Authors Accepted Manuscript

Effect of machining parameters in ultrasonic vibration cutting Chandra Nath, M. Rahman PII: DOI: Reference: To appear in: S0890-6955(08)00025-4 doi:10.1016/j.ijmachtools.2008.01.013 MTM 2243 International Journal of Machine Tools & Manufacture 7 November 2007 21 January 2008 25 January 2008
www.elsevier.com/locate/ijmactool

Received date: Revised date: Accepted date:

Cite this article as: Chandra Nath and M. Rahman, Effect of machining parameters in ultrasonic vibration cutting, International Journal of Machine Tools & Manufacture (2008), doi:10.1016/j.ijmachtools.2008.01.013 This is a PDF le 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 galley proof before it is published in its nal citable 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.

Effect of machining parameters in ultrasonic vibration cutting Chandra Nath, M. Rahman* Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576

First author: Chandra Nath, Institute: National University of Singapore, Singapore Postal address: Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576 Tel: +65-6516 4644 Fax: +65-6779 1459 E-mail: g0500294@nus.edu.sg

*Corresponding author: Prof. M. Rahman

Institute: National University of Singapore, Singapore Postal address: Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576 Tel: +65-6516 2168 Fax: +65-6779 1459

E-mail: mpemusta@nus.edu.sg

c A

e c

d te p

n a

s u

ri c

t p

Abstract Ultrasonic vibration cutting (UVC) method is an efficient cutting technique for difficult-tomachine materials. It is found that the UVC mechanism is influenced by three important parameters: tool vibration frequency, tool vibration amplitude and workpiece cutting speed that determine the cutting force. However, the relation between the cutting force and these three parameters in the UVC is not clearly established. This paper presents firstly the mechanism how these parameters effect the UVC. With theoretical studies, it is established that tool-workpiece contact ratio (TWCR) plays a key role in the UVC process where the increase in both the tool vibration parameters and the decrease in the cutting speed reduce the TWCR that in turn reduces both cutting force and tool wear, improves surface quality and prolongs tool life. This paper also experimentally investigates the effect of cutting parameters on cutting performances in the cutting of Inconel 718 by applying both the UVC and the conventional turning (CT) methods. It is observed that the UVC method promises better surface finish and improves tool life in hard cutting at low cutting speed as compared to the CT method. The experiments also show that the TWCR, when investigating the effect of cutting speed, has a significant effect on both the cutting force and the tool wear in the UVC method which substantiates the theoretical findings.

Keywords: Ultrasonic vibration cutting; Tool vibration parameters, Cutting speed; Toolworkpiece contact ratio; Cutting tool life.

c A

e c

d te p

n a

s u

ri c

t p

1. Introduction

Ultrasonic vibration cutting (UVC) method is a more effective cutting process over conventional turning (CT) in terms of cutting force, cutting instability, tool blunting, tool wear, chip generation, surface finish and so on, to machine difficult-to-cut materials such as Ni- and Ti-based super alloys, hardened steels, optical glasses, ceramics, tungsten carbides, etc. [1-17].This method improves productivity by saving the manufacturing time by about 510 % as well as the machining cost by about 30% of the cost of parts as required in the deburring process of precision parts [18].

Along with the usual parameters for the CT method, two additional parameters: tool vibration frequency and vibration amplitude, are considered in ultrasonic vibration cutting (UVC) system that assist to improve cutting quality and to increase remarkable tool life by lowering, mainly, the cutting force and improving the dynamic cutting stability. Extensive theoretical research [12-16], simulation [17] and experimental results [8-12, 19, 20] for the UVC method mention that the lower cutting force is due to a considerable reduction of friction between the tool and workpiece [17, 19] and the separating or pulse cutting characteristic of the tool [12-14, 18, 21]. Previous experimental reports also indicated that the UVC method performs better at low cutting speeds [6-8, 18-22] and at both high tool vibration frequency and amplitude [23]. Moreover, the UVC mechanism during the tool-workpiece interaction is basically related to these above three parameters [3, 8, 13, 15, 20]. Therefore, the cutting force is directly related to these three parameters. However, the relationship between the cutting force and these three parameters has not yet been established. Furthermore, V. I. Babitsky et al [2, 6] justified and compared the axial surface profile and roundness profile for the cutting of Inconel 718 by applying both the CT and the UVC

c A

e c

d te p

n a

s u

ri c

t p

methods. However, the ability of the cubic boron nitride (CBN) tool for this excellent tough and creep-rupture resistance alloy and the effect of cutting parameters on the cutting performance by applying the UVC method have not yet been studied. As a hard and tough cutting inexpensive tool material, CBN is the best choice, next only to the diamond tools. The authors in this paper investigate firstly the mechanism of the effect of these three parameters in the UVC process. It is theoretically understood that the amount of the cutting force is directly related with these three parameters that establish two key factors: toolworkpiece contact ratio and tool-workpiece relative speed (hereafter, called TWCR and TWRS, respectively), for this intermittent cutting technique. This paper focuses mainly on controlling the first key factor, TWCR. It is found that the increase in both the tool vibration frequency and the vibration amplitude, and the decrease in the workpiece cutting speed reduce the TWCR. As the TWCR decreases, the non-cutting time of the tool increases that decreases the cutting force and enhances both increased tool life and improved cutting quality.

This paper also investigates the effect of cutting parameters such as cutting speed and feed rate in cutting Inconel 718. The cutting quality is evaluated in terms of three cutting force components, tool wear, chip formation and surface roughness for both the UVC and the CT methods. Additionally, it observes and discusses the behavior of tool failure and the formation of the chips, at different cutting conditions, by means of scanning electron microscopy (SEM). Finally, a comparative study has been configured to show the advantages of the UVC method over the CT method. It is concluded that the UVC method performs better than the CT method up to a certain cutting speed range. However, beyond this range, the CBN tools catastrophically failed for machining of Inconel 718 in the UVC method. The authors consider that this type of failure during ultrasonic cutting is due to a number of consecutive high impacts between the tool and the workpiece for long durations at a comparatively high cutting speed. The higher the cutting speed, the larger the TWCR

c A

e c

d te p

n a

s u

ri c

t p

is. Therefore, the cutting speed has a significant effect on cutting force and tool wear in the UVC method that substantiates the theoretical findings. Finally, it is remarked that the UVC method offers better cutting performance in hard and tough cutting materials at low cutting speed and at high tool vibration frequency and amplitude.

2. Theory

2.1 Study of ultrasonic vibration cutting mechanism Fig. 1 shows an illustration of a UVC system where a piezoelectric transducer (PZT) is configured into the tool shank to excite the tool in any desired direction corresponding to three conventional axes. In this study, a tangential or cutting directional type UVC system is considered as implemented by most of the previous researchers [3, 8-16, 20]. The rest of the cutting system in this technique is same as conventional turning set up.

Fig. 1. Schematic of ultrasonic vibration cutting

The cutting principle of the UVC method was explained by the previous researchers [8, 12, 13, 15, 20]. The tool oscillates at an ultrasonic frequency f (i.e. vibration period,

with a very small vibration amplitude a where the workpiece rotates with a constant cutting speed v c . Accordingly in Fig. 2, the cutting edge starts to vibrate from the origin O and then cuts the workpiece material during the interaction periods t ab , t ab .. .

c A

e c

d te p

n a

s u

ri c

t p

T = 1/ f

Fig. 2. Pulse cutting state in the UVC method [15].

The displacement of vibrating tool is described by:

x = a sin t = a sin 2ft

(1)

where x and are the displacement and the angular velocity of the tool, respectively. Thus the tool vibration speed is vt = x = a cos t that varies from a minimum of & any peak or valley to a maximum of
(vt ) max = a = 2af
(vt ) min = 0

at

at midpoint of its either upward or

downward motion, except the initial tool speed at the origin O. It was realized [3, 8, 12, 13, 15, 20] that the ultrasonic cutting is satisfied if a > vc , otherwise it becomes a conventional cutting process. Therefore, the critical cutting speed in the UVC method is defined as
(vc ) cr = a = 2af = (vt ) max

Fig. 2 also illustrates three following basic equations that govern the UVC system:

v c + a cos t b = 0

(at t = t b )

a sin (T + t a ) a sin t b = vc (T + t a t b )
r = tc / T

where r is the tool-workpiece contact ratio (TWCR).

Eqs.(2)-(4) formulate a final equation as obtained by [13]:

v c (1 r ) = 2af sin r cos[cos 1 (v c /(2af )) r ]

c A

e c

d te p
(for

n a

s u
(2) (3) (4)

ri c

t p

2 af > v c )

(5)

Therefore, it is clear from the final Eq. [5] that the term TWCR during ultrasonic cutting is dependent on three vital parameters; f, a and vc. Since this technique cuts the workpiece for a certain period in each vibration cycle, the cutting force in this method also should be TWCR ( = t c / T ) times of that for the continuous cutting [11, 14]. That means a lower value

of TWCR can decrease the cutting force in the UVC system. The above Eq. [5] directly indicates that the TWCR can be lowered by controlling those three important parameters. In the following sections 2.2-2.4, the effect of these factors are studied theoretically first. Then the effect of the third factor, workpiece cutting speed, is justified in the experiment in section 4.1.

2.2 The effect of tool vibration frequency

Suppose two UVC systems vibrate at two different frequencies f1 = 20 kHz and f2 = 35 kHz with the same amplitude a = 15 m. Hence the tool vibration period T1 is higher than T2. Fig. 3(a) combines both the UVC systems including the tool displacement curves and their corresponding pulse cutting states as a function of time.

Fig. 3. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration frequencies (Subscripts: 1, 2 are for 20 kHz, 35 kHz, respectively), (b) Relation between TWCR and tool vibration frequency, f.

Now let a workpiece, rotating with a constant cutting speed vc , engages at a1 and disengages at b1 during the upward and downward motion, respectively, of the tool for the low frequency UVC system. Thus the system for this case follows the same contact period tc1 as t b1 a1 , t b1 a1 , t b1a1 ,... in the consecutive vibration cycles. Similarly, the high frequency tool for the same workpiece cutting speed follows the contact period tc2 as
t b2 a2 , t b2 a2 , t b2 a2 ,... . It is clear from the figure that tc1 > tc2. But since T1 > T2, it can not

c A

e c

d te p

n a

s u

ri c

t p

directly be determined whether

r1 > r2

or vice versa.

Fig. 3(b) plots the TWCR against the tool vibration frequency using the Eq. [5] where a = 15 m and vc = 15 m/min. Two different values of TWCR, r1 = 0.2162 and r2 = 0.1600 can

be found for f1 = 20 kHz and f2 = 35 kHz, respectively. Thus, it is clear that the TWCR for a low frequency tool is higher than for a high frequency tool in the UVC system. Therefore, the tool cutting area experiences for a short duration of pulsating cutting force when using a high frequency tool. Though the number of contact between the tool and the workpiece always will be higher for a relatively high frequency cutting tool, it was experimentally reported [23] that the increase of tool vibration frequency in the UVC system improves cutting quality and prolongs tool life that agrees the theoretical studies.

2.3 The effect of tool vibration amplitude

Again suppose another two tangential UVC systems operate two different tool vibration amplitudes a1 = 10 m and a2 = 25 m where the frequency f = 20 kHz is fixed. Thus the tool vibration period T is same for both the systems. Fig. 4(a) illustrates a combined diagram of two different displacement curves and corresponding pulse cutting states against the time cycle for these two systems. If a workpiece rotates with a constant cutting speed vc for both the conditions, then the tool of small vibration amplitude interacts with the workpiece earlier at c1 and separates from the workpiece later at d1 as compared to the tool of large vibration amplitude. In this manner, let the tool for the former case follows the tool-workpiece contact time t c1 as
t d1 c1 , t d1 c1 ,... in the successive passes where the tool for the later case follows t c 2 as

c A

e c

d te p

n a

s u

ri c

t p

t d 2 c2 , t d 2 c2 ,... . It is clearly seen from Fig. 4(a) that t c1 > t c 2 . Thus r1 > r2 because T1 = T2 .

Fig. 4. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration amplitudes (Subscripts: 1, 2 are for 10 m, 25 m, respectively), (b) Relation between TWCR and tool vibration amplitude, a.

Fig. 4(b) shows the TWCR against the tool vibration amplitude by using the final Eq.[5] where f = 20 kHz and vc = 15 m/min. Two different values of TWCR, r1 = 0.2710 and r2 = 0.1644 can be picked out at a1 = 10 m and a2 = 25 m respectively where r1 > r2 . Therefore, it is obvious that the TWCR for a tool of a high vibration amplitude is lower than for a tool of a small vibration amplitude. As the number of contacts between the tool and the workpiece is same for both the tool conditions [see Fig. 4(a)], the tool for the second condition favors better cutting performance in the UVC system. Y.-L Zhang [23] experimentally investigated that the increase of tool vibration amplitude in the UVC system improves cutting quality and saves tool life which substantiates the theoretical findings.

2.4 The effect of workpiece cutting speed

Fig. 5(a) shows a tool displacement diagram and corresponding pulse cutting state against time for a single tool following f = 20 kHz and a = 15 m. Let two cutting speeds vc1 and
vc 2 for the workpiece be considered where, vc1 < vc 2 . Since vc1 is the smallest, the workpiece

for this case holds with the tool later at p1 and then separates earlier at q1 during the upward and downward motion of the tool respectively. Accordingly, the tool-workpiece in this condition follows the contact period t c1 as t q p , t q p ,... in the consecutive vibration cycles.
1 1 1 1

c A

e c

d te p

n a

s u

ri c

t p

On the other hand, the workpiece of cutting speed vc 2 engages with the tool earlier at p2 and disengages later at q2 with following the tool-workpiece contact period t c 2 as t q2 p2 , t q2 p2 ,... . From the following Fig. 5(a), it is clear that t c1 < t c 2 and hence frequency was applied to the cutting tool.
r1 < r2

, because same

Fig. 5. UVC process: (a) Tool displacement and pulsating cutting force against time at f = 20 kHz and a = 15 m. (Subscripts: 1, 2 are for low and high cutting speed, respectively), (b) Relation between TWCR and workpiece cutting speed, vc .

The curve for TWCR vs the workpiece cutting speed is plotted in Fig. 5(b) by using again the final Eq. [5]. If two different cutting speeds, 20 m/min and 40 m/min, are considered then the TWCR is found to about 0.2536 for 20 m/min as compared to about 0.3803 for 40 m/min. Therefore, the TWCR for a low workpiece cutting speed in the UVC system is lower than that for a high workpiece cutting speed. That means the tool experiences a short duration of the pulsating cutting force when applying low cutting speed.

Moreover, the TWRS in the UVC method increases with the increase in cutting speed that definitely effects cutting quality in machining. Previous experimental works [6-8, 1822] indicated that the UVC method improves cutting quality and saves tool life at low cutting speed values. Therefore, low values of cutting speed are suggested to be used for the UVC technique.

3. Experimental set up and procedure

All the cutting tests were conducted with a modern CNC lathe machine Okuma LH35-N. One end of the workpiece was held tightly in the three-jaw chuck and the other end was supported by the lathe centre. A Sonic impulse SB-150 device containing a PZT [see Fig. 6] was used to vibrate the tool tip in the tangential direction. The available power source supplied for the device was AC 100V with 50 ~ 60 Hz frequency that consumes 260 VA of electric power. Finally, a fresh CBN tool insert (600 type) was mounted on the cutter head in each test.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 6. Photograph of a tool holder containing a PZT and tool insert in the UVC method.

Table 1 presents the experimental conditions used for both the CT and the UVC methods where Table 2 and Table 3 show physical and mechanical properties of both the workpiece Inconel 718 and the CBN tool materials, respectively. When switching off the generator of the vibration device, the cutting becomes CT. Upon switching on, the device provides the frequency of about 19 kHz and the amplitude of about 15 m. Therefore, the maximum vibrating speed of the tool tip can be calculated as
(vt ) max = 2af

= 107.4 m/min.

In order to maintain separating type vibration cutting, the cutting speeds in all the operations were chosen to be less than this critical speed.

Table 1: Experimental conditions

Table 2: Properties (at RT) of workpiece Inconel 718 Table 3: Properties of CBN tool inserts used

In regular intervals (about 2 minutes) of one-pass, machining was stopped in order to observe and measure output parameters such as tool flank wear width, chip formation, and surface roughness. A KISTLER 3-Component Tool Dynamometer was used to measure the cutting force components for tangential, radial and axial directions. For analysis, the force signals were measured via a Graphtec chart recorder. Also, the width of the tool wear along the flank, VB was measured with a Toolmakers Microscope whilst the topography of the tool wear and the formation of chips were examined with a Scanning Electron Microscope (SEM). Moreover, a surface analyzer, Surtronic 10, was used to measure the average surface roughness, Ra , throughout the experiments.

c A

e c

d te p

n a

s u

ri c

t p

4. Results and discussions

4.1.The effect of cutting speed on cutting force and on tool wear

Fig. 7(a) shows the effect of cutting speeds on the cutting force components for both the cutting techniques at a feed rate of 0.1 mm/rev. It is observed that all the force components for the UVC method are reduced up to 15-25% that is required for the CT method at all the cutting speeds. Another finding is that, for both the cutting methods, the thrust force component is the highest among all the components followed by the tangential component and then the axial component.

Fig. 7. Effect of cutting speed in both the cutting methods: (a) Cutting force components, (b) Tool flank wear width, VB after 10 minutes of cutting.

Fig. 7(b) plots the tool flank wear width, VB against the cutting speeds after 10 minutes of cutting and Fig. 8 illustrates the CBN tool wear characteristics with the following SEM photographs for various cutting conditions by applying both the cutting techniques. Though highest tool wear rate was seen, the cutting operation was continued throughout all the cutting speeds in the CT method. In contrast, the wear rate in the UVC method was negligible at cutting speed up to 10 m/min. However, beyond this speed such as 15 m/min and 20 m/min, the tool nose as well as the cutting edge experienced high wear rate and eventually failed after 4 minutes of machining. Since the tools were worn out within short time at 15 m/min and 20 m/min by applying the UVC method, the wear values for those two cutting speeds are not shown in Fig. 7(b). Figs. 8 (a)-(f) were captured after 10 minutes of machining while Figs. 8 (g)-(h) were taken after 4 minutes of operation.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 8. The SEM photographs of tool wear characteristics at different cutting conditions. (a)-(d): CT method; (e)-(h): UVC method.

Fig. 8 also shows that the CT method continuously generated BUE (Built-Up-Edge) that left the pits and debris on the tool rake and flank faces. These pits and debris sticking on the tool cutting area always increase the cutting force and hence increase the tool wear. In contrast, a very small amount of BUE and pits was produced in the UVC method. Therefore, Figs. 7-8 reveal that the removal of BUE, the separating cutting characteristic, the consequent reduction of the surface tearing during UVC [19], aerodynamic lubrication [1, 19] and the generation of comparatively thinner and even chips [Fig. 11] could be the main reasons of producing both the lowest cutting force and the tool wear at less than 15 m/min in the UVC method.

However, in the UVC method, a sudden increase of force components was observed when the cutting speed shifts from 10 m/min to 15 m/min. This is because both the TWCR and the TWRS increase with the increase of the cutting speeds as discussed with Figs. 5 (a)-(b). Ignoring the TWRS factor in this study, the TWCRs for 10 m/min and 15 m/min are found to be about 0.1739 and 0.2163, respectively. Hence the tool-workpiece contact time is reduced by more than 4% in each vibration cycle if the cutting speed reduces from 15 m/min to 10 m/min. That means the tool engaged with the workpiece for a relatively long duration at high cutting speed and also got attacked from a number of consecutive high mutual radial and tangential impacts during the upward motion of the tool. As Inconel 718 has excellent toughness, ultimate transverse strength, yield strength and creep-rupture resistance (see Table 2), and a relatively high tool relief angle (110) was used in the tests, it

c A

e c

d te p

n a

s u

ri c

t p

is assumed that the CBN tool material, which is next to diamond in hardness, could not sustain these high impacts for a long duration in this method. It is seen in Fig. 10(b) that the CBN tool started to wear off at the beginning of cutting at a comparatively high cutting speed of 12.5 m/min or beyond. Though the failure is insignificant at a cutting speed of 10 m/min (see Figs. 8(e)-(f)) even after 10 minutes of cutting, it is seen in Figs. 8(g)-(h) that the failure firstly started with fracture (or abrasion) at the tool nose-flank area and it increased with the increase in cutting speed. Because of worsening the tool condition, the measured cutting force components were found to be higher immediately when the cutting speeds shift toward 12.5 m/min. Due to the long duration of tool-workpiece interaction at these speeds (i.e. high TWCR) as compared to 10 m/min, the high number of impacts in vibration cutting led to fatigue failure of the tool just after 4 minutes of cutting at either 15 m/min or at 20 m/min. The tool failed along with the cutting edge, starting from the tool nose area as seen in the Fig. 8. Since the failure was very severe, it was considered to be a catastrophic failure. Therefore, the UVC method results in long tool life at low cutting speed.

4.2 The effect of feed rate on cutting force and on tool wear

Fig. 9(a) reveals that the cutting force increases with the feed rate in both the processes, which means that the UVC method accords with the same rule in the CT method. However, all the force components, at almost all the feed rates, of the UVC process are reduced approximately to about 12-20% of that of the CT process. Moreover, the cutting force increase rate with the feed rate is insignificant in the UVC process unlike in the CT process.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 9. Effect of feed rate in both the cutting methods at a cutting speed of 10 m/min: (a) Cutting force components, (b) Tool flank wear width VB after 10 minutes of cutting.

Fig. 9(b) shows that the tool flank wear width in the CT method increases suddenly when the feed rate is increased from 0.025 mm/rev to 0.05 mm/rev and then it maintains a steady increase rate. In contrast, in the UVC method, the wear increases linearly with a very small slope throughout the whole range of the feed rates. Thus it is clear that the tool wear rate in the CT method is significantly higher than in the UVC method.

Figs. 9(a)-(b) also demonstrate that the lower cutting force components in the UVC method reduced the tool wear rate that lengthens the tool life. At all the feed rates, it is observed that the tool wear in the UVC method is reduced to about 12-14 % of that in the CT method. According to these results, it is predicted that the tool life for the UVC method is almost 7-8 times higher than that for the CT method.

4.3 Tool wear vs cutting time

Figs. 10 (a)-(b) illustrate that the tool wear rate in the UVC method is significantly lower than in the CT method up to the cutting speed of 10 m/min. According to the following experimental results, it is observed that the tool life in UVC of Inconel 718 is almost 4-8 times higher than that in CT up to that cutting speed limit. This is because both the TWCR and the TWRS are low at low cutting speeds in the UVC method. Thus a small value of both the TWCR and the TWRS results in low tool wear rate.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 10. Tool flank wear width, VB against cutting time at a feed rate of 0.1 mm/rev for (a) CT method and (b) the UVC method.

However, as discussed earlier, due to the high value of both the TWCR and the TWRS beyond the cutting speed of 10 m/min, the tools eventually failed just after 4 minutes of cutting in the UVC method. On the other hand, though 100% TWCR causes a rise in temperature and different wear mechanisms at the tool-workpiece contact areas and elastic and plastic deformations occur to cause fast tool wear; cutting still could be continued in the CT method at high speed because no impact is faced by the tool edge in this method. The above studies thus reveal that the CT method limits the repeatability of the tool which in turn increases the production costs. On the other hand, the UVC method limits the use of high cutting speeds with hard and tough cutting like Inconel 718.

4.4 Analysis of chip formation

Fig. 11 shows the SEM photographs of chips at different feed rates for both the cutting methods. It can be easily observed that the CT method generated thick, uneven, short and cracked chips whereas the UVC method produced comparatively thin, smooth and long chips.

Fig. 11. SEM photographs of the chips produced at different cutting conditions by both the cutting methods: (a)-(b) CT method, (c)-(d) UVC method.

c A

e c

d te p

n a

s u

ri c

t p

The generation of thicker, uneven and cracked chips is always unfavorable to achieve a high quality machining because these types of chips negatively affect the tool cutting area to have non-uniform friction and generation of high temperatures, high cutting forces, high regenerative chatter, and rapid tool wear which shorten the tool life. In contrast, non-

continuous interaction between the tool and the workpiece in the UVC method generated thin, smooth and long chips that did not affect the tool life significantly. 4.5 The effect of cutting speed and feed rate on surface roughness Fig. 12(a) shows the roughness values Ra against the cutting speeds that were taken after 10 minutes of machining with both the cutting methods. Since the tools were worn out in the UVC method after 4 minutes of cutting at 15 m/min and 20 m/min, the Ra values for those conditions were not considered. It is observed that the Ra values increase with the increase in cutting speeds where the increase rate for the UVC method is very insignificant, unlike that for the CT method.

Fig. 12. Average surface roughness values, Ra in both the cutting methods after 10 minutes of cutting: (a) against cutting speeds at a feed rate of 0.1 mm/rev, (b) against feed rates at a cutting speed of 10 m/min.

It is also seen from Fig. 12(b) that a minimum surface roughness Ra value in the CT method was 2.4 m for the cutting of Inconel 718, whereas it was 0.6 m in the UVC method. Furthermore, the Ra values with the UVC technique did not cross 0.8 m at the maximum feed rate of 0.1 mm/rev. Thus the Ra values in the UVC process do not increase markedly with the feed rates, as it does in the CT process.

Figs. 12(a)-(b) also reveal that the surface finish with the UVC method is improved by about 75-85% over the CT method. Therefore, a high quality surface finish for tough cutting could be achieved with the UVC method.

c A

e c

d te p

n a

s u

ri c

t p

The above studies demonstrate that, since the TWCR is 100% in the CT method, the generation of thick, uneven and severe cracked chips, BUE, high cutting force components,

frictional heat, and high cutting instability, etc. deteriorated the machined surface and finally produced a rough and coarse surface. On the contrary, only a fraction of TWCR and consequent reduction of the surface tearing in the UVC method reduced the cutting force, frictional force and frictional heat and produced comparatively sharper fine chips that have less influence on the machined surface. Thus the surface finish obtained in the UVC method is regular and smooth, which is much better than that in the CT method. It is also observed that both the TWCR and the TWRS in the UVC method influence the cutting quality for tough cutting superalloy Inconel 718. For example, since the TWCR and the TWRS are the lowest for 5 m/min as compared to other cutting speeds (see section 2 for the TWCR values), the surface finish is better for this cutting speed. Thus, as low as the values of these key parameters, the cutting quality distinctly improves.

4.6 Comparative analysis between CT and UVC method

Lastly, Fig. 13 presents a brief comparison between the CT and the UVC methods based on the above experimental findings. Four different output parameters; the tangential and radial cutting force components, the flank wear width and the Ra values at a feed rate of 0.1 mm/rev were taken into account for this comparison. Since the UVC method performs remarkably better up to a cutting speed of 10 m/min in cutting of Inconel 718, this speed was selected as best suited for this analysis.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 13. Comparative analysis of cutting performances between the CT and the UVC methods at selected cutting speed of 10 m/min.

The above chart shows that the UVC method, in all the cases, promises better cutting performance than the CT process. Therefore, it is concluded that the UVC method does not

only attain high quality cutting of difficult-to-machine materials, but it also raises the tool life distinctly and saves machining cost. 5. Conclusions

The effect of tool vibration frequency, tool vibration amplitude and workpiece cutting speed in the UVC method were studied theoretically. Also the effect of cutting parameters on a superalloy Inconel 718 with CBN tools was investigated thoroughly by applying the UVC methods. In order to compose a comparative analysis, the cutting conditions applied in the UVC method were also considered for the CT method. The cutting force components, the tool flank wear width, the chip formation and the surface roughness were justified as the output parameters. Based on the theoretical studies and experimental results achieved, the following conclusions can be compiled:

1. In the UVC technique, the cutting quality depends mainly on two important factors: TWCR and TWRS. The cutting mechanism shows that the TWCR relies on three independent key parameters; the tool vibration frequency, the tool vibration amplitude and the workpiece cutting speed.

2. To achieve high quality cutting, the TWCR should be kept as low as possible. The value of TWCR can be lowered by increasing both the tool vibration frequency and amplitude, as well as by decreasing the workpiece cutting speed. 3. The test results show that the cutting force for the UVC method was required to about 12-25 % of that for the CT method in cutting of Inconel 718. 4. The tool flank wear in the UVC method was found to be about 12-25 % of that in CT method when cutting up to 10 m/min. Accordingly, the tool life with the UVC method is increased by at least 4-8 times over the CT method.

c A

e c

d te p

n a

s u

ri c

t p

5. A minimum Ra value of 0.6 m was achieved with the UVC method whereas 2.4 m was achieved with the CT method for the same cutting condition. Hence, the cutting quality with the UVC method was improved by about 75-85% over the CT method. 6. However, beyond the cutting speed 10 m/min, the CBN tools catastrophically failed after 4 minutes of machining applying the UVC method. This type of failure may be due to high TWCR when a high cutting speed was used. A number of consecutive high impacts between the tool and the workpiece with high TWCR induced to cause fast tool wear that agrees with the theoretical study on the effect of the third factor; the effect of workpiece cutting speed.

7. To conclude, the UVC method has been found to be a suitable technique to achieve high quality finish surfaces for Inconel 718; though low cutting speed range should be maintained in this method.

c A

e c

d te p

n a

s u

ri c

t p

References

1. M. Zhou, X. J. Wang, B. K. A. Ngoi, J. G. K. Gan, Brittle-ductile transition in the diamond cutting of glasses with the aid of ultrasonic vibration, J. Mater. Process. Technol. 121 (2002) 243-251. 2. V.I. Babitsky, A.V. Mitrofanov, V.V. Silberschmidt, Ultrasonically assisted turning of aviation materials: simulations and experimental study, Ultrasonics 42 (2004) 81-86. 3. M. Xiao, K. Sato, S. Karube, T. Soutome, The effect of tool nose radius in ultrasonic vibration cutting of hard metal, Int. J. Mach. Tools Manuf. 43 (2003) 1375-1382. 4. G.F. Gao, B. Zhao, F. Jiao, C.S. Liu, Research on the influence of the cutting conditions on the surface microstructure of ultra-thin wall parts in ultrasonic vibration cutting, J. Mater. Process. Technol. 129 (2002) 66-70.

5. A.V. Mitrofanov et al, Finite element simulations of ultrasonically assisted turning, Computational Materials Science 32 (2005) 463-471. 6. V.I. Babitsky, A.N. Kalashnikov, A. Meadows, A.A.H.P Wijesundara, Ultrasonically assisted turning of aviation materials, J. Mater. Process. Technol. 132 (2003) 157-167. 7. M. Jin, M. Murakawa, Development of a practical ultrasonic vibration cutting tool system, J. Mater. Process. Technol. 113 (2001) 342-347. 8. J.-D. Kim, I.-H. Choi, Micro surface phenomenon of ductile cutting in the ultrasonic vibration cutting of optical plastics, J. Mater. Process. Technol. 68 (1997) 89-981. 9. C. Nath, M. Rahman, S.S.K. Andrew, A study on ultrasonic vibration cutting of low alloy steel, J. Mater. Process. Technol. 192-193 (2007) 159-165.

c A

e c

d te p

n a

s u

ri c

t p

10. L. Balamuth, Ultrasonic assistance to conventional metal removal, Ultrasonics, July 1966, 125-130. 11. C.S. Liu, B. Zhao, G.F. Gao, F. Jiao, Research on the characteristics of the cutting force in the vibration cutting of a particle-reinforced metal matrix composites SiCp/Al, J. Mater. Process. Technol. 129 (2002) 196-199. 12. J. Kumabe et al, Ultrasonic superposition vibration cutting of ceramics, Precision Engineering, 0141-6359/89/020071-07. 13. M. Xiao, S. Karube, T. Soutome, K. Sato, Analysis of chatter suppression in vibration cutting, Int. J. Mach. Tools Manuf. 42 (2002) 1677-1685.

14. J. Kumabe, M. Hachisuka, Super-precision cylindrical machining, Precision Engineering, 0141-6359/84/020067-06.

15. M. Xiao, Q. M. Wang, K. Sato, S. Karube, T. Soutome, H. Xu, The effect of tool geometry on regenerative stability in ultrasonic vibration cutting, Int. J. Mach. Tools Manuf. xx (2005) 1-8.

16. A.V. Mitrofanov, V.I. Babitsky, V.V. Silberschmidt, Finite element analysis of ultrasonically assisted turning of Inconel 718, J. Mater. Process. Technol. 153-154 (2004) 233-239.

17. A.V. Mitrofanov, N. Ahmed, V.I. Babitsky, V.V. Silberschmidt, Effect of lubrication and cutting parameters on ultrasonically assisted turning of Inconel 718, J. Mater. Process. Technol. 162-163 (2005) 649-654. 18. C. Ma, E. Shamoto, T. Moriwaki, Y. Zhang, L. Wang, 2005, Suppression of burrs in turning with ultrasonic elliptical vibration cutting, Int. J. Mach. Tools Manuf. 45 (2005) 1295-1300. 19. R.C. Skelton, Effect of ultrasonic vibration on the turning process, Int. J. Mach. Tool Des. Res. Vol. 9, pp. 363-374, Pergamon Press 1969, Printed in Great Britain.

c A

e c

d te p

n a

s u

ri c

t p

20. J.-D. Kim, E.-S. Lee, A study of the ultrasonic-vibration cutting of carbon-fiber reinforced plastics, J. Mater. Process. Technol. 43 (1994) 259-277. 21. A.V. Mitrofanov et al, Finite element simulations of ultrasonically assisted turning, Computational Materials Science 28 (2003) 645-653. 22. J.-D. Kim, I.-H. Choi, Characteristics of chip generation by ultrasonic vibration cutting with extremely low cutting velocity, The International Journal of Advanced Manufacturing Technology, Volume 14, Number 1 / January, 1998. 23. Y.-L. Zhang, Z.-M. Zhou, Z.-H. Xia, Diamond turning of titanium alloy by applying ultrasonic vibration. Transactions of the Nonferrous Metals Society of China. Vol. 15, no. Special 3, pp. 279-282. Nov. 2005.

c A

e c

d te p

n a

s u

ri c

t p

Caption of tables: Table 1: Experimental conditions Table 2: Properties (at RT) of workpiece Inconel 718 Table 3: Properties of CBN tool inserts used

c A

e c

d te p

n a

s u

ri c

t p

Table 1: Experimental conditions Workpiece Material Diameter Length Tool Material Rake angle Relief angle Approach angle Nose radius Cutting conditions Depth of cut Feed rate Cutting speed Vibration conditions Frequency Amplitude Inconel 718 175 mm 600 mm CBN (BN250) +100 110 300 0.4 mm 0.10 mm

0.025 - 0.1 mm/rev 5 - 20 m/min

19 1.5 kHz 15 m

c A

e c

d te p

n a

s u

ri c

t p

Table 2: Properties (at RT) of workpiece Inconel 718

Density Melting temp range Avg. thermal exp. coeff. Specific heat Thermal conductivity Ultimate tensile strength Yield strength (0.2% off.) Elongation in 50 mm Elastic modulus (tension) Hardness

8.19 g/cm3 1260-1336 0C 13.0 m/m.K 435 J/kg.K 11.4 W/m.K 1240 MPa 1036 MPa 12% 211 GPa 36 HRC

c A

e c

d te p

n a

s u

ri c

t p

Table 3: Properties of CBN tool inserts used

CBN contents CBN grain size Binder Poissions ratio Thermal conductivity Thermal stability Hardness (GPa)

8590 (vol.%) 35 (m) Co etc. 0.22 100130 (W/m K) 1270 (K in air)

3540 (at room temp.) 12 (at 1273 K)

c A

e c

d te p

n a

s u

ri c

t p

Captions of figures:

Fig. 1. Schematic of ultrasonic vibration cutting.

Fig. 2. Pulse cutting state in the UVC method.

Fig. 3. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration frequencies (Subscripts: 1, 2 are for 20 kHz, 35 kHz, respectively), (b) Relation between TWCR and tool vibration frequency, f.

Fig. 4. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration amplitudes (Subscripts: 1, 2 are for 10 m, 25 m, respectively), (b) Relation between TWCR and tool vibration amplitude, a.

Fig. 5. UVC process: (a) Tool displacement and pulsating cutting force against time at f = 20 kHz and a = 15 m. (Subscripts: 1, 2 are for low and high cutting speed, respectively), (b) Relation between TWCR and workpiece cutting speed, vc .

Fig. 6. Photograph of a tool holder containing a PZT and tool insert in the UVC method.

Fig. 7. Effect of cutting speed in both the cutting methods: (a) Cutting force components, (b) Tool flank wear width, VB after 10 minutes of cutting.

c A

e c

d te p

n a

s u

ri c

t p

Fig. 8. The SEM photographs of tool wear characteristics at different cutting conditions. (a)-(d): CT method; (e)-(h): UVC method.

Fig. 9. Effect of feed rate in both the cutting methods at a cutting speed of 10 m/min: (a) Cutting force components, (b) Tool flank wear width VB after 10 minutes of cutting.

Fig. 10. Tool flank wear width, VB against cutting time at a feed rate of 0.1 mm/rev for (a) CT method and (b) the UVC method.

Fig. 11. SEM photographs of the chips produced at different cutting conditions by both the cutting methods: (a)-(b) CT method, (c)-(d) UVC method.

Fig. 12. Average surface roughness values, Ra in both the cutting methods after 10 minutes of cutting: (a) against cutting speeds at a feed rate of 0.1 mm/rev, (b) against feed rates at a cutting speed of 10 m/min.

Fig. 13. Comparative analysis of cutting performances between the CT and the UVC methods at selected cutting speed of 10 m/min.

c A

e c

d te p

n a

s u

ri c

t p

Figures:

Fig. 1. Schematic of ultrasonic vibration cutting

c A

e c

d te p

n a

s u

ri c

t p

x = a sin t
a
ta tc tb ta tc

Fig. 2. Pulse cutting state in the UVC method [15].

c A

e c

d te p

n a

s u

ri c

t p

1 0.8 0.6

R C W T

0.4 0.2 0 0

20, 0.2162

10

Tool frequency, f (kHz)

(a)

Fig. 3. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration frequencies (Subscripts: 1, 2 are for 20 kHz, 35 kHz, respectively), (b) Relation between TWCR and tool vibration frequency, f.

c A

e c

d te p

n a

s u
15

20

ri c
25

t p
30

35, 0.1600

35

40

(b)

1 0.8 0.6

R C W 0.4 T
0.2 0 0 5

10, 0.2710 25, 0.1644

Vibration amplitude, a (micron)

10

15

20

25

30

35

40

(a)

Fig. 4. UVC process: (a) Tool displacement and resultant cutting force for two different tool vibration amplitudes (Subscripts: 1, 2 are for 10 m, 25 m, respectively), (b) Relation between TWCR and tool vibration amplitude, a.

c A

e c

d te p

n a

s u

ri c
(b)

t p

1 0.8 0.6

R C W 0.4 T
0.2 0 0

20, 0.2536 40, 0.3803

20

Cutting speed, v (m/min)

40

60

80

100

120

(a)

(b)

Fig. 5. UVC process: (a) Tool displacement and pulsating cutting force against time at f = 20 kHz and a = 15 m. (Subscripts: 1, 2 are for low and high cutting speed, respectively), (b) Relation between TWCR and workpiece cutting speed, vc .

c A

e c

d te p

n a

s u

ri c

t p

PZT PZT holder Tool insert

Fig. 6. Photograph of a tool holder containing a PZT and tool insert in the UVC method.

c A

e c

d te p

n a

s u

ri c

t p

80 70 60 50 40 30 20 10 0 2.5 5 7.5 10 12.5 15 17.5 20 Cutting speed, m/min

Fr in CT Ft in CT Fa in CT Fr in UVC Ft in UVC Fa in UVC

Cutting forces, N

(a)
0.5
Flank wear width, mm

0.4 0.3 0.2 0.1 0

CT

UVC

c A

e c
5

d te p
(b)

n a

s u

ri c

t p

7.5 10 12.5 15 Cutting speeds, m/min

20

Fig. 7. Effect of cutting speed in both the cutting methods: (a) Cutting force components, (b) Tool flank wear width, VB after 10 minutes of cutting.

CT Method

UVC Method

Failure starts at tool nose

(a) 10 m/min, 0.05 mm/rev, CT

(e) 10 m/min, 0.05 mm/rev, UVC

No failure at nose-flank

(b) 10 m/min, 0.1 mm/rev, CT

(f) 10 m/min, 0.1 mm/rev, UVC

c A

e c

(c) 15 m/min, 0.05 mm/rev, CT

d te p

Fracture area

n a

s u

ri c

t p

(g) 15 m/min, 0.05 mm/rev, UVC Catastrophic failure

Fracture

(d) 15 m/min, 0.1 mm/rev, CT

(h) 15 m/min, 0.1 mm/rev, UVC

Fig. 8. The SEM photographs of tool wear characteristics at different cutting conditions. (a)-(d): CT method; (e)-(h): UVC method.

70 60 50 40 30 20 10 0 0 0.025 0.05 0.075 0.1 Feed rates, mm/rev

Fr in CT Fr in UVC Ft in CT Ft in UVC Fa in CT Fa in UVC

Cutting forces, N

(a)

Flank wear width, mm

0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0

CT UVC

Fig. 9. Effect of feed rate in both the cutting methods at a cutting speed of 10 m/min: (a) Cutting force components, (b) Tool flank wear width, VB after 10 minutes of cutting.

c A

e c

d te p
0.025

m
(b)

n a

s u
0.1

ri c

t p

0.05 0.075 Feed rate, mm/rev

0.5
Tool wear, mm

5 m/min 7.5 m/min 10 m/min 12.5 m/min 15 m/min 20 m/min 0 2 4 6 8 10 Cutting time, min

0.4 0.3 0.2 0.1 0

(a)
1
Tool wear, mm

5 m/min 7.5 m/min 10 m/min 12.5 m/min 15 m/min 20 m/min 0 2 4 6 8 10 Cutting time, min

0.8 0.6 0.4 0.2 0

Fig. 10. Tool flank wear width, VB against cutting time at a feed rate of 0.1 mm/rev for (a) CT method and (b) the UVC method.

c A

e c

d te p
(b)

n a

s u

ri c

t p

(a) 10m/min, 0.05 mm/rev, CT

(c) 10 m/min, 0.05 mm/rev, UVC

(b) 10 m/min, 0.1 mm/rev, CT

(d) 10 m/min, 0.1 mm/rev, UVC

Fig. 11. SEM photographs of the chips produced at different cutting conditions by both the cutting methods: (a)-(b) CT method, (c)-(d) UVC method.

c A

e c

d te p

n a

s u

ri c

t p

Surface roughness, micron

6.0 5.0 4.0 3.0 2.0 1.0 0.0 5

CT UVC

7.5 10 12.5 15 Cutting speeds, m/min

20

(a)

Surface roughness, micron

6 5 4 3 2 1 0 0.025 CT UVC

Fig. 12. Average surface roughness values, Ra in both the cutting methods after 10 minutes of cutting: (a) against cutting speeds at a feed rate of 0.1 mm/rev, (b) against feed rates at a cutting speed of 10 m/min.

c A

e c

d te p

0.05 0.075 Feed rate, mm/rev

n a
(b)

s u

ri c

t p

0.1

70 60 50 40 30 20 10 0
1 2

CT UVC

1) T. force, N; 2) R. force, N; 3) T. wear, mm X 100; and 3) Roughness Ra, micron X 10

Fig. 13. Comparative analysis of cutting performances between the CT and the UVC methods at selected cutting speed of 10 m/min.

c A

e c

d te p

n a

s u

ri c

t p