International Journal of Machine Tools & Manufacture 51 (2011) 512–519

Contents lists available at ScienceDirect

International Journal of Machine Tools & Manufacture

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

Dynamic modelling of shear band formation and tool-tip vibration in

ultra-precision diamond turning
H. Wang, S. To n, C.Y. Chan, C.F. Cheung, W.B. Lee
State Key Laboratory in Ultra-precision Machining Technology, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong, PR China

a r t i c l e i n f o a b s t r a c t

Article history: This paper proposes a dynamic model to correlate the two basic physical phenomena in ultra-precision
Received 17 October 2010 diamond turning, i.e. the formation of adiabatic shear band (ASB) and high frequency tool-tip vibration
Received in revised form (HFTTV). The conventional approach explains the former using a static model without consideration of
23 February 2011
the latter. In this paper, a dynamic model is developed to reflect how the ASB and the HFTTV
Accepted 28 February 2011
Available online 8 March 2011
interactively affect each other. To illustrate the validity of this model, a novel experimental method is
proposed and the effect of HFTTV on cutting force, surface roughness and chip morphology of ASBs is
Keywords: discussed in terms of the variation of strain rates.
Adiabatic shear band & 2011 Elsevier Ltd. All rights reserved.
Tool-tip vibration
Dynamic model
Ultra-precision diamond turning

1. Introduction which is independent of the cutting speed in face turning, where

the cutting speed is proportional to the radial distance from
Elastic strain induced shear band and high frequency tool-tip the centre of the workpiece at a constant spindle speed. Since the
vibration are two important physical phenomena in the micro- HFTTV is directly related to the load drop induced by the
cutting process related to chip morphology and machining formation of adiabatic shear band, a definite correlation between
dynamics. These phenomena have always been studied separately the cutting speed and adiabatic shear bands does not exist in
since there has never been a model before to successfully link terms of a dynamic system of the micro-cutting process. There-
together these phenomena in the research fields of ultra-precision fore, the cutting speed was introduced by most of the quasi-static
machining. models as an explanation for the formation of adiabatic shear
Adiabatic shear bands (ASBs) result from the competition bands. However, the effect of cutting speed is almost always
between thermal softening due to the plastic heating effect and overemphasised because the quasi-static model is not realistic.
strain hardening in deformed materials. In machining, the thick- The effect of strain rate on the micro-cutting process is one of
ness and spacing of adiabatic shear bands are usually analysed in the critical issues that have not yet been fully understood. Different
terms of the cutting velocity or cutting speed [1–3]. The common models have been proposed to calculate the mean or average shear
assumptions were a constant cutting speed [3] under quasi-static strain rate in the primary deformation zone [8–11]. Astakhov [12]
conditions [4] and/or with a perfectly plastic material [5]. In the reviewed the prior research on strain rate and concluded that
diversity of previously proposed models, the thickness and the the average shear strain rate in the shear zone lies in the range of
spacing of shear bands are approximately inversely proportional 103–105 s  1 or even higher. According to the regimes of mechan-
to the imposed strain rate or cutting speed. Molinari et al. [3] ical loading response presented by Wasley [13] a dynamic model
conducted a series of cutting experiments to identify the relation- should be established instead of a quasi-static one, where the
ship between the occurrence of adiabatic shear bands and cutting elastic and plastic stress waves should be considered. In a recent
speed in orthogonal cutting of Ti–6Al–4V. In the speed range from study, the effect of strain rate in metal cutting has been studied by
0.01 to 73 m/s, the fully formed adiabatic shear bands were FEM analysis [14], where the variation of strain rate was imple-
observed only at velocities of 12–36 m/s. However, in our pre- mented by changing the cutting speed of the rigid body movement
vious studies [6,7], the high frequency tool-tip vibration (HFTTV) of the cutting tool. The strain rate effect also received enough
was identified as an intrinsic property of the dynamic process, attention in the development of constitutive models for work
materials, such as the Johnson–Cook model and the models
proposed by Dudzinski and Molinari [15] and Shirakashi
n
Corresponding author. Tel.: þ852 27666587; fax: þ852 27647657. et al. [16]. With such constitutive models, the strain rate effect
E-mail address: [email protected] (S. To). can be studied even with static models, such as those developed by

0890-6955/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmachtools.2011.02.010
 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519 513

Budak and Ozlu [17] and Dirikolu et al. [18]. However, a static with an integral scaling factor (k).
model overlooks the physical phenomenon of the micro-cutting pffiffiffiffiffiffiffiffi k0 y0
process in ultra-precision diamond turning. d ¼ 6 2m ð1Þ
Vat0
Based on our prior research findings, high frequency tool-tip
vibration (HFTTV) is a dominant factor in the formation of shear Vs0 Vs1 V cos a
Vc ¼ V sin j, V¼ ¼ ð2Þ
bands and contributes to the fluctuation of cutting force, heat 2 2 cosðjaÞ
generation and other related problems. In recent research, tool- pffiffiffiffiffiffiffiffi
tip vibration in this high frequency range was successfully applied pffiffiffiffiffiffiffiffi k0 y0 sin j 12 2mk0 y0 cosðjaÞ
tB ¼ t2 t1 ¼ d sin j=Vc ¼ 6 2m ¼
by mode control techniques to improve the surface roughness [19] VVc at0 at0 cos a V 2
in their experiments; however, the mechanism has not been fully ð3Þ
studied. In this paper, the relationship between HFTTV and elastic
strain induced shear bands is first established. Without the 2p 2 2 3
tA ¼ ½m2 kcð1aÞ2 y0 =ð1þ 1=mÞb g_ t^ 0 a2 1=4 ð4Þ
popular assumptions on quasi-static process and perfectly plastic V sin j
deformation, a dynamic model is proposed to further understand
tA lA
different aspects. Zt ¼  pV ð5Þ
tB lB

1
fASB ¼ ¼ k fttv ðkZ 1Þ ð6Þ
tA þ tB

2. Transient analysis for shear band formation

3. Dynamic model for shear band formation

Following the assumptions adopted by Dinzart and Molinari [4],
a quasi-static model during the final stage is employed to study the
3.1. Dynamic model
relationship between cutting velocity and thickness and spacing of
shear bands first. Different from Molinari et al. [3], the effect of
HFTTV and elastic strain is further studied using transient analysis In this section a dynamic model is developed based on our
based on the steady-state model (Fig. 1). previously proposed pendulum model [7], to reveal the intrinsic
First, the thickness of shear band (d) is determined by Eq. (1) connection between the theory and model of HFTTV and ASBs.
with the applied velocity (V*) as defined by Dinzart and Molinari [4], Several assumptions are taken in this paper. The dynamic model
where m is the strain rate sensitivity factor, k0 is the heat is applied after the full formation of the first ASB (L1, Fig. 2(1)).
conductivity, a is the thermal softening parameter, t0 is the shear
flow resistance and y0 denotes the initial temperature. The relation-
ship between the velocity parameters is expressed by Eq. (2) as
derived by Molinari et al. [3]. Therefore, by combining Eqs. (1) and
(2), the time for the formation of shear band (tB) can be obtained by
Eq. (3). On the other hand, the time for elastic deformation to
achieve a large volume of deformation is determined by Eq. (4),
which is directly derived from the equation for the shear band
spacing [1,2]. Based on Eqs. (3) and (4), the employed parameter Zt
in our prior work [6] can be further calculated as in Eq. (5), which
links the time ratio with the cutting speed following a proportional
relationship. Thus, at a higher cutting speed, the thickness of shear
bands will decrease and the spacing between shear bands increases
relatively. Moreover, the relationship between the frequencies of
HFTTV (fttv) and the formation of ASBs (fASB) can be found in Eq. (6)

Fig. 1. Schematic of the transient model for orthogonal cutting. Fig. 2. Dynamic model of HFTTV and ASBs.
514 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519

The frequency of ASBs is assumed to be the same as that of 3 in Fig. 2, provided that the vibrations of HFTT and the formation
HFTTV, i.e. k¼1 in Eq. (6). Solely the situation in which HFTTV is of ASBs are in phase.
in phase with the cyclic formation of ASBs is illustrated here. The
effect of phase shift will be further explained in the following 3.2. Effect of equivalent cutting velocity
section.
In Fig. 2, the tool tip of the diamond cutter is shown as a
The significance of HFTTV on the equivalent cutting velocity is
pendulum with a constant cutting speed (V). The tool tip is modelled
that it can induce a sharp increase in the cutting velocity in an
as a rigid bob with mass (M). The extra velocity (DV7) due to the
extremely short interval of time, which does not affect a large
HFTTV is super-posed on the cutting speed (V), and the sign (7) of
volume of work material but does effectively modify the stress
the super-posed velocity is determined by its phase. In stage
distribution in the primary deformation zone, in favour of the
1 of Fig. 2, the rigid bob accelerates in the same direction as that
formation of ASBs and a reduced cutting force. An explanation for
of the cutting speed. As a result, the actual strain rate imposed on the
this is based on the subsequent analysis of stress wave propaga-
work material exceeds that provided solely by the cutting speed. In
tion in this section.
stage 2, a thin layer of work material directly in contact with the rake
The equivalent cutting velocity is confined within its max-
face of the tool tip experiences the stage of pre-compression. In this
imum (vmax) and minimum (vmin) as denoted in Fig. 3. In the
stage, the elastic energy on the tool tip is transferred to the chip root
dynamic region of Wasley’s regimes of mechanical loading
by the pre-compressed layer as shown in the shaded rectangular
response [13], wave propagation should be considered. The
area. The effects of pre-compression are (i) displacement and
speeds of elastic wave (ce) and plastic wave (cp) propagation can
bending moment of the chip above the depth of cut (ap), due to
be calculated by Eq. (7), where r is the density of work material,
the effect of free edge and (ii) elastic strain concentration at both the
E is Young’s modulus and E0 is defined in the plastic region. Since
free edge (point A) and chip root (point B), which will further lead to
E0 5E, cp 5ce.
the initiation and propagation of ASBs, as explained in our prior sffiffiffiffiffiffiffiffiffiffiffiffi
work [10]. At the moment immediately before the initiation of ASBs, pffiffiffiffiffiffiffiffiffi 1 @sP pffiffiffiffiffiffiffiffiffi ffi
the maximum cutting force on the tool tip is achieved and the ce ¼ E=r, cp ¼  E0 =r ð7Þ
r @eP
superposed velocity vanishes. In stage 3 the formation of ASB
abruptly reduces the stress capacity of the elastically strained The relationship between the equivalent cutting velocity and
material due to the effect of work softening, which results in a load wave propagation speed is shown in Fig. 3. The HFTTV induced
drop on the cutting force. At this stage, an opposite speed is increase in cutting speed makes the equivalent cutting velocity
superposed in the negative direction to the cutting speed. exceed the speed of plastic wave propagation (ce) in stage 1 of Fig. 2,
The three stages of a complete cycle of HFTTV with regard to i.e. vmax 4cp, which reduces the plastic strain to a much smaller
the formation of ASBs lead to a cyclic behaviour on the tool-tip volume in primary deformation zone. This is also evidenced in the
displacement as well as the cutting speed, as shown in Fig. 3. The review of the principle of high speed machining in the work of Zhou
broken line illustrates the ideal tool-tip displacement and the et al. [20]. On the other hand, the elastic wave propagates at a much
solid line shows the actual displacement. Regions A and B faster rate, i.e. ce bvmax 4cp, so the elastic strain concentration at
correspond to the time notations tA and tB in Fig. 1. Considering points A and B in Fig. 3 can be accomplished at an infinite rate as
the cutting speed and the imposed velocity by HFTTV, the assumed. The equivalent cutting velocity decreases to its minimum
equivalent cutting velocity is qualitatively plotted in the sub- (vmin) during the pre-compression period (stage 2 in Fig. 2), which is
coordinate system in Fig. 3. In region A the cyclic fluctuation on possibly lower than cp. However, ASB occurs at this moment and the
the equivalent cutting velocity experiences stages 1 and 2, as load-drop takes place on the cutting force in stage 3. The fast
plotted in Fig. 2. The velocity in region B corresponds to stage equivalent cutting velocity or higher strain rate ensures a thinner
pre-compression layer, i.e. the bulk material experiences elastic
deformation only, which in turn reduces the energy loss on the
pendulum bob or the tool tip in reality.

4. Experimental verification

4.1. Experiment setup

(1) Experiment A: verification of HFTTV and its effect on dynamic

cutting force and surface roughness.
As one of the most important commercialised magnesium
alloys, Mg–9Al–1Zn alloy (AZ91) in the shape of a rectangular
plate is employed as the workpiece material in this verifica-
tion experiment (Fig. 4). The AZ91 plate sample is concen-
trically mounted on a cylinder base that is fixed on the
spindle of ultra-precision diamond turning machine (Opto-
form 30 from Rank Pneumo, USA). The face turning is
performed at a depth of cut (ap) of 20 mm, a feed rate of
20 mm min  1 and a spindle speed of 8000 rpm. The tool nose
radius is 2.503 mm and the rake angle is 151. A force
transducer Kistler 9252A is mounted with a pre-loading force
under the tool shank. A 14-bit multifunction Data Acquisition
(DAQ) card (PCI-6132, National Instrument) is installed on a
PC workstation to convert the analogue signals of the cutting
Fig. 3. Cyclic displacement of tool tip and the transient cutting speed. forces. With a Kistler 5011B charge amplifier, the converted
 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519 515

Fig. 4. Plate workpiece of Mg alloy—AZ91 in Experiments A and B1.

Fig. 5. Optical images of the machined surface at different sample locations by Olympus BX60 (a) and (b), and Alicona IFM G4 (c) in s1 and (d) in s2 (Experiment A).

signals are recorded at the sampling rate of 1 MHz. The In Experiment B2, to further demonstrate the effect of material
signals of the cutting force are further processed using fast factor on HFTTV and effect of HFTTV on formation of ASBs,
Fourier transform (FFT) and studied with power spectrum aluminium single crystal samples are employed in the micro-
density (PSD) analysis. cutting process. Rod samples of aluminium single crystal on
Based on the geometry of the workpiece surface, continuous (0 1 1), (0 0 1) and (1 1 1) planes are prepared with a diameter
face turning can be achieved only approximately at a radius of of 12.7 mm. With a round machining surface instead of a
half the width of the workpiece surface from its centre. rectangular surface as in Experiment A, continuous face turning
Intermittent cutting is performed at a distance from the centre is ensured. Without introducing further excitation into the micro-
larger than this radius, shown as zout in Fig. 4. In the experi- cutting process, the effect of material properties on HFTTV and
ment, three sample locations s1, s2a and s2b are selected in the chip formation is studied. The face turning is performed at a
inner zone (zin) and the outer zone close to and far from the depth of cut of 10 mm and a feed rate of 20 mm min  1 with a
transition zone (ztrans), respectively. At each sample location, spindle speed of 8000 rpm. The tool nose radius is 2.503 mm and
signals of cutting force are acquired and profiles of the the rake angle is 151. The measurement methods of cutting force
machined surface are examined. The surface roughness is and PSD analysis are the same as those in Experiment A. Mean-
evaluated by a non-contact 3D optical measuring system while, the chips generated in face turning of samples on different
Alicona IFM G4. crystallographic planes are collected to examine under SEM. The
(2) Experiment B: effect of HFTTV on chip formation with ASBs. micrographs are taken on the free surface of chips with SEM.
In Experiment B1, the same setup and work material AZ91 alloy
are employed as in Experiment A. At locations s1, s2a and s2b as 4.2. Results and discussion
denoted in Fig. 4, chips generated in the micro-cutting process
are collected to study the effect of HFTTV on formation of ASBs The images of the machined surface are taken by an optical
using a scanning electron microscope (SEM, JEOL 6490). The microscope Olympus BX60 as shown in Fig. 5(a) and (b). The
micrographs are taken on the free surface of chips with SEM. surface roughness in sample locations s1 and s2 are measured by
516 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519

Alicona IFM G4 with a profile length of 140 mm, in accordance other researchers. Therefore, the cutting speed cannot fully
with ISO4288:1996. The average roughness (Ra) is 12 nm in s1, explain the formation of ASBs in this regime and the boost
where the turning marks can be observed clearly, and 6.7 nm of strain rate induced by HFTTV provides a better alternative
in s2. explanation.
The force signals in the cutting direction in the three sample The effect of HFTTV on chip formation with ASBs is revealed in
locations s1, s2a and s2b are trimmed and plotted sequentially in the micrographs of free surface of chips obtained in Experiment B1,
Fig. 6. In the sample location s1, continuous cutting is performed. where a lamellar feature (as defined in the models of Black [21] and
High excitation induced vibration is observed in the force signal
at location s2b. The transition zone is found close to the boundary
between s1 and s2a, where the periodic excitation from the impact
between the tool tip and the edge of the plate workpiece starts to
take effect. In Fig. 7, the normalised mean values of the measured
forces at different sample locations are shown with their standard
deviation. Due to the effect of excitation, the standard deviation is
the highest at s2b, which also leads to an increase in the mean
force. The total force or the nominal cutting force directly
acquired by the force sensor is actually the vibrational force, a
major portion of which is irrelevant to the material removal
process, i.e. irrelevant to the relative displacement in the tool–
chip interface. Therefore, the true cutting force is calculated based
on the force resulting from the cyclic interaction between the
work material and the cutting tool. As shown in Fig. 7, the trend of
the true cutting force is significantly different from that of the
nominal cutting force in that (i) the standard deviation remains at
the same level in different sample locations and (ii) the effect of
strain rate on the reduction of cutting force is revealed. Thus, it is
of paramount importance, in the study of force signals, to clarify
the concepts of vibrational force and cutting force. With the data
of the true cutting force, the effect of HFTTV on the reduction of
the mean cutting force is clearly observed in Fig. 7. Fig. 7. Normalised mean vibrational force and cutting force at different sample
Power spectrum density (PSD) analysis is performed on the locations.
signals of cutting force. The data are shown in the overlaid curves
(Fig. 8) to reveal the frequency shift in the range of 12–14 kHz. Due
to the high excitation and the lesser effect of process damping under
a much higher strain rate (the curve of s2b), the four primary peaks
can be regarded as natural frequencies of the tool tip, which
also sets up a datum line for further study. Corresponding to the
identified characteristic twin peaks in the prior work [7], the peaks
P2s2b and P3s2b are traced for analysis in this paper. According to the
analysis in Section 3.2, in the case of mid-excitation, a reduction of
cutting speed results in a lower strain rate and a more significant
effect of process damping takes place, which leads to a left shift of
the peaks P2s2b and P3s2b to the lower frequencies, shown as peaks P2s2a
and P3s2a in the curve s2a of Fig. 8. In the continuous cutting (curve
s1), no excitation is present since there is no impact between tool tip
and the edge of the rectangular plate workpiece in each rotational
cycle. Thus, the observable peaks are the damped ones that are
directly related to the material removal process, i.e. the character-
istic twin peaks (CTPs).
Based on the spindle speed and the distance from the work-
piece centre to the sample locations in our turning experiment,
the maximum cutting speed is calculated as 0.67 m/s at the
sample location s2b, which is far slower than the velocity range Fig. 8. Power spectrum density (PSD) analysis of cutting force signals in the three
12–36 m/s employed in the cutting trials by Molinari et al. [3] and sample locations s1, s2a and s2b (Experiment A).

Fig. 6. Measured cutting forces in sample locations of s1, s2a and s2b (Experiment A).
 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519 517

Von Turkovich [22]) is observed with shear band spacing (Ls) as adopted by many other researchers [23–25]. Comparing the average
denoted in Fig. 9(c). On the free surface of chips, these lamella values of lamella spacing in different sample locations, the external
features are in relation to the features of adiabatic shear bands that excitation at location s2a (Fig. 9(b)) and s2b (Fig. 9(c)) plays an
can be observed on the cross-section of chips (Fig. 1 of, Ref. [6]). important role to increase the thickness of the lamella structure, i.e.
Such measurement method of shear band spacing has also been a larger value of shear band spacing Ls is obtained while a smaller

Fig. 9. SEM images of free surface of chips generated at different locations in face Fig. 10. SEM images of the free surface of chips from an aluminium single crystal
turning; (a) s1, (b) s2a and (c) s2b (Experiment B1). on various planes; (a) (0 1 1), (b) (0 0 1) and (c) (1 1 1) (Experiment B2).
518 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519

In this paper, the dynamic model is established and a new

experimental technique is proposed to verify the theoretical
model and analysis. The major findings are summarised as below:

(i) The cutting speed is not the real reason for the formation of
ASBs in serrated chips. The HFTTV is identified in the
experiments to trigger the initiation of ASBs in terms of the
boost on the strain rate in the cyclic relative tool–chip
displacement during the micro-cutting process. A dynamic
model is proposed to study the cyclic variation of the cutting
speed and strain rate.
(ii) This paper proposes a novel experimental method to exam-
ine different levels of applied excitation on a single work-
piece in ultra-precision diamond turning. Based on an
analysis of natural frequencies of the structure of a cutting
tool, damped frequencies and characteristic twin peaks
(CTPs), the evolution of the frequencies reveals both the
structural properties of cutting tools and the factors of the
material removal process.
(iii) The method for the unbiased evaluation of cutting force is
Fig. 11. Power spectrum density (PSD) analysis of cutting force in face turning of studied, in which the true cutting force is extracted from the
aluminium single crystals on planes (0 1 1), (0 0 1) and (1 1 1) (Experiment B2).
signal of vibrational force to exclude the effect of the
excessive applied excitations. The experimental results prove
value of Ls is found in the continuous cutting of location s1 (Fig. 9(a)). that HFTTV reduces the mean cutting force.
Furthermore, it is observed that the lamella structures are more
(iv) In the micro-cutting process, the physical phenomena such as
regular in locations with external excitation (s2a and s2b), where the the ASB–HFTTV system can be studied only with a dynamic
peaks at natural frequencies are pronouncedly excited as shown
model considering different factors such as material, machine-
in Fig. 8. tool dynamics and process factors. The factor of elasticity is
The mechanical properties of the work material are another
indispensable and plays as important a role as the factor of
key factor in surface generation and chip formation. Material plasticity in the material removal process. Previous studies
factors become even more significant in the micro-cutting process
using static models oversimplified the analysis.
because the effect of crystallographic orientation needs to be
taken into consideration. In Experiment B2, the effect of orienta-
tion of aluminium single crystals on both HFTTV and ASBs is Acknowledgements
verified in Fig. 10. For the planes of (0 1 1), (0 0 1) and (1 1 1), the
values of mean shear band spacing (Ls) are 1.5, 0.8 and 0.5 mm, The authors would like to express their sincere thanks to the
respectively. Greater shear band spacing corresponds to a higher research committee of The Hong Kong Polytechnic University and
process damping effect induced on the tool–chip and tool–work the Research Grants Council of the Hong Kong Special Adminis-
interfaces, where the former is shown in Fig. 2 and the latter has trative Region of the People’s Republic of China for their financial
been discussed in detail in our previous work [7]. As a result, the support of the research work (Project code: PolyU5391/08E).
greater damping effect leads to a more pronounced decrease in
transient cutting speed as plotted in the v–t coordinate system References
of Fig. 3; therefore, a larger elastically deformed volume is formed
in chip flow, which in turn increases the shear band spacing. The [1] T.W. Wright, H. Ockendon, A scaling law for the effect of inertia on the
consistent results on PSD analysis of cutting force is shown in formation of adiabatic shear bands, International Journal of Plasticity 12
Fig. 11, where the characteristic twin peaks (CTPs) are denoted (1996) 927–934.
[2] A. Molinari, Collective behavior and spacing of adiabatic shear bands, Journal
with arrows and an illustrative presentation of density distribu- of the Mechanics and Physics of Solids 45 (1997) 1551–1575.
tion of the peaks in the frequency range of interest is given to the [3] A. Molinari, C. Musquar, G. Sutter, Adiabatic shear banding in high speed
left of CTPs. It is also in line with the experimental results of our machining of Ti–6Al –4V: experiments and modelling, International Journal
of Plasticity 18 (2002) 443–459.
previous work [7], where the ratio of CTPs can be correlated with
[4] F. Dinzart, A. Molinari, Structure of adiabatic shear bands in thermo-
the effect of process damping and a significant damping effect viscoplastic materials, European Journal of Mechanics — A/Solids 17 (1998)
shifts the density distribution to the left, i.e. increases the ratio of 923–938.
[5] T.W. Wright, H. Ockendon, A model for fully formed shear bands, Journal of
P1/P2 as shown in Fig. 11. Therefore the experimental data
the Mechanics and Physics of Solids 40 (1992) 1217–1226.
quantitatively verify that the formation of ASBs is affected by [6] H. Wang, S. To, C.Y. Chan, C.F. Cheung, W.B. Lee, Elastic strain induced shear
HFTTV, which is also sensitive to the mechanical properties of bands in the microcutting process, International Journal of Machine Tools and
work materials. Manufacture 50 (2010) 9–18.
[7] H. Wang, S. To, C.Y. Chan, C.F. Cheung, W.B. Lee, A theoretical and experi-
mental investigation of the tool-tip vibration and its influence upon surface
generation in single-point diamond turning, International Journal of Machine
Tools and Manufacture 50 (2010) 241–252.
5. Conclusions [8] B.T. Chao, G.H. Bisacre, The effect of speed and feed on the mechanics of
metal cutting, Proceedings of the Institution of Mechanical Engineers 165
(1951) 1.
In conclusion, high frequency tool-tip vibration (HFTTV) and
[9] D. Drucker, An analysis of the mechanics of metal cutting, Journal of Applied
the formation of adiabatic shear bands (ASBs) are intrinsically Physics 20 (1949) 1013.
well-related physical phenomena, which can be regarded as a [10] D. Kececioglu, Shear strain rate in metal cutting and its effect on shear flow
system (ASB–HFTTV) in the micro-cutting process of ultra-preci- stress, ASME Journal of Engineering for Industry 80 (1958) 158.
[11] P.L.B. Oxley, Rate of strain effect in metal cutting, ASME Journal of Engineer-
sion diamond turning. The study of their relationship proposes ing for Industry 84 (1963) 335.
new requirements on the modelling and experimental methods. [12] V.P. Astakhov, Metal Cutting Mechanics, CRC Press LLC, 1998 pp.95–123.
 H. Wang et al. / International Journal of Machine Tools & Manufacture 51 (2011) 512–519 519

[13] R.J. Wasley, Stress Wave Propagation in Solids: An Introduction, Marcel [19] V. Ostasevicius, R. Gaidys, J. Rimkeviciene, R. Dauksevicius, An approach based
Dekker, Inc., New York, 1973 pp.19–24. on tool mode control for surface roughness reduction in high-frequency
[14] J.P. Davim, C. Maranha~ o, A study of plastic strain and plastic strain rate in vibration cutting, Journal of Sound and Vibration 329 (2010) 4866–4879.
machining of steel AISI 1045 using FEM analysis, Materials and Design 30 [20] L. Zhou, J. Shimizu, A. Muroya, H. Eda, Material removal mechanism beyond
(2009) 160–165. plastic wave propagation rate, Precision Engineering 27 (2003) 109–116.
[15] D. Dudzinski, A. Molinari, A modeling of cutting for viscoplastic materials, [21] J.T. Black, Flow stress in metal cutting, Transactions of the ASME, Journal of
International Journal of Mechanical Sciences 39 (1997) 369–389. Engineering for Industry 101 (1979) 403–415.
[16] T. Shirakashi, K. Maekawa, E. Usui, Flow stress of low carbon steel at high [22] B.F. Von Turkovich, Shear stress in metal cutting, Transactions of the ASME,
temperature and strain rate (parts I–II), Bulletin of the Japan Society of Journal of Engineering for Industry 92 (1970) 151–157.
Precision Engineering 17 (1983) 161–172. [23] J. Barry, G. Byrne, D. Lennon, Observations on chip formation and acoustic
[17] E. Budak, E. Ozlu, Development of a thermomechanical cutting process model emission in machining Ti–6Al–4V alloy, International Journal of Machine
for machining process simulations, CIRP Annals—Manufacturing Technology Tools and Manufacture 41 (2001) 1055–1070.
57 (2008) 97–100. [24] J. Barry, G. Byrne, Chip formation, acoustic emission and surface white layers in
[18] M.H. Dirikolu, T.H.C. Childs, K. Maekawa, Finite element simulation of chip hard machining, CIRP Annals—Manufacturing Technology 51 (2002) 65–70.
flow in metal machining, International Journal of Mechanical Sciences 43 [25] M.A. Davies, T.J. Burns, C.J. Evans, On the dynamics of chip formation in machining
(2001) 2699–2713. hard metals, CIRP Annals—Manufacturing Technology 46 (1997) 25–30.