Journal of Mechanics Engineering and Automation 3 (2013) 441-448 D DAVID PUBLISHING

Chip Formation in Micro-cutting

François Ducobu, Edouard Rivière-Lorphèvre and Enrico Filippi

Machine Design and Production Engineering Department, FPMs (Faculty of Engineering), UMONS (University of Mons), Mons
B-7000, Belgium

Received: March 29, 2013 / Accepted: April 25, 2013 / Published: July 25, 2013.

Abstract: The miniaturisation context leads to the rise of micro-machining processes. Micro-milling is one of the most flexible and fast
of them. Although it is based on the same principles as macro-cutting, it is not a simple scaling-down of it. This down-sizing involves
new phenomena in the chip formation, such as the minimum chip thickness below which no chip is formed. This paper presents a
review of the current state of the art in this field from an experimental and a numerical point of view. A 2D finite element model is then
developed to study the influence of the depth of cut on the chip formation. After the model validation in macro-cutting, it highlights the
phenomena reported in literature and allows to perform a minimum chip thickness estimation.

Key words: Chip formation, micro-cutting, minimum chip thickness, orthogonal cutting, saw-toothed chip, Ti6Al4V.

1. Introduction between micro and macro-cutting are presented, as

Micro-milling is a micro-manufacturing technology
by removal of material with a miniature cutting tool. It
allows to produce pieces and features ranging from
several mm to several µm, which is very interesting in
the current context of miniaturization. Micro-milling is
the most flexible and fastest way to produce complex
tridimensional micro-forms including sharp edges and
a good surface finished in many materials: metallic
alloys, composites, polymers and ceramics [1-2]. Its
applications are very varied. Micro-injection moulds,
watch components, optical devices, components for the
aerospatial, biomedical and electronic industry are a
few examples.
The micro-cutting phenomenon is not a direct scaling
of macro-cutting, although micro-milling is based on
macro-milling. The down-sizing of the process implies
some changes in the cutting phenomenon.
This paper provides the current state of the art of
chip formation and minimum chip thickness in
micro-cutting from an experimental and numerical
point of view. The main differences and difficulties
Corresponding author: François Ducobu, Ph.D., teaching
assistant, research fields: machining by chips removal and
machine design. E-mail: Francois.Ducobu@umons.ac.be.
well as experimental and numerical work in this field.
A numerical model is thus developed to study the chip
formation in micro-cutting.
The paper is organized as followed: Section 2 deals
with micro-cutting chip formation; Section 3 presents
the numerical model; Section 4 is about the model
validation in macro-cutting; Section 5 presents the
results in micro-cutting and it ends with the
conclusions at Section 6.
2. Chip Formation
2.1 Minimum Chip Thickness
The main difference between macro and
micro-cutting cutting processes concerns the chip
formation involving the so-called “minimum chip
thickness” phenomenon [1]. It can be quickly
explained like this: in micro-cutting the depth of cut
reduces dramatically and reaches the same order of
magnitude as the tool edge radius. If it becomes smaller
than a certain depth, i.e., the minimum chip thickness,
no chip is formed as shown in Fig. 1.
Depending on the material, Chae et al. [1] evaluate it
between 5% and 38% of the tool edge radius.
442 Chip Form
mation in Mic
cro-cutting

Fig. 1 Scheematic representation of th he minimum chip

thickness in orthogonal
o cuttting (r: edge raadius of the toool, h:
Fig.. 2 Schematicc representatioon of the negattive rake anglee
depth of cut, hm: minimum chip
c thickness)), inspired from
m Ref.
in orthogonal cuttting, inspired ffrom Ref. [5].
[1].
Ducobu
D et al.. [4] show thhat the introdu uction of thee
The minimum chip thhickness leadds to a risingg of
sizee effect and the minimum m chip thick kness in theirr
slipping forcces and plougghing of the machined
m surrface
dynnamic model leads
l to an inncrease in the cutting forcee
(highlightedd by Bissacco et al. [3]), coontributing inn the
andd vibrations. They
T even beecome too high comparedd
increase of cutting
c forcess, burrs formaation and surrface
to th
he required precision.
p
roughness.
milling, Ducobu et al. [4]] show that when
In micro-m w 2.4 Influence of the Machinedd Material
the feed perr tooth is sm
maller than thhe minimum chip
Moreover
M at this scale, tthe microstruucture of thee
thickness, seeveral tool rootations are needed
n to forrm a
matterial takes immportance andd the granulaar structure off
chip. They also
a show thaat the stabilityy of the operaation
the workpiece material
m musst be taken into
i account..
can be feed dependent, coontrary to maacro-milling.
Indeed, in micro-cutting, as thhe dimensions of the depthh
2.2 Negativee Rake Angle of cut,
c the tool or o feature to pproduce are often
o smallerr
thann the grain size of the macchined material, its naturee
The macroo-cutting assuumption stating that the toool is andd micro-granuular structuree have to be taken intoo
sharp, comppletely cuts thhe surface andd generates chips
c account, as highllighted in Reffs. [1, 5]. Theerefore, it cann
is not valid in micro-cuttting. This is due
d to the higghly no longer
l be connsidered as hoomogeneous anda isotropic,,
effective neggative rake anngle caused by
b the small depth
d conntrary to the assumption
a m
made in macro o-cutting. Thee
of cut being of the same order
o of magnnitude as the tool miccrostructure ofo the machineed material taakes thereforee
edge radius (Fig. 2). a grreat importannce in micro-ccutting.
A highly negative rake angle leadss to ploughinng of Chae
C et al. [1] and Doornfeld et all. [5] reportt
the machineed surface annd elastic spring back off the variiations in cutting forcess and vibraations duringg
workpiece. The spring back
b fraction occurring unnder miccro-machiningg due to the laack of homog geneity of thee
the flank faace leads to friction, raissing the specific worrkpiece granuular structuree. This leads to variationss
cutting energgy. in cutting coonditions (hhardness in particular)..
Modifying the cuutting conditiions or the maachine designn
2.3 Size Effeect
is not
n a solutionn to eliminatee them, as theey are due too
At a smalll depth of cuut, Filiz et al. [6] observedd the the nature of thee machined m material. Finaally, averagedd
so-called “siize effect”: a decrease in the depth off cut cuttting coefficieents from maccro-cutting caannot be usedd
leads to a nonlinear
n inccrease in thee specific cuttting anyy more.
energy. Minnimum chip thickness
t andd specific cuttting
2.5 Finite Elemeent Modellingg
energy are thus closely related. Thee specific cuttting
energy coulld be an inddicator makinng it possiblle to Up
U to now, veery few finitee element moddelling aboutt
detect channges (from slipping
s to shearing)
s in the miccro-cutting caan be foundd in the literrature. Nonee
cutting mechhanism and too monitor thee process. deaaling with the Ti6Al4V titaanium alloy were
w found.
 Chip Form
mation in Mic
cro-cutting 4433

A 2D ALE (Arbittrary Lagranngian Eulerrian)
orthogonal cutting
c finite element
e moddel was develooped
by Woon et al. [7] in ordeer to study thee influence off the
tool edge radius on chip c formatioon. This model
m An
considers thhe workpiece material (AISI 4340 steel) as
homogeneouus and the toool is modelled as a perfeectly
w and withoout edge radiuus. The results of
rigid solid with
their researcch show that thhe chip is formed by extruusion
along the toool edge radiius when thee depth of cuut is
lower than a breaking vaalue and conffirm that the tool
cannot be coonsidered shaarp in micro-ccutting.
Liu et all. [8-9] deveeloped a 2D D strain graddient
plasticity-baased finite ellement modeel in which they
introduce material
m strengthening mechanisms to
explain thee size effeect in miccro-cutting. The
simulations were perform med with an aluminium alloy a
(Al5083-H116) exhibitinng a small straain rate hardenning
exponent, thhus minimizzing strain raate effects. Two T geoometry and booundary condditions of thee model. Thee
main strengtthening factoors were conssidered: the sttrain
gradient strrengthening and the cuttting temperaature
decrease in the
t secondaryy shear zone, with decreasse in
the depth of o cut. It shows that thee strain graddient
strengtheninng is the dominant mechanism in
micro-cuttinng conditionss. The tool edge radiuss of
course playss a role in the size effect buut only a part of
o it.
In order to study thhe influence of the grannular
structure of the machined material onn chip formattion,
Simoneau et e al. [10] developed
d a 2D Lagranngian
orthogonal cutting heterrogeneous (A AISI 1045 stteel)
finite elemennt model withh a sharp tool.. They observved a
new chip foormation mecchanism: thee softest mateerial
(ferrite) is extruded between
b the hardest grrains
(pearlite). Thhey called it a “quasi-shearr extrusion chhip”.
This shows that it is cruucial to model the workppiece
material as heterogeneoous in microo-cutting of this
material.
Lag grangian orthhogonal cutting model is i developedd
usinng the finitte element method thaanks to thee
commmercial software ABAQU US/Explicit v6.8.
v
A importantt characteristic of this model m is itss
valiidity not onlyo in miccro-cutting but b also inn
maccro-cutting. ThisT allows to study chaanges in thee
cuttting mechanism from maccro- to micro-cutting withh
onee single moddel. The abillity to form saw-toothedd
chipps in macro-ccutting is onee of the requirements (andd
diffficulties) introoduced by thee multi-scale aspect of thee
mod del.
The
T model onnly takes intoo account the area close too
the cutting edge of the tool, w where the chhip is formed..
Thee cutting tooll is modelledd with a finitee edge radiuss
andd it is taken innto account thanks to the fine mesh off
the workpiece. Its I rake and clearance an ngles are set,,
resppectively, to 15° and 2°. Fig. 3 presen nts the initiall
cuttting speed, Vc, is equall to 75 m/m min. For thee
maccro-cutting caase, the mesh is made of arround 19,0000
threee- and four-nnode linear ellements.
The
T workpiecee material is th the titanium allloy Ti6Al4V V.
Its behaviour is described w with the TAN NH (TANgentt
Hyp perbolic) laww [11], a Johnsson-Cook law w modified too
takee the strain sooftening into account. Thee model doess
not take the grannular structuree of Ti6Al4V V into accountt
andd constitutes thereforee a simpliification off
miccro-cutting. The
T tool materrial is tungsten n carbide andd
its behaviour
b is described
d by a linear elastic law. Due too
the Lagrangian formulation,
f a chip separaation criterionn
based on an “erooding elementt” method is introduced
i inn

3. Numeriical Model
In order too study the innfluence of thhe depth of cuut on
chip formaation in ortthogonal cuttting, numerrical
simulations are carriedd out. A 2D 2 plane sttrain Fig.. 3 Boundaryy conditions an
nd initial geomeetry.
444 Chip Form
mation in Mic
cro-cutting

the model to make chhip formationn possible. This T

physical chhip separationn criterion is i based on the
temperature dependent tensile failuure of Ti6Al4V.
Coulomb’s friction
f is useed to model friction
fr at the tool
chip interfacce and all thee friction eneergy is conveerted
into heat, which is usually
u assuumed [12]. The
a tool initiaal temperaturee is set to 25 °C.
workpiece and
Only conducction is conssidered and all a the parts faces
f
are adiabaticc.

4. Model Validation
V in Macro-ccutting
Fig.. 4 Experimental setup.
The resuults are first validated in i macro-cuttting
conditions through
t the comparison of the modeelled (Taable 1) is of the same ordder of magn
nitude thoughh
saw-toothedd chip morphhology and cuttingc forcees to smaaller than thhe experimenntal one. Thiis differencee
experimentaal cutting reesults in thhe same cuttting cou
uld be explaiined by the pparameters choice
c of thee
conditions (ddepth of cut ofo 280 µm). TAN
NH law [11]..
Experimeents were performed
p o a lathe in
on Although
A the gap in magnnitude is largeer for the FF F
orthogonal cutting
c condittions. As shoown in Fig. 4, the (feeed force), sim
milar observattions are mad de and a linkk
workpiece specimen
s is a shaft compporting flangees in cann be drawn bettween its cycclic evolution and the teethh
the form off successive slices (diam meter 60 mm m) of formmation (morreover when CF is max ximal, FF iss
equal thicknness (2 mm). The tool and tool holder werew minnimal, and viice-versa). T The gap in th he feed forcee
custom madde by SECO in order to provide p the same
s mag gnitude mighht be caused byy the high inffluence of thee
characteristiics as the nummerical modeel. The tool width
w friction at the toool chip interrface, which is
i difficult too
(6 mm) is larger
l than thhat of the diisks. The cuttting meaasure and moodel.
processes arre performed in a plunge cutting modee on
5. Results
R in Micro-cutti
M ing
each slice inn dry cutting conditions.
c The tool was fixed
f
on the lathhe in a way to provide a high rigiddity, The
T model iss then used to highlight the specificc
particularly in the cuttinng direction. The chips werew feattures of micrro-cutting preesented in thee first part off
collected to be observedd latter on a microscopee. In the paper. A minnimum chip tthickness pred diction basedd
order to avoiid workpiece displacemennts and vibratiions, on the
t numericall results is finnally performmed.
a tailstock was
w used. For
F a determ mined materrial, the min nimum chipp
The morpphology of thhe modelled chip (Fig. 5aa) is thicckness depennds on the ddepth of cut (h) and thee
very close too the experimmental one (F Fig. 5b). For each
e cuttting edge radius of the toool (r). Eight h//r decreasingg
tooth, a slippping band iss formed in thhe primary shear ratios have beeen consideredd in order to t study thee
zone, as exxpected. It vanishes
v as the tool mooves inflluence of thee depth of cuut on the ch hip formationn
forward, inittiating the toooth formationn. process, the cuttiing edge radiuus remaining constant: h/rr
Fig. 6 shoows that the CF
C (cutting foorce) has a cyyclic = 14 4, 5, 2, 1, 0.55, 0.25, 0.125 and 0.05.
t formationn of a saw-toothed
evolution. This is due to the From
F these obbservations it can be conclu
uded that thiss
chip: a drop in the force happens duriing the formaation mod del is able to reproducee the chip formation
f off
of a tooth, as
a already shoown by Bäkeer et al. [13]. The Ti6Al4V in orthogonal macro-cuttin ng with a
force exhibits seven cyclles corresponding to the seeven satiisfactory fidelity. The inflluence of the depth of cutt
teeth formedd. The magnnitude of thee simulated force f on chip
c formatioon can therefoore be studied
d.
 Chip Form
mation in Mic
cro-cutting 4455

(a) (b)
Fig. 5 (a) Modelled
M and (b
b) experimentaal chip morphoologies.

rakee angle is neggative and onnly the value of

o the cuttingg
edg
ge radius of thhe tool influennces the chip formation.
The
T decrease of the deptth of cut leaads to a chipp
form
mation meechanism eevolving away a from
m
maccro-cutting: under
u h/r = 00.25 the material seems too
be less
l sheared than
t pushed aand deformed d by the tool..
Thiis was also nuumerically nooticed by Wo oon et al. [7]..
Acccording to theem such a chhip is formed by extrusionn
alon
ng the edge raadius of the ttool.
For
F h/r valuess under 0.25,, no chip is formed
f and a
Fig. 6 Macrro-cutting forcees. smaall amount off material acccumulates in
n front of thee
Table 1 Commparison betw ween modelled d and experimeental tooll. This smalll amount groows when thee tool movess
RMS (root mean
m square) foorces values. forw
ward until it reaches a thhickness greaater than thee
Experiments Modellingg Differencee
Value min
nimum chip thhickness and is then remo
oved from thee
(N·mm-1) (N·mm-1) (%)
Cutting force 344 218 37 worrkpiece. The fact that the chip is only formed
f whenn
Feed force 130 46 65 the accumulationn of cutting thhickness is hiigher than thee
min
nimum chip thickness haas also been observed byy
5.1 Chip Moorphology
Duccobu et al. in Ref. [4] and Vogler et al. in Ref. [14].
Figs. 7 annd 8 show chipc formatioon and machhined The
T fading off the primaryy shear zone occurs whenn
material defformation forr the eight h/rr ratios valuees. A h/r decreases andd it cannot bee distinguisheed any longerr
chip is undooubtedly form med when h/r = 5 to 0.5, which
w wheen h/r is smalller than 0.255.
is not true foor h/r = 0.25 to 0.05. The chip morphology Inn micro-cuttiing the cuttinng tool can no
n longer bee
goes from saaw-toothed too no chip inclluding segmented connsidered as sharp
s and thhe h/r value has a greatt
and nearly continuous chip. When h/r < 1, the effecctive inflluence on thee chip formattion. It can beb concludedd
446 Chip Form
mation in Mic
cro-cutting

Fig.. 9 Cutting foorces ratio.

thuss observed: an
a inversion between cuttting and feedd
forcces has occurrred. These oobservations are
a similar too
thosse made expeerimentally bby Jun et al. [15]. With a
Fig. 7 Chip morphologies for 14 to 1 h/rr values (Von Mises
M
3
stresses contoours (10 Pa)). feedd to cutting forces ratio criterion vallue of 2, thee
minnimum chip thhickness valuue would be between
b 5 µm
m
and
d 10 µm.

5.3 Specific Cuttting Energy

The
T evolutions of the rratio between the meann
specific cutting energy
e from ssimulations divided
d by thee
expperimental cuttting energy ffor the macro--cutting case,,
callled the mean normalized
n sppecific cutting
g energy, cann
be seen
s in Fig. 10 for the eighht simulated h/r
h ratios.
A nonlinear riser in the sspecific cuttinng energy iss
Fig. 8 Chip morphologiess for 0.5 to 0.005 h/r values (Von
(
notiiced when thee depth of cutt decreases, in n accordancee
Mises stressess contours (103 Pa)).
withh the previouusly presentedd size effect phenomenon.
p .
that changess in the chip formation meechanism happpen Its evolution beecomes exponential aroun nd 5-10 µm..
o the h/r ratiio situated beetween 0.125 and
for a value of Acccording to the t size efffect, the min nimum chipp
0.25. thicckness value isi thus of the order of 5-10 0 µm.

5.2 Cutting Forces

F 5.4 Workpiece Elastic
E Recoveery

As the h/rr ratio decreaases, the teethh are less marrked During
D the cuttting process,, an elastic recovery of thee
and tend to disappear reesulting in foorces becominng a worrkpiece is observed. Plotting it versus the depth off
constant vallue. The evollution of the ratio of the feed cut in Fig. 11 shoows a nonlineear increase of o it for smalll
force to the cutting
c force when the h/r ratio decreases is dep
pths of cut. Thhis rise (from
m about 0.45% % of h for 2800
shown in Figg. 9. µm to about 25% % of h for 1 µm) contributees to increasee
The more the h/r ratio decreases, thhe more the foorces the feed force, the slippingg forces and the specificc
ratio increasses. When thhe h/r ratio becomes
b smaaller cuttting energy. The
T elastic reecovery of th he workpiecee
than the uniit, the feed foorce becomess greater thann the couuld thereforre be an inddicator to evaluatee thee
cutting force. A change in the cuttinng mechanism m is minnimum chip thickness
t valuue. Accordin ng to Fig. 11,,
 Chip Form
mation in Mic
cro-cutting 4477

thicckness phenoomenon, whiich is review wed from ann

expperimental andd numerical ppoint of view in this paper..
Inn order to model the chip formaation, a 2D D
Laggrangian planne strain orthogonal cuttin ng model hass
beeen developed using the finnite element method. Thee
nummerical resultts highlight thhe phenomenna reported inn
literrature and brring a numerrical contribuution to theirr
commprehension.
An
A estimationn of the minim mum chip thicckness is alsoo
perfformed, leadding to a vaalue around 25% of thee
Fig. 10 Meaan specific cuttting energy. cuttting edge radiius of the toool.

Acknowledgm
ments
The
T authors gratefully accknowledge the FEDER R
finaancial supportt (SINUS Prooject) allowing to carry outt
the experimentall tests.

References
[1] J. Chae, S.S.
S Park, T. Freiheit, Inv vestigation off
micro-cuttingg operation, Inteernational Journnal of Machinee
Tools and Maanufacture 45 (22006) 313-332..
[2] H. Weule, V. Huntrup, H. Trrischler, Micro--cutting of steell
to meet new requirements
r inn miniaturization
n, Annals of thee
Fig. 11 Elasstic recovery off the workpiecce.
CIRP 50 (20001) 61-64.
[3] G. Bissacco, H.N. Hansen, L. De Chiffre,, Size effect onn
the minimum m chip thicknness value is less than 10 µm:
surface generration in microm milling of hardened tool steel,,
the elastic sppring back evvolution becoomes exponenntial Annals of thee CIRP 55 (2006) 593-596.
under this vaalue. [4] F. Ducobu, E. E Rivière-Lorrphèvre, E. Fillippi, Dynamicc
simulation ofo the microo-milling proccess includingg
5.5 Minimum
m Chip Thickkness Predictiion minimum chiip thickness andd size effect, Key
K Engineeringg
Materials 5044-506 (2012) 12269-1274.
Each metthod results in i a probablee minimum chip [5] D. Dornfeld,, S. Min, Y. T Takeuchi, Receent advances inn
thickness vaalue. It is obvvious that thhe minimum chip mechanical micromachining
m g, Annals of the CIRP 55 (20066)
thickness is less a precisee and single value
v than a raange 745-768.
[6] S. Filiz, C. Conley,
C M. Waasserman, O. Ozdoganlar,
O Ann
of values with
w unclear limits. The result of alll the
experimental investigation of micro-maachinability off
criteria prevviously presennted is that thhe minimum chip copper 101 using tungstten carbide micro-endmills,
m ,
thickness vaalue is of the order
o of 5 µmm (25% of r) with
w International Journal of Maachine Tools an nd Manufacturee
a lower limit around 2.5 µm (12.5% of o r) and an uppper 47 (2007) 1088-1100.
[7] K. Woon, M. M Rahman, F F. Fang, K. Neo, K. Liu,,
limit inferioor to 10 µm m (50% of r). This ordeer of
Investigations of tool edge radiu us effect inn
magnitude iss confirmed in i Refs. [6, 144]. micromachinning: A FEM sim mulation appro oach, Journal off
Materials Proocessing Technoology 167 (200 07) 316-337.
6. Conclussions [8] K. Liu, S.N N. Melkote, Finnite element analysis
a of thee
influence of tool
t edge radiuus on size effecct in orthogonall
The dowwn-sizing of the milling process induuces
micro-cuttingg process, Internnational Journall of Mechanicall
some changees in the cuttiing phenomennon. One of them
t Sciences 49 (2007)
( 650-6600.
is the chip formation, involving
i thee minimum chip [9] K. Liu, S.N. Melkote, Materrial strengthenin ng mechanismss
448 Chip Formation in Micro-cutting
and their contribution to size effect in micro-cutting,
Transactions of the ASME 128 (2006) 730-738.
[10] A. Simoneau, E. Ng, M.A. Elbestawi, The effect of
microstructure on chip formation and surface defects in
micro-scale, mesoscale, and macroscale cutting of steel,
Annals of the CIRP 55 (2006) 97-102.
[11] M. Calamaz, D. Coupard, F. Girot, A new material model
for 2D numerical simulation of serrated chip formation
when machining titanium alloy Ti6Al4V, International
Journal of Machine Tools and Manufacture 48 (2008)
275-288.
[12] M. Nasr, E.-G. Ng, M. Elbestawi, Effects of workpiece
thermal properties on machining-induced residual
stresses-thermal softening and conductivity, Proceedings
of the Institution of Mechanical Engineers, Part B: Journal
of Engineering Manufacture 221 (2007) 1387-1400.
[13] M. Bäker, J. Rösler, C. Siemers, A finite element model
for high speed metal cutting with adiabatic shearing,
Computers and Structures 80 (2002) 495-513.
[14] M.P. Vogler, R.E. DeVor, S.G. Kapoor, On the modeling
and analysis of machining performance in
micro-endmilling, Part II: Cutting force prediction,
Journal of Manufacturing Science and Engineering 126
(2004) 695-705.
[15] M.B.G. Jun, R.E. DeVor, S.G. Kapoor, Investigation of
the dynamics of micro-endmilling, Part II: Model
validation and interpretation, Journal of Manufacturing
Science and Engineering 128 (2006) 901-912.