CUTTING EDGE RADIUS EFFECTS ON DIAMOND-COATED TOOLS

Feng Qin and Kevin Chou Mechanical Engineering Department The University of Alabama Tuscaloosa, AL Dustin Noland and Raymond G. Thompson Vista Engineering, Inc. Birmingham, AL

Wangyang Ni Kennametal Inc. Latrobe, PA

KEYWORDS

rate decreases at a higher feed. (3) The combined effects lead to complex wear behavior of diamond-coated tools with different edge radii.

Cutting Force, Deposition Stresses, Diamond-Coated Tools, Edge Radius, Tool Wear

INTRODUCTION

ABSTRACT

Diamond coatings using technologies such as chemical vapor deposition (CVD) have been attempted to replace costly polycrystalline cutting tools in machining of abrasive lightweight materials. It has been shown that coating delamination, often occurred prematurely, is the major failure mode of CVD diamond coated tools (Chou 2005). High thermal and mechanical loading during machining and insufficient coating adhesion result in coating failure, and then the exposed substrate encounters massive deformation and catastrophic wear. Moreover, in CVD processes, the thermal mismatch between the coating and substrate materials generates high residual stresses in the tool. Diamond coatings, with a smaller thermal expansion coefficient, receive a compressive residual stress, on the order of GPa; however, the substrates, generally cobalt (Co)-cemented tungsten carbide (WC), receive a tensile stress. Such a high level of deposition residual stresses may impact coating functions (Kitamura 2003). Moreover, around any geometric changes, such as edges, the local stress fields may be severely altered (Gunnars 1996).

Cutting edge geometry has compound effects on diamond-coated tools. Specifically, the edge radius affects the deposition stresses and machining loads in an opposite way. Therefore, there may exist an edge radius for optimal tool life. Moreover, such an optimal radius may be different from one cutting condition to another, as the edge radius effects on machining loads are dependent upon machining parameters. In this study, tungsten carbide cutting inserts with different edge radii were investigated. Finite element simulations were used to evaluate the interface stresses along the rounded cutting edge. The inserts of different edge radii were diamond coated. The coated tools were further tested in composite machining for tool wear evaluations and cutting forces analysis. The major findings are summarized as follows. (1) The deposition residual stresses decreases noticeably with increasing of the edge radius. (2) Increasing the edge radius will increase cutting forces; however, the increasing

Transactions of NAMRI/SME

653

Volume 37, 2009

While the cutting edge geometry affects the deposition stresses around the tool tip, it also complicates the thermal and mechanical loads imposed during machining. The radius edge can significantly affect the chip formations and cutting processes, especially when the uncut chip thickness is close to or smaller than the edge radius, known as the “size effect” (Schimmel 2002; Ranganath, 2007). Effects of the edge radius on chip formations, friction conditions, and part surface integrity, etc., have been widely studied (Thiele 2000; Tian 2004; Nasr 2007; Fang 2008). Most studies indicated that the normal force increases with the edge radius and is much more sensitive to the edge radius size, compared to the cutting component.

film deposition and during subsequent machining. In this study, WC-Co cutting inserts with different edge radii were investigated. First, finite element (FE) modeling was used to evaluate the interface stresses around the cutting edge at different edge radii. Second, commercial WC-Co cutting inserts were prepared with different edge radii. The inserts were then diamond coated and further tested in composite machining with cutting force analyzed and tool wear evaluated. The goal was to analyze how the cutting edge geometry affects the coating tool wear due to deposition stresses and machining loads at different conditions. DEPOSITION STRESS ANALYSIS

Literature of cutting edge effects on coating tool performance is limited. Bouzakis et al. (2003) studied the wear behavior of physical vapor deposition (PVD) coatings on cemented carbide inserts with various cutting edge radii in milling. The authors claimed that the cutting edge radius increasing can lead to a longer tool life. Rech et al. investigated the effects of edge radius of PVD coated tools upon chip formation and tool stresses in orthogonal cutting of steel. The authors reported that there exists an optimum cutting edge radius that minimizes tool stresses, especially within the coating layer, and prolongs the tool life (Rech 2004). Almeida et al. (2005) investigated edge preparations of diamond-coated tools in machining hard metals. Three types of edge conditions, i.e., up-sharp, chamfer, and hone were tested. The edge conditions were found significant to the machining forces, wear pattern, and tool life, etc. The authors reported that coating delamination occurred first at the honed tool. Hu et al. (2007) investigated stress evolutions, 2D analysis, in a diamond-coated tool from deposition to machining with simplified machining load conditions. It was reported that at a low feed, increasing the edge radius will reduce the maximum circumferential normal stress. However, the edge radius seems to have minor effects at a high feed.

Model and Simulations Diamond-coated cutting tools were modeled in CAD software (Pro/Engineer) according to the actual geometry: square-shaped inserts that are o 12.7 mm wide and 3.2 mm thick with an 11 relief angle and a 0.8 mm corner radius. Different edge radii were incorporated into the models. The diamond coating on the substrate has a uniform thickness, 5 µm, at the rake and the relief faces, extending to at about 1.6 mm from the substrate bottom. The CAD models of the tool (substrate and coating) were then imported into FE software, ANSYS, for thermal stress simulations. Due to the symmetry, only a quarter model is needed for analysis. The element used for structural analysis was Solid45. Meshing was generated in the coating first using the default setting and the edge area was refined. The substrate was then meshed using the default setting. Static structural analysis with thermal strains considered was conducted. A uniform deposition temperature of 800°C was set as the initial condition and a room temperature of 25°C was set as the final temperature. Linear-elastic material models independent of temperatures were used for both diamond and WC. The elasticity, Poisson’s ratio, and thermal expansion coefficient of diamond (Heath 1986) and WC (Amirhagni 2001) used were 1200 GPa, 0.07, 2.5 µm/(m·K), and 620 GPa, 0.24, 5.5 µm/(m·K), respectively. The boundary conditions used in the mechanical analysis included two symmetric planes and a fully constrained point at the bottom corner of the quarter model. After the model setup, structural analysis was executed to obtain displacement, strain, and stress data.

It is not clear how the cutting edge affects the machining performance of diamond-coated tools. Currently, the fabrication of coated tools uses off-the-shelf substrates and the tool edge geometry has not been integrated into the coating tool design. To effectively use diamondcoated tools, it is necessary to understand the combined effects of the edge radius due to the

Transactions of NAMRI/SME

654

Volume 37, 2009

Figure 2 compares the interface stresses around the edge for different edge radii. The abscissa in Figure 2 is the normalized distance (by the arc length of the rounded edge) from 0 to 1, where 0 is the beginning of edge rounding at rake and 1 is the end of rounding at the relief surface. First, high stress concentrations due to the edge sharpness can be noted. For the radial normal stress (σr), the maximum reduces from 0.93 GPa for 5 µm to 0.2 GPa for 65 µm edge radius, Table 1. The large edge radius also gives smooth stress gradients along the edge. On the other hand, the circumferential normal stress (σθ) is in a close range, though the sharp edge tends to have a large gradient. Also noted is that the minimum stress increases then decreases with increasing the edge radius.

Further, the stress data along the interface was extracted and then transformed into the local polar coordinate to evaluate the interface stresses along the cutting edge, including three components: the radial normal stress (σr), the circumferential normal stress (σθ), and the shear stress (τrθ) that are a function of the location. The procedures were detailed in Renaud et al. (2008). Results Figure 1 shows the stress contours (σz component, one axis parallel to the rake) in diamond-coated inserts with different edge radii. It can be noted that the stress in the coating, away from the edge is about 3.0 GPa in compression, which is consistent with the biaxial stress analysis. Moreover, the stress distribution around the tool edge shows considerable stress concentration for the 5 µm radius tool.

(a)

σr (GPa)

1

(a) z

5 µm 15 µm 30 µm 65 µm

(0.1 GPa)

0.5

x

z y 0 0

(b)

0.2

0.4 0.6 0.8 Normalized distance

1

0 0

0.2

0.4

0.6

0.8

1

-0.5 5 µm 15 µm 30 µm 65 µm

-1

σθ (GPa)

(b)

-1.5 -2 -2.5 -3 Normalized distance

FIGURE 2. EDGE RADIUS EFFECTS ON INTERFACE STRESSES: (a) σr AND (b) σθ. TABLE 1. MAXIMUM AND MINIMUM INTERFACE STRESSES AT DIFFERENT EDGE RADII (UNIT: GPa).

FIGURE 1. EXAMPLES OF STRESS CONTOURS (σz) AROUND CUTTING EDGE: (a) 5 µm AND (b) 65 µm EDGE RADIUS.

Max σr Min σθ

Transactions of NAMRI/SME

655

5 µm 0.93 -2.84

15 µm 0.59 -2.58

30 µm 0.35 -2.77

65 µm 0.2 -2.89

Volume 37, 2009

MACHINING INVESTIGATION

methane, was inserted to the gas mixture to obtain nanostructures by preventing cellular growth. The pressure was about 90 Torr and the substrate temperature was about 800°C. The coated inserts were further inspected by the interferometer to measure the edge radius and to estimate the coating thickness. Figure 4 is examples of cutting edge images of coated tools. The coating thickness was estimated between 5 and 8 µm. Surface roughness, Ra, was about 0.5 µm for coated tools.

Experimental Details The substrates used for diamond coating experiments were also square-shaped inserts (SPG422) as in the deposition stress analysis. The insert material selected was fine-grain WC with 6 wt.% cobalt (K68 from Kennametal). Four levels of edge radii were evaluated: nominally, 5 µm, 15 µm, 30 µm, and 65 µm. The edge radius of cutting inserts prior to coating was measured by a white-light interferometer, NT1100 from Veeco Metrology. Measurement results indicate that the edge radii were: averagely, 3.79 µm, 13.7 µm, 29.8 µm, and 66.4 µm. Figure 3 shows examples of cutting edge images from the interferometer. Surface textures of the inserts were also assessed by the interferometer with surface roughness analyzed. It was shown that the surface roughness of inserts was in a similar range, 0.29 µm to 0.32 µm of Ra.

(a)

(a) (b)

(b) FIGURE 4. CUTTING EDGE IMAGES OF COATED TOOLS: (a) 5 µm AND (b) 65 µm.

A computer numerical control lathe, Hardinge Cobra 42, was used to perform machining experiments, outer diameter turning, to evaluate the tool wear of diamond-coated tools. With the tool holder used (CSRNL-164D), the diamondo coated cutting inserts formed a 0 rake angle, o o 11 relief angle, and 75 lead angle. The workpieces were round bars made of A359/SiC20p composite. Two machining conditions were used; one is 4 m/s and 0.05 mm/rev, and the other is 1.3 m/s and 0.15 mm/rev. The depth of cut was set at 1 mm. Machining was conducted at room temperature without coolant. For each machining condition, two tests were repeated. During machining testing, the cutting inserts were periodically inspected by optical

FIGURE 3. CUTTING EDGE IMAGES: (a) 5 µm AND (b) 65 µm EDGE RADIUS.

For the coating process, diamond films were deposited using a high-power microwave plasma-assisted CVD process. A gas mixture of methane in hydrogen was used as the feedstock gas. Nitrogen, maintained at a certain ratio to

Transactions of NAMRI/SME

656

Volume 37, 2009

microscopy to measure flank wear-land. Worn tools after testing were also examined by scanning electron microscopy (SEM). In addition, cutting forces were monitored during machining using a Kistler dynamometer.

Force (N)

170

Force (N)

70

-30

0

10

20

30

40

50

60

70

80

Time (s)

(b)

220

Fr Ft Fa

Force (N)

170

120

70

20

-30

0

10

20

30

40

50

60

70

80

Time (s)

FIGURE 6. CUTTING FORCES AT 1.3 M/S AND 0.15 MM/REV FOR (a) 5 µm AND (b) 65 µm EDGE RADIUS.

Fr Ft Fa

100

Figure 7 compares cutting force increases with the edge radius at two different machining conditions. The force ratios were obtained by normalizing with the forces from the 5 µm radius tools. It is noted that (1) the normal force, Fn (resultant of Fr and Fa) show a higher rate compared to the cutting force (Ft), and (2) the increasing rate is mush greater at a small feed.

70

40

10 0

120

20

Figure 5 shows cutting forces during initial cutting (first pass) at 4 m/s and 0.05 mm/rev for 2 different edge radii. The force values of all 3 components: tangential (Ft), radial (Fr) and axial (Fa) were reasonably steady during the entire pass. For the 65 µm radius tool, cutting forces were higher than those from the sharp tool, in particular, the radial component. Figure 6, on the other hand, shows cutting forces at 1.3 m/s and 0.15 mm/rev for different edge radii. The force increasing due to the edge hone is less than that in the 4 m/s and 0.05 mm/rev condition. It can be noted that for the sharp tools, cutting forces had a step increase during cutting. This may be caused by the high deposition stresses combined with the high mechanical load. (a)

Fr Ft Fa

Cutting Forces

130

(a)

220

10

20

30

40

50

60

70

80

-20 Time (s)

130

2

(b)

Fr Ft

Force ratio

Force (N)

1.5

Fa

100

70

Fn (4, 0.05) Fn (1.3, 0.15) Ft (4, 0.05) Ft (1.3, 0.15)

1

40

0.5

10

0 5 0

10

20

30

40

50

60

70

80

15

30

65

Edge radius (µm)

-20 Time (s)

FIGURE 5. CUTTING FORCES AT 4 m/s AND 0.05 MM/REV FOR (a) 5 µm AND (b) 65 µm EDGE RADIUS.

Transactions of NAMRI/SME

FIGURE 7. CUTTING FORCE RATIO COMPARISONS (NORMALIZED BY FORCE AT 5 µm); PARAMETER PAIR DENOTES CUTTING SPEED (M/S) AND FEED (MM/REV).

657

Volume 37, 2009

Tool Wear

It is also noted that for the sharp tools (5 µm and 15 µm), the tool life at 1.3 m/s and 0.15 mm/rev is shorter than that at 4 m/s and 0.05 mm/rev, which is consistent with observations from the previous studies because the mechanical effect seems to be more dominant to delamination onset. However, for the large hone tools (30 µm and 65 µm), the tool life at 1.3 m/s and 0.15 mm/rev is longer than that at 4 m/s and 0.05 mm/rev, contradictory to the sharp tool cases, implying that the alleviation of deposition stresses by the rounded edge may outweigh the added machining loads due to the enlarged edge radius.

Figure 8 shows tool wear, flank wear-land width (VB), along cutting time at the 4 m/s and 0.05 mm/rev condition for different edge radii. Results of two replicates are shown. In general, the tools showed a gradual increase of tool wear followed by an abrupt increase of wear-land in one or two passes. It is believed that, during that specific passes, coating delamination occurred and resulted in rapid wear of the exposed substrate material. Tool wear and the onset of coating delamination (abrupt wear increase) are dependent on the edge radius. The tools with 5 µm and 15 µm substrate edge radii have similar wear growth curves. 65 µm radius tools show slightly greater wear resistance and delay of delamination-induced catastrophic wear. Surprisingly, the 30 µm radius tools result in the poorest tool life.

1.2

VB (mm)

0.9

2

5 15 30 65 S

VB (mm)

1.5

5 15 30 65 S

µm µm µm µm i

µm µm µm µm

0.6

0.3

0 0

1

3

6

9

12

15

18

21

24

Cutting Time (min)

FIGURE 9. TOOL WEAR DEVELOPMENT AT 1.3 M/S AND 0.15 MM/REV.

0.5

Figure 10 shows examples of worn tool images (from SEM) of two different edge radii after machining testing. Flank wear-land is the major wear pattern. Moreover, inserts with both large and small edge radii show similar wear features, a large wear-land once coating being delaminated and substantial metal deposits for the high feed condition.

0 0

3

6

9

12

Cutting Time (min)

FIGURE 8. TOOL WEAR DEVELOPMENT AT 4 M/S AND 0.05 MM/REV.

On the other hand, Figure 9 shows flank wear-land width (VB) vs. cutting time at the 1.3 m/s and 0.15 mm/rev condition for different edge radii. The tools with 5 µm and 15 µm substrate edge radii have a rather rapid linear wear growth and the shortest tool life. The 5 µm tool also shows a high initial wear which might be caused by the high deposition stress and result in significant force increasing (Figure 6). The 30 µm tools show a somewhat better tool life, but most strikingly, 65 µm radius tools demonstrate significantly delay of abrupt wear. A general trend can be noted; the larger the edge radius, the better the wear resistance among the edge radii tested. Using 0.5 mm VB as the life criterion, 65 µm radius tools have an average of ~20 min tool life vs. ~3 min for 5 µm radius tools.

Transactions of NAMRI/SME

(a)

658

Volume 37, 2009

(b)

m/s and 0.05 mm/rev. At 1.3 m/s and 0.15 mm/rev, Ra is 1.39 µm vs. 1.16 µm, in average, for 5 µm and 65 µm tools, respectively.

CONCLUSIONS In this study, WC-Co cutting inserts were investigated into the edge radius effects on the deposition residual stresses by numerical methods as well as on cutting forces and tool wear by experiments. Finite element simulations were used to evaluate the interface stresses along the cutting edge at different edge radii. In addition, commercial WC-Co inserts were prepared with different edge radii. The inserts were diamond coated, under identical conditions, with a thickness of 5 to 8 µm. The coated tools with different substrate edge radii were further tested in machining of composite bars. Two different cutting conditions: (4 m/s and 0.05 mm/rev) and (1.3 m/s and 0.15 mm/rev), were tested. Machining forces were monitored and analyzed against the edge radius and tool flank-wear was measured and evaluated and wear features were examined by SEM.

(c)

The major findings can be summarized as follows. (1) The deposition residual stresses, primarily the radial normal component (σr), decreases significantly with the increase of the edge hone. The maximum reduces from 0.93 GPa for 5 µm to 0.2 GPa for 65 µm edge radius. (2) Increasing the edge radius will increase cutting forces, mainly the radial and axial components; moreover, the increasing rate decreases at a higher feed. (3) The combined effects above result in complex wear behavior of diamond-coated tools with different edge radii. In particular, at the 1.3 m/s and 0.15 mm/rev condition, a 65 µm hone gives a tool life over 5 times longer than 5 µm sharp tools, though tools of either radii have a similar tool wear results at the 4 m/s and 0.05 mm/rev condition.

(d)

FIGURE 10. WORN TOOL SEM IMAGES AT 4 M/S AND 0.05 MM/REV: (a) 5 µm AND (b) 65 µm EDGE RADIUS, AND AT 1.3 M/S AND 0.15 MM/REV: (c) 5 µm AND (d) 65 µm EDGE RADIUS.

ACKNOWLEDGMENTS This research is supported by NSF, Grant #: 0728228. Veeco Metrology provides partial funding for the equipment support. C. Ibrahim and A. Klasterka conducted tool edge radius measurements.

Part surface finish produced by different edge-radius tools, at the first cutting pass, was also measured by a stylus profilometer. The results show a similar surface roughness range between different edge radii: averagely, 0.85 µm and 0.99 µm of Ra for 5 µm and 65 µm tools at 4

Transactions of NAMRI/SME

659

Volume 37, 2009

REFERENCES

Nasr, M.N.A., E.-G. Ng, and M.A. Elbestawi (2007). “Modelling the Effects of Tool-edge Radius on Residual Stresses when Orthogonal Cutting AISI 316L,” International Journal of Machine Tools and Manufacture, Vol. 47(2), pp. 401-411.

Almeida, F.A., F.J. Oliveira, M. Sousa, A.J.S. Fernandes, J. Sacramento, and R.F. Silva (2005). “Machining Hardmetal with CVD Diamond Direct Coated Ceramic Tools: Effect of Tool Edge Geometry,” Diamond and Related Materials, Vol. 14(3-7), pp. 651-656.

Ranganath, S., A.B. Campbell, and D.W. Gorkiewicz (2007). “A Model to Calibrate and Predict Forces in Machining with Honed Cutting Tool or Inserts,” International Journal of Machine Tools and Manufacture, Vol. 47(5), pp. 820-840.

Amirhaghi, S., H.S. Reehal, R.J.K. Wood, and D.W. Wheeler (2001). “Diamond Coatings on Tungsten Carbide and Their Erosive Wear Properties,” Surface and Coatings Technology, Vol. 135, pp. 126-138.

Renaud, A., J. Hu, F. Qin, and Y.K. Chou (2008). "Numerical Simulations of 3D Tool Geometry Effects on Deposition Residual Stresses in Diamond Coated Cutting Tools,” Proceedings of the 2008 International Manufacturing Science and Engineering Conference, Oct. 7-10, 2008, Evanston, IL, 2008, MSEC2008-72204.

Bouzakis, K.-D., N. Michailidis, G. Skordaris, S. Kombogiannis, S. Hadjiyiannis, K. Efstathiou, G. Erkens, S. Rambadt, I. Wirth (2003). “Optimization of the Cutting Edge Roundness and Its Manufacturing Procedures of Cemented Carbide Inserts, to Improve Their Milling Performance after a PVD Coating Deposition,” Surface and Coating Technology, Vol. 163-164, pp.625-630.

Rech, J., Y.-C. Yen, H. Hamdi, T. Altan, K.D. Bouzakis (2004). “Influence of Cutting Edge Radius of Coated Tool in Orthogonal Cutting of Alloy Steel.” Materials Processing and Design: Modeling, Simulation and Applications, NUMIFORM 2004, pp. 1402-1407.

Chou, Y.K. and J. Liu (2005). "CVD Diamond Tool Performance in Composite Machining," Surface and Coatings Technology, Vol. 200, pp. 1872-1878.

Schimmel, R.J., W.J. Endres, and R. Stevenson (2002). “Application of an Internally Consistent Material Model to Determine the Effect of Tool Edge Geometry in Orthogonal Machining,” Transactions of ASME, Journal of Manufacturing Science and Engineering, Vol. 124, pp. 536-543.

Fang, N. and L.S. Xiong (2008). "Determination of Friction and Material-flow Boundary Condition on the Tool Round Cutting Edge," Transactions of NAMRI/SME, Vol. 36, pp. 413-420. Gunnars, J. and A. Alahelisten (1996). “Thermal Stresses in Diamond Coatings and Their Influence on Coating Wear and Failure,” Surface and Coatings Technology, Vol. 80(3), pp. 303312.

Thiele, J.D., S.N. Melkote, R.A. Peascoe, and T.R. Watkins (2000). “Effect of Cutting-edge Geometry and Workpiece Hardness on Surface Residual Stresses in Finish Hard Turning of AISI 52100 Steel,” Transactions of ASME, Journal of Manufacturing Science and Engineering, Vol. 122, pp. 642-649.

Heath, P.J. (1986). “Properties and Uses of Amborite,” Industrial Diamond Review, Vol. 46, pp. 120-127.

Tian, Y. and Y.C. Shin (2004). “Finite Element Modeling of Machining of 1020 Steel Including the Effect of Round Cutting Edge,” Transactions of NAMRI/SME, Vol. 32, pp. 111-118.

Hu, J., Y.K. Chou, and R.G. Thompson (2007). "On Stress Analysis of Diamond Coating Cutting Tools,” Transactions of NAMRI/SME, Vol. 35, pp. 177-184.

Yen, Y.-C., A. Jain, and T. Altan (2004). “A Finite Element Analysis of Orthogonal Machining Using Different Tool Edge Geometries,” Journal of Materials Processing Technology, Vol. 146(1), pp. 72-81.

Kitamura, T., H. Hirakata, and T. Itsuji (2003). “Effect of Residual Stress on Delamination from Interface Edge between Nano-film,” Engineering Fracture Mechanics, Vol. 70(15), pp. 2089-2101.

Transactions of NAMRI/SME

660

Volume 37, 2009