications related to the friction and the flow stress models.
3D FEM studies can be utilized for the investigation of similar parameters and
characteristics of machining. Liu et al. [14] implemented 3D FEM in their study
related to 17-4PH stainless steel machining. The developed 3D model enabled the
investigation of the cutting performance of various micro-grooved tools. Studies that
involve textured tools benefit greatly from the use of 3D FEM, since the complexity of
the tool models cannot be easily approximated otherwise. Similar studies [15, 16] for
the machining of AISI1045 steel and Ti6Al4V alloy, use FEM in three dimensions to
examine the machinability under certain conditions with micro-grooved and textured
tools. The increasing number of 3D studies in the past few years [17–23] proves the
necessity of the 3D FEM in machining. Tzotzis et al. [24] utilized 3D FEM to predict
the developed machining forces during AISI4140 turning with ceramic tools. The
authors developed a 3D FE model implementing tool CAD models of high precision,
for the study of the tool nose influence on the generated cutting forces. Xu et al. [25]
utilized 3D FE techniques for the modeling of titanium composites drilling, in order
to provide an understanding of the interface damage progression.
In this work, a number of key points related to the techniques used in FEM-based
studies of machining were extracted from recent advances in the field. Therefore, an
effort was made to compare different approaches utilized in 2D and 3D FEM-based
researches, focusing mainly on the modeling of the machining forces and torque, the
1.2 Comparison Between 2D and 3D FEM Analysis Techniques 3
chip formation, the residual stresses, the generated temperatures, and the progression
of tool wear.
1.2 Comparison Between 2D and 3D FEM Analysis

Techniques
Either 2D or 3D, the studies that implement FEM focus on the investigation of
aspects that affect machining processes in order to analyze their mechanisms. A
high number of variables involved in machining can be examined via FEM, but the
number of dimensions used enables different approaches each time.
1.2.1 Evaluation of Cutting Forces and Torque
Evaluation of the forces generated during machining is a topic widely discussed

among researchers, since it can provide valuable insight into the tool wear and
performance, as well as the machinability of a material. Most studies related to this
topic include an examination of the cutting conditions’ influence on the produced
machining forces, which is possible to be carried out with either 2D or 3D FE models.
The difference is that a 2D model can approximate only the cutting force F c and the
thrust force F p , whereas a 3D can simulate the feed force F f as well. Despite this
difference, the 2D model cannot be considered obsolete by any means, since feed
force is not that important when compared to both the cutting and the thrust force.
In fact, the measured feed force values will always be lower compared to the other
two force components. Thus, it does not contribute much to the generated resultant
cutting force. However, when it is required to evaluate feed force as well, the 3D
model is the only option. Finally, Fig. 1.1a illustrates an example force analysis on
the plane, whereas Fig. 1.1b depicts the equivalent analysis in three dimensions.
Because 2D simulations run much faster compared to 3D ones, many researchers
[26–28] utilized this advantage to perform preliminary studies in two dimensions.
Such studies are used for evaluation purposes of several aspects involved in the
development of the FE model, such as the friction coefficients, the flow stress, and
the damage progression constants.
Similarly, cutting torque can be determined by both 2D and 3D FE models.
However, a 3D model can deliver more accurate results at the cost of long running
times. Torque can provide valuable information on the required machine power and
allow the engineers to assess the ongoing machining process. A number of studies
focused on the study of drilling, utilized 3D FE modeling in order to output combined
results related to the cutting forces and torque with the chip formation [29, 30] or
temperature distribution with the chip formation [31]. In contrast, Matsumura and
Fig. 1.1 Machining force

analysis in 2D (a) and 3D (b)
Tamura [32] established a hybrid simulation model, which implemented 2D simu-

lations for cutting force prediction. The simulations focused on the energy analysis,
where the three-dimensional chip flow was modeled on the plane, including both the
cutting and the chip flow directions.
Concluding, the selection of the appropriate method for the assessment of the
cutting forces and torque depends to a great extent on the computational resources
and the influence of the assumptions made during a 2D analysis.
1.2.2 Chip Formation and Dimensions Assessment
Regarding the understanding of the chip formation mechanisms, the utilization of 3D

FE modeling clearly provides an edge over the equivalent 2D method. In such studies,
the implementation of the third dimension is critical, since it provides a more accurate
approximation of both the chip formation and morphology. The chip flow in two
dimensions can be predicted with reasonable accuracy. A number of 2D studies [6,
10, 11, 33] focused on the examination of the strain distribution and reported findings
on the chip flow of acceptable reliability, approximating many types of chip such
as continuous, discontinuous, with built-up-edge and serrated. However, according
to Thepsonthi and Özel [34], the 2D chip tends to be accumulated in front of the
tool’s edge, due to the lack of the third dimension. Hence, the 2D chip flow is based
on assumptions that negatively affect the accuracy of the results. Especially when
studying more complex processes such as drilling and milling, the implementation of
the third dimension is imperative as shown in several studies [34–37]. Chip breakage
[16] is a topic that cannot be easily examined without the implementation of the third
dimension. A 3D study can reveal both the breakage points, as well as the critical
dimensions and morphology at these points.
Figure 1.2a illustrates an example of chip formation at the initiation of the cutting
process in 3D turning, whereas Fig. 2b illustrates the equivalent process in 2D.
Moreover, Fig. 1.2c depicts the chip formation during 3D milling. It is evident that
the first two images present some similarities, the fact that proves the possibility to
approximate the chip flow to some extent. However, Fig. 1.2c points out that more
complex machining processes produce equally complex chips, making their study
less efficient in two dimensions. In addition, 3D studies enable the examination of
burr formation [38, 39], as well as allow the consideration of the tool run-out [26]
and other machine errors, which cannot be taken into account in 2D. Furthermore,
it is possible to study the performance of the cutting tool, regardless of its shape
and geometrical characteristics [14, 40]. Finally, it is highlighted that the 3D FE
modeling allows the investigation of more aspects related to the chip flow, compared
to the 2D one.
1.2.3 Examination of Residual Stresses
To access the residual stresses induced during the machining of materials, many
researchers prefer a combination of 2D and 3D FE modeling. It is common to utilize
2D simulations for the identification of the loading zones, whereas 3D simulations
are used for the fundamental modeling of the residual stresses. This technique leads
to reduced simulation times, without compromising the reliability of the results.
Rami et al. [27] managed to replace the tool-chip interface with the generated ther-
momechanical loadings. By using 2D FEM, the authors identified the depth where
the residual stresses state appears and next, implemented 3D FEM in their study
to obtain more realistic results by applying the loadings on a 3D model. Addition-
ally, Salvati and Korsunsky [41] analyzed the microscale residual stresses in Electric
Discharge Machining (EDM) with the aid of thermomechanical FEM solely in a
planar domain. It must be noted, however, that the identification of the stress field can
be realized completely in 3D without the need for 2D simulations [42, 43]. According
Fig. 1.2 Example chip flow in 3D turning (a), 2D turning (b), and 3D milling (c)
to Attanasio et al. [44], the use of the 3D FEM model enables the study of complex
tool geometries, as well as the identification of the optimal cutting parameters that
can yield the lowest possible residual stress values. Valiorgue et al. [42] simulated
the residual stress without the chip removal process, by applying the equal to the
thermomechanical loadings onto the cut surface in three dimensions, simulating a
multiple-pass cutting process.
1.2.4 Temperature and Tool-Wear Assessment
Cutting temperatures are directly related to the generated tool wear, making the
prediction of these variables critical. Even though both methods are able to yield
results for the temperatures and tool wear, there are cases where 3D is a one-way
solution, whereas, in other cases, a combination between the two can be more effec-
tive. Similar to the variables mentioned in the previous sections, 3D modeling is the
preferred method over 2D for temperatures generated during more complex processes
such as drilling and milling. In addition, when tools with special geometries are
involved, the 2D method simply cannot yield acceptable results, since too many
assumptions and simplifications must be made. According to Ebrahimi et al. [12],
the length where contact exists between the primary cutting edge and the workpiece
is considered to be the same as in a 2D analysis. Therefore, the authors combined

2D and 3D FEMs for the prediction of the chip morphology and the cutting forces
respectively. On the other hand, some studies [4, 6] preferred to use solely the 2D
method to model tool wear progression with respect to typical tool characteristics
such as rake angle and nose radius. However, the majority of the researchers, resort
to the 3D method for the development of FE models that involve complex processes
and tool geometries, in addition to more realistic approaches.
Malakizadi et al. [45] investigated the flank tool wear and the developed temper-
atures during turning in three dimensions, considering several cutting conditions, by
utilizing dense 3D mesh distributions on the tool’s areas of interest. Majeed et al.
[31] utilized 3D FE modeling to assess the temperature allocation on the chip and
the test piece during drilling, with an aim to evaluate tool life. Similarly, Lotfi et al.
[46] examined the connection between the built-up heat and the tool flank wear in
drilling with the 3D method. Magalhães et al. [47] approximated the temperature
distribution and the tool rake face wear, on the rounded edges of the tool, discretized
with multiple chamfers. Moreover, Lotfi et al. [48] utilized 3D modeling for the
prediction of the temperature interface and tool wear rates, among other variables,
during turning with both coated and uncoated tools. Especially for the case of the
coated tools, the models included the full geometrical aspects that can be captured
only in three dimensions.
Figure 1.3 illustrates sample temperature results for 3D drilling (Fig. 1.3a), 2D
turning (Fig. 1.3b), and 3D turning as well (Fig. 1.3c). In specific, Fig. 1.3a depicts the
temperature distribution along the cutting edges of the twist drill in three dimensions.
It is noted that the temperature changes with respect to the depth of penetration, since
the contact volume is analogous to the depth. Moreover, because a 2D model cannot
simulate the contact volume of the tool, it is not possible to approximate the equiv-
alent developed temperatures. Figure 1.3a, b compare the temperature distribution
across the surface of the generated chip in two and three dimensions accordingly, for
identical cutting conditions and at similar cutting step. Despite the similarities that
can be found between the two methods, regarding the temperature values, as well
as the distribution zones, the 2D method cannot highlight the volumetric boundaries
of each one of these zones, in contrast to the 3D one. The same limitations and
restrictions can be observed when investigating tool wear as well.
Fig. 1.3 Example temperature distribution along a twist drill’s edges in 3D (a), on the 2D chip
during turning (b), and on the 3D chip, respectively (c)
1.3 Conclusions
This study introduced a number of points that derive from the comparison of the
approaches made with 2D and 3D FEM during standard machining processes. The
variables that are affected by the application of the FEM techniques, and are discussed
in the present chapter, are the cutting forces, torque, temperature distribution on the
cutting edges, as well as on the chip surface, the tool wear development, the chip
flow and the generated stresses. By collecting and comparing information found
in the literature, the following conclusions can be drawn. Additionally, Table 1.1
summarizes some key points related to the benefits and the limitations that emerge
when using FEM in these types of studies, that were identified during the literature
review.
• The 2D FEM constitutes an effective tool for preliminary simulations and
sensitivity studies due to the simple setup and short run times.
• In contrast, the 3D FEM is a more resource-intensive method compared to the
2D one, the fact that is responsible for the long simulation times. In spite of
this disadvantage, the 3D FEM provides more realistic results when compared
to equivalent 2D analyses, and in some cases is the only way to generate certain
output.
1.3 Conclusions 9
Table 1.1 Summary of the advantages and disadvantages of 2D and 3D FEM

2D FEM 3D FEM
(+) • Model setup is simple • Utilizes the full tool geometry, thus,
• Run times are short enabling the use of complex tools
• Is effective for trial-and-error testing • Provides realistic chip formation, which
• Provides accurate approximation of most includes several phenomena that occur
types of chip morphology during machining such as chip breakage,
• Can be used to identify the residual tool run-out, and burr formation
stress’s depth • Enables the study of every chip dimension
• Can output the generated temperatures, • Enables the prediction of all force
flank, and rake tool wear with acceptable components
accuracy • Allows the full simulation of the residual
stresses
• Can output volumetric tool wear
• Enables the study of the temperature
distribution zones in three dimensions
(−) • Utilizes simplified tool geometry, thus, • Depends heavily on the computational
only the rake angle, the clearance angle, resources
and the nose radius participate in the • Run times can become very long
process • Model setup is more complicated
• The study takes place in the plane, compared to 2D
neglecting several important • Some output variables are difficult to be
characteristics of the process measured
• Prediction of the forces is not as accurate
as in 3D
• The prediction of the cutting forces and torque is more accurate in 3D due to the
modeling of the full tool geometry and contact region.
• Modeling of the chip morphology can be determined accurately with both
methods. However, the 3D chip flow can yield results that include parameters
which cannot be easily implemented in a 2D analysis, such as the burr formation,
chip breakage, and the complete dimensions of the chip.
• The residual stresses, in terms of appearance depth and domain, can be identi-
fied by both methods with success. To acquire a deeper understanding of this
phenomenon, however, the three-dimensional analysis seems a more efficient
approach.
• Finally, prediction of the temperatures that develop on the tool-chip interface
can be carried out reliably by both methods. The same applies for the tool wear,
with respect to both flank and rake regions. Nevertheless, the implementation of
the third dimension, enables the use of the complete geometry of a tool, thus,
allowing the examination of the full contact domain between the cutting tool and
the material.
References
1. Cepero Mejías F, Curiel Sosa JL, Phadnis VA (2018) Machining 2D FEM study of long fibre
reinforced unidirectional laminates using a linear continuum damage technique. 6th Conf
Comput Mech
2. Cepero-Mejías F, Curiel-Sosa JL, Zhang C, Phadnis VA (2019) Effect of cutter geometry
on machining induced damage in orthogonal cutting of UD polymer composites: FE study.
Compos Struct 214:439–450. https://doi.org/10.1016/j.compstruct.2019.02.012
3. Zhou Y, Sun H, Li A et al (2019) FEM simulation-based cutting parameters optimization in
machining aluminum-silicon piston alloy ZL109 with PCD tool. J Mech Sci Technol 33:3457–
3465. https://doi.org/10.1007/s12206-019-0640-3
4. Peng B, Bergs T, Klocke F, Döbbeler B (2019) An advanced FE-modeling approach to improve
the prediction in machining difficult-to-cut material. Int J Adv Manuf Technol 103:2183–2196.
https://doi.org/10.1007/s00170-019-03456-0
5. Mishra SK, Ghosh S, Aravindan S (2019) Performance of laser processed carbide tools for
machining of Ti6Al4V alloys: a combined study on experimental and finite element analysis.
Precis Eng 56:370–385. https://doi.org/10.1016/j.precisioneng.2019.01.006
6. Tagiuri ZAM, Dao TM, Samuel AM, Songmene V (2022) A numerical model for predicting
the effect of tool nose radius on machining process performance during orthogonal cutting of
AISI 1045 steel. Materials (Basel) 15. https://doi.org/10.3390/ma15093369
7. Parida AK, Maity K (2019) Analysis of some critical aspects in hot machining of Ti-5553
superalloy: experimental and FE analysis. Def Technol 15:344–352. https://doi.org/10.1016/j.
dt.2018.10.005
8. González G, Segebade E, Zanger F, Schulze V (2019) FEM-based comparison of models
to predict dynamic recrystallization during orthogonal cutting of AISI 4140. Procedia CIRP
82:154–159. https://doi.org/10.1016/j.procir.2019.04.061
9. Chandra Mouli G, Prakash Marimuthu K, Jagadeesha T (2020) 2D finite element analysis
of inconel 718 under turning processes. In: IOP conference series: materials science and
engineering, pp 1–11
10. Yaich M, Ayed Y, Bouaziz Z, Germain G (2020) A 2D finite element analysis of the effect of
numerical parameters on the reliability of Ti6Al4V machining modeling. Mach Sci Technol
24:509–543. https://doi.org/10.1080/10910344.2019.1698606
11. Xu X, Outeiro J, Zhang J et al (2021) Machining simulation of Ti6Al4V using coupled Eulerian-
Lagrangian approach and a constitutive model considering the state of stress. Simul Model Pract
Theory 110. https://doi.org/10.1016/j.simpat.2021.102312
12. Ebrahimi SM, Hadad M, Araee A, Ebrahimi SH (2021) Influence of machining conditions on
tool wear and surface characteristics in hot turning of AISI630 steel. Int J Adv Manuf Technol
114:3515–3535. https://doi.org/10.1007/s00170-021-07106-2
13. Prakash Marimuthu K, Kumar CSC, Prasada HPT (2018) 2D finite element thermo-mechanical
model to predict machining induced residual stresses using ALE approach. Mater Today Proc
5:11780–11786. https://doi.org/10.1016/j.matpr.2018.02.147
14. Liu G, Huang C, Su R et al (2019) 3D FEM simulation of the turning process of stainless steel
17–4PH with differently texturized cutting tools. Int J Mech Sci 155:417–429. https://doi.org/
10.1016/j.ijmecsci.2019.03.016
15. Mishra SK, Ghosh S, Aravindan S (2018) 3D finite element investigations on textured tools with
different geometrical shapes for dry machining of titanium alloys. Int J Mech Sci 141:424–449.
https://doi.org/10.1016/j.ijmecsci.2018.04.011
16. Wu L, Liu J, Ren Y et al (2020) 3D FEM simulation of chip breakage in turning AISI1045
with complicate-grooved insert. Int J Adv Manuf Technol 108:1331–1341
17. Mishra DK, Verma AK, Arab J et al (2019) Numerical and experimental investigations
into microchannel formation in glass substrate using electrochemical discharge machining.
J Micromech Microeng 29:1–13
18. Wan M, Ye X-Y, Wen D-Y, Zhang WH (2019) Modeling of machining-induced residual stresses.
J Mater Sci 54:1–35. https://doi.org/10.1007/s10853-018-2808-0
References 11
19. Lotfi M, Amini S, Akbari J (2020) Surface integrity and microstructure changes in 3D elliptical
ultrasonic assisted turning of Ti–6Al–4V : FEM and experimental examination. Tribol Int
151:106492. https://doi.org/10.1016/j.triboint.2020.106492
20. Park CI, Wei Y, Hassani M et al (2019) Low power direct laser-assisted machining of carbon
fibre-reinforced polymer. Manuf Lett 22:19–24. https://doi.org/10.1016/j.mfglet.2019.10.001
21. Tzotzis A, Tapoglou N, Verma RK, Kyratsis P (2022) 3D-FEM approach of AISI-52100 hard
turning: modelling of cutting forces and cutting condition optimization. Machines 10:74. https://
doi.org/10.3390/machines10020074
22. Tzotzis A, Efkolidis N, Oancea G, Kyratsis P (2021) FEM-based comparative study of square &
rhombic insert machining performance during turning of AISI-D3 steel. Int J Mod Manuf
Technol 13:143–148. https://doi.org/10.54684/ijmmt.2021.13.2.143
23. He YL, Davim JP, Xue HQ (2018) 3D progressive damage based macro—mechanical FE
simulation of machining unidirectional FRP composite. Chinese J Mech Eng. https://doi.org/
10.1186/s10033-018-0250-5
24. Tzotzis A, García-Hernández C, Huertas-Talón J-L, Kyratsis P (2020) Influence of the nose
radius on the machining forces induced during AISI-4140 hard turning: a CAD-based and 3D
FEM approach. Micromachines 11:798. https://doi.org/10.3390/mi11090798
25. Xu J, Lin T, Li L et al (2022) Numerical study of interface damage formation mechanisms in
machining CFRP/Ti6Al4V stacks under different cutting sequence strategies. Compos Struct
285:115236. https://doi.org/10.1016/j.compstruct.2022.115236
26. Davoudinejad A, Tosello G, Parenti P, Annoni M (2017) 3D finite element simulation of micro
end-milling by considering the effect of tool run-out. Micromachines 8:1–20. https://doi.org/
10.3390/mi8060187
27. Rami A, Kallel A, Sghaier S et al (2017) Residual stresses computation induced by turning of
AISI 4140 steel using 3D simulation based on a mixed approach. Int J Adv Manuf Technol
28. Pittalà GM, Monno M (2010) 3D finite element modeling of face milling of continuous chip
material. Int J Adv Manuf Technol 47:543–555. https://doi.org/10.1007/s00170-009-2235-0
29. Tzotzis A, Markopoulos AP, Karkalos NE et al (2021) FEM based investigation on thrust
force and torque during Al7075-T6 drilling. In: IOP conference series: materials science and
engineering, p 012009
30. Dou T, Fu H, Li Z et al (2019) Prediction model, simulation, and experimental validation on
thrust force and torque in drilling SiCp/Al6063. Int J Adv Manuf Technol 103:165–175. https://
doi.org/10.1007/s00170-019-03366-1
31. Majeed A, Iqbal A, Lv J (2018) Enhancement of tool life in drilling of hardened AISI 4340
steel using 3D FEM modeling. Int J Adv Manuf Technol 95:1875–1889. https://doi.org/10.
1007/s00170-017-1235-8
32. Matsumura T, Tamura S (2015) Cutting simulation of titanium alloy drilling with energy
analysis and FEM. Procedia CIRP 31:252–257. https://doi.org/10.1016/j.procir.2015.03.045
33. Davim JP, Maranhão C (2009) Study on plastic strain and plastic strain rate in machining of
steel AISI 1020 using FEM analysis. Mater Des 30:160–165. https://doi.org/10.1016/j.matdes.
2008.04.029
34. Thepsonthi T, Özel T (2015) 3-D finite element process simulation of micro-end milling Ti-
6Al-4V titanium alloy: experimental validations on chip flow and tool wear. J Mater Process
Tech 221:128–145. https://doi.org/10.1016/j.jmatprotec.2015.02.019
35. Tzotzis A, García-Hernández C, Huertas-Talón J-L, Kyratsis P (2020) FEM based mathematical
modelling of thrust force during drilling of Al7075-T6. Mech Ind 21:415. https://doi.org/10.
1051/meca/2020046
36. Tzotzis A, García-Hernández C, Huertas-Talón J-L, Kyratsis P (2020) 3D FE modelling of
machining forces during AISI 4140 hard turning. Strojniški Vestn—J Mech Eng 66:467–478.
https://doi.org/10.5545/sv-jme.2020.6784
37. Lotfi M, Amini S, Aghaei M (2018) 3D FEM simulation of tool wear in ultrasonic assisted
rotary turning. Ultrasonics 88:106–114. https://doi.org/10.1016/j.ultras.2018.03.013
38. Davoudinejad A, Tosello G, Annoni M (2017) Influence of the worn tool affected by built-
up edge (BUE) on micro end-milling process performance: a 3D finite element modeling
investigation. Int J Precis Eng Manuf 18:1321–1332. https://doi.org/10.1007/s12541-017-
0157-6
39. Guo YB, Dornfeld DA (2000) Finite element modeling of burr formation process in drilling
304 stainless steel. J Manuf Sci Eng Trans ASME 122:612–619. https://doi.org/10.1115/1.128
5885
40. Kyratsis P, Tzotzis A, Markopoulos A, Tapoglou N (2021) CAD-based 3D-FE modelling
of AISI-D3 turning with ceramic tooling. Machines 9:4. https://doi.org/10.3390/machines9
010004
41. Salvati E, Korsunsky AM (2020) Micro-scale measurement & FEM modelling of residual
stresses in AA6082—T6 Al alloy generated by wire EDM cutting. J Mater Process Tech
275:1–12. https://doi.org/10.1016/j.jmatprotec.2019.116373
42. Valiorgue F, Rech J, Hamdi H et al (2012) 3D modeling of residual stresses induced in finish
turning of an AISI304L stainless steel. Int J Mach Tools Manuf 53:77–90. https://doi.org/10.
1016/j.ijmachtools.2011.09.011
43. Özel T, Ulutan D (2012) Prediction of machining induced residual stresses in turning of titanium
and nickel based alloys with experiments and finite element simulations. CIRP Ann Manuf
Technol 61:547–550. https://doi.org/10.1016/j.cirp.2012.03.100
44. Attanasio A, Ceretti E, Giardini C (2009) 3D FE modelling of superficial residual stresses
in turning operations. Mach Sci Technol 13:317–337. https://doi.org/10.1080/109103409032
37806
45. Malakizadi A, Gruber H, Sadik I, Nyborg L (2016) An FEM-based approach for tool wear
estimation in machining. Wear 368–369:10–24. https://doi.org/10.1016/j.wear.2016.08.007
46. Lotfi M, Amini S, Al-Awady IY (2018) 3D numerical analysis of drilling process: heat, wear,
and built-up edge. Adv Manuf 6:204–214. https://doi.org/10.1007/s40436-018-0223-z
47. Magalhães FC, Ventura CEH, Abrão AM, Denkena B (2020) Experimental and numerical
analysis of hard turning with multi-chamfered cutting edges. J Manuf Process 49:126–134.
https://doi.org/10.1016/j.jmapro.2019.11.025
48. Lotfi M, Jahanbakhsh M, Farid AA (2016) Wear estimation of ceramic and coated carbide tools
in turning of Inconel 625: 3D FE analysis. Tribol Int 99:107–116. https://doi.org/10.1016/j.tri
boint.2016.03.008
Chapter 2
Fundamentals of 3D Finite Element
Modeling in Conventional Machining
Abstract The present chapter introduces the fundamentals of Finite Element

Method (FEM) applied in standard machining processes such as turning, drilling,
and milling. The most broadly used models for material flow, friction, and material
separation and evolution are presented. In addition, a short introduction is included
to meshing, element types, contact interface, and boundary conditions. Finally, the
most recent modifications and advances in the various models implemented in FEM
are presented, as well as the state of the art, especially in three dimensions.
Keywords 3D FEM · 3D simulation · Machining · Turning · Drilling · Milling ·

Cutting forces · Chip geometry · Cutting temperature
2.1 Introduction
Computers are widely used in engineering to simulate optimal prototype design

solutions that meet established performance criteria. Various computing tools are
often used in setting, evaluating, and testing the design goals and specifications of a
product or system. It is expected that in the near future, computer-based modeling
methods and tools will simply require a description of the expected behavior or
structure of a design idea, in order to perform the simulation with an aim to evaluate
the idea in a fully automated manner.
Computer-aided engineering (CAE) is a technology that utilizes computer systems
to analyze the geometry of a system, allowing the engineer to simulate and study
the behavior, as well as the performance of a model, in order to optimize certain
parameters. CAE systems are available for a wide range of analyses, such as stress,
thermal, and vibration. Several CAE software, such as ANSYS™, ABAQUS™, and
ALTAIR™ can calculate the resulting stresses of a complex geometry, by determining
loads using fundamental static equations and numerical methods. They can also
be used to analyze heat transfer problems, fluid flow, electromagnetism, and other
multidimensional problems.
Many physical problems involve complex geometries, different materials,
complex boundaries, and initial conditions. To solve complex problems, engineers

14 2 Fundamentals of 3D Finite Element Modeling in Conventional Machining
must resort to numerical methods, especially when analytical methods cannot be

of assistance. In general, any practical problem can be idealized as a mechanical
problem, as shown in Fig. 2.1. Especially engineering problems are models of phys-
ical states, expressed mathematically. A great number of models contain differential
equations, which are derived using fundamental laws of physics (e.g., Newton’s
laws) and principles, such as the conservation of energy and mass, and momentum,
with a set of limits and initial conditions. For simple problems, it is possible to find a
specific solution in analytical form, where the solution is given via multiple variables
and parameters, which are already available in Static Engineering, Strength of Mate-
rials, Fluid Engineering, and other fields of engineering. However, for problems
of increased complexity, numerical methods are more appropriate. These include
the Finite Difference Method (FDM), where derivatives are replaced by differential
equations, the Finite Element Method (FEM), and the Boundary Element Method
(BEM), which uses integrated formulations to produce a system of algebraic equa-
tions. With FEM, the whole analysis domain or geometry is discretized. This chapter
deal with the implementation of FEM in machining processes, a method that meets
an increased degree of acceptance in the past few years.
Fig. 2.1 Solution flow chart of an engineering problem

2.1 Introduction 15
2.1.1 FEM-Based Studies in Machining
Despite the fact that FEM-based analyses usually take a long time to complete and
require a great deal of computer resources, it is a preferred method to simulate
complex processes such as most machining operations. With the advent of more
advanced software and the technological evolution of computers, it is possible
to simulate most conventional and non-conventional processes in a relatively
straightforward way.
The development of more specialized software, oriented towards machining and
forming, such as DEFORM™, FORGE™, and AdvantEdge™, enabled the investi-
gation of the machining processes in both two and three dimensions. Earlier, Ceretti
et al. [1] modified the code of DEFORM™-2D to study the orthogonal cutting process
of steel. To achieve reasonable accuracy, several changes were made to the mate-
rial separation process model. Similarly, Klocke et al. [2] studied the turning of
AISI-1045 steel with the aid of 2D FEM, at high speeds. The authors indicated
that despite the fact that several assumptions were made regarding the model, the
results were in accordance with their experimental testing. Maranhão and Davim [3]
modeled the behavior of stainless steel during machining, to investigate the influ-
ence of friction on standard parameters that are involved. Moreover, Xie et al. [4]
implemented commercially available software to predict the wear of turning tools.
By using FEM, the authors managed to calculate the chip formation and heat transfer,
which were then used to determine the wear on the tool, according to the applied
cutting conditions.
Besides typical industrial materials such as steel and aluminium, FEM-based
study of machining in two dimensions can be extended to more advanced materials
such as titanium [5, 6] and nickel alloys [7, 8]. These studies focus on the surface
quality and integrity of the machined material, residual stresses, thermal behavior,
chip morphology, and more.
Furthermore, a number of 3D studies in machining exist in literature, dealing
with both conventional and non-conventional processes. An early example is the
study on hard turning by Guo and Liu [9]. The authors established a 3D FE model
to study the machinability of hardened steel with ultra-high strength tools. Karpat
and Özel [10] investigated turning with 3D simulations to predict generated forces
and temperatures with regard to the microgeometry of the tools. Moreover, a set
of experiments was conducted for comparison. Buchkremer et al. [11] proposed
a calculation method, which details how the chip characteristics at the breakage
point and the thermomechanical properties on the uncut surface are related. The
necessary parameters were obtained by implementing experimentally generated chip
geometries into the methodology. Similar studies in turning [12–14], investigate the
generated cutting forces, chip geometry, temperature distribution, tool wear, and
more.
FEM can be applied to an equally important machining process such as drilling.
Gao et al. [15] proposed a three-dimensional FEM model of drilling with a twist drill.
In the study, the scraps forming process, the cutting forces, and the cutting tempera-
tures are analyzed. Majeed et al. [16] studied the effect of the developed temperatures
on tool life, during drilling steel 4140. The authors determined the temperatures at
the drill bit and compared them with the equivalent experimental ones. Oezkaya
et al. [17] investigated flow drilling with the aid of FEM. Their paper presented a
three-dimensional FE model for predicting the performance of modified tools during
flow drilling of AlSi10Mg. The study includes the performance of both preheated
and non-preheated tools in comparison with experimental results. Additional studies
demonstrate the implementation of FEM when modeling the drilling operation and
its usefulness when exploring several performance parameters [18–20].
Similar to turning and drilling, milling is one of the most used processes in
industry. Hence, several studies deal with the investigation of milling based on FEM.
Maurel-Pantel et al. [21] described the simulation process of 304L stainless steel
shoulder milling with FEM. A number of experiments were carried out for valida-
tion purposes, which delivered the generated milling forces. Davoudinejad et al. [22]
presented a 3D FEM approach for the micro-milling process of Al6082-T6. The
built-up edge (BUE) and its effects on the process performance were examined in
terms of the chip flow, burr formation, and cutting forces. Moreover, experiments
were conducted with ultra-high precision for comparison purposes. Gao et al. [23]
developed a 3D Coupled Eulerian–Lagrangian (CEL) model for the simulation of
aluminium end-milling processes. The chip flow was predicted with the proposed
model, and all the stages of the formation were compared with experimental data.
As seen in similar studies [24–27], the examination parameters that draw the atten-
tion of the researchers are the milling forces, the temperature distribution, the chip
geometry, the tool wear, the residual stresses, the machining errors, and more.
It must be noted that FEM has applicability to non-conventional processes also,
especially as computer technology advances and computational time is decreased.
Electrical Discharge Machining (EDM) is one of the standard non-conventional
machining methods that can benefit from FEM. A number of studies indicate that it
is possible to study several aspects of the process with significant accuracy, meaning
that FEM is a reliable method for non-conventional machining processes as well.
Such studies [28–31] usually investigate the behavior of hard-to-machine materials
such as metallic glasses via EDM and ultrasonic-assisted machining, as well as the
performance of the electrodes, the machinability of various materials, and the effects
of the machining parameters on the machined surface.
The present chapter introduces the implementation of FEM in standard machining
processes and summarizes most of the available models and techniques for meshing,
boundary conditions, material flow stress, friction, and material separation. Addi-
tionally, an effort is made to include indicative studies and the latest trends in the
field.
2.2 Finite Element Modeling in Machining 17
2.1.2 Advantages and Limitations of FEM
It is not easy to determine whether 2D or 3D modeling is a better method, since both

have their advantages and disadvantages. Most of the time it depends on the type of
study. Thus, in some cases, 2D may prove to be an ideal solution, whereas, in other
cases, 3D is the only way to study a phenomenon.
Regarding the 2D analysis, it enables the use of analytical rigid parts that reduce
calculation time and can be used to simplify the mechanical problem, making it ideal
for preliminary simulations. However, it cannot be used to simulate more complicated
phenomena and geometries. On the other hand, a 3D analysis produces results closer
to the real conditions, as it implements the full geometry of a cutting tool. 3D analyses
are ideal for complicated simulations but are sensitive to computer resources.
In general, FE modeling provides the user with several advantages:
• The full machining process can be visualized.
• Many machining aspects and variables can be thoroughly examined.
• Experimental testing can be reduced.
• Safety procedures can be eliminated.
• Costs related to production, as well as production times, can be minimized.
• A great deal of manufacturing failures and errors can be predicted.
• Finally, a number of machining results can be accurately predicted.
As someone would expect, this method is followed by some limitations also:
• Idealization of real-life objects cannot be exact for very complex shapes.
• The method yields approximate solutions.
• The computation is very costly. For instance, as the mesh becomes finer, the
requirement for computer resources grows. Thus, there is a computational limit
based on the current computer resources.
• Time needed for solving the problems, increases with the degree of fineness of
the mesh, hindering the prospect of solving a model in full.
2.2 Finite Element Modeling in Machining
2.2.1 Meshing and Element Types
During the discretization process of a model, the volume or the surface is divided into
a discrete number of smaller elements, which are called finite elements. Elements
can be one-dimensional if they are simple lines in the plane, two-dimensional if they
are shapes in the plane (e.g., a triangle), usually called shell elements, and three-
dimensional if they are solids in space (e.g., tetrahedron, hexahedron, and pyramid).
During the solution process, the equations that govern the behavior of a system are
applied to each of the individual elements of a model. One-dimensional elements are
used to discretize lengthy parts with a uniform cross section, such as shafts, beams,
and others. Shell elements (two-dimensional) are usually used for models that have
a small and uniform thickness (e.g., sheet-metal parts), while the solid elements
(three-dimensional) are used in complex parts, which have a complex geometry
(e.g., cast objects). Finally, 2D and 3D elements can be first-order (linear) or second-
order (parabolic), with the second-order elements having a greater number of nodes
than the first-order, making them higher quality elements. Figure 2.2 illustrates the
different types of elements.
As shown in Fig. 2.2, nodes are the vertices of the forming elements of a grid.
Also, they are the points of an element with which it contacts and connects with
neighboring elements. Their number depends on the type of element, so for the first
order, a triangular shell element will have three nodes, while a hexahedral solid
element will have eight nodes.
Correspondingly for the second order, the triangle will have 6, whereas the hexa-
hedron 20. Figure 2.3 depicts typical examples of first- and second-order elements,
respectively.
Modern computer simulation systems enable the selection of the desirable element
type and the edit of the mesh parameters. In addition, it is possible to use the embedded
to the software mesh generator, which automatically selects the appropriate type of
elements according to the problem. For example, once the user selects to process a
Fig. 2.2 Standard types of finite elements

Fig. 2.3 Number of nodes based on the element’s order
problem in two dimensions with symmetrical geometry, the software suggests the
use of the optimal type of elements, which in this case is the plane element.
One of the most important mesh parameters is density, which refers to the size
of the elements that will be generated within a boundary. Thus, the total number of
elements in an area can be determined by their size. Most Finite Element Analysis
(FEA) systems usually allow the user to either set the number of elements upfront
or define the minimum/maximum size of an element. With the latter option, the
mesh size is calculated according to either the minimum or the maximum size of
an element. Secondary parameters that affect the mesh density are the sample grid
resolution and the critical point tolerances. These parameters are considered to have
lesser effect on the density, compared to the element size (Fig. 2.4).
Fig. 2.4 Typical mesh size parameters

Fig. 2.5 Sample meshing

for a typical machining
process
Besides element size, the minimum element number in a circle or arc and the size
ratio between neighboring elements are equally important mesh parameters. It is
noted that not all elements within an area have the same size. Moreover, it is possible
to set various element sizes for different regions within an area. This technique is
usually implemented when a local refinement is desired, such as at the machined
surface of the workpiece during machining (Fig. 2.5). The local mesh can be defined
with the specified size ratio. The DEFORM™-3D software suggests a size ratio of
7:1 [14, 32, 33], irrespective of the machining operation. Some studies on turning [3,
13, 34] point out that 4:1 to 5:1 is an acceptable ratio, whereas a number of works
in drilling [16, 35] applied a 10:1 ratio. More complex parts that are used during
milling, were discretized with ratios in the range of 10:1–20:1 [36, 37]. In any case,
the minimum element size (MES) is set to values in the order of fractions of an mm
and is based on the experience and expertise acquired with FEA simulations. It is
noted that especially for the DEFORM™-3D system, as a rule of thumb, a sufficient
MES is usually equal to 25% of the feed for turning, or 50% of the feed per tooth
for drilling and milling [38–41].
Early examples of 3D FEM-based studies in machining [9, 42], where the local
mesh refinement was utilized, prove that a higher mesh density is usually applied at
the area of interest, which is the contact area of the tool-work pair. This technique
allows for increased precision and resolution of the geometry that participates in
the machining operation. In addition, the field variables such as strain, temperature,
and material damage become more accurate. However, this benefit comes with a
drawback; The computational time that is required to solve the problem increases
significantly, following a geometric progression pattern. Therefore, it is desirable for
regions, where large deformations occur, to have a considerable number of small-
sized elements. Contrarily, a coarser mesh (fewer small elements) is preferable in
regions where the gradients of the state variables are negligible. The latter can greatly
improve the conservation of computational resources. Finally, it is pointed out that
a mesh with a density below an acceptable range may cause problems. More specif-
ically, a low-quality mesh around corner edges or in regions with localized surface
Fig. 2.6 Example tool model meshing
effects such as high damage, will most likely interfere with the remeshing process
and cause mesh degradation.
Regarding the tool model meshing, a similar practice is usually followed. Hence,
the area around the tooltip or cutting edge that is in close proximity with the unma-
chined plane of the workpiece, is discretized with a denser mesh compared to the
rest surface of the tool. A number of studies [13, 16, 34, 43] indicate that a 4:1
size ratio is adequate for the solution process, regardless of the machining operation
and software used. In most cases, the tool is modeled to be rigid for simplification
purposes of the under-study problem. As such, the nodes of the tool model cannot be
deformed, contributing this way to the optimized simulation times. In the event that
deformation effects must be investigated on the tool, it can be modeled as plastic.
Either way, the complexity of the tool geometry can significantly affect the gener-
ated results, in addition to the completion time. It is pointed out that a rigid tool
still should be meshed with this technique, in order to serve the thermal analysis.
Figure 2.6 illustrates an example meshing with a 4:1 ratio for a turning insert and a
twist drill model as well.
2.2.2 Mesh Adaptivity
Remeshing is the most common method to prevent extreme element distortion during
modeling of machining. There are three types of mesh adaptivity: h-adaptivity, p-
adaptivity, and r-adaptivity. The first one is considered a general-use method and
it enables the change of the element size, allowing this way the alteration of the
mesh density and the node connectivity. The second type is less common since
it is not effective in metal machining modeling. It increases the precision of the
solution by changing the order of the interpolation polynomials. Finally, the third
type is implemented in the Arbitrary Lagrangian–Eulerian (ALE) formulation and
is capable of relocating the nodes. Therefore, the shape of the mesh elements is
optimized without affecting the connectivity.
Since large plastic deformations occur during machining, it is important that the
conditions of the model’s initial mesh are continuously checked. In the event that
mesh becomes unusable, the remeshing process is triggered, replacing the distorted
mesh with a new undistorted one. Moreover, this process interpolates the field vari-
ables such as strain, temperature, node velocity, and damage from the unusable mesh
to the new one. Most FEA software systems embed an algorithm that performs a
check if certain criteria are met during the simulation process and handle both the
remeshing and variable interpolation procedures at the same time. One standard
criterion is the interference depth between the tool and the machined part, where the
penetration depth of the primary into the slave object is monitored. A much higher
value of penetration depth than the one needed will result in distorted elements,
whereas a much lower value will increase the number of the remeshing processes,
negatively affecting the run time. Another criterion for triggering the remeshing is
the occurrence of the negative Jacobian matrix, which renders the mesh unusable. It
is noteworthy that the remeshing technique is applied to the newly cut surface and the
generated chip as well, to improve the convergence of the solution [44] and is based
on the strain and strain rate values [24]. Modern versions of various FEA systems
deal with the mesh distortion problem with local remeshing, rather than with a full
replacement of the mesh.
Lastly, there are three mathematical formulations that are used in continuum
FEM: the Lagrangian, the Eulerian, and the ALE. Practically, the primary models
are considered to be two, since the ALE is a combination of the Lagrangian and
Eulerian. The updated Lagrangian formulation [16, 35, 44] is regarded as a major
advancement in meshing techniques, allowing for improved computational times
[45]. Additionally, the ALE formulation [17, 34, 46] enables the nodal adjustment,
which allows the element nodes to adapt independently, preventing the occurrence
of large deformations and maintaining the mesh topology. Nowadays, the Eule-
rian formulation is not the preferred choice of researchers in machining, because it
requires the chip geometry to be predetermined and does not allow chip separation.
In spite of the aforementioned disadvantages, this method employs a fixed space
mesh, which completely disables the severe distortion problem [47].
Element deletion
As damage in the material initiates, it will evolve and, eventually, the material will
fracture. The criteria implemented in FEM that are used to model the chip separation
and damage evolution are discussed in Sect. 2.7. Modern FEA systems include an
option, as part of the remeshing procedure, that enables the deletion of extremely
distorted elements to maintain the integrity of the mesh. Elements that exceed a
defined critical damage value are removed from the mesh, making this method an
effective way of modeling crack propagation. Due to the fact that element deletion
may generate sharp edges at the area of interest, it is possible to select the degree of
element deletion. Finally, coupled with local remeshing, compose an effective tool for
the generation of a realistic chip. Figure 2.7 illustrates an example of chip generation
Fig. 2.7 Material separation and chip generation
and evolution during modeling of machining with DEFORM™-3D ver.12, where a

remeshing technique is locally used, deletion for elements of specific size is enabled
and Cockcroft-Latham criterion is implemented with the equivalent to the material
damage constant.
2.2.3 Tool-Workpiece Representation
Accurate representation of the tool-work interface is of great importance in FEM-

based studies in machining. In most cases, the workpiece is a simplified part, which is
easy to be designed in a CAD software. The tools, on the other hand, are usually more
complex and require certain techniques to be accurately modeled. One source for
tool models is the manufacturer’s website that usually contains several tool models.
Unfortunately, most vendors do not upload accurate representations of their tools,
but rather simplified ones. Even though a tool model with less information would
contribute towards the improvement of the simulation time, the generated results
would not be as reliable as it should. Even small misinterpretations in the geometry
may alter significantly the results [48]. Therefore, for more accurate representations
of complex tools, such as twist drills, specialized drills, and end-mills, an effective
method is 3D scanning [21, 35, 44]. This method requires a 3D scanner and the
equivalent piece of software that is able to convert the scanning process into a cloud
of points and finally into a usable file format such as STL. Nonetheless, the method
requires hardware and resources that are not always available. As a final resort, it
is possible to design the tool in a CAD system, if all the critical geometric aspects
of the tool are available. Moreover, it is feasible to implement programming to
automatically generate parametrical tool models [49, 50], increasing this way the
accuracy and improving the calculation process.
Figure 2.8 compares the geometry and the design intent between a turning insert
model from the vendor and the equivalent parametric model designed in a CAD
system. In specific, it is shown that the origin point of the downloaded model is
not ideally positioned, making the insertion of the model into the FEA system more
complicated. Furthermore, it does not include a honing feature on the cutting edge,
reducing this way the efficiency of the simulation and the reliability of the generated
Fig. 2.8 Model comparison with regard to the FEM aspects of a sample CNGA120408 tool
results [49]. Finally, lower tessellation in general, let alone of a critical geometrical
aspect of the tool, tends to reduce the simulation precision. In addition, the size of the
STL file for each of the two models is of the same order. Concluding, the generated
CAD model contains extra geometrical features and thus implements a great deal of
insert variations.
2.2.4 Boundary Conditions
The conditions that express the borders between the tool and the workpiece are
usually applied in such a way so that the tool is able to advance towards the fixed
workpiece or the tool is constrained from translating along the feed path and at the
same time is allowed to rotate (for processes such as milling and drilling). In any case,
the movement of the tool and the workpiece is related to each other and equal to the
cutting speed. The standard parameters that are set during the boundary conditions
setup are the cutting speed, the feed, the cutting direction, the depth and the width
of cut where applicable, as well as the initial contact zone between the tool and the
work, and the heat exchange.
Figure 2.9 illustrates the set of the basic boundary conditions for three typical
machining procedures, drilling with a twist drill (Fig. 2.9a), turning with an insert
(Fig. 2.9b), and face milling with an insert (Fig. 2.9c).
By observing Fig. 2.9a it is shown that the twist drill translates along the Z axis
(feed) towards the workpiece while rotating around the Z axis (spindle speed) at the
same time, similar to the real procedure. Moreover, workpiece is fixed so that all
nodes at the bottom and on the periphery of the cylinder are fixed, allowing only
the distortion of the nodes inside the pre-made hole. Figure 2.9b depicts the cutting
path that the turning insert follows in order to perform the cut. Since the radius of
the work is drawn on the Z axis, the cutting trajectory will be on the YZ plane. In
this case, the cutting speed and feed are converted to tool simultaneous movement
along the Y and Z axes. The work is fixed on the side and along the bottom so that
Fig. 2.9 Boundary conditions for three typical machining processes; drilling (a), turning (b), and
face milling (c)
the velocities of the nodes on both the X and Z axes are equal to zero. In contrast
to the real conditions, the workpiece is constrained, whereas the tool is the object
that moves. At last, Fig. 2.9c shows the cutting path of the milling insert on the XY
plane and the fixation of the work in a similar manner to the turning workpiece.
In this process, the kinematics of the simulation are identical to the real ones, with
the exception that only one tooth participates in the cut for simplification purposes.
As long as the steady state is achieved, it is possible to design the workpiece as an
orthogonal part, further simplifying the process. This applies to both turning and
face milling or similar processes. However, the addition of the circularity to the part
makes the setup more realistic and allows the full study of the procedure.
Analysis domain
A focused analysis area is often used [24, 32, 38, 51] to simplify the study of the
cutting process. This way the phenomena that occur during machining can be studied
to an extent without sacrificing vast amounts of resources and time. CAD systems are
an ideal tool to extract the necessary geometrical aspects of the tool-work framework,
position the tool and examine the kinematics of the system [43].
Figure 2.10 presents an example analysis domain for turning extracted with the
aid of a commercially available CAD software. In Fig. 2.10a, the domain is an arc
that belongs to the diameter of the work. In this instance, the tool uses a constant
speed (U x ), usually expressed in mm/s, to replicate feed by moving linearly towards
the uncut surface and a constant angular velocity (ωx ) for the revolution around the
X axis. By projecting the arc-shaped part on a 2D plane, the center and the radius of
the generated sector will be matched to the ones of the workpiece. Hence, the angular
velocity in rad/s can be determined as ω = 2π f , with the frequency f given in rps.
Moreover, the simulation time can be calculated by simply dividing the angle of the
Fig. 2.10 Analysis domain simplification for turning of arc-shaped (a) and rectangular work (b)
arc, plus any extra travel by the angular velocity. For example, for the machining of an
arc-shaped part of 45° belonging to a workpiece with 100 mm diameter, at a constant
cutting speed equal to 100 m/min, the angular velocity is calculated approximately
33.33 rad/s, and the simulation time is close to 24 ms.
The second case (Fig. 2.10b) is simpler since no revolution is set. Here, the tool
may perform two translations with constant speed, one towards the −X axis (feed)
and one towards the +Y axis (cutting direction) with a speed equal to U y .
Heat exchange
Thermal modeling is an equally critical approximation to all other modeling proce-
dures implemented in numerical simulations. Equation 2.1 calculates the conductive
heat transfer Qt between the contact surfaces (chip and tool rake face) with a heat
transfer coefficient h between material and environment that have temperature T w
and T 0 , respectively. Hence, heat from the cut material is transferred to the tool.
Q t = h(Tw − T0 ) (2.1)
Typical assumptions in heat transfer modeling are the fact that 100% of the friction
work is converted to heat, with 90% of the work converted to heat and 10% stored
in the material [35]. The heat flux Qd generated by the plastic distortion of the work
ẇ p can be determined with Eq. 2.2, which takes into account the fraction of the
mechanical energy losses to heat and the material density ρ.
Fig. 2.11 Heat exchange schematic in FEM-based study
f ẇ p
Qd = (2.2)
ρ
The heat generated due to the frictional forces Qf is given by Eq. 2.3 as a rate,
where ηf is the friction-based work conversion coefficient, τ f denotes the frictional
stresses, and us is the sliding velocity between the cutting tool and the part.
Q f = η f τ f us (2.3)
Finally, Fig. 2.11 shows the relationship between the surfaces that come in contact,
as well as the heat flux between the materials and the environment, while machining
a workpiece, as modeled with the FEM.
2.2.5 Material Flow Stress Modeling
To achieve reasonable and reliable results, the use of material parameters as close
to reality as possible is important. Since the phenomena that take place during
machining are complex, properties such as flow stress, yield strength, fraction strain,
friction, and elastic constants, as well as thermo-physical properties such as density,
Poisson’s ratio, thermal conductivity, and thermal capacity, should be determined
with the best possible accuracy for a wide range of strains, strain rates, tempera-
tures, and pressures. Depending on the type of machining operation, the tools, and
the materials that take place in the process, the occurrence of extreme conditions is
expected; strains of 100–700%, strain rates up to 10−6 s, temperatures at the range
of 500–1400 °C, heating rates that can reach values close to106 °Cs−1 and pressures
that may build up towards 3GPa [52].
Nowadays, a decent number of flow stress models exist, as well as the equivalent
constants for a variety of known materials spanning from aluminium alloys, steel,
and titanium to thermoplastics. The power law equation is one of the simplest models
for flow stress. It is an elastoplastic model ideal for studying the relationship between
flow stress and strain at low temperatures and strain rates. Equation 2.4 represents the
model, which calculates flow stress in relation to strain, strain rate, and temperature.
σ = C(T )ε n(T ) (2.4)
where σ is the flow stress usually expressed in MPa, C refers to the strength coeffi-
cient, T is the temperature, ε denotes the strain, whereas n is the exponent expressing
strain hardening. The next parameters, C and n are in function with the temperature.
Oxley [53] proposed a modified power law equation for the description of the
material flow stress, by implementing the velocity-modified temperature into the
model. Equation 2.5 shows the modified model, whereas Eq. 2.6 represents the
modified temperature.
σ = C(T mod )ε n(T mod ) (2.5)

ε̇
T mod = T 1 − u log (2.6)
ε̇0
where σ is the flow stress, T and T mod are the temperature and the modified temper-
ature, respectively, C refers to the strength coefficient, ε denotes the strain, whereas
ε̇ and ε̇ 0 the strain rate and the default strain rate in s−1 , respectively, n is the strain
hardening factor, and finally, u is a constant. Both the coefficient for strength and the
strain hardening exponent are in function with the modified temperature.
Maekawa et al. [54] included the strain path-dependent loading effect, in their
proposed model. Equation 2.7 presents the model, where the integral is assigned for
the history effect. To implement this model in FEM, it is required to reduce it to a
simpler form.
⎡ ⎤N
M m −m
ε̇ ε̇ ⎢ ε̇ N
⎥
e−kT
kT
σ =C ⎣ eN dε ⎦ (2.7)
1000 1000 1000
ε,T ≡ε̇
where σ is the flow stress, T is the temperature, C is the strength coefficient, ε denotes
the strain, whereas ε̇ is the strain rate, M is the strain rate sensitivity, and k, N, and
m are constants.
A broadly accepted model is the one proposed by Johnson and Cook [55], which is
used for stress analysis under extreme conditions that occur in deformations, followed
by high strain rates and temperatures. The model was modified in the past decade
by many researchers, to meet the needs of different applications [56–62] involving

machining, metal forming, material testing, and structural crashworthiness for a range
of materials such as steel, aluminium, and titanium alloys.
Equation 2.8 represents the generalized form of the Johnson–Cook equation,
which is divided into three terms. The first term considers the strain, the second
one the strain rate, and the third takes into account the temperature. It is shown that
each term embeds each of the three aforementioned parameters separately. As long
as the constants are experimentally determined, the model can be used for a variety
of materials.

ε̇ θ − θ0 m
σ = A + Bε n 1 + C ln 1− (2.8)
ε̇0 θm − θ0
A, B, and C denote the material constants that relate to the stresses and strains
accordingly, n and m are the strain hardening exponent and a coefficient related to
temperature respectively. Finally, θ and θ 0 are the reference and bulk temperature
accordingly, whereas θ m is the material melting temperature.
The constitutive model developed by Zerilli and Armstrong [63] is based on the
dislocation mechanics theory and takes into account the structure of metals. There-
fore, authors suggested two models, one for each type of metal structure. Equa-
tions 2.9 and 2.10 correspond to the BCC and FCC structure, respectively, including
the plastic strain hardening contribution.

σ = c0 + c1 exp −c3 T + c4 T ln ε̇ + c5 ε n (2.9)

σ = c0 + c2 ε n exp −c3 T + c4 T ln ε̇ (2.10)
where c0 to c5 and n are constants, that can be experimentally determined for a variety
of metals.
It is noteworthy that many FEA systems allow the expression of the material
plastic behavior in a tabular form. Despite the fact that material plastic behavior
is a non-linear phenomenon, a tabular data form enables the linear or logarithmic
connection between strain, strain rate, and temperature. Therefore, any material can
be represented as a function of these three parameters, as long as the data are available,
making the tabular data form a versatile and robust approach. Equation 2.11 shows
the model based on tabular data of strain, strain rate, and temperature.

σ = f ε, ε̇, T (2.11)
where σ equals to the effective flow stress, ε is the effective strain, ε̇ is the effective
strain rate, and T is the temperature.
Table 2.1 Sample flow stress models for 3D studies of steel

A (MPa) B (MPa) C n m Material-process Source
546 487 0.25 0.027 0.631 AISI1045 turning [11]
553.1 600.8 0.234 0.0134 1.0 AISI1045 drilling [64]
595 580 0.023 0.133 1.03 AISI4140 turning [65]
598 768 0.0137 0.2092 0.807 AISI4140 turning [12]
950 725 0.015 0.375 0.625 AISI4340 turning [16]
688.17 150.82 0.043 0.336 2.77 AISI52100 turning [66]
Table 2.2 Sample flow stress models for 3D studies of aluminium alloys
546 678 0.024 0.71 1.56 AA7075-T6 drilling [67]
527 575 0.017 0.72 1.61 AA7075-T6 drilling [68]
214.25 327.7 0.00747 1.31 0.504 AA6082-T6 milling [26]
370.4 1789.37 0.0128 0.73315 1.5282 AA6061-T6 milling [69]
324 114 0.002 0.42 1.34 AA6061 turning [70]
Table 2.3 Sample flow stress models for 3D studies of Ti6Al4V

1098 1092 0.014 0.93 1.1 Drilling, milling [71, 72]
1000 780 0.033 0.47 1.02 Milling [25]
1080 1007 0.01304 0.6349 0.77 Milling [73]
782.7 498.4 0.028 0.28 1 Drilling [74]
862.5 683.1 0.012 0.35 1 Turning [75]
Material constants
The accuracy of the flow stress model is a factor that greatly affects the generated
results. The model constants are obtained via experimental testing and usually require
fine-tuning depending on the process studied, the conditions, and the tool-workpiece
geometries. The next Tables 2.1, 2.2, and 2.3 present a comparison between sample
constitutive models for a number of typical industry-standard materials such as steels,
aluminium alloys, and titanium accordingly.
2.2.6 Friction Modeling and Contact Description
Friction between two bodies is a complex phenomenon where extreme pressures and
temperatures are generated. Especially for machining, three distinct heat generation
Fig. 2.12 The deformation zones in metal cutting
zones are present, the primary zone where deformation occurs, also called shear zone,
where the elastoplastic deformation occurs, and the maximum heat is generated. The
secondary deformation also called the friction zone, where plastic deformation is
present, and the rest of the heat is produced between the moving chips and the tool,
due to friction. And finally, the tertiary zone, which is the tool-workpiece interface
zone where elastic deformation occurs. In this zone, the heat generated is minimal
compared to the sum of the other two. Figure 2.12 illustrates the three deformation
zones during metal cutting.
A number of models that deal with the friction description are available in the
literature. Coulomb’s law is the most commonly used model, providing an adequate
estimation of the friction that occurs in the sliding subzone. This model (Eq. 2.12)
describes the relationship between frictional and normal stress via a coefficient.
τ f = μσn (2.12)
where τ f is the stress, μ the coefficient for friction, and σ n the normal stress. This
formula is easy to implement, however it does not yield very accurate results, espe-
cially after a certain critical value of the normal stress. Despite this fact, it is still
used [24, 65, 76] for its simplicity and the adequate approximation it offers under
certain conditions (i.e. low cutting speeds for turning) in the sliding region, which is
of great importance in machining.
According to Zorev [77] the interface where the chip and the tool are in contact
(secondary zone) can be discretized into two subregions, the sticking and the sliding.
The author stated that the frictional stress in the shear zone cannot exceed the shear
yield strength. Thus, the frictional stress in the sticking subzone can be determined
in relation to the material’s shear yield strength as indicated by Eq. 2.13.
σ
τf = m√ (2.13)
3
where m is the shear friction constant and σ the effective stress. The combination
of the two models can provide a better approximation for friction, considering that
the frictional stresses in the sticking region can be calculated by the material’s shear
strength, and in the sliding region they increase based on Coulomb’s formula. Many
simulation systems allow the implementation of a hybrid friction model, where the
user can input both shear and Coulomb friction constants.
Advancement of friction models led to the description of the intermediate region
(transitional subzone) that connects the sticking to the sliding subzone. Usui and
Shirakashi [78] proposed a model based on Zorev’s assumptions, which implements
non-linearity for the stress expression. Their empirical description is represented by
Eq. 2.14.

τ f = k 1 − e−(μσn /k) (2.14)
where k is the material flow stress and μ the friction coefficient as used in Coulomb’s
law, which can be calculated by Eq. 2.15 after determining the feed force F f and the
cutting force F c experimentally, in addition to the rake angle γ . The formula is based
on Merchant’s analysis [79] and does not take into account the effect of ploughing.
A modification by Albrecht [80] is available that considers the condition of the tool.
F f + Fc tan γ
μ= (2.15)
Fc − F f tan γ
Childs et al. [81] proposed a refined version of Eq. 2.14, for a smoother transition
between the two subzones by implementing two factors, m and n, as shown in Eq. 2.16.
The first coefficient ensures that the frictional stresses do not exceed flow stress at
high values of normal stress and the latter is responsible for the smooth transition
from the sticking to the sliding subzone. Both coefficients can be calculated with
experimental testing (split tool technique).

n 1/n
τ f = mk 1 − e−(μσn /mk) (2.16)
Another approach is the one from Iwata et al. [82] where the hardness of the
material is taken into account, as well as the pressure that is developing between the
newly created surface and the tool face. The empirical formula is shown in Eq. 2.17
where H V is the hardness in the Vickers scale and p is the pressure usually expressed
in MPa.

HV 0.07μp
τf = tanh (2.17)
0.07 HV
Additional models were proposed by researchers in the past few years. Sekhon
and Chenot [83] derived a computationally convenient form from Norton’s law,
which utilizes the relative sliding velocity, given between the tool and the chip.
Yang and Liu [84] proposed a stress-based friction model in which frictional and
normal stresses are related with the aid of fourth-order polynomial. Zemzeni et al.
[85] investigated the possibility of identifying a friction model and determining a
coefficient for the tool, as well as the workpiece and chip area of interest when dry-
machining AISI4142 steel. Similar works [86, 87] led to the identification of friction
parameters for machining modeling widening the range of materials and conditions.
Moreover, Palanisamy et al. [88] examined recently the friction modeling of Ti6Al4V
machining, in addition to the constitutive modeling.
Friction coefficients
Despite that Coulomb’s law is oversimplified in terms of the approximation degree
of the frictional forces, it is used in many FEM studies, both 2D and 3D. As already
discussed, FEM investigations on processes with complexities, require more sophis-
ticated models for friction in order to generate accurate results. Table 2.4 summa-
rizes the implementation of friction models by a number of studies focused on the
machining of typical materials in three dimensions. Most works implement a static
Coulomb coefficient, however, a hybrid model, sometimes with variable values, is
expected to yield results of better accuracy. Arrazola et al. [89] proposed a new
approach for friction modeling, by using variable coefficients, and reported increased
accuracy compared to static modeling.
Table 2.4 Friction modeling in 3D machining

Model Coefficient Process Material Source
Coulomb μ = 0.3 Turning Ti6Al4V [75]
Coulomb μ = 0.5 Milling Ti6Al4V [27]
Coulomb μ = 0.7 Milling AA6082-T6 [26, 90]
Coulomb μ = 0.17 Turning AA2024-T351 [91]
Coulomb μ = 0.6 Turning AISI-D2 [92]
Coulomb μ = 0.577 Turning AISI-4140 [43]
Coulomb μ = 0.32 Turning AISI-4140 [65]
Coulomb μ = 0.8 Milling AISI-306L [21]
Shear m = 0.82 Turning AISI-1045 [93, 94]
Hybrid μ = 0.7 Drilling Ti6Al4V [95]
m = 0.9
Hybrid μ = 0.5 Milling Aluminium 6000 [24]
m = 1.0
2.2.7 Material Separation
As soon as the tool model begins to penetrate the workpiece, the sum of the mathemat-
ical formulae for contact, friction, flow stress, material separation, etc., that constitute
the machining process, comes into effect. Material separation due to damage, is the
stage at which the elements forming the cut surface, begin to separate so that the
undeformed chip can be shaped. Similar to friction, flow stress, etc., modeling the
chip separation and evolution is a complex task.
When a ductile material starts losing its ability to resist deformation, structural
failure begins to occur according to stiffness degradation. Moreover, when the stiff-
ness no longer exists, the material fails. Figure 2.13 illustrates a typical stress–strain
curve visualizing the stress–strain response of a ductile material. The line between
points (a) and (b), represents the linear elastic zone. Curve (b–c) is characterized
by the plastic yielding with strain hardening. Damage initiates at point (c) and from
that point on, the material begins to lose its load-carrying capacity. Finally, point (d)
translates into total fracture. The region between points (b) and (c) can be modeled
using a flow-stress model, whereas the rest of the curve past point (c) requires a
damage criterion.
A well-established criterion for damage modeling is the Johnson–Cook shear
failure model [96], which depends on the strain rate, temperature, pressure, and
equivalent stress. Equation 2.18 represents the aforementioned criterion.
σ
m
ε f = D1 + D2 exp D3 [1 + D4 ln ε̇][1 + D5 T ] (2.18)
σ
where εf is the equivalent strain to fracture, D1 –D5 are the material damage constants,
σ m is the average given by the three normal stresses, σ̃ is the equivalent stress,
ε̇ is the strain rate, and T the temperature. The material damage parameters are
usually determined experimentally with tensile [97–99] or impact tests [100–102]
for a variety of standard materials such as carbon steels and aluminium alloys.
An equally important criterion of ductile fracture is the one presented by Cockcroft
and Latham [103], which considers the maximum normal stress that is operating.
Fig. 2.13 Uniaxial

stress–strain response for
ductile specimen
Equation 2.19 is the normalized version of the criterion proposed by the authors, that
describes the occurrence of fracture when the integral is equal to the constant value
Dc , for a given strain rate and temperature.
ε f
σmax
Dc = dε pl (2.19)
σ
0
where Dc is the material constant value, σ max denotes the maximum stress, whereas
σ represents the equivalent stress. Additionally, εf is the upper limit of the integral,
representing the fracture strain and εpl is the plastic strain. A modified adaptation of
the criterion was used recently by Razanica et al. [104].
Other notable works on the damage initiation formula that can be implemented
in FEM and are based on the concepts of effective stress and equivalent strain, come
from McClintock [105], Rice and Tracey [106], Brozzo et al. [107], Chandrakanth
and Pandey [108] and Obikawa et al. [109].
Damage evolution
The damage progression is based on the fracture energy that is dissipated during
the process [110]. Once the damage criterion is fulfilled, it is necessary that a new
approach is used for the description of the material separation evolution, since the
energy dissipated during the deformation decreases with remeshing. As such, a stress-
displacement response is necessary to minimize mesh dependency that the stress–
strain relation produces. Fracture energy Gf in a totally fractured material is given
by Eq. 2.20 [5, 111] with respect to the fracture toughness K c , the elastic modulus
E, and Poisson’s ratio ν, which can be used to reduce the mesh dependency.
(1 − ν 2 )
G f = (K c )2 (2.20)
E
Exponential damage behavior is considered to exist between the chip and the
tool path zone, as described in Eq. 2.21. The formula is set in such a way that the
energy dissipated during material separation is close to the fracture energy Gf , and the
varying stiffness degradation D draws close to unity asymptomatically at an infinite
equivalent plastic displacement, where σ and u are the equivalent stress and plastic
displacement, respectively.
⎛ ⎞
u f
⎜ σ ⎟
D = 1 − exp⎝− du ⎠ (2.21)
Gf
0
In the case that stiffness degradation attains unity, the flow stress tensor σ for the
failed element can be described by Eq. 2.22.
σ = (1 − D)σ (2.22)
2.2.8 Tool Wear
Prediction modeling of the tool wear phenomenon has been studied by many
researchers. Usually, these models approximate the rate of material volume loss
on the contact face, such as the rake or the flank face, with respect to the contact area
and the temporal development of the wear effect. Furthermore, it is required that the
material characteristics and the cutting parameters are included in the models. One
of the first approaches to approximate the wear mechanism between flat surfaces
was made by Archard [112]. The author established a simple equation to describe
sliding-based wear, according to the asperity contact theory, which was implemented
in many studies related to cutting tool wear, with modifications.
Palanikumar and Davim [113] developed a formulation to predict tool wear during
machining of composite components. Marksberry and Jawahir [114] modified a
model from previous work to be used with near-dry machining. Their model for dry
machining is shown in Eq. 2.23.
(1/n c )
km VR
T = TR n n (2.23)
f 1d 2 V
where T is the tool wear, T R denotes the tool wear reference for 1 min, V is the
cutting speed, whereas V R is the reference cutting speed when tool wear is measured
for one minute, nc is a factor for the coating effect, f is the depth of cut, d the tool
nose radius, and finally, k, n1 , and n2 are empirical constants. An early study on tool
wear was carried out by Takeyama and Murata [115]. Hao et al. [116] analyzed the
tool wear schema and mechanisms for Inconel 718 dry cutting. Zhang et al. [117]
researched the wear for diamond tools used in precision machining and investigated
ways of implementing the equivalent modeling to analytical and numerical methods.
More recent studies [118–122] focus on the prediction of tool wear with the aid of
methods such as neural networks, machine learning, and fuzzy logic. Typical steels,
aluminium, and titanium alloys are the materials of choice for many works. It is
noteworthy that the model from Usui et al. [123] for crater and flank wear, that
takes into consideration the interface pressure and temperature, as well as the sliding
velocity, can be implemented in FEM with success. Equation 2.24 represents the tool
wear over time as an integral.
dW
= apvs exp(−b/T ) (2.24)
dt
2.3 Typical FEM-Based Results 37
where W is the wear, p is the contact pressure, vs the sliding velocity, T denotes the
temperature at the contact area and, lastly, a and b are experimentally determined
coefficients.
2.3 Typical FEM-Based Results
FEM-based studies can generate a variety of predicted state variables and parameters
such as cutting forces, torque, temperature distribution, residual stresses, and tool
wear. The accuracy provided by the use of FEM is acceptable; however, it strongly
depends on the models used for the rheology of the material, the friction, and so
on. According to Melkote et al. [124], a tendency to predict lower thrust forces in
FE metal cutting is present, probably due to the oversimplification of the friction
coefficient values, pointing out the importance of friction modeling.
2.3.1 Cuttings Forces, Torque, and Residual Stresses
The cutting forces prediction is crucial for all machining processes since it constitutes
an indicator for the stress of the machine, the quality of the machined part, as well
as the longevity of the tool. Higher values of forces than expected, usually lead to
faster tool wear, frequent machine maintenance, and production of lower quality
parts. Torque is a parameter of great interest as well, since it is directly related to the
machining power. Thus, knowledge of the produced torque during an operation can
provide the user with useful insight into the amount of power that is required, the cost,
and so on. Furthermore, prediction of the stresses that remain on the microstructure
of the metal after the machining operation, enables the assessment of the material’s
surface integrity. The degree of residual stresses that appear is a parameter that affects
the quality of the machined part directly.
With this in mind, many studies [14, 61, 68, 75] include the assessment of the
predicted forces and other parameters for optimization purposes. By implementing
statistical techniques and machine learning, it is possible to determine the optimal
cutting conditions, such as depth of cut, tool geometrical aspects, feed, and cutting
speed for a specific scope. The numerical results are often compared with analogous
experimental data for verification. Experimental testing is the most effective way
to calibrate the developed FE model and fine-tune the corresponding factors that
participate in the modeling procedure. Multi-component dynamometers and data
acquisition systems are the mean for acquiring the force components, as well as
torque. Sreeramulu et al. [34] modeled the oblique cutting procedure of Al7075-T6
with coated tools, in DEFORM™-3D, with respect to feed, cutting speed, and depth
of cut, generating results for the cutting force. Oezkaya and Biermann [125] worked
on the modeling of AISI1045 steel tapping by simulating the process in DEFORM™-
3D. They examined the generated torque with tool production in mind and developed
a mathematical model for the generation of torque during a total load cycle. Rami
et al. [65] investigated the ability to determine the residual stresses during machining
of AISI4140 steel with a FE model developed in ABAQUS™. Moreover, the authors
simulated several tool passages under various conditions and examined the effect of
the thermo-mechanical loadings, as well as of the feed. Similar studies [20, 126, 127]
exhibit the advantages of FEM in machining modeling and the ability to investigate
the aforementioned parameters.
2.3.2 Chip Morphology, Temperature Distribution, and Wear
Temperature allocation on both the tool and the part, as well as the chip geometric
characteristics are considered to be factors of equal significance to tool wear and
deflection. Thermally induced errors contribute a large proportion of the total
machining error. Temperature-based factors affect friction conditions and altogether
influence the generated chip morphology and the rate of tool wear. Pittalà and Monno
[24] investigated the chip temperature and the cutting forces generated while face-
milling aluminium 6000. The 3D FE model of this case was built with DEFORM™-
3D. The authors performed a friction sensitivity analysis to improve the model. Lotfi
et al. [35] studied the drilling process of AISI1045 with the aid of FEM with regard to
the heat built-up, the tool wear, and the BUE. Attanasio et al. [93] developed a 3D FE
model for studying the turning of AISI-1045 within a range of f and V c . The chosen
software was DEFORM™-3D. The authors focused on the tool wear investigation
by utilizing a modified tool wear model, coupled with experiments. Paktinat and
Amini [128] made a comparison between the conventional drilling procedure and
the drilling of Al7075 assisted ultrasonically. In a similar manner to previous studies,
they developed a FE model in three dimensions for the examination of the developed
stresses, the BUE formation, as well as the chip morphology, and the temperature
distribution. Analogous works [26, 129, 130] focused on three-dimensional modeling
and simulation of machining proceedings were carried out, expanding the range of
knowledge in the field, coupled with experimental work in order to investigate vari-
ables similar to the ones mentioned in this paragraph. It should be mentioned that
a number of papers are available in the bibliography that focus on other factors and
actions that are involved in machining, such as the effect of cutting fluids, the tool
runout, the minimization of machining errors, and more.
Figure 2.14a presents the formation of the chip during dry drilling of aluminium
alloy 7075 with a 5-mm HSS twist drill, while Fig. 2.14b illustrates the temperature
distribution on the drill’s surfaces that are in contact with the work, with the aid of
the color gradient. It is evident that by observing the figures it is possible to extract
useful information for the tool performance and the evolution of the process.
References 39
Fig. 2.14 Example chip generation (a) and tool-tip temperature distribution and b in drilling
2.4 Conclusions and Perspectives
This chapter introduces the most important aspects of FE modeling in machining

by summarizing concepts on the representation of the tool-workpiece interface, the
meshing process, the set of boundary conditions, the friction phenomenon, and the
material separation. Furthermore, it includes several works in the field from the early
years until recently, as well as references on standard models and formulae that can
be implemented by modern FEA software, to model most machining processes in
three dimensions with adequate precision.
It is concluded that 3D FEM in machining is a highly valued tool that provides
several advantages, especially compared to 2D FEM, in the investigation of the
machinability of materials and the performance of the processes. However, it should
be pointed out that despite the advantage of the computational resources, 3D FEA
still requires hefty amounts of run-time. Moreover, many models for material
flow, damage, etc., are not optimized for three-dimensional studies. Therefore, it
is imperative that new models and modifications are developed for 3D applications.
Finally, it is anticipated that 3D models will increase in number with time,
especially with the advent of more sophisticated technologies and computational
resources, expanding the field of work to non-conventional methods on aerospace,
energy, and biomechanics applications.
References
1. Ceretti E, Fallböhmer P, Wu WT, Altan T (1996) Application of 2D FEM to chip formation in

orthogonal cutting. J Mater Process Technol 59:169–180. https://doi.org/10.1016/0924-013
6(96)02296-0
2. Klocke F, Raedt H-W, Hoppe S (2001) 2D-FEM simulation of the orthogonal high speed
cutting process. Mach Sci Technol 5:323–340. https://doi.org/10.1081/MST-100108618
3. Maranhão C, Paulo Davim J (2010) Finite element modelling of machining of AISI 316 steel:
numerical simulation and experimental validation. Simul Model Pract Theory 18:139–156.
https://doi.org/10.1016/j.simpat.2009.10.001
4. Xie LJ, Schmidt J, Schmidt C, Biesinger F (2005) 2D FEM estimate of tool wear in turning
operation. Wear 258:1479–1490. https://doi.org/10.1016/j.wear.2004.11.004
5. Yaich M, Ayed Y, Bouaziz Z, Germain G (2017) Numerical analysis of constitutive coefficients
effects on FE simulation of the 2D orthogonal cutting process: application to the Ti6Al4V.
Int J Adv Manuf Technol 93:283–303. https://doi.org/10.1007/s00170-016-8934-4
6. Calamaz M, Coupard D, Girot F (2008) A new material model for 2D numerical simulation of
serrated chip formation when machining titanium alloy Ti-6Al-4V. Int J Mach Tools Manuf
48:275–288. https://doi.org/10.1016/j.ijmachtools.2007.10.014
7. Rana K, Rinaldi S, Imbrogno S et al (2016) 2D FE prediction of surface alteration of Inconel
718 under machining condition. Procedia CIRP 45:227–230. https://doi.org/10.1016/j.procir.
2016.02.346
8. Liu L, Wu M, Li L, Cheng Y (2020) FEM simulation and experiment of high-pressure
cooling effect on cutting force and machined surface quality during turning Inconel 718.
Integr Ferroelectr 206:160–172. https://doi.org/10.1080/10584587.2020.1728637
9. Guo YB, Liu CR (2002) 3D FEA modeling of hard turning. J Manuf Sci Eng Trans ASME
124:189–199. https://doi.org/10.1115/1.1430678
10. Karpat Y, Ozel T (2008) Process simulations for 3D turning using uniform and variable micro-
geometry PCBN tools. Int J Mach Mach Mater 4:26–38. https://doi.org/10.1504/IJMMM.
2008.020908
11. Buchkremer S, Klocke F, Veselovac D (2016) 3D FEM simulation of chip breakage in metal
cutting. Int J Adv Manuf Technol 82:645–661. https://doi.org/10.1007/s00170-015-7383-9
13. Hu HJ, Huang WJ (2013) Effects of turning speed on high-speed turning by ultrafine-grained
ceramic tool based on 3D finite element method and experiments. Int J Adv Manuf Technol
67:907–915. https://doi.org/10.1007/s00170-012-4535-z
turning: modelling of cutting forces and cutting condition optimization. Machines 10:74.
https://doi.org/10.3390/machines10020074
15. Gao X, Li H, Liu Q et al (2011) Simulation of stainless steel drilling mechanism based
on Deform-3D. Adv Mater Res 160–162:1685–1690. https://doi.org/10.4028/www.scientific.
net/AMR.160-162.1685
16. Majeed A, Iqbal A, Lv J (2018) Enhancement of tool life in drilling of hardened AISI 4340
steel using 3D FEM modeling. Int J Adv Manuf Technol 95:1875–1889. https://doi.org/10.
1007/s00170-017-1235-8
17. Oezkaya E, Hannich S, Biermann D (2019) Development of a three-dimensional finite element
method simulation model to predict modified flow drilling tool performance. Int J Mater Form
18. Nagaraj M, Kumar AJP, Ezilarasan C, Betala R (2019) Finite element modeling in drilling of
Nimonic C-263 alloy using deform-3D. C Comput Model Eng Sci 118:679–692. https://doi.
org/10.31614/cmes.2019.04924
19. Attanasio A, Ceretti E, Outeiro J, Poulachon G (2020) Numerical simulation of tool wear in
drilling Inconel 718 under flood and cryogenic cooling conditions. Wear 458–459:203403.
https://doi.org/10.1016/j.wear.2020.203403
20. Tzotzis A, García-Hernández C, Huertas-Talón J-L, Kyratsis P (2020) FEM based mathemat-
ical modelling of thrust force during drilling of Al7075-T6. Mech Ind 21:415. https://doi.org/
10.1051/meca/2020046
21. Maurel-Pantel A, Fontaine M, Thibaud S, Gelin JC (2012) 3D FEM simulations of shoulder
milling operations on a 304L stainless steel. Simul Model Pract Theory 22:13–27. https://doi.
org/10.1016/j.simpat.2011.10.009
0157-6
References 41
23. Gao Y, Ko JH, Lee HP (2018) 3D coupled Eulerian-Lagrangian finite element analysis of end
milling. Int J Adv Manuf Technol 98:849–857. https://doi.org/10.1007/s00170-018-2284-3
25. Wu HB, Zhang SJ (2014) 3D FEM simulation of milling process for titanium alloy Ti6Al4V.
Int J Adv Manuf Technol 71:1319–1326. https://doi.org/10.1007/s00170-013-5546-0
10.3390/mi8060187
27. Ji C, Li Y, Qin X, et al (2015) 3D FEM simulation of helical milling hole process for titanium
alloy Ti-6Al-4V. 1733–1742. https://doi.org/10.1007/s00170-015-7323-8
stresses in AA6082-T6 Al alloy generated by wire EDM cutting. J Mater Process Tech 275:1–
12. https://doi.org/10.1016/j.jmatprotec.2019.116373
29. Quarto M, Urso GD, Giardini C, Maccarini G (2020) FEM model development for the
simulation of a micro-drilling EDM process. Int J Adv Manuf Technol 106:3095–3104
30. Ming W, Zhang S, Zhang G et al (2022) International journal of heat and mass transfer progress
in modeling of electrical discharge machining process. Int J Heat Mass Transf 187:1–32.
https://doi.org/10.1016/j.ijheatmasstransfer.2022.122563
31. Liu C, Duong N, Jahan MP et al (2019) Experimental investigation and numerical simula-
tion of of bulk metallic glass with focus on crater sizes bulk metallic glass with with focus
focus on on crater crater sizes sizes metallic glass. In: Procedia manufacturing. Elsevier B.V.,
Pennsylvania, pp 275–286
machining forces during AISI 4140 hard turning. Strojniški Vestn J Mech Eng 66:467–478.
33. Mathivanan A, Swaminathan G, Sivaprakasam P et al (2022) DEFORM 3D simulations and
Taguchi analysis in dry turning of 35CND16 steel. Adv Mater Sci Eng 2022:1–10
34. Sreeramulu D, Rao CJ, Sagar Y, Venkatesh M (2018) Finite element modeling and machining
of Al 7075 using coated cutting tools. Mater Today Proc 5:8364–8373. https://doi.org/10.
1016/j.matpr.2017.11.530
36. Ucun İ, Aslantas K, Bedir F (2016) Finite element modeling of micro-milling: numerical
simulation and experimental validation. Mach Sci Technol 20:148–172. https://doi.org/10.
1080/10910344.2016.1147650
37. Ibrahim M, El A (2019) Investigation of finite element modelling on thin- walled machining
of Ti6Al4V using DEFORM-3D investigation of finite element modelling on thin-walled
machining of Ti6Al4V using DEFORM-3D. https://doi.org/10.1088/1742-6596/1167/1/
012002
010004
39. Scientific Forming Technologies Corporation (2016) DEFORM V11.3 (PC) documentation
40. Senthil Kumar N, Tamizharasan T (2012) Optimization of cutting insert geometry using
DEFORM-3D: numerical simulation and experimental validation. Int J Simul Model 11:65–
76. https://doi.org/10.2507/IJSIMM11(2)1.200
41. Gardner J, Dornfeld D (2006) Finite element modeling of drilling using DEFORM. Berkeley
Univ Calif 1–8
42. Guo YB, Dornfeld DA (2000) Finite element modeling of burr formation process in drilling
304 stainless steel. J Manuf Sci Eng Trans ASME 122:612–619. https://doi.org/10.1115/1.
1285885
44. Lotfi M, Jahanbakhsh M, Farid AA (2016) Wear estimation of ceramic and coated carbide
tools in turning of Inconel 625: 3D FE analysis. Tribol Int 99:107–116. https://doi.org/10.
1016/j.triboint.2016.03.008
45. Markopoulos A (2013) Finite element method in machining processes. Springer
46. Llanos I, Villar JA, Urresti I, Arrazola PJ (2009) Finite element modelling of oblique
machining using an arbitrary Lagrangian-Eulerian formulation. Mach Sci Technol 13:385–
406. https://doi.org/10.1080/10910340903237921
47. Agmell M (2018) Applied FEM of metal removal and forming, 1st edn. Studentlitteratur,
Lund
48. Kheireddine AH, Lu T, Jawahir IS, Hamade RF (2013) An FEM analysis with experimental
validation to study the hardness of in-process cryogenically cooled drilled holes in Mg. 8:588–
593. https://doi.org/10.1016/j.procir.2013.06.156
49. Tzotzis A, García-Hernández C, Huertas-Talón JL, Kyratsis P (2020) CAD-based automated
design of FEA-ready cutting tools. J Manuf Mater Process 4:1–14. https://doi.org/10.3390/
jmmp4040104
50. Vijayaraghavan A, Dornfeld DA (2007) Automated drill modeling for drilling process
simulation. J Comput Inf Sci Eng 7:276–282. https://doi.org/10.1115/1.2768091
51. Kyratsis P, Tzotzis A, Manavis A (2021) Computational design and digital fabrication. In:
Kumar S, Rajurkar KP (eds) Advances in manufacturing systems. Springer, Singapore, pp
1–16
52. Arrazola PJ, Özel T, Umbrello D et al (2013) Recent advances in modelling of metal machining
processes. CIRP Ann Manuf Technol 62:695–718. https://doi.org/10.1016/j.cirp.2013.05.006
53. Oxley PLB (1989) The mechanics of machining: an analytical approach to assessing
machinability
54. Maekawa K, Shirakashi T, Usui E (1983) Flow stress of low carbon steel at high temperature
and strain rate. II: flow stress under variable temperature and variable strain rate. Bull Japan
Soc Precis Eng 17:167–172
55. Johnson GR, Cook WH (1983) A constitutive model and data for materials subjected to
large strains, high strain rates, and high temperatures. In: Proceedings of the 7th international
symposium on ballistics, pp 541–547
56. Lin YC, Chen XM, Liu G (2010) A modified Johnson-Cook model for tensile behaviors of
typical high-strength alloy steel. Mater Sci Eng A 527:6980–6986. https://doi.org/10.1016/j.
msea.2010.07.061
57. Wang X, Huang C, Zou B et al (2013) Dynamic behavior and a modified Johnson-Cook
constitutive model of Inconel 718 at high strain rate and elevated temperature. Mater Sci Eng
A 580:385–390. https://doi.org/10.1016/j.msea.2013.05.062
58. Song W, Ning J, Mao X, Tang H (2013) A modified Johnson-Cook model for titanium matrix
composites reinforced with titanium carbide particles at elevated temperatures. Mater Sci Eng
A 576:280–289. https://doi.org/10.1016/j.msea.2013.04.014
59. He J, Chen F, Wang B, Zhu LB (2018) A modified Johnson-Cook model for 10%Cr steel at
elevated temperatures and a wide range of strain rates. Mater Sci Eng A 715:1–9. https://doi.
org/10.1016/j.msea.2017.10.037
60. Duc-Toan N, Tien-Long B, Dong-Won J et al (2015) A modified Johnson-Cook model to
predict stress-strain curves of boron steel sheets at elevated and cooling temperatures. High
Temp Mater Process 31:37–45. https://doi.org/10.1515/htmp.2011.127
61. Li S, Sui J, Ding F et al (2021) Optimization of milling aluminum alloy 6061–T6 using
modified Johnson-Cook model. Simul Model Pract Theory 111:102330. https://doi.org/10.
1016/j.simpat.2021.102330
62. Seo J-M, Jeong S-S, Kim Y-J et al (2021) Modification of the Johnson–Cook model for the
strain rate effect on tensile properties of 304/316 austenitic stainless steels. J Press Vessel
Technol 144. https://doi.org/10.1115/1.4050833
63. Zerilli FJ, Armstrong RW (1987) Dislocation-mechanics-based constitutive relations for
material dynamics calculations calculations. J Appl Phys 61:1816–1825
References 43
64. Nan X, Xie L, Zhao W (2016) On the application of 3D finite element modeling for small-
diameter hole drilling of AISI 1045 steel. Int J Adv Manuf Technol 84:1927–1939. https://
doi.org/10.1007/s00170-015-7782-y
65. Rami A, Kallel A, Sghaier S et al (2017) Residual stresses computation induced by turning
of AISI 4140 steel using 3D simulation based on a mixed approach. Int J Adv Manuf Technol
66. Umer U, Kishawy H, Mian SH et al (2021) 3D-modeling of hard turning using self-propelled
rotary tools. Procedia CIRP 102:103–108. https://doi.org/10.1016/j.procir.2021.09.018
67. Tzotzis A, Markopoulos A, Karkalos N, Kyratsis P (2020) 3D finite element analysis of
Al7075-T6 drilling with coated solid tooling. In: MATEC web of conferences, pp 1–6
68. Yaşar N (2019) Thrust force modelling and surface roughness optimization in drilling of AA-
7075: FEM and GRA. J Mech Sci Technol 33:4771–4781. https://doi.org/10.1007/s12206-
019-0918-5
69. Li G, Liu M, Zhao S (2021) Reduced computational time in 3D finite element simulation of
high speed milling of 6061–T6 aluminum alloy. Mach Sci Technol 25:558–584
70. Ojal N, Copenhaver R, Cherukuri HP et al (2022) A realistic full-scale 3D modeling of turning
using coupled smoothed particle hydrodynamics and finite element method for predicting
cutting forces. J Manuf Mater Process 6. https://doi.org/10.3390/jmmp6020033
71. Krishnakumar P, Sripathi J, Vijay P, Ramachandran KI (2016) Finite element modelling and
residual stress prediction in end milling of Ti6Al4V alloy. In: IOP conference series: materials
science and engineering
73. Rao B, Dandekar CR, Shin YC (2011) An experimental and numerical study on the face
milling of Ti-6Al-4V alloy: tool performance and surface integrity. J Mater Process Technol
211:294–304. https://doi.org/10.1016/j.jmatprotec.2010.10.002
74. Li Y, Yang Y, Wang Y, Gao F (2022) Study on the simulation and experiment of ultrasonic-
assisted vibration drilling of Ti6Al4V. J Phys Conf Ser 2242.https://doi.org/10.1088/1742-
6596/2242/1/012011
75. Sahu NK, Andhare AB (2019) Prediction of residual stress using RSM during turning of
Ti–6Al–4V with the 3D FEM assist and experiments. SN Appl Sci 1:1–14. https://doi.org/10.
1007/s42452-019-0809-5
76. Tzotzis A, Efkolidis N, Oancea G, Kyratsis P (2021) FEM-based comparative study of
square & rhombic insert machining performance during turning of AISI-D3 steel. Int J Mod
Manuf Technol 13:143–148. https://doi.org/10.54684/ijmmt.2021.13.2.143
77. Zorev NN (1963) Inter-relationship between shear processes occurring along tool face and
shear plane in metal cutting. Int Res Prod Eng 49:143–152
78. Usui E, Shirakashi T (1982) Mechanics of machining—from descriptive to predictive theory,
on the art of cutting metals 75 years later. ASME PED 7:13–35
79. Merchant ME (1945) Mechanics of the metal cutting process. II. Plasticity conditions in
orthogonal cutting. J Appl Phys 16:318–324
80. Albrecht P (1960) New developments in the theory of the metal-cutting process: Part I. The
ploughing process in metal cutting. J Eng Ind 82:348–357. https://doi.org/10.1115/1.3664242
81. Childs THC, Maekawa K, Obikawa T, Yamane Y (2000) Metal machining: theory and
applications. Butterworth-Heinemann
82. Iwata K, Osakada K, Terasaka Y (1984) Process modeling of orthogonal cutting by the rigid-
plastic finite element method. J Eng Mater Technol 106:132–138. https://doi.org/10.1115/1.
3225687
83. Sekhon GS, Chenot JL (1993) Numerical simulation of continuous chip formation during
non-steady orthogonal cutting. Eng Comput 10:31–48
84. Yang X, Liu CR (2002) A new stress-based model of friction behavior in machining and its
significant impact on residual stresses computed by finite element method. Int J Mech Sci
44:703–723
85. Zemzemi F, Rech J, Ben SW et al (2008) Identification of a friction model at

tool/chip/workpiece interfaces in dry machining of AISI4142 treated steels. 9:3978–3990.
https://doi.org/10.1016/j.jmatprotec.2008.09.019
86. Sun Y, Chen T, Qiong C, Shafai C (2016) A comprehensive experimental setup for identifica-
tion of friction model parameters. Mech Mach Theory 100:338–357. https://doi.org/10.1016/
j.mechmachtheory.2016.02.013
87. Behera BC, Ghosh S, Rao PV (2018) Modeling of cutting force in MQL machining environ-
ment considering chip tool contact friction. Tribol Int 117:283–295. https://doi.org/10.1016/
j.triboint.2017.09.015
88. Palanisamy NK, Lorphèvre ER, Gobert M et al (2022) Identification of the parameter values
of the constitutive and friction models in machining using EGO algorithm: application to
Ti6Al4V. Metals (Basel) 12:1–21
89. Arrazola PJ, Ugarte D, Domínguez X (2008) A new approach for the friction identification
during machining through the use of finite element modeling. Int J Mach Tools Manuf 48:173–
183. https://doi.org/10.1016/j.ijmachtools.2007.08.022
90. Davoudinejad A, Parenti P, Annoni M (2017) 3D finite element prediction of chip flow, burr
formation, and cutting forces in micro end-milling of aluminum 6061–T6. Front Mech Eng
91. Asad M, Mabrouki T, Ijaz H et al (2014) On the turning modeling and simulation: 2D and 3D
FEM approaches. Mech Ind 15:427–434. https://doi.org/10.1051/meca/2014045
92. Hu HJ, Huang WJ (2014) Tool life models of nano ceramic tool for turning hard steel based
on FEM simulation and experiments. Ceram Int 40:8987–8996. https://doi.org/10.1016/j.cer
amint.2014.01.095
93. Attanasio A, Ceretti E, Rizzuti S et al (2008) 3D finite element analysis of tool wear in
machining. CIRP Ann Manuf Technol 57:61–64. https://doi.org/10.1016/j.cirp.2008.03.123
94. Attanasio A, Ceretti E, Giardini C (2009) 3D FE modelling of superficial residual stresses
in turning operations. Mach Sci Technol 13:317–337. https://doi.org/10.1080/109103409032
37806
95. Ucun İ, Aslantas K, Özkaya E, Cicek A (2017) 3D numerical modelling of micro-milling
process of Ti6Al4V alloy and experimental validation. Adv Mater Process Technol 3:250–260.
https://doi.org/10.1080/2374068X.2016.1247343
96. Johnson GR, Cook WH (1985) Fracture characteristics of three metals subjected to various
strains, strain rates, temperatures and pressures. Eng Fract Mech 21:31–48
97. Murugesan M, Dong-Won J (2019) Johnson Cook material and failure model parameters esti-
mation of AISI-1045 medium carbon steel for metal forming applications. Materials (Basel)
12:1–18. https://doi.org/10.3390/ma12040609
98. Bal B, Karaveli KK, Cetin B, Gumus B (2019) The precise determination of the Johnson–
Cook material and damage model parameters and mechanical properties of an aluminum
7068-T651 alloy. J Eng Mater Technol 141. https://doi.org/10.1115/1.4042870
99. Zhang D, Shangguan Q, Xie C, Liu F (2015) A modified Johnson–Cook model of dynamic
tensile behaviors for 7075–T6 aluminum alloy. J Alloys Compd 619:186–194. https://doi.org/
10.1016/j.jallcom.2014.09.002
100. Stopel M, Skibicki D (2018) Determination of the Johnson-Cook damage parameter D 4 by
Charpy impact testing. Mater Test 60:974–978
101. Banerjee A, Dhar S, Acharyya S et al (2015) Determination of Johnson cook material and
failure model constants and numerical modelling of Charpy impact test of armour steel. Mater
Sci Eng A 640:200–209. https://doi.org/10.1016/j.msea.2015.05.073
102. Wang X, Shi J (2013) International journal of impact engineering validation of Johnson-Cook
plasticity and damage model using impact experiment. Int J Impact Eng 60:67–75. https://
doi.org/10.1016/j.ijimpeng.2013.04.010
103. Cockcroft MG, Latham DJ (1968) Ductility and the workability of metals. J Inst Met 96:33–39
104. Razanica S, Malakizadi A, Larsson R et al (2020) FE modeling and simulation of machining
Alloy 718 based on ductile continuum damage. Int J Mech Sci 171:105375. https://doi.org/
10.1016/j.ijmecsci.2019.105375
References 45
105. McClintock FA (1968) A criterion for ductile fracture by the growth of holes. J Appl Mech
35:363–371. https://doi.org/10.1115/1.3601204
106. Rice JR, Tracey DM (1969) On the ductile enlargement of voids in triaxial stress fields. J
Mech Phys Solids 17:201–217
107. Brozzo P, Deluca B, Rendina R (1972) A new method for the prediction of formability limits
in metal sheets. In: Proc. 7th biennal Conf. IDDR
108. Chandrakanth S, Pandey PC (1995) An isotropic damage model for ductile material. Eng
Fract Mech 50:457–465
109. Obikawa T, Sasahara H, Shirakashi T, Usui E (1997) Application of computational machining
method to discontinuous chip formation. J Manuf Sci Eng 119:667–674. https://doi.org/10.
1115/1.2836807
110. He YL, Davim JP, Xue HQ (2018) 3D progressive damage based macro-mechanical FE
simulation of machining unidirectional FRP composite. Chin J Mech Eng. https://doi.org/10.
1186/s10033-018-0250-5
111. Zhang YC, Mabrouki T, Nelias D, Gong YD (2011) Chip formation in orthogonal cutting
considering interface limiting shear stress and damage evolution based on fracture energy
approach. Finite Elem Anal Des 47:850–863. https://doi.org/10.1016/j.finel.2011.02.016
112. Archard JF (1953) Contact and rubbing of flat surfaces. J Appl Phys 24:981–988. https://doi.
org/10.1063/1.1721448
113. Palanikumar K, Paulo Davim J (2007) Mathematical model to predict tool wear on the
machining of glass fibre reinforced plastic composites. Mater Des 28:2008–2014. https://
doi.org/10.1016/j.matdes.2006.06.018
114. Marksberry PW, Jawahir IS (2008) A comprehensive tool-wear/tool-life performance model
in the evaluation of NDM (near dry machining) for sustainable manufacturing. Int J Mach
Tools Manuf 48:878–886. https://doi.org/10.1016/j.ijmachtools.2007.11.006
115. Takeyama H, Murata R (1963) Basic investigation of tool wear. J Eng Ind 85:33–37. https://
doi.org/10.1115/1.3667575
116. Hao Z, Gao D, Fan Y, Han R (2011) New observations on tool wear mechanism in dry
machining Inconel718. Int J Mach Tools Manuf 51:973–979. https://doi.org/10.1016/j.ijm
achtools.2011.08.018
117. Zhang SJ, To S, Zhang GQ (2017) Diamond tool wear in ultra-precision machining. Int J Adv
Manuf Technol 88:613–641. https://doi.org/10.1007/s00170-016-8751-9
118. Peng B, Bergs T, Schraknepper D et al (2019) A hybrid approach using machine learning to
predict the cutting forces under consideration of the tool wear. Procedia CIRP 82:302–307.
https://doi.org/10.1016/j.procir.2019.04.031
119. Wang J, Li Y, Zhao R, Gao RX (2020) Physics guided neural network for machining tool
wear prediction. J Manuf Syst 57:298–310. https://doi.org/10.1016/j.jmsy.2020.09.005
120. Xu X, Wang J, Zhong B et al (2021) Deep learning-based tool wear prediction and its applica-
tion for machining process using multi-scale feature fusion and channel attention mechanism.
Meas J Int Meas Confed 177:109254. https://doi.org/10.1016/j.measurement.2021.109254
121. Li Y, Wang J, Huang Z, Gao RX (2022) Physics-informed meta learning for machining tool
wear prediction. J Manuf Syst 62:17–27. https://doi.org/10.1016/j.jmsy.2021.10.013
122. Seeholzer L, Krammer T, Saeedi P, Wegener K (2022) Analytical model for predicting tool
wear in orthogonal machining of unidirectional carbon fibre reinforced polymer (CFRP).
Springer, London
123. Usui E, Shirakashi T, Kitagawa T (1984) Analytical prediction of cutting tool wear. Wear
100:129–151. https://doi.org/10.1016/0043-1648(84)90010-3
124. Melkote SN, Grzesik W, Outeiro J et al (2017) Advances in material and friction data for
modelling of metal machining. CIRP Ann Manuf Technol 66:731–754. https://doi.org/10.
1016/j.cirp.2017.05.002
125. Oezkaya E, Biermann D (2017) Segmented and mathematical model for 3D FEM tapping
simulation to predict the relative torque before tool production. Int J Mech Sci 128–129:695–
708. https://doi.org/10.1016/j.ijmecsci.2017.04.011
126. Wei L, Wang D (2019) Comparative study on drilling effect between conventional drilling and
ultrasonic-assisted drilling of Ti-6Al-4V/Al2024-T351 laminated material. Int J Adv Manuf
Technol 103:141–152. https://doi.org/10.1007/s00170-019-03507-6
127. Ucun İ (2016) 3D finite element modelling of drilling process of Al7075-T6 alloy and exper-
imental validation. J Mech Sci Technol 30:1843–1850. https://doi.org/10.1007/s12206-016-
0341-0
128. Paktinat H, Amini S (2017) Ultrasonic assistance in drilling: FEM analysis and experimental
approaches. Int J Adv Manuf Technol 92:2653–2665. https://doi.org/10.1007/s00170-017-
0285-2
129. Ezilarasan C, Senthil VS, Velayudham A (2014) heoretical predictions and experimental
validations on machining the Nimonic C-263 super alloy. Simul Model Pract Theory 40:192–
207. https://doi.org/10.1016/j.simpat.2013.09.008
130. Malakizadi A, Gruber H, Sadik I, Nyborg L (2016) An FEM-based approach for tool wear
estimation in machining. Wear 368–369:10–24. https://doi.org/10.1016/j.wear.2016.08.007
Chapter 3
FEM-Based Study of AISI52100 Steel
Machining: A Combined 2D and 3D
Approach
Abstract In this research, the cutting forces that are induced during AISI52100
turning and the geometrical characteristics of the generated chips are studied by
means of combined 2D and 3D Finite Element (FE) analyses. A number of 2D
simulation tests were performed according to a design of experiments in order to
evaluate the accuracy of three friction models, as well as a flow stress model at a
given range of cutting conditions. Upon the comparison of the numerical results with
the equivalent experimental ones, in terms of the thrust and cutting force, the model
with the best fit was selected and, consequently, the establishment of a 3D FE model
was achieved based on the 2D concepts. The three force components, as well as the
chip’s geometrical aspects, were studied and visualized.
Keywords AISI52100 machining · FEM · DEFORM™ · ANSYS™ · Thrust

force · Cutting force · Chip geometry
3.1 Introduction
Nowadays, it is evident that the utilization of Finite Element Method (FEM) in

machining is an integral part of the studying procedure. A range of studies that
are available in the bibliography, indicate that FEM is an indispensable tool for
the accurate investigation of several parameters that are involved in machining and
manufacturing, that in some cases would not be possible to study otherwise. The
minimization of the experimental testing, the cost reduction, and the repeatability,
are only some of the benefits that are expected from the use of FEM.
Typical manufacturing processes such as drilling, milling, and turning, as well
as more modern operations such as Electrical Discharge Machining (EDM), can be
studied with the aid of FEM, for a variety of materials. Three-dimensional Finite
Element (FE) studies specifically seem to gain ground compared to 2D works.
However, it is noted that two-dimensional tools are still used by many researchers,
since they are considered to be effective for a wide range of applications, as well as
time efficient. Material plastic behavior and friction modeling are two of the most

48 3 FEM-Based Study of AISI52100 Steel Machining: A Combined 2D …
important parameters that greatly affect the numerical results and are usually thor-
oughly investigated by the researchers. The most common material flow stress model
is the one proposed by Johnson and Cook [1], applied to virtually any process. Several
other models are also used, either modified or with minor adjustments. On the other
hand, friction modeling is an aspect that is frequently debated due to the nature of
the phenomena that occur during machining. Coulomb, shear, or hybrid models are
typically used with constant coefficients after a calibration procedure, while, in some
cases, more complex models are coupled with variable coefficients.
Xiong et al. [2] investigated the performance of a metal matrix composite in
terms of typical cutting and chip geometric parameters. The study was based on
a 2D concept with Johnson–Cook flow stress modeling. The work by Salvati and
Korsunsky [3] focused on the characterization of the residual stresses yielded during
the EDM cutting of AA6082-T6. The authors performed two passes, one main cut
and one trim cut, examined the surface morphology, and carried out the equivalent
experiment in a simulated environment. Liu et al. [4] evaluated the performance of
microtextured tools during machining of stainless steel 17-4PH. A 3D FE model
was developed for this purpose and several tool textures were tested. In addition
to the cutting forces, chip morphology was also studied. Similarly, Lotfi et al. [5]
investigated the ultrasonic-assisted turning of AISI4140 steel by using a 3D model
and performing experimental work. The authors focused on the tool wear analysis of
the process. Kyratsis et al. [6] investigated the influence of typical conditions on the
cutting forces induced while turning AISI-D2 tool steel, simulated by a 3D FE model.
Studies related to drilling, usually focus on investigating the generated thrust force,
cutting torque, chip characteristics and evolution, residual stresses, temperatures, tool
wear, built-up-edge (BUE), and more, as can be seen in the next papers [7–12]. In a
similar manner, investigations that are based on milling processes [13–17], tend to
examine the machinability of industry-related materials, the parameters that affect the
performance of the cutting tools and the performance of the available cutting fluids
and cooling methods in general. In most cases, the ultimate purpose of examining
the influence of several parameters on the process, is the prediction of these factors
and the development of models that can forecast the behavior of the tool-workpiece
system.
Current work presents a 3D FE model for AISI52100 steel machining, with the
combined implementation of a CAD-based and a two-dimensional FE setup. The
first method enabled the quick and accurate derivation of the tool geometry aspects,
as well as the machining area of interest, whereas the latter, allowed the relatively
straightforward calibration of the rheological and the friction model by comparing
the generated 2D simulation runs results, with equivalent experimental ones. The
summarized workflow of the study can be seen in Fig. 3.1.
3.2 Materials and Methods 49
Fig. 3.1 The workflow of the present study
3.2 Materials and Methods
3.2.1 Machining Process Framework
A CAD-based setup resembling the tool-part interaction was utilized prior to the
development of the three-dimensional FE model, in order to facilitate the extraction
of the required parameters that affect the process, as well as to verify the posi-
tioning. Setups that rely on CAD systems are an effective way to design and plan
the machining operation without the need for time-consuming and costly experi-
mental setups. Moreover, such a setup allows the user to effortlessly change any
parameter that participates in the process. In addition, basic assembly and collision
detection features are available in most CAD systems, including freeware versions.
Fig. 3.2 The CAD-based framework of the machining process
On the other hand, experimental working is of course an integral part of the modeling
procedure, since it is required for calibration purposes.
The under-study process is the external longitudinal turning of a Ø56 cylindrical
bar made by AISI52100 steel, with Cubic Boron Nitride (CBN) tools. The ISO code
for the tool is SNGA120408T01020, meaning that it is a square-shaped negative
insert, with a 0.8 mm nose radius and honed cutting edge with 0.1 mm width and
20° angle. The equivalent number for the toolholder is PSBNR2525M12. Turning
with standardized tools and toolholders provides several advantages such as accu-
rately fixed cutting angles and interchangeability. The cutting angles were extracted
during the CAD modeling procedure; the tool cutting edge angle (KAPR) is 75°, the
clearance angle is 6° and, finally, both the rake (GAMO) and the inclination (LAMS)
angle are equal to −6°. Figure 3.2a illustrates the CAD assembly for the workpiece,
the tool, and its toolholder. The model of the toolholder was downloaded, since it is
available online by many manufacturers, whereas both the insert and the work were
modeled in SolidWorks™ 2021. Especially the tool model was designed with the aid
of an applet [18]. Figure 3.2b depicts the most important geometric characteristic of
the tool.
3.2.2 Preliminary FE Model Assessment
The machining parameters applied to the current study were chosen according to the
recommendations of several manufacturers for this type of inserts, that are intended
for hardened material cutting. Therefore, the selected values of cutting speed are 125,
176, and 246 m/min, respectively, the feed is equal to 0.08, 0.12, and 0.16 mm/rev
accordingly, and finally the depth of cut is 0.45 mm for all tests, as seen in Table 3.1.
Table 3.1 The cutting

Level V c (m/min) f (mm/rev) ap (mm)
parameters and the levels
used in the study +1 246 0.16 0.45
0 176 0.12 0.45
−1 125 0.08 0.45
The nine tests correspond to the three levels of the two factors. However, several 2D
FE tests were carried out before attempting to run the 3D tests, in order to calibrate
the model.
The two-dimensional simulations were choseThe cutting parameten for the short
run time and minimal preparation work. Figure 3.3 illustrates the generalized 2D
model used in this study, along with the boundary conditions. The two-dimensional
setup was developed in the ANSYS™ module, Explicit Dynamics 2021 R2. Each
test was allowed to run until the steady state was achieved, thus up to approximately
1 mm, and lasted for about one hour and a half. This way, excessive chip accumulation
in the cutting edge of the tool was avoided, making the comparison of the thickness
between the 2D and 3D results possible, as discussed later in Sect. 3.3.2. Both the
rheological and friction models were calibrated according to experimental data that
are available in the literature [19], with the following method.
Flow stress modeling
A tabular data format was used in this study, by extracting the values of flow stress
from the diagram of Fig. 3.4, which is tuned for the machining of AISI52100, at
similar conditions [20]. This way the flow stress is represented as a function of strain,
strain rate, and temperature. The tabular data format is considered to be a robust
method for representing the material flow stress, as long as an adequate number of
data points are available. Each curve of strain rate was converted to a column in the
table and each strain value to a line. Thus, the flow stress value that corresponds to
the points of strain (0.05, 0.1, 0.2, 0.4, 0.7, 1.0, 2.0, and 5.0) for each strain rate
Fig. 3.3 The 2D FE model setup

Fig. 3.4 The flow stress

chart for AISI52100 steel at
20 °C
curve at the given temperature, were extracted. Minor adjustments were made by
performing a small number of 2D simulations, to tune the simulated values as close
to the test results as possible, for the applied depth of cut.
The properties, regarding mechanical and thermal behavior, of the tool-work mate-
rials, are presented in Table 3.2, where the thermal properties for the workpiece are
expressed with respect to the temperature. It should be noticed that the tool was set to
behave as rigid, however, its thermal properties were set so that the thermal analysis
can be carried out.
Friction modeling
In a similar manner to the flow stress calibration, friction model was calibrated
according to the experimental results. Due to the fact that in 2D simulations feed force
is neglected, the comparison was done between the experimental and the simulated
thrust (F p ) and cutting (F c ) forces. At first, two sets that contain nine tests each were
carried out to examine the influence of the shear and the Coulomb friction coefficient,
respectively. The three levels for each coefficient are depicted in Table 3.3. It is
noted that the initial coefficients were selected based on the default values of several
software used in FEA, such as DEFORM™, LS-DYNA™, and ABAQUS™, and
the literature [22]. Furthermore, the number of simulations for each set was defined
according to an L9 Taguchi orthogonal array design, instead of the full factorial
design that implements 27 runs, to reduce the workload without compromising the
fidelity of the model. Table 3.4 includes the tests in their run order and the parameters
involved.
A comparison between the results was performed upon completion of the tests. In
general, the runs yielded results with underestimated cutting forces. Moreover, the
Table 3.2 Standard

Mechanical properties AISI52100 CBN
properties for the AISI52100
work [21] and CBN tool [22] E [GPa] 210 Rigid
ρ [kg/m3 ] 7,850
ν 0.30
Thermal properties AISI52100 CBN
C [J/kg K] 278 → 93 °C 20,000
324 → 316 °C
579 → 649 °C
718 → 871 °C
α [°C−1 ] 11.9 × 10−6 4.5
k [W/mK] 24.57 → 149 °C 60
24.4 → 349 °C
24.23 → 477 °C
24.75 → 604 °C
Table 3.3 The friction

Level m μ
coefficients used in the
preliminary tests +1 0.8 0.50
0 0.6 0.35
−1 0.4 0.20
Table 3.4 The 2D

Run V c (m/min) f (mm/rev) m µ
preliminary tests
1 125 0.08 0.4 0.20
2 125 0.12 0.6 0.35
3 125 0.16 0.8 0.50
4 176 0.08 0.6 0.35
5 176 0.12 0.8 0.50
6 176 0.16 0.4 0.20
7 246 0.08 0.8 0.50
8 246 0.12 0.4 0.20
9 246 0.16 0.6 0.35
initial value of both the shear and the Coulomb coefficient predicted the cutting forces
more accurately. Afterward, an evaluation of a hybrid model was done by selecting
the coefficient values that fitted the experimental values the best. The conclusion was
that a hybrid model with a shear constant equal to m = 0.6 and a coefficient of μ =
0.35 generated error, equal to 18.7%, being the lowest value. In order to evaluate the
coefficients, the relative error between the experimentally generated cutting forces
and the FE results was calculated for all the possible combinations and utterly the
Fig. 3.5 Comparison graph for the different friction models and coefficient values tested
total resultant error, expressed in percentage, was generated with the aid of Eq. 3.1.
Where ∆x is the relative error calculated by the numerical and the experimentally
acquired value of thrust force and cutting force, a represents the number of the runs
for each combination (i.e., a = 3 for friction coefficient m = 0.6) and, finally, b is the
number of the cutting force components, which is equal to two. Figure 3.5 visualizes
the comparison between the selected coefficients by presenting the total error.
|
|∑
| a ∑ b
| ∆x 2
| i=1 j=1
err ortotal = × 100 (3.1)
a×b
3.2.3 Numerical Modeling of the Turning Process in Three

Dimensions
For the development of the 3D-FE model, DEFORM™-3D ver. 12.0 was utilized,
which implements the Lagrangian modeling method coupled with an embedded
remeshing technique for the treatment of the mesh distortion during the chip gener-
ation procedure. The criterion applied during the remeshing affects the size control
according to the proportions of the elements around the distorted mesh.
It is a common practice to simplify the area where the cutting process takes
place in order to shorten the simulation run time. The CAD-based setup allowed the
extraction of a simplified version of the workpiece (Fig. 3.6a). Therefore, the work
was modeled as an arc-shaped, one-eighth fraction of the full bar. To further improve
the simulation time, a predefined cut surface was shaped on the workpiece model.
Regarding the behavior of the models, the workpiece was set to be deformable and
approximately 46,000–91,000 tetrahedron elements comprised its mesh, based on the
applied value of feed, whereas the tool was set to be non-deformable and discretized
with a high number of elements, approximately 50,000. Since the tool was set to act
rigidly, the mesh size had a minimal effect on the results. However, a denser mesh on
the tool and especially on the cutting tip produces a better visual representation of the
process, without sacrificing computational resources [20]. The size of the smallest
element in the workpiece’s mesh was defined to be 25% of the value of feed [23],
thus ranging between 0.02 and 0.04 mm for this study. Furthermore, to optimize the
mesh at the contact area, the small-to-large size ratio of mesh elements was set to
7:1 (Fig. 3.6b) for the work and 4:1 (Fig. 3.6c) for the tool, respectively.
As a result of the simplification process, the workpiece did not rotate around axis
Z as it would in real time, rather it was fixed allowing the tool to follow the trajectory
illustrated in Fig. 3.6a to achieve the cut. Finally, to represent the heat exchange of the
workpiece-tool system, due to both convection and conduction, with the environment
(T ambient = 20 °C), the equivalent coefficients were applied. For dry cutting, their
values are as follows, hconv = 20 W/(m2 × °C) and hcond = 4.5 × 104 W/(m2 × °C),
respectively [21].
To approximate the damage process, where the material separation occurs, the
normalized Cockcroft-Latham criterion [24] was applied. Equation 3.2 represents
the criterion, which in general is a simplified formula that relates the material damage
constant to both the maximum and the effective stress.
Fig. 3.6 The 3D FE model setup; the analysis domain (a), the meshed part (b), and the tool (c)
/ε f
σmax
Dc = dε pl (3.2)
σ
0
Dc represents a material constant that enables the fracturing, σ max denotes the
maximum tensile stress and σ̃ the effective stress. The limit of the integral, εf repre-
sents the limit fracture strain, whereas εpl the plastic strain. In addition, the fraction
of the plastic work that is converted to heat, was set to 90%, so that the remaining
10% was allowed to be stored in the workpiece.
3.3 Results and Findings
3.3.1 Machining Forces Evaluation
The average run time of each of the 3D simulations was about 10 h with a deviation
of a couple of hours depending on the feed value, on an AMD Ryzen 3.6 GHz CPU,
96 GB RAM, and SSD technology hard disk. It was observed that the fraction of
the workpiece used, was enough to allow the steady state to occur in a total 8 ms
cut. Therefore, the force values versus the time were extracted for each of the runs
and the mean value was calculated for all three force components after applying a
first-order exponential smoothing on the time series. Next, a comparison was made
with the experimentally acquired values. The comparison graphs of Fig. 3.7 visualize
the results of the three cutting-force elements. In specific, Fig. 3.7a illustrates the
results for the thrust force, Fig. 3.7b for the cutting force, and Fig. 3.7c for the force
at the feed direction. It is evident that the agreement levels are high, especially for
the cutting force, which is the component that contributes the most to the resultant
force. The relative error comparing the numerical results with the experimental range
from −11.2 to 26.6% for the thrust force F p , −5.9 to 16.9% for the cutting force
F c, and 14.6 to 28.7% for the feed component F f . The relatively increased error that
is observed among the feed values is possibly related to the fact that the calibration
was done based on 2D tests. As already discussed, two-dimensional tests neglect the
feed force, thus an underestimation is expected. In any case, the feed force is the
component with the smallest degree of contribution to the resultant machining force.
Regarding the 3D tests, it is noted that they were run in the same order as the
2D ones. By studying Fig. 3.7, the following observations can be made. First, the
cutting force F c contributes the most to the resultant force F resultant , as proven by
solving Eq. 3.3 and then calculating the ratio between the squares of F c and F resultant .
The contribution percentage varied between 67.5 and 76.6%. As an example, thrust,
cutting, and feed force for numerical test number 2 are equal to approximately 169N,
358N, and 103N, respectively. Thus, the resultant force is equal to 409.1N, yielding
a contribution percentage for F c close to 76.6%.
3.3 Results and Findings 57
Fig. 3.7 Comparison graphs between the experimental and the simulated machining forces; thrust
force (a), cutting force (b), and feed force (c)
(
Fresultant = F p2 + Fc2 + F 2f (3.3)
Next, it is obvious that the feed acts increasingly on all machining force compo-
nents. On the contrary, any changes in the cutting speed values do not affect the gener-
ated forces significantly. In fact, both thrust and feed force exhibit light changes in
their values, with cutting force being an exception, since it is evident that an increase
in cutting speed reduces cutting force value noticeably. Moreover, the most impactive
feed value is equal to 0.16 mm/rev. For instance, the percentage of increase for the
cutting force, when shifting from 0.08 to 0.16 mm/rev, reaches 15.2, 23.2, and 47.4%
for cutting speed equal to 125, 176, and 246 m/min accordingly.
3.3.2 Chip Geometry Evaluation
Chip morphology possesses a key role in determining material machinability and

tool-wear prediction when machining. The geometric characteristics of the gener-
ated chip enable the understanding of the influence of several parameters on the
machining operation. Feed, cutting speed, depth-of-cut, and cutting angles are some
of the most important factors that affect the performance of the tool and the overall
machinability. FEM allows the modeling of the machining operation with respect to
the aforementioned parameters, for prediction purposes of several results that some
decades ago could not be predicted with another way, except with experimental
working.
In the present study, the chip formation and morphology were examined with the
aid of both the 2D and 3D simulations and evaluated in terms of the agreement level
between the numerical values and analytical calculations. Specifically, the deformed
chip thickness hc and the chip width b, as well as the shear angle ϕ were the parameters
that were evaluated. The chip thickness was measured directly in the 2D FE model
and indirectly for the 3D model as described later on, whereas the width was measured
only in the three-dimensional model. Additionally, the shear angle was calculated
with Eq. 3.4 [25] and measured with the aid of the 3D model as well. The shear angle
ϕ, usually expressed in degrees, is determined by the rake angle γ and the cutting
ratio r c , which essentially is the ratio of the undeformed chip thickness h0 to the
deformed hc . Thus, it is valid that rc = h 0 / h c .
( )
rc cos γ
ϕ = tan−1 (3.4)
1 − rc sin γ
The chip thickness can be expressed in mm as well, but it is common to use

µm. To determine the undeformed chip thickness with respect to feed f and major
cutting-edge angle κ r , it is possible to use Eq. 3.5 [26].
h 0 ≈ f sin κr (3.5)
Fig. 3.8 The chip geometry evaluation schematic; chip width in 3D (a), chip thickness in 3D (b),
and chip thickness, as well as shear angle in 2D (c)
To avoid assuming that the chip width is constant across its length, assumption
that is common for 2D strategies, the width was measured between multiple opposite
nodes of the chip (Fig. 3.8a) at approximately 1 mm length of cut and then the
mean value was calculated. In a similar manner, the chip thickness was measured on
multiple points across its section to determine the mean value.
To facilitate the measurement and ensure adequate accuracy, the workpiece model
was exported to the CAD system (Fig. 3.8b). The machined workpiece model was
sectioned in the middle of the chip’s width, across the XY plane, revealing this way
the planar chip morphology. By observing Fig. 3.8b it is evident that the curling of
the chip is similar to the one of the 2D simulation. In addition, the chip thickness
was measured in the 2D FE model, by recording a number of measurements along
the periphery of the prescribed circle that was formed around the curling of the
chip (Fig. 3.8c). Finally, the shear angle was measured with the aid of the two-
dimensional model as seen in Fig. 3.8c. Moreover, it was calculated with Eq. 3.4 by
using as inputs the known rake angle, the cutting-edge angle, and the feed, in addition
to the simulated chip thickness. It is noted that the measurements and calculations
of Fig. 3.8 regard the conditions used in test number 2.
To visualize the comparison between the generated chip thickness and the shear
angle, the results and calculations were summarized in the charts of Fig. 3.9. Chip
width was excluded from the comparison study since it did not exhibit any particular
interest. The reason behind this decision is the fact that the parameters that affect the
chip width such as the cutting-edge angle, the depth-of-cut, and the tool-nose radius
remain constant for all tests, thus no significant variation was expected in the results.
Fig. 3.9 The deformed chip thickness graph (a) and the shear angle graph (b)
Figure 3.9a illustrates the chip thickness yielded in both 2D and 3D simulations.
It is evident that the values exhibit a high level of accordance, meaning that the two-
dimensional tests are ideal when a quick assessment of the chip thickness is required.
As shown in the graph, it is concluded that the feed has a significant impact on the
chip thickness, acting increasingly. The same applies to the cutting speed, but the
level of influence is lower compared to the effect of the feed. Mhamdi et al. [27]
reported similar findings during their experimental work under cutting conditions of
the same magnitude (feed between 0.05 and 0.2 mm/rev and cutting speed ranging
from 50 to 250 m/min) for AISI-D2 steel. Specifically, the authors determined the
chip thickening ratio, which increases with the rise of the feed value. An increasing
trend is also present in Fig. 3.9b, where it is shown that the shear angle is influenced
increasingly by higher values of feed. Despite the presence of a few wider values of
error in this diagram, the trend remains clear.
References 61
3.4 Conclusions
Concluding, the present study highlighted the importance of the establishment of a

three-dimensional FE model as well as the value of 2D simulations and CAD-based
tools when studying the effects of the cutting parameters in machining. Moreover,
evaluated a range of friction coefficients and their influence on the development of a
numerical model. In terms of the generated machining forces and the chip geometry,
some useful conclusions can be drawn.
• The FE model yielded results close to the experimental findings with adequate
levels of error.
• All machining force components are greatly affected by higher levels of feed. This,
however, is not true for the cutting speed since its effect is marginal, especially
for both the thrust and feed force. The effect on the cutting force, on the other
hand, is slightly more significant.
• Especially the feed value equal to 0.16 mm/rev boosts the forces significantly.
• Furthermore, it is shown that the cutting force F c (tangential) is the dominant
component.
• Finally, both the feed and the cutting speed act increasingly on the produced
deformed chip thickness, especially the feed.
• The same applies to the estimated shear angle, with the exception of the cutting
speed, which seems to have a negligible effect on the angle.
References
1. Johnson GR, Cook WH (1985) Fracture characteristics of three metals subjected to various
strains, strain rates, temperatures and pressures. Eng Fract Mech 21:31–48
2. Xiong Y, Wang W, Jiang R et al (2018) Mechanisms and FEM simulation of chip formation in
orthogonal cutting in-situ TiB2/7050Al MMC. Materials (Basel) 11:1–19. https://doi.org/10.
3390/ma11040606
stresses in AA6082- T6 Al alloy generated by wire EDM cutting. J Mater Process Tech 275:1–
12. https://doi.org/10.1016/j.jmatprotec.2019.116373
4. Liu G, Huang C, Su R et al (2019) 3D FEM simulation of the turning process of stainless steel
17–4PH with differently texturized cutting tools. Int J Mech Sci 155:417–429. https://doi.org/
10.1016/j.ijmecsci.2019.03.016
010004
7. Kheireddine AH, Lu T, Jawahir IS, Hamade RF (2013) An FEM analysis with experimental
validation to study the hardness of in-process cryogenically cooled drilled holes in Mg. 8:588–
593. https://doi.org/10.1016/j.procir.2013.06.156

1051/meca/2020046
10. Gao X, Li H, Liu Q et al (2011) Simulation of stainless steel drilling mechanism based on
Deform-3D. Adv Mater Res 160–162:1685–1690. https://doi.org/10.4028/www.scientific.net/
AMR.160-162.1685
machining forces during AISI 4140 Hard Turning. Strojniški Vestn. J Mech Eng 66:467–478.
14. Soo SL, Dewes RC, Aspinwall DK (2010) 3D FE modelling of high-speed ball nose end
milling. Int J Adv Manuf Technol 50:871–882. https://doi.org/10.1007/s00170-010-2581-y
15. Afazov SM, Ratchev SM, Segal J (2012) Prediction and experimental validation of micro-
milling cutting forces of AISI H13 steel at hardness between 35 and 60 HRC. Int J Adv Manuf
Technol 62:887–899. https://doi.org/10.1007/s00170-011-3864-7
16. Man X, Ren D, Usui S et al (2012) Validation of finite element cutting force prediction for end
milling. Procedia CIRP 1:663–668. https://doi.org/10.1016/j.procir.2012.05.019
17. Li S, Sui J, Ding F et al (2021) Optimization of milling aluminum alloy 6061–T6 using modified
Johnson-Cook model. Simul Model Pract Theory 111:102330. https://doi.org/10.1016/j.sim
pat.2021.102330
18. Tzotzis A, García-Hernández C, Huertas-Talón JL, Kyratsis P (2020) CAD-based automated
design of FEA-ready cutting tools. J Manuf Mater Process 4:1–14. https://doi.org/10.3390/
jmmp4040104
19. Bouacha K, Athmane M, Mabrouki T, Rigal J (2010) Statistical analysis of surface roughness
and cutting forces using response surface methodology in hard turning of AISI 52100 bearing
steel with CBN tool. Int J Refract Met Hard Mater 28:349–361. https://doi.org/10.1016/j.ijr
mhm.2009.11.011
turning: modelling of cutting forces and cutting condition optimization. Machines 10:74. https://
doi.org/10.3390/machines10020074
21. Scientific Forming Technologies Corporation (2016) DEFORM V11.3 (PC) Documentation
22. Arrazola PJ, Özel T (2008) Numerical modelling of 3D hard turning using arbitrary Lagrangian
Eulerian finite element method. Int J Mach Mach Mater 3:238–249
24. Cockcroft MG, Latham DJ (1968) Ductility and the workability of metals. J Institue Met
96:33–39
25. El-Hofy H (2018) Fundamentals of machining processes: conventional and nonconventional
processes. CRC Press
26. Agmell M (2018) Applied FEM of metal removal and forming, 1st edn. Studentlitteratur, Lund
27. Mhamdi MB, Ben SS, Boujelbene M, Bayraktar E (2013) Experimental study of the chip
morphology in turning hardened AISI D2 steel. J Mech Sci Technol 27:3451–3461. https://
doi.org/10.1007/s12206-013-0869-1
Chapter 4
Experimental and 3D Numerical Study
of AA7075-T6 Drilling Process
Abstract Finite Element Method (FEM) in machining is a technique that is widely

accepted by the research community in the past few years, because it offers increased
accuracy, robust results, and simplified test repeatability. The 3D-FE modeling of
7075-T6 aluminium alloy drilling is being presented in this study, with the use of
commercial Finite Element Analysis (FEA) software, namely DEFORMTM . The
drilling process was simulated according to typical cutting parameters such as cutting
speed and feed. The approximation of the process was achieved by implementing
the most critical aspects, including flow stress of the material, tool geometry, friction
behavior, and proper meshing. Prior to the development of the Finite Element (FE)
model, an identical set of drilling tests was conducted with a CNC machine. More-
over, the results (thrust force and cutting torque) were outputted via a dynamometric
system. The yielded numerical and experimental results demonstrated an increased
agreement, with the relative error varying at reasonable levels for both cutting force
and torque.
Keywords AA7075 drilling · FEM · DEFORMTM · Thrust force · Cutting

torque · Temperature distribution · Chip geometry
4.1 Introduction
A significant number of studies in machining are responsible for the technological

advancement of the field. Modern industries depend heavily on this technology for
the manufacturing of a wide range of products, making any research tool in machining
an invaluable asset. Statistical tools and mathematical models are among these tools
that are used until today.
Budak et al. [1] investigated the mechanics of cutting techniques for the prediction
of cutting force components during milling. Moreover, the authors experimentally
verified the milling cutting force coefficients prediction when milling Ti6Al4V alloy
with respect to chatter, eccentricity, cutting conditions, and cutter geometry. Liu
et al. [2] studied the influence of tool-nose radius and tool wear on the residual stress

64 4 Experimental and 3D Numerical Study of AA7075-T6 Drilling Process
formation during machining of bearing steel. The results were presented for a variety
of cutting conditions with CBN tools, on bearing steel.
Especially for drilling and other conventional processes, many experimental inves-
tigations make use of well-established statistical methodologies such as the Response
Surface Methodology (RSM), the Taguchi method, and the Artificial Neural
Networks (ANN), which of course are extended to non-conventional machining
processes as well. Kao et al. [3] investigated inverted drilling of Al7075 alloy to
examine several parameters such as hole roundness, surface roughness, and hole
enlargement. The authors applied a gray-Taguchi methodology for optimization
purposes of the aforementioned characteristics. Kyratsis et al. [4] implemented RSM
in their work related to drilling of Al7075. Both the thrust force and torque, based on
standard cutting conditions, were studied and the equivalent mathematical models
were derived for prediction purposes. Similar works [5, 6] were realized with the
aid of RSM and ANN for drilling. The aim of these works was to investigate the
effect of several factors on drilling force and torque, in addition to the develop-
ment of prediction models for rapid and reliable calculation of the aforementioned
parameters.
Finite Element (FE) modeling is another well-established methodology that is
widely used by researchers in the field of machining, despite the fact that the model
preparation, as well as the computational stages, are time-consuming, in addition to
the need for increased computational power.
Davim et al. [7] examined the performance of diamond and cemented tools during
the machining of 7075 aluminium alloy by utilizing a commercial Finite Element
Analysis (FEA) software. Additionally, they studied the behavior of the alloy in
terms of machinability. Maranhão and Davim [8] modeled the machining process
of AISI-316 steel by means of FEA and determined the influence of the friction on
several crucial machining parameters, for instance, cutting forces and temperature,
strains, shear stresses, as well as residual stresses. Studies that are oriented towards
drilling [11–13] have used the FE method as well, for developing models in three
dimensions that can be used to study the influence of similar specifications. Moreover,
Nan et al. [14] worked on the three-dimensional FEM model for the micro-drilling
process of AISI-1045 steel. The authors carried out experimental testing to validate
the model, studying at the same time the produced thrust force, torque, and chip
morphology. In some cases [9, 10], the use of CAD-based techniques is observed as
an alternative to the FEA. In this type of study, the programming via the software’s
interface is used to develop a code that can calculate the related parameters, such as
cutting forces, torque, and chip geometry, with increased accuracy. The code usually
comprises Boolean operations that are responsible for considering the tool geometry,
the cutting conditions, and the workpiece to solve the typical formulas that govern
the machining process.
The present study deals with the AA7075 drilling via 3D-FE modeling, in which
typical cutting parameters, within a standard range, were taken into account. Lastly,
the model results were validated via experimental testing, proving the reliability and
accuracy of the model.
4.2.1 Layout of Experimental Testing
Prior to developing the numerical model, nine physical tests were conducted to study
the effect of standard cutting conditions on the drilling of Al7075-T6. The selected
cutting conditions are three levels of cutting speed and feed as seen in Fig. 4.1.
Moreover, the used cutting tool is a Ø10 carbide drill with designation number
B041A10000CPG (KC7325 grade), which is a general-purpose carbide drill, able to
cut aluminum and its alloys. Additionally, the cutting tool is coated with a double
layer of TiN/TiAlN. Figure 4.1 illustrates the geometrical aspects of the physical
drill and compares it with the CAD model used in the numerical study.
The workpiece used for the drilling operations is an orthogonal plate made of
aluminium alloy 7075-T6. The most important physical properties of the alloy are
included in Table 4.1.
The experimental work was carried out with the VF-1 CNC machining center.
Furthermore, a type 9123C force measurement device was utilized along with the
equivalent charge amplifier and data acquisition system, for the cutting force and
torque measurement. To dissipate the generated heat, a typical oil-based coolant was
used. Figure 4.2a contains the aforementioned parts and components. Furthermore,
Fig. 4.2b, c include two sample diagrams indicating force and torque development
versus time, respectively. Both diagrams were generated with the aid of Dynoware
software type 2825D-02, outputting a mean thrust force value equal to approximately
Fig. 4.1 Physical tool and CAD model comparison

Table 4.1 AA7075-T6 and

Mechanical properties AA7075-T6 WC/Co
WC mechanical and thermal
properties [15, 16] E [GPa] 71.7 Rigid
ρ [kg/m3 ] 2,810
ν 0.33
Thermal properties AA7075-T6 WC/Co
C [J/kgK] 960 150
α [o C−1 ] 22 × 10−6 5 × 10−6
k [W/mK] 130 59
430N and a mean torque value close to 1800Nmm. It should be noted that the equiv-
alent force and torque versus time diagrams for the simulated drilling tests were run
for only a few milliseconds. As explained in Sect. 4.2.2, it is very time-consuming
to run a simulation until the drill tip fully penetrates the workpiece. For this reason,
most duplications were run until a steady state was reached and then halted to save
time.
4.2.2 Finite Element Layout
In order to visualize and study the drilling process in three dimensions, DEFORMTM -
3D software was used, specialized in Finite Element Analysis (FEA). 9 simulation
tests were carried out in accordance with the cutting conditions applied during the
experimental phase. The Finite Element (FE) setup was based upon an example that
is already available by DEFORMTM , whereas the tool model was obtained via the
KENNAMETALTM online tool library. Despite the fact that the tool CAD model was
already available by the manufacturer, some changes had to be made for the model to
be properly used. One such change is the conversion of the model to STL file format.
Regarding the FE setup, the kinematics of the process, as well as the tool-work
interface were derived from the example as can be seen in Fig. 4.3. Moreover, the
example was run to verify that it produces reasonable results; thrust forces, torque,
chip morphology, and cut interface temperature. To reduce simulation time, the runs
were halted when cutting force steady state was achieved. Furthermore, the round
workpiece was designed with an alteration compared to the original from the example.
That is a drill spot at the center, which enables a more rapid achievement of the steady
state. This happens due to the fact that as soon as the cutting edge starts to rotate, a
full chip fragment is removed.
In addition, it was important to set a reasonable time step for the simulations, as it
directly affects the completion time, as well as the quality of the visualized process.
In this case, the time step was set to 5.24 × 10−5 s. To calculate the time step, the
cutting speed must be converted to rotational speed of the drill in rounds per second.
Next, the time the drill requires to complete a rotation can be derived. Finally, the
Fig. 4.2 The CNC machine coupled with the data acquisition system and dynamometer (a), a
sample force-time diagram (b), and torque-time diagram (c)
calculated time and the number of steps per revolution, the tool performs, can be
used to determine the simulation time step. The number of the steps should be close
to 360 or more [17] so that one step equals to no more than one degree. It is obvious
that more steps lead to better accuracy but significantly increase run time.
The part was modeled to act deformingly with a mesh of approximately 147,000
elements. Furthermore, the remeshing technique embedded in DEFORM™ was
applied to maintain a mesh with a reasonable number of elements at the area of
interest. It is noted that the size of the mesh changes dynamically as soon as the
cutting process begins. Despite the fact that the round shape can keep the number of
elements relatively low, a denser mesh at the center countermeasures this advantage,
since the denser the mesh, the longer the simulation times. It is recommended to set
the refined mesh in such a manner so that the largest element should not exceed 50%
of the feed. For the present study, a 10:1 ratio was applied. On the contrary, the drill
was modeled to behave rigidly and allowed the minimum number of elements to be
produced. However, the mesh at the area of the drill tip was tuned with a 4:1 ratio,
since it is in the direct proximity of the uncut surface of the part.
The simulation of the aluminium alloy flow stress during the drilling numerical
tests was achieved with the Johnson–Cook model in its generalized form, which is
very common in such situations where high deformations occur, along with high
strain and temperatures. Equation 4.1 represents the model.

ε̇ T − T0 m
σ = A + Bεn 1 + C ln 1− (4.1)
ε̇0 Tm − T0
Fig. 4.3 The drilling process FE model setup

Table 4.2 Material model constants [18]

A (MPa) B (MPa) C n m T 0 (°C) T m (°C)
546 678 0.024 0.71 1.56 20 635
Each one of the parameters of the aforementioned formula represents a specific

component of the material model. Specifically, A, B, and C are material coefficients,
based on the stress and strain properties of the material. ε represents the plastic strain,
n the constant related to the strain hardening, whereas m to the thermal softening, ε̇ is
the strain-rate and ε˙0 is the reference rate. Finally, T and T 0 represent the default and
bulk temperature accordingly, whereas T m is the temperature where the workpiece
melts. The equivalent for the AA7075-T6 constants is presented in Table 4.2, as found
in the literature for similar conditions, with reference strain rate s−1 . To reduce the
produced errors, slight calibration changes were made according to a small number
of trial-and-error runs.
The material separation process occurs when the bond between the nodes of the
workpiece is broken. This process can be determined via a damage model. In the
present case, the normalized Cockcroft-Latham criterion [19] was selected since it
is widely used by many researchers when studying the separation process on similar
materials. Equation 4.2 represents the criterion.
ε f
σmax
Dc = dε pl (4.2)
σ
0
Dc is a constant related to the material fracture, σ max is the maximum stress, whereas
σ̃ represents the effective stress. Finally, εf and εpl are the fracture and the normal
strain, respectively.
Regarding the interaction between the tool and the produced chip, a hybrid model
[20] was implemented to simulate friction in a simple fashion. In most cases, the
approximation of the friction that develops at the friction zone where sliding occurs
is adequate. However, an approximation of both the sliding and the sticking zone can
yield better results. Thus, the simulations can benefit from the use of a hybrid model
as long as the friction coefficients yield reliable results. The generated frictional
stresses around the tip of the drill can be determined by Eq. 4.3. This area has a more
sticking behavior; therefore, the frictional stress is assumed to be analogous to the
weaker material’s shear strength [21].
σy
τ f = kτ √ (4.3)
3
In Eq. 4.3, τ f represents the stress due to friction, k τ denotes a constant and σ y is
the uniaxial yield stress for the workpiece.
On the other hand, friction across the sliding zone can be calculated by Coulomb’s
law. This simplified formula can be used for most metal materials.
τ f = μσn (4.4)
where τ f is the friction-based stress, μ the coefficient, and σ n the normal stress at
the tool-work interface.
Both constant coefficients, the shear and Coulomb were set close to the default
values of the software, which are m = 0.7 and μ = 0.6, according to the findings of
similar studies [20, 22] for the same material.
Consequently, the boundary conditions for the model were set. First of all, the side
walls of the test piece were constrained in this fashion so that all nodes would remain
fixed (U x = U y = U z = 0). Next, the heat exchange conditions were appointed to
all surfaces of the test piece. Specifically, a coefficient for the coolant was assigned,
to approximate convection, with a typical value of 2 × 103 W/(m2 × °C). [17]. In
addition, the drill was allowed to revolve around the feed axis and translate in a
similar manner as the example. Finally, a relationship was assigned for the contact
between the part and the formed chips to approximate any possible collision between
the material and the chips.
4.3 Results and Findings
4.3.1 Cutting Forces and Torque Analysis
To make the comparison possible, both the experiments and the simulations were
carried out under the same cutting conditions. The next values of cutting speed were
used: 50, 100, and 150 m/min. The same applied for feed, having values ranging from
0.15 to 0.25 mm/rev. The produced results were compared so that the drilling process
could be analyzed within the specified range of conditions. Figure 4.4 contains six
graphs that summarize the aforementioned analysis. First of all, the agreement level
for the compared numerical and experimental results was found to be rather high,
considering that the relative error ranges between −15 and 9% for the thrust force
and between −8 and 13% for the torque.
It is shown that feed acts increasingly for both force F z and cutting torque M z .
According to the numerical results, the percentage of increase for the torque, when
shifting from the low level to the middle one, is 46.2, 26.2, and 30.3% for cutting
speed equal to 50, 100, and 150 m/min, respectively. The equivalent percentages for
the force are 27.7, 33.1, and 46.3% accordingly. Similarly, any increasing change in
cutting speed affects both components increasingly, but the effect is weaker compared
to the one generated by the feed.
Fig. 4.4 Experimental versus numerical results for thrust force and torque
4.3.2 Chip Morphology Analysis
Chip formation during drilling of AA7075-T6 exhibits similar patterns between the
experimental process and the simulations. This fact can be observed in Fig. 4.5a,
which depicts the formation of the simulated chip at the first level of feed and speed.
In addition, a sample of the physical chip is illustrated for the same conditions. The
conical shape for both cases (experimental and simulated) was noticeable, which
remained the same for all test runs no matter the change in cutting speed or feed.
Furthermore, the diameter of the curling was measured at various points of the chip’s
length to receive a mean value. It was measured at approximately 5.233 mm for the
simulated and 5.965 mm for the experimental one (Fig. 4.5b). The experimental
Fig. 4.5 The simulated chip evolution (a), sample experimental chip dimensions (b), and simulated
chip thickness measurement (c)
chip dimensions were measured with calipers and an electronic image taken from
a microscope, whereas the simulated ones were with the aid of a CAD system, as
shown in Fig. 4.5c. Specifically, the deformed model of the workpiece was exported
to the CAD environment and sectioned across the plane of interest. In this way,
the chip thickness was revealed, and several measurements were taken in order to
generate an accurate mean value. The conditions applied were the same for both
measurements, specifically for the simulation, the chip was measured as soon as
full curling was formed. The percentage of agreement was calculated as high as
approximately 85–89%. Similar findings were observed for the chip thickness as
well.
4.3.3 Temperature Distribution Analysis
Finally, the temperature distribution on the drill contact area was examined by means
of FEA. This parameter is subject to the heat transfer that takes place between the
drill and the work. A typical stage of the heat distribution was determined to be when
the full length of the cutting edge was engaged and when the thermal steady state was
achieved, which was equal to approximately 0.033 and 0.15 mm penetration depth,
respectively. The temperature distribution for both stages is illustrated in Fig. 4.6 for
an example run with the low value of both cutting speed and feed. It is seen that the
maximum temperature for both stages (77 and 129 °C) were located on the cutting
lips corner (Fig. 4.6a, b, respectively).
Fig. 4.6 Temperature at the tools’ cutting lip corner at full engagement (a) and at the occurrence
of the steady state (b)
By selecting cutting speed and feed at higher levels, a slight increase in tempera-
tures was observed, probably due to the higher frictional forces that generate as the
tool rotates faster.
4.3.4 Concluding Remarks
Concluding, the present study proposed a FE model for AA7075 drilling in three
dimensions. Typical parameters in three consecutive values, with the same drill,
were the used cutting conditions, leading to nine simulation runs. An equal number of
experiments were performed to investigate the fidelity of the FE model. By comparing
the results yielded by the experiments and simulation tests, a good level of agreement
was found, meaning that the model can produce reliable results within the applied
range of conditions. In general, the error for both cutting forces and torque was
determined to be acceptable. Moreover, the morphology of the produced chip was
considered to be similar for all cases studied.
Finally, the next conclusions can be deduced from the analysis:
• The FE model can be effortlessly changed, in terms of the cutting parameters,
material constants, and tool geometry, to extend the evaluation of the effects of
the cutting conditions on similar materials.
• Any increase in the feed rate can affect significantly the thrust force and the torque
as well, specifically leading to an increase.
• On the contrary, changing the cutting speed does not seem to greatly affect thrust
force or cutting torque.
• As expected, the conical shape of the generated chips is maintained regardless of
the cutting parameters.
References
1. Budak E, Altintas¸ Y, Armarego (1996) Prediction of milling force coefficients from orthogonal
cutting data. J Manuf Sci Eng 118:216–224. https://doi.org/10.1115/1.2831014
2. Liu M, Takagi JI, Tsukuda A (2004) Effect of tool nose radius and tool wear on residual stress
distribution in hard turning of bearing steel. J Mater Process Technol 150:234–241. https://doi.
org/10.1016/j.jmatprotec.2004.02.038
3. Kao JY, Hsu CY, Tsao CC (2019) Experimental study of inverted drilling Al-7075 alloy. Int J
Adv Manuf Technol 102:3519–3529. https://doi.org/10.1007/s00170-019-03416-8
4. Kyratsis P, Garcia-Hernandez C, Vakondios D, Antoniadis A (2016) Thrust force and torque
mathematical models in drilling of Al7075 using the response surface methodology. In: Davim
JP (ed) Design of experiments in production engineering. Springer International Publishing,
Cham, pp 151–164
5. Kyratsis P, Markopoulos A, Efkolidis N et al (2018) Prediction of thrust force and cutting
torque in drilling based on the response surface methodology. Machines 6:24. https://doi.org/
10.3390/machines6020024
6. Garcia-Hernandez C, Marín R, Talón J et al (2016) WEDM manufacturing method for noncir-
cular gears using CAD/CAM software. Strojniški Vestn J Mech Eng 62:137–144. https://doi.
org/10.5545/sv-jme.2015.2994
7. Davim JP, Maranhão C, Jackson MJ et al (2008) FEM analysis in high speed machining
of aluminium alloy (Al7075-0) using polycrystalline diamond (PCD) and cemented carbide
(K10) cutting tools. Int J Adv Manuf Technol 39:1093–1100. https://doi.org/10.1007/s00170-
007-1299-y
9. Kyratsis P, Bilalis N, Antoniadis A (2011) CAD-based simulations and design of experiments
for determining thrust force in drilling operations. Comput Des 43:1879–1890. https://doi.org/
10.1016/j.cad.2011.06.002
10. Kyratsis P, Tapoglou N, Bilalis N, Antoniadis A (2011) Thrust force prediction of twist drill
tools using a 3D CAD system application programming interface. Int J Mach Mach Mater
10:18–33. https://doi.org/10.1504/IJMMM.2011.040852
1051/meca/2020046
12. Tzotzis A, Markopoulos AP, Karkalos NE et al (2021) FEM based investigation on thrust
force and torque during Al7075-T6 drilling. In: IOP conference series: materials science and
engineering, p 012009
13. Tzotzis A, Markopoulos A, Karkalos N, Kyratsis P (2020) 3D finite element analysis of Al7075-
T6 drilling with coated solid tooling. In: MATEC web of conferences, pp 1–6
14. Nan X, Xie L, Zhao W (2016) On the application of 3D finite element modeling for small-
diameter hole drilling of AISI 1045 steel. Int J Adv Manuf Technol 84:1927–1939. https://doi.
org/10.1007/s00170-015-7782-y
15. Jafarzadeh E, Movahhedy MR, Khodaygan S (2018) Prediction of machining chatter in milling
based on dynamic FEM simulations of chip formation. Adv Manuf 6:334–344. https://doi.org/
10.1007/s40436-018-0228-7
16. MatWeb L (2019) MatWeb Material Property Data: {http://www.matweb.com}. www.matweb.
com. Accessed 29 Sep 2019
18. Brar NS, Joshi VS, Harris BW (2009) Constitutive model constants for Al7075-T651 and
Al7075-T6. AIP Conf Proc 1195:945–948. https://doi.org/10.1063/1.3295300
19. Cockcroft MG, Latham DJ (1968) Ductility and the workability of metals. J Institue Met
96:33–39
References 75
20. Ucun I (2016) 3D finite element modelling of drilling process of Al7075-T6 alloy and exper-
imental validation. J Mech Sci Technol 30:1843–1850. https://doi.org/10.1007/s12206-016-
0341-0
21. Agmell M (2018) Applied FEM of metal removal and forming, 1stedn. Studentlitteratur, Lund
22. Melkote N, Grzesik W, Outeiro J et al (2017) Advances in material and friction data for
modelling of metal machining. CIRP Ann Manuf Technol 66:731–754. https://doi.org/10.1016/
j.cirp.2017.05.002
Chapter 5
3D Finite Element Simulation of CK45
Steel Face-Milling: Chip Morphology
and Tool Wear Validation
Abstract Nowadays, three-dimensional Finite Element (FE) modeling is used more

often for the investigation of the machining processes, due to the increased approx-
imation it provides and the fact that a number of variables cannot be examined by
other means. The present work introduces a 3D FE model of CK45 (AISI1045) steel
face-milling. In specific, it proposes a model for full immersion face-milling with
a head mill. Moreover, it focuses on the comparison of the simulated chip forma-
tion process with experimental results, as well as the investigation of the tool wear in
terms of the generated temperatures. The results of this study exhibit good agreement
results for the generated chip morphology and dimensions. In addition, it is pointed
out that tool wear is directly connected to cutting-edge temperatures and chip flow.
Keywords AISI1045 · Face milling · FEM · DEFORM · Tool wear · Chip

formation · Tool temperature
5.1 Introduction
Milling is one of the most versatile processes used in the industry. Aerospace, vehicle,
construction, and biomedical are only a few of the industries that utilize milling to
produce a number of products such as mechanical parts, molds, implants, and more.
Therefore, investigation of the process and its parameters is often the topic of many
studies. The tools and techniques that are usually applied in such investigations
involve experimental testing, statistical processing, and soft computing, as well as
numerical modeling. Especially Finite Element (FE) modeling, constitutes a robust
tool for the examination of several factors that participate in the milling process
and in other machining procedures as well. Maurel-Pantel et al. [1] investigated the
machinability of AISI304L stainless steel by means of the Finite Element Method
(FEM). The authors examined the generated cutting forces and stresses and validated
the results by comparing the numerical results with equivalent experimental ones.
Gao et al. [2] presented a Coupled Eulerian–Lagrangian (CEL) model in 3D, for
the simulation of AA6061-T6 end-milling, validated by experimental work. Two
milling methods were tested, with the milling forces and the chip production being

78 5 3D Finite Element Simulation of CK45 Steel Face-Milling: Chip …
the objective investigation parameters. As seen in example studies that relate to other
conventional machining processes such as drilling and turning [3–7], the typical
investigation parameters are the generated chip morphology, the machining forces,
and torque, the tool wear and life, the residual stresses, as well as the temperature
generation across the tool cutting edge or the tool-workpiece contact interface.
Micro-milling is another subject of increased interest, where FE modeling is appli-
cable. In such delicate operations, tools of very small diameter are used, usually in the
range of a few hundred microns. For example, Thepsonthi and Özel [8] studied three
cases of micro-milling, the full-immersion of tool, and the half-immersion during
standard milling techniques. The chip formation, the tool wear, the temperature allo-
cation across the cutting-edge, and the cutting-edge roundness effect on the produced
forces, were the main objective of the study. Similarly, Ucun et al. [9] worked on a
three-dimensional approximation of micro-milling of Ti-6Al-4V alloy. The authors
focused their study on the cutting forces generation, the chip aspects, and the tool
stress. Additionally, experiments were carried out to verify the model. Similar works
[10, 11] are available in the literature, focusing on parameters that cannot be easily
examined via experiments, confirming the need for FE models. Especially models
in three dimensions are considered to be tools of increased precision when dealing
with the simulation of more complex processes.
Face milling is the most common milling operation and can be performed with
a variety of tools, each one designed for a specific operation. Pittalà and Monno
[12] developed a 3D FE model and focused on the examination of the developed
milling forces. A sensitivity study and an inverse method were utilized for the iden-
tification of the friction coefficients and the material constants respectively. Soo
et al. [13] presented a Lagrangian-based, 3D FE model for simulating end-milling
of Inconel718 superalloy with ball-nosed tools, at high speeds. The study included
cutting force predictions, validated against corresponding experimental results, as
well as tool-chip temperatures. Rao et al. [14] carried out an experimental analysis
focused on the required energy and surface quality measurements, as well as cutting-
tool metrics. The authors supplemented the experiments with a FEM-based tool wear
model in terms of the contact interface temperature and stress, as well as the chip
velocity. Nieslony et al. [15] developed a FE model in three dimensions for inves-
tigation purposes of the cutting power and specific cutting energy as to the rotation
angle of the mill and the chip thickness. Similar works [16–18] studied face milling
and several conditions that affect the process, such as burr formation, chip thickness,
and cutting forces.
This study utilizes well-established models and methods to develop a 3D FE model
for CK45 (AISI1045) steel face-milling with coated tools. Moreover, the recom-
mended by the manufacturer cutting conditions were applied. The study focuses
mainly on the chip formation mechanisms, the tool wear, and temperature gener-
ation. In addition, the results regarding the chip development and tool wear were
validated with experiments.
5.2.1 Experimental Framework
The experimental work was carried out on a four-axis HAAS VF-1 milling center
without the use of coolant. The used workpiece is a CK45 steel plate with 150 mm
length, 120 mm width, and 20 mm thickness. To perform the face milling on the
plate, a general-purpose shell mill and the equivalent inserts were selected, with ISO
designation numbers 50A04RS45SE14EG and SEPT1404AESNGB2, respectively.
According to the manufacturer, the selected cutter system is designed for steel, stain-
less steel, and several high-temperature alloys. Moreover, the inserts are carbide
with a 3.5 microns TiAlN + TiN coating. A mill arbor, with code 40.340.22 by
HAIMER™, was used to clamp the shell mill to the machine. Finally, the machining
conditions are available in the manufacturer’s catalogue for the CK45 steel, which
belongs to the ISO P2 material group. Thus, the cutting speed was set to 230 m/min
and the feed to 0.36 mm/tooth. The milling process was done with full immersion
of the head at depth of cut equal to 2 mm, lower than the maximum allowed. The
effective cutting diameter is 50 mm.
Figure 5.1 illustrates the complete set of experimental work. Specifically, Fig. 5.1a
depicts the assembly of the mill cutter, with a more detailed inspection of the milling
insert, Fig. 5.1b shows the CNC machine and, lastly, Fig. 5.1c illustrates the plate
that was used as a workpiece.
5.2.2 Face-Milling CAD-Based Setup
The equivalent of the experimental setup was reproduced with the aid of a CAD
system. This technique allows the examination of the setup in a simple and repeatable
manner. Hence, in the present work, the positioning of the tool on the test piece was
achieved, the angles involved in the milling procedure, as well as the force vectors,
were extracted, and the FE setup was facilitated with respect to the CAD-based
model.
The workpiece was designed in the CAD system in such a way that it would
resemble a part with an already cut region, based on the cutter’s effective diameter,
as well as the insert’s shape and the selected cutting conditions (Fig. 5.2a). It is noted
that the milling insert has an entering angle (KAPR) equal to 45° and an axial rake
angle of 18°, with no inclination in the radial direction. Figure 5.2b illustrates the
dimensions of the used tool. To further simplify the setup, only one of the four inserts
was implemented.
Fig. 5.1 The milling cutter assembly, with a closer view of the insert (a), the CNC milling center
(b), and the used workpiece (c)
Fig. 5.2 The setup of the milling process in the CAD environment, with details (a) and the mill
insert dimensions (b)
5.2.3 Numerical Modeling of the Face-Milling Process
Analysis of domain and boundary conditions

The milling process FE model was set with the aid of DEFORM™-3D ver. 12 soft-
ware that uses an updated Lagrangian formulation. First, the workpiece was designed,
as seen in Fig. 5.3a, based on the CAD-based setup. Hence, it was designed according
to the effective cut diameter, the depth of cut, the radius of the insert’s cutting edge,
and the entering angle. A sectioned sketch of the work is illustrated in Fig. 5.3b.
It is shown that it was sketched as an arc with an already cut region, to minimize
the simulation time. Moreover, it was set to be plastic and was initially meshed
with approximately 32,000 tetrahedral elements, based on the feed value. Since the
simulated material separation process is identical to all machining processes, a rule
of thumb is to use elements with a minimum size, equal to a portion of the feed
value [19, 20]. The size ratio of 7:1 for the elements was applied on the contact
area, which combined with the local remeshing criteria, shifted the total number of
elements to 100,000. The tool, on the other hand, was meshed with roughly 125,000
elements. Even though the tool was set to behave as rigid, it was necessary to use
a high number of mesh elements to achieve a better approximation of the thermal
behavior of the tool, as well as to calculate its wear more accurately. It must be noted
that the majority of the elements belong to the coating’s mesh, which was crucial to
the wear calculation. Due to the subtle thickness of the coating (Fig. 5.3c), the size
of the elements had to be very small. In addition, a 10:1 size ratio was used to refine
the areas around both the primary and the secondary cutting edges (Fig. 5.3c). The
primary cutting edge is the one that cuts most of the formed chip volume, whereas the
secondary is the one that separates the chip from the bottom of the part. As discussed
in Sect. 5.3.2, it is evident that less wear is induced on the secondary cutting edge.
Finally, the detail of Fig. 5.3b depicts the average chip section that is expected to be
formed due to the shape of the tool and the feed value.
The boundary conditions for the model were set according to the milling kine-
matics. The workpiece was fixed on its bottom and side, as shown in Fig. 5.3b. In
contrast, both a translation and a rotation were applied on the tool so it would be
possible to follow the trajectory illustrated in Fig. 5.3a. This way, it was ensured that
both the tool rotation and the table feed are simulated. Furthermore, the boundary
conditions related to the heat transferred among the tool, the work, and the environ-
ment were attributed. The constants for both types of heat transfer were set to 20 and
4.5 × 104 W/(m2 × °C), respectively.
Material and friction modeling
To model the material’s behavior under the conditions that occur during machining, a
tabular data format was used according to temperatures from 20 °C (ambient temper-
ature) to the softening temperature of the material. Moreover, the strain values [21]
used range from 0 to 5, for strain rates equal to 1, 10, 102 , 103 , 104 , 105 , and 5 × 105 ,
respectively. Therefore, the flow stress was modeled with respect to the temperature,
the strain, and the strain rate for the aforementioned range of values. A number of
Fig. 5.3 The milling process FE model setup: the analysis domain (a), the workpiece section (b),
and the cutting-tool model (c)
extra points were interpolated in the table to yield more smooth flow stress curves.
The approximation of the damage process and material separation, the normalized
Cockcroft-Latham criterion [22] was applied, which links a material damage constant
with both the maximum and the effective stress. Moreover, the mechanical work-
to-heat conversion factor was set to 0.9, with the rest of the work absorbed by the
workpiece.
In addition, Table 5.1 includes the standard properties of the part material used in
the present study, as found in the software’s material library. It is noted that the thermal
properties are expressed with respect to the temperature. Similar to the flow stress
values, intermediate points were interpolated to fill the tabular data format for these
properties. Finally, Table 5.2 displays the thermal properties of the tool material,
as well as the coating, since they were taken into account during the calculations
involved in the thermal analysis.
In order to describe the frictional phenomenon that occurs between the tool and
the part, a hybrid approach was followed. Specifically, the shear friction was modeled
with a constant coefficient equal to m = 1 and the Coulomb coefficient was deter-
mined by the levels of the relative sliding velocity. According to the findings of Binder
et al. [24], the friction coefficient between AISI1045 steel and coated carbides is close
Table 5.1 CK45 steel thermomechanical properties [21]

Mechanical properties
E (GPa) 212.0 207.0 (100 °C) 192.0 164.0 (600 °C) 69.44
(20 °C) (300 °C) (1500 °C)
ρ (kg/m3 ) 7,850
ν 0.3
Thermal properties
Specific heat (J/kgK) Thermal expansion × 10−6 Thermal conductivity
(°C−1 ) (W/mK)
361.9 (20 °C) 11.9 (20 °C) 41.7 (20 °C)
389.4 (100 °C) 12.5 (100 °C) 43.4 (100 °C)
445.9 (300 °C) 13.6 (300 °C) 41.4 (300 °C)
610.7 (600 °C) 14.9 (600 °C) 34.1 (600 °C)
610.7 (1500 °C) 14.9 (1500 °C) 34.1 (1500 °C)
Table 5.2 Tool thermal

Thermal properties Carbide Coating
properties [21, 23]
Heat capacity [J/kgK] 150 150
Thermal expansion 5× 10−6 9.2 × 10−6
[°C−1 ]
Thermal conductivity 100 °C 40.15 12.61
[W/mK] 300 °C 48.55 14.01
500 °C 56.95 15.41
700 °C 65.35 16.81
900 °C 73.75 18.21
to 0.3 for sliding velocities that range from 1000 to 1500 mm/s. To identify the rela-
tive sliding velocity, a 3D simulation with a smaller workpiece was run. After the
generation of a full chip curling (Fig. 5.4) and the achievement of the thermomechan-
ical steady state, the relative sliding velocity for the present case was determined to
be between 1000 and 1700 mm/s. Therefore, the Coulomb coefficient equal to μ =
0.3 was preserved.
Fig. 5.4 The preliminary simulation run showing the sliding velocities at the direction of cut
5.3 Results and Discussion
5.3.1 Chip Formation Analysis and Temperature Distribution

Evaluation
A 3D chip flow analysis is more efficient compared to the equivalent 2D, since
these models are limited in the XY plane. Because of the plane limitation, the chip
accumulates in front of the tool’s cutting edge. Therefore, the chip flow predicted by
a two-dimensional model is less accurate. Moreover, a three-dimensional simulation
yields a more realistic flow, and it is possible to measure all dimensions of the chip.
Figure 5.5a illustrates the chip flow during the cut, produced by the fully immersed
tool. The generated chip’s section is found to be as expected (see Fig. 5.3b), based
on the feed value and the 45° corner angle. Additionally, the phases where the chip
formation was observed, reveal that the chip tends to form a small spring with one or
two curls, which then slightly straightens (ϕ = 100°) and finally breaks into single
curl chips, probably due to the contact with the workpiece.
The temperature rise on the tool’s cutting edges was examined by collecting
the maximum temperature and its zone, as well as the temperature distribution.
Figure 5.5b shows the temperature rise along the cutting edges with respect to the
revolution of the tool. It is shown that the area where the heat spreads, match the two
wear zones, which is reasonable since these areas are in contact with the workpiece. In
addition, it is evident that the temperature rises quickly as soon as the full engagement
5.3 Results and Discussion 85
Fig. 5.5 The simulated chip morphology and chip flow in 3D (a) and the temperature distribution
on the tool’s cutting edges (b)
of the tool occurs. This denotes a lower level of heat transferred between the cutting
edges and the chip, due to the smaller contact zone compared to other types of
inserts. This is reasonable considering that the 45° corner angle produces relatively
thin chips, thus the contact area is smaller.
Figure 5.6 presents a sample chip experimentally generated, compared to the
equivalent simulated one. The parameters that are taken into account are the chip
curling diameter, the deformed chip width, and thickness. The measurements of the
aforementioned dimensions were performed on a number of chips. Especially the chip
thickness, was measured on multiple points across the curling as seen in Fig. 5.6a, in
order to derive an average value, since the tool run-out that is present in the system
does not allow the generation of a uniform undeformed chip thickness. The same
measurement technique was utilized for the simulated chip (Fig. 5.6b), within the
CAD environment. As seen in Fig. 5.6b, the sample chip model was exported to
the CAD system, where it was sectioned to achieve more accurate measurements. In
general, it is shown that there are similarities between the simulated and the real chip.
Moreover, the agreement percentage for all three dimensions, curl radius, width, and
thickness, was found to be close to 85, 88, and 78%, respectively. It must be noted
that the simulated chip was determined to be thicker compared to the real ones during
all measurements.
5.3.2 Tool Wear Assessment
The Usui et al. [25] model was utilized to predict the adhesive-based tool wear by
using the results from the FEM runs, since the variables that participate in the model
Fig. 5.6 The experimental (a) and the simulated (b) sample chip formation
can be determined via FEM-based simulations. In specific, the thermomechanical

load generated in the tool-work contact interface can be converted to nodal tool wear.
In the present case, the average wear depth along the primary and the secondary
cutting edges of the tool was determined at discrete points of time. Equation 5.1
represents the tool wear W with respect to time (dW /dt). In other words, the wear
represents the wear volume with respect to time and area of contact. Where p is the
contact interface pressure, vs represents the sliding velocity developed by the chip,
T represents the interface temperature and, finally, a and b are coefficients related to
the workpiece and the tool material, respectively, defined experimentally.
dW
= apvs exp(−b/T ) (5.1)
dt
The modeling variables, specifically the pressure, the sliding velocity, and the
temperature generated at the area of interest, were extracted from the FEM simu-
lations, whereas the values of the two constants were calculated considering the
cutting data acquired after the full engagement of the mill head to the workpiece and
the completion of three passes, thus approximately after 20 s. Moreover, they were
compared to the constants used by studies [26, 27] that examined similar materials,
to verify that they are of the same magnitude. It should be noted that the intention of
this part of the study is to approximate the tool wear zones, rather than the tool wear
rate. As seen in Fig. 5.7a, the tool wear is located at two distinct zones, the primary,
and the secondary zone. The primary zone corresponds to the primary cutting edge,
which is in contact with the material being removed, whereas the secondary zone
corresponds to the secondary cutting edge, which is in contact with the bottom surface
of the machined part. Once the full engagement of the tool occurs, the secondary
cutting edge scratches continuously the bottom surface. However, due to the fact
5.4 Concluding Remarks 87
Fig. 5.7 The tool wear zones (a), the simulated wear pattern (b), and the predicted average tool
wear depth (c)
that no chip is generated, the tool wear is rather small compared to the wear that is
induced in the primary zone. The FEM simulations yielded a wear pattern similar to
the actual one, as can be seen in the sample of Fig. 5.7b.
Finally, Fig. 5.7c illustrates the average wear values that are formed along the
cutting edge. This graph shows a rapid tool wear depth as soon as the engagement
begins, but this increase is quickly limited. Furthermore, it is indicated that the wear
progress is continuous but not its rate, rather it increases as the tool constantly wears
with the cutting distance. However, it is evident that the increase occurs at a slower
pace.
5.4 Concluding Remarks
In this work, a 3D-FE model for the face-milling of CK45 steel was developed by
implementing well-established methodologies and models. The developed model
focused on the examination of the chip formation, the cutting-edge temperature
distribution, and the tool wear. The results of the study can be related to the following
remarks:
• The generated chip follows a typical curl pattern. As the cut advances, the chip
formed resembles a small spring with a couple of curls. Due to the contact with
the workpiece, the chip slightly unfolds and breaks into smaller pieces. Moreover,
the chip thickness is relatively small, with the average value being lower than the
feed value.
• Tool wear appears in two zones matching the two cutting edges of the tool. Most
of the wear occurs across the primary cutting edge, due to the higher contact
length compared to the secondary cutting edge, which scratches the bottom cut
surface.
• A rapid increase in the tool wear is evident as soon as the tool begins to cut,
followed by a milder and at the same time, constant increase. The same applies
to the generated maximum temperature, which is located on the primary cutting
edge. This is reasonable since the temperature is related to the wear and both are
influenced by the contact area.
References
1. Maurel-Pantel A, Fontaine M, Thibaud S, Gelin JC (2012) 3D FEM simulations of shoulder

milling operations on a 304L stainless steel. Simul Model Pract Theory 22:13–27. https://doi.
org/10.1016/j.simpat.2011.10.009
2. Gao Y, Ko JH, Lee HP (2018) 3D coupled Eulerian-Lagrangian finite element analysis of end
milling. Int J Adv Manuf Technol 98:849–857. https://doi.org/10.1007/s00170-018-2284-3
3. Buchkremer S, Klocke F, Veselovac D (2016) 3D FEM simulation of chip breakage in metal
cutting. Int J Adv Manuf Technol 82:645–661. https://doi.org/10.1007/s00170-015-7383-9
1051/meca/2020046
6. He YL, Davim JP, Xue HQ (2018) 3D progressive damage based macro-mechanical FE simu-
lation of machining unidirectional FRP composite. Chin J Mech Eng. https://doi.org/10.1186/
s10033-018-0250-5
9. Ucun İ, Aslantas K, Özkaya E, Cicek A (2017) 3D numerical modelling of micro-milling
process of Ti6Al4V alloy and experimental validation. Adv Mater Process Technol 3:250–260.
https://doi.org/10.1080/2374068X.2016.1247343
0157-6
10.3390/mi8060187
References 89
13. Soo SL, Dewes RC, Aspinwall DK (2010) 3D FE modelling of high-speed ball nose end
milling. Int J Adv Manuf Technol 50:871–882. https://doi.org/10.1007/s00170-010-2581-y
14. Rao B, Dandekar CR, Shin YC (2011) An experimental and numerical study on the face
milling of Ti-6Al-4V alloy: tool performance and surface integrity. J Mater Process Technol
211:294–304. https://doi.org/10.1016/j.jmatprotec.2010.10.002
15. Nieslony P, Grzesik W, Bartoszuk M, Habrat W (2016) Analysis of mechanical characteristics
of face milling process of Ti6Al4V alloy using experimental and simulation data. J Mach Eng
16:1–9
16. Sartkulvanich P, Sahlan H, Altan T (2007) A finite element analysis of burr formation in face
milling of a cast aluminum alloy. Mach Sci Technol 11:157–181. https://doi.org/10.1080/109
10340701339895
17. Chan CH, Kasim MS, Izamshah R et al (2017) Analysis of face milling performance on Inconel
718 using FEM and historical data of RSM. In: IOP conference series: materials science and
engineering, pp 1–11
18. Cai S, Yao B, Feng W, Cai Z (2019) An improved cutting force prediction model in the milling
process with a multi-blade face milling cutter based on FEM and NURBS. Int J Adv Manuf
Technol 104:2487–2499
010004
machining forces during AISI 4140 hard turning. Strojniški Vestn J Mech Eng 66:467–478.
22. Cockcroft MG, Latham DJ (1968) Ductility and the workability of metals. J Inst Met 96:33–39
23. Akbar F, Sheikh M, Mativenga P (2009) On the heat partition properties of (TiAl)N Compared
with TiN coating in high speed machining. Proc Inst Mech Eng Part B J Eng Manuf 223:363–
375. https://doi.org/10.1243/09544054JEM1365
24. Binder M, Klocke F, Lung D (2015) Tool wear simulation of complex shaped coated cutting
tools. Wear. https://doi.org/10.1016/j.wear.2015.01.015
25. Usui E, Shirakashi T, Kitagawa T (1984) Analytical prediction of cutting tool wear. Wear
100:129–151. https://doi.org/10.1016/0043-1648(84)90010-3
26. Attanasio A, Ceretti E, Fiorentino A et al (2010) Investigation and FEM-based simulation of
tool wear in turning operations with uncoated carbide tools. Wear 269:344–350. https://doi.
org/10.1016/j.wear.2010.04.013
27. Courbon C, Fabre D, Methon G et al (2021) CIRP annals—manufacturing technology a 3D
modeling strategy to predict efficiently cutting tool wear in longitudinal turning of AISI 1045
steel. CIRP Ann Manuf Technol 70:57–60. https://doi.org/10.1016/j.cirp.2021.04.071

3D FEA Simulations in Machining

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

3D FEA Simulations in Machining

Uploaded by

Copyright:

Available Formats

SpringerBriefs in Applied Sciences and Technology

Manufacturing and Surface Engineering

Manufacturing and Surface Engineering

ISSN 2191-530X ISSN 2191-5318 (electronic)

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023

1 A Comparative Study Between 2D and 3D Finite Element

3 FEM-Based Study of AISI52100 Steel Machining: A Combined

Keywords 2D FEM · 3D FEM · Machining · Cutting forces · Residual stresses ·

The Finite Element Method (FEM) is a well-known tool, implemented often in

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

a polycrystalline diamond (PCD) tool to acquire the optimal finishing parameters

1.2 Comparison Between 2D and 3D FEM Analysis

1.2.1 Evaluation of Cutting Forces and Torque

Evaluation of the forces generated during machining is a topic widely discussed

Fig. 1.1 Machining force

Tamura [32] established a hybrid simulation model, which implemented 2D simu-

1.2.2 Chip Formation and Dimensions Assessment

Regarding the understanding of the chip formation mechanisms, the utilization of 3D

1.2.3 Examination of Residual Stresses

1.2.4 Temperature and Tool-Wear Assessment

is considered to be the same as in a 2D analysis. Therefore, the authors combined

Table 1.1 Summary of the advantages and disadvantages of 2D and 3D FEM

Abstract The present chapter introduces the fundamentals of Finite Element

Keywords 3D FEM · 3D simulation · Machining · Turning · Drilling · Milling ·

Computers are widely used in engineering to simulate optimal prototype design

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 13

must resort to numerical methods, especially when analytical methods cannot be

Fig. 2.1 Solution flow chart of an engineering problem

2.1.1 FEM-Based Studies in Machining

2.1.2 Advantages and Limitations of FEM

It is not easy to determine whether 2D or 3D modeling is a better method, since both

2.2 Finite Element Modeling in Machining

2.2.1 Meshing and Element Types

Fig. 2.2 Standard types of finite elements

Fig. 2.3 Number of nodes based on the element’s order

Fig. 2.4 Typical mesh size parameters

Fig. 2.5 Sample meshing

Fig. 2.6 Example tool model meshing

2.2.2 Mesh Adaptivity

Fig. 2.7 Material separation and chip generation

and evolution during modeling of machining with DEFORM™-3D ver.12, where a

2.2.3 Tool-Workpiece Representation

Accurate representation of the tool-work interface is of great importance in FEM-

2.2.4 Boundary Conditions

Fig. 2.11 Heat exchange schematic in FEM-based study

2.2.5 Material Flow Stress Modeling

σ = C(T )ε n(T ) (2.4)

σ = C(T mod )ε n(T mod ) (2.5)

by many researchers, to meet the needs of different applications [56–62] involving

Table 2.1 Sample flow stress models for 3D studies of steel

Table 2.3 Sample flow stress models for 3D studies of Ti6Al4V

2.2.6 Friction Modeling and Contact Description

Fig. 2.12 The deformation zones in metal cutting

Table 2.4 Friction modeling in 3D machining

2.2.7 Material Separation

Fig. 2.13 Uniaxial

2.2.8 Tool Wear

2.3 Typical FEM-Based Results

2.3.1 Cuttings Forces, Torque, and Residual Stresses

2.3.2 Chip Morphology, Temperature Distribution, and Wear

2.4 Conclusions and Perspectives

This chapter introduces the most important aspects of FE modeling in machining

1. Ceretti E, Fallböhmer P, Wu WT, Altan T (1996) Application of 2D FEM to chip formation in

85. Zemzemi F, Rech J, Ben SW et al (2008) Identification of a friction model at