International Journal of Machine Tools & Manufacture

Transcription

1 International Journal of Machine Tools & Manufacture 58 (2012) Contents lists available at SciVerse ScienceDirect International Journal of Machine Tools & Manufacture journal homepage: An innovative approach to monitor the chip formation effect on tool state using acoustic emission in turning M.S.H. Bhuiyan n, I.A. Choudhury, Y. Nukman Department of Engineering Design and Manufacture, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia article info Article history: Received 26 August 2011 Received in revised form 3 February 2012 Accepted 3 February 2012 Available online 8 March 2012 Keywords: Chip formation Chip formation effect Tool state investigation Tool wear Acoustic emission abstract Chip formation in metal cutting is inevitable and has a remarkable effect on tool state and therefore on the tool life. The work presented here introduces a new technique to independently monitor the chip formation effect on the tool state. This has been done by separating the chip formation events from the rest of the frequencies of occurrences. A customized tool holder and sensor setup are designed and integrated with the conventional tool holder to capture the signals from chip formation independently during turning. The signals taken by acoustic emission (AE) sensor represent the effect of chip formation on the tool state. The frequencies remaining below the transient offset signal are mostly coming from the tool wear and plastic deformation of the workmaterial. It has been observed that the acoustic emission is more susceptible to entire occurrences in turning. The time domain signal and corresponding frequency response can predict the tool state effectively. From raw AE signals and their RMS values, the tool wear and plastic deformation are observed to increase with the increase of cutting speed, feed rate and depth of cut. However, the tool wear has been found to decrease with chip breakage even at higher cutting speed and feed rate, and this has been verified by measuring the tool wear. The chip formation frequency has been found to vary between 68.3 khz and khz while the maximum intensity was observed at 97.7 khz. & 2012 Elsevier Ltd. All rights reserved. 1. Introduction A machining process includes a number of occurrences like tool wear, tool chipping, tool breakage, tool failure, chip formation, chip breakage, process interruption and some other irregularities and so on. The entire phenomena have their particular effect on the process and also on the tool state. The different occurrences during metal cutting affect the cutting tool condition and disturb the process stability. In conventional turning, the chip formation is the only way to remove material from the workpiece to shape the workmaterial into a desired dimension. The chip formation has a significant influence on the cutting tool condition depending on the formation mechanism and its geometry. The mechanism of chip formation, the types of chips, the separation and rate of removal of chips, the energy content and temperature of chips and the rubbing action of chips with the tool face determine the tool wear. Dolinsek and Kopac [1] have pointed out that the heat generation and the chip formation force during metal cutting dominate the tool wear. The chip formation not only produces the tool wear, in some cases it causes breakdown of n Corresponding author. Tel.: þ ; fax: þ addresses: sayem_um@yahoo.com (M.S.H. Bhuiyan), imtiaz@um.edu.my (I.A. Choudhury), nukman@um.edu.my (Y. Nukman). the cutting tool. Tool wear increases power consumption and tool breakdown interrupts the operation and affects the product s quality. To avoid the difficulties and uncertainties associated with the chip formation in machining, effective monitoring system is necessary. 2. Chip formation in metal cutting and monitoring 2.1. Chip formation in metal cutting The chip formation is a certain happening in metal cutting to remove material from the workpiece. Therefore, its effect on the cutting tool is unavoidable. This is associated with a complicated interaction of plastic and elastic deformation of the workmaterial within a small region known as shear zone. Jared and Dow [2] have pointed out that the interaction in shear zone ultimately defines both the geometry and motion of the generated chips in metal cutting. However, the cutting speed affects the shape and dimensions of the chip formation zone, and therefore controls the chip formation types and also the chip geometry. Kim and Kweun [3] observed that as the cutting speed increases, chips become thinner. The reason has been explained by Tsai [4] as in high cutting speed the controlled chip region decreases and consequently the chip thickness is reduced. Depending on the cutting /$ - see front matter & 2012 Elsevier Ltd. All rights reserved. doi: /j.ijmachtools

2 20 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) conditions, both the steady state and the cyclic types of chips are generated in turning. It is reported from Astakhov [5] that the steady state type chips generate from the blunt tool-tip and are continuous in nature. In continuous chip formation, Dolinsek and Kopac [1] have found that at slow cutting speeds, adhesion and abrasion are the main tool wear mechanisms whereas abrasion and chemical wear are most likely at high cutting speeds. The cyclic types of chips are discontinuous, wavy and saw toothed in shape and are caused by built-up-edge (BUE). Kishawy and Wilcox [6] have pointed out that the saw-toothed chip formation is caused from the periodic cracking, and not from the adiabatic shear. On the other hand, the adiabatic shear is involved with the micro-crack region and it has a specific effect on the chip flow. Shaw and Vyas [7] observed that, with the thinning of microcrack region, the chip moves up the tool face and changes the types of tool wear. Davies et al. [8] have observed that among the entire chip types, the segmented chip is typically produced from conventional turning and its effect on the cutting tool is the most. Besides, Chen et al. [9] have empirically established that the segmented chip formation force is less whereas the generated heat is more than the continuous chip formation. From the observation of Balaji and Ghosh [10], for a particular tool geometry, the cutting speed typically controls the onset of strain localization and thus the segmented chip formation. Depending on the cutting conditions and tool geometry, the generated chips flow in different direction to dispose. According to Jared and Dow [11], the chip flow direction or the flow angle only defines its initial direction of motion however, that might be changed by the external force in later stage. With the change of chip flow angle the stress concentration in chip as well as its geometry are reportedly altered. According to Shamoto and Aoki [12], the chip flow in a grooved tool basically consists of side-flow and backflow, whereas the chip curl consists of up-curl and side-curl. During the chip removal, the friction occurs at the sticking and the sliding region of the tool chip interface. According to Kilic and Raman [13], the uniform shear stress in the sticking region decreases with a power law in the sliding zone. The cutting tool is thereby affected from the sticking and sliding of chips during formation. Balaji and Ghosh [10] have investigated that the sticking regions are the most heat-affected regions due to the intense friction between the underside of the chip and the tool s groove. They have found an increase in cutting force and contact length with the decrease of chip curvature. On the other hand, Arcona and Dow [14] have reported that the change in the cutting force indicates a change in cutting tool geometry and workpiece surface finish. Besides, the change in cutting force indicates different types of chip generation in machining. The various types chip formation and disposal affects the tool in different way and to different extent depending on their types, geometry and flow angle Chip formation monitoring Although, monitoring of metal cutting process is common, independent monitoring of a particular occurrence, i.e., chip formation is hardly reported. In metal cutting, a number of researchers have reported monitoring of the chip formation during tool state investigation. Both the direct and indirect sensing methods have been used to make the monitoring possible. Direct methods consist of laser, optical, and ultra-sonic sensors which provide a direct measurement. These methods are still very expensive and difficult to apply in the machining process environment. In contrast, indirect methods are more economical, and are based on sensors which represent the machining state by sensing cutting forces, vibrations, temperatures, current consumption, etc. Basically, four sensors: dynamometers, accelerometers, AE sensors and current sensors have been widely used to monitor machining systems and the associated occurrences [15]. Gandarias et al. [16] have used laser sensor to investigate the tool breakage in monitoring the cutting tool state in milling. Tool condition was monitored by focusing a laser beam with a spot size of 50 mm constantly on the cutter, and at the same time directing the beam reflection towards a receiver. As the laser beam is required to reflect back from the object to be monitored; therefore application of laser sensor in monitoring chip formation like transient occurrences in a clumsy machining environment is really very difficult, and therefore not being used yet. Barry and Byrne [17] have monitored chip formation with AE sensor in turning. They have investigated the nature of AE signal corresponding to the continuous and saw tooth chip formation at various depths of cut. The observation result revealed that, during continuous chip formation the AE RMS varied between 0.05 V and 0.1 V; and during saw tooth chip formation, AE RMS contained at least one order of magnitude greater. Pawade and Joshi [18] have used AE signal to determine the quality of machined surface in high-speed turning of Inconel 718. The energy, number of counts, and mean frequency amplitude of the AE signals were used to evaluate the dependence of the machining deformation on the surface generation mechanism. From their observation, the AE signal variations is found to be useful to correlate abnormal events during machining, such as higher thrust forces, chip form variation, and surface anomalies produced in machining. Teti et al. [19] have measured different force components to investigate the chip formation types in metal cutting. The measured cutting forces were demonstrated with six different variables (feed force, radial force, cutting force, resultant force and the force ratio between the axial and radial force). The different variables and Partial Least Squares (PLS) method have been used to correlate the signals of three cutting force components with the corresponding chip types. The experiment showed that at low feed rates, the force ratio became more effective than the other variables considered, and at relatively larger feeds the resultant force was more significant. Somkiat [20] have used dynamic cutting force to investigate the chip formation types in turning. From their observation, the dynamic cutting force has its own characteristic pattern in each cutting situation of continuous chip formation, broken chip formation, and the chatter. When the chips were continuous, the dynamic feed force was relatively small in amplitude and the power spectrum density (PSD) was relatively large at low frequency range, typically less than 50 Hz. On the other hand, the dynamic component of the feed force, among the three force components, was relatively large in amplitude when the chips were broken into pieces, and the relatively large PSD appeared in the frequency range, which corresponded to the chip breaking frequency. Tanhjitsitcharoen and Moriwaki [21] have used various cutting force signals to identify the chip forms, especially the favorable and unfavorable chip types. From their observation, the dynamic components of three cutting forces were small in amplitude when the chips were continuous. However, when the chips were broken, the dynamic feed force component becomes larger. Hence x, y, and z components of cutting force were larger in broken chips than in the continuous chips. From the literature, the force dynamometer is an extensively used sensor in chip formation monitoring. Despite, the force dynamometer can effectively senses the continuous chip formation; it has limitation in sensing segmented chip formation. As the frequency of segmented chip formation is greater than 10 khz, it is beyond the response of conventional dynamometers [17].Onthe other hand, the strain energy released during saw tooth chip formation is the dominant source of AE signal. Therefore, chip formation monitoring using AE sensor is more effective.

3 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) Acoustic emission in monitoring the chip formation The acoustic emission (AE) is the transient elastic wave generated by the rapid release of energy from a localized source or sources within a material. The AE sources belong to the state of stress inside the material, and the energy release is associated with the abrupt redistribution of internal stresses. The plastic deformation, phase transformations, vacancy coalescence and decohesion of inclusions and fracture are key sources of acoustic emission. However, according to Ravindra et al. [22], only the plastic deformation and fracture have major significance in metal cutting. From the study of Dornfeld [23], the possible AE sources in metal cutting are; (i) plastic deformation of the work material during the cutting process, (ii) plastic deformation in the chip, (iii) frictional contact between the tool flank face and the workpiece resulting in flank wear, (iv) frictional contact between the tool rake face and the chip resulting in crater wear, (v) collisions between chip and tool, (vi) chip s breakage and (vii) tool fracture. Due to strong dependency of acoustic emission on process variables, it is widely used to keep the tool under close watch during the machining. According to Li [24], the major advantage of using AE to monitor the tool condition is that the frequency range of the AE signal is much higher than that of the machine vibrations and environmental noises and does not interfere with the cutting operation. From the study of Xavier and Sampathkumar [25], the AE sensors of frequency range 30 khz to 2 MHz are mostly used in tool condition monitoring. However, the microphones or accelerometers are used to measure the frequencies below 20 khz or near audio frequencies of acoustic emission. The AE signals from conventional tool setup are very stochastic and squiggle to look. The signal is complex and it contains a range of numerous frequencies from different occurrences in turning. A number of frequencies generated from different sources including the chip formation are incorporated in the same output signal. The raw signals are complex and are required to be processed to extract the significant features from them. The AE parameters are used to correlate the cutting conditions, tool geometry, tool wear, plastic deformation, chip formation etc with the captured signal. Li [26] has found that the tool breakage and chip breakages are easily identifiable by sudden bursts of AE, followed by large changes in AE intensity. On the other hand, the tool wear and plastic deformation are more difficult to detect because of the low intensity and the dependency of AE signals on process parameters. The entire features need to be considered recognizing a change in sensing the tool condition to make the monitoring more effective. Inasaki [27] has used the typical pattern recognition analysis to make it possible and to monitor the tool failure resulting from wear and fracture. Emel and Kannatey-Asibu Jr [28] have used the class mean criteria to choose the best feature of signal before adopting the pattern recognition system. Ravindra et al. [22] have mentioned that the best features are the ones that carry more information regarding the status of the cutting tool. The AE signal can be classified into two types, continuous and burst. According to Chiou and Liang [29], the continuous-type AE signals are associated with plastic deformations of the workmaterial and chips and tool wear during metal cutting. However the burst-type signals are observed during crack growth inside the material. Moriwaki and Okushima [30] have found that, the tool fracture, chip breaking, chip impacts or chip tangling generate a burst-type AE signals. As the chip formation is accomplished with crack growth, chip breakage and chip removal, it is believed that the chip formation produces a transient burst AE signal during metal cutting. The RMS signal shows the energy content of the raw signal whereas the signal frequency is more specific to any particular occurrence. The RMS is effective to sense the tool wear and plastic deformation but not suitable for detecting the tool breakage, chip formation and other burst type signal producing events. Therefore, the combined application of RMS and frequency analysis could identify the occurrences and illustrate the signals more effectively. This paper attempts to isolate the effect of chip formation on tool state by separating the chip formation occurrences from other incidences in turning. Signals of chip formation are captured separately with the help of a dummy tool setup attached to the main tool holder to independently monitor its influence on cutting tool state. The ultimate target of this work is to isolate the signal component according to different occurrences from the AE signal pattern. The possible correlation of types of chip formation with the signal pattern and frequency will be investigated in this paper. 3. Materials and method The conventional tool setup and the way of signal acquisition are not able to serve the purpose of this work. To investigate the chip formation effect on the tool wear, it is necessary to separate the AE signal arising from chip formation only and capture it independently. It is therefore required to modify the conventional tool setup to materialize the objective. The raw signal, its RMS and frequency analysis are used to analyze the output of the sensor from the modified dummy tool setup to make it more usable Modified tool and sensor setup A special tool setup is designed and fabricated to make it possible to separate the transient signal generated from chip formation only. One dummy tool setup that has been replicating the conventional tool setup is designed and integrated into the conventional tool setup. The dummy tool holder and tool-insert arrangement has been designed and fabricated to aid the AE signal to follow about the same path of transmission from sources to the sensor. The dummy tool-insert and tool holder arrangement is mounted over the main tool-insert and tool holder arrangement. The dummy arrangement is set in such a way with respect to the main tool setup that it cannot come in contact with the workpiece while the main tool cuts the material. However, the chips that are released during metal cutting would touch the dummy tool insert as it leaves the workpiece. To avoid mutual vibration between the main and dummy tool holders, rubber insulation is placed in between. The placement of rubber insulation has helped to dampen the low-frequency signal components arising from plastic deformation and tool wear. Besides the AE sensor and the data acquisition system allow the signal above 50 khz to pass to storage. It is expected that the combined effect of rubber insulation and data acquisition system could successfully make the dummy setup signal independent. A piezoelectric AE sensor is placed on the dummy tool holder to capture the acoustic emission generated during cutting. This is placed on the dummy tool holder as close as possible to the spot of collision between chip and dummy tool-insert. The whole setup is shown in Fig. 1 in details. The performance of the dummy arrangement has been tested and responses are satisfactory. The AE signals from the new modified tool-sensor setup carries only the transient signal. Since the chip touches the dummy tool-insert as it forms, breaks and is removed; the signal obtained from the new setup shows the chip formation occurrence only corresponding to the different cutting conditions. As the sensor is placed on the dummy tool holder, it never comes in contact with the main toolholder assembly and the sensor transient AE signal does not include the tool fracture signal. Fig. 1(a) shows schematic of the

4 22 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) Fig. 1. The setup for separating the chip formation occurrences (a) schematic view, (b) real view. Fig. 2. The AE signal measuring chain in metal cutting. modified tool holder setup while Fig. 1(b) shows a real view of the setup during cutting. This approach would be potential to monitor the tool state more effectively The AE signal acquisition technique The procedure of AE signal acquisition in the present study is illustrated in Fig. 2. The piezoelectric AE sensor, type: KISTLER 8152B is placed as close as possible to the tool chip contact zone, i.e. on the dummy tool shank. The AE sensor has a frequency range from 50 khz to 1 MHz. The sensor hold-down force of several Newtons is used to ensure good contact and to minimize the coupling thickness. Because of high impedance of the AE sensor, it must be directly connected to a coupler to make the signal compatible. The coupler type that is used in the experiment is KISTLER-5125B which contains a buffer amplifier and a high-pass (HP) filter of limit 50 khz. The HP filter in the coupler allows the signal above 50 khz to pass to the next module DEWE-43. The DEWE-43 module containing a low-pass filter of limit 1000 khz which cuts off the signal beyond 1000 khz and allows the remaining signal to the storage for further processing. Low frequency noise components, which are inevitably present in AE signal, are considered not to be correlated with the occurrences. Therefore, those components should be eliminated (by high-pass filter) at the earliest possible stage of signal processing to enable usage of full amplitude range of the equipment. However, the very high frequencies components are also cut off (by low-pass filter) to avoid electric sparks or aliasing. Finally, the signal comes out from the DEWE-43 module is amplified and then digitized before storing for further analysis Experimental details The turning operation is performed on a COLCHESTER VS MASTER3250, 165 mm 1270 mm Gap bed Center Lathe machine. The work-piece used is a round bar (diameter 92 mm length Table 1 Cutting conditions. Cutting speed, v (m/min) : 130, 150, 175, 190 Feed rate, f (mm/rev) : 0.28, 0.32, 0.50 Depth of cut, a p (mm) : 1,2 760 mm) of ASSAB-705, a medium carbon steel of hardness HB It contains carbon (0.35%), chromium (1.40%), iron (95.95%), manganese (0.70%), molybdenum (0.20%), and nickel (1.40%) by weight. The TiN coated carbide, type: TNMG160408N-GU tool insert and PTGNR 2020K-16 tool holder assembly are used as main tool arrangement. For the dummy arrangement of tool, the mild steel tool holder is used whereas the tool-insert is the same. The cutting conditions, tooling material and workpiece combination have considerable effect on the tool wear. Continuous dry cut is conducted in the experimental investigation. A set of experimental runs have been conducted at different cutting conditions. The tool life has been measured every 2 minutes for a period of 16 minutes at each cutting condition. The acoustic emission signals have been captured at every cut and the signals are correlated with corresponding tool wear and chip formation. The experimental conditions are given in Table Result and discussion The AE performs exceptionally well in tool conditions monitoring in machining. It can extract more information about the occurrences thus aiding to assess the tool state more accurately. Even though, the raw signal is stochastic, squiggle and difficult to interpret, it becomes more useful as it is analyzed. The RMS of AE represents the energy content of the raw signal which is an indication of intensity of the occurrences. Both the raw and RMS signal of AE are capable of giving significant information about the occurrences. The frequency study could provide more specific information about particular

5 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) occurrences. The raw AE signals are the resultant illustration of all occurrences that take place during turning. In normal turning some major occurrences dominate the signal patterns. The tool wear, the plastic deformation of workmaterial and the chip formation are the key source of the AE signal. For the conventional setup of AE sensor, the continuous low amplitude pattern of raw signals represents the tool wear and plastic deformation whereas the transient burst pattern of signal typically depict the chip formation occurrences. For tool breakage the burst signals amplitude will be higher and momentarily the signal is lost for a short period of time. Both types of signal coexist in the raw AE signal. Because of its random and complex shape, it is quite difficult to exactly measure the pattern of low amplitude continuous signals until separation is possible. Fig. 3(a) represents a raw AE signal captured at cutting speed of 130 m/min, feed rate of 0.28 mm/rev and depth of cut of 1 mm during shut down of machine. Fig. 3(b) and (c) represents a raw AE signal and its RMS signal respectively, captured at cutting speed of 190 m/min, feed rate of 0.32 mm/rev and depth of cut of 2 mm from the conventional tool setup in turning. From Fig. 3(b), the continuous and transient patterns of the AE signals are more apparent while the instant of tool breakage is also spotted clearly on the AE signal. From Fig. 3(c), it is apparent that the energy content of the RMS signals fluctuates all along the operation. From Fig. 3(a) and (b), the difference between the signals arising from tool breakage and machine shut down is very obvious. In case of tool breakage, the amplitude of RMS AE signal AE signal during shutting down the machine Continuous type low amplitude pattern of AE signal Transient type high amplitude pattern of AE signal AE signal from tool breakage AE RMS signal at the moment of tool breakage Fig. 3. AE signal captured form conventional setup (a) raw AE signal captured during machine shut down, (b) raw AE signal captured during tool breakage and (c) RMS signal of Fig. 3(b). AE signal from conventional tool setup Mean line axis at Y value 0 AE signal from dummy tool setup Offset of the mean line from 0 Axis at Y value 0 Fig. 4. The raw AE signal taken from (a) the conventional tool setup and (b) the dummy tool setup in turning.

6 24 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) momentarily increases and the signal immediately diminishes. However for machine shut down, the signal characteristics are different and the signal is found to diminish gradually. The raw AE signal of Fig. 4(a) and (b) have been captured at cutting speed 130 m/min, feed rate 0.28 mm/rev and depth of cut of 1 mm from the conventional and the dummy setup, respectively. The raw AE signal, a type of complex wave is captured from the conventional tool setup has a fluctuation around a mean value which is zero in this case. However, the signals from the dummy tool holder which is to capture only the chip formation, removal, and breakage is clearly offset from the zero mean axis. The signals from the conventional setup show all the occurrences taking place and contain various frequencies and energies. On the contrary, the signal from the dummy setup coming from chip formation has the only high-energy frequency components and is therefore offset. The low energy frequency components arising from the plastic Fig. 5. (a) RMS signal of Fig. 6(a) and (b) RMS signal of Fig. 6(b). Fig. 6. The raw AE signal captured from the dummy tool setup at feed rate of 0.32 mm/rev, depth of cut of 1 mm and at four different cutting speeds of (a) 130 m/min; (b) 150 m/min; (c) 175 m/min and (d) 190 m/min.

7 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) deformation of workmaterial and tool wear are clearly absent in the signal captured from the dummy tool holder setup. The values remaining below the offset of transient signal indirectly represent the continuous, low amplitude components and therefore, indicate the tool wear and plastic deformation. However, the offset signal represents the intensity of chip formation occurrences during turning. This is because the signal from the dummy setup isolates the chip formation frequencies from the whole domain. The RMS signals of Fig. 5(a) and (b) represents the energy content of the corresponding raw AE signal of Fig. 4(a) and (b), respectively. From Fig. 5(a) and (b), the average amplitude of dummy setup RMS signal is V while that from the conventional setup is V. Since the dummy setup signal contains only the chip formation frequencies which are higher in energy content, the resultant RMS is considerably higher in amplitude. On the other hand, the conventional setup signal consisting of low and high frequency components contain the entire occurrences, the resultant RMS signal is essentially low in amplitude. The frequency levels of these signals are also found to differ from each other. From the signal of conventional tool setup, the frequencies are observed to fluctuate between 50 khz and 720 khz whereas from the dummy tool-holder setup, it varies between 71.3 khz and 201 khz. From Fig. 6(a) through (d), the AE signals of dummy tool setup are found to offset at every cutting condition mainly depending on the cutting speeds. The offset signals indicate the intensity of the chip formation on the tool state. From Fig. 6(a) to (c), both the amplitude and the offset of signals are found to simultaneously increase with the increase of cutting speed. However, an exception is observed from Fig. 6(d). At the highest cutting speed (190 m/min), the amplitude and the offset of the signal are found to decrease. The increase in amplitude of the dummy setup signal indicates the increase of intensity of chip formation affecting the tool state. For Fig. 6(a) to (d), the average amplitude of corresponding RMS AE signal has been measured and found to be V, V, V and V, respectively. The same fluctuation of signals has been observed from the RMS values of raw AE signals. With the increase of cutting speed, the chip formation rate increases, the chip formation types and the geometry also change. Fig. 7 represents the different chips formation collected (from dummy tool setup) at different cutting conditions in turning. Fig. 7. Various chips formation at different cutting conditions. (a) at v=130 m/min, f=0.32 mm/rev and ap=1 mm, (b) at v=150 m/min, f=0.32 mm/rev and ap=1 mm, (c) at v=175 m/min, f=0.32 mm/rev and ap=1 mm, (d) at v=190 m/min, f=0.32 mm/rev and ap=1 mm, (e) at v=175 m/min, f=0.28 mm/rev and ap=1 mm, (f) at v=175 m/min, f=0.50 mm/rev and ap=1 mm, (g) at v=150 m/min, f=0.28 mm/rev and ap=1 mm and (h) at v=150 m/min, f=0.28 mm/rev and ap=1 mm.

8 26 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) Fig. 7(a) to (d) represent the chips formation at different cutting speeds. From Fig. 7(a) (c) the chips are observed to be continuous, helical and serrated. With the increase of cutting speed, the chips are found to transform from wavy to the saw toothed type and the chip curvature is reduced. Therefore, the resulting cutting force and the contact length of chips with the cutting tool are increased [10]. Besides, both the thermal and the frictional effect of saw toothed chips on the tool-insert are most significant. The corresponding tool wear and the plastic deformation are therefore increased, and these are apparent on the signal of Fig. 6(a) through (c). However, the chip formation at highest cutting speed of Fig. 7(d) has shown not much change in types except the breakage. With the chip breakage, the developed stress inside the material is released and the plastic deformation is decreased. Therefore, the resultant effect on the cutting tool is reduced, the signal amplitude and the amplitude remaining below the offset are decreased. The corresponding flank wear of cutting tool has been measured by taking the tool insert off from tool holder at the end of every cut. A magnification of 40X has been used to capture the image of flank wear by light source microscope, Model: I CAMSCOPE(G). From the image, the average flank wear has been measured using a measuring software, Measure IT. The tool wear values are mm, mm, mm and mm for cutting conditions given in Fig. 7(a) through (d), respectively. From the analysis, it is observed that the chip breakage reduces the tool wear. The frequency analysis of the corresponding chip formation signal has shown a change as the cutting speed changes. The range of frequency fluctuation lies between khz and 191 khz for Fig. 7(a); 97.7 khz and khz for Fig. 7(b); 97.7 khz and khz for Fig. 7(c) and 97.7 khz and khz for Fig. 7(d). Although, no major difference has been observed between the chips from the conventional and dummy tool setups; in some cases slight distortion is found in chip formation. This is possibly generated from the collision between the chip and the dummy tool-insert during removal. As the chip touches the dummy tool setup only after leaving the main tool insert which is the object of interest; it does not have any influence on its formation. From Fig. 8(a) and (b), with the increase of feed rate, the amplitude and the offset of AE signal are observed to increase. The increase in amplitude indicates the increased effect of chip formation on the cutting tool. The increase in offset indicates the increase in plastic deformation and the tool wear during the chip formation. However, an exception is observed at the highest feed rate (0.50 mm/rev) where both the amplitude and the offset are found to decrease. The decrease in amplitude indicates a drop in the chip formation effect whereas the decrease in offset indicates a decrease in plastic deformation and the tool wear. For Fig. 8(a) to (c), the average amplitude of corresponding RMS AE signal has been monitored, and found to be V, V and V respectively. The same fluctuation of signals has been observed from the RMS values of raw AE signals. Fig. 7(c), (e) and (f) represents the corresponding chips formation at different feed rates. From Fig. 7(e) and (c), with the increase of the feed rate the chips are transformed from wavy to saw toothed type. The width of chip is increased and therefore, there is an increase in chip-tool contact length and in cutting AE signal components from chip formation Tool wear + Plastic deformation region Mean line axis at Y value 0 Fig. 8. The raw AE signal captured from the dummy tool setup at cutting speed of 175 m/min, depth of cut of 1 mm and at three different feed rates of (a) 0.28 mm/rev, (b) 0.32 mm/rev and (c) 0.50 mm/rev.

9 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) Fig. 9. The raw AE signal captured from the dummy tool setup at cutting speed of 150 m/min, feed rate of 0.28 mm/rev and at two different depths of cut of (a) 1 mm and (b) 2 mm. force [10]. Additionally, the thermal effect of serrated chip formation is more damaging to the cutting tool than the steady types. Therefore the corresponding chip formation effect, plastic deformation and the tool wear are found to increase with the increase of feed rate. However, in case of Fig. 7(f) an exception is noticed where the chips are found to break. In this case, the stresses are released, the corresponding cutting force and the plastic deformation are decreased and the resultan effect on the cutting tool is reduced. The tool wear values are mm, mm and mm for Fig. 7(e), Fig. 7(c) and (f), respectively. From the frequency analysis, the frequency fluctuation lies between 97.7 khz and khz for Fig. 7(e), 97.7 khz and khz for Fig. 7(c) and 97.7 khz and khz for Fig. 7(f). From Fig. 9(a) and (b), both the amplitude and the offset of the AE signal are observed to increase with the increase of depth of cut. The tool-workpiece contact area increases with the increase of depth of cut; therefore the cutting force, the plastic deformation and the resultant tool wear are found to increase. For Fig. 9(a) and (b), the average amplitude of corresponding RMS AE signal has been monitored and found to be V and V respectively. The same conclusion can be made from the RMS values of signals. Fig. 7(g) and (h) represents the chips formation at different depths of cut corresponding to signals. From Fig. 7(g) and (h), the chips are found to be continuous and cyclic at both cutting conditions. However, with the increase of depth of cut from 1 mm to 2 mm, the chips are observed to transform from wavy to the saw toothed type. Therefore, the stress inside material is increased and the corresponding effect on the tool wear and the plastic deformation increases. The measured tool wear are mm and mm for Fig. 7(g) and (h), respectively. From the frequency study, it is observed that the frequencies fluctuate between 68.3 khz and khz for Fig. 7(g) and for Fig. 7(h), the variation lies between khz and khz. The positive aspects of the dummy tool-holder setup in capturing the AE signal to monitor the chip formation effect on the tool state are: (i) the effect of chip formation on the tool state is very obvious and could monitor independently without any difficulty, however that is impossible for a conventional setup signal. (ii) from the average value of RMS signal, one can predict the tool wear and plastic deformation which is impossible from the conventional setup signal of Fig. 5(b). 5. Conclusions The tool wear measure has a strong dependency on the cutting condition. The chip formation has a remarkable effect on the tool life. The dummy setup has provided a means to monitor the chip formation effect independently and predict the rate of tool wear and plastic deformation of workmaterial from the AE signal. The dummy setup, the raw AE signal, its RMS and the frequency study of captured AE signal are able to describe and correlate the output signal with the tool state. The offset domain of the dummy setup signal represents the plastic deformation of workmaterial and the rate of tool wear together. The offset signal shows the impact of chip formation on the tool state. All the cutting parameters (cutting speed, feed and depth of cut) have significant effect on the chip formation and tool wear. Based on AE signal response, the important findings of this work are: The plastic deformation of material, rate of tool wear and chip formation types vary with cutting conditions. The chip formation is mostly affected by the change of cutting speed followed by the depth of cut and the feed rate. The rate of tool wear and plastic deformation of workmaterial increase with the increase of cutting speed, feed rate and depth of cut until the chip breakage. The tool wear decreases with the increase of chip breakage. With the increase of cutting speed from 130 m/min to 175 m/min at constant feed rate and depth of cut, the average amplitude of RMS AE signals has been observed to increase from V to V. The corresponding tool wear has increased from mm to mm. However, the exception is found at highest cutting speed (190 m/min) where the average amplitude of RMS AE signal has dropped from V to V. The corresponding tool wear has been found to decrease from mm to mm and the chips were found to break. At constant cutting speed and depth of cut and with the increase of feed rate from 0.28 mm/rev to 0.32 mm/rev, the average amplitude of RMS AE signal has been found to increase from V to V. The corresponding tool wear has increased from mm to mm. Both the average amplitude of RMS AE signal and the tool wear were found to decrease with chip breakage even at highest feed rate (0.50 mm/rev). The average amplitude of RMS AE signal has been observed to decrease

10 28 M.S.H. Bhuiyan et al. / International Journal of Machine Tools & Manufacture 58 (2012) from V to V while the corresponding tool wear has been found to reduce from mm to mm. With the increase of depth of cut from 1 mm to 2 mm and at constant cutting speed and feed rate, the average amplitude of RMS AE signal has been observed to increase from V to V. The corresponding tool wear increased from mm to mm. For entire cutting conditions, the chip formation frequency has been found to vary between 68.3 khz and khz while the maximum intensity was observed at 97.7 khz. Acknowledgement The authors would like to thank the UMRG (RG028/09AET), University of Malaya for providing the funds to carry out this work. References [1] S. Dolinsek, J. Kopac, Acoustic emission signals for tool wear identification, Wear (1999) [2] B.H. Jared, T.A. Dow, Investigation of the direction of chip motion in diamond turning, Precision Engineering 25 (2001) [3] J.D. Kim, O.B. Kweun, A chip-breaking system for mild steel in turning, International Journal of Machine Tools and Manufacture 37 (1997) [4] C.L. Tsai, Analysis and prediction of cutting forces in end milling by means of a geometrical model, International Journal of Advanced Manufacturing Technology 31 (2007) [5] V.P. Astakhov, The assessment of cutting tool wear, International Journal of Machine Tools and Manufacture 44 (2004) [6] H.A. Kishawy, J. Wilcox, Tool wear and chip formation during hard turning with self-propelled rotary tools, International Journal of Machine Tools and Manufacture 43 (2003) [7] M.C. Shaw, A. Vyas, The mechanism of chip formation with hard turning steel, CIRP Annals Manufacturing Technology 47 (1998) [8] M.A. Davies, Y. Chou, On chip morphology, tool wear and cutting mechanics in finish hard turning, CIRP Annals Manufacturing Technology 45 (1996) [9] L. Chen, T.I. EL-Wardany, W.C. Harris, Modelling the effects of flank wear land and chip formation on residual stresses, CIRP Annals Manufacturing Technology 53 (2004) [10] A.K. Balaji, R. Ghosh, Performance-based predictive models and optimization methods for turning operations and applications: Part 2 Assessment of chip forms/chip breakability, Journal of Manufacturing Processes 8 (2006) [11] B.H. Jared, T.A. Dow, Investigation of the direction of chip motion in diamond turning, Precision Engineering 25 (2001) [12] E. Shamoto, T. Aoki, Control of chip flow with guide grooves for continuous chip disposal and chip-pulling turning, Cirp Annals Manufacturing Technology 60 (2011) [13] D.S. Kilic, S. Raman, Observations of the tool chip boundary conditions in turning of aluminum alloys, Wear 262 (2007) [14] C. Arcona, T.A. Dow, A new technique for studying the chip formation process in diamond turning, Precision Engineering 18 (1996) [15] J. Abellan-Nebot, F. Romero Subirón, A review of machining monitoring systems based on artificial intelligence process models, International Journal of Advanced Manufacturing Technology 47 (2010) [16] E. Gandarias, S. Dimov, D.T. Pham, A. Ivanov, K. Popov, R. Lizarralde, New methods for tool failure detection in micromilling, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 220 (2006) [17] J. Barry, G. Byrne, Cutting tool wear in the machining of hardened steels: Part I: Alumina/TiC cutting tool wear, Wear 247 (2001) [18] R.S. Pawade, S.S. Joshi, Analysis of acoustic emission signals and surface integrity in the high-speed turning of Inconel 718, Journal of Engineering Manufacture 228 (2012) [19] R. Teti, I.S. Jawahir, K. Jemielniak, T. Segreto, S. Chen, J.I. Kossakowska, Chip form monitoring through advanced processing of cutting force sensor signals, CIRP Annals Manufacturing Technology 55 (2006) [20] T. Somkiat, In-process monitoring and detection of chip formation and chatter for CNC turning, Journal of Materials Processing Technology 209 (2009) [21] S. Tangjitsitcharoen, T. Moriwaki, Intelligent monitoring and identification of cutting states of chips and chatter on CNC turning machine, Journal of Manufacturing Processes 10 (2008) [22] H.V. Ravindra, Y.G. Srinivasa, R. Krishnamurthy, Acoustic emission for tool condition monitoring in metal cutting, Wear 212 (1997) [23] D. Dornfeld, S. Liang, Tool wear detection using time series analysis of acoustic emission, Journal of Engineering for Industry-Transactions ASME 111 (3) (1989) [24] X. Li, A brief review: acoustic emission method for tool wear monitoring during turning, International Journal of Machine Tools and Manufacture 42 (2002) [25] J.F. Xavier, S. Sampathkumar, Condition monitoring of turning process using AE sensor, Proceedings of ICME, Dhaka, Bangladesh.(2005). [26] X. Li, Detection of tool flute breakage in end milling using feed-motor current signatures, IEEE/ASME Transactions on Mechatronics 6 (2001) [27] I. Inasaki, Application of acoustic emission sensor for monitoring machining processes, Ultrasonics 36 (1998) [28] E. Emel, E. Kannatey-Asibu Jr, Acoustic emission and force sensor fusion for monitoring the cutting process, International Journal of Mechanical Sciences 31 (1989) [29] R.Y. Chiou, S.Y. Liang, Analysis of acoustic emission in chatter vibration with tool wear effect in turning, International Journal of Machine Tools and Manufacture 40 (2000) [30] T. Moriwaki, K. Okushima, Detection for cutting tool fracture by acoustic emission measurement, CIRP Annals 29 (1980)