Elliptical vibration cutting of hardened die steel with coated carbide tools

Elliptical vibration cutting of hardened die steel with coated carbide tools

Accepted Manuscript Title: Elliptical Vibration Cutting of Hardened Die Steel with Coated Carbide Tools Author: Hiroshi Saito Hongjin Jung Eiji Shamot...

2MB Sizes 2 Downloads 31 Views

Recommend Documents

No documents
Accepted Manuscript Title: Elliptical Vibration Cutting of Hardened Die Steel with Coated Carbide Tools Author: Hiroshi Saito Hongjin Jung Eiji Shamoto PII: DOI: Reference:

S0141-6359(16)00006-4 http://dx.doi.org/doi:10.1016/j.precisioneng.2016.01.004 PRE 6333

To appear in:

Precision Engineering

Received date: Revised date: Accepted date:

11-6-2015 18-12-2015 4-1-2016

Please cite this article as: Saito H, Jung H, Shamoto E, Elliptical Vibration Cutting of Hardened Die Steel with Coated Carbide Tools, Precision Engineering (2016), http://dx.doi.org/10.1016/j.precisioneng.2016.01.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

*Highlights (for review)

Ac

ce pt

ed

M

an

us

cr

ip t

Highlights:  A transferability of cutting edge profile to workpiece was evaluated.  Diamond coated tool has higher transferability compared with conventional TiN coated tool especially at a small pick feed of 3 μm or 5 μm.  Cutting mechanics was discussed by analyzing ploughing force components, friction coefficient between cutting tool and workpiece, and observing chips.  Wear resistance of diamond coated tool and TiN coated tool was evaluated by conducting tool life tests.

Page 1 of 26

Elliptical Vibration Cutting of Hardened Die Steel with Coated Carbide Tools Hiroshi Saito*, Hongjin Jung**, Eiji Shamoto**

cr

ip t

* Yamagata Research Institute of Technology, 2-1 Matsuei 2-chome, Yamagata-shi, Yamagata-ken 990-2473, Japan Tel: +81-23-644-3222, Fax: +81-23-644-3228 [email protected]

an

us

** Nagoya University, Furo-cho, Chikusa-ku, Nagoya-shi, Aichi-ken 464-8601, Japan [email protected] [email protected] Tel: +81-52-789-2705, Fax: +81-52-789-5305

Ac

ce pt

ed

M

Abstract: Elliptical vibration cutting of hardened die steel with coated carbide tools is examined in this research in order to achieve low-cost high-precision machining. Diamond coated tools are applied because of superior hardness of their polycrystalline diamond coating and its low manufacturing cost. TiN coated tools are also tested, since they are widely used for conventional machining of steels. Machinability of hardened die steel by the elliptical vibration cutting with coated carbide tools is discussed in three aspects in this study, i.e. transferability of cutting edge profile to cut surface, cutting force, and tool life. The transferability is evaluated quantitatively by calculating correlation coefficients of measured roughness profiles. It is clarified that the diamond coated tools have high transferability which leads to diffraction of light on the surface machined at micro-scale pick feed. Total cutting forces including ploughing components are measured at various feed rates, and then shearing components and ploughing components are separated utilizing linear regression. The measured results indicate, for example, that the all forces become considerably smaller only when elliptical vibration is applied to the TiN coated tool without cutting fluid. It is also found that this considerable reduction of forces interestingly corresponds to higher friction coefficient, which is identified from the ploughing components. Tool life tests are carried out by various machining methods, i.e. elliptical vibration/ordinary wet/dry cutting with diamond/TiN coated tools. The result shows, for example, that the flank wear is smallest in the wet elliptical vibration cutting with the diamond coated tool. Keywords: diamond coated carbide tool, elliptical vibration cutting, transferability, cutting force, tool life

Page 2 of 26

1. INTRODUCTION

Ac

ce pt

ed

M

an

us

cr

ip t

Development of high precision and cost effective machining method for hardened die steel has been demanded especially in die and mold manufacturing. Hardened die steel is generally finished after hardening to prevent geometrical change and surface oxidation due to the heat treatment. However, the hardness and the brittleness of the hardened steel make it difficult to achieve high quality surface by using conventional machining methods. Single crystalline diamond tools are suitable for ultraprecision machining since they have superior mechanical properties such as high hardness and nano-order sharpness of cutting edges. On the other hand, the use of the single crystalline diamond tools is restricted to non-ferrous materials such as oxygen-free copper, electroless nickel or plastics, and they cannot be used for ferrous materials because of rapid tool wear [1]. Song et al. reported that the diamond tool wear is highly dependent on the tool–workpiece contact time, and that the wear can be greatly reduced when the contact time is less than 0.3 ms [2]. Elliptical vibration cutting, which is one of the intermittent cutting processes, has succeeded in suppressing the wear of the single crystalline diamond tools when machining ferrous materials, and it realized ultraprecision machining of hardened die steels [3-6]. However, the cost of the single crystalline diamond tools is higher than ordinary cutting tools made of sintered carbide or high speed steel. To overcome this problem, alternative low cost cutting tools have been examined. For example, Rahman et al. achieved sub-micron surface roughness of hardened steel by applying the elliptical vibration cutting with PCD (Poly-crystalline diamond) tools [7]. The elliptical vibration cutting with PCD tools is tested on other hard-to-machine materials such as sintered carbide as well [8]. Although the manufacturing cost can be reduced by using the PCD tools compared with the single crystalline diamond tools, the PCD tools are still more expensive than the ordinary cutting tools. Two kinds of coated carbide tools, i.e. diamond coated tool and TiN (titanium nitride) coated tool, are applied to the elliptical vibration cutting of hardened die steel to achieve its low-cost high-precision machining in this research. The diamond coated tool has thin diamond layer which is coated on carbide tool by CVD (Chemical Vapor Deposition) process. It has superior mechanical properties such as high hardness and low coefficient of friction [9-11]. Moreover, the manufacturing cost of the diamond coated carbide tool is lower than other diamond tools [12]. The TiN coated tool is widely used in dies and molds industry, and the characteristics of the TiN coated tool is thought to be different from the diamond coated tool (for example, the cutting edge of the TiN coated tool is generally sharper than that of the diamond coated tool). In order to clarify performance of the coated carbide tools, various experiments are conducted and machinability of hardened die steel by the elliptical vibration cutting with the coated carbide tools is discussed as follows. Transferability of cutting edge profiles to workpiece surfaces is evaluated quantitatively by utilizing correlation coefficients of roughness profiles. Moreover, mechanics of elliptical vibration cutting of the hardened die steel with the coated tools are discussed based on analysis of measured cutting forces and observation of finished surfaces and chips. Influence of elliptical vibration and cutting fluid on the tool wear is also examined by conducting various tool life tests.

Page 3 of 26

At last, cutting performances and characteristics of these coated carbide tools are discussed based on analysis of the experimental results.

ip t

2. METHODS FOR EXPERIMENTS AND EVALUATIONS

Ac

ce pt

ed

M

an

us

cr

2.1. Workpices and cutting tools A typical hardened die steel of Stavax (AISI 420 modified) with hardness of 54 HRC is used for workpieces. The size of the workpiece is 6040 mm, and it is fixed on the dynamometer (Kistler, 9256C) to measure cutting forces. Two kinds of coated carbide tools, diamond coated tools and TiN (titanium nitride) coated tools, are tested in this study. The nose radii of the both coated tools is 0.4 mm, and their base materials are sintered carbide. Coated carbide tools have cutting edge roundness according mainly to thickness of coatings. Figure 1(a) shows cross sections perpendicular to the cutting edges measured with an optical surface profiler (Zygo, NewView7300). The cutting edge roundness is evaluated by calculating the best fit circular arc. The measurements and their evaluations are conducted 9 times for each kind of coated tools, and the mean edge radii of the diamond coated tools and the TiN coated tools are identified to be 16.3 μm and 5.8 μm respectively. Accuracy of the cutting edge profile is also an important factor because its roughness is transferred to the finished surface of the workpiece. However, it is difficult to measure the cutting edge profile engaged to the workpiece. Instead of measuring the cutting edge profile, surface roughness of the flank face along the cutting edge is measured as shown in Fig. 1(b). The roughness Rt of the diamond coated tool and the TiN coated tool are identified to be about 750 nm and 500 nm, respectively.

Figure 1(a)

Page 4 of 26

ip t cr us an

Figure 1(b)

ed

M

Figure 1; Cross sections and surface roughness of coated carbide tools. (a); Cross sections of coated carbide tools (R: radius of circular arc fitted to cross section curve of cutting edge by least-squares method). (b); Surface roughness of the flank face measured along the cutting edge.

Ac

ce pt

2.2. Fundamental planing experiments The elliptical vibration cutting process can be illustrated schematically in Fig. 2. The elliptical vibration is applied so that the process becomes intermittent. The chip is pulled up during cutting while the tool moves down without cutting. The pulling force reduces or reverses the tool-chip friction force and decreases chip thickness, cutting force or energy [3]. The intermittent process and the reduced energy result in practically long tool life [2]. Experimental setup for fundamental planing experiments is shown in Fig. 3. Elliptical vibration cutting device (Taga Electric, EL-50Σ) is equipped on an ultraprecision machine tool (Nagase Integrex, N2C-53US4N4). The cutting tools are vibrated at a frequency of about 40 kHz and an amplitude of 4 μmp-p. Cutting oil (Palace chemical Co. Ltd., Nano-cut 9) is supplied to the cutting point with compressed air through a mist nozzle. The same cutting oil is used in other cutting experiments in this research. The cutting conditions are summarized in Table 1.

Page 5 of 26

ip t cr

ce pt

ed

M

an

us

Figure 2; Schematic illustration of elliptical vibration cutting process.

Figure 3; Experimental setup for fundamental planing experiments. Table 1; Cutting conditions of fundamental planing experiments

Ac

Machining conditions Depth of cut (μm) 10 Pick feed (μm) 3,5,10,20,40,60 Cutting speed (m/min) 1 Cutting fluid Oil mist Elliptical vibration conditions Frequency (kHz) About 40 Amplitude (μmp-p) 4 Phase shift (deg.) 90 (Circle)

2.3. Evaluation of finished surfaces The finished surfaces are evaluated here, since they are fundamentally important in precision machining. First, surface roughness Rt of the finished surfaces is measured by using an optical surface profiler (Zygo, NewView7300). The surfaces are also observed

Page 6 of 26

M

an

us

cr

ip t

with a DIC (Differential interference contrast) microscopy (Carl Zeiss, LSM5 PASCAL). These methods are effective to analyze surface roughness and burr formation on finished surfaces. On the other hand, the authors could not find a good method to quantitatively evaluate transferability of cutting edge profiles to the finished surfaces. Therefore, a new method is proposed in this study to evaluate the transferability by utilizing correlation coefficients of roughness profiles. Measured roughness profiles at a pick feed of 3 μm are shown in Fig. 4. Periodical cutter marks can be observed more clearly in the roughness profile obtained with the diamond coated tool than in that obtained with the TiN coated tool. This result implies that the diamond coated tool has higher transferability of the cutting edge profile to the finished surface. To evaluate this transferability quantitatively, the correlation coefficients of the roughness profiles are calculated. It is desirable to calculate the correlation coefficients between the cutting edge profile and the surface roughness profile. However, it is difficult to measure the cutting edge profile which generates the finished surface. Thus, instead of the cutting edge profile a segment x={x1,x2,...,xi,...,xn}, whose length is 50 times as long as the pick feed, is selected from the roughness profile as shown in Fig. 5. Then, the correlation coefficients of the segment with different segment yj={y1+j,y2+j,...,yi+j,...,yn+j}, where j=0,1,2…, are calculated as follows:

Ac

ce pt

ed

where and represent mean values of x and yj, respectively. Eq. (1) is normalized by dividing the correlation coefficients by standard deviations of x and yj. Thus, if yj is equal to x (it means perfect repeatability of the feed marks or perfect transferability), the correlation coefficient becomes 1. Fig. 6 shows the correlation coefficients calculated at a pick feed of 3 μm. The correlation coefficient of the diamond coated tool changes periodically and the peak in each period is about 0.5, while that of the TiN coated tool is about 0.1. The correlation coefficient shows similarity between the two segments, and hence it has peaks at the same period as the pick feed. Therefore, the transferability of the cutting tools is evaluated by averaging the 20 local maximum values, where the transferability is quantitated from 0 to 1 and 1 represents perfect transferability (cutting edge profile is perfectly transferred to the finished surface).

Figure 4(a)

Page 7 of 26

ip t cr

ce pt

ed

M

an

Figure 4; Roughness profiles at pick feed of 3 μm. (a) Finished with diamond coated tool. (b) Finished with TiN coated tool.

us

Figure 4(b)

Ac

Figure 5; Definition of {x} and {y}

Figure 6; Correlation coefficients at pick feed of 3 μm.

Page 8 of 26

M

an

us

cr

ip t

2.4. Evaluation of cutting forces Cutting forces are measured and analyzed here to understand cutting mechanics. Cutting forces include two kinds of components, i.e. shearing components due to shearing process at the shear zone and friction process at the rake face, and ploughing components due to ploughing or rubbing process between the flank face and the finished surface. They are seperated by analyzing cutting forces measured at various feed rates. Moreover, friction coefficient μ between the cutting tools and the workpiece is estimated from the ploughing components. Then, those values are utilized to discuss the cutting mechanics in this study. First, another series of planing experiments is conducted for this evaluation under the conditions listed in Table 2. Pick feeds of 2, 4, 6, 8, 10 μm (pick feed of 5μm is also selected to sample the chips) and a large depth of cut of 100 μm are selected, so that length of the cutting edge engaged with the workpiece is almost constant and the uncut chip thickness is changed dominantly. The ploughing components are dependent on the cutting edge length, and thus the ploughing components can be separated from the cuting forces. Fig. 7 shows a typical example of the forces measured at various pick feeds in the elliptical vibration cutting with the diamond coated tool and the cutting fluid. The linear relations shown by the solid lines are identified by the least squares method, and their inclinations represent the shearing components while their intersections with the vertical axis are the ploughing components. In this example, the ploughing components are identified as Fep = 0.644 N, Fef = 1.334 N and Fet = 3.129 N in the principal, feed and thrust directions respectively.

ed

Table 2; Cutting conditions for evaluation of cutting forces

Ac

ce pt

Machining conditions Depth of cut (μm) 100 Pick feed (μm) 2, 4, (5), 6, 8, 10 Cutting speed (m/min) 1 Cutting fluid Oil mist

Page 9 of 26

Figure 7; Identification of ploughing components and shearing components from cutting forces measured at various pick feeds in elliptical vibration cutting with diamond coated tool and cutting fluid.

cr

ip t

Fig. 8 illustrates the ploughing components Fep, Fef, and Fet acting on the cutting edge. Since Fep is understood to be friction force between the flank face and the finished surface, friction coefficient μ is estimated as follows:

ed

M

an

us

Note that the total normal force can be approximated by vector composition of Fef and Fet, as the nose radius is much larger than the cutting edge length and variation of normal force direction along the curved cutting edge is small at the selected conditions. For example, the friction coefficient can be estimated as μ = 0.189 by substituting the above-identified ploughing components into Eq. (2). Next, six series of the cutting tests are conducted by corresponding six machining methods shown in Table 3, i.e. elliptical vibration/ordinary wet/dry cutting with diamond/TiN coated tools. Each series of tests is conducted under the conditions listed in Table 2. Note that ordinary cutting with diamond coated tools is not conducted because of rapid tool wear. The chips are sampled at a pick feed of 5 μm, and they are observed with a SEM (Scanning electron microscope, FEI, Quanta400). Surface profiles are measured with the optical surface profiler (Zygo, NewView7300) in these series of experiments, too. Table 3; Various machining methods and their symbols. Elliptical vibration On On On On Off Off

ce pt

Coating of tool Diamond coating Diamond coating TiN coating TiN coating TiN coating TiN coating

Cutting fluid Wet Dry Wet Dry Wet Dry

Ac

Symbol D-On-Wet D-On-Dry T-On-Wet T-On-Dry T-Off-Wet T-Off-Dry

Page 10 of 26

Figure 8; Schematic illustration of ploughing force components acting on cutting edge.

Ac

ce pt

ed

M

an

us

cr

ip t

2.5. Tool life tests In order to investigate influence of the elliptical vibration and the cutting fluid on tool wear, tool life tests are conducted by the various machining methods listed in Table 3, except T-Off-Dry. Different ultraprecision machine tool (Nagase Integrex, NIC-300) and elliptical vibration device (Taga Electric, new type of EL-50Σ) are employed in this series of tests. The elliptical vibration conditions are same as those in Table 1. The workpiece is fixed on the dynamometer (Kistler, 9256C) to measure cutting forces. Experimental setup is shown in Figure 9. Depth of cut, pick feed and cutting speed are fixed to be 100 μm, 5 μm and 1m/min, respectively. The tool life tests are continued until the cutting forces become much larger than their initial values or change suddenly due to possible delamination of coating. After the tool life tests, tool damages are observed with a measuring microscope (Nikon, MM-40/L3U).

Figure 9; Experimental setup for tool life tests.

3. RESULTS AND DISCUSSIONS 3.1. Surface roughness and transferability of cutting edge profiles Fig. 10 shows the surface roughness Rt measured at various pick feeds listed in Table 1. It is found that the surface roughness obtained with the TiN coated tools is lower or better than that obtained with the diamond coated tools at every pick feed. Good quality surfaces with roughness of less than 0.5 μm Rt are achieved with the TiN coated tool at the pick feeds of 3, 5 and 20 μm, while the surface roughness obtained

Page 11 of 26

ip t

with the diamond coated tool is higher than 0.8 μm. The reason for this higher surface roughness is thought to be caused by higher roughness of the cutting edge represented in Fig. 1(b). Fig. 11 shows the correlation coefficients calculated at various pick feeds listed in Table 1. It shows that the transferability of cutting edge profile of the diamond coated tool is always better than that of the TiN coated tool. Especially at the low pick feeds of 3 and 5 μm, the correlation coefficients of the diamond coated tool, 0.54 and 0.75, are 6 and 2.4 times higher than those of the TiN coated tool, 0.09 and 0.31, respectively.

Ac

ce pt

ed

M

an

us

cr

Photographs of finished surfaces taken with the DIC microscopy are shown in Fig. 12. Some burrs are observed at the same spatial periods as pick feeds on the surfaces finished with the diamond coated tool. It is considered that the burrs are formed due to the large cutting edge radius of 16.3 μm shown in Fig. 1(a). On the other hand, irregular scratches and micro asperities are observed on the surfaces finished with the TiN coated tool. This suggests that adhesion is likely to occur sometimes between the flank face and the workpiece material. Fig. 13 shows a photograph of two workpieces finished with the diamond coated tool and the TiN coated tool at a pick feed of 5 μm. Diffraction of light can be observed only on the surface finished with the diamond coated tool, because of regular feed marks resulted from better transferability of cutting edge profile.

Figure 10; Surface roughness Rt measured at various pick feeds.

Page 12 of 26

ip t cr

Ac

ce pt

ed

M

an

us

Figure 11; Correlation coefficients obtained at various pick feeds.

Figure 12; Photographs of finished surfaces at pick feed of 5 μm and 10 μm.

Page 13 of 26

ip t cr us

an

Figure 13; Surfaces of hardened die steel finished with TiN coated tool and diamond coated tool (depth of cut: 10 μm, pick feed: 5 μm).

Ac

ce pt

ed

M

3.2. Ploughing force components and friction coefficient of cutting tools Fig. 14 shows measured cutting forces at a typical pick feed of 6 μm. The results show that the all cutting forces become small when the elliptical vibration is turned “On”. This is because the dynamometer does not respond to the dynamic cutting forces varying at the ultrasonic frequency and detects the average cutting forces, and because the instantaneous forces are also reduced in the elliptical vibration cutting as described in Section 2.2. Comparing the coatings in the elliptical vibration cutting “On”, the diamond coating “D” results in higher cutting forces, especially in the thrust direction. It is considered that these higher forces are caused by the dull cutting edge shown in Fig. 1(a). The maximum uncut chip thickness is calculated geometrically to be 3.94 μm under the cutting conditions shown in Table 2 (pick feed: 6 μm). This value is smaller than the cutting edge radii of the diamond coated tools (16.3 μm) and the TiN coated tools (5.8 μm). Thus, the actual rake angle of the diamond coated tool had a large negative value, which leads to the large cutting forces, especially in the thrust direction. It should be noted that the dry elliptical vibration cutting with the TiN coated tool “TOn-Dry” results in the smallest cutting forces, especially in the thrust direction. This is an interesting phenomenon. Generally, the all cutting forces become larger in dry cutting than in wet cutting because of no lubrication effect. The reason is discussed in the following sections.

Page 14 of 26

ip t cr us

an

Figure 14; Measured cutting forces at a pick feed of 6 μm.

Ac

ce pt

ed

M

Figure 15 shows the ploughing components extracted from the measured total cutting forces. The ploughing components have a similar tendency to the cutting forces shown in Fig. 14; they decrease when the elliptical vibration is “On”, they further decreases with the TiN coating, and they become smallest in the case of “T-On-Dry”. The main difference is that the ploughing components are relatively large in the thrust and the feed directions, since they are generated by the ploughing or rubbing process between the flank face and the finished surface. Fig. 15 also shows the friction coefficient μ estimated by Eq. (2). It indicates that the diamond coated tool “D” results in low friction coefficient around 0.2 regardless of the cutting fluid. The reason may be the low friction property of the diamond coating itself [10-11]. In contrast, the TiN coated tool has a low friction coefficient of 0.2 only in the wet elliptical vibration cutting “T-On-Wet”, while it has large friction coefficients of greater than 0.56 in the other cases of “T-On-Dry”, “T-Off-Wet” and “T-Off-Dry”. This result suggests that the cutting fluid or its vapor infiltrates into the interface between the flank face and the workpiece, although the period of the elliptical vibration is very short (about 25 μs). In addition, it suggests that the TiN coating basically has a large friction coefficient with the steel workpiece, and that this large friction may correspond to the scratches and the worse transferability discussed in the previous section. The infiltration will not succeed in the whole interface, and the adhesion will occur in some tiny regions, where the scratches will grow up and then disappear leading to the poor transferability.

Page 15 of 26

ip t cr us

an

Figure 15; Ploughing force components and friction coefficients.

Ac

ce pt

ed

M

SEM photographs of some typical chips are shown in Fig. 16. In order to understand the cutting mechanics, chip width and thickness are measured from those photographs, and the results are plotted in Figs. 17 and 18. They show average values of more than three data. The chip formed by the dry elliptical vibration cutting with the TiN coating “T-On-Dry” is the longest, and the chip thickness of about 3 μm is the smallest accordingly. This corresponds to the above-mentioned interesting phenomenon, i.e. the shearing and the ploughing components decrease considerably as the cutting fluid is turned off and the friction coefficient increases. This must not happen in the ordinary cutting. The reason for this phenomenon can be explained as follows. The chip is pulled up in the elliptical vibration cutting, especially just before the tool is separated from the chip in each vibration cycle, as shown in Chapter 2. It means that the chip-rake friction is reversed in the chip formation process. This reversal leads to the opposite tendency. Unlike the ordinary cutting, the chip-rake friction enhances the chip flow in the elliptical vibration cutting, and hence the higher friction leads to the smaller chip thickness and the smaller cutting forces. Since the diamond coated tool has similar friction coefficient regardless of the cutting fluid, above-mentioned phenomenon, i.e. decrease of chip thickness or decrease of ploughing components when the cutting fluid is turned off, is not observed. The measured chip widths are shown in Fig. 18 with numbers of the tools used for the tests. The theoretical chip width of 291.6 μm is derived geometrically using the nose radius, the depth of cut and the pick feed. The measured widths roughly agree with the theoretical one, and their difference will be caused mainly by form error of the cutting edges and elastic deformation due to the ploughing components. The former error should be the same when using the same cutting tools. Therefore, the difference between “T-On-Wet” and “T-On-Dry” should be caused mainly by the latter elastic deformation due to ploughing components. The reason for this larger width in “T-OnDry” can be explained as follows. As mentioned above, the chip is pulled up strongly in “T-On-Dry”, and the more workpiece material is removed as the chip, i.e. the stagnant

Page 16 of 26

Ac

ce pt

ed

M

an

us

cr

ip t

point [13] moves down and the ploughed material decreases as illustrated in Fig. 20. This leads to wider cutting region and the larger chip width. The considerably low ploughing components of less than 0.6 N (see Fig. 15) also support this explanation. The widths of chips formed by the ordinary cutting, “T-Off-Wet” and “T-Off-Dry”, are relatively large as shown in Fig. 18. As shown in Fig. 16, the whole images of chips generated by “T-Off-Wet” and “T-Off-Dry” show that the width changes in the chip flow direction, and that their average widths may be smaller. This implies that the uncut chip thickness and the width of cutting region change accordingly. Fig. 19 shows profiles of surfaces finished by the six different methods. It shows that the surface profiles in the cutting direction fluctuate largely in “T-Off-Wet” and “T-Off-Dry”. This agrees with the above-mentioned fluctuation of chip width. On the other hand, a mirror quality surface with a small surface roughness of 0.044 μm Rt in the cutting direction is obtained by the dry elliptical vibration cutting with the TiN coating “T-On-Dry”. This corresponds to the small cutting and ploughing forces. The roughness in the cutting direction obtain with the diamond coated tool “D-On-Wet” and “D-On-Dry” are not very different from that obtained by “T-On-Wet”, but the surfaces generated with the diamond coated tool has burrs due to the dull cutting edge.

Page 17 of 26

us

cr

ip t

Figure 16; SEM photographs of chips (depth of cut: 100 μm, pick feed: 5 μm, cutting speed: 1 m/min).

ce pt

ed

M

an

Figure 17; Thickness of chips measured from SEM photographs.

Ac

Figure 18; Width of chips measured from SEM photographs.

Page 18 of 26

ip t cr us an M Ac

ce pt

ed

Figure 19; Surface profile of finished surface (depth of cut: 100 μm, pick feed: 5 μm, cutting speed: 1 m/min).

Figure 20; Schematic illustration of ploughing effect decreased by elliptical vibration cutting.

Page 19 of 26

Ac

ce pt

ed

M

an

us

cr

ip t

3.3. Tool life tests and tool damages Figure 21 shows the cutting forces measured during the tool life tests and plotted against the nominal cutting distance, and Fig. 22 shows photographs of the cutting edges after the tool life tests. As shown in Fig. 21(a), the results of diamond coated tools are similar between the wet and the dry machining up to the nominal cutting distance of 350 m. This agrees with the other results of similar cutting forces and ploughing force components (Figs. 14 and 15), friction coefficients (Fig. 15), chips (Figs. 16-18) and surfaces (Fig. 19). The cutting forces drop suddenly at 350 m in the wet machining and at 650 m in the dry machining, where drastic changes on the cutting edges are thought to be occurred at these points. Fig. 22 indicates that these drops are accompanied by sudden decrease of depth of cut due to delamination of diamond coating. Although the delamination occurred, constant low cutting forces are observed in the wet machining “D-On-Wet”. The delamination in the “D-On-Wet” occurs mainly on the rake face, and the diamond coating on the flank face almost remains as shown in Fig. 22. Thus, the diamond coating on the flank face can work as a new cutting edge after the delamination, since the coating thickness is larger than the uncut chip thickness under the cutting conditions. The constant low cutting forces support this explanation. Comparing the flank wear after the tool life tests (see Fig. 22), the wear in “D-On-Wet” is surprisingly small. These facts indicate that the diamond coated tool has superior wear resistance against the die steel in the elliptical vibration cutting like the single crystalline diamond tools [3-6]. Suppression of the delamination is the next challenge to be solved. The TiN coated tool has fairly good wear resistance only in the wet elliptical vibration cutting “T-On-Wet” as shown in Fig. 21(b), where the cutting forces gradually increase and become large similarly to those in the dull diamond coated tools. The final flank wear becomes much larger than that in “D-On-Wet”, but it is much smaller than that in the ordinary cutting “T-On-Wet” considering the cutting distance. The dry elliptical vibration cutting “T-On-Dry” does not show good performance. The forces increase rapidly around 200 m as shown in Fig. 21(b), and significant tool damage is observed after the cutting of 216 m in Fig. 22. Thus, “T-On-Dry” is not practical although it shows the best performances in the other aspects of the forces (Figs. 14 and 15) and the surface quality (Fig. 19) when the tool is new. Fig. 21(c) shows the cutting forces in the wet ordinary cutting with the TiN coated tool. The forces increase with the nominal cutting distance, and the thrust force rises up from 15 N to 32 N at 100 m. The cutting edge has significant damage after the short cutting distance of 177 m (Fig. 22). The rapid tool wear corresponds to the high friction coefficient of 0.58 (Fig. 15). These all tendencies of wear progress (Figs. 21 and 22) correspond to the friction coefficients (Fig. 15) in the same TiN coated tools, i.e. tool life is long when the friction coefficient is low.

Page 20 of 26

ip t cr us

ce pt

ed

M

an

Figure 21 (a)

Ac

Figure 21 (b)

Page 21 of 26

ip t cr us

Figure 21 (c)

ce pt

ed

M

an

Figure 21; Cutting forces plotted against nominal cutting distance. (a): Diamond coated tools, (b): TiN coated tools (elliptical vibration: On), (c): TiN coated tools (elliptical vibration: Off)

Ac

Figure 22; Photographs of cutting edges after tool life tests (bracketed numbers show nominal cutting distances). To summarize the above results, advantages and disadvantages of the both coated tools as well as the single crystalline diamond tool are summarized in Table 4. First, the TiN coated tool has high friction property with the workpiece, which results in interesting phenomenon; i.e. cutting forces especially thrust force become small in the dry elliptical vibration cutting since the chip is pulled up strongly. This characteristic of small cutting forces can be advantage in machining low stiffness workpieces such as thin wall and long slender workpieces. However, the results of the tool life tests in Section 3.3 suggest that the high friction property accelerates tool wear. Considering practical use with rigid structures, the TiN coated tool should be used in the wet cutting in order to reduce the tool wear. Next, the diamond coated tool in the wet elliptical vibration cutting represents superior potential; i.e. the low friction property which leads to the high transferability, high wear resistance against the die steel. However, surface

Page 22 of 26

roughness is large, and cutting forces are high. Delamination of the coating on the rake face after cutting of several hundred meters is also a minor problem. In summary, no cutting tool is perfect at present. In the future, the authors expect that it may be possible to solve the problems of the diamond coated tool, for example by developing an efficient method to sharpen the cutting edge.

cr

 Cheap  Fairly good surface roughness can be obtained in the wet elliptical vibration cutting  Cutting forces especially thrust force become small in the dry elliptical cutting because of the high friction property

Disadvantages  Bad transferability of cutting edge profile to workpiece  Low wear resistance especially in the dry elliptical vibration cutting

us

Advantages

an

Cutting tool TiN coated tool

ip t

Table 4; Advantages and disadvantages of cutting tools in elliptical vibration cutting of die steel.

 Lower manufacturing cost compared with other diamond tools  Good transferability of cutting edge profile to workpiece  Superior wear resistance against die steel

Single crystalline diamond tool

 Mirror surface finish  Superior wear resistance against die steel

 Higher surface roughness of finished surface  Large cutting forces  Coating delamination on rake face after cutting of several hundred meters  Expensive

ce pt

ed

M

Diamond coated tool

Ac

Finally, the results obtained here are applied to practical machining of an ornamental surface of hardened die steel as a demonstration. Fig. 23 shows the check pattern surface finished by the wet elliptical vibration cutting with the diamond coated tool “D-On-Wet”. The surface is machined in the two perpendicular directions, and the regular feed marks with high transferability generates clear diffraction of light on the finished surface. This cutting technology can be used in dies and molds industry to produce high quality surfaces.

Page 23 of 26

ip t cr us an M ed

ce pt

Figure 23; Photograph of check pattern surface finished by the wet elliptical vibration cutting with diamond coated tool (Stavax with hardness of 54 HRC, pick feed: 10 μm).

4. CONCLUSION

Ac

An experimental study on the elliptical vibration cutting of hardened die steel with diamond coated tools and TiN coated tools is carried out. A method for analyzing transferability of cutting edge profiles to finished surfaces utilizing correlation coefficients of roughness profiles is proposed. Machinability of hardened die steel by the elliptical vibration cutting with the coated tools is evaluated in terms of surface roughness, shearing and ploughing components, friction coefficient between the flank face and the workpiece, chip fomation and tool life as well as proposed correlation coefficient. The following conclusions can be drawn: a) By applying the wet elliptical vibration cutting, surface roughness Rt of less than 0.5 μm is achieved with the TiN coated tool at pick feeds of 3, 5, 20 μm, while surface roughness obtained with the diamond coated tool is higher than 0.8 μm due to high roughness of its flank face and large radius of its cutting edge roundness.

Page 24 of 26

ed

M

an

us

cr

ip t

b) The transferability of the diamond coated tools is higher than that of the TiN coated tools. The transferability values of the diamond coated tools are 2.4 to 6 times higher than those of the TiN coated tools especially at low pick feeds of 3 and 5 μm. The high transferability of the diamond coated tools generates regular feed marks, which lead to clear diffraction of light. c) The friction coefficient between the diamond coated tools and the finished steel surfaces is about 0.2, regardless of the cutting fluid. In contrast, the friction coefficient with the TiN coated tools changes from 0.20 in wet machining to 0.56 in dry machining when the elliptical vibration is applied. This suggests that the cutting fluid or its vapor infiltrates into the interface between the flank face and the finished surface when they separate in each vibration cycle. d) It is found that higher friction coefficient leads to thinner chips, lower cutting forces, lower ploughing components and lower surface roughness in the elliptical vibration cutting, in contrary to the ordinary cutting. This unique phenomennon can be observed when the hardened die steel is machined in the dry elliptical vibration cutting with the TiN coated tools. The reason is explained by reversal of chip-rake friction due to the elliptical vibration. Although a mirror quality surface is obtained, this dry machining is not practical because the high friction coefficient results in the rapid tool wear. e) It is clarified from the tool life tests that the diamond coated tools have superior wear resitance although delamination of the diamond coating occurs especially in the rake face after cutting distance reaches some extent, e.g. 350 m or 650 m.

ce pt

ACKNOWLEDGMENTS

Ac

The diamond coated carbide tools used in the experiments were specially manufactured by OSG Corporation, and the research was supported partially by the Knowledge Hub of Aichi, “Development of Ultra-Precision/High-Functional Processing Technology for Hard-to-Process Materials”. The authors would like to express deepest appreciation for the supports.

REFERENCES

[1] S. Shimada, H. Tanaka, M. Higuchi, T. Yamaguchi, S. Honda, K. Obata, Thermochemical wear mechanism of diamond tool in machining of ferrous metals. Annals of the CIRP 53 (2004) 57-60. [2] Y. Song, K. Nezu, C. Park, T. Moriwaki, Tool wear control in single-crystal diamond cutting of steel by using the ultra-intermittent cutting method, International Journal of Machine Tools and Manufacture 49 (2009) 339-343. [3] E. Shamoto, T. Moriwaki, Study on Elliptical Vibration Cutting, CIRP 43 (1994) 35-38.

Page 25 of 26

Ac

ce pt

ed

M

an

us

cr

ip t

[4] E. Shamoto and T. Moriwaki, Ultraprecision Diamond Cutting of Hardened Steel by Applying Elliptical Vibration Cutting, CIRP 48 (1999) 441-444. [5] E. Shamoto, N. Suzuki, T. Moriwaki, Y. Naoi, Development of Ultrasonic Elliptical Vibration Controller for Elliptical Vibration Cutting, CIRP 51 (2002) 327-330. [6] Y. Wang, N. Suzuki, E. Shamoto, Q. Zhaoa, Investigation of tool wear suppression in ultraprecision diamond machining of die steel, Precision Engineering 35 (2011) 677-685. [7] X. Zhang, A. Senthil Kumar, M. Rahman, C. Nath, K. Liu, Experimental study on ultrasonic elliptical vibration cutting of hardened steel using PCD tools, Journal of Materials Processing Technology 211 (2011) 1701–1709. [8] C. Nath, M. Rahman, K. Soon Neo, Machinability study of tungsten carbide using PCD tools under ultrasonic elliptical vibration cutting, International Journal of Machine Tools and Manufacture 49 (2009) 1089–1095. [9] J. Karner, M. Pedrazzini, I. Reineck, M.E. Sjo¨strand, E. Bergmann, CVD diamond coated cemented carbide cutting tools, Materials Science and Engineering 209 (1996) 405-413. [10] C.D. Torres, P.J. Heaney, A.V. Sumant, M.A. Hamilton, R.W. Carpick, F.E. Pfefferkorn, Analyzing the performance of diamond-coated micro end mills, International Journal of Machine Tools and Manufacture 49 (2009) 599-612. [11] T.H. Huang, C.T. Kuo, T.S. Lin, C.S. Chang, Wear behaviour of various diamondcoated cutting tools under different deposition conditions, Diamond and Related Materials 2 (1993) 928–932. [12] J.P. Dismukes, J.V. Busch, N.V. Nallicheri, K.R. Walton, Economic Assessment of CVD Diamond Synthesis Technologies, Materials Science Monographs 73 (1991) 635–644. [13] Y. Wang, E. Shamoto, Elliptical Vibration Cutting of Hardened Steel with Large Nose Radius Single Crystal Diamond Tool, International Journal of Automation Technology, 8 (2014) 820-826.

Page 26 of 26