- Email: [email protected]
Modelling and simulation of effect of ultrasonic vibrations on machining of Ti6Al4V
Accepted Manuscript Modeling and Simulation of Effect of Ultrasonic Vibrations on Machining of Ti6Al4V Sandip Patil, Shashikant Joshi, Asim Tewari, Suhas S. Joshi PII: DOI: Reference:
S0041-624X(13)00270-9 http://dx.doi.org/10.1016/j.ultras.2013.09.010 ULTRAS 4676
To appear in:
Ultrasonics
Please cite this article as: S. Patil, S. Joshi, A. Tewari, S.S. Joshi, Modeling and Simulation of Effect of Ultrasonic Vibrations on Machining of Ti6Al4V, Ultrasonics (2013), doi: http://dx.doi.org/10.1016/j.ultras.2013.09.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Modeling and Simulation of Effect of Ultrasonic Vibrations on Machining of Ti6Al4V Sandip Patil, Shashikant Joshi, Asim Tewari, Suhas S. Joshi1 Department of Mechanical Engineering Indian Institute of Technology Bombay, Powai, Mumbai – 400 076 India
Abstract: The titanium alloys cause high machining heat generation and consequent rapid wear of cutting tool edges during machining. The ultrasonic assisted turning (UAT) has been found to be very effective in machining of various materials; especially in the machining of “difficult-to-cut” material like Ti6Al4V. The present work is a comprehensive study involving 2D FE transient simulation of UAT in DEFORM framework and their experimental characterization. The simulation shows that UAT reduces the stress level on cutting tool during machining as compared to that of in continuous turning (CT) barring the penetration stage, wherein both tools are subjected to identical stress levels. There is a 40-45 % reduction in cutting forces and about 48% reduction in cutting temperature in UAT over that of in CT. However, the reduction magnitude reduces with an increase in the cutting speed. The experimental analysis of UAT process shows that the surface roughness in UAT is lower than in CT, and the UATed surfaces have matte finish as against the glossy finish on the CTed surfaces. Microstructural observations of the chips and machined surfaces in both processes reveal that the intensity of thermal softening and shear band formation is reduced in UAT over that of in CT. Keywords: Machining, titanium, ultrasonic vibrations, finite element simulation, chip microstructure, cutting forces
1.
Introduction
Titanium is recognised for its strategic importance as a unique lightweight, high strength, structurally efficient metal for critical and high performance aircraft such as jet engine and airframe components. These alloys are termed as ‘difficult-to-machine’ materials but have high utility in aerospace, medical and other manufacturing sectors. Ultrasonically assisted turning (UAT) is an advanced machining technique, in which up to 30000 low energy vibro-impacts are superimposed on the cutting tool. This technique of UAT was developed in 1960s in which high frequency ultrasonic vibrations are superimposed 1
Corresponding Author: Ph: +91 22 2576 7527; Email: [email protected]
1
on the conventional movement of cutting tool, providing a range of benefits in machining hard metal alloys [1]. Various parameters of UAT such as vibration frequency, amplitude of vibrations, direction of ultrasonic vibrations, friction at the tool tip-workpiece interface and cutting speed, affect the level of cutting forces generated during the process. The projected benefits of UAT are summarized as below: •
An improvement in the dynamic cutting stability [2].
•
A reduction in the cutting forces and improvement in tool life [3] [2] [4].
•
A reduction in surface roughness of machined surface [2] [5].
•
A reduction in residual stresses in the machined workpiece [6].
•
A reduction in cutting temperature [6].
Use of ultrasonic vibrations in machining was introduced about 24 years ago [7]. Ultrasonic vibration on the cutting tool during machining of ceramic reduced cutting tool and work temperature. Tensile loading of ceramic due to vibration was considered as a reason for smooth and efficient cutting [8]. A significant reduction in chip thickness and cutting forces was observed by imparting a synchronised vibration to the cutting tool in cutting and chip flow direction [7]. With UAT, a large increase in MRR of the order of 2-3 times was observed in machining Ti15-3-3-3 and Ni-625 [9] [10]. Further, an increase in MRR and surface finish in machining
titanium alloy was observed by combining UAT with external
heating of the workpiece [11]. Several researchers have studied the effect of such vibrations assisted techniques to machine ‘difficult-to-cut’ materials such as Inconel 718. However, research has been scarce on its application and effective machining of Ti6Al4V. Babitsky et al. [4] [12] found that the UAT method performs better at lower cutting speed using high frequency and high amplitude of vibration. Koshimizu [5] observed a reduction in cutting forces under UAT of Ti6Al4V. The author also found that the vibration assisted technique holds good for Ti6Al4V, only when the cutting speed was 30% below the maximum vibrating velocity of the cutting tip. Nath and Rahman [2] studied the effect of machining parameters during UAT of Inconel 718. Muhammad et al. [1] developed a 3D FE (Finite Element) model to compare the cutting forces during UAT and CT for Ti-15. An increase in the amplitude and frequency of vibration has reduced cutting forces to 25% and 36% respectively. An increase in cutting tool velocity and a reduction in tool-work contact was attributed as a reason for such a large reduction in the cutting forces. Tang et al. [13] developed a 2D FE model for Ti6Al4V and observed a 2
higher shear angle beside a reduction in the cutting temperature and friction coefficient, as compared to CT. Ahmed et al. [4] developed a 2D FE model to study the stresses under UAT of Inconel 718. Mitrofanov et al. [6] developed a 2D FE model to study the plastic and residual strains in the machined layer and compared the results with nano-indentation tests on Inconel 718. Amini and Nategh [14] studied the effect of parameters viz. vibration amplitude, depth of cut, feed rate and cutting speed on surface roughness of the workpiece machined under UAT. A numerical 3D FEM model showed five time increase in MRR in orthogonal machining of Ti15-3-3-3 [15]. Sharma and Dogra [16] showed substantial decrease in cutting forces and improvement in surface finish during UAT of modern high-grade nickel-based alloys C263 and Inconel-718. Most of the above studies concentrated on the influence of UAT on cutting forces and surface finish. However, inadequate attention has been paid to study the influence of UAT on the microanalysis of segmentation and associated microstructural changes. The present research work analyses and compares the UAT and CT of Ti6Al4V in terms of cutting forces, cutting temperature, surface quality assessment, chip morphology, and chip microstructure. Both modelling and experimentation technique were used to study the above mentioned parameters.
2.
FE Model and Experimentation with UAT
2.1 FE simulation of UAT A 2D FE model for orthogonal cutting was developed in DEFORM TM in which a cutting tool of tungsten carbide was assumed to be rigid and workpiece of Ti6Al4V to be plastic. Fig. 1 shows a scheme of relative movement of the workpiece and cutting tool. The ultrasonic vibrations with a frequency of 20 kHz and amplitude of 20
were imposed on the
movement of cutting tool in the direction of cutting velocity.
3
Fig. 1 Relative movement of the workpiece and cutting tool in orthogonal ultrasonic assisted turning Dimensions of the workpiece used in the simulation were 2.5 mm long and 0.8 mm high. The depth of cut (uncut chip thickness) of 0.1 mm was used during all the experiments. Number of segments produced during machining workpiece over a length of 2.5 mm is around 50. In general, it is felt that this study over a length of 2.5 mm is in the steady-state machining region. The model was developed under plain strain conditions and consists of approximately 5000 four node quadrilateral elements initially. The number of elements increases during the simulation to about 6000 due to re-meshing. The size of the element in the process zone is 2
. The cutting forces were observed to vary between 3 to 4% with an increase in the
number of elements. Thus, it is concluded that the results are fairly independent of the number of mesh elements. In the model (see Fig. 2a), the workpiece moves with a constant velocity, which corresponds to cutting speed
and equals to 300 mm/s. Kinematic boundary conditions providing this
type of movement are applied to the bottom side of the workpiece whereas, the top surface is free. Kinematic condition for the side AB, AD and BC of the workpiece is
and
. This is shown in the Fig. 2a. The cutting tool is considered to be rigid with a rake angle of 7° and it was immovable for the simulations of conventional turning process. However, the tool vibrates harmonically in the direction of cutting velocity during the simulation of ultrasonic assisted turning [3] as given by: , where,
, the frequency
= 20 kHz and amplitude
However, the cutting tip vibration velocity =
and
= 20 = 2513
. (300 mm/s).
This is the condition of separation of the tool from the chip during each cycle of ultrasonic vibrations [6].
4
Direction of vibration
Tool B
A
Workpiece
C
D a.
b. TM
Fig. 2a,b 2D Finite element model in DEFORM a. meshing and boundary condition for workpiece b. movement control for cutting tool during UAT
In DEFORM, the ultrasonic vibrations were imposed on the cutting tool by defining the velocity of tool as a function of time. Fig. 2b shows the movement control defined for the cutting tool in DEFORM during UAT. A Johnson- Cook material model accounting for the strain rate sensitivity has been used for Ti6Al4V. It is given as below:
where,
,
strain rate = 2000
;
and
is plastic strain rate and
is reference plastic
are room and melting temperature and
is strain
rate sensitivity of material, respectively. Table 1 gives the value of material parameters
,
used in DEFORM simulation [17]. Normally, it is difficult to obtain
material properties at such a high strain (4 to 6) and strain rate ( the material properties obtained at strain rate of 2000
, therefore,
from Ref. [17] were used in the
UAT simulation in this work. Table 1 Material parameters used for simulation [17] Strain
rate
(
724.7
683.1
0.47
0.035
1
2000
5
2.2 Experimental Set up for UAT The experimental results were needed for two purposes: i.
To validate the simulation results such as cutting forces, cutting temperatures, etc.
ii.
To analyse the machined surface quality effects that could not be understood from the
simulation. An elaborate experimental set up was developed for the UAT of Ti6Al4V. The set up preparation includes design and fabrication of vibrating tool assembly which consists of transducer, booster, horn or tool holder and a tool tip, etc. Figs. 3a,b show the experimental set up developed for UAT. Cutting tool
Horn
Booster Converter Cutting tool
a.
b.
Fig. 3a,b Set up of Ultrasonic assisted turning a. converter and booster arrangement b. Horn and cutting tip. Ultrasonic vibration tool assembly consists of an ultrasonic generator (SG-22-2000W-20 kHz), a piezoelectric transducer or converter to convert the electrical signal from generator to mechanical vibrations, a booster which will boost these vibrations, since the amplitude of vibrations would be very less at the face of the transducer. These vibrations are carried to the cutting tip using a horn. The assembly was mounted on a jig plate (Mild steel) with three supports mounted on it which hold the tool assembly at proper places. A frequency of 20 kHz and amplitude of 20
was used for the experimentation. To calibrate the UAT system,
6
amplitude of vibration of 20
was set in the generator using a potentiometer and
controlling the voltage. Also, it was verified with the help of dial indicator. The vibrations were given to the tool in the feed direction. A hollow pipe of titanium (Ti6Al4V) of 93 mm outer diameter and 1 mm thickness was used as a workpiece for orthogonal turning for this experimentation. Dry machining was carried out for both UAT and CT. Depth of cut of 0.1 mm was used during machining. The difference in diameter before and after machining was measured with coordinate measuring machine to evaluate the variation in the depth of cut due to UAT. No significant change in the depth of cut ( was observed. Therefore, the variation in the cutting forces was attributed due to ultrasonic vibrations. When the cutting tip is vibrated ultrasonically in the direction of feed, then the following limitation is imposed: Feed direction: Feed velocity where,
= feed mm/rev and
Vibrating tip velocity ( )
(2)
= spindle RPM.
In all the experiments, feed velocity used was 0.1 mm/s, which was much lower than vibrating cutting tip velocity of 2513 mm/s. This satisfies the condition of ultrasonic vibration assisted cutting. An ultrasonic vibration in the feed direction improves the machining productivity and is known as sweep cutting [4]. Table 2 gives detailed specifications of workpiece, cutting tool, processing parameters used for the experiments. Table 2 Experimental specifications Ti6Al4V, outer dia. Workpiece 93mm, thickness 1mm KENNAMETAL Cutting tool (DNMG 150608 KC 9225) Cutting Speed 10, 20, 30 m/min Feed rate 0.1mm/rev Depth of cut 0.1mm Vibration parameters Coolant
3.
20 kHz, 20 Dry
Analysis of UAT using simulation
3.1 Mechanism of cutting under UAT This section explains the mechanism of cutting Ti6Al4V in the present work. Various phenomena and stages involved in the mechanism of cutting appear analogous to machining 7
Inconel 718 under UAT as described in an early reference on this topic [6]. Figs. 4a-d show characteristic stages of vibration during simulation of a single cycle of ultrasonic vibrations. The total simulation time during UAT was 0.007 . This includes 140 cycles of vibrations at the frequency of 20 kHz. The cutting process during a single cycle of vibration could be divided into four main stages as shown in Figs. 4a-d. The corresponding chip-tool interface simulation of conventional turning (CT) is shown in Fig. 5a.
Chip
Chip Tool
Tool
Workpiece
Workpiece
a. Stage 1 approach
b. Stage 2 chip contact
Chip
Tool
Tool
Workpiece
Workpiece
c. Stage 3 penetration
d. Stage 4 unloading
Chip Tool
Effective stresses (MPa)
Fig. 4a-d Stress distribution at each step during simulation a. approach b. chip contact c. penetration d. unloading 1450 1400 1350 1300 1250 1200 1150 1100 1050 1000
Conventional Ultrasonic assisted
10 20 30 Cutting speed (m/min)
Workpiece a.
b.
8
Fig. 5 a,b Effective stress distribution during single cycle of vibration ( : 20 m/min, : 0.1 mm/rev, 20 kHz, 20 ) a. stress distribution in UAT b. Effective stresses vs cutting speed
The stress state during UAT is inherently transient and changes for different stages of the ultrasonic vibrations cycle. Also, the stress fluctuates due to formation of shear bands. Therefore, an average effective stress over the time domain was considered for the analysis. Figs. 4a-d show various stages of a single ultrasonic vibration cycle during UAT. During the first stage of cycle, cutting tool approaches the chip, where the effective stresses go up to 215 MPa. In the second stage, the tool comes in contact with the chip, where effective stresses rise up to 1050 MPa. In the third stage, the tool comes in full contact with the chip and penetration begins, where the stresses rise up to 1360 MPa, which are similar in magnitude to that of the CT process, see Fig. 5b. Thus, the stress-state during penetration stage is somewhat analogous to CT. The stresses above 1000 MPa are evident in the primary and secondary deformation zones. During unloading cycle of vibrations, the magnitude of the effective stresses again reduces to 253 MPa. Thus, the stress distribution during both approach and unloading stage (Fig. 4a and d) remains fairly quasistatic. This change in the magnitude of the effective stresses during the cycle of vibrations leads to a reduction in the mean stress during UAT over that of in CT. The reduction in mean stresses during UAT leads to a reduction in interaction forces between cutting tool and workpiece. Fig. 5b shows the effect of an increase in the cutting speed on the average effective stresses generated during the two turning processes. As the cutting speed increases, the frequency of separation of the cutting tool from the chip reduces. Therefore, the difference between the effective stresses generated during UAT and CT decreases. At a cutting speed of 10 m/min, the difference between the effective stresses is around 17%. However, it decreases to 16% and 7%, when the cutting speed is increased to 20 m/min and 30 m/min, respectively. Thus, at higher cutting speeds, application of ultrasonic vibrations may not be very effective.
3.2 Effective strain During UAT, about 27% reduction in an average effective strain is observed as compared with CT, see Fig. 6.
9
10
CT UAT
Strain
8 6 4 2
0 72 307 396 561 742 848 987 1105 1216 1406 1501 1611 1803
0
Time(µs) Fig. 6 Average effective strain during CT and UAT
This reduction in effective strain leads to lesser strain hardening in UAT thereby, lowering the cutting forces, comparing with that of in CT.
3.3 Effect of shear friction (m) In the simulation, initially a constant chip-tool interface shear friction used was 0.6. In the later part, the value of shear friction was changed to 0, 0.4, and 0.8 to understand its effect on cutting and thrust forces during both CT and UAT. Figs. 7a,b show the effect. 140
Conventional Ultrasonic assisted
140
Thrust force (N)
Cutting force (N)
160
120 100 80 60
Conventional Ultrasonic assisted
120 100 80 60 40
40 0
0.4 0.6 0.8 Shear friction (m) a.
0
0.4 0.6 0.8 Shear friction (m) b.
10
Fig. 7 a,b Effect of shear friction (m) on cutting force and thrust force ( : 20 m/min, 0.1 mm/rev, 20 kHz, : 20 ) a. Cutting force v/s shear friction (m) b. Thrust force v/s shear friction
When the shear friction is changed from zero to a value of 0.8, there is an average reduction of 43% in cutting forces and 48% in thrust forces during UAT over that of in CT. It may be noted that with an increase in the shear friction (m), both cutting and thrust forces increase.
3.4 Cutting Temperature (Simulation) One of the most important parameters in metal cutting is temperature in the cutting zones. An increase in cutting temperature can affect the cutting process in terms of thermal softening i.e. reduction in yield stress and other material properties such as Young’s modulus, coefficient of thermal expansion, specific heat and thermal conductivity, etc. The excessive heating can lead to the tool wear and a reduction in the tool life. The simulation results giving the maximum chip-tool interface temperature attained at different cutting speeds during UAT and CT are shown in Fig. 8. During UAT, the cutting tool separates from the chip within each cycle of ultrasonic vibrations. Such intermittent contacts lead to a reduction in the total time for thermal conduction between the tool and chip and cooling due to convective heat transfer to the environment [9]. Thus, as shown in Fig. 8, there is a reduction in maximum cutting temperature attained during UAT than CT. At a cutting speed of 10 m/min, the difference between maximum cutting temperature attained during UAT and CT is 48%, which reduces to 24% at a cutting speed of 20 m/min, and to 16 % at a cutting speed of 30 m/min. Thus, as the cutting speed increases, the difference between the maximum cutting temperature during CT and UAT decreases.
11
Machining temperature (°C)
390
Conventional Ultrasonic assisted
350 310 270 230 190 150 110
10 20 30 Cutting speed (m/min) Fig. 8 Maximum cutting temperatures during UAT and CT
3.5 Cutting Temperature Validation Measurement of cutting temperature was done during both CT and UAT processes by using a thermal imaging camera, FLIR P 640. The emissivity of Ti6Al4V was determined at machining temperature experimentally. Ti6Al4V material was heated to 300 °C in a furnace. The temperature was measured with a thermocouple and the thermal camera at the same time. Emissivity of material in the thermal camera was set such that, the temperature readings from the thermocouple and the camera are identical. The emissivity of Ti6Al4V thus obtained was 0.34 at 300 °C. Figs.9a,b show the images captured by thermal imaging camera giving maximum machining temperature attained after 120
under UAT and CT. With UAT, a 25% reduction in the
Work
Maximum temperature
Maximum temperature Tool
Tool
Machining temperature (°C)
maximum temperature reached during machining was observed, see Figs.9 a,b.
500 400 300 200
CT- experimental CT-predicted UAT-experimental UAT-predicted
100 0 10
20
30
Cutting speed (m/min) a. CT
b. UAT
c.
Fig. 9 a-c Comparison of temperature in cutting zone at 20 m/min, : 0.1 mm/rev, 20 kHz, 20 µm under a. CT b. UAT c. Cutting speed Vs machining temperature 12
As shown in Fig. 9c, with an increase in the cutting speed from 10 m/min to 20 m/min, a 20% reduction in maximum cutting temperature was recorded during UAT over that of in CT. Further, a reduction of 17% is noted when cutting speed is increased from 20 m/min to 30 m/min. However, at a cutting speed of 10 m/min, the cutting temperature under UAT was higher than that of during CT. At the cutting speed of 10 m/min, feed velocity is 0.05 mm/s, which is quite less than the vibrating tip velocity 2513 mm/s. This increases the vibration energy generated at the cutting tip. This may cause higher cutting temperature during UAT. The difference between the measured values of cutting temperature and those obtained from the simulation is shown in Fig. 9c. It is observed that the experimental data of maximum cutting temperature generated during both UAT and CT match well with the corresponding simulation data, with an error of 24% at a cutting speed of 20 m/min. At 20 m/min, a higher cutting temperature was observed compared with the other two cutting speeds. At a cutting speed of 10 m/min, higher vibration energy reduces the cutting temperature due to aerodynamic lubrication. On the other hand, at a cutting speed of 30 m/min, higher contact time increases heat conduction, which prevents rise in the machining temperature.
3.6 Cutting Forces (Simulation) Imposing ultrasonic vibrations on the cutting tool during conventional turning process leads to a reduction in cutting forces. This is due to a reduction in tool-workpiece contact ratio (TWCR) [2]. Thus, an intermittent machining during UAT increases the non-cutting time of the tool, which decreases the cutting forces and helps enhance the tool life. In UAT, the vibrating tool causes the forces to fluctuate over a wide range hence, only the average values of cutting forces were used for comparison. As shown in Fig. 10 a,b, about 40 % decrease in cutting and thrust forces each was observed during UAT as compared to that of in CT.
13
Conventional Ultrasonic assisted
160
130
Thrust forces (N)
Cutting forces (N)
180
140 120 100 80
Conventional Ultrasonic assisted
110
60
90 70 50
10 20 30 Cutting speed (m/min) a.
10 20 30 Cutting speed (m/min) b.
Fig. 10 a,b Influence of cutting speed on a. cutting forces and b. thrust forces under CT and UAT At a cutting speed of 10 m/min, 37 % and, at 30 m/min about 31% decrease in cutting forces was observed during UAT over that of during CT. However, a maximum of 44 % reduction in cutting forces and 50% reduction in thrust forces was observed during UAT over that of in CT at a cutting speed of 20 m/min. Similarly, a reduction by about 48% at a cutting speed of 10 m/min, and 25% at a cutting speed of 30 m/min is observed in thrust forces during UAT over that of during CT. This reduction in cutting and thrust forces during UAT is due to the reduction in the friction between the tool and work piece observed as a result of pulsating cutting characteristic of the cutting tool [2]. In addition to the reduction in friction, avoidance of built up edge formation and aerodynamic lubrication due to vibrations were the other factors that reduces the forces during UAT. At higher cutting speed of 30 m/min, contact between tool-work increases. This causes a rise in temperature at work-tool interface, see Fig. 8. Higher heat generated due to an increase in the cutting speed causes formation of built up edge.
Further, there is a reduction in
aerodynamic lubrication due to an increase in the contact between tool-work [4]. Therefore, at a cutting speed of 30 m/min, a significant reduction in the cutting forces was not observed as compared with that at the cutting speed of 20 m/min. On the other hand at lower cutting speed of 10 m/min, higher material strength due to low heat generated during UAT increases the cutting forces. The lower heat generation is due to reduction in the contact time, which reduces the friction between work-tool interfaces [3]. This is evident from the Fig. 8, which shows a lower machining temperature at cutting speed of 10 m/min. 14
The tool workpiece contact ratio (TWCR) is lesser at lower cutting speed and increases with an increase in cutting speed during UAT. However, the earlier work suggested that UAT method improves cutting quality and tool life at lower cutting speed. Hence, a lower cutting speed is suggested during the UAT process.
The selection of appropriate vibration
parameters such as frequency and amplitude of vibrations are critical for UAT. The effect of an increase in vibration frequency on cutting and thrust forces is shown in Fig. 11a. The effect of an increase in amplitude of vibrations on cutting and thrust forces is shown in Fig. 11b. Similar trends of variation in amplitude and frequency of vibration on the cutting forces
200 180 160 140 120 100 80 60 40
Cutting force(CT) Thrust force(CT) Cutting force(UAT) Thrust force(UAT)
Forces (N)
Forces (N)
were observed for machining of Inconel 718 under UAT [18].
200 180 160 140 120 100 80 60 40
10 20 30 Frequency of vibration (KHz)
Cutting force(CT) Thrust force(CT) Cutting force(UAT) Thrust force(UAT)
10 20 30 Amplitude of vibration (µm)
a.
b.
Fig. 11 a, b Effect of change in vibration parameters on cutting and thrust forces during UAT and CT ( : 20 m/min, 0.1 mm/rev, 20 kHz, 20 ) a. force v/s vibration frequency b. force v/s amplitude of vibrations There is about 13 % decrease in cutting forces and 27 % decrease in thrust forces when the frequency of vibration is increased from 10 kHz to 20 kHz. Further increase in the vibrating frequency from 20 kHz to 30 kHz leads to 14% decrease in cutting forces and 10% decrease in thrust forces. Fig. 8b shows similar effect on cutting forces when the amplitude of vibration is increased from 10
to 30
.
A decrease in cutting force by 17 % and thrust force by 33 % was observed when the amplitude of vibrations is increased from 10 of vibration from 20
to 30
to 20
. Further, an increase in amplitude
leads to a 8 % decrease in cutting forces and a 4 %
decrease in thrust forces.
15
3.7 Validation of cutting forces Dynamometer to measure the cutting forces could not be mounted on lathe due to ultrasonic set up. Therefore, an indirect method to measure cutting and thrust forces was adopted. The cutting forces were evaluated by measuring the deflection of cutting tool holder or horn using dial gauge of least count 1
in both cutting as well as in thrust directions. Using the basic
equations for deflection of solid round bar of titanium, cutting forces were evaluated as given below:
where, N,
is deflection in mm measured using dial indicator given in Table 3,
is length of the bar = 140 mm,
N/mm2 , and
is Modulus of Elasticity = 110
is second moment of inertia which is calculated by
, where,
is force in
is diameter of the
round bar = 27 mm. Table 3 shows the average deflection values measured below the tool in both the directions. About ten readings were taken for each of the processes. Table 3 Average deflection measured and machining forces Deflection Cutting Thrust (N) direction (N) direction ( ( CT 21 23 65 71 UAT
15
16
46
49
It is observed that the evaluated forces match reasonably well with the simulation values. The error being approximately 22% for cutting forces and 5% for thrust forces.
4.
Experimental analysis of UAT
It includes analysis of machined surfaces and morphology of chips generated during UAT and CT.
4.1 Surface quality Surface finish of machined workpiece is extremely sensitive to changes in machining process. Hence, it was used as a criterion to identify the special characteristics during both UAT and CT process. The roughness of machined surface was measured in terms of (average surface roughness) during both UAT and CT. ‘Veeco Surface Profilometer’ was 16
used to measure the
value. The perimeter of the workpiece was cut into five equal parts
and five surface roughness measurements were performed on the cylindrical surface. Figs. 12a,b shows the visual difference in both conventional and ultrasonic machined surfaces. Conventionally machined surfaces have a glossy finish, see Fig.12a. On the other hand, ultrasonic vibration assisted machined surfaces have a matte finish, see Fig.12b.
Glossy finish
a. at CT
Matte finish
b. at UAT
c. at CT
d. at UAT
Fig. 12a-d Surface quality assessment under UAT and CT machined surfaces ( 20 m/min, 0.1 mm/rev, : 20 kHz, : 20 ) Machined surface under a. CT b. UAT; 3D topography under c. CT d. UAT Figs.12 c-d show the 3D micro-topography for both conventional and ultrasonic turned surface at the same cutting conditions. It is observed that the tool marks on the machined surfaces are on higher side during CT. On the other hand, the tool marks on UAT surfaces are smooth and much uniform. A reduction in contact time between the tool and workpiece during UAT leads to a change in material deformation process [2]. This reduces the cutting temperature and tool wear, which ultimately improves the surface quality of the machined workpiece. The average surface roughness that for UAT surface it is 0.6
value measured for CT surface is 1
and
. Thus, a 40% reduction in surface roughness is observed on
UAT surfaces than the CT surfaces. Figs. 13a,b show machined surface images taken using white light interferometer for both CT and UAT surfaces. However, Figs. 13 c,d show optical 17
images for the same. As shown in Fig. 13b and d, during UAT process, material is removed such that it produces ‘fish scale’ like structure on the surface [5]. The ultrasonic vibrations form a high frequency vibro-impact process, which increases the dynamic stiffness of the lathe-tool-workpiece system as a whole and improves the accuracy of turning [4]. Also, it abolishes the built-up edge (BUE) formed during UAT at low cutting speed that reduces the surface roughness.
Fish scale structure 20
a. CT
b. UAT
d. UAT
c. CT
Fig. 13a-d Machined surfaces -White light interferometer and optical images (CT and UAT) Fig. 14a shows the effect of an increase in the cutting speed on average surface roughness during both UAT and CT. Fig. 14b shows the variation in hardness from the machined
2.5
5.5
Conventional Ultrasonic assisted
2
Hardness (GPa)
Avg. surface roughness (Ra)
surfaces.
1.5 1 0.5
5
Conventional Ultrasonic assisted
4.5 4 3.5 3 2.5
0 10
20
30
Cutting speed (m/min) a.
5 15 25 35 45 Distance from machined surface (µm) b.
Fig. 14a,b Effect of ultrasonic vibrations on machined surface a. Avg. surface roughness vs cutting speed b. variation in hardness from machined surface ( : 0.1 mm/rev, 20 kHz, 20 ) As shown in Fig. 14a, at a cutting speed of 20 m/min, which is 14% of maximum vibrating velocity of the cutting tip, the UAT process produces a good surface finish than that of during 18
CT. A 40 % improvement in surface finish was noted during UAT over that of during CT at a cutting speed of 20 m/min. Thus, the above studies demonstrate that, during CT, 100% tool workpiece contact ratio (TWCR) leads to a generation of higher cutting forces, frictional heat, and higher cutting instability, etc. which causes deterioration of machined surface finish and finally produces rough and coarse surface finish. On the other hand, a reduced tool workpiece contact ratio (TWCR) for UAT leads to a generation of lower cutting forces, and frictional heat, which results into smooth and regular machined surface that is better than that of in CT [2]. To study the influence of UAT and CT processes on the microstructure of surface layers of treated materials, nano-indentation tests were carried out using a Nano indenter (model TI900). In this experiment, a three sided pyramid diamond probe was used to indent the surface of the specimen with a constant force of 10 mN, with the indentation depth on a nanometre scale. Indentation was done at 5 indentations was kept 10
from the machined surface and distance between two to minimise the influence of indentation on the material
properties. The periphery of the machined surface during both UAT and CT process was cut into equal parts and then nano-indentations were carried out on machining affected region of sample as indicated in Fig. 15a. Fig.14b shows nano-indentation results on surfaces machined using CT and UAT processes. It is seen that the hardened surface thickness for UATed is lower than that of for the CTed. Average hardness value on UAT surface is 3.98 GPa, which is 16% less than the average hardness of CT surface layer i.e. 4.74 GPa. Figs. 15b,c shows the microstructure of the machined cross-section surface of the specimen, where the nanoindentation tests were carried out for both UAT and CT machined surfaces. 20 µm Machining Affected 39 µm zone
Machining affected zone
Workpiece a.
Region of nano indentation b. CT
Machining Affected zone Region of nano indentation c. UAT
Fig. 15 a-c Optical images for the cross-sectional surfaces during UAT and CT ( m/min, : 0.1 mm/rev, 20 kHz, : 20 )
20
It is observed that, the grains in CTed surface are highly deformed as compared to those in UAT. This can be correlated with the results obtained from nano-indentation tests [6]. As the 19
highly deformed grains make the surface much harder, the CT generates a relatively harder surface layer. On the other hand, less deformed grains are evident on UAT machined surface, which indicates lower machined surface hardness. The shear strain values of the deformed grains are measured using image analyser software. The depth for the deformed grains in UAT surface was 20
and that of in CT surface was 39
. The maximum strain
measured on the deformed grains on UATed surface was 1.27, and that for CTed surface was 2.9. Therefore, a 56% reduction in the shear strain in the deformed grains is observed during UAT over that of in CT.
4.2 Analysis of machined chips In this section, quantitative data on chip morphology obtained during UAT and CT machining are presented. The section also includes chip thickness, segment spacing, width of segment and shear band width. This data were obtained from SEM and optical images of chips generated during machining using an image analyser software.
4.2.1 Chip Segmentation Analysis of chip segmentation and segment geometry gives valuable data on the distribution of stresses on shear plane, chip forming mechanism, material properties at the time of deformation during machining process, etc. A change in chip morphology indicates the cutting forces on the tool and surface integrity of work piece which in turns influences machinability of titanium alloys. Various morphological features obtained during CT and UAT have been analysed, see Table 4 for the details.
20
Table 4 Significance of chip geometry on machining characteristics during UAT and CT
Terminologies
a. Details of chip geometry Chip segmentation
UAT
CT 132
Improvement in UAT 18% reduction
Significance Lower chip compression factor Higher normal stresses and crack size
Chip thickness
95
Segment pitch
52
27
48% increase
Chip length per softening
49
26
45% increase
Less thermal softening, Less shear band formation
Segment Width
48
26
45% increase
More shear band spacing
Crack angle
37
49
23% decrease
Less brittle nature of material
Included angle
39
58
23% decrease
Less plastic deformation
Chip compression factor
1.1
1.3
18% reduction
Minimum plastic deformation
Shear angle
34°
16°
50% increase
Less cutting forces
It is evident from the table that UAT reduces the severity of the machining process on Ti alloys as compared to CT. Figs. 16 a,b show SEM images of chips obtained at CT and UAT. The chip thickness measured for CTed chip was 132
and that for UATed it was 95
.
Therefore, an 18 % reduction in chip thickness was observed in UAT process over that in CT process. 21
95
132
Maximum chip thickness
Chip thickness (µm)
145 135
CT UAT
125 115 105 95
Maximum chip thickness
10
a. CT
b. UAT
20 30 Cutting speed (m/min) c.
Fig.16 a-c Chip morphology under a. CT b. UAT c.cutting speed Vs chip thickness
60
1.3 1.2 1.1 1
Conventional Ultrasonic assisted
Segment width (µm)
Chip compression factor
1.4
50 40 30 20 10
0.9 10 20 30 Cutting speed (m/min) a.
Conventional Ultrasonic assisted
10 20 30 Cutting speed (m/min) b.
Fig.17 a,b Influence of cutting speed on chip geometry at CT and UAT a. chip compression factor Vs cutting speed b. segment width Vs cutting speed It is observed that at the three cutting speeds, chip thickness under UAT was lower than CT, see Fig. 16c. With an increase in cutting speed from 10 m/min to 20 m/min, there is a 10 % increase in chip thickness in UAT and CT. However, with an increase in the cutting speed from 20 m/min to 30 m/min, UAT experienced 13% rise in chip thickness as compared to the 4% rise in the CT. A continuous interaction between tool and workpiece during CT produces thicker chips. On the other hand, non-continuous interaction between tool and workpiece during UAT produces thin chips. The generation of thicker, uneven, cracked chips are not favourable for high quality machining as it results in non-uniform friction, high cutting temperature, cutting forces and rapid tool wear [2]. From the chip thickness a chip compression factor is evaluated. It is obtained by dividing chip thickness with uncut chip thickness, which is feed per revolution in orthogonal turning. 22
Chip compression ratio indicates plastic deformation during machining. The ultimate objective of machining process is to separate a certain layer from rest of work piece with minimum possible plastic deformation and thus the energy consumption. With an increase in the cutting speed, a reduction in chip compression factor is noted in UAT over that of in CT, see Fig. 17a. At a cutting speed 10 and 20 m/min, the chip compression factor for UAT is 16% of that of for CT. However, at a cutting speed of 30 m/min, it is only about 7% of that of for CT. Thus, at higher cutting speed, the consequent increase in the rate of plastic deformation is unlikely to make any difference between UAT and CT. Segment width is another chip geometry parameter which represents distance between two thermal softening or shear band. It is observed that a reduction in cutting temperature during UAT leads to a reduction in thermal softening and hence increases the segment width. During CT, with an increase in the cutting temperature at higher cutting speed, the thermal softening increases, this reduces the segment width, see Fig. 17b. At a cutting speed of 30 m/min both CT and UAT experience higher cutting temperature. Therefore, a decrease in segment width was recorded at this cutting speed than that of at 10 and 20 m/min. Influence of CT and UAT process on shear angle has been analysed. Measurement of shear angle from the SEM images of chip is as illustrated in Fig. 18 a,b. An increase in shear angle was observed in machining under UAT. This further confirms that under UAT, cutting forces reduce. In addition, segment width in UAT appears non-uniform as illustrated in Fig. 18 a,b. The formation of shear band in UAT is not regular, which causes formation of segments of non-uniform width. However, a strange phenomenon was seen during UAT process that as shown in Figs. 18a-c, the chip segmentation does not begin at the edge of the chips as in case of CT, see Fig. 18c. It begins at a distance of 27 m/min. This distance is about 70
from the chip edge (tool face) at the cutting speed of 20 for the cutting speed of 30 m/min.
23
Features: Low shear angle Uniform segment width
Segment width
Features High shear angle Non-Uniform segment width Segment width
27
a. CT
b. UAT
Fig. 18 a,b Chip morphology at a cutting speed of 20 m/min under a. CT b. UAT
4.2.2 Chip Microstructure Shear bands in the chip microstructure represent localized deformation of material in small zones. The surrounding area of shear band remains almost unaffected. Due to shear localization, the tool is subjected to changes in cutting forces, which adversely affects the tool life and integrity of the machined work surfaces [19] [20]. Therefore, an analysis of shear band formation in chips obtained in UAT and CT has been carried out. The microstructure of chips obtained with and without ultrasonic vibrations to the cutting tool at a cutting speed of 20 and 30 m/min, and a feed rate of 0.1 mm/rev is as shown in Figs. 19 a-d.
Shear band
Elongated grains with no shear band
Highly elongated grains a. CT at 20 m/min
Highly elongated grains
b. UAT at 20 m/min
Elongated grains with no shear band
Shear band c. CT at 30 m/min
d. UAT at 30 m/min
24
Fig.19 a-d comparison of chip microstructure at a, c. conventional and b, d. ultrasonic assisted machining In CT, a shear band of 3
width is observed, see Figs.19a,c. On the other hand, in UAT,
shear bands in the chips are not visible, see Figs. 19b,d.
During CT, since there is a
continuous interaction between tool and work, heat continues to generate which leads to a localized thermal softening. This causes formation of shear bands. However, in case of UAT, since there is an intermittent cutting, it leads to less heat concentration and hence no shear bands are observed. As shown in Figs. 20 a,b, a crack length along the shear plane at CT was 16.5 UAT was 34.5
and that at
. Larger crack length in UAT shows predominance of fracture in segment
formation [21]. However, during CT, a reduction in crack length in the chips shows smaller intensity of fracture in segment formation. Here, segments are formed predominantly by shear band [22].
Crack length =16.5
a. CT at 20 m/min
Crack length =34.5 b. CT at 20 m/min
Fig. 20 a,b Size of crack at free surface of chip ( : 20 m/min, :0.1 mm/rev, :20 µm) a. under CT b. under UAT
:20 kHz,
It is observed that grains in the CT chips are more elongated than that in UAT, see Figs. 19 a,d. These elongated and small width grains make the chip harder which in turn exerts significant abrading forces on the cutting tool. However, during the UAT, grains in the chip are less elongated. Thus, the chip appears strain relieved and therefore, less hardened at UAT. These chips exert relatively less reactive forces on the face of cutting tool. Conclusions
25
A 2D FE model of UAT has been developed in DEFORM. The simulation of single cycle of vibration shows that the cutting process is divided into four stages – approach, tool-chip contact, penetration and unloading. A cutting tool in UAT is subjected to the same stress level as that of in CT only in the penetration stage. During other stages, the tool in UAT is subjected to considerably lower stresses than in CT. Similar observations were made by Mitrofanov et al. 2003 in machining Inconel 718 under UAT.
As the cutting speed increases from 10 to 30 m/min, the difference between effective stresses in CT and UAT reduces from 17% to 6%. Thus, application of ultrasonic vibrations may not be very effective at higher cutting speeds.
The simulation shows that a reduction in TWCR (tool-work contact ratio) in UAT and the transient process conditions during UAT, leads to a considerable (45-50%) reduction in the cutting forces in UAT than that of in CT. Also, the selection of appropriate frequency and amplitude of ultrasonic vibrations during UAT reduces the cutting forces.
Further, a reduction in contact time between tool and chip, and the consequent convective cooling of tools during UAT reduce the cutting temperature during UAT. As the cutting speed increases from 10 m/min to 30 m/min, the difference between maximum cutting temperature attained during UAT and CT is reduced from 48% to 16%.
The experimental assessment of UAT process on titanium alloy shows that the UATed surfaces show matte appearance while the CTed show glossy appearance. However, the average machined surface roughness (
) in UAT is lower than that of
during CT.
The microstructure along the cross-sections of machined surfaces shows that the grains along the UATed surfaces are deformed lesser than those along the CTed surface cross-sections.
The chips produced during UAT are thin and continues. However, CT generates thick and uneven chips. The intensity of thermal softening is reduced during UAT over that of in CT. This indicates a reduction in the shear band formation in UATed chips than that of in CTed chips.
Acknowledgement
26
The authors gratefully acknowledge the partial support provided for this work by National Centre for Aerospace Innovation and Research, IIT Bombay, a Dept. of Science and Technology- Government of India, The Boeing Company and IIT Bombay Collaboration. Author also acknowledges Roop Telsonic Ultrasonic Ltd. Mumbai for providing ultrasonic generator and its required setting for the experiments. References 1.
2. 3.
4.
5. 6.
7. 8. 9. 10.
11. 12.
13. 14.
15. 16.
R. Muhammad, N. Ahmed, M. Demiral, A. Roy, V.Silberschmidt, Computational Study of Ultrasonically-Assisted Turning of Ti alloys, Advanced Materials Research, Vol. 223 (2011) pp. 30-36. C. Nath, M.Rahman, Effect of machining parameters in ultrasonic vibration cutting, International Journal of Machine Tools and Manufacture, Vol. 48 (2008) pp. 965-974. V.I.Babitsky, A. Mitrofanov, V. Silberschmidt, Ultrasonically assisted turning of aviation materials: simulations and experimental study, Journal of Ultrasonics, Vol. 42, (2009) pp. 81-86. V. Babitsky, A. Kalashnikov, A. Meadows, A. Wijesundara, Ultrasonically assisted turning of aviation materials. Journal of Materials Processing Technology, Vol. 132 ( 2003) pp. 157-167. S.Koshimizu, Ultrasonic Vibration assisted cutting of Titanium alloy, Key Engineering Materials, Vols. 389-390 (2009) 277-282. A. Mitrofanov, V. Babitsky, V. Silberschmidt, Finite element simulations of ultrasonically assisted turning, Computational Materials Science, Vol. 28 (2003) pp. 645-653. Shamoto, Moriwaki, Study on Elliptical Vibration Cutting, CIRP Annals, Vol. 43/1,(1994) pp. 35-38. J.Kumabe, Ultrasonic superposition vibration cutting of ceramics, Precision Engineering, Vol. 041. A. Maurotto, A. Roy, V. Babitsky, V. Silberschnidt, Analysis of machinability of Tiand Ni-based alloys, Solid State Phenomena , Vol. 188 (2012) pp. 330-338. A. Maurotto, R. Muhammad, A. Roy, V. Babitsky, V. Silberschmidt. Comparing machinability of Ti-15-3-3-3 and Ni-625 alloys in UAT, Procedia CIRP , Vol.1, (2012) pp. 330-335. R. Muhammad, A. Maurotto, A. Roy, V. Silberschmidt, Hot ultrasonically assisted turning of β-Ti alloy, Procedia CIRP, Vol. 1(2012) pp. 336-341. S.Voronia, V. Babitsky, Auto resonant control strategies of loaded ultrasonic transducer for machining applications, Journal of Sound and Vibration, Vol. 313 (2008) pp. 395-417. Z.Tang, C. Liu, J. Yi. FE simulation of ultrasonic vibration orthogonal cutting of Ti6Al-4V, Advanced Materials Research, Vols. 97-101 (2010) pp. 1933-1936. A. Chegini, J. Akbari, A. Rajabnejad, Ultrasonically assisted turning of NiTi based shape memory alloy.5th International Conference and Exhibition on Design and Production of Machines and Dies/Molds, 18-21 June 2009 Pine Bay Hotel Kusadasi, Aydin, Turkey R. Muhammad, N. Ahmed, A. Roy, V. Silberschmidt, Numerical Modelling of Vibration-Assisted Turning of Ti-15333, Procedia CIRP , Vol. 1 (2012) pp. 347-352. V.S.Sharma, M. Dogra, N.M Suri, Advances in the turning process for productivity improvement –a review, Journal of Engineering Manufacture, (2008) pp. 222:1417.
27
17. 18. 19. 20. 21.
22.
23. 24.
Woei-Shyan Lee, Chi-Feng Lin, Plastic deformation and fracture behaviour of Ti6Al 4V alloy, Material Science and Engineering A, 241 (1998) 48-59. G. Boothroyds, W. Knight. Fundamental of Machining and Machine Tool. New york : Marcel Dekker Inc., 1989. N. Ahmed, A.V. Mitrofanov, V.I. Babitsky, V.V. Silberschmidt, Analysis of forces in ultrasonically assisted turning, J. Sound Vibrat., Vols. 308 (3-5) (2007) pp. 845-854. E. O. Ezugwu, J. Bonney, An overview of the machinability of aeroengine alloys., Journal of Materials Processing Technology, Vol. 134 (2003) pp. 233–253. S. Sun, M. Brandt, M.S. Dargush, Characteristics of cutting forces and chip formation in machining. 2009, International Journal of Machine Tools and Manufacture , Vol. 49 (2009) pp.561–568. A. Vyas, M. Shaw, Mechanics of saw-tooth chip formation in metal cutting . 1999, Journal of Manufacturing Science and Engineering—Transactions of the ASME, Vol. 211 (1999) pp.163–172. H. Zhen-Bin, R. Komanduri, On a Thermo mechanical model of shear instability in Machining, Annals of the ClRP, Vol. 44/1, (1995) pp. 69-74. Robert W.Cahn, P. Haasen. Physical Metallurgy. Sara Bwgedmtstraat : ELSEVIER SCIENCE B.V., 1996.
28
Highlights •
A comprehensive 2D FE transient simulation of ultrasonic assisted turning (UAT) is presented.
•
Simulation shows four stages of chip-tool interaction in UAT and reduced stress level on tool.
•
A 45-50% reduction in cutting forces and 48% reduction in cutting temperature is observed in UAT.
•
The experimental analysis of the UAT process shows that machined surfaces have matte finish.
•
Microstructural study shows thermal softening and shear band formation in chips reduce in UAT.
29
S0041-624X(13)00270-9 http://dx.doi.org/10.1016/j.ultras.2013.09.010 ULTRAS 4676
To appear in:
Ultrasonics
Please cite this article as: S. Patil, S. Joshi, A. Tewari, S.S. Joshi, Modeling and Simulation of Effect of Ultrasonic Vibrations on Machining of Ti6Al4V, Ultrasonics (2013), doi: http://dx.doi.org/10.1016/j.ultras.2013.09.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Modeling and Simulation of Effect of Ultrasonic Vibrations on Machining of Ti6Al4V Sandip Patil, Shashikant Joshi, Asim Tewari, Suhas S. Joshi1 Department of Mechanical Engineering Indian Institute of Technology Bombay, Powai, Mumbai – 400 076 India
Abstract: The titanium alloys cause high machining heat generation and consequent rapid wear of cutting tool edges during machining. The ultrasonic assisted turning (UAT) has been found to be very effective in machining of various materials; especially in the machining of “difficult-to-cut” material like Ti6Al4V. The present work is a comprehensive study involving 2D FE transient simulation of UAT in DEFORM framework and their experimental characterization. The simulation shows that UAT reduces the stress level on cutting tool during machining as compared to that of in continuous turning (CT) barring the penetration stage, wherein both tools are subjected to identical stress levels. There is a 40-45 % reduction in cutting forces and about 48% reduction in cutting temperature in UAT over that of in CT. However, the reduction magnitude reduces with an increase in the cutting speed. The experimental analysis of UAT process shows that the surface roughness in UAT is lower than in CT, and the UATed surfaces have matte finish as against the glossy finish on the CTed surfaces. Microstructural observations of the chips and machined surfaces in both processes reveal that the intensity of thermal softening and shear band formation is reduced in UAT over that of in CT. Keywords: Machining, titanium, ultrasonic vibrations, finite element simulation, chip microstructure, cutting forces
1.
Introduction
Titanium is recognised for its strategic importance as a unique lightweight, high strength, structurally efficient metal for critical and high performance aircraft such as jet engine and airframe components. These alloys are termed as ‘difficult-to-machine’ materials but have high utility in aerospace, medical and other manufacturing sectors. Ultrasonically assisted turning (UAT) is an advanced machining technique, in which up to 30000 low energy vibro-impacts are superimposed on the cutting tool. This technique of UAT was developed in 1960s in which high frequency ultrasonic vibrations are superimposed 1
Corresponding Author: Ph: +91 22 2576 7527; Email: [email protected]
1
on the conventional movement of cutting tool, providing a range of benefits in machining hard metal alloys [1]. Various parameters of UAT such as vibration frequency, amplitude of vibrations, direction of ultrasonic vibrations, friction at the tool tip-workpiece interface and cutting speed, affect the level of cutting forces generated during the process. The projected benefits of UAT are summarized as below: •
An improvement in the dynamic cutting stability [2].
•
A reduction in the cutting forces and improvement in tool life [3] [2] [4].
•
A reduction in surface roughness of machined surface [2] [5].
•
A reduction in residual stresses in the machined workpiece [6].
•
A reduction in cutting temperature [6].
Use of ultrasonic vibrations in machining was introduced about 24 years ago [7]. Ultrasonic vibration on the cutting tool during machining of ceramic reduced cutting tool and work temperature. Tensile loading of ceramic due to vibration was considered as a reason for smooth and efficient cutting [8]. A significant reduction in chip thickness and cutting forces was observed by imparting a synchronised vibration to the cutting tool in cutting and chip flow direction [7]. With UAT, a large increase in MRR of the order of 2-3 times was observed in machining Ti15-3-3-3 and Ni-625 [9] [10]. Further, an increase in MRR and surface finish in machining
titanium alloy was observed by combining UAT with external
heating of the workpiece [11]. Several researchers have studied the effect of such vibrations assisted techniques to machine ‘difficult-to-cut’ materials such as Inconel 718. However, research has been scarce on its application and effective machining of Ti6Al4V. Babitsky et al. [4] [12] found that the UAT method performs better at lower cutting speed using high frequency and high amplitude of vibration. Koshimizu [5] observed a reduction in cutting forces under UAT of Ti6Al4V. The author also found that the vibration assisted technique holds good for Ti6Al4V, only when the cutting speed was 30% below the maximum vibrating velocity of the cutting tip. Nath and Rahman [2] studied the effect of machining parameters during UAT of Inconel 718. Muhammad et al. [1] developed a 3D FE (Finite Element) model to compare the cutting forces during UAT and CT for Ti-15. An increase in the amplitude and frequency of vibration has reduced cutting forces to 25% and 36% respectively. An increase in cutting tool velocity and a reduction in tool-work contact was attributed as a reason for such a large reduction in the cutting forces. Tang et al. [13] developed a 2D FE model for Ti6Al4V and observed a 2
higher shear angle beside a reduction in the cutting temperature and friction coefficient, as compared to CT. Ahmed et al. [4] developed a 2D FE model to study the stresses under UAT of Inconel 718. Mitrofanov et al. [6] developed a 2D FE model to study the plastic and residual strains in the machined layer and compared the results with nano-indentation tests on Inconel 718. Amini and Nategh [14] studied the effect of parameters viz. vibration amplitude, depth of cut, feed rate and cutting speed on surface roughness of the workpiece machined under UAT. A numerical 3D FEM model showed five time increase in MRR in orthogonal machining of Ti15-3-3-3 [15]. Sharma and Dogra [16] showed substantial decrease in cutting forces and improvement in surface finish during UAT of modern high-grade nickel-based alloys C263 and Inconel-718. Most of the above studies concentrated on the influence of UAT on cutting forces and surface finish. However, inadequate attention has been paid to study the influence of UAT on the microanalysis of segmentation and associated microstructural changes. The present research work analyses and compares the UAT and CT of Ti6Al4V in terms of cutting forces, cutting temperature, surface quality assessment, chip morphology, and chip microstructure. Both modelling and experimentation technique were used to study the above mentioned parameters.
2.
FE Model and Experimentation with UAT
2.1 FE simulation of UAT A 2D FE model for orthogonal cutting was developed in DEFORM TM in which a cutting tool of tungsten carbide was assumed to be rigid and workpiece of Ti6Al4V to be plastic. Fig. 1 shows a scheme of relative movement of the workpiece and cutting tool. The ultrasonic vibrations with a frequency of 20 kHz and amplitude of 20
were imposed on the
movement of cutting tool in the direction of cutting velocity.
3
Fig. 1 Relative movement of the workpiece and cutting tool in orthogonal ultrasonic assisted turning Dimensions of the workpiece used in the simulation were 2.5 mm long and 0.8 mm high. The depth of cut (uncut chip thickness) of 0.1 mm was used during all the experiments. Number of segments produced during machining workpiece over a length of 2.5 mm is around 50. In general, it is felt that this study over a length of 2.5 mm is in the steady-state machining region. The model was developed under plain strain conditions and consists of approximately 5000 four node quadrilateral elements initially. The number of elements increases during the simulation to about 6000 due to re-meshing. The size of the element in the process zone is 2
. The cutting forces were observed to vary between 3 to 4% with an increase in the
number of elements. Thus, it is concluded that the results are fairly independent of the number of mesh elements. In the model (see Fig. 2a), the workpiece moves with a constant velocity, which corresponds to cutting speed
and equals to 300 mm/s. Kinematic boundary conditions providing this
type of movement are applied to the bottom side of the workpiece whereas, the top surface is free. Kinematic condition for the side AB, AD and BC of the workpiece is
and
. This is shown in the Fig. 2a. The cutting tool is considered to be rigid with a rake angle of 7° and it was immovable for the simulations of conventional turning process. However, the tool vibrates harmonically in the direction of cutting velocity during the simulation of ultrasonic assisted turning [3] as given by: , where,
, the frequency
= 20 kHz and amplitude
However, the cutting tip vibration velocity =
and
= 20 = 2513
. (300 mm/s).
This is the condition of separation of the tool from the chip during each cycle of ultrasonic vibrations [6].
4
Direction of vibration
Tool B
A
Workpiece
C
D a.
b. TM
Fig. 2a,b 2D Finite element model in DEFORM a. meshing and boundary condition for workpiece b. movement control for cutting tool during UAT
In DEFORM, the ultrasonic vibrations were imposed on the cutting tool by defining the velocity of tool as a function of time. Fig. 2b shows the movement control defined for the cutting tool in DEFORM during UAT. A Johnson- Cook material model accounting for the strain rate sensitivity has been used for Ti6Al4V. It is given as below:
where,
,
strain rate = 2000
;
and
is plastic strain rate and
is reference plastic
are room and melting temperature and
is strain
rate sensitivity of material, respectively. Table 1 gives the value of material parameters
,
used in DEFORM simulation [17]. Normally, it is difficult to obtain
material properties at such a high strain (4 to 6) and strain rate ( the material properties obtained at strain rate of 2000
, therefore,
from Ref. [17] were used in the
UAT simulation in this work. Table 1 Material parameters used for simulation [17] Strain
rate
(
724.7
683.1
0.47
0.035
1
2000
5
2.2 Experimental Set up for UAT The experimental results were needed for two purposes: i.
To validate the simulation results such as cutting forces, cutting temperatures, etc.
ii.
To analyse the machined surface quality effects that could not be understood from the
simulation. An elaborate experimental set up was developed for the UAT of Ti6Al4V. The set up preparation includes design and fabrication of vibrating tool assembly which consists of transducer, booster, horn or tool holder and a tool tip, etc. Figs. 3a,b show the experimental set up developed for UAT. Cutting tool
Horn
Booster Converter Cutting tool
a.
b.
Fig. 3a,b Set up of Ultrasonic assisted turning a. converter and booster arrangement b. Horn and cutting tip. Ultrasonic vibration tool assembly consists of an ultrasonic generator (SG-22-2000W-20 kHz), a piezoelectric transducer or converter to convert the electrical signal from generator to mechanical vibrations, a booster which will boost these vibrations, since the amplitude of vibrations would be very less at the face of the transducer. These vibrations are carried to the cutting tip using a horn. The assembly was mounted on a jig plate (Mild steel) with three supports mounted on it which hold the tool assembly at proper places. A frequency of 20 kHz and amplitude of 20
was used for the experimentation. To calibrate the UAT system,
6
amplitude of vibration of 20
was set in the generator using a potentiometer and
controlling the voltage. Also, it was verified with the help of dial indicator. The vibrations were given to the tool in the feed direction. A hollow pipe of titanium (Ti6Al4V) of 93 mm outer diameter and 1 mm thickness was used as a workpiece for orthogonal turning for this experimentation. Dry machining was carried out for both UAT and CT. Depth of cut of 0.1 mm was used during machining. The difference in diameter before and after machining was measured with coordinate measuring machine to evaluate the variation in the depth of cut due to UAT. No significant change in the depth of cut ( was observed. Therefore, the variation in the cutting forces was attributed due to ultrasonic vibrations. When the cutting tip is vibrated ultrasonically in the direction of feed, then the following limitation is imposed: Feed direction: Feed velocity where,
= feed mm/rev and
Vibrating tip velocity ( )
(2)
= spindle RPM.
In all the experiments, feed velocity used was 0.1 mm/s, which was much lower than vibrating cutting tip velocity of 2513 mm/s. This satisfies the condition of ultrasonic vibration assisted cutting. An ultrasonic vibration in the feed direction improves the machining productivity and is known as sweep cutting [4]. Table 2 gives detailed specifications of workpiece, cutting tool, processing parameters used for the experiments. Table 2 Experimental specifications Ti6Al4V, outer dia. Workpiece 93mm, thickness 1mm KENNAMETAL Cutting tool (DNMG 150608 KC 9225) Cutting Speed 10, 20, 30 m/min Feed rate 0.1mm/rev Depth of cut 0.1mm Vibration parameters Coolant
3.
20 kHz, 20 Dry
Analysis of UAT using simulation
3.1 Mechanism of cutting under UAT This section explains the mechanism of cutting Ti6Al4V in the present work. Various phenomena and stages involved in the mechanism of cutting appear analogous to machining 7
Inconel 718 under UAT as described in an early reference on this topic [6]. Figs. 4a-d show characteristic stages of vibration during simulation of a single cycle of ultrasonic vibrations. The total simulation time during UAT was 0.007 . This includes 140 cycles of vibrations at the frequency of 20 kHz. The cutting process during a single cycle of vibration could be divided into four main stages as shown in Figs. 4a-d. The corresponding chip-tool interface simulation of conventional turning (CT) is shown in Fig. 5a.
Chip
Chip Tool
Tool
Workpiece
Workpiece
a. Stage 1 approach
b. Stage 2 chip contact
Chip
Tool
Tool
Workpiece
Workpiece
c. Stage 3 penetration
d. Stage 4 unloading
Chip Tool
Effective stresses (MPa)
Fig. 4a-d Stress distribution at each step during simulation a. approach b. chip contact c. penetration d. unloading 1450 1400 1350 1300 1250 1200 1150 1100 1050 1000
Conventional Ultrasonic assisted
10 20 30 Cutting speed (m/min)
Workpiece a.
b.
8
Fig. 5 a,b Effective stress distribution during single cycle of vibration ( : 20 m/min, : 0.1 mm/rev, 20 kHz, 20 ) a. stress distribution in UAT b. Effective stresses vs cutting speed
The stress state during UAT is inherently transient and changes for different stages of the ultrasonic vibrations cycle. Also, the stress fluctuates due to formation of shear bands. Therefore, an average effective stress over the time domain was considered for the analysis. Figs. 4a-d show various stages of a single ultrasonic vibration cycle during UAT. During the first stage of cycle, cutting tool approaches the chip, where the effective stresses go up to 215 MPa. In the second stage, the tool comes in contact with the chip, where effective stresses rise up to 1050 MPa. In the third stage, the tool comes in full contact with the chip and penetration begins, where the stresses rise up to 1360 MPa, which are similar in magnitude to that of the CT process, see Fig. 5b. Thus, the stress-state during penetration stage is somewhat analogous to CT. The stresses above 1000 MPa are evident in the primary and secondary deformation zones. During unloading cycle of vibrations, the magnitude of the effective stresses again reduces to 253 MPa. Thus, the stress distribution during both approach and unloading stage (Fig. 4a and d) remains fairly quasistatic. This change in the magnitude of the effective stresses during the cycle of vibrations leads to a reduction in the mean stress during UAT over that of in CT. The reduction in mean stresses during UAT leads to a reduction in interaction forces between cutting tool and workpiece. Fig. 5b shows the effect of an increase in the cutting speed on the average effective stresses generated during the two turning processes. As the cutting speed increases, the frequency of separation of the cutting tool from the chip reduces. Therefore, the difference between the effective stresses generated during UAT and CT decreases. At a cutting speed of 10 m/min, the difference between the effective stresses is around 17%. However, it decreases to 16% and 7%, when the cutting speed is increased to 20 m/min and 30 m/min, respectively. Thus, at higher cutting speeds, application of ultrasonic vibrations may not be very effective.
3.2 Effective strain During UAT, about 27% reduction in an average effective strain is observed as compared with CT, see Fig. 6.
9
10
CT UAT
Strain
8 6 4 2
0 72 307 396 561 742 848 987 1105 1216 1406 1501 1611 1803
0
Time(µs) Fig. 6 Average effective strain during CT and UAT
This reduction in effective strain leads to lesser strain hardening in UAT thereby, lowering the cutting forces, comparing with that of in CT.
3.3 Effect of shear friction (m) In the simulation, initially a constant chip-tool interface shear friction used was 0.6. In the later part, the value of shear friction was changed to 0, 0.4, and 0.8 to understand its effect on cutting and thrust forces during both CT and UAT. Figs. 7a,b show the effect. 140
Conventional Ultrasonic assisted
140
Thrust force (N)
Cutting force (N)
160
120 100 80 60
Conventional Ultrasonic assisted
120 100 80 60 40
40 0
0.4 0.6 0.8 Shear friction (m) a.
0
0.4 0.6 0.8 Shear friction (m) b.
10
Fig. 7 a,b Effect of shear friction (m) on cutting force and thrust force ( : 20 m/min, 0.1 mm/rev, 20 kHz, : 20 ) a. Cutting force v/s shear friction (m) b. Thrust force v/s shear friction
When the shear friction is changed from zero to a value of 0.8, there is an average reduction of 43% in cutting forces and 48% in thrust forces during UAT over that of in CT. It may be noted that with an increase in the shear friction (m), both cutting and thrust forces increase.
3.4 Cutting Temperature (Simulation) One of the most important parameters in metal cutting is temperature in the cutting zones. An increase in cutting temperature can affect the cutting process in terms of thermal softening i.e. reduction in yield stress and other material properties such as Young’s modulus, coefficient of thermal expansion, specific heat and thermal conductivity, etc. The excessive heating can lead to the tool wear and a reduction in the tool life. The simulation results giving the maximum chip-tool interface temperature attained at different cutting speeds during UAT and CT are shown in Fig. 8. During UAT, the cutting tool separates from the chip within each cycle of ultrasonic vibrations. Such intermittent contacts lead to a reduction in the total time for thermal conduction between the tool and chip and cooling due to convective heat transfer to the environment [9]. Thus, as shown in Fig. 8, there is a reduction in maximum cutting temperature attained during UAT than CT. At a cutting speed of 10 m/min, the difference between maximum cutting temperature attained during UAT and CT is 48%, which reduces to 24% at a cutting speed of 20 m/min, and to 16 % at a cutting speed of 30 m/min. Thus, as the cutting speed increases, the difference between the maximum cutting temperature during CT and UAT decreases.
11
Machining temperature (°C)
390
Conventional Ultrasonic assisted
350 310 270 230 190 150 110
10 20 30 Cutting speed (m/min) Fig. 8 Maximum cutting temperatures during UAT and CT
3.5 Cutting Temperature Validation Measurement of cutting temperature was done during both CT and UAT processes by using a thermal imaging camera, FLIR P 640. The emissivity of Ti6Al4V was determined at machining temperature experimentally. Ti6Al4V material was heated to 300 °C in a furnace. The temperature was measured with a thermocouple and the thermal camera at the same time. Emissivity of material in the thermal camera was set such that, the temperature readings from the thermocouple and the camera are identical. The emissivity of Ti6Al4V thus obtained was 0.34 at 300 °C. Figs.9a,b show the images captured by thermal imaging camera giving maximum machining temperature attained after 120
under UAT and CT. With UAT, a 25% reduction in the
Work
Maximum temperature
Maximum temperature Tool
Tool
Machining temperature (°C)
maximum temperature reached during machining was observed, see Figs.9 a,b.
500 400 300 200
CT- experimental CT-predicted UAT-experimental UAT-predicted
100 0 10
20
30
Cutting speed (m/min) a. CT
b. UAT
c.
Fig. 9 a-c Comparison of temperature in cutting zone at 20 m/min, : 0.1 mm/rev, 20 kHz, 20 µm under a. CT b. UAT c. Cutting speed Vs machining temperature 12
As shown in Fig. 9c, with an increase in the cutting speed from 10 m/min to 20 m/min, a 20% reduction in maximum cutting temperature was recorded during UAT over that of in CT. Further, a reduction of 17% is noted when cutting speed is increased from 20 m/min to 30 m/min. However, at a cutting speed of 10 m/min, the cutting temperature under UAT was higher than that of during CT. At the cutting speed of 10 m/min, feed velocity is 0.05 mm/s, which is quite less than the vibrating tip velocity 2513 mm/s. This increases the vibration energy generated at the cutting tip. This may cause higher cutting temperature during UAT. The difference between the measured values of cutting temperature and those obtained from the simulation is shown in Fig. 9c. It is observed that the experimental data of maximum cutting temperature generated during both UAT and CT match well with the corresponding simulation data, with an error of 24% at a cutting speed of 20 m/min. At 20 m/min, a higher cutting temperature was observed compared with the other two cutting speeds. At a cutting speed of 10 m/min, higher vibration energy reduces the cutting temperature due to aerodynamic lubrication. On the other hand, at a cutting speed of 30 m/min, higher contact time increases heat conduction, which prevents rise in the machining temperature.
3.6 Cutting Forces (Simulation) Imposing ultrasonic vibrations on the cutting tool during conventional turning process leads to a reduction in cutting forces. This is due to a reduction in tool-workpiece contact ratio (TWCR) [2]. Thus, an intermittent machining during UAT increases the non-cutting time of the tool, which decreases the cutting forces and helps enhance the tool life. In UAT, the vibrating tool causes the forces to fluctuate over a wide range hence, only the average values of cutting forces were used for comparison. As shown in Fig. 10 a,b, about 40 % decrease in cutting and thrust forces each was observed during UAT as compared to that of in CT.
13
Conventional Ultrasonic assisted
160
130
Thrust forces (N)
Cutting forces (N)
180
140 120 100 80
Conventional Ultrasonic assisted
110
60
90 70 50
10 20 30 Cutting speed (m/min) a.
10 20 30 Cutting speed (m/min) b.
Fig. 10 a,b Influence of cutting speed on a. cutting forces and b. thrust forces under CT and UAT At a cutting speed of 10 m/min, 37 % and, at 30 m/min about 31% decrease in cutting forces was observed during UAT over that of during CT. However, a maximum of 44 % reduction in cutting forces and 50% reduction in thrust forces was observed during UAT over that of in CT at a cutting speed of 20 m/min. Similarly, a reduction by about 48% at a cutting speed of 10 m/min, and 25% at a cutting speed of 30 m/min is observed in thrust forces during UAT over that of during CT. This reduction in cutting and thrust forces during UAT is due to the reduction in the friction between the tool and work piece observed as a result of pulsating cutting characteristic of the cutting tool [2]. In addition to the reduction in friction, avoidance of built up edge formation and aerodynamic lubrication due to vibrations were the other factors that reduces the forces during UAT. At higher cutting speed of 30 m/min, contact between tool-work increases. This causes a rise in temperature at work-tool interface, see Fig. 8. Higher heat generated due to an increase in the cutting speed causes formation of built up edge.
Further, there is a reduction in
aerodynamic lubrication due to an increase in the contact between tool-work [4]. Therefore, at a cutting speed of 30 m/min, a significant reduction in the cutting forces was not observed as compared with that at the cutting speed of 20 m/min. On the other hand at lower cutting speed of 10 m/min, higher material strength due to low heat generated during UAT increases the cutting forces. The lower heat generation is due to reduction in the contact time, which reduces the friction between work-tool interfaces [3]. This is evident from the Fig. 8, which shows a lower machining temperature at cutting speed of 10 m/min. 14
The tool workpiece contact ratio (TWCR) is lesser at lower cutting speed and increases with an increase in cutting speed during UAT. However, the earlier work suggested that UAT method improves cutting quality and tool life at lower cutting speed. Hence, a lower cutting speed is suggested during the UAT process.
The selection of appropriate vibration
parameters such as frequency and amplitude of vibrations are critical for UAT. The effect of an increase in vibration frequency on cutting and thrust forces is shown in Fig. 11a. The effect of an increase in amplitude of vibrations on cutting and thrust forces is shown in Fig. 11b. Similar trends of variation in amplitude and frequency of vibration on the cutting forces
200 180 160 140 120 100 80 60 40
Cutting force(CT) Thrust force(CT) Cutting force(UAT) Thrust force(UAT)
Forces (N)
Forces (N)
were observed for machining of Inconel 718 under UAT [18].
200 180 160 140 120 100 80 60 40
10 20 30 Frequency of vibration (KHz)
Cutting force(CT) Thrust force(CT) Cutting force(UAT) Thrust force(UAT)
10 20 30 Amplitude of vibration (µm)
a.
b.
Fig. 11 a, b Effect of change in vibration parameters on cutting and thrust forces during UAT and CT ( : 20 m/min, 0.1 mm/rev, 20 kHz, 20 ) a. force v/s vibration frequency b. force v/s amplitude of vibrations There is about 13 % decrease in cutting forces and 27 % decrease in thrust forces when the frequency of vibration is increased from 10 kHz to 20 kHz. Further increase in the vibrating frequency from 20 kHz to 30 kHz leads to 14% decrease in cutting forces and 10% decrease in thrust forces. Fig. 8b shows similar effect on cutting forces when the amplitude of vibration is increased from 10
to 30
.
A decrease in cutting force by 17 % and thrust force by 33 % was observed when the amplitude of vibrations is increased from 10 of vibration from 20
to 30
to 20
. Further, an increase in amplitude
leads to a 8 % decrease in cutting forces and a 4 %
decrease in thrust forces.
15
3.7 Validation of cutting forces Dynamometer to measure the cutting forces could not be mounted on lathe due to ultrasonic set up. Therefore, an indirect method to measure cutting and thrust forces was adopted. The cutting forces were evaluated by measuring the deflection of cutting tool holder or horn using dial gauge of least count 1
in both cutting as well as in thrust directions. Using the basic
equations for deflection of solid round bar of titanium, cutting forces were evaluated as given below:
where, N,
is deflection in mm measured using dial indicator given in Table 3,
is length of the bar = 140 mm,
N/mm2 , and
is Modulus of Elasticity = 110
is second moment of inertia which is calculated by
, where,
is force in
is diameter of the
round bar = 27 mm. Table 3 shows the average deflection values measured below the tool in both the directions. About ten readings were taken for each of the processes. Table 3 Average deflection measured and machining forces Deflection Cutting Thrust (N) direction (N) direction ( ( CT 21 23 65 71 UAT
15
16
46
49
It is observed that the evaluated forces match reasonably well with the simulation values. The error being approximately 22% for cutting forces and 5% for thrust forces.
4.
Experimental analysis of UAT
It includes analysis of machined surfaces and morphology of chips generated during UAT and CT.
4.1 Surface quality Surface finish of machined workpiece is extremely sensitive to changes in machining process. Hence, it was used as a criterion to identify the special characteristics during both UAT and CT process. The roughness of machined surface was measured in terms of (average surface roughness) during both UAT and CT. ‘Veeco Surface Profilometer’ was 16
used to measure the
value. The perimeter of the workpiece was cut into five equal parts
and five surface roughness measurements were performed on the cylindrical surface. Figs. 12a,b shows the visual difference in both conventional and ultrasonic machined surfaces. Conventionally machined surfaces have a glossy finish, see Fig.12a. On the other hand, ultrasonic vibration assisted machined surfaces have a matte finish, see Fig.12b.
Glossy finish
a. at CT
Matte finish
b. at UAT
c. at CT
d. at UAT
Fig. 12a-d Surface quality assessment under UAT and CT machined surfaces ( 20 m/min, 0.1 mm/rev, : 20 kHz, : 20 ) Machined surface under a. CT b. UAT; 3D topography under c. CT d. UAT Figs.12 c-d show the 3D micro-topography for both conventional and ultrasonic turned surface at the same cutting conditions. It is observed that the tool marks on the machined surfaces are on higher side during CT. On the other hand, the tool marks on UAT surfaces are smooth and much uniform. A reduction in contact time between the tool and workpiece during UAT leads to a change in material deformation process [2]. This reduces the cutting temperature and tool wear, which ultimately improves the surface quality of the machined workpiece. The average surface roughness that for UAT surface it is 0.6
value measured for CT surface is 1
and
. Thus, a 40% reduction in surface roughness is observed on
UAT surfaces than the CT surfaces. Figs. 13a,b show machined surface images taken using white light interferometer for both CT and UAT surfaces. However, Figs. 13 c,d show optical 17
images for the same. As shown in Fig. 13b and d, during UAT process, material is removed such that it produces ‘fish scale’ like structure on the surface [5]. The ultrasonic vibrations form a high frequency vibro-impact process, which increases the dynamic stiffness of the lathe-tool-workpiece system as a whole and improves the accuracy of turning [4]. Also, it abolishes the built-up edge (BUE) formed during UAT at low cutting speed that reduces the surface roughness.
Fish scale structure 20
a. CT
b. UAT
d. UAT
c. CT
Fig. 13a-d Machined surfaces -White light interferometer and optical images (CT and UAT) Fig. 14a shows the effect of an increase in the cutting speed on average surface roughness during both UAT and CT. Fig. 14b shows the variation in hardness from the machined
2.5
5.5
Conventional Ultrasonic assisted
2
Hardness (GPa)
Avg. surface roughness (Ra)
surfaces.
1.5 1 0.5
5
Conventional Ultrasonic assisted
4.5 4 3.5 3 2.5
0 10
20
30
Cutting speed (m/min) a.
5 15 25 35 45 Distance from machined surface (µm) b.
Fig. 14a,b Effect of ultrasonic vibrations on machined surface a. Avg. surface roughness vs cutting speed b. variation in hardness from machined surface ( : 0.1 mm/rev, 20 kHz, 20 ) As shown in Fig. 14a, at a cutting speed of 20 m/min, which is 14% of maximum vibrating velocity of the cutting tip, the UAT process produces a good surface finish than that of during 18
CT. A 40 % improvement in surface finish was noted during UAT over that of during CT at a cutting speed of 20 m/min. Thus, the above studies demonstrate that, during CT, 100% tool workpiece contact ratio (TWCR) leads to a generation of higher cutting forces, frictional heat, and higher cutting instability, etc. which causes deterioration of machined surface finish and finally produces rough and coarse surface finish. On the other hand, a reduced tool workpiece contact ratio (TWCR) for UAT leads to a generation of lower cutting forces, and frictional heat, which results into smooth and regular machined surface that is better than that of in CT [2]. To study the influence of UAT and CT processes on the microstructure of surface layers of treated materials, nano-indentation tests were carried out using a Nano indenter (model TI900). In this experiment, a three sided pyramid diamond probe was used to indent the surface of the specimen with a constant force of 10 mN, with the indentation depth on a nanometre scale. Indentation was done at 5 indentations was kept 10
from the machined surface and distance between two to minimise the influence of indentation on the material
properties. The periphery of the machined surface during both UAT and CT process was cut into equal parts and then nano-indentations were carried out on machining affected region of sample as indicated in Fig. 15a. Fig.14b shows nano-indentation results on surfaces machined using CT and UAT processes. It is seen that the hardened surface thickness for UATed is lower than that of for the CTed. Average hardness value on UAT surface is 3.98 GPa, which is 16% less than the average hardness of CT surface layer i.e. 4.74 GPa. Figs. 15b,c shows the microstructure of the machined cross-section surface of the specimen, where the nanoindentation tests were carried out for both UAT and CT machined surfaces. 20 µm Machining Affected 39 µm zone
Machining affected zone
Workpiece a.
Region of nano indentation b. CT
Machining Affected zone Region of nano indentation c. UAT
Fig. 15 a-c Optical images for the cross-sectional surfaces during UAT and CT ( m/min, : 0.1 mm/rev, 20 kHz, : 20 )
20
It is observed that, the grains in CTed surface are highly deformed as compared to those in UAT. This can be correlated with the results obtained from nano-indentation tests [6]. As the 19
highly deformed grains make the surface much harder, the CT generates a relatively harder surface layer. On the other hand, less deformed grains are evident on UAT machined surface, which indicates lower machined surface hardness. The shear strain values of the deformed grains are measured using image analyser software. The depth for the deformed grains in UAT surface was 20
and that of in CT surface was 39
. The maximum strain
measured on the deformed grains on UATed surface was 1.27, and that for CTed surface was 2.9. Therefore, a 56% reduction in the shear strain in the deformed grains is observed during UAT over that of in CT.
4.2 Analysis of machined chips In this section, quantitative data on chip morphology obtained during UAT and CT machining are presented. The section also includes chip thickness, segment spacing, width of segment and shear band width. This data were obtained from SEM and optical images of chips generated during machining using an image analyser software.
4.2.1 Chip Segmentation Analysis of chip segmentation and segment geometry gives valuable data on the distribution of stresses on shear plane, chip forming mechanism, material properties at the time of deformation during machining process, etc. A change in chip morphology indicates the cutting forces on the tool and surface integrity of work piece which in turns influences machinability of titanium alloys. Various morphological features obtained during CT and UAT have been analysed, see Table 4 for the details.
20
Table 4 Significance of chip geometry on machining characteristics during UAT and CT
Terminologies
a. Details of chip geometry Chip segmentation
UAT
CT 132
Improvement in UAT 18% reduction
Significance Lower chip compression factor Higher normal stresses and crack size
Chip thickness
95
Segment pitch
52
27
48% increase
Chip length per softening
49
26
45% increase
Less thermal softening, Less shear band formation
Segment Width
48
26
45% increase
More shear band spacing
Crack angle
37
49
23% decrease
Less brittle nature of material
Included angle
39
58
23% decrease
Less plastic deformation
Chip compression factor
1.1
1.3
18% reduction
Minimum plastic deformation
Shear angle
34°
16°
50% increase
Less cutting forces
It is evident from the table that UAT reduces the severity of the machining process on Ti alloys as compared to CT. Figs. 16 a,b show SEM images of chips obtained at CT and UAT. The chip thickness measured for CTed chip was 132
and that for UATed it was 95
.
Therefore, an 18 % reduction in chip thickness was observed in UAT process over that in CT process. 21
95
132
Maximum chip thickness
Chip thickness (µm)
145 135
CT UAT
125 115 105 95
Maximum chip thickness
10
a. CT
b. UAT
20 30 Cutting speed (m/min) c.
Fig.16 a-c Chip morphology under a. CT b. UAT c.cutting speed Vs chip thickness
60
1.3 1.2 1.1 1
Conventional Ultrasonic assisted
Segment width (µm)
Chip compression factor
1.4
50 40 30 20 10
0.9 10 20 30 Cutting speed (m/min) a.
Conventional Ultrasonic assisted
10 20 30 Cutting speed (m/min) b.
Fig.17 a,b Influence of cutting speed on chip geometry at CT and UAT a. chip compression factor Vs cutting speed b. segment width Vs cutting speed It is observed that at the three cutting speeds, chip thickness under UAT was lower than CT, see Fig. 16c. With an increase in cutting speed from 10 m/min to 20 m/min, there is a 10 % increase in chip thickness in UAT and CT. However, with an increase in the cutting speed from 20 m/min to 30 m/min, UAT experienced 13% rise in chip thickness as compared to the 4% rise in the CT. A continuous interaction between tool and workpiece during CT produces thicker chips. On the other hand, non-continuous interaction between tool and workpiece during UAT produces thin chips. The generation of thicker, uneven, cracked chips are not favourable for high quality machining as it results in non-uniform friction, high cutting temperature, cutting forces and rapid tool wear [2]. From the chip thickness a chip compression factor is evaluated. It is obtained by dividing chip thickness with uncut chip thickness, which is feed per revolution in orthogonal turning. 22
Chip compression ratio indicates plastic deformation during machining. The ultimate objective of machining process is to separate a certain layer from rest of work piece with minimum possible plastic deformation and thus the energy consumption. With an increase in the cutting speed, a reduction in chip compression factor is noted in UAT over that of in CT, see Fig. 17a. At a cutting speed 10 and 20 m/min, the chip compression factor for UAT is 16% of that of for CT. However, at a cutting speed of 30 m/min, it is only about 7% of that of for CT. Thus, at higher cutting speed, the consequent increase in the rate of plastic deformation is unlikely to make any difference between UAT and CT. Segment width is another chip geometry parameter which represents distance between two thermal softening or shear band. It is observed that a reduction in cutting temperature during UAT leads to a reduction in thermal softening and hence increases the segment width. During CT, with an increase in the cutting temperature at higher cutting speed, the thermal softening increases, this reduces the segment width, see Fig. 17b. At a cutting speed of 30 m/min both CT and UAT experience higher cutting temperature. Therefore, a decrease in segment width was recorded at this cutting speed than that of at 10 and 20 m/min. Influence of CT and UAT process on shear angle has been analysed. Measurement of shear angle from the SEM images of chip is as illustrated in Fig. 18 a,b. An increase in shear angle was observed in machining under UAT. This further confirms that under UAT, cutting forces reduce. In addition, segment width in UAT appears non-uniform as illustrated in Fig. 18 a,b. The formation of shear band in UAT is not regular, which causes formation of segments of non-uniform width. However, a strange phenomenon was seen during UAT process that as shown in Figs. 18a-c, the chip segmentation does not begin at the edge of the chips as in case of CT, see Fig. 18c. It begins at a distance of 27 m/min. This distance is about 70
from the chip edge (tool face) at the cutting speed of 20 for the cutting speed of 30 m/min.
23
Features: Low shear angle Uniform segment width
Segment width
Features High shear angle Non-Uniform segment width Segment width
27
a. CT
b. UAT
Fig. 18 a,b Chip morphology at a cutting speed of 20 m/min under a. CT b. UAT
4.2.2 Chip Microstructure Shear bands in the chip microstructure represent localized deformation of material in small zones. The surrounding area of shear band remains almost unaffected. Due to shear localization, the tool is subjected to changes in cutting forces, which adversely affects the tool life and integrity of the machined work surfaces [19] [20]. Therefore, an analysis of shear band formation in chips obtained in UAT and CT has been carried out. The microstructure of chips obtained with and without ultrasonic vibrations to the cutting tool at a cutting speed of 20 and 30 m/min, and a feed rate of 0.1 mm/rev is as shown in Figs. 19 a-d.
Shear band
Elongated grains with no shear band
Highly elongated grains a. CT at 20 m/min
Highly elongated grains
b. UAT at 20 m/min
Elongated grains with no shear band
Shear band c. CT at 30 m/min
d. UAT at 30 m/min
24
Fig.19 a-d comparison of chip microstructure at a, c. conventional and b, d. ultrasonic assisted machining In CT, a shear band of 3
width is observed, see Figs.19a,c. On the other hand, in UAT,
shear bands in the chips are not visible, see Figs. 19b,d.
During CT, since there is a
continuous interaction between tool and work, heat continues to generate which leads to a localized thermal softening. This causes formation of shear bands. However, in case of UAT, since there is an intermittent cutting, it leads to less heat concentration and hence no shear bands are observed. As shown in Figs. 20 a,b, a crack length along the shear plane at CT was 16.5 UAT was 34.5
and that at
. Larger crack length in UAT shows predominance of fracture in segment
formation [21]. However, during CT, a reduction in crack length in the chips shows smaller intensity of fracture in segment formation. Here, segments are formed predominantly by shear band [22].
Crack length =16.5
a. CT at 20 m/min
Crack length =34.5 b. CT at 20 m/min
Fig. 20 a,b Size of crack at free surface of chip ( : 20 m/min, :0.1 mm/rev, :20 µm) a. under CT b. under UAT
:20 kHz,
It is observed that grains in the CT chips are more elongated than that in UAT, see Figs. 19 a,d. These elongated and small width grains make the chip harder which in turn exerts significant abrading forces on the cutting tool. However, during the UAT, grains in the chip are less elongated. Thus, the chip appears strain relieved and therefore, less hardened at UAT. These chips exert relatively less reactive forces on the face of cutting tool. Conclusions
25
A 2D FE model of UAT has been developed in DEFORM. The simulation of single cycle of vibration shows that the cutting process is divided into four stages – approach, tool-chip contact, penetration and unloading. A cutting tool in UAT is subjected to the same stress level as that of in CT only in the penetration stage. During other stages, the tool in UAT is subjected to considerably lower stresses than in CT. Similar observations were made by Mitrofanov et al. 2003 in machining Inconel 718 under UAT.
As the cutting speed increases from 10 to 30 m/min, the difference between effective stresses in CT and UAT reduces from 17% to 6%. Thus, application of ultrasonic vibrations may not be very effective at higher cutting speeds.
The simulation shows that a reduction in TWCR (tool-work contact ratio) in UAT and the transient process conditions during UAT, leads to a considerable (45-50%) reduction in the cutting forces in UAT than that of in CT. Also, the selection of appropriate frequency and amplitude of ultrasonic vibrations during UAT reduces the cutting forces.
Further, a reduction in contact time between tool and chip, and the consequent convective cooling of tools during UAT reduce the cutting temperature during UAT. As the cutting speed increases from 10 m/min to 30 m/min, the difference between maximum cutting temperature attained during UAT and CT is reduced from 48% to 16%.
The experimental assessment of UAT process on titanium alloy shows that the UATed surfaces show matte appearance while the CTed show glossy appearance. However, the average machined surface roughness (
) in UAT is lower than that of
during CT.
The microstructure along the cross-sections of machined surfaces shows that the grains along the UATed surfaces are deformed lesser than those along the CTed surface cross-sections.
The chips produced during UAT are thin and continues. However, CT generates thick and uneven chips. The intensity of thermal softening is reduced during UAT over that of in CT. This indicates a reduction in the shear band formation in UATed chips than that of in CTed chips.
Acknowledgement
26
The authors gratefully acknowledge the partial support provided for this work by National Centre for Aerospace Innovation and Research, IIT Bombay, a Dept. of Science and Technology- Government of India, The Boeing Company and IIT Bombay Collaboration. Author also acknowledges Roop Telsonic Ultrasonic Ltd. Mumbai for providing ultrasonic generator and its required setting for the experiments. References 1.
2. 3.
4.
5. 6.
7. 8. 9. 10.
11. 12.
13. 14.
15. 16.
R. Muhammad, N. Ahmed, M. Demiral, A. Roy, V.Silberschmidt, Computational Study of Ultrasonically-Assisted Turning of Ti alloys, Advanced Materials Research, Vol. 223 (2011) pp. 30-36. C. Nath, M.Rahman, Effect of machining parameters in ultrasonic vibration cutting, International Journal of Machine Tools and Manufacture, Vol. 48 (2008) pp. 965-974. V.I.Babitsky, A. Mitrofanov, V. Silberschmidt, Ultrasonically assisted turning of aviation materials: simulations and experimental study, Journal of Ultrasonics, Vol. 42, (2009) pp. 81-86. V. Babitsky, A. Kalashnikov, A. Meadows, A. Wijesundara, Ultrasonically assisted turning of aviation materials. Journal of Materials Processing Technology, Vol. 132 ( 2003) pp. 157-167. S.Koshimizu, Ultrasonic Vibration assisted cutting of Titanium alloy, Key Engineering Materials, Vols. 389-390 (2009) 277-282. A. Mitrofanov, V. Babitsky, V. Silberschmidt, Finite element simulations of ultrasonically assisted turning, Computational Materials Science, Vol. 28 (2003) pp. 645-653. Shamoto, Moriwaki, Study on Elliptical Vibration Cutting, CIRP Annals, Vol. 43/1,(1994) pp. 35-38. J.Kumabe, Ultrasonic superposition vibration cutting of ceramics, Precision Engineering, Vol. 041. A. Maurotto, A. Roy, V. Babitsky, V. Silberschnidt, Analysis of machinability of Tiand Ni-based alloys, Solid State Phenomena , Vol. 188 (2012) pp. 330-338. A. Maurotto, R. Muhammad, A. Roy, V. Babitsky, V. Silberschmidt. Comparing machinability of Ti-15-3-3-3 and Ni-625 alloys in UAT, Procedia CIRP , Vol.1, (2012) pp. 330-335. R. Muhammad, A. Maurotto, A. Roy, V. Silberschmidt, Hot ultrasonically assisted turning of β-Ti alloy, Procedia CIRP, Vol. 1(2012) pp. 336-341. S.Voronia, V. Babitsky, Auto resonant control strategies of loaded ultrasonic transducer for machining applications, Journal of Sound and Vibration, Vol. 313 (2008) pp. 395-417. Z.Tang, C. Liu, J. Yi. FE simulation of ultrasonic vibration orthogonal cutting of Ti6Al-4V, Advanced Materials Research, Vols. 97-101 (2010) pp. 1933-1936. A. Chegini, J. Akbari, A. Rajabnejad, Ultrasonically assisted turning of NiTi based shape memory alloy.5th International Conference and Exhibition on Design and Production of Machines and Dies/Molds, 18-21 June 2009 Pine Bay Hotel Kusadasi, Aydin, Turkey R. Muhammad, N. Ahmed, A. Roy, V. Silberschmidt, Numerical Modelling of Vibration-Assisted Turning of Ti-15333, Procedia CIRP , Vol. 1 (2012) pp. 347-352. V.S.Sharma, M. Dogra, N.M Suri, Advances in the turning process for productivity improvement –a review, Journal of Engineering Manufacture, (2008) pp. 222:1417.
27
17. 18. 19. 20. 21.
22.
23. 24.
Woei-Shyan Lee, Chi-Feng Lin, Plastic deformation and fracture behaviour of Ti6Al 4V alloy, Material Science and Engineering A, 241 (1998) 48-59. G. Boothroyds, W. Knight. Fundamental of Machining and Machine Tool. New york : Marcel Dekker Inc., 1989. N. Ahmed, A.V. Mitrofanov, V.I. Babitsky, V.V. Silberschmidt, Analysis of forces in ultrasonically assisted turning, J. Sound Vibrat., Vols. 308 (3-5) (2007) pp. 845-854. E. O. Ezugwu, J. Bonney, An overview of the machinability of aeroengine alloys., Journal of Materials Processing Technology, Vol. 134 (2003) pp. 233–253. S. Sun, M. Brandt, M.S. Dargush, Characteristics of cutting forces and chip formation in machining. 2009, International Journal of Machine Tools and Manufacture , Vol. 49 (2009) pp.561–568. A. Vyas, M. Shaw, Mechanics of saw-tooth chip formation in metal cutting . 1999, Journal of Manufacturing Science and Engineering—Transactions of the ASME, Vol. 211 (1999) pp.163–172. H. Zhen-Bin, R. Komanduri, On a Thermo mechanical model of shear instability in Machining, Annals of the ClRP, Vol. 44/1, (1995) pp. 69-74. Robert W.Cahn, P. Haasen. Physical Metallurgy. Sara Bwgedmtstraat : ELSEVIER SCIENCE B.V., 1996.
28
Highlights •
A comprehensive 2D FE transient simulation of ultrasonic assisted turning (UAT) is presented.
•
Simulation shows four stages of chip-tool interaction in UAT and reduced stress level on tool.
•
A 45-50% reduction in cutting forces and 48% reduction in cutting temperature is observed in UAT.
•
The experimental analysis of the UAT process shows that machined surfaces have matte finish.
•
Microstructural study shows thermal softening and shear band formation in chips reduce in UAT.
29