study on near dry machining of aluminum alloys

83

Study on Near Dry Machining of Aluminum Alloys∗

Hiromi YOSHIMURA∗∗, Toshimichi MORIWAKI∗∗∗, Nobuo OHMAE∗∗∗, Tetsuo NAKAI∗∗∗∗,Toshiro SHIBASAKA∗∗∗, Hiroshi KINOSHITA∗∗∗, Makoto MATSUI† and Manabu SHIMIZU∗∗

Series of orthogonal cutting tests of aluminum alloys have been carried out to investigatethe chip formation process and adhesion of the work material to the rake face of the cuttingtool under near dry cutting conditions. Almost no adhesion of the work material was observedover a wide range of cutting speed tested when the aluminum alloy was cut with the sintereddiamond cutting tool. On the other hand, large amount of adhesion of the work materialwas observed at low cutting speeds, when cut with the carbide and DLC-coated tools. Theamount of the adhered material is reduced with an increase in the cutting speed. The averagecoefficient of friction on the rake face increases almost in linear relation with the amount ofadhesion. The chemical components of the adhered layer were also analyzed by EDS andAES techniques.

Key Words: Near Dry Machining, Aluminum Alloys, Adhesion, Coefficient of Friction

1. Introduction

In order to reduce CO2 emissions and waste in ma-chining operations, it is essential to reduce the use of cut-ting oil that accounts for substantial portions of the energyconsumption and production of industrial waste. The dryor near dry machining(1) – (5) is therefore the key technol-ogy to reduce the use of cutting oil. The most progressingproduction processes which employ the dry or near drymachining are gear cutting, or hobbing(6) – (9), and deephole drilling(10) in the area of steel cutting. On the otherhand, the near dry machining has not been much applied tocutting of aluminum alloys(11) – (13), because the aluminumadheres to the cutting edge of the tool, which leads tobreakage of the tools due to jammed chips and degrada-tion of machining accuracy. The aims of the present studyare to clarify the cutting phenomenon in the near dry ma-chining of aluminum alloys, and to establish guidelines for

∗ Received 28th October, 2005 (No. 05-4219)∗∗ Engine Production Engineering Div., Toyota Motor Cor-

poration, 1, Toyota-cho, Toyota, Aichi 471–8571, Japan.E-mail: hiromi [email protected]

∗∗∗ Department of Mechanical Engineering, Kobe University,1–1, Rokkodai, Nada, Kobe 657–8501, Japan.E-mail: [email protected]

∗∗∗∗ Headquarters for Innovative Cooperation and Develop-ment, Kobe University, 1–1, Rokkodai, Nada, Kobe 657–8501, Japan

† Mechatronics Systems Div., Toyota Motor Corporation,Teiho Plant 7, Teiho-cho, Toyota, Aichi 471–8571, Japan

appropriate machining method and selection of the toolmaterials and the machining conditions most suitable fornear dry machining.

Series of orthogonal cutting tests of aluminum alloyswere carried out here to study the effects of the tool ma-terial, the cutting speed, the rake angle and the amount ofthe cutting oil supplied in MQL upon the formation of ad-hered layer of aluminum on the rake face of the cuttingtool as well as the roughness of the cut surface.

2. Experimental Method and Conditions

Orthogonal cutting of aluminum alloy was carried outon an engine lathe by employing the longitudinal feed asshown in Fig. 1. The solid workpiece of aluminum alloy,equivalent to JIS AC2B, is machined to a tube shape asshown in Fig. 2 prior to the cutting tests, and cut with var-

Fig. 1 Photograph of experimental setup

JSME International Journal Series C, Vol. 49, No. 1, 2006

84

Fig. 2 Schematic illustration of experimental setup

Table 1 Composition of workpiece, wt%

Table 2 Test conditions

ious types of cutting tools of triangular inserts. Table 1shows the chemical composition of the aluminum work-piece.

The orthogonal chip formation process was filmedfrom the side by employing a high-speed motion cameraas schematically illustrated in Fig. 2. The cutting forcecomponents were measured with a quartz type tool dy-namometer, KISTLER Type 9257B. Table 2 gives the testconditions for cutting tests. Three cutting tool materialswere selected for the cutting tests, which are sintered car-bide tool, diamond like carbon coated (DLC) tool and sin-tered diamond tool.

3. Experimental Result and Discussions

3. 1 Observation of adhesionFigures 3 and 4 show typical photographs of cutting

Fig. 3 Formation and falling-off of adhesion on cementedcarbide tool (take angle 0◦, cutting speed 100 m/min,feed rate 0.1 mm/rev, quantity of oil mist supplied20 cc/h)

Fig. 4 Formation of adhesion on cemented carbide tool(rake angle 20◦, cutting speed 100 m/min, feed rate0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

zone taken with the high-speed motion camera when cut-ting with the carbide tool at a cutting speed of 100 m/min.The line drawings are also added to illustrate the cuttingzone clearly. Adhesion can be observed for both rake an-gles of 0◦ and 20◦. In the case of the rake angle of 0◦

as shown Fig. 3, adhesion accumulates as a deposited ma-terial, or so called build-up edge. This adhesion repeatsa cyclic of growing up to around 0.2 mm and dropping.When the rake angle is over 20◦ as shown Fig. 4, the ad-hesion does not grow much, but the adhesion of a thin androughly even layer is observed over the rake face similar toso called Belag. As for the growth and fallout of the build-up edge, it is not clearly seen, but the adhesion seems to betaking place continuously. Thin layer of adhesion is alsoobserved when the cemented carbide tool with a rake an-gle of 20◦ is employed at cutting speeds of 300 m/min and

Series C, Vol. 49, No. 1, 2006 JSME International Journal

85

Fig. 5 SEM photographs of adhered layers on cutting tool face (rake angle 20◦, cutting speed100 m/min, feed rate 0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

Fig. 6 SEM and EDS analyses of sintered diamond tool (rake angle 20◦, cutting speed100 m/min, feed rate 0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

1 000 m/min. The contact length of the chip with the rakeface is short in this cutting edge. Almost no adhesion canbe observed when the sintered diamond tool with a rakeangle of 20◦ is employed at a cutting speed of 100 m/min.

Figure 5 shows examples of SEM photographs of theadhered layers on the rake faces of cemented carbide tooland sintered diamond tool after cutting at a cutting speedof 100 m/min. The adhered layer is accumulated on thecemented carbide tool in the form of a layer flowing fromthe cutting edge in the chip flow direction. On the otherhand, the adhesion is observed only near the cutting edgeof the sintered diamond tool. It is considered that the ad-hesion depends on the unevenness of the cutting edge inthe case of the diamond tool. In order to investigate thecomponents of adhered layer on the cutting edge of the ce-mented carbide tool, an EDS, or energy dispersive X-rayspectrometer, analysis was performed over the rake faceand the flank face. The aluminum was found to be de-posited on the rake face, all the way from the edge to theboundary of the adhered layer, but no adhesion is observedon the flank face.

Figure 6 shows an SEM image depicting the part nearthe cutting edge of the sintered diamond tool, and the re-sults of the EDS analysis. The sintered diamond tool was

Table 3 Selected conditions for cutting tests

washed in 99.5 vol.% ethanol by applying ultrasonic vi-bration for five minutes, and was dried in air before SEMand EDS analyses. The figure reveals the presence ofsmall amount of aluminum even in the area without de-position.

3. 2 Effect of tool materials and cutting conditionsThe effect of tool materials and cutting conditions

on the formation of metal adhesion and finished sur-face roughness were investigated under selected condi-tions summarized in Table 3. The state of adhesion ofaluminum workpiece on the tool face was examined af-ter cutting for 2 mm in the feed direction. The amount ofadhered layer is represented by the width of the adhered


86

Fig. 7 Adhesion layer width on tool

Fig. 8 Relationship between cutting speed and metal adhesionwidth (rake angle 20◦, feed rate 0.1 mm/rev, quantity ofoil mist supplied 20 cc/h)

layer on the rake face of the cutting tool as illustrated inFig. 7.

Figure 8 shows the relationship between the cuttingspeed and the adhesion widths for different tool materi-als tested. Almost no adhesion is observed when cuttingwith the sintered diamond tool. On the other hand, largeamount of adhesion is observed when cutting with the ce-mented carbide tool and DLC-coated tool, but the amountof adhesion decreases as the cutting speed is increased. Itis presumed that the decrease in the adhesion layer widthcaused by an increase in the cutting speed depends on thecutting temperature, but this aspect needs further investi-gation in the future.

Figure 9 shows the relationship between the cuttingspeed and the finished surface roughness. The surfaceroughness is maintained at approximately 3 µm in Rywhen cut with the sintered diamond tool, while it is muchimproved with the DLC-coated tool and the cementedcarbide tool at higher cutting speeds. The finished sur-face roughness is improved according to a decrease in theamount of adhesion. It is presumed that the repeated for-mation and falling-off of adhered metal degraded the char-acteristics of the finished surface. Based on the charac-

Fig. 9 Relationship between cutting speed and finished surfaceroughness (rake angle 20◦, feed rate 0.1 mm/rev,quantity of oil mist supplied 20 cc/h)

Fig. 10 Relationship between average coefficient of frictionbetween tool rake face and chip and adhesion (rakeangle 20◦, feed rate 0.1 mm/rev, quantity of oil mistsupplied 20 cc/h)

teristics of the finished surface, it is understood that goodsurface quality is obtained regardless the tool material pro-vided that the cutting speed is 300 m/min or more.

Figure 10 shows the relationship between the averagecoefficient of friction between the rake face and the chipand the adhesion width. The average coefficient of frictionwas estimated based on the cutting force components mea-sured. The average coefficient of friction and the amountof adhesion remain almost proportional to each other, ir-respective of the tool materials and cutting speeds. Noadhesion is formed when the coefficient of friction is 0.5or less. The average coefficient of friction is that of thefriction between the adhesion surface and chips, and it ispresumed to be larger than the one between the cuttingtool and chips. Moreover, the magnitude of the coefficientof friction becomes smaller as the adhesion layer widthdecreases, namely, the cutting speed rises.

Figure 11 shows the effect of feed rate on the adhesionwidth. The amount of adhesion remains almost negligiblewith the sintered diamond tool regardless of the feed rate,while it increases with an increase in the feed rate in thecases of the cemented carbide tool and DLC-coated tool.


87

Fig. 11 Effect of feed rate on adhesion width (rake angle 20◦,cutting speed 100 m/min, quantity of oil mist supplied20 cc/h)

Fig. 12 Effect of rake angle and rake face roughness onadhesion width (cutting speed 100 m/min, feed rate0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

Fig. 13 Effect of coolant on adhesion width (rake angle 20◦,cutting speed 100 m/min, feed rate 0.1 mm/rev)

Additional cutting experiments were conducted withthe cemented carbide tool by varying the conditions shownin Table 3. Figure 12 shows the effects of the rake angleand the rake face roughness on the adhesion width. Fig-ures 13–15 show the effects of the lubrication conditions.It is understood from this experiment that the adhesion is

Fig. 14 Effect of supply direction of mist on adhesion width(rake angle 20◦, cutting speed 100 m/min, feed rate0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

Fig. 15 Effect of mist flow rate on adhesion width (rake angle20◦, cutting speed 100 m/min, feed rate 0.1 mm/rev)

not improved even when the cutting fluid is applied, or inwet cutting. It is slightly improved when small amount ofmist is present. It is understood that the adhesion is notmuch reduced by changing the lubrication conditions andthe roughness of the tool rake face as long as the cementedcarbide tool is employed.

3. 3 Detailed analysis of observed layerAn AES, or Auger Electron Spectrometer, analysis

was performed where the adhered layer was formed. Fig-ure 16 shows the results of AES analysis of the cementedcarbide tool, and Fig. 17 shows those of the sintered dia-mond tool.

Figure 16 shows the results of investigation at theinterface between the adhered metal and the cutting tool.The left-hand part of the figure shows the results of ananalysis performed on the components of Al, Si, W, C,and O, while removing the adhered metal by sputtering.In the case illustrated above, the interface is consideredto be present at around 45 minutes of sputtering time. Atthis point, aluminum decreases gradually, and this phe-nomenon seems to be dependent on the surface roughnessof the observed area. No oxygen is found in the neigh-borhood of the cutting edge of the cemented carbide tool


88

Fig. 16 AES, or Auger Electron Spectrometer, analysis of adhered layer on cemented carbidetool (rake angle 20◦, cutting speed 100 m/min, feed rate 0.1 mm/rev, quantity of oilmist supplied 20 cc/h)

Fig. 17 AES analysis of adhered layer on sintered diamond tool (rake angle 20◦, cutting speed100 m/min, feed rate 0.1 mm/rev, quantity of oil mist supplied 20 cc/h)

between the rake face of the tool and the adhered layer.And almost no Si nor O was detected. The cemented car-bide tool and the aluminum are welded to each other attheir interface. On the other hand, the oxygen is found atthe tail of the adhered layer, which is 0.4 mm away fromthe cutting edge, and it is likely that some oxide may have

been deposited at the interface between the rake face andthe adhered layer.

In the case of the sintered diamond tool (refer toFig. 17), there is no contamination by oxide nor carbides atthe interface, which is similar to the case near the cuttingedge of the cemented carbide tool.


89

When adhesion is formed, it is presumed that oxygenis present in the proximity of the point of separation be-tween the cutting tool and the chips, and that the oxideplays some role in formation of the adhesion.

4. Conclusions

Series of orthogonal cutting tests of aluminum alloywere carried out with cemented carbide, DLC-coated andsintered diamond cutting tools for a wide range of cuttingconditions. The chip formation process was observed byemploying a high-speed motion camera, and the adheredlayers of aluminum on the face of the cutting tools werecarefully analyzed by EDS and AES techniques. The fol-lowing remarks are concluded;

( 1 ) When the aluminum alloy is cut with the sintereddiamond tool, almost no adhesion of the work material isobserved on the rake face of the tool over a wide cuttingspeed range tested, and good surface is obtained.

( 2 ) The aluminum is adhered on the rake face of thetool when cut with the carbide tool, and the amount of theadhered layer is reduced with an increase in the cuttingspeed. The roughness of the cut surface is improved ac-cordingly. The surface roughness is kept at almost 2µmin Ry for a cutting speed range from 300 to 1 000 m/min,which is better than that obtained by the sintered diamondtool. The DLC-coated tool shows similar results withthose obtained by the carbide tool.

( 3 ) The nominal coefficient of friction on the rakeface obtained by the cutting force components measuredincreases almost in linear relation with the amount of ad-hesion. When the coefficient of friction is 0.5 or less, al-most no adhesion is observed.

( 4 ) The adhered substance on the face of the carbidetool consists of deposit of the newly generated surface ofaluminum in the proximities of the cutting edge, but as itgoes further from the cutting edge, oxygen is detected andis deposited in the form of an oxide.

References

( 1 ) Klocke, F. and Eisenblattere, G., DRY Cutting, Annalsof the CIRP, Vol.46, No.2 (1997), pp.519–526.

( 2 ) Sutherland, J.W., Kulur, V.N. and King, N.C., An Ex-perimental Investigation of Air Quality in Wet and DryTurning, Annals of the CIRP, Vol.49, No.1 (2000),pp.61–64.

( 3 ) Min, S., Inasaki, I., Fujimura, S., Wada, T., Suda, S.and Wakabayashi, T., A Study on Tribology in MinimalQuantity Luburication Cutting, Annals of the CIRP,Vol.54, No.1 (2005), pp.105–108.

( 4 ) Wakabayashi, T., Sato, H. and Inasaki, I., Turning Us-ing Extremely Small Amount of Cutting Fluid, JSMEInt. J., Ser. C, Vol.41, No.1 (1998), pp.143–148.

( 5 ) Suda, S., Wakabayashi, T., Inasaki, I. and Yokota,H., Multifunctional Application of Synthetic Ester toMachine Tool Lubrication Based on MQL MachiningLubricants, Annals of the CIRP, Vol.53, No.1 (2004),pp.61–64.

( 6 ) Yoshimura, H., Miyazaki, K., Ono, T. and Nakamura,M., Development of Dry Hobbing, TOYOTA TechnicalReview, (in Japanese), Vol.52, No.2 (2002), pp.78–83.

( 7 ) Komori, M., Sumi, M. and Kubo, A., Method of Pre-venting Cutting Edge Failure of Hob Due to ChipCrush, JSME Int. J., Ser. C, Vol.47, No.4 (2004),pp.1140–1148.

( 8 ) Matsuoka, H., Ono, H. and Tsuda, Y., Wear Reduc-tion Effects of Ca Sulfonate on Coated Tools Hobbing,JSME Int. J., Ser. C, Vol.45, No.3 (2002), pp.821–830.

( 9 ) Matsuoka, H., Tsuda, Y. and Ono, H., Fundamental Re-search on Hobbing of Austempered Ductile Iron Gear,JSME Int. J., Ser. C, Vol.46, No.2 (2003), pp.1160–1170.

(10) Yoshimura, H., Ecological Manufacturing by ToyotaMotors, Proceeding of HSS·FORUM Conference/Smart Solution for Metal Cutting, (2005), pp.1–9.

(11) Ng, E-G, Szablewski, D., Dumitrescu, M., Elbestawi,M.A. and Sokolowski, J.H., High Speed Face Millingof Aluminium Silicon Alloy Casting, Annals of theCIRP, Vol.53, No.1 (2004), pp.69–72.

(12) List, G., Nouari, M., Gehin, D., Gomez, S., Manaud,J.P., Le Petitcorps, Y. and Girot, F., Wear Behaviourof Cemented Carbide Tools in Dry Machining of Alu-minum Alloy, Wear, 259 (2005), pp.1177–1189.

(13) Nagata, M., Morita, H., Yamada, T. and Iwasa, M.,Approach to MQL Cutting by Using High LubricationCutting Fluid, Proceeding of the LEM21, Vol.1, No.05-204 (2005), pp.127–132.


study on near dry machining of aluminum alloys

Documents