Effect of Cutting Conditions on Tool Performance in CBN Hard Turning

Effect of Cutting Conditions on Tool Performance in CBN Hard Turning

Journal of Journal Manufacturing of Manufacturing ProcessesProcesses Vol. 7/No. Vol. 1 7/No. 1 2005 2005 Effect of Cutting Conditions on Tool Perform...

167KB Sizes 8 Downloads 51 Views

Journal of Journal Manufacturing of Manufacturing ProcessesProcesses Vol. 7/No. Vol. 1 7/No. 1 2005 2005

Effect of Cutting Conditions on Tool Performance in CBN Hard Turning Yong Huang, Dept. of Mechanical Engineering, Clemson University, Clemson, South Carolina, USA Steven Y. Liang, Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA

Pa Pt R T

Hardness of abrasive particle Tool hardness Tool nose radius Average temperature along tool-workpiece interface Vc Cutting speed VB Flank wear length w Width of cut ␣ Rake/chamfer angle, taken as a positive value for simplicity ␥ Clearance angle σ Average normal stress along tool-workpiece interface

Abstract Cubic boron nitride (CBN) cutting tools are commonly used for single-point turning of hardened materials. Their performance is of importance for hard turning to be a viable technology in view of the high cost of CBN cutting tools and the cost of downtime for tool change. Based on the calibrated tool wear model in finish turning hardened 52100 bearing steel, tool performance is evaluated as a function of cutting conditions based on the flank wear criterion. A satisfactory match has been reached when comparing model predictions and experimental measurements. Further, an analysis of variance was carried out to investigate statistical significance among cutting conditions, and it shows that cutting speed plays a dominant role in determining the tool performance in terms of tool life, followed by feed and depth of cut, and overall tendencies agree with predictions from the general Taylor tool life equation as well as experimental observations. The proposed methodology in evaluating the tool performance can help to optimize the hard turning process and, eventually, help hard turning to be a viable technology.

Introduction Because hard turning offers possible benefits over traditional form grinding in the context of lower equipment costs, shorter setup time, fewer process steps, greater part geometry flexibility, and the elimination of cutting fluid use, as noted in König et al. (1984), König, Berktold, and Koch (1993), and Tönshoff, Arendt, and Ben Amor (2000), it has received increasing attention as an alternative to the grinding process. Of the presently available cutting tool materials, cubic boron nitride (CBN) is the most commonly used for the finish hard turning process. The ability to predict the CBN tool performance under various cutting conditions is of great importance to the overall process optimization of finish hard turning. In general, because tool flank wear length (wear land) is closely related to part finish, surface integrity, and cutting forces and power, researchers have regarded the flank wear length as the tool life criterion or as an important index to evaluate tool performance, including the case of hard turning (Takatsu, Shimoda, and Otani 1983; Abrao, Wise, and

Keywords: CBN, Cutting Conditions, Tool Performance, Flank Wear, Hard Turning

Nomenclature a Hardness constant d Depth of cut f Feed K Constant K abrasion Process-related dimensionless abrasive wear coefficient K adhesion Process-related adhesive wear coefficient Process-related diffusive wear coefK diff ficient KQ Constant related with activation energy for diffusion n Constant This paper is an original work and has not been previously published except in the Transactions of NAMRI/SME, Vol. 32, 2004.

10

Journal of Manufacturing Processes Vol. 7/No. 1 2005

dVB (cot γ + tan α) R × = dt ⎡⎣VB ( R − VB tan γ ) ⎤⎦

Aspinwall 1995; Dewes and Aspinwall 1996). So, tool performance is evaluated based on the flank wear progression in this study. As is generally accepted, different wear mechanisms coexist during CBN hard turning, and the dominant mechanism may be different under different cutting conditions even using the same tool/ workpiece combination. In hard turning, not only tool geometry and cutting conditions but also the tool and workpiece material properties play strong roles in influencing tool wear mechanisms, as shown by Chou and Barash (1995). The main wear mechanisms in turning hardened steel using CBN tools are considered to be a combination of abrasion, adhesion, and diffusion, as shown by Huang (2002). The objective of this study is to investigate the overall tool performance in terms of tool life as a function of cutting conditions—that is, cutting speed, feed, and depth of cut—based on the flank wear criterion. A methodology is proposed here to study the tool performance in finish turning hardened 52100 bearing steel using a low CBN content tool, which is commonly used as a cutter in hard turning.

⎫ ⎧ ⎛ Pan −1 ⎞ aT K K ⎪ abrasion ⎜ n ⎟ VcVBσ + K adhesion e Vc σ ⎪ ⎪ ⎪ ⎝ Pt ⎠ ⎬ ⎨ KQ ⎪ ⎪ − ⎪⎭ ⎪⎩+ K diff VcVBe T + 273

(1)

The sharp cutting edge wears rapidly at the beginning of its use, which is called the break-in period. This period happens within the first few minutes of cutting and is followed by a uniform steady-state flank wear period. The wear rate during the break-in period is random in nature and is not of interest in this study. The above flank wear rate model [Eq. (1)] is aimed at specifying the flank wear rate in the steady-state wear period after the break-in period. Therefore, it does not accommodate the singularity point for the perfectly sharp tool. In applying this proposed wear rate model, the process information such as average temperature and average normal stress along the tool-workpiece interface is required as the model inputs. Although feed and depth of cut do not explicitly show up in Eq. (1), they are the indispensable information to estimate the required process information for tool wear modeling. The required average temperature and stress information can be determined based on the given cutting conditions, as follows. Given the cutting conditions, namely, cutting speed, depth of cut, and feed, the process information, such as cutting forces, shear angle, and the shear flow stress, for a given tool-workpiece combination can be estimated from Oxley’s predictive machining theory (Oxley 1989) or its modified method (Huang and Liang 2003a). To apply the aforementioned process modeling approach for the three-dimensional oblique hard turning application, the equivalent two-dimensional cutting geometry, including the equivalent cutting edge normal rake angle, α*n , the equivalent inclination angle, i*, and the equivalent side cutting edge angle, Cs* , should be first calculated based on the 3-D oblique cutting geometry transformation model (Oxley 1989; Arsecularatne, Mathew, and Oxley 1995; Arsecularatne and Mathew 2000). As reviewed by Huang (2002), as the tool wears neither the shear angle nor the chip thickness changes noticeably in both conventional cutting and practical hard turn-

Tool Wear Modeling Unified Approach to Modeling Flank Wear Progression Numerous models have been proposed to describe the general volume loss and/or wear rate for different wear mechanisms, including applications in metal cutting, such as some notable work on abrasive wear (Rabinowicz, Dunn, and Russell 1961), adhesive wear (Archard 1953; Shaw and Dirke 1956; Kannatey-Asibu 1985), and diffusive wear (Loladze 1981 and Kannatey-Asibu 1985). Unfortunately, in modeling tool wear, only one or two assumed dominant wear mechanism(s) are documented, as noted in Usui, Shirakashi, and Kitagawa (1978); KannateyAsibu (1985); Kramer and Judd (1985); and Kramer (1986). Huang and Liang (2004) have contributed an effort to predict the flank wear rate in CBN hard turning by modeling the effects of abrasion, adhesion, and diffusion in hard turning as a function of cutting conditions, tool geometry, and material properties, as follows:

11

Journal of Manufacturing Processes Vol. 7/No. 1 2005

ing. Under practical hard turning conditions, the normal stress and the rubbing force due to flank wear are modeled based on the process information of chip formation and the flank wear length as in Huang (2002). Also, average temperature can be calculated based on the process information of the chip formation process, rubbing force, and flank wear length as discussed by Huang and Liang (2003b). For simplicity, the rubbing heat source is considered as a plane heat source in this study of thermal modeling. With this process information, then the tool flank wear rate can be calculated for every specified tool flank wear length as Eq. (1) with known and/or calibrated wear coefficients. Based on the estimated wear rate, the new flank wear length after a time step can be predicted. As a geometric change feedback, the updated flank wear length is used to predict the new average normal stress and new average temperature along the tool-workpiece interface. The iteration goes on until the tool flank wear length reaches its life criterion. Figure 1 shows the flow of the calculation logic. This proposed unified approach is applicable to any types of tools, ranges of conditions, and material systems. It can also be further extended to other machining processes regardless of tool geometries and cutting conditions. The coefficients of the discussed wear rate model [Eq. (1)], Kabrasion, Kadhesion, a, Kdiff, and KQ, need to be known or calibrated based on experimental data, and they change with different tool-workpiece combinations. The low CBN content tool is commonly adopted in finish hard turning. For simplicity, the model here considers the case where there is only one kind of abrasive particle existing, which is cementite in hard turning when using the low CBN content tool.

Equivalent 2D geometry based on cutting conditions Process modeling (Huang and Liang, 2003a)

Other process constants and calibrated wear coefficients Process info.

Stress model (Huang, 2002)

σ

Temperature model (Huang and Liang, 2003b)

T

Flank wear rate model (Equ. (1))

dVB dt

Update mechanism Updated VB

Figure 1 Unified Approach to Modeling Flank Wear Progression in Hard Turning

mm wide edge chamfer and a 0.8 mm nose radius. The tool holder used was a Kennametal DCLNR164D (ISO DCLNR-164D). No cutting fluid was applied. Flank wear length was measured using an optical microscope (Zygo NewView 200). Under the aforementioned conditions, the calibrated flank wear model can be expressed as follows, which was determined based on the experimental data under conditions 5, 9, and A of Tables 1 and 2 (Huang and Liang 2004): dVB (cot γ + tan α) R × = dt ⎡⎣VB ( R − VB tan γ ) ⎤⎦ ⎫ ⎧ ⎛ Pan −1 ⎞ ⎪ ⎪0.0295K ⎜ n ⎟ VcVBσ ⎝ Pt ⎠ ⎪ ⎪ ⎪ ⎪ −14 9.0313×10 −4 T Vc σ ⎬ ⎨+1.4761 × 10 e ⎪ 20460 ⎪ − ⎪+5.7204 × 106 V VBe T + 273 ⎪ c ⎪ ⎪ ⎭ ⎩

(2)

The units used are in SI for the right side of the above equations and µm/minute for the left side. This calibrated model [Eq. (2)] is further utilized to evaluate the tool performance by varying cutting speed, feed, and depth of cut, respectively, under the tool manufacturer recommended gentle cutting conditions. To verify this proposed modeling approach, several cutting tests were carried out as follows. Cutting tests have been designed and performed based on the test matrix, which is a standard central composite design with an alpha value of 1.414, and the center point (0,0) is determined based on the

Effect of Cutting Conditions on Tool Performance Based on Tool Wear Progression Tool Flank Wear Rate and Experimental Preparation Hardened AISI 52100 bearing steel with a hardness 62 HRC was machined on a horizontal lathe (Hardinge T-42) using a low CBN content tool insert (Kennametal KD050). The geometry of KD050 is specified by Kennametal CNGA-432S0420 (ISO CNGA120408S01020), with a negative 20° and 0.1

12

Journal of Manufacturing Processes Vol. 7/No. 1 2005

Table 1 Test Matrix (standard central composite design)

recommended range from the tool manufacturer, as noted by Dawson (2002). A typical radial depth of cut required to remove the stock allowance was suggested as 0.203 mm, which was used with this test matrix. The test conditions are shown in Table 1. To investigate the effect of depth of cut, the wear progressions were also measured by changing the depth of cut according to the cutting conditions shown in Table 2. Conditions 4, 7, and 11 are the same cutting conditions (with a cutting speed of 2.29 m/s, a feed of 0.114 mm/rev, and a depth of cut of 0.203 mm) in this experimental design; so are conditions 5 and C. The experimental data from cutting tests have been utilized to calibrate and verify the proposed approach in evaluating the tool performance. The experimental data of conditions 5, 9, and A of Tables 1 and 2 were used to calibrate the flank wear rate model [Eq. (2)], and in the following sections the calibrated model is further applied to investigate the tool performance in a much broader range of cutting conditions based on this experimental design. Tool life criteria based on the flank wear length need to be specified in evaluating the tool performance. Based on the experimental tool life measurements when the tools failed under the investigated set of cutting conditions, different tool life criteria (150, 110, and 125 µm) have been chosen for the following investigations. For real applications, these tool life criteria may not be adopted because their main purpose here is for the model evaluation. Due to the size of the experimental data set, no uncertainty characterization is offered here. Because the flank wear rate is not a constant during a flank wear progression, the estimated tool life may be little different due to the simulation time steps selected. By considering the tradeoff between the computational load and the prediction accuracy, the simulation step here is chosen as 0.25 minute, 0.5 minute, or 1 minute depending on the investigated cutting conditions. Based on the experimental observations, the flank wear rate turned out to be uniform after the wear land was larger than 10 µm, which is picked up as a sign of the end of break-in period for simplicity. Tool flank wear length after the breakin period is assumed as 10 µm in this study. The effects of cutting speed, feed, and depth of cut on tool performance have been investigated, re-

Condition Index 1 2 3 4 5 6 7 8 9 10 11

Speed (m/s) 3.05 1.52 3.05 2.29 1.52 3.36 2.29 2.29 2.29 1.21 2.29

Feed (mm/rev) 0.152 0.152 0.076 0.114 0.076 0.114 0.114 0.061 0.168 0.114 0.114

Depth of Cut (mm) 0.203 0.203 0.203 0.203 0.203 0.203 0.203 0.203 0.203 0.203 0.203

Table 2 Test Matrix with Respect to Depth of Cut (Condition C here is same as Condition 5 in Table 1)

Condition Index A B C

Feed (mm/rev) 0.076 0.076 0.076

Speed (m/s) 1.52 1.52 1.52

Depth of Cut (mm) 0.102 0.152 0.203

spectively, and compared with the experimental investigation in the following sections. Effect of Cutting Speed By specifying a typical feed (0.114 mm/rev) and depth of cut (0.203 mm), the relationship between cutting speed and tool performance in terms of tool life is plotted in Figure 2. Here the tool life criterion is specified as 150 µm and the simulation step is 1 minute. Conditions 4, 7, and 11 are the same cutting conditions (cutting speed = 2.29 m/s) and the measured wear rates under these conditions were repeatable (Huang 2002), so only one data point is utilized here. Because early microchipping was observed under condition 6 (cutting speed = 3.36 m/s), the associated wear progression measurements are not utilized in evaluating the proposed methodology here. It can be seen from this figure that cutting speed plays a very important role in determining tool performance. A very satisfactory match between the simulation results and the experimental measurements has been observed. Effect of Feed By specifying a typical cutting speed (2.29 m/s) and depth of cut (0.203 mm), the relationship between feed and tool performance in terms of tool

13

Journal of Manufacturing Processes Vol. 7/No. 1 2005

50

45 Simulation result Experimental data

40

40

30

Tool life (min)

Tool life (min)

35

25 20

1.4

1.6

1.8

2 2.2 2.4 Cutting speed (m/s)

2.6

2.8

3

15 0.1

3.2

Figure 2 Effect of Cutting Speed on Tool Life Based on Flank Wear Criterion (feed = 0.114 mm/rev and depth of cut = 0.203 mm)

10 9 8 7 6 5

0.08

0.1

0.12 Feed (mm/rev)

0.14

0.16

0.14

0.16 0.18 Depth of cut (mm)

0.2

0.22

depth of cut and tool performance in terms of tool life is plotted in Figure 4. Here the tool life criterion is specified as 125 µm and the simulation step is 0.25 minute. Although the model simulation shows a general tendency that tool life decreases when depth of cut increases based on the flank wear criterion, there are noticeable errors when comparing the measurements at small depths of cut (0.102 mm and 0.152 mm). The maximum error percentage is 38% when depth of cut is 0.102 mm. This disagreement is mainly attributed to two possible reasons: (1) The model required process information, such as average temperature and average stress along the toolworkpiece, is not properly estimated under such shallow cuts with a large negative rake angle and large nose radius CBN cutting tool (Huang and Liang 2003c). (2) It may also come from the inaccurate wear coefficients calibrated using estimated process information (Huang and Liang 2004). The predictions are expected to be improved by providing accurate process information and wear coefficients.

Simulation result Experimental data

11

0.12

Figure 4 Effect of Depth of Cut on Tool Life Based on Flank Wear Criterion (cutting speed = 1.52 m/s and feed = 0.076 mm/rev).

12

Tool life (min)

30

20

10

4 0.06

35

25

15

5 1.2

Simulation result Experimental data

45

0.18

Figure 3 Effect of Feed on Tool Life Based on Flank Wear Criterion (cutting speed = 2.29 m/s and depth of cut = 0.203 mm)

life is plotted in Figure 3. Here the tool life criterion is specified as 110 µm and the simulation step is 0.5 minute. As expected, it can be seen that tool life decreases when feed increases based on the flank wear criterion. The match between the simulation results and the experimental measurements is satisfactory, and there is a noticeable error when feed is 0.168 mm/rev. The error percentage is 80% under these cutting conditions. This disagreement is attributed to the early CBN tool fracture under such aggressive conditions (cutting speed 2.29 m/s and feed 0.168 mm/rev) because the tool lasted no longer than 5 minutes.

Investigation of Effect of Cutting Conditions Based on Analysis of Variance As discussed before, tool performance is affected by cutting speed, feed, and depth of cut, collectively, as cutting conditions. These cutting conditions (factors) and interacting effects of these conditions have their influence on the tool performance as evaluated based on tool life. To establish statistical significance among cutting conditions, a three-factor (cutting speed, feed, and depth of cut), two-level, and fullfactorial design of experiments, which is shown in

Effect of Depth of Cut By specifying a typical cutting speed (1.52 m/s) and feed (0.076 mm/rev), the relationship between

14

Journal of Manufacturing Processes Vol. 7/No. 1 2005

Table 3 Experimental Design and Tool Life Predictions

Cutting Speed (m/s) 1.21 1.21 1.21 1.21 3.05 3.05 3.05 3.05

Feed (mm/rev) 0.061 0.168 0.061 0.168 0.061 0.168 0.061 0.168

Depth of Cut (mm) 0.102 0.102 0.203 0.203 0.102 0.102 0.203 0.203

Table 4 ANOVA Table for Tool Performance Based on Tool Life Criterion

Predicted Tool Life (min.) 27.5 26.5 27.0 25.0 8.0 5.0 7.0 3.5

Source Cutting speed Feed Depth of cut Error Total

F 1815.00 24.07 5.40

P 0.000 0.008 0.081

Response Surface Response surfaces are usually drawn to further appreciate the effect of cutting conditions on the tool performance in terms of tool life. Because the depth of cut is always specified by the stock allowance, only feed and cutting speed are usually selectable in most finish hard turning cases. Based on the proposed modeling approach (Figure 1) and the calibrated tool wear model [Eq. (2)], a response surface is generated by changing cutting speed and feed within the range of interest. This response surface is shown in Figure 5. Here the tool life criterion is also specified as 110 µm and the simulation time step is specified as 0.5 minute. Again, it can be seen that cutting speed impacts the tool performance more significantly than feed does. Such response surfaces can help designers and workers choose optimal cutting conditions in machining hardened materials.

Conclusions In this study, tool performance has been evaluated in terms of tool life based on the flank wear criterion as a function of cutting conditions, that is, cutting speed, feed, and depth of cut. Based on the discussed approach, tool performance can be reasonably evaluated when adopting the flank wear progression criterion. The output of the ANOVA table shows that cutting speed plays a dominant role in determining the tool performance in CBN hard turning of hardened 52100 bearing steel, followed by feed and depth of cut. This tendency agrees with predictions from the general Taylor tool life equation and typical experimental observations.

1.07

= 172

MS 850.78 11.28 2.53 0.47

where H0 = 60, H is the workpiece hardness expressed in HRC scale, and T is the tool life in minutes. Equation (3) clearly states that the cutting speed is the dominant role in determining the tool life, followed by feed and depth of cut. The comparable results have been reported by Dawson (2002) as well.

Table 3, was conducted. Two levels are selected based on the recommended range of the tool insert manufacturer. The replicate of each run is one. Tool life shown in Table 3 is estimated by the proposed approach as outlined in Figure 1. Here the tool life criterion is specified as 110 µm and the simulation time step is specified as 0.5 minute. The technique of analysis of variance (ANOVA, full balanced) is applied to investigate the main effects from three independent cutting conditions (factors) based on tool life predictions. The resulting ANOVA table is given in Table 4. Degrees of freedom (DF), mean square (MS), F-distribution value (F), and P-value (P) associated with each factor are shown in Table 4. The P-value can be interpreted as the probability that the factor is not significant. It is shown from P-values in Table 4 that if the one-sided confidence interval is set at 99%, both cutting speed and feed have significant effects on tool life, and cutting speed plays a dominant role in determining tool life because the P-value of cutting speed is smaller than those of feed and depth of cut, and correspondingly, the F-value is larger. Although there are noticeable errors in estimating tool life by changing the depth of cut as shown in Figure 4, it is still believed that the effect of depth of cut is least significant due to the associated largest P-value from this ANOVA table. These conclusions agree with typical experimental observations and results from Taylor’s tool life equations proposed in Poulachon, Moisan, and Jawahir (2001) and Dawson (2002) in turning hardened 52100 steel. Poulachon, Moisan, and Jawahir found that tool life can be expressed as follows: ⎛ H ⎞ VcT 0.285 d 0.112 f 0.335 ⎜ ⎟ ⎝ H0 ⎠

DF 1 1 1 4 7

(3)

15

Journal of Manufacturing Processes Vol. 7/No. 1 2005

tion of Mechanical Engineers, Part C, Journal of Mechanical Engg. Science (v217, n11), pp1195-1208. Huang, Y. and Liang, S.Y. (2003c). “Force modeling in shallow cuts with large negative rake angle and large nose radius tools – application to hard turning.” Int’l Journal of Advanced Mfg. Technology (v22, n9-10), pp626-632. Huang, Y. and Liang, S.Y. (2004). “Modeling of CBN tool flank wear progression in finish hard turning.” ASME Journal of Mfg. Science and Engg. (v126, n1), pp98-106. Kannatey-Asibu Jr., E. (1985). “A transport-diffusion equation in metal cutting and its application to analysis of the rate of flank wear.” Journal of Engg. for Industry (v107), pp81-89. König, W.; Hochschule, T.; Komanduri, R.; Schenectady, D.; and Tönshoff, H.K. (1984). “Machining of hard materials.” Annals of the CIRP (v33, n2), pp417-427. König, W.; Berktold, A.; and Koch, K.F. (1993). “Turning vs. grinding.” Annals of the CIRP (v42, n1), pp39-43. Kramer, B.M. and Judd, P.K. (1985). “Computational design of wear coatings.” Journal of Vacuum Science and Technology A (vA3, n6), pp2439-2444. Kramer, B.M. (1986). “Predicted wear resistances of binary carbide coatings.” Journal of Vacuum Science and Technology A (vA4, n6), pp2870-2873. Loladze, T.N. (1981). “Of the theory of diffusion wear.” Annals of the CIRP (v30, n1), pp71-76. Oxley, P.L.B. (1989). Mechanics of Machining, an Analytical Approach to Assessing Machinability. West Sussex, England: Ellis Horwood Ltd. Poulachon, G.; Moisan, A.; and Jawahir, I.S. (2001). “Tool-wear mechanisms in hard turning with polycrystalline cubic boron nitride tools.” Wear (v250), pp576-586. Rabinowicz, E.; Dunn, L.A.; and Russell, P.G. (1961). “A study of abrasive wear under three-body conditions.” Wear (v4), pp345-355. Shaw, M.C. and Dirke, S.O. (1956). “On the wear of cutting tools.” Microtechnic (v10, n4), pp187-193. Takatsu, S.; Shimoda, H.; and Otani, K. (1983). “Effect of CBN content on the cutting performance of polycrystalline CBN tools.” Int’l Journal of Refractory Metals and Hard Materials (v2, n4), pp175-178. Tönshoff, H.K.; Arendt, C.; and Ben Amor, R. (2000). “Cutting of hardened steel.” Annals of the CIRP (v49, n2), pp547-566. Usui, E.; Shirakashi, T.; and Kitagawa, T. (1978). “Analytical prediction of three dimensional cutting process, part 3: cutting temperature and crater wear of carbide tool.” Journal of Engg. for Industry (v100), pp236-243.

Longest tool life at the lowest feedrate and cutting speed investigated

Figure 5 Relationship Between Tool Life and Both Cutting Speed and Feed

The proposed methodology can help improve the state of the art of the cutting condition optimization in hard turning. Eventually it will help hard turning to be a viable technology. This modeling approach can also be further extended to appreciate the tool performance of other machining processes. References Abrao, A.M.; Wise, M.L.H.; and Aspinwall, D.K. (1995). “Tool life and workpiece surface integrity evaluations when machining hardened AISI 52100 steels with conventional ceramic and PCBN tool materials.” SME Technical Paper MR95-159. Dearborn, MI: Society of Manufacturing Engineers. Archard, J.F. (1953). “Contact and rubbing of flat surfaces.” Journal of Applied Physics (v24), pp981-988. Arsecularatne, J.A.; Mathew, P.; and Oxley, P.L.B. (1995). “Prediction of chip flow direction and cutting forces in oblique machining with nose radius tools.” Proc. of the Institution of Mechanical Engineers (v209B), pp305-315. Arsecularatne, J.A. and Mathew, P. (2000). “Oxley modeling approach, its applications and future directions.” Machining Science and Technology (v4, n3), pp363-397. Chou, Y. and Barash, M.M. (1995). “Review on hard turning and CBN cutting tools.” SME Technical Paper MR95-214. Dearborn, MI: Society of Manufacturing Engineers. Dawson, T. (2002). “Machining hardened steel with polycrystalline cubic boron nitride cutting tools.” PhD thesis. Atlanta: Georgia Institute of Technology. Dewes, R.C. and Aspinwall, D.K. (1996). “The use of high speed machining for the manufacture of hardened steel dies.” Transactions of NAMRI/SME (v24). Dearborn, MI: Society of Manufacturing Engineers, pp21-26. Huang, Y. (2002). “Predictive modeling of tool wear rate with application to CBN hard turning.” PhD thesis. Atlanta: Georgia Institute of Technology. Huang, Y. and Liang, S.Y. (2003a). “Cutting forces modeling considering the effect of tool thermal property-application to CBN hard turning.” Int’l Journal of Machine Tools and Manufacture (v43, n3), pp307-315. Huang, Y. and Liang, S.Y. (2003b). “Modeling of cutting temperature distribution under tool flank wear effect.” Proc. of the Institu-

Authors’ Biographies Yong Huang is assistant professor of mechanical engineering at Clemson University. He received his BS degree in mechatronics engineering from Xidian University (China), MS degrees in mechanical engineering from Zhejiang University (China) and the University of Alabama, and his MS in electrical and computer engineering and PhD in mechanical engineering from the Georgia Institute of Technology. His current research interests are manufacturing process modeling, optimization, monitoring, and control. Steven Y. Liang is a professor and Woodruff Faculty Fellow of Mechanical Engineering at the Georgia Institute of Technology. He was the founding director of Georgia Tech’s Precision Machining Research Consortium and currently serves as director of the manufacturing education program and associate director of the Manufacturing Research Center at Georgia Tech. Dr. Liang’s research interests lie in micro/nano scale manufacturing, ultraprecision machining, and diagnostics/prognostics of machinery, and he has authored more than 200 book chapters, archival journal papers, and professional conference articles in these areas.

16