A Numerical Investigation and Experimental

Validation on Chip Morphology of Aluminum Alloy

6061 during Orthogonal Machining
Sohail Akram, Husain Imran, Mushtaq Khan, Aamir Mubashar, Salman Warsi, Umair Riaz
School of Mechanical and Manufacturing Engineering NUST
Islamabad, Pakistan
E-mail: [email protected]

Abstract— In this research work, Aluminum alloy Al6061-T6 life, and machining efficiency[3]. Many researchers have
was machined in order to investigate the chip morphology at studied the chip mechanism for different materials. Lida Zhu et
different cutting parameters. Chip morphology with different al. [3] presented a novel experimental turn-milling orthogonal
shapes and sizes affects various aspects such as cutting force, machining process in which both workpiece and tool were
chatter stability and cutting temperature which in turn influences rotating, to investigate 3D chip morphology of Al6061-T6.
workpiece surface quality, tool life, and machining efficiency. Chip length, chip thickness and micro-hardness variations were
Orthogonal cutting experiments were carried out with the studied at cutting speed of cutter 78-312m/min, cutting speed
CCMW 09 T3 04-H13A uncoated carbide tool in the cutting of workpiece 1.25-5.0m/min and axial feed velocity of 20-
speed range of 250–1,000 m/min and feed rate of 0.1–0.4 mm/r to
80mm/min and found that chip length contrary to conventional
obtain the chip thicknesses and the shear angles. Commercially
available software Abaqus/Explicit was used to simulate a 2D
turning decreased with the increased cutting speed of cutter
orthogonal model. The simulation results were found to be in whereas it increased with increase of axial feed velocity. Chip
close agreement with the experimental results and showed that thickness decreased with the increase of cutting speed of cutter
chip thickness decreased with increasing cutting speed whereas and increased with the increase of axial feed velocity which
shear angle increased with the increase of cutting speed. ultimately affected the feed rate. Moreover, micro-hardness
was found higher near the machined surface compared to
Keywords: Finite Element Method, Machining, Orthogonal immediate subsurface whereas it was found relatively stable in
Cutting, chip morphology. the bulk chip. Also, average micro-hardness was lower in the
continuous chip compared to serrated chip. J. Shi and C. R. Liu
I. INTRODUCTION [4] simulated a FEM model in ABAQUS/Explicit in order to
High speed machining is widely used in manufacturing predict chip production and phase transformation during
industry to remove material at higher rates by reducing orthogonal machining of hardened AISI 52100 steel (62HRC)
production lead-times, having lower cutting forces, higher using Polycrystalline Cubic Boron Nitride (PCBN) tools and
precision and excellent surface finish [1].Recently Aluminum simulated the formation of segmental chips and predicted the
alloys, especially 6xxx series due to medium strength, transformations in the phases by matching the increased cutting
corrosion resistance, formability, weldability and lower cost are temperature needed for phase transformation with the white
expected to provide substitution for steels by providing great layer areas present on the actual chips. Ravi Shankar et al. [5]
improvements in energy economy, recyclability and life-cycle analyzed the grain size and micro-hardness of chips produced
cost [2]. Aluminum alloy Al 6061-T6 being a non-ferrous by machining Al6061-T6 at low velocity of 10mm/s and tool
metal is extensively used in aircraft, submarine and automotive rake angle of +50 and -200 and found the microstructure of
industry due to good toughness characteristics, excellent chips to be composed of equi-axed grains with mean size
corrosion resistance, good acceptance of applied coatings, ranging from 70-100nm whereas the Vicker hardness to be
relatively high strength, good workability, good finishing 150± 3HV and 158±3HV, respectively. The grain size and
characteristics, easy weldability and due to wide availability. Vickers hardness of the bulk workpiece material were found to
be ~75µm and 110kg/mm2. Daochun Xu et al. [6] developed a
Types of chips in metal cutting depend upon the workpiece theoretical model for high-speed orthogonal cutting of Al6061-
material, machining parameters and tool geometry and play a T6 to calculate important chip parameters of chip formation
vital role in the optimization of cutting process and surface and including chip thickness, shear angle, friction angle, length of
subsurface integrity of workpiece material. Chip formation and shear plane, width of first shear zone, shear slip distance, tool-
its morphology, generally accepted as the result of shear chip contact length, cutting and thrust forces at cutting speed of
deformation in the shear zone and tool-chip friction, are 100-1900m/min and feed rate of 0.06-0.15mm/r. Increasing
important characteristics of metal cutting and provide speed decreased chip thickness, friction angle, length of shear
important information on the machining process itself. Chip plane, width of first shear zone, tool-chip contact length,
morphology having different shapes and sizes influences cutting and thrust forces whereas shear angle and shear slip
various aspects such as machining temperature, chatter stability distance increased with the increase of speed. Moreover,
and required cutting forces which affects surface quality, tool

978-1-5090-0645-8/16/$31.00 ©2016 IEEE

331
increasing speed produced ribbon chips whereas curling chips damage and failure model and a thermal analysis incorporating
were produced with the increase of feed. the heat dissipated from inelastic deformation energy and
friction were used in the fully coupled thermal-mechanical FE
Discontinuous chips in high-speed machining of AISI 4340 model. The results of the chips produced in the current FE
(32HRc) were simulated and experimentally verified by Y.B. model were first validated against the experimental results
Guo and David W. Yen [7] using Johnson–Cook (JC) plasticity reported by Daochun Xu et al. [6] and were found in close
model for workpiece material properties and material crack agreement. The FEM model was then modified according to
initiation and propagation using the Johnson–Cook damage present study.
model and found that discontinuous chips were produced due
to the internal crack initiation and propagation in front of the III. 2D MATERIAL MODELING:
tool and above the cutting edge instead of the free surface.
Since machining being an extremely dynamic process with
In this study, a fully coupled thermal-mechanical FE model remarkable changes occurring in geometry, Abaqus/Explicit
was developed in ABAQUS/Explicit to investigate the chip software has been selected as it takes temperatures and
morphology of Al6061-T6 at cutting speed of 250-1000m/min displacements as nodal variables during coupled temperature-
and feed rate of 0.1-0.4mm/rev. Yujie Chen [8] had compared displacement calculations which results in better mesh aspect
explicit and implicit schemes with CAE results in terms of ratio by using adaptive remeshing technique. The numerical
CPU time cost,, Von-Mises stress and internal energy from model was developed in ABAQUS/Explicit with Al6061-T6 as
history field output for one element in orthogonal cutting and workpiece material and elastic carbide tool. Tool and
found that total internal energy calculated from the above workpiece geometry, material properties and mesh were
mentioned 03 methods didn’t show significant difference. generated in the modules. When the model is completed,
Therefore, explicit scheme was preferred over implicit scheme ABAQUS/CAE generates an input file that is submitted to the
due to convergence problems of implicit scheme as CPU time ABAQUS analysis product which generates desired output
was more to calculate whereas Von-Mises stress (Field output) results in post-processing. The adaptive meshing referred to as
and internal energy (history output i.e. to capture the whole Arbitrary Lagrangian-Eulerian (ALE) analysis that combines
deformation process instead of final results at the end of the the features of pure Lagrangian analysis and pure Eulerian
time step as in field output). Since the machining process is a analysis were employed here. ALE adaptive meshing was used
short time process with extreme non-linear plastic deformation, to maintain a high-quality mesh throughout the analysis, even
converge is a potential issue with implicit scheme, therefore when large deformation or loss of material occurs, by allowing
explicit scheme was preferred in the current research. the mesh to move independently of the material [9].
The objective of the present study is to investigate the chip Johnson-Cook (JC) material model, suitable for the cases
morphology in the work-piece (Al-6061-T6) and its sensitivity having strain-rate variation over a large range (102-106 s-1),
to cutting speed and feed rate by using numerical and had been employed for predicting large plastic deformation
experimental methods. fulfilling material damage and failure criteria. JC constitutive
model describes the flow stress of the workpiece material in
II. NUMERICAL SIMULATION terms of strain, strain-rate and thermal effects. JC constitutive
Understanding the fundamentals of machining operations model gives equivalent stress in plastic region by (1):
through experimental studies generally has some limitations
e.g. cost & time. Machining exhibits large non-linearity due to ̅ −
large plastic deformations, damage initiation and material =( + ̅ ) 1+ 1− (1)
̅ −
failure beyond elastic limit. Machining operation also involves
frictions and complex contact phenomenon between work-
Where the first term in the above equation shows stain
piece and tool material which plays a vital role in predicting the
effect, second term shows strain-rate effect whereas the last
behavior of cutting simulation. ABAQUS/Explicit deals with
term shows thermal effect. Moreover, is the equivalent stress,
such complex mechanism in an efficient way, therefore a large
̅ the equivalent plastic strain, n the work-hardening exponent,
number of researchers deal with this FEM code in order to
simulate machining phenomenon to predict various parameters ̅ the plastic strain rate, ̅ the reference strain rate (1.0 s-1), A
of work-piece e.g. chip morphologies, cutting forces involved, is the initial yield stress, B is the hardening modulus, , C the
thermal effects and surface integrity along with tool wear of strain rate dependency coefficient, m is the thermal softening
actual machining. Therefore, numerical modeling and coefficient, Troom is the room temperature and Tmelt is the
simulation is usually carried out to reduce experimentation or melting temperature. The constants in the above equation are
rather eliminate it. A FEM model was developed in the present usually obtained by means of dynamic Hopkinson bar tensile
research to simulate the chip formation during orthogonal test over a wide range of temperature. Table I shows the values
machining of Al6061-T6 using uncoated carbide insert to of different parameters used in Johnson-Cook material
investigate the effects of different cutting velocities and feed constitutive model which were taken from the literature. The
rates on the chip morphology in the work piece material. Johnson–Cook damage model based on shear failure was used
in conjunction with the Johnson–Cook plasticity model and is
A chip separation criterion based on critical equivalent based on the value of equivalent plastic strain at element
plastic strain; a Coulomb’s law for the friction at the tool/chip integration points. Failure is assumed to occur when the
interface; a Johnson-Cook material constitutive suitable for damage parameter exceeds one. JC model damage and failure
problems where strain-rates vary over a large range; JC follows a cumulative damage law and is given as (2):

332
 ∆ ̅
= (2)
̅

Where, ∆ε shows the increment of the equivalent plastic

strain and ε is the equivalent strain at failure. According to the
Johnson-Cook damage model, the cumulative strain ∆ε is
updated at every analysis increment. ̅ is given by (3):

̅
̅ = + 1+
̅
− Fig. 1. Geometry illustrating workpiece, tool, chip and shear angle
× 1+ (3)
−
workpiece and tool. The workpiece was fixed in x-
Where the first term is pressure dependent, second term is direction as well as y-direction whereas the tool was allowed to
strain-rate effect and the last term is thermal effect. The values move in the x-direction only with the prescribed velocity while
of damage constants (D_(i, where 1≤ i ≤ 5)) are taken from keeping the y- direction fixed. The results of chip morphology
the literature as given in Table II. from simulation were obtained and then compared with the
experimental results. Table III shows the experimental and
Friction between tool and chip is one of the most complex simulated chip morphology at v = 1000m/min and f = 0.1-
phenomenons of machining process. It not only influences the 0.4mm/rev.
tool wear and quality of machined surface but also chip
geometry, cutting temperature and power required to remove a IV. EXPERIMENTATION
certain volume of metal. In this research the simplest friction Experiments were performed to validate the developed
model i.e. Coulomb’s friction model was used to model finite simulation model. Orthogonal tube turning method was used to
element simulation. The Model is as shown in Fig.1. simulate the orthogonal machining process. Orthogonal
machining is basically a 2D machining operation and serves as
The aim was to predict the chip thickness and shear angle
the guideline for development of models related to forces,
in the machined component under various cutting speeds and
stresses, energy and other aspects of the machining process
feed-rates while keeping the tool geometry and cutting depth
constant. The Boundary conditions were applied for the both Workpiece is plastically deformed in orthogonal machining
operation in shear plane to produce the chip. For this purpose,
TABLE I. JOHNSON COOK MATERIAL CONSTANTS FOR AL-6061-T6 Al 6061 pipes with uniform wall thickness of 4mm were cut
[10],[11]
with a cutting tool that was fed in a perpendicular direction to
Material Property Value the workpiece. The cutting edge of the tool was sharp and its
Density 27000 kg/m3
length exceeded the wall thickness of pipes. Under this cutting
Young's Modulus 68.9 GPa
arrangement, the width of cut is essentially equal to the wall
Poisson's ratio, ν 0.33
thickness of the pipe and the undeformed chip thickness is
A 324 MPa
equal to the cutting feed of the tool. Table IV shows the
B 114 MPa
chemical properties of Al 6061 workpieces and Table V shows
the specifications of Sandvik supplied cutting tool used in the
N 0.42
experimentation.
C 0.002
̅ 1 sec-1 Table VI shows the cutting condition used for experimental
planning. Full factorial design of experiments (DOE) was used
TABLE II. MATERIAL MODEL FAILURE CONSTANTS [10],[11] to develop 16 experimental trials. Each trial was repeated three
times to ensure the repeatability of the experiments. All the
D1 -0.77 trials were performed in dry cutting environment.
D2 1.45 During the experimentation, chips of each cutting condition
D3 -0.47 were collected and processed for metallographic analysis
through optical microscope. Thickness and form for every chip
D4 0.0
was noted. Parameters like chip thickness ratio (r) and shear
D5 0.0 angle (φ) were computed. Chip thicknesses were
experimentally calculated through optical microscope after
grinding and polishing the chip samples. Shear angles were
calculated through (4):

333
 TABLE III. CHIP MORPHOLOGY AT V=1000M/MIN, F= 01-0.4MM/REV

S.No. Cutting Parameters Chips obtained through experimentation Simulated chips

Cutting speed = 1000m/min

1.
Feed-rate = 0.1mm/rev

Cutting speed = 1000m/min

2.
Feed-rate = 0.2mm/rev

Cutting speed = 1000m/min

3.
Feed-rate = 0.3mm/rev

Cutting speed = 1000m/min

4.
Feed-rate = 0.4mm/rev

334
 TABLE IV. CHEMICAL PROPERTIES OF AL 6061-T6
Si Fe Cu Mn Mg Cr Zn Ti Al = = (5)
% 0.62 0.22 0.29 0.07 1.1 0.18 0.01 0.01 Bal.

TABLE V. SPECIFICATIONS OF CUTTING TOOL AND HOLDER V. RESULTS AND DISCUSSION

Specifications Description The thickness and shear angle of each thickness was
Tool Holder SCACL 1616 K 09-S measured 05 times at different points for averaging and are
Insert CCMW 09 T3 04-H13A shown in the Tables VII and VIII. With the cutting speed of
Type Uncoated without chip breaker 250-10000m/min and feed of 0.1–0.4mm/r, chip thickness
Cutting Speed (m/min) 250-1000 ranged from 191 to 762μm experimentally whereas it came
hex (mm) ≈ fn (mm/rev) 0.1-0.4 out to be 158 to 670μm. Chip thickness was found to
decrease with the increase of cutting speed and increased
TABLE VI. LEVELS OF CUTTING PARAMETERS USED FOR DESIGN OF with increasing feed because the change of either cutting
EXPERIMENTS
speed or feed would result in the change of shear angle. The
Parameters Level 1 Level 2 Level 3 Level 4 shear angle was calculated by substituting the chip thickness,
Feed (mm/rev) 0.1 0.2 0.3 0.4 cutting thickness, and rake angle into Eq. 4 and Eq. 5. The
Cutting Speed (m/min) 250 500 750 1000
results are shown in table VIII. With 250-1000m/min cutting
speed and 0.1-0.4mm/r feed, the variation in shear angle was
= (4) ranged from 22.37° to 35.01° experimentally whereas
1− simulated results ranged from 25°-35°.
Here is the rake angle of the tool and is defined as Unlike chip thickness, shear angles were increased with
chip thickness ratio and is expressed as the ratio of feed or increasing cutting speed and feed rate. The measured %
un-deformed chip thickness “ ” to actual thickness of chip errors are also shown in Table VII and VIII. Discrepancies in
" ", Rake angle of the inserts in this research work was 0o. the results were mainly due to subjectivity of measuring
Table III shows the microscopic images of some of the chips techniques at different locations of chips and since the values
collected experimentally during the orthogonal machining at were in microns.
various cutting parameters. Finally, (4) above for shear angle
can be rewritten as (5): TABLE VIII. EXPERIMENTAL AND SIMULATED RESULTS OF SHEAR
ANGLE

TABLE VII. EXPERIMENTAL AND SIMULATED RESULTS OF CHIP Shear

THICKNESS Cutting Shear angle % error
S.No. Feed angle
Speed (Simulated) (Approx.)
(Exp,)
Chip Chip
S Cutting % error
Feed Thickness Thickness 1 250 0.1 24.44 25 2.29
.No. Speed (Approx.)
(Exp) (Simulated)
2 500 0.1 25.46 26 2.12
1 250 0.1 220 175 20.45
3 750 0.1 26.57 28 5.38
2 500 0.1 210 167 20.48
4 1000 0.1 27.63 31.5 14.01
3 750 0.1 200 165 17.50
5 250 0.2 22.37 27 20.70
4 1000 0.1 191 158 17.28
6 500 0.2 29.05 29.5 1.55
5 250 0.2 486 390 19.75
7 750 0.2 29.40 30.5 3.74
6 500 0.2 360 360 0.00
8 1000 0.2 34.14 32 6.27
7 750 0.2 355 343 3.38
9 250 0.3 26.15 28 7.07
8 1000 0.2 295 323 9.49
10 500 0.3 31.63 30 5.15
9 250 0.3 611 520 14.89
11 750 0.3 32.83 31 5.57
10 500 0.3 487 485 0.41
12 1000 0.3 33.63 32.5 3.36
11 750 0.3 465 476 2.37
13 250 0.4 27.70 28.5 2.89
12 1000 0.3 451 461 2.22
14 500 0.4 32.17 32 0.53
13 250 0.4 762 670 12.07
15 750 0.4 34.18 34.5 0.94
14 500 0.4 636 625 1.73
16 1000 0.4 35.01 35 0.03
15 750 0.4 589 587 0.34
16 1000 0.4 571 561 1.75

335
 VI. CONCLUSIONS [9] N. Sawarkar, “Finite Element based Simulation of Orthogonal
Cutting Process to Determine Residual Stress Induced,” pp. 33–38,
This study showed that: 2014.
[10] D. R. Lesuer, G. J. Kay, and M. M. LeBlanc, “Modeling large-strain,
Finite element modeling is helpful to simulate the high-rate deformation in metals,” Third Bienn. Tri-Laboratory Eng.
predicted chip morphology by using a JC material and JC Conf. Model. Simul., pp. 3–5, 2001.
damage model. Under the selected range of cutting velocities [11] L. E. Schwer, “Aluminum plate perforation: a comparative case study
and feed-rates, the chip thickness decreased with the increase using Lagrange with erosion, multi-material ALE, and smooth
of cutting speed and increased with the increase of feed rate. particle hydrodynamics.,” 7th Eur. LS-DYNA Conf., 2009. Eason, B.
Shear angles were also found to be sensitive to cutting Noble, and I.N. Sneddon, “On certain integrals of Lipschitz-Hankel
type involving products of Bessel functions,” Phil. Trans. Roy. Soc.
parameters and increased with the increase of cutting speed London, vol. A247, pp. 529-551, April 1955.
and feed rate. FEM model and experimental results were in
close agreement especially at higher cutting speed.
The difference in experimental and simulated results was
attributed mainly to the subjectivity of measuring technique
at different chip locations and extremely small values of chip
thickness, in microns.
However, a quantitative relationship between the chip
parameters, cutting forces, tool wear-rate and surface/surface
damage characterization under different cutting conditions
needs a further study in the future in order to have a
comprehensive predictive metal cutting model which can
correlate surface/subsurface integrity with the cutting
parameters for an efficient and reliable machined component.

ACKNOWLEDGMENT
The authors would like to thank Higher Education
Commission, Pakistan for the funding under the indigenous
scholarship programme at National University of Science
and Technology (NUST), Islamabad, Pakistan.

REFERENCES
[1] J. D. P. Velásquez, A. Tidu, B. Bolle, P. Chevrier, and J.-J.
Fundenberger, “Sub-surface and surface analysis of high speed
machined Ti–6Al–4V alloy,” Mater. Sci. Eng. A, vol. 527, no. 10–11,
pp. 2572–2578, 2010.
[2] S. H. Lee, Y. Saito, T. Sakai, and H. Utsunomiya, “Microstructures
and mechanical properties of 6061 aluminum alloy processed by
accumulative roll-bonding,” Mater. Sci. Eng. A, vol. 325, no. 1–2, pp.
228–235, 2002.
[3] L. Zhu, X. Jin, and C. Liu, “Experimental investigation on 3D chip
morphology properties of rotary surface during orthogonal turn-
milling of aluminum alloy,” Int. J. Adv. Manuf. Technol., 2015.
[4] J. Shi and C. R. Liu, “On predicting chip morphology and phase
transformation in hard machining,” Int. J. Adv. Manuf. Technol., vol.
27, no. 7–8, pp. 645–654, 2006.
[5] M. Ravi Shankar, S. Chandrasekar, W. D. Compton, and A. H. King,
“Characteristics of aluminum 6061-T6 deformed to large plastic
strains by machining,” Mater. Sci. Eng. A, vol. 410–411, pp. 364–
368, 2005.
[6] D. Xu, P. Feng, W. Li, Y. Ma, and B. Liu, “Research on chip
formation parameters of aluminum alloy 6061-T6 based on high-
speed orthogonal cutting model,” Int. J. Adv. Manuf. Technol., vol.
72, no. 5–8, pp. 955–962, 2014.
[7] Y. B. Guo and D. W. Yen, “A FEM study on mechanisms of
discontinuous chip formation in hard machining,” J. Mater. Process.
Technol., vol. 155–156, pp. 1350–1356, 2004.
[8] Y. Chen, “Prediction of Subsurface Damage During Machining
Nickel-Based Superalloys,” 2013.

336