International Journal of Machine Tools & Manufacture 41 (2001) 22712285

Friction and cutting forces in cryogenic machining of Ti6Al4V

Abstract Conventional cutting uid serves both as a coolant and lubricant. In cryogenic machining, liquid nitrogen (LN2) is recognized as an effective coolant due to its low temperature; however, its lubrication properties are not well known. The focus of this study was to investigate how the friction between the chip and the tool is affected by focused jetting LN2 to the cutting point in machining Ti6Al4V. Results of cutting force measurements indicated that the cold strengthening of titanium material increased the cutting force in cryogenic machining, but lower friction reduced the feed force. A mathematical model was developed to convert the measured 3D forces in oblique cutting into the normal and frictional force components on the tool rake face, and then to calculate the effective friction coefcient. It was found that the friction coefcient on the toolchip interface was considerably reduced in cryogenic machining. Increased shear angle and decreased thickness of the secondary deformation zone, ndings from a chip microstructure study, offer further evidence that friction is reduced. 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Lubrication; Cryogenic machining; Friction coefcient; Titanium machining; Cutting force

1. Introduction Liquid nitrogen (LN2) as a cryogenic coolant for machining titanium or its alloys has been investigated widely and proven to be capable of signicantly improving tool life [17]. Generally, this improved machinability is attributed to a reduced diffusion-related wear mechanism and less degraded tool hardness or a reduced yielding strength at lower temperatures. Never before has LN2 been considered as a lubricant in the metal cutting process. Typically when referring to lubricants, people immediately think about oil, a substance that has high viscosity and sticks to
* Corresponding author. Tel.: +1-212-854-2957; fax: +1-212-954-3304. E-mail address: [email protected] (S.Y. Hong).

0890-6955/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 8 9 0 - 6 9 5 5 ( 0 1 ) 0 0 0 2 9 - 3

2272

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

the metals surface to form a boundary layer to reduce friction. Unlike oil, LN2 has low viscosity, does not adhere to metal surfaces easily due to its non-wetting tendency, and evaporates quickly into the air. These qualities suggest that LN2 would not be a good lubricant. However, the authors believe that if used wisely, LN2 can also effectively lubricate. The objective of this study is to investigate the effects of LN2 on friction at the toolchip interface in cryogenic machining. The material used in this study was Ti6Al4V, a typical titanium alloy widely used in the aerospace and defense industry and well-known as a difcult-to-machine material. This highly abrasive material quickly wears the tool. Because LN2 has different properties than the conventional cutting uid, the method of applying it to the cutting process must be adjusted to gain the maximum benet from LN2. In this study, the economical cryogenic machining approach [6,8] was adopted for the experimental setup. This approach features a LN2 delivery nozzle system that sprays LN2 only to a localized zone of the tool rake and/or the tool ank in well-controlled jets, instead of ooding the general cutting area. Many reasons support the decision to adopt this approach. Most types of tool wear in machining Ti64, e.g. diffusive wear, abrasive wear and adhesive wear, result from heat generated in the cutting process. This heat results from shearing on the shear plane and friction on the toolchip interface, which dramatically elevate the temperature. However, this intense temperature change is localized to only a millimeter area of the tool faces. Therefore, to achieve an adequate cooling effect with minimized coolant owrate, it sufces to apply the coolant to the localized tool faces with well-controlled and focused jets, instead of ooding the general cutting area. Minimizing coolant consumption by applying it selectively is important economically because, unlike the conventional cooling uids, LN2 cannot be circulated inside the machine tool. As LN2 is released into normal atmospheric pressure and absorbs heat during the cutting process, it quickly evaporates. On the other side, to maximize the cooling effect, it is essential to position the LN2 nozzle as closely as possible to the cutting edge for the LN2 to reach and cool the points of heat generation. To effectively lubricate, LN2 should be injected to the friction areas between the chip and rake face and between the tool and the newly machined surface. Frequently, a liquid or gas cushion will form between the chip and the tool by injecting LN2 to the chiptool interface area, enhancing the cooling effect. With extremely high pressure in the cutting zone, it is impossible to have a gas/uid cushion that would fully separate the chiptool contact in the sticking zone. However, the normal stress in the sliding zone tends to drop to zero when the chip leaves the tool face. Using a chipbreaker to lift the chip helps the LN2 reach the chiptool interface, further enhancing the LN2s lubrication effect. Therefore, in our experimental design, the LN2 delivery nozzle was built in with the chipbreaker. Generally, lubricants are broadly dened as materials which are used for one or more of the following purposes: to reduce friction, to prevent wear, to prevent adhesion, to aid in distributing the load, to cool the moving elements, and to prevent corrosion. LN2 was already known for its capacity to cool and to help reduce tool wear in cryogenic machining. In this study, reduction of the coefcient of friction was used as evidence that LN2 is a good lubricant in cryogenic machining. Findings from the study of the chip microstructure (e.g. the formation and thickness of the secondary deformation zone (SDZ) is a result of the toolchip friction) offered helpful clues about friction.

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2273

2. Mathematical model for toolchip friction evaluation To evaluate the effective friction coefcient on the toolchip interface, the normal and frictional forces on the tool rake must be determined from measured cutting forces. The cutting force components are usually measured experimentally by a three-dimensional dynamometer in directions of cutting, feeding and thrusting for turning operations. For an orthogonal cutting, the normal and frictional forces can be calculated from the cutting force measurements in terms of the rake angle in a two-dimensional hodograph [9]. Yet, in industry, most cutting processes are performed in oblique tool congurations, characterized by inclining the angle of the major cutting edge and by simultaneous engagement of both the major cutting edge and tool nose radius in the process. For an oblique cutting, a complicated three-dimensional geometrical transformation must be formulated to calculate the normal and frictional forces from the measured cutting force components. This is usually done by introducing an effective rake angle that is dened in the plane containing the surface cutting velocity and the chip ow direction [10]. Unfortunately, engaging of the tool nose radius in the cutting process may cause the chip ow direction to vary with the depth of cut. Therefore, most toolmakers in practice do not use the effective rake angle to specify the tool geometry. Fig. 1 shows a three-dimensional hodograph formulated for evaluating the normal and frictional forces of an outside diameter turning operation, where x, y and z coordinates are dened in the direction of feed, cutting and thrust, respectively. In the practice of geometric tool specications adopted by most tool makers for carbide tool inserts, tool nose angle d is given, and the inclination angles i and i for the major and minor cutting edges are specied and dened in the yz and x y planes, respectively. An angle q, made between the projection of the major cutting edge on the xz plane and the direction of feed is given to specify the mounting of the tool insert on the tool holder.

Fig. 1. Hodograph for friction force evaluation.

2274

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

In the following formulation, it is assumed that the direction of the resultant friction on the tool rake is coincident with that of the chip ow and there are sufcient clearance angles for both the major and minor tool anks; therefore, the friction on the tool anks can be ignored. This assumption holds true when the tool is fresh and has sharp edges (i.e. without any wear). In ABD, AOD and AOB, we have ABADtan i (ABD) AOAD/sin q (AOD) ABAOtan h (AOB) From which the effective inclination angle for the major cutting edge, h, can be obtained as htan1(tan isin q) (1) Similarly, from OTS, ORQ and TSRQ, we obtain the effective inclination angle for the minor cutting edge, h, as follows htan1(tan i cos q) (2) where q is the angle made by the direction of feed and the projection of the minor cutting edge on the xz plane. The unit normal, eN, to the rake BOQ can be determined from a vector multiplication of the unit vectors eOB and eOQ, which are dened on the major and minor cutting edges, as eNeOBeOQ eOB{cos h cos q, sin h, cos h sin q}T eOQ{cos h cos q, sin h, cos h sin q} eLMeNeOB
T

(3)

Inside the rake face BOQ, the unit normal eLM to the major cutting edge OB can be written as (4)

This makes a local coordinate system eNeOBeLM with eN and eOB. The chip ow direction can be represented in eNeOBeLM as elocal={0, sin b, cos b}T, where b is the chip ow angle. b is not equal to the effective inclination angle i of the major cutting edge because the tool nose is also engaged in cutting. In our research, the chip ow angle b was measured by inspecting the wear marks on the tool rake. Obviously, the unit vector eF in the chip ow direction can be calculated in xyz from elocal by the following coordinate transformation eF([eN,eOB,eLM]T)1elocal (5)

Therefore, with the unit vectors of the rake normal and chip ow direction available, the resultant normal force N and friction force F on the tool rake can be determined as NPeN FPeF P{FF,FC,FT} (6)

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2275

where FF, FC, FT are the cutting force components in the direction of feeding, cutting and thrusting, respectively, which were measured from a three-dimensional dynamometer in the current research. The effective friction coefcient on the tool rake is dened as m=F/N.

3. Experimental study 3.1. Setup for LN2 cooling A schematic LN2 delivery system developed for turning operations is shown in Fig. 2. The nozzle allows the LN2 jets to be applied selectively to the tool rake face or to the ank face, or simultaneously to both. These two tool faces are the areas where most crater wear and ank wear occur, respectively. Integrating the nozzle into the design of the chipbreaker helps to lift the formed chip from the tool rake face so that the LN2 jet can reach the toolchip interface. The LN2 jet applied at a high pressure may form a liquid or gas cushion on the toolchip interface, enhancing its effectiveness as a lubricant. Fig. 3 illustrates the LN2 delivery nozzle assembly for turning operations, with the tool holder omitted for clarity. Fig. 4 is a photo of the tool/nozzle assembly. By adjusting the position of the supply head on the chipbreaker, either one nozzle (the primary nozzle) or two nozzles (both the primary and secondary nozzle) can be activated to apply LN2 to the tool rake only or simultaneously to the tool rake and ank. Thermally insulated tubing, made of low thermal conductivity stainless steel, and jacketed in a high vacuum, was used to transfer the LN2 from the delivery tank into the inlet of the supply head. A cryogenic turbine ow meter was used to monitor the volumetric LN2 ow rate. The volumetric LN2 ow rate was 0.625 l/min for rake cooling (using the primary nozzle), 0.53 l/min for

Fig. 2. Schematic of nozzle for localized LN2 delivery.

2276

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

Fig. 3.

Assembly of two-nozzle LN2 delivery system.

Fig. 4. Photo of the nozzles and tool assembly.

ank cooling (using the secondary nozzle), and 0.814 l/min for both rake and ank cooling (two nozzles on).

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2277

3.2. Machining test Cutting tests were performed for both dry and cryogenic cutting of Ti6Al4V on a 30 HP (22.4 kW), slant-bed CNC turning machine, Cincinnati-Milacron Cinturn 1407C. The CNC controller is capable of keeping the cutting speed constant despite the workpiece diameter by automatically adjusting the RPM of the spindle. Typical cutting parameters for machining Ti64 in industry are used: depth of cut, 1.27 mm (0.05 inches); feed, 0.254 mm (0.010 inches) and cutting speeds ranging from 1 m/s, 1.5 m/s, and 2 m/s to 2.5 m/s (200 ft/min, 300 ft/min, 400 ft/min, and 500 ft/min). The cutting tool used for the test was an uncoated CNMA432-K68 insert (equivalent to C3 class or ISO K05-K20, M10-M20), which is commonly used in the US aerospace industry for titanium machining. This rhombic at insert with a tool holder MCLNL-164C (both from Kennametal Inc.) creates a geometrical conguration of the inclination angles of the major cutting edge and the minor cutting edge i=i=5, insert nose angle d=80, and side angle of the major cutting edge q=95 (see Fig. 1). Cutting force components, Fc, FF, and FT in the direction of main cutting, feeding and thrusting, respectively, were measured with fresh tool inserts using a Kistler 3D dynamometer. 3.3. Experiment results 3.3.1. Cutting forces For the cutting speed of 1.5 m/s (300 ft/min), the cutting forces are compared in Fig. 5 among: (1) dry cutting, (2) cutting with a secondary nozzle releasing LN2 to the ank surface of the tool, (3) cutting with primary nozzle with primary nozzle injecting LN2 to between the chip and the tool rake face, and (4) cutting with a combination of primary and secondary nozzles. The studys

Fig. 5. Cutting forces in different cooling condition at 1.5 m/s (300 ft/min).

2278

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

Table 1 Highest temperatures (C) at cutting tool insert [11] Dry cutting Measured Theoretical 865 1072 Flank cooling 447 526 Rake cooling 335 437 Both rake and ank 208 265

ndings show that all cryogenic machining approaches have higher main cutting force and thrust force. These outcomes are consistent with our ndings on cutting temperatures in a previous study [11], summarized in Table 1. The more liquid nitrogen is used the lower the cutting temperature. As our material study on cryogenic machining shows, when the temperature is lower the workpiece material becomes stronger and harder [12], and in turn, the cutting forces increase. However, it is interesting to note that in cryogenic machining the feed force, which is closely related to the frictional force of the chip acting on the tool, decreased. The lowest feed force is for cooling the rake using only the primary nozzle. At lower temperatures, the material became harder and less sticky, and the friction force smaller, especially cooling the rake by releasing LN2 to the chip and tool interface area. Under these conditions, LN2 may form a gas and liquid cushion enhancing its lubrication effect. The feed force reduction indicated that in the direction of chip ow the friction force reduction was more than the cutting force increase. When both nozzles were on, it has the benet of the lubrication cushion effect, therefore it has lower feed force than ank cooling. However, the two-nozzle mode had higher feed force than the primary nozzle mode, because of higher cutting force resulting from lower material temperature, which increased the material strength. In the selection of cooling strategy, cooling only the ank was a poor choice because it did not provide a cushion effect or sufciently cool the cutting tool. It has a high tendency to precooling the workpiece and to increase the hardness and strength of the workpiece material. Therefore, subsequent studies will focus more on rake cooling using the primary LN2 nozzle and on both rake and ank cooling using two nozzles. Extensive cutting force tests were conducted to investigate the inuence of cutting speed. Tables 25 list the cutting forces measured at cutting speed from 1 m/s to 2.5 m/s (200500 ft/min). In each test, all the cutting forces decreased as the cutting speed increased whether using dry cutting, cryogenic rake cooling or cryogenic machining using both nozzles. It can be explained that because the cutting temperature is higher at higher cutting
Table 2 Cutting forces in dry cutting of Ti64 Cutting speed V, m/s 1.0 1.5 2.0 2.5 Cutting force Fc, N 1317 1268 1201 1103 Thrust force F t, N 516 476 418 374 Feed force F f, N 596 560 503 489 Normal force N, N 1343 1290 1228 1134 Friction force F, N 725 619 507 456 Friction coeff. m 0.54 0.48 0.41 0.40

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285 Table 3 Cutting forces in cryogenic cutting when using only the rake nozzle Cutting speed V, m/s 1.0 1.5 2.0 2.5 Cutting force Fc, N 1348 1308 1250 1156 Thrust force F t, N 538 520 489 431 Feed force F f, N 560 507 467 458 Normal force N, N 1437 1343 1295 1183 Friction force F, N 458 363 325 249 Friction coeff. m 0.32 0.27 0.25 0.21

2279

Table 4 Cutting forces in cryogenic cutting when using both rake and ank nozzles Cutting speed V, m/s 1.0 1.5 2.0 2.5 Cutting force Fc, N 1388 1361 1285 1210 Thrust force F t, N 565 543 503 476 Feed force F f, N 543 516 454 423 Normal force N, N 1569 1417 1396 1282 Friction force F, N 375 325 307 269 Friction Coeff. m 0.24 0.23 0.22 0.21

Table 5 Forces in cryogenic machining Ti64 when using only the ank nozzle Cutting speed V, m/s 1.5 Cutting force Fc, N 1290 Thrust force F t, N 494 Feed force F f, N 547 Normal force N, N 1312 Friction force F, N 498 Friction coeff. m 0.38

speed [11], the workpiece material strength drops. This agrees with general understanding in metal cutting that higher speed cutting consumes less energy per unit of mass. 3.3.2. Friction coefcient Table 2 lists the cutting forces measured in the cutting test and the calculated normal forces and frictional forces for dry cutting. Table 3 lists them for cryogenic cutting using only the primary rake nozzle, and Table 4 lists them for cryogenic cutting with both rake and ank cooling. Obviously applying LN2 to the tool rake alone can signicantly reduce the gross friction coefcient. At a cutting speed of 1.5 m/s (300 ft/min), the friction coefcient for cooling the tool rake with LN2 (0.27) is smaller than the friction coefcient (0.38) for ank cooling by LN2 (calculated in Table 5). Cryogenic cutting at 1.5 m/s (300 ft/min) with both rake and ank cooling yields the lowest friction, 0.23, at the same speed.

2280

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

Fig. 6 shows the friction coefcient versus cutting speed for three cutting conditions: (1) dry cutting (without any coolant), (2) primary nozzle on (LN2 jet applied only to the tool rake) and (3) two nozzles on (LN2 jets applied simultaneously to both the tool rake and ank). For all cutting speeds, the data consistently show that the friction coefcient is in the order of high to low as: dry cuttingcryogenic ank coolingcryogenic rake coolingboth rake and ank cooling. Compared with the temperature study [13], the effectiveness of friction reduction for different cooling approaches is in a similar order as the cooling results. 4. Discussion 4.1. Coefcient of friction For the cryogenic cutting, the low friction coefcient results from the reduced temperature at the toolchip interface, which helps in maintaining a high level of hardness on the chip face adjacent to the tool rake, as indicated by the Ti6Al4V materials test [12]. That prevents the strong adhesion between the tool rake and chip face as occurred in dry cutting. With the help of a chipbreaker, the LN2 jet, applied at a high pressure of about 1.75106 Pa (250 psi) at the nozzle outlet, may also tend to penetrate into the toolchip interface and therefore reduce the metal-tocarbide contact friction. In all test conditions, the friction coefcient decreased as surface cutting speed increased. In other words, the friction coefcient decreased as the sliding speed of the chip on the tool rake increased. Since the toolchip interface temperature increases with the cutting speed for both dry and cryogenic cuttings, the reduction in friction coefcient with cutting speed can be attributed to the chips hot softening, which again reduced the yielding strength in the chips SDZ and

Fig. 6. Friction coefcients on toolchip interface vs. cutting speed.

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2281

lessened the chips resistance to sliding with the tool rake. However, with an additional jet applied to the tool ank (i.e. using two nozzles), the temperature on the toolchip interface is further reduced [11] and the hot-softening of the chip in relation to the increased cutting speed becomes negligible, leading to no considerable changes in the friction coefcient at increased cutting speed. Generally a lubricant is broadly dened as a material used for one or more of the following purposes: to reduce friction, to prevent wear, to prevent adhesion to aid in distributing the load, to cool the moving elements, and to prevent corrosion. LN2 was already known for its cooling capacity and ability to help reduce tool wear in cryogenic machining. In this study, reduction of the coefcient of friction is evidence that LN2 is a good lubricant in cryogenic machining. 4.2. Inuence of chipbreaker position The LN2 delivery nozzle used in this study was to reach and effectively cool on the localized hot-zones in the cutting area using a minimal amount of LN2. As shown in Fig. 2, the chipbreaker (with an integrated nozzle) lifted the chip off the rake face allowing the LN2 jet to direct LN2 to the toolchip interface. Therefore, the chipbreaker should be positioned close to the tool edge. But if positioned too close to the tool edge, the chipbreaker can act as a part of the tool rake and signicantly change the tool rake geometry, increasing the friction force component on the tool rake face. To maximize the lubrication effect of LN2, the nozzle/chipbreaker must be strategically positioned. Cutting tests at 1.5 m/s were performed to determine the best position for the chipbreaker when cryogenically cooling using the primary nozzle, as described by l and l in Fig. 7. The angle l was adjusted so that the LN2 jet could be directed toward the toolchip contact area and a proper chipbreaking was observed. A fresh tool insert was used for testing each position of the chipbreaker as listed in Table 6. The test results were again converted into the effective coefcient of friction and plotted in Fig. 8. The friction coefcient listed for dry cutting was obtained by switching off the LN2 supply.

Fig. 7. Chipbreaker/nozzle positioning.

2282

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

Table 6 Positions of chipbreaker for testing Position No. 1 l (mm) l (deg.) 1.75 15 2 1.50 15 3 1.25 15 4 1.25 20 5 1.25 10 6 1.00 10

Fig. 8.

Friction coefcient at different chipbreaker positions.

When testing cryogenic cooling using the primary nozzle, Position No. 3 produces the lowest friction coefcient on the toolchip interface. Although Position No. 1 and No. 2 show much improvement over dry cutting, the friction coefcient is not as low as Position No. 3. Because of the relatively large distance l between the chipbreaker and the tool edge, the chipbreaker was unable to lift the chip, which limited the LN2 jet to reach closely to the high temperature point on the tool rake. Position No. 6 was too close to the cutting edge, causing the chip to block the primary nozzle and the wedge-shaped chipbreaker to change the effective tool rake angle. 4.3. Metallurgical evidence of friction reduction Friction between the tool and the chip in metal cutting can inuence primary deformation and change the shear angle. It also causes secondary deformation in the change of grain structure, a layer where the lines of maximum grain elongation are curved. Therefore, the chip micrograph can indicate the level of frictional deformation, thus the lubrication effect of LN2.

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2283

Fig. 9.

Microscopic structure of the chip, dry cutting at speed 1.5 m/s and feed 0.254 mm.

Figs. 9 and 10 show the microscopic structure of the chip samples for dry cutting at 1.5 m/s (300 ft/min) and for cryogenic cutting using the primary nozzle with the same cutting parameters. One of the most distinguishable differences between the two chip samples is the signicantly reduced SDZ thickness from cryogenic cooling. Reduced SDZ thickness results from (1) the lowered toolchip interface temperature, which makes the chip less plastic, and from (2) the reduced frictional force, which is due to less adhesion between the tool and chip. Therefore, the reduction in SDZ thickness induced by the cryogenic cooling is evidence of reduced friction between the tool rake and the chip.

Fig. 10. Microscopic structure of the chip, cryogenic cutting using only the primary nozzle under the same cutting parameter as in Fig. 9.

2284

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

5. Conclusion Liquid nitrogen is an effective lubricant in cryogenic machining if used correctly. This lubrication effect is evidenced by the reductions of feed force, effective coefcient of friction between the chip and the tool face, and the thickness of the secondary deformation layer in the chip microstructure. Cryogenic machining tends to increase the cutting force because the work material becomes harder and stronger at lower temperature. However, the lower temperature makes the material less sticky, reducing the frictional force inherent in the cutting process. A clever approach of cryogenic lubrication would be to reduce LN2 spray to the workpiece while enhancing cooling at the toolchip interface. LN2 is highly effective when it is released to the chiptool interface via the primary nozzle. This effectiveness can be further enhanced by positioning the chipbreaker/nozzle so it can lift the chip, allowing LN2 to reach the area and to potentially form a uid cushion. Acknowledgements The authors wish to express their gratitude for nancial support of the cryogenic machining project by the National Science Foundation (Grant No. DMI-9528710) and the Edison Materials Technology Center (Project No. CT-32), and for the shared resources available through the joint partnership of Cincinnati-Milacron, General Motors-Delco-Chassis, Timken Company, Kennametal Inc., GE Aircraft Engines, Air Products and Chemicals, BOC Group, Vortec Co., A.F.Leis Co., Enginetics Co., Abrasive-Form, Inc., and Gem City Engineering. Special thanks are also due to Atul Saksena for his help in designing the delivery line and nozzles. References
[1] K. Uehara, S. Kumagai, Chip formation, surface roughness and cutting force in cryogenic machining, Ann. CIRP 17 (1) (1968) 6874. [2] K. Uehara, S. Kumagai, Characteristics of tool wear in cryogenic machining, J. Jpn. Soc. Precis. Eng. 35 (9) (1969) 7377. [3] A.R. Machado, J. Wallbank, Machining of titanium and its alloysa review, Proc. Inst. Mech. Eng., Part B: J. Eng. Manufact. 204 (B1) (1990) 5360. [4] A.D. Fillipi, R. Ippolite, Facing milling at 180C, Ann. CIRP 19 (2) (1971) 214215. [5] O.W. Dillon, R.J. De Angels, W.Y. Lu, J.S. Gunasekera, J.A. Deno, Some results from cryogenic cutting processes, J. Mater. Shap. Technol. 8 (1) (1990) 2329. [6] S.Y. Hong, Economical cryogenic machining for high speed cutting of difcult-to-machine materials, in: Proceedings of First International Conference on Manufacturing Technology, Hong Kong, 2729 December, 1991. [7] Z.Y. Wang, K.P. Rajurkar, J. Fan, Turning Ti6Al4V with cryogenic cooling, Trans. NAMRI/SME 24 (1996) 38. [8] S.Y. Hong, Cryogenic machining, US Patent No. 5,901,623, 11 May 1999. [9] G. Boothroyd, W.A. Knight, Fundamentals of machining and machine tools, 2nd ed., Marcel Dekker, 1989, pp. 8691. [10] E. Usui, A. Hirota, M. Masuko, Analytical prediction of three dimensional cutting process, Part 1. Basic cutting model and energy approach, Trans. ASME, J. Eng. Ind. 100 (1) (1978) 222228.

S.Y. Hong et al. / International Journal of Machine Tools & Manufacture 41 (2001) 22712285

2285

[11] Y. Ding, S.Y. Hong, A study of the cutting temperatures in machining processes cooled by liquid nitrogen, Tech. Pap. NAMRI (1995) 114119. [12] Z. Zhao, S.Y. Hong, Cooling strategies for cryogenic machining from a materials viewpoint, J. Mater. Eng. Perform. 1 (5) (1992) 669678. [13] S. Hong, Y. Ding, Cooling approaches and cutting temperatures in cryogenic machining of Ti6Al4V, Int. J. Machine Tools Manufact. 41 (10) (2001) 14171437.