Precision Engineering 62 (2020) 1–9

Contents lists available at ScienceDirect

Precision Engineering
journal homepage: http://www.elsevier.com/locate/precision

Study on machinability of additively manufactured and conventional

titanium alloys in micro-milling process
F. Hojati a, *, A. Daneshi a, B. Soltani a, B. Azarhoushang a, D. Biermann b
a
Institute of Precision Machining (KSF), Hochschule Furtwangen University, Tuttlingen, Germany
b
Institute of Machining Technology (ISF), Technische Universit€
at Dortmund, Dortmund, Germany

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

Keywords: Capability of Additive Manufacturing (AM) technology in the production of complex parts with high flexibility
Micro milling has led to the growing interest in their application as an alternative for conventional manufacturing processes.
Additive manufacturing Despite the outstanding benefits of the AM process, due to their poor surface quality, the precision parts pro­
EBM
duced by this method generally need to be machined, ground, or polished. This paper addresses the machin­
Specific cutting energy
Surface quality
ability of AM Ti6Al4V titanium alloy parts in the micro-milling process with a specific focus on cutting forces,
Burr formation specific cutting energy, burr formation, and surface quality. Additive parts were produced by Electron Beam
Melting (EBM) technique and were compared with the extruded Ti6Al4V parts in the micro-milling process. No
significant difference could be observed in the cutting forces of both materials at chip thicknesses between 7.4
and 37.3 μm, despite the higher hardness of the EBM Ti6Al4V compared to the extruded Ti6Al4V. However,
micro-milling of the EBM parts produced finer surfaces. Cutting forces and specific cutting energies of EBM parts
were less than those of extruded parts at minimal chip thicknesses (lower than 7.4 μm). Continuous wavy-type
burrs were formed in micro-milling of the EBM Ti6Al4V and were larger than those of extruded Ti6Al4V.

1. Introduction In spite of their advantages, the components produced by AM

Titanium alloys are widely used in aerospace, aeronautical, auto­
mobile, chemical and medical industries due to their superior mechan­
ical and material properties. Low thermal conductivity, high chemical
reactivity, the ability to maintain strength at high temperature and se­
vere work hardening of titanium and its alloys lead to high machining
temperature and cutting forces. Additionally, the possibility of phase
transformation of titanium alloys during the machining complicates
their cutting process even further.
One of the common methods for the production of titanium parts has
been the forging process for many years. Recently, the interest in the
manufacturing of different components made of titanium alloys
employing Additive Manufacturing (AM) techniques is tremendously
increasing. Horn and Harrysson [1] provided a report regarding the
current state of additive manufacturing techniques and their applica­
tions. According to their report, the parts in AM are formed layer by
layer from metal powders by a heat source in an arbitrary direction.
Therefore, a high degree of freedom can be achieved through AM pro­
cesses, which provides the possibility to produce complex parts with low
material wastage and high productivity.
generally have poor surface quality and are in most cases only near net
shaped. To achieve the desired surface quality and part accuracy, several
investigations have already been carried out mainly on Ti6Al4V as the
commonly used titanium alloy in additive manufacturing. Alexander
et al. [2] studied the Ti6Al4V alloy produced by EBM and reported a
degree of anisotropy in terms of microstructure and material properties
between the horizontal and vertical planes. They also investigated the
machinability of raw and heat-treated EBM thin-wall parts. In their
work, the application of Hot Isostatic Pressing (HIP) enhanced the ho­
mogeneity of the material structure in all cross-section planes and made
the mechanical properties of EBM parts isotropic. Although, the HIP
process had almost no influence (only 3–4%) on the resulted surface
properties. Sartori et al. [3] conducted an investigation in turning of
Ti6Al4V parts manufactured by EBM and Direct Metal Laser Sintering
(DMLS). The additively induced material properties through DMLS
showed higher hardness and lower thermal conductivity in comparison
with EBM and wrought material. This difference led to the deepest crater
in the tool in turning of DMLS under dry condition. Concerning the tool
wear, thermal and mechanical properties of EBM, DMLS and wrought
material, Sartori et al. [3] concluded that EBM and DMLS have the best

* Corresponding author.
E-mail address: hofa@hs-furtwangen.de (F. Hojati).

https://doi.org/10.1016/j.precisioneng.2019.11.002
Received 31 October 2019; Accepted 1 November 2019
Available online 4 November 2019
0141-6359/© 2019 Elsevier Inc. All rights reserved.
F. Hojati et al.
and worst machinability respectively. This indicates that the application
of different AM techniques causes the variation in mechanical and ma­
terial properties and eventually machinability of the parts. In this re­
gard, Milton et al. [4] also carried out an investigation in the milling of
Selective Laser Melting (SLM) Ti6Al4V parts. The axial forces of SLM
parts were 22% higher than those regarding the conventional hot rolled
alloy parts. Additionally, no variation in cutting force between the built
samples in three different planes was reported for SLM parts. With
respect to the surface integrity, the material microstructure was not
changed after machining. In terms of the comparison of cutting force
between SLM and conventionally produced parts, the obtained results
by Milton et al. [4] are in good agreement with the findings of Polishetty
et al. [5]. They studied the influence of cutting speed and feed rate on
cutting forces and surface roughness in turning of SLM Ti6Al4V. The
SLM part induced higher cutting forces compared to the conventional
parts. Moreover, it was reported that a finer surface was achieved for
SLM parts due to their higher hardness and brittle nature. Le Coz et al.
[6] studied the machinability of SLM parts in comparison with cast alloy
parts. They also confirmed that the machining forces (cutting and feed
forces) of SLM Ti6Al4V parts are higher than those of conventional parts
in the micro-turning process. No difference in chip morphology between
SLM and conventional parts was detected, and the micro-cutting did not
highly influence the surface integrity of the machined parts. Bonaiti
et al. [7] studied the micro-milling of additive Ti6Al4V parts manufac­
tured by Laser Engineering Net Shaping (LENS). The AM parts induced
lower cutting forces despite their higher hardness. The results oppose
the founding of other researchers who studied the machining of SLM
parts. Hence, the machining of additively manufactured parts would not
always lead to the higher cutting forces compared with conventionally
produced parts. In fact, the machinability depends on the applied AM
techniques in the production of parts.
Altintas and Jin [8] addressed the corresponding issues of
micro-cutting. In spite of its advantages, some issues arise in
micro-cutting in comparison with macro-cutting that must be taken into
account to assure the robustness of the process for industrial applica­
tions. The uncut chip thickness in micro-cutting is small, and the ratio of
uncut chip thickness to the cutting edge radius is much lower than that
in macro-cutting. In fact, the size of the chip thickness with respect to the
edge radius determines the removal mechanism and transition from
ploughing to shearing/cutting. Lai et al. [9] explained in their paper that
in micro-cutting, the edge radius acts as a rake face with a high negative
rake angle and the material in front of the tool is conducted in two di­
rections. They stated that the material above the stagnation point flows
up to the rake face and become a part of the chip whereas the material
under the stagnation point flows through the flank face to the machined
surface leading to the elastic recovery. Thus, the stagnation point dif­
ferentiates the deformation mechanism. As discussed by Chae et al. [10],
a critical value as the minimum uncut chip thickness (hmin) could be
defined under which no chip is formed, and ploughing predominates.
The effective cutting process takes place with a chip thickness higher
than hmin. Moreover, Simoneau et al. [11] mentioned that the reduction
of chip thickness contributes to a non-linear increasement of specific
cutting energy as a result of increased flow stress in the cutting zone.
This phenomenon is called the size effect. Due to the size effect, the burr
formation is more complicated in the micro-cutting compared with the
macro-cutting. Another issue that should be taken into account
regarding the micro-cutting is the tool deflection. Uriarte et al. [12]
proposed the mechanistic modelling for prediction of micro-milling
cutting forces than was used for estimation of tool deflection in
machining of H13 steel hardened to 60 HRC with two-flute carbide
micro end mills with diameters from 0.1 to 0.4 mm. They reported the
good agreement between simulation and experimental results. Other
researchers also addressed these issues and tried to come up with the
solution. Arnaiz-Gonz� alez et al. [13] applied two types of Artificial
Neural Networks ANNs, multilayer perceptron (MLP) and radial basis
functions (RBF), in order to predict dimensional errors for the inclined
2
Precision Engineering 62 (2020) 1–9
surfaces manufactured by ball-end-milling. As concluded, both ANN
methods indicate the better accuracy compared to the semi-empirical
models. Lo’pez de Lacalle et al. [14] proposed the diagnostic tool and
applied it in three cases: validation of mechanistic cutting force model,
detection of unexpected tool engagement and investigation of static and
dynamic milling problems. They confirmed that the proposed model has
capability to be used in manufacturing of complex parts, detection of
potential problems and process optimization.
According to the obtained results from different studies, a prediction
about the influence of AM processes on the machinability of the material
is not possible, and each AM process has a special effect on the additively
produced components. So, it is highly demanded to put more effort into
studying the effect of different AM methods on the machinability of the
materials. Only few studies were allocated to deal with the challenges in
micro-milling of additive parts. In this regard, the current investigation
aims to examine the micro-milling of EBM Ti6Al4V parts with respect to
the cutting forces, specific cutting energy, surface quality, and burr
formation. The influence of the EBM process on the machinability of
Ti6Al4V is evaluated through a comparison with extruded parts. The
size effect phenomenon in micro-cutting as one of the most important
issues is also evaluated with respect to the specific cutting energy.
Moreover, the burr formation for both extruded and EBM Ti6Al4V is
analyzed and correlated with specific cutting energy and the size effect.
2. Material
The objective titanium alloy in this investigation is Ti6Al4V. The
chemical composition of Ti6Al4V is provided in Table 1. In the current
study, Ti6Al4V titanium alloy is produced by Electron Beam Melting
(EBM) as one of the additive manufacturing techniques and its
machinability is compared with conventional Ti6Al4V manufactured by
extrusion process.
The samples were polished and etched (Etchant Kroll’s reagent,
1–3 ml HF, 2–6 ml HNO3, 100 ml DI water, 30 s) prior to micro-milling.
As illustrated in Fig. 1, the microstructure of the extruded Ti6Al4V is
much coarser than EBM Ti6Al4V. This difference highlights the effect of
the manufacturing method on the material microstructure. According to
the obtained values from micro-hardness measurement, the EBM
Ti6Al4V with 450HV has higher hardness in comparison with extruded
Ti6Al4V with 350HV. Keist and Palmer [16] studied the relation be­
tween the micro-hardness and the corresponding mechanical properties
of additively manufactured Ti6Al4V. They proposed the linear relation
between Vickers micro-hardness, yield strength, and Ultimate Tensile
Strength UTS. Therefore, it is expected that the mechanical properties of
EBM in terms of yield strength and UTS are higher than those for
Extruded titanium because of the higher hardness of EBM parts that is
attributed to their finer material microstructure.
3. Experiment
The experimental tests were carried out by a high-precision 5-axis
CNC machining center (KERN Pyramid Nano). The width of cut ae and
axial depth of cut ap were kept constant and equal to 0.3 mm and 1 mm,
respectively. The feed per tooth fz and cutting speed vc were the varying
input parameters, and cutting force Fc, specific cutting energy uc, burr
formation and surface quality were the examined parameters. Table 2
summarizes the input parameters.
Fig. 2 presents the experimental setup. The EBM and extruded
Ti6Al4V samples were in a block form with a dimension of
Table 1
Chemical composition of Ti6Al4V [15].
Chemical composition (wt%)
Al V C Fe O N H Ti
6 4 0.03 0.1 0.15 0.01 0.003 Bal
F. Hojati et al. Precision Engineering 62 (2020) 1–9

Fig. 1. Microstructure of a) EBM b) extruded Ti6Al4V.

4.8 kHz and 4.6 kHz in x, y and z direction respectively and the
Table 2
measuring area between 250 N and 250 N in all directions) with a
Experimental parameters.
sampling frequency of 300 kHz. For each test, the planar resultant force
Cutting speed Feed per Uncut chip Width of Axial depth was calculated from the measured force components in x and y di­
vc [m/min] tooth fz [μm/ thickness [μm] cut ae [mm] of cut ap
rections. To prevent the influence of tool wear on the experimental re­
tooth] [mm]
sults, the tests were randomly carried out. Moreover, each test was
35 1.25 0.9 0.3 1
repeated minimum two times, and the low variation between obtained
2.5 1.8
5 3.7 results for each test confirms the good repeatability of the experiment. In
10 7.4 order to consider the influence of tool deflection, the actual radial depth
15 11.2 of cut was also measured by the taster. According to the measurements,
20 14.9 the difference between theoretical and actual radial depth of cut is less
25 18.6
50 37.3
than 10 μm that shows approximately 3% deviation. The qualitative and
60 1.25 0.9 quantitative analysis of the workpiece surface were performed by a
2.5 1.8 confocal microscope (μsurf mobile plus - Nanofocus) at three positions of
5 3.7 the sidewall (the beginning, the middle and the end). Additionally,
10 7.4
Scanning Electron Microscopy (SEM- XL30 ESEM Philips) and optical
15 11.2
20 14.9 microscopy (Keyence VHX 5000) were used for burr formation analysis.
25 18.6
50 37.3 4. Results and discussion
80 1.25 0.9
2.5 1.8
5 3.7
As explained, different mechanical properties and microstructures
10 7.4 were detected for EBM and extruded parts. In this section, it is aimed to
15 11.2 see how the differences influence the machinability of these parts with
20 14.9 respect to the cutting force, surface quality, and burr formation.
25 18.6
50 37.3
4.1. Cutting force

30 � 20 � 2 mm and 30 � 20 � 10 mm, respectively. In this study, the
side wall of the samples was micro-milled through down milling strategy
with three flutes micro-end milling tool (WN 3142 R–N – made by
company Gühring, PVD Coating TiAlN, shaft DIN 6535-HA/HB) with
1.8 mm diameter under a wet condition (utilizing oil as coolant lubri­
cant). The cutting forces were measured by a piezoelectric dynamometer
(type 9256C2 from Kistler Company with Eigen frequencies 4 kHz,
Fig. 3 indicates the influence of cutting speed on milling forces at
different uncut chip thicknesses hcu for both extruded and EBM Ti6Al4V.
The uncut chip thickness hcu is calculated as follow [17]:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
hcu ¼
2fz ae D ae 2
(1)
D
where ae, D, and fz denote the radial depth of cut, tool diameter and feed

Fig. 2. Experimental Setup for micro-milling test.

3
F. Hojati et al.
Fig. 3. Influence of chip thickness and cutting speed for a) EBM and b)
extruded part at hcu>7.4 μm, and c) comparison of milling forces for EBM and
extruded parts at hcu<7.4 μm.
per tooth, respectively. According to Fig. 3-a and Fig. 3-b, minor dif­
ferences between milling forces for EBM and extruded Ti6Al4V can be
observed at uncut chip thicknesses between 7.4 and 37.3 μm despite the
higher hardness and yield strength of EBM parts. This phenomenon was
also observed by other researchers. Abbasi et al. [18] studied the in­
fluence of microstructure and hardness on the machinability of Ti6Al4V
with different heat treatments in the milling process. They concluded
that the alloy microstructure has an effective influence on cutting forces
in addition to the material hardness. The coarse lamellar α structure of
as-received Ti6Al4V compared to the fine lamellar αþβ structure of
heat-treated Ti6Al4V cooled in air resulted in higher cutting forces in
spite of the lower hardness of as-received Ti6Al4V. Also, Denkena and
Grove [19] investigated the microstructural effects on the machinability
of Ti6Al4V in the turning process and reported higher cutting force for
coarser lamellar microstructures, although their corresponding tensile
4
Precision Engineering 62 (2020) 1–9
strength is lower than that of finer lamellar microstructures. The results
in the mentioned studies are in good agreement with the current
investigation. The identical milling forces despite the different strengths
of the extruded and EBM material can be due to the fact that the total
force results from the energy which is consumed for pure cutting, plastic
deformation, and friction. On the one side, the higher strength of EBM
Ti6Al4V (due to its fine microstructure) in comparison with extruded
Ti6Al4V can lead to higher energy consumption for pure cutting. On the
other side, higher ductility and toughness of extruded parts (due to their
coarse microstructure) increase the degree of plastic deformation in the
chip formation process that is followed by higher energy being
consumed for plastic deformation during the chip formation process.
Consequently, higher strength of EBM parts and more ductile behavior
of extruded Ti6Al4V have both led to the same resultant milling forces in
hcu > 7.4 μm. In the case of hcu < 7.4 μm, the resultant forces of extruded
parts are higher than those for EBM parts as shown in Fig. 3-c. This could
be attributed to the effect of negative rake angle and consequently
higher plastic deformation when the chip thickness becomes smaller
than the cutting edge radius. This will be described more in detail in
section 4.
According to the constitutive law [20], the material yield point is
influenced by strain, strain rate, and temperature. During the machining
process, the interaction between these three factors affect the yield point
and consequently, the cutting force. At constant chip thickness hcu,
increasing the cutting speed leads to higher strain rates. The strain
hardening and thermal softening are two phenomena acting against
each other during a machining process which have a decisive effect on
the cutting force. No significant change in cutting force could be
observed by an increase in the cutting speed from vc ¼ 35 m/min to
vc ¼ 80 m/min at chip thicknesses below 19 μm. Thereby, it could be
supposed that the intensity of hardening and softening are the same at
different cutting speeds in performed experiments. Reduction in force by
increasing cutting speed could be only observed at hcu ¼ 37.3 μm for
both materials, which could be attributed to a dominating thermal
softening effect at higher chip thicknesses.
In additive manufacturing, each part is formed layer by layer in an
arbitrary direction so-called Built-Up Direction (BUD). Therefore, a
difference between mechanical and material properties of manufactured
parts in parallel and normal directions to the BUD may be expected due
to the material anisotropy In this regard, the micro-hardness measure­
ment was carried out in two different directions (normal and parallel to
the BUD) to evaluate the degree of anisotropy for EBM parts. The same
micro-hardness values show that the EBM parts have isotropic me­
chanical properties. Additionally, in these two directions, milling ex­
periments were conducted at different cutting speeds vc and uncut chip
thicknesses hcu, and no remarkable differences between cutting forces
were observed, as illustrated in Fig. 4.
4.2. Specific cutting energy
From the force components in x and y directions (Fx and Fy), the
specific cutting energy uc at different cutting parameters (chip thickness
and cutting speed) was calculated as below [21]:
Z tc qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
vc
uc ¼ Fx 2 þ Fy 2 dt (2)
Vrem 0
where vc, Vrem, and tc are respectively cutting speed, removal chip vol­
ume and the cutting time. Fig. 5 illustrates the relation between the
uncut chip thickness and specific cutting energy for both EBM and
extruded Ti6Al4V at different cutting speeds, vc ¼ 35, 60 and 80 m/min.
Accordingly, the influence of the size effect in the micro-milling process
can be clearly observed. Once the uncut chip thickness decreases, the
specific cutting energy non-linearly increases. For hcu < 7.4 μm, the chip
thickness is less than or comparable to the cutting edge radius of the
utilized milling tool. In this condition, the rake angle is negative, and
F. Hojati et al. Precision Engineering 62 (2020) 1–9

then the plastic deformation is considerable compared to the pure cut­

Fig. 4. Comparison between the resultant forces when milling in normal and
parallel to BUD.
ting in the chip formation process. As a result, the specific cutting en­
ergies increase. However, the proportion of the pure cutting in hcu >
7.4 μm is more significant than the plastic deformation and therefore,
lower specific cutting energies are expected. For hcu < 7.4 μm, the
specific cutting energies of extruded Ti6Al4V are 5–15% higher than
those of the EBM parts. As explained in section 4.1, the ductile behavior
of extruded parts with coarse lamellar microstructure is higher than that
of EBM parts with fine microstructure. In the chip formation process, the
material with high ductility requires more energy to reach the fracture
and be separated from the workpiece. So, larger plastic deformation of
extruded parts is expected particularly for hcu < 7.4 μm where the
plastic deformation and ploughing are dominant and may intensify the
influence of ductile behavior on energy consumption. As a consequence,
higher specific cutting energies resulted in extruded Ti6Al4V. In the case
of hcu > 7.4 μm where the pure cutting predominates over the plastic
deformation and ploughing, the minimum specific energy of uc � 2.5 J/
mm3 was obtained, and it is nearly the same for both EBM and extruded
Ti6Al4V. This indicates less effect of ductile behavior of extruded
Ti6Al4V on chip formation when hcu > 7.4 μm. In fact, the same specific
cutting energies for these two materials resulted from the balance of
higher strength of EBM Ti6Al4V and higher toughness of extruded
Ti6Al4V.

4.3. Surface quality

Fig. 5. Influence of the chip thickness on the specific cutting energy at a)
vc ¼ 35 m/min, b) vc ¼ 60 m/min, c) vc ¼ 80 m/min.
The EBM and extruded parts were machined, and produced surfaces
were compared. As illustrated in Fig. 6, the EBM parts show finer sur­
faces in most cases in comparison with extruded Ti6Al4V. The obtained
results are in good agreement with those reported by Milton et al. [4]
and Polishetty et al. [5] as described in the introduction section. This
could be attributed to the higher brittle behavior of EBM parts with
450HV Hardness compared to the extruded parts with 350HV. More
precisely, the material in front of the cutting edge either moves toward
the rake face to form the chip or is squeezed under the flank face, which
in turn results in the degradation of the machined surface. Lai et al. [9]
performed a molecular dynamic simulation to study the critical rake
angle in nanometric cutting of copper material and explained that the
material separation takes place at the stagnation point. In the case of
extruded Ti6Al4V, the stagnation point stands at a higher level
compared to the EBM parts, which contributes to the conduction of more
material under the tool and eventually more ploughing. Consequently,
the more brittle characteristics of the EBM parts leads to the less pressed
material through the ploughing on the machined surface leading to the
finer surfaces.
At cutting speeds vc ¼ 35 m/min and vc ¼ 80 m/min, the surface
roughness parameters (Sa and Sz) decrease with increasing uncut chip
thickness from hcu ¼ 0.9 μm to hcu ¼ 3.7 μm for EBM Ti6Al4V. Beyond
hcu ¼ 3.7 μm, the surface roughness increases with chip thickness up to
hcu ¼ 37.3 μm. Chen et al. [22] studied the minimum undeformed chip
thickness and size effect in micro-milling of potassium dihydrogen
phosphate crystal. They also obtained the same trend for surface
roughness by changing the undeformed chip thickness and reported this
behavior as an influence of the size effect on the surface roughness. The
decrease in the surface roughness (from hcu ¼ 0.9 μm to hcu ¼ 3.7 μm) is
related to the reduction of the ploughing effect. Where the uncut chip
thickness is higher than the cutting edge radius of the tool, the cutting
behavior seems similar to macro-milling and an increase in feed per
tooth (or undeformed chip thickness) results in higher surface roughness
(from hcu ¼ 3.7 μm to hcu ¼ 37.3 μm). The influence of the size effect on
the surface roughness can also be seen for the extruded parts. In micro
milling and for uncut chip thicknesses higher than the critical uncut chip
thickness hcu-crt, the pure cutting is a dominant mechanism while
ploughing significantly takes place for the uncut chip thicknesses less
than hcu-crt. Determination of hcu-crt with respect to the surface rough­
ness for both EBM and extruded Ti6Al4V is planned for future

5
F. Hojati et al.
Fig. 6. Surface roughness for EBM and extruded parts.
investigations.
Fig. 7 provides a comparison between the micro-milled surface of
EBM and extruded parts at cutting speed of vc ¼ 35 m/min and chip
thickness of hcu ¼ 0.9 μm, 3.7 μm and 37.3 μm. The increase of chip
thickness from 0.9 μm to 3.7 μm has not led to a remarkable change in
the surface topography. However, the surface topography has been
dramatically altered at hcu ¼ 37.4 μm. Feed marks on the machined
surface with hcu ¼ 37.3 μm can be clearly observed. High surface
roughness at hcu ¼ 37.3 μm is mainly associated with these feed marks.
The waviness of extruded parts is observable also, particularly at
hcu ¼ 37.3 μm.
Fig. 8 provides a comparison between EBM and extruded parts with
respect to the waviness. Obviously, the waviness of the extruded parts is
higher than the EBM parts in most cases. The particular high waviness of
the extruded part at hcu ¼ 37.3 μm may be attributed to the lower
hardness of the extruded part compared to the EBM part.
4.4. Burr formation
Various factors such as material, tool geometry, and machining pa­
rameters, affect the burr size and burr formation. It should be noted that
the view direction of formed burr is perpendicular to the surface “A”
shown in Fig. 9 for the following provided images.
Fig. 10 indicates the correlation between the specific cutting energy,
uncut chip thickness, and burr formation. At lower chip thicknesses
where the ploughing could be dominant and specific cutting energy is
larger, burrs were formed considerably. Whereas, no burr could be
6
Precision Engineering 62 (2020) 1–9
observed at the higher chip thicknesses, due to the reduced portion of
ploughing.
Nakayama et al. [23] investigated the burr formation mechanism to
come up with the solution for the improvement of machined part
quality. According to their findings, the main factor in the burr forma­
tion is the amount of shear deformation in the cutting zone. Higher shear
deformation causes wider and thicker burr. Therefore, any factor
influencing shear deformation affects the burr dimension indirectly.
Since the plastic deformation in the machining of EBM parts is less than
extruded parts (due to their higher hardness), it could be expected that
the narrower and thinner burr is formed in the machining of EBM parts,
leading to smaller and discontinuous burr compared to the extruded
parts.
Fig. 11 compares the burr formation of the EBM and extruded
Ti6Al4V at vc ¼ 35 m/min and hcu ¼ 0.9 μm. According to the classifi­
cation by Chern [24], the generated burr in micro-milling of extruded
Ti6Al4V is categorized as the wavy-type. In the case of EBM Ti6Al4V, the
formed burr is discontinuous and much smaller than the burr sizes of the
extruded parts. The irregularities on the EBM part surface are present up
to the depth of 70 μm. These irregularities may interrupt the burr for­
mation during the machining process that results in discontinuous burrs.
The hardness of EBM parts and the irregularities on their surface (as
produced surface) can be considered as the main source of difference
between the burr formation of EBM and extruded parts.
After removing the irregularities from the surface of EBM parts
(Surface A in Fig. 9) through a preliminary milling pass, an additional
test was carried out at hcu ¼ 0.9 μm and vc ¼ 35 m/min. Fig. 12-(a-c)
indicates the burr formation of EBM parts without the presence of ir­
regularities on the surface. The formed burr is significant and continuous
and has a wavy-shape, which is even more pronounced than the burrs of
the extruded parts. Thereby, it can be concluded that the irregularities
on the surface of additive parts have a significant influence on the burr
formation. According to Bonaiti et al. [7] who studied the machinability
of LENS parts in micro-milling, the burr formation was significant in the
AM materials while no or little burrs was detected for conventional
parts. For additive parts, they found a direct relation between the
amount of burr formation and material hardness. This is in good
agreement with the obtained result in the current work. The higher burr
formation of the additive parts despite their higher hardness, indicates
that other factors associated with the microstructure of the material may
play an essential role in the burr formation.
5. Conclusion
The current investigation was allocated to study and compare the
machinability of EBM Ti6Al4V parts with conventional parts produced
by the extrusion process based on cutting forces, specific cutting energy,
surface quality, and burr formation. The obtained results are summa­
rized as below:
� The higher strength of EBM Ti6Al4V due to its finer microstructure
and higher ductility of extruded Ti6Al4V owing to its coarser
microstructure resulted in almost the same cutting forces at chip
thicknesses between 7.4 and 37.3 μm while a 5–15% lower cutting
forces were observed for EBM parts at hcu < 7.4 μm. The cutting
force was remarkably influenced by the chip thickness. Higher chip
thicknesses resulted in higher cutting forces. Moreover, a high
deformation rate (or strain rate) through increasing the cutting speed
at constant chip thickness has not led to the higher cutting forces for
both EBM and extruded Ti6Al4V. Therefore, additively manufac­
tured and conventional parts had the same material response at
different strain and strain rates. Reduction in force by increasing
cutting speed could be only observed at hcu ¼ 37.3 μm for both ma­
terials, which could be attributed to a dominating thermal softening
effect at higher uncut chip thicknesses.
F. Hojati et al. Precision Engineering 62 (2020) 1–9

Fig. 7. Confocal microscopy of the machined surface of EBM (a–c) and extruded (d–f) parts at cutting speed of vc ¼ 35 m/min.

Fig. 9. The direction view of the burr with respect to the milling direction.
Fig. 8. The waviness of EBM and extruded parts.

specific cutting energy could be indicated. It is stated that a high

� The influence of the size effect in the micro-milling of EBM and ploughing effect through the reduction of uncut chip thickness could
extruded Ti6Al4V was evaluated with respect to the specific cutting be contributed to the considerable increase of the specific cutting
energy. A correlation between the uncut chip thickness and the

7
F. Hojati et al. Precision Engineering 62 (2020) 1–9

� The surface quality of machined parts was analyzed based on 3D
� The considerable burr formation at low uncut chip thicknesses and
Fig. 10. The correlation between specific cutting energy and burr formation for
extruded Ti6Al4V.
energy. At chip thicknesses smaller than 7.4 μm, the specific energies
of extruded parts were 5–15% higher than the EBM parts. This can be
attributed to the considerable effect of ductile behavior (plastic
deformation) of extruded parts on the energy of chip formation. At
hcu> 7.4 μm, the pure cutting regime predominates over the plastic
deformation and the minimum specific energy of uc � 2.5 J/mm3 was
obtained for both materials.
roughness parameters (Sa and Sz). As an example, at vc ¼ 80 m/min
and hcu, ¼ 0.9, 3.7 and 37.3 μm, Sa values of EBM parts are 0.75, 0.58
and 1.2 μm, respectively that are lower than those for extruded
parts. In most cases, Sa and Sz values of extruded parts are higher
than those in EBM parts. However, EBM parts indicate in some cases
higher surface roughness. But in general, it can be concluded that the
surface quality of EBM parts is in better condition compared to the
extruded parts. Due to the higher hardness of EBM parts and
consequently their higher brittle characteristic, less plastic flow on
the machined surface of the EBM parts produces the finer surface
(lower roughness and waviness values compared to the extruded
parts). The correlation between surface roughness and specific cut­
ting energy was also discussed and is planned to be studied in detail
for the future works.
simultaneously high specific cutting energies could be correlated to a
high portion of the ploughing regime. Once the uncut chip thickness

Fig. 11. Comparison of burr formation between EBM (a–c) and extruded (d–f) Ti6Al4V at hcu ¼ 0.9 μm and vc ¼ 35 m/min.

8
F. Hojati et al.
Fig. 12. A comparison of burr formation between EBM (a–c) and extruded (d–f) Ti6Al4V at hcu ¼ 0.9 μm and vc ¼ 35 m/min after removing irregularities from the
EBM part surface.
increases, the pure cutting becomes the dominant material removal
mechanism, which is accompanied by less burr formation. The
formed burr was wavious and continuous in the case of the extruded
parts, whereas the discontinuity in burrs was observed for the EBM
parts due to the irregularity on the surface. The burr formation of
EBM Ti6Al4V (after removing the irregularities) was more than the
extruded parts in spite of the higher hardness of EBM Ti6Al4V.
Acknowledgment
Thanks to CoHMed (BMBF) for the fund/Company Gühring for
providing the tools/Arcam for providing EBM manufactured parts/Prof.
Volker Bucher for SEM.
References
[1] Horn TJ, Harrysson OLA. Overview of current additive manufacturing technologies
and selected applications. Sci Prog 2012;95(Pt 3):255–82.
[2] Alexander I, Vladimir G, Petr P, Mihail K, Yuriy I, Andrey V. Machining of thin-
walled parts produced by additive manufacturing technologies. Procedia CIRP
2016;41:1023–6. https://doi.org/10.1016/j.procir.2015.08.088.
[3] Sartori S, Moro L, Ghiotti A, Bruschi S. On the tool wear mechanisms in dry and
cryogenic turning Additive Manufactured titanium alloys. Tribol Int 2017;105:
264–73. https://doi.org/10.1016/j.triboint.2016.09.034.
[4] Milton S, Morandeau A, Chalon F, Leroy R. Influence of finish machining on the
surface integrity of Ti6Al4V produced by selective laser melting. Procedia CIRP
2016;45:127–30. https://doi.org/10.1016/j.procir.2016.02.340.
[5] Polishetty A, Shunmugavel M, Goldberg M, Littlefair G, Singh RK. Cutting force
and surface finish analysis of machining additive manufactured titanium alloy Ti-
6Al-4V. Procedia Manuf. 2017;7:284–9. https://doi.org/10.1016/j.
promfg.2016.12.071.
[6] Le Coz G, Fischer M, Piquard R, D’Acunto A, Laheurte P, Dudzinski D. Micro
cutting of Ti-6Al-4V parts produced by SLM process. Procedia CIRP 2017;58:
228–32. https://doi.org/10.1016/j.procir.2017.03.326.
[7] Bonaiti G, Parenti P, Annoni M, Kapoor S. Micro-milling machinability of DED
additive titanium Ti-6Al-4V. Procedia Manuf. 2017;10:497–509. https://doi.org/
10.1016/j.promfg.2017.07.104.
[8] Altintas Y, Jin X. Mechanics of micro-milling with round edge tools. CIRP Annals
2011;60(1):77–80. https://doi.org/10.1016/j.cirp.2011.03.084.
[9] Lai M, Zhang XD, Fang FZ. Study on critical rake angle in nanometric cutting. Appl
Phys A 2012;108(4):809–18. https://doi.org/10.1007/s00339-012-6973-8.
[10] Chae J, Park SS, Freiheit T. Investigation of micro-cutting operations. Int J Mach
Tool Manuf 2006;46(3–4):313–32. https://doi.org/10.1016/j.
ijmachtools.2005.05.015.
Precision Engineering 62 (2020) 1–9
[11] Simoneau A, Ng E, Elbestawi MA. Chip formation during microscale cutting of a
medium carbon steel. Int J Mach Tool Manuf 2006;46(5):467–81. https://doi.org/
10.1016/j.ijmachtools.2005.07.019.
[12] Uriarte L, Azc� arate S, Herrero A, Lopez de Lacalle LN, Lamikiz A. Mechanistic
modelling of the micro end milling operation. Proc Inst Mech Eng B J Eng Manuf
2008;222(1):23–33. https://doi.org/10.1243/09544054JEM837.
[13] Arnaiz-Gonz� alez A,
� Fern�andez-Valdivielso A, Bustillo A, L�
opez de Lacalle LN. Using
artificial neural networks for the prediction of dimensional error on inclined
surfaces manufactured by ball-end milling. Int J Adv Manuf Technol 2016;83(5–8):
847–59. https://doi.org/10.1007/s00170-015-7543-y.
[14] Lacalle LL de, Lamikiz A, S� anchez JA, Bustos IF de. Recording of real cutting forces
along the milling of complex parts. Mechatronics 2006;16(1):21–32. https://doi.
org/10.1016/j.mechatronics.2005.09.001.
[15] F04 Committee. Specification for wrought titanium-6aluminum-4vanadium alloy
for surgical implant applications (UNS R56400). West Conshohocken, PA: ASTM
International. https://doi.org/10.1520/F1472-08E01.
[16] Keist JS, Palmer TA. Development of strength-hardness relationships in additively
manufactured titanium alloys. Mater Sci Eng A 2017;693:214–24. https://doi.org/
10.1016/j.msea.2017.03.102.
[17] Black SC, editor. Principles of engineering manufacture. third ed. London: Arnold;
1996.
[18] Abbasi SA, Feng P, Ma Y, Zhang J, Yu D, Wu Z. Influence of microstructure and
hardness on machinability of heat-treated titanium alloy Ti-6Al-4V in end milling
with polycrystalline diamond tools. Int J Adv Manuf Technol 2016;86(5–8):
1393–405. https://doi.org/10.1007/s00170-015-8245-1.
[19] Denkena B, Grove T. The effect of microstructure on the machinability of Ti-6Al-
4V. In: Venkatesh V, editor. Proceedings of the 13th World conference on titanium:
sponsored by titanium committee of the structural materials division of the
minerals, metals & materials society (TMS), held august 16-20, 2015, manchester
grand hyatt, san diego, California, USA/edited by Vasisht Venkatesh [and more].
Hoboken, New Jersey: Wiley; 2016. p. 905–10.
[20] Kushner V, Storchak M. Modelling the material resistance to cutting. Int J Mech Sci
2017;126:44–54. https://doi.org/10.1016/j.ijmecsci.2017.03.024.
[21] Lauro CH, Brand~ ao LC, Carou D, Davim JP. Specific cutting energy employed to
study the influence of the grain size in the micro-milling of the hardened AISI H13
steel. Int J Adv Manuf Technol 2015;81(9–12):1591–9. https://doi.org/10.1007/
s00170-015-7321-x.
[22] Chen N, Chen M, Wu C, Pei X, Qian J, Reynaerts D. Research in minimum
undeformed chip thickness and size effect in micro end-milling of potassium
dihydrogen phosphate crystal. Int J Mech Sci 2017;134:387–98. https://doi.org/
10.1016/j.ijmecsci.2017.10.025.
[23] Nakayama K, Arai M. burr formation in metal cutting. CIRP Annals 1987;36(1):
33–6. https://doi.org/10.1016/S0007-8506(07)62547-5.
[24] Chern GL. Analysis of burr formation and breakout in metal cutting [PhD].
Berkeley: University of California; 1993.