A new approach to investigate tool condition using …

A new approach to investigate tool condition using ... based on its types. The cutting tool ... to detect tool wear and fracture in single point...

0 downloads 13 Views 2MB Size
Int J Adv Manuf Technol (2012) 61:465–479 DOI 10.1007/s00170-011-3722-7

ORIGINAL ARTICLE

A new approach to investigate tool condition using dummy tool holder and sensor setup Md. Sayem Hossain Bhuiyan & Imtiaz Ahmed Choudhury & Nukman Yusoff

Received: 14 October 2010 / Accepted: 24 October 2011 / Published online: 6 December 2011 # Springer-Verlag London Limited 2011

Abstract The industrial demand for automated machining systems to enhance process productivity and quality in machining aerospace components requires investigation of tool condition monitoring. The formation of chip and its removal have a remarkable effect on the state of the cutting tool during turning. This work presents a new technique using acoustic emission (AE) to monitor the tool condition by separating the chip formation frequencies from the rest of the signal which comes mostly from tool wear and plastic deformation of the work material. A dummy tool holder and sensor setup have been designed and integrated with the conventional tool holder system to capture the time-domain chip formation signals independently during turning. Several dry turning tests have been conducted at the speed ranging from 120 to 180 m/min, feed rate from 0.20 to 0.50 mm/rev, and depth of cut from 1 to 1.5 mm. The tool insert used was TiN-coated carbide while the work material was high-carbon steel. The signals from the dummy setup clearly differ from the AE signals of the conventional setup. It has been observed that time-domain signal and corresponding frequency response can predict the tool conditions. The rate of tool wear was found to decrease with chip breakage even at higher feed rate. The tool wear and plastic deformation were viewed to decrease with the increased radius of chip curvature and thinner chip M. S. H. Bhuiyan (*) : I. A. Choudhury : N. Yusoff Department of Engineering Design and Manufacture, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia e-mail: [email protected] I. A. Choudhury e-mail: [email protected] N. Yusoff e-mail: [email protected]

thickness even at the highest cutting speed, and these have been verified by measuring tool wear. The chip formation frequency has been found to be within 97.7 to 640 kHz. Keywords Tool state monitoring . Plastic deformation . Chip formation . Acoustic emission . Tool wear

1 Introduction Machining is embedded from the beginning of manufacturing to shape the material into the desired dimensional object. Each machining process has some exclusive individuality, which is involved with specific type of operation. The entire particular phenomena have their own effect on the process and also on the tool state. The effect on the tool could be mechanical, chemical, thermal, and abrasive, but all these results in tool wear, tool fracture, and breakdown. Tool wear causes extra power consumption and leads to inaccurate tolerances. Tool breakdown interrupts the operation and affects the product quality. To prevent tool failure and uncertainties, effective monitoring system should have to be in place for keeping the tool under surveillances. Currently, wide ranges of monitoring systems are available for tool condition monitoring. Some of them use the direct monitoring method, and some other use indirect method, whereas both drive to correlate system output signal with occurrences more accurately. All these methods represent the entire happenings during operation and are able to characterize the output signal into different occurrences according to the pattern of the signal. The assumption is use to interpret the signal and to correlate the signal pattern with the sources of occurrences. However, definitely physical visualization will be more effective and acceptable than the assumption. This is a prerequisite to know all about the tentative causes of tool state derogation before adapting a

466

monitoring system. The temperature or cutting speed is the major factor affecting wear mechanisms [1]. At slow cutting speeds, adhesion and abrasion are the main wear mechanisms. Abrasion and chemical wear is essential at high cutting speeds, especially in continuous chip formation [2]. The heat generation and the chip formation dominate the tool wear much. In metal cutting, the material removal from the workpiece is done by chip formation and is accomplished into two steps: chip generation and chip disposal/removal. Chip’s generation is also carried out into two stages: plastic deformation and crack growth. Each step and stage of material removal has its individuality and its own effect on the tool state. So the tool state monitoring by observing the chip removal might be an effective technique. It has a different level of effect on the tool face depending on chips’ energy content and its types as well [3]. The effect of chip formation depends on several things like how does the chip form, its type, how does it separate, how (speedy/slowly) is it removed, the level of energy content, the temperature content, and the intensity of chip impact on the tool face and so on. The plastic deformation of the work material and progressive tool wear comprise continuous but comparatively low level of energy. The crack growth, chip formation, and chip removal contain considerably high levels of energy; however, they recur with certain frequency. Thereby the effect on the tool state is like the burst and discontinuous. In addition, the chip formation and the removal are dominated by the cutting conditions. The increase in cutting speed comes with thin chips [3]. The corresponding effect of chip formation on the tool face is reduced and hence the wear. At the high cutting speed, the chip thickness reduces because with the increase of cutting speed, the controlled chip region decreases [4]. The chip formation has its own effect on tool state based on its types. The cutting tool experiences different level of wear from different chip geometry. With the decrease of chip curvature, the cutting force, the chip thickness, and the contact length are increased [5]. The consequent effect of tool geometry on cutting tool is increased and thus the tool wear. The different types of chip formation from the workpiece affect both the flank and rake faces of tool insert and propagate wear. The chip abrades much of the tool insert, and it also carries away most of the heat generated from metal cutting during its removal [6]. The chip formation and all other occurrences in machining require to be well investigated to avoid any undesirable catastrophic tool failure during machining. This paper describes a promising technique to separate the chip formation signal from the whole sources of occurrences. The purpose of this work is to distinguish and differentiate the signal component according to different occurrences from the complex signal pattern. An attempt has been made to incorporate a dummy tool insert in the tool holder and capture chip formation, removal, and breakage signal along with acoustic emission (AE) signal

Int J Adv Manuf Technol (2012) 61:465–479

from the cutting tool. AE signal from the cutting zone comprises of tool wear, plastic deformation of work material, chip formation and removal, and tool fracture. This study will aid to investigate the tool condition more effectively and would help to develop a skilled control system. The setup has been designed only for the experimental purpose and at this point not for practical use. The objective is to separate the different patterns of the signal with corresponding occurrence before incorporating it in the practical application. The investigation of the tool state would be much more effective and meaningful if the pattern and frequency of all sources are identified.

2 Acoustic emission in tool condition monitoring The AE is the transient elastic wave generated by the rapid release of energy from a localized source or sources within a material. During the AE process, a stress wave is generated and propagated through the material. This effect appeared as plastic deformation, phase transformations, vacancy coalescence, and decohesion of inclusions and fracture, which are sources of acoustic emission. However, only the plastic deformation and fracture have major significance in metal cutting [7]. Dornfeld pointed out that the possible AE sources referring to stress waves generated by the sudden release of energy in deforming material during metal cutting are (a) plastic deformation of the workpiece during cutting process, (b) sharing of the chip, (c) frictional contact between the tool flank face and the workpiece resulting in flank wear, (d) frictional contact between the tool rake face and the chip resulting in crater wear, (e) collisions between chip and tool, (f) chip breakage, and (g) tool fracture [8]. Figure 1 shows that all the above sources are associated with three different cutting zones, namely the primary deformation zone (shear zone), the secondary deformation zone (tool–chip interface), and

Fig. 1 AE signal generation at tool/chip interface [11]

Int J Adv Manuf Technol (2012) 61:465–479

the tertiary zone (rubbing and friction between the tool flank and newly machined surface). However, from Cohen’s [9] experimental results, it is confirmed that the primary deformation zone/shear zone is the largest source of AE and produces in excess of 75% of the total AE signal. The AE derived from turning of metals consists of continuous and transient signals, which have distinctly different characteristics. Continuous signals are associated with sharing in the primary zone and wear on the tool flank and rake face, while burst or transient signals result from either tool fracture or from chip formation (chip breakage, chip–tool collision). Therefore, AE sources given in (a) to (d) generate continuous AE signals, while sources from (e) to (g) generate transient AE signals [10], since the AE is a very high-frequency signal and generated from the object’s internal structure changes. Therefore, the possibility of the signal disturbance by surrounding noise is considerably less. 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. Some research has shown that AE has been successfully used in laboratory tests to detect tool wear and fracture in single point turning operations [9]. The AE signal captured from a conventional setup is a complex waveform, and it consists of numerous basic signals coming from different occurrences during machining. A wide range of signals is included in the same output makes it hardly readable. The frequencies due to plastic deformation, tool wear, breakage, chip formation, chip removal, process interruption, collision among tool–workpiece, tool–chip, chip–workpiece, etc. exist in raw signal. So further processing is required to extract the features from the signal. Usually, the pattern recognition analysis of AE signals is used to trace and extract the feature from the signal. The features extracted from the signals are the power of the signal, the power of the residual signal, and the autoregressive parameters of the AE signal. The cutting process is stochastic in nature, and a number of process parameters like material properties, tool geometry, etc. can vary during machining. Diagnosis based on few features only by fixing a threshold value is inadequate and undependable for tool wear monitoring. Hence, all the features need to be considered recognizing a change in the progressive wear of the tool and making the sensing system more efficient. This is done by a diagnostic system based on a pattern recognition technique [7]. Before adopting the pattern recognition system, feature reduction is done by selecting only the best features using feature selection criteria, i.e., class mean criteria [12]. It is generally agreed that the continuous-type AE signals are associated with plastic deformations and tool wear during metal cutting, while the burst-type signals are observed during crack

467

growth inside the material. Additionally, tool fracture, chip breaking, chip impacts, or chip tangling generate a bursttype AE signals [13–15]. As the chip formation is accomplished with crack growth, chip break, and chip removal, it is believed that the chip formation produces a transient burst AE signal during metal cutting. The RMS value of burst AE signal due to tool fracture depends on fracture area [16]. However, the AE signals associated with catastrophic tool failure are not influenced significantly by the engagement and disengagement of the tool, depth of cut and feed rate [15]. It is reported that the change in the AE RMS value at the point of tool failure is not significant, especially during interrupted cutting [17]. So the use of raw AE and other modulated form of the signals are rather meaningful than AE RMS values in interrupted cutting. Though the amplitude of chip formation signals is not the highest, it appears as a burst.

3 Materials and method The conventional tool setup and signal processing methods are not able to serve the purpose of this work. To investigate the chip formation signal from the complex waveform, it is necessary to separate and capture the signal independently. To materialize the objective, the conventional tool holder setup has to be modified. The raw AE signal and its RMS are used to illustrate the sensor output coming from the modified tool– sensor setup. 3.1 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 occurrences. One dummy tool setup that has been replicating the conventional tool setup is designed and integrated to the conventional tool setup. The dummy tool holder and tool insert arrangement has designed and fabricated to aid the AE signal to follow about the same path of transmission from sources to a sensor. Even though there would be some deviation of signal got from the difference of material properties of the tool holders, for both locations of sensor, the AE signals transmit from sources to the sensor through a tool insert and tool holder as well. The dummy tool insert and tool holder arrangement is mounted over the main tool insert and tool holder arrangement. The dummy tool holder is clamped using clamping bolts. The dummy insert is in a bit off position with respect to the main tool setup so that it cannot come in contact with the workpiece during the cutting operation. However, the chips that are released during metal cutting would come in contact with the dummy insert as it leaves the workpiece. A rubber insulation of 2 mm is placed in

468

Int J Adv Manuf Technol (2012) 61:465–479

Fig. 2 The setup for separating and capturing chip removal signal: a schematic view and b real view

between the main and dummy tool holder to avoid the mutual vibration effect on the signal. The rubber piece has helped to damp the low-frequency components from plastic deformation and tool wear, besides the AE sensor and the data acquisition (DAQ) system let the signal above 50 kHz to pass to storage. It is considered that the combine effect of rubber and DAQ could successfully make the dummy setup signal independent and uninfluenced. A piezoelectric AE sensor is placed on the dummy toll holder to sense 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 sketched in Fig. 3 in details. The performance of the dummy arrangement is tested and has got a satisfactory response. The AE signals from the new modified tool– sensor setup carries only the transient signal. Since the chip strike, the dummy tool insert forms, breaks, and is removed; the signal obtained from the new setup shows the chip formation occurrences corresponding to the different cutting conditions. As the sensor is placed on the

Fig. 3 The AE signal measuring chain in metal cutting

dummy tool holder and it never comes in contact with the main tool holder assembly, the sensor transient AE signal does not include the tool fracture signal. Figure 2a shows schematic of the modified tool holder setup while Fig. 2b shows a real view of the setup during cutting. This approach is not yet a customized technique; however, it would be a potential approach to monitor the tool state more effectively. 3.2 The AE signal acquisition technique The procedure of AE signal acquisition during metal cutting follows the pattern schematically illustrated in Fig. 3. The KISTLER 8152B AE-piezoelectric sensor mounted on the dummy tool shank is placed as close as possible to the chip removal zone. The AE sensor has a frequency range from 50 kHz to 1 MHz. The sensor holddown force of several Newtons is used to ensure good contact and to minimize the coupling thickness. Because of high impedance of the sensor, it must be directly connected to a coupler which contains a buffer

Int J Adv Manuf Technol (2012) 61:465–479

469

Table 1 Cutting conditions Cutting conditions

Cutting speed (m/min)

Feed rate (mm/rev)

Depth of cut (mm)

1

120

0.32

1

2

120

0.32

1.5

3 4

120 120

0.50 0.50

1 1.5

5

120

0.20

1

6 7

120 150

0.20 0.32

1.5 1 1.5

frequency noise components, which are inevitably present in AE signal, are considered not to be correlated with the occurrences and hence useless. Besides, it requires much energy to amplify such low frequencies and thus affect the useful band of the signal to amplify properly. Therefore, those components should be eliminated (using high-pass filter) at the earliest possible stage of signal processing to enable usage of full amplitude range of the equipment. The low-pass filter is used to filter out the high-frequency noise components to avoid electric sparks or aliasing of frequencies. The filtered AE signal is then amplified and digitized before storing for further processing.

8

150

0.32

9

150

0.50

1

10 11

150 150

0.50 0.20

1.5 1

3.3 Experimental details The turning operation is performed on a COLCHESTER VS MASTER3250 165×1,270-mm gap bed center lathe. The workpiece is a round bar (85 mm diameter and 760 mm long) of ASSAB-705, medium carbon steel (hardness HB 270–310). By weight, it contains carbon (0.35%), chromium (1.40%), iron (95.95%), manganese (0.70%), molybdenum (0.20%), and nickel (1.40%). The TiN-coated carbide, type: TNMG 16 04 08-PM tool insert and PTGNR 2020K-16 tool holder assembly, is used as a main tool arrangement. For the dummy arrangement of the tool, a mild steel tool holder is used, whereas the tool insert is the same. The experiment has been conducted in dry cutting mode for this investigation. The cutting conditions and tool–workpiece combination has a significant role on tool wear. Continuous cut is conducted for every 2 min to remove the material from the work piece. The conditions under which the experiment is carried out are tabulated below in Table 1.

12

150

0.20

1.5

13

180

0.32

1

14

180

0.32

1.5

15 16 17

180 180 180

0.50 0.50 0.20

1 1.5 1

18

180

0.20

1.5

amplifier. A KISTLER-5125B-type coupler and a DEWE-43 module are used in this experiment. The coupler allows the signal to pass through a high-pass filter and cut off below 50kHz frequencies. Then the signal feeds into the DEWE-43 module which cut off very high frequencies above 1,000 kHz. The coupler and DEWE-43 module jointly acts as a band-pass filter which has a low cutoff frequency of 50 kHz and a high cutoff frequency of 1,000 kHz. The necessary modification of the raw signal is undertaken inside this module. Low-

Fig. 4 a Raw AE signal taken from the conventional setup showing different incidences during turning at cutting speed 180 m/min, feed rate 0.32 mm/rev, and depth of cut 1.5 mm; b RMS of the same signal

470

Int J Adv Manuf Technol (2012) 61:465–479

Fig. 5 The raw AE signals a from conventional tool setup and b from dummy tool setup without any insulation between the main and dummy tool holder

4 Result and discussion The tool condition monitoring becomes fruitful when the AE signal representation is comprehensive and unambiguous. The AE can extract more information from the monitored occurrences after analyzing the raw AE signal. The raw AE signal is the representation of all incidences described in signal during turning. The RMS shows the average energy content in the raw signal. The raw AE, its frequency level, and RMS AE are capable of giving some significant information about the occurrences. The raw AE signals are very squiggle and stochastic in nature. It is a resultant illustration of all occurrences that takes place during turning. During normal turning, some major occurrences like the plastic deformation, tool wear, tool

Fig. 6 a RMS signal of Fig. 5a and b RMS signal of Fig. 5b

fracture, and chip formation all have significant effect on tool state and hence on the AE signal. For the conventional setup of AE sensor, the continuous, low amplitude pattern of raw signals represents plastic deformation and tool wear, whereas the burst and discrete pattern of signal represent the chip formation and tool fracture occurrences. For tool breakage, the burst signal amplitude will be higher and momentarily the signal is lost for a short interval. Both types of signal pattern co-exist in the raw AE signal. Because of its random shape, it is quite difficult to exactly measure the magnitude of continuous-type low-amplitude pattern from the raw signals until separation is possible. The dummy tool setup is used to separate the chip formation frequencies from others. After separation, the continuous and the transient pattern of AE signal stand separate and are easily distinguishable. The

Int J Adv Manuf Technol (2012) 61:465–479

471

Fig. 7 Raw AE signal a taken from the conventional tool setup at cutting speed 120 m/min, feed rate 0.32 mm/rev, and depth of cut 1.5 mm; b captured from the dummy tool setup with insulation at cutting speed 120 m/min, feed rate 0.32 mm/rev, and depth of cut 1.5 mm

continuous, low amplitude AE signal represents the plastic deformation and tool wear, whereas the transient AE signal corresponds to chip formation. The raw AE signals captured from the system are more meaningful to correlate with the occurrences.

enclosed in the circle representing the tool breakdown incidence, however, is not obvious from the raw signal. From Fig. 4b, it is apparent that the energy content of the RMS signals fluctuates all along the operation. There is a sudden drop and die down of the signal at the moment of tool breakdown.

4.1 AE signal pattern The raw AE signal of Fig. 4a represents all the occurrences or sources during turning. Two types of pattern are more distinct in the signal: (a) the continuous and low amplitude pattern and (b) the transient and burst pattern. The signal

Fig. 8 a RMS signal of Fig. 7a; b RMS signal of Fig. 7b

4.2 AE signals without insulation between the main tool holder and the dummy tool holder The raw AE signal of Fig. 5a is captured at a random cutting condition from the conventional tool setup in

472

Int J Adv Manuf Technol (2012) 61:465–479

Fig. 9 a Raw AE signal taken from dummy tool setup at cutting speed 120 m/min, feed rate 0.32 mm/rev, and depth of cut 1 mm; b RMS of the same signal

turning. Figure 5b is taken at the same cutting condition from the dummy tool setup, when there was no insulation between the main and dummy tool holders. The signal of Fig. 6a, b represents the RMS form of signal Fig. 5a, b, respectively. From Fig. 5a, b, it is obvious that there is no remarkable difference between the signals except the amplitude. The RMS signals of Fig. 6a, b represent the energy content of corresponding raw AE signal. From Fig. 6a, the energy content of the AE signal captured from the conventional setup fluctuates between 0.0019 and 0.4837 V while that

from the dummy setup (refer Fig. 6b), the fluctuation is between 0.0052 and 0.1713 V. As there is no insulation between the tool holders, almost the entire frequency of AE signal from the main tool holder is transmitted to the dummy tool holder. However, due to system loss, some of signal frequencies might be missing and thereby the amplitude of the dummy setup AE signal is different from the amplitude of signal taken from the main tool setup. Therefore, to eliminate the effect of interaction, insulation is needed between the dummy setup and the main tool holder.

Fig. 10 a Raw AE signal taken from dummy tool setup at cutting speed 120 m/min, feed rate 0.32 mm/rev, and depth of cut 1.5 mm; b RMS of the same signal

Int J Adv Manuf Technol (2012) 61:465–479

4.3 AE signal with insulation between the main tool holder and the dummy tool holder Figure 7a, b shows that the raw AE signals are captured at the same cutting conditions from the conventional tool setup and dummy tool setup, respectively. The signals are captured with insulation provided between the main tool holder and the dummy one. Figure 8a, b shows the RMS AE signals of Fig. 7a, b, respectively. The raw AE signal, a type of complex wave that 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 (with insulation) which is supposed 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 (with insulation) coming from chip formation has the only high-energy frequency components and is therefore offset. Comparing Figs. 5b and 7b, it is clear

473

that the rubber insulation placed between the conventional and dummy tool holder has performed successfully and eliminated interaction effect. Similar observations have been made with regard to the RMS signals (Figs. 6b and 8b). Based on the analysis, the values remaining below the offset of transient signal indirectly represent the continuous, low amplitude components and, therefore, the tool wear and plastic deformation. This is because the signal from the new setup isolates the chip formation components from the whole domain. The offset signal represents the chip formation occurrences which have high-energy content. Figures 9a, b and 10a, b represent raw AE and its RMS signal at two different depths of cut with same cutting speed and feed rate, respectively. From these figures, it is obvious that the amplitude of signal increases with the increase of depth of cut at constant speed and feed rate. At 1 mm depth of cut, the RMS values lie between 1.064 and 3.005 V; while at 1.5 mm depth of cut, the RMS values fluctuate between 1.0046 and 3.4533 V. This is because, with the increase of depth of cut, the chip thickness is

Fig. 11 The raw AE signal captured from the dummy tool setup at cutting speed 150 m/min, depth of cut 1 mm, and with three different feed rates a at 0.20 mm/rev, b at 0.32 mm/rev, and c at 0.50 mm/rev

474

increases, and therefore, it affects both the chip formation and the corresponding AE signal. From the frequency analysis, it is observed that, at the cutting speed of 120 m/min, feed rate 0.32 mm/rev, and depth of cut 1 mm, the frequency of chip formation lies between 97.7 and 430 kHz. However, with the change of depth of cut to 1.5 mm, it fluctuates from 97.7 to 420 kHz. Figure 11a–c represents the AE signal at constant cutting speed and depth of cut with three different feed rates. All the AE signals are captured from the dummy tool setup with insulation between the main and dummy tool holder. Figure 12 shows the corresponding RMS signal of Fig. 11. From Fig. 12a, it is observed that at feed rate 0.20 mm/rev, the RMS values of the AE signal lies between 2.7205 and 3.3149 V while the frequency of chip formation varies from 97.7 to 270 kHz. However, as the feed rate increases, both the RMS and frequencies response are mixed in nature. At feed rate of 0.32 mm/rev, the RMS value fluctuates within 2.178 and 6.8387 V and the frequency from 97.7 to 640 kHz. At the highest feed rate of 0.5 mm/rev, it is expected that both RMS value and frequency of chip formation would increase. The RMS values at 0.50 mm/rev fluctuates between 3.8211 to 3.3149 V while the frequency lies within 97.7 to 570 kHz, which are smaller than that at 0.32 mm/rev. Because of chip breakage, RMS value and

Int J Adv Manuf Technol (2012) 61:465–479

frequency response at 0.50 mm/rev are found smaller than that at 0.32 mm/rev. The continuous chip is maintained until the feed rate of 0.32 mm/rev. From Fig. 12a–c, it is apparent that the RMS values of tool wear and plastic deformation decrease even at high feed rate due to chip breakage. Figure 13 represents the AE signals at constant feed rate and depth of cut with three different cutting speeds. The AE signals are taken from the dummy tool setup with insulation between the main and dummy tool holder. Figure 14 shows the corresponding RMS signal of Fig. 13. At cutting speed of 120 m/min with constant feed rate and depth of cut, the RMS value lies in between 0.9966 and 3.9533 V while the frequency varies from 97.7 to 430 kHz. As the cutting speed changes to 150 m/min, the RMS vales fluctuate between 1.6708 and 5.306 V and the frequency between 97.7 and 640 kHz. At the highest cutting speed 180 m/min, the RMS values fluctuate between 2.0449 and 5.3064 V whereas the frequency response lies within 97.7 and 190 kHz. The RMS value and frequency response at cutting speed 180 m/min are decreased from those at 150 m/min indicating chips breakage leading to the reduction of plastic deformation and tool wear. The advantages of the dummy setup in capturing the AE signal and the RMS due to chip formation are:

Fig. 12 a RMS signal of Fig. 11a, b RMS signal of Fig. 11b, and c RMS signal of Fig. 11c

Int J Adv Manuf Technol (2012) 61:465–479

475

Fig. 13 a The raw AE signal captured from the dummy tool setup at feed rate 0.32 mm/rev, depth of cut 1 mm, and for three different cutting speeds a at cutting speed 120 m/min, b at cutting speed 150 m/min, and c at cutting speed 180 m/min

(a) From the RMS signal, one can predict the tool wear and plastic deformation which is impossible from the signal shown in Fig. 7a of a conventional setup. (b) The signal from the conventional setup cannot distinguish the effect of chip formation while this is clearly visible from the signal of the dummy setup.

4.4 Chip formation using the conventional tool setup The chips of Fig. 15 are collected from the conventional tool setup in turning. Figure 15a–f shows the various chips formed at different cutting conditions indicated in the figures. The chips are observed to be continuous, helical, straight, and serrated. Figure 15a, b shows the chips generated at constant cutting speed of 120 m/min, feed rate 0.32, and with different depths of cut of 1 and 1.5 mm, respectively. From Fig. 15a, b, it is noticed that with the increase of depth of cut, the chip transforms from wavy to the saw toothed type, and the chip curvature is reduced as well. With the decrease of chip curvature, the cutting force,

the chip thickness, and the contact length are increased. Besides, both the thermal and the frictional effect of saw toothed chips on the tool insert are most significant. Therefore, the corresponding effect on the tool state is found to increase. For Fig. 15a, b, the energy contents of the corresponding AE signals are 3.005 and 3.4500 V, respectively (refer Figs. 9b and 10b). The flank wear is measured for the same period of time. The measured tool wear during chip formation for Fig. 15a is 0.0094 mm and for Fig. 15b is 0.0162 mm. The AE signals as well as the measured of tool wear have shown a similar response to the change of chip formation types. Figure 15c–e shows the chips formed at constant cutting speed 150 m/min, depth of cut 1 mm, and with three different feed rates of 0.20, 0.32, and 0.50 mm/rev, respectively. With the increase of the feed rate, the chips are transformed gradually from wavy to serrated, the chip curvature is changed, and at the highest feed rate (0.50 mm/rev), the chip breakage occurs. For Fig. 15c–e, the energy contents of the corresponding AE signals are 3.3149, 6.8387, and 5.5329 V, respectively (refer Fig. 12). The more remarkable observation is that the energy content of AE

476

Int J Adv Manuf Technol (2012) 61:465–479

Fig. 14 a RMS signal of Fig. 13a, b RMS signal of Fig. 13b, and c RMS signal of Fig. 13c

signal increases as the chip transforms from wavy to serrated and as the chip curvature decreases. However, there is exception when the chip is broken. The energy content of the corresponding AE signal is decreased with the chip breakage. The corresponding tool wears are measured, the tool wear at feed rate of 0.20 mm/rev is 0.0133 mm, at feed rate of 0.32 mm/rev is 0.0432 mm, and at feed rate of 0.50 mm/rev is 0.0263 mm. The tool wear has shown an analogous change to the AE signal and chip formation occurrences as well. Figure 15a, d, and f shows the chips formed at constant feed rate of 0.32 mm/rev and depth of cut of 1 mm with three different cutting speeds of 120, 150, and 180 m/min, respectively. The chips are continuous at all cutting speed; however, at cutting speed 150 m/min, the chips are segmented and a reduction in chip curvature is observed. As the speed changes from 150 to 180 m/min, the chip thickness decreases while the radius of curvature increases. Therefore, the corresponding effect on the tool is decreased. For Fig. 15a, d, and f, the energy contents of the corresponding AE signals are 3.005, 6.8387, and 4.6289 V, respectively (refer Fig. 14). From the energy analysis of the AE signal, it is seen that the energy content of the AE signal has followed the chip formation occurrences and have changed with the chip formation types. The energy level of the AE signal is appeared to rise with the increase of cutting

speed. However, at the highest cutting speed (180 m/min), a drop is observed. This is because of the increased radius of chip curvature and thinner chip thickness. The corresponding tool wears for Fig. 15a is 0.0094 mm, for Fig. 15d is 0.0432 mm, and for Fig. 15f is 0.0198 mm. In the case of the highest cutting speed 180 m/min, the increase in the cutting speed leads to a decrease in the chip formation zone. Therefore, the corresponding plastic deformation and the resultant tool wear are also reduces. From the above explanation, the response of AE signals comports with the tool wear and the chip formation in turning. 4.5 Chip formation using the dummy tool setup Figure 16a–f shows the corresponding chip types that have been generated from the new modified setup under the same cutting conditions. The chip’s morphology at different cutting conditions is observed to be identical with the chips generated from the conventional setup; however, in some cases, slight distortion of chip is noticed. Comparing Figs. 15 and 16, no major change either in chip formation or in chip curl or in chip breakage is noticed, since the chip curl is controlled by radius of chip curvature and thus by the depth of cut and that was being unaffected by the dummy setup anyway. Therefore, not much change in chip

Int J Adv Manuf Technol (2012) 61:465–479

477

Fig. 15 Chips from the conventional tool setup at different cutting conditions

curls is seen. Besides, the tool insert used in the experiment has chip breaker which breaks the chip and helps it to curl before making any contact with the insert placed on the dummy setup. As in the modified setup, the dummy insert is placed 2 mm above the main insert, and also at an offset distance from the cutting plane, there is no possibility that the dummy insert would influence the chip shape. The gap 2 mm is because of rubber insulation placed in between the main and dummy tool holders. Referring Fig. 2, this could be clearer. However, the chips (curled/straight) that formed would touch the dummy tool insert along with leaving the tool contact point. Thereby, some distortion is observed in the chip construction. As the main tool insert is our object of interest, trivial distortion in chip construction is ignored and is considered not affecting the tool state and the cutting conditions as well. Figure 17a–f shows the corresponding flank wear values at cutting conditions given in either Fig. 15 or Fig. 16 at which the chips are collected. A new tool insert has been used for the every cutting condition until failure. The image of tool wear until failure is not presented in this paper as the chips collected and presented here are for the initial tool wear only. The tool flank wear has been measured by taking the tool insert off from tool holder at the end of every cut. A magnification of ×40 has been used to capture the image of flank wear by an light source microscope, model I CAMSCOPE(G). From the captured image, the average flank wear has been measured using a measuring software, Measure IT.

5 Conclusions The plastic deformation and tool wear measure have a strong dependency on the cutting condition. The chip formation also has a remarkable effect on the tool life. The following conclusion could be made based on the experimental investigations: &

&

&

&

The dummy setup could successfully separate the chip formation occurrences from the whole domain. The new technique is capable of independently monitoring the chip formation effect and also the plastic deformation and tool wear. Visualizing the raw AE signal, its frequency range, and AE RMS, one can investigate as well as predict the tool condition. The information about the chip formation types corresponding to different cutting conditions has made the monitoring more effective. The AE signal from the dummy setup describes the tool wear, plastic deformation, and chip formation occurrences more clearly and without any ambiguity. The values remaining below the offset of the transient AE signal of the dummy setup presents the plastic deformation and tool wear. The offset signal shows the chip formation occurrences in turning. The plastic deformation, tool wears, and chip formation types vary with cutting conditions. The frequency of chip formation varies from 97.7 to 640 kHz for the entire cutting conditions, whereas the maximum chip

478

Int J Adv Manuf Technol (2012) 61:465–479

Fig. 16 Chips from the dummy tool setup at different cutting conditions

&

formation frequency has been recorded at cutting speed of 150 m/min, feed rate of 0.32 mm/rev, and depth of cut of 1 mm. Increasing the feed rate from 0.32 to 0.50 mm/rev at a constant cutting speed of 150 m/min and a depth of cut of 1 mm, the tool wear has decreased from 0.0432 to 0.0263 mm, and the chips were found to break at the highest feed. The amplitudes of RMS AE signals also drop from 6.8387 to 5.5329 V. On the other hand, at constant feed rate 0.32 mm/ rev and depth of cut 1 mm, with the increase of

Fig. 17 Tool flank wear at different cutting conditions

&

cutting speed from 150 to 180 m/min, the chip thickness has decreased while the radius of curvature increased. The resultant tool wear has reduced from 0.432 to 0.0198 mm. The energy level of the corresponding AE signal has also dropped from 6.8387 to 4.6289 V. With the drop of amplitude in the chip formation frequency, the band remaining below the RMS signals also shrinks indicating a reduction in plastic deformation and tool wear rate as well.

a

b VBaverage= 0.0094 mm

c

VBaverage= 0.0162 mm

d VBaverage= 0.0133 mm

e

VBaverage= 0.0432 mm

f VBaverage= 0.0263 mm

VBaverage= 0.0198 mm

Int J Adv Manuf Technol (2012) 61:465–479

& &

This investigation shows that the chip breakage reduces the rate of tool wear progression even at higher feed rate. The tool wear has decreased with the decrease of chip thickness and increase of radius of chip curvature.

Hence, the dummy setup has made it possible to predict tool wear progression. The next challenge is to separate the frequency of plastic deformation from the tool wear so that on line tool condition monitoring becomes much more effective.

479

6.

7.

8.

9.

10. Acknowledgments The authors would like to thank the UMRG, University of Malaya for providing the funds to carry out this work. 11.

References 1. Loffler FHW (1994) Systematic-approach to improve the performance of PVD coatings for tool applications. Surf Coat Technol 68:729–740 2. Dolinsek S, Kopac J (1999) Acoustic emission signals for tool wear identification. Wear 225:295–303 3. Kim JD, Kweun OB (1997) A chip-breaking system for mild steel in turning. Int J Mach Tools Manuf 37:607–617 4. Tsai CL (2007) Analysis and prediction of cutting forces in end milling by means of a geometrical model. Int J Adv Manuf Technol 31:888–896 5. Balaji AK, Ghosh R (2006) Performance-based predictive models and optimization methods for turning operations and applications:

12.

13. 14.

15. 16. 17.

part 2—assessment of chip forms/chip breakability. J Manuf Process 8:144–158 Kalpakjian S, Schmid SR (2001) Fundamentals of cutting. In: Kalpakjian S, Schmid SR (eds) Manufacturing engineering and technology, 4th edn. Prentice Hall, Upper Saddle River, pp 534–568 Ravindra HV, Srinivasa YG, Krishnamurthy R (1997) Acoustic emission for tool condition monitoring in metal cutting. Wear 212:78–84 Liang S, Dornfeld D (1989) Tool wear detection using time series analysis of acoustic emission. J Eng Ind Trans ASME 111:199– 205 Saini DP, Park YJ (1996) A quantitative model of acoustic emissions in orthogonal cutting operations. J Mater Process Technol 58(4):343–350 Li X (2002) A brief review: acoustic emission method for tool wear monitoring during turning. Int J Mach Tools Manuf 42:157– 165 Xavier JF, Sampathkumar S (2005) Condition monitoring of turning process using AE sensor. Proceedings of ICME 2005, Dhaka, Bangladesh, 28–30 December 2005, AM-20 Emel E, Kanoatey-Asibu E (1988) Toot failure monitoring in turning by pattern recognition analysis of AE signals. Trans ASME J Eng Ind 110:137–145 Inasaki I (1998) Application of acoustic emission sensor for monitoring machining processes. Ultrasonics 36:273–281 Chiou RY, Liang SY (2000) Analysis of acoustic emission in chatter vibration with tool wear effect in turning. Int J Mach Tools Manuf 40:927–941 Moriwaki T, Okushima K (1980) Detection for cutting tool fracture by acoustic emission measurement. CIRP Ann 29:35–40 Lan MS, Dornfeld DA (1984) In-process tool fracture detection. J Eng Mater Technol 106:111–118 Jemielniak K, Otman O (1998) Catastrophic tool failure detection based on acoustic emission signal analysis. CIRP Ann 47:31–34