- Email: [email protected]
High-Speed Milling of Dies and Molds in Their Hardened State
High-speed Milling of Dies and Molds in Their Hardened State M. A. Elbestawil (l), L. Chenz, C. E. Beczel, T. I. El-Wardanyl ’Intelligent Machine and Manufacturing Research Center, McMaster University, Hamilton, Ont., Canada *United Technologies Research Center, East Hartford, Connecticut, USA Received on January 9,1997
Abstract This paper presents an experimental investigation of high speed milling of dies and molds. Several critical issues involved with the high speed milling of H13 tool steel of hardness up to 55 HR have been studied and explained from a detailed analysis of experimental observations. The experiments were performed using several grades of PCBN ball-nose end mills with various edge preparations. The effect of different process parameters on the tool performance and the surface finish produced was also investigated. The cutting parameters involved were; cutting speeds in the range of 220 to 1320 m/min, feed variation from 0.0254 to 0.1 mm/tooth, axial depth of cut from 0.625 up to 2 mm, and radial width of cut of 0.254 mm. During the preliminary experimental investigation, the tilt angle was kept constant at 10 degrees. Several tests were conducted to study the effect of the different tool path directions on the cutting tool performance. Dry and wet cutting conditions were used and the effect of coolant on the tool life was also determined. The optimum cutting conditions have been specified based on the modes of tool failure, tool life and surface integriv produced. Keywords:
milling, dies and molds, Cubic Boron Nitride (CBN)
Introduction Dies and molds are utilized to produce a large number of discrete components to near net and final shape [l]. The functional parts of the die often include sculptured surfaces. The intricate geometry of these sculptured surfaces and the relatively high degree of hardness of the die materials used necessitates the use of advanced technology in their manufacture. This technology employs new NC tool path generation methods and high speed milling of heat treated die steels using CNC milling machines equipped with proper controlling sensors (21. With the development of super-hard cutting materials such as cubic boron nitride (CBN), poly crystalline cubic boron nitride (PCBN) and the like, the technology of high speed machining (HSM) of hardened steels has created considerable interest to several leading die and mold manufacturers. However, this technology is rather new and a lack of process knowledge and appropriate cutting edge geometry have impeded the broad application of HSM thus far [3]. It is believed that the use of HSM in the die and mold industry can reduce the machining time (4, 51, produce an improved workpiece surface quality and also provide longer tool life [6). Several issues involving high speed milling of dies and molds have been recently studied; optimization of the tool paths generated and cutting conditions, the metal removal efficiency, the tool life and process stability [2 to 91. Ball end milling cutters are used extensively in machining parts with sculptured surfaces. There has been significant research reported in modelling the mechanics of ball end milling [lo,I l l aiming for a longer tool life, where breakage of the tool shank, and chipping of ball shaped helical flutes are the critical factors in ball end milling of die alloys [lo1. PCBN ball end mill tools have been recently used in high speed milling of dies and molds. The workpiece materials were heat treated alloy steels of hardness 30 to 45 HR, [6,12].The suggested cutting speeds were between 500 to 1000 m/min. Also, ultra high spindle speeds (50K1 OOK rpm) allow higher feed rates of up to 10 m/min. This shortens the machine cycle or tightens the pick feeds to improve the surface finish produced. However, the integrity
Annals of the ClRP Vol. 46/1/1997
of the surface produced has yet to be investigated. At such high speeds, the nature of the chip formation is completely modified. The chip formation mechanism is expected to vary as the workpiece hardness increases. Also, different tool wear mechanisms may be observed at this high cutting speed due to the amount of heat generated and dissipated during cutting. This paper examines the cutting performance of PCBN tools during high speed semi finish and finish milling of H 13 tool steel of hardness up to 55 HR,. The type of tool wear, the dominant wear mechanism, and tool failure modes are studied. The effect of different process parameters on the cutting forces, chip morphology, and metallurgical surface integrity produced are also presented. Experimental Procedure . The workpiece material was H13 tool steel, heat treated to 45 and 55 HR,. The workpieces were prepared in the form of 50x100~300mm blocks. The cutting tools were 12.7 mm diameter single tooth ball nose end mills to avoid the influence of tool run-out on wear measurements. Two different grades of PCBN cutting edges brazed on solid carbide shanks were used. Grade 1 was a high volume fraction CBN (90%) with metallic binder and grade 2 was a low volume fraction CBN (65%) with ceramic binder. Three different edge preparations (sharp, honed with a 0.025 mm radius, and chamfered with a 20’ angle) were used. The rake angle was kept at -lo’,according to [6.12]. A 10’ tilt angle was used as recommended by several investigators [4,6]. The pick feed (fp) was kept constant for all cutting tests at 0.25 mm. The cutting speeds (V) were evaluated for the highest point of engagement on the ball end mill in the range of 220 to 1320 dmi n. Table 1 lists the selected variables and their respective values. The cutting force components (Fx. Fy, Fz) were measured using a force dynamometer. The length of each cutting path was approximately 300 mm and at the end of each 10” pass the tool was examined for chipping, breakage, notch wear, or material sticking on the tool rake face. The tool wear was measured using a tool maker’s microscope. The parameters measured to represent the progress of the wear were; average and maximum tool wear
57
Table 1 Cutting parameters used in the experiment Parameter Spindle speed (N) (revlmin) Feed (f) (mmltooth) Axial depth of cut (ADC) (mm)
Values
1 1 1 :3: 1 10,000
60,000
0.025
0.05
0.625
1.125
Type of cut
remain undeformed at the tool exit, because of the thinning phenomenon. Examining the chip cross section microstructure reveals an extensive shear zone with subsurface plastic flow. The expected saw tooth chips (characteristic of hard machining) were not observed because the cracks initiated at the free surface of the workpiece were terminated immediately due to plastification.
Flood coolant
PCBN grades Edge preparation ~
~~
I
sharp
Workpiece hardness HR, Cutting path
I
45 Q
honed
I
K-land
I
a) Top of chip b) Bottom of chip Figure 2 Chips produced at 10,000 rpm.
55
to fp
!I to fp
VB, and VB,, respectively. The position of the maximum tool wear on the cutting edge was also recorded. The surface roughness of the workpiece was measured at several locations along the length of cut using a portable stylus-type instrument. The deviation of the milled surface from the normal surface was measured for evaluating the effect of the tool wear on the geometry of the surface produced. To study the chip geometry, some chips were collected from each cutting test and mounted in an epoxy metallurgical mount, polished, etched, and examined by optical microscopy at magnifications lOOX and 1600X. In addition, some of these polished and mounted samples were also examined by Scanning Electron Microscopy (SEM) using back scattered electrons analysis to determine the region of plastic deformation. Samples of the machined surface at the start of cutting and at the end of tool life were taken from the block for examination of surface integrity under the optical and scanning electron microscope. Experimental Results and Discussion 1. ChiD formation durino hioh m e e d milling Owing to the limited period of engagement of the
cutting edge, the chip produced was short, completely segmented, and of variable thickness caused by the ball end mill geometry. The thinning of the chip in both axial and radial directions is illustrated in Figure 1.
figure 3 shows an SEM image of the chips produced when a spindle speed of 60,000 rpm is used under the same cutting conditions. The high heat generated at these speeds caused the thinner part of the chip to melt and form small spheres (refer to Figure 3). Fig. 3-b illustrates the recrystallization of this melt upon quenching. The diameter of !hese spheres are some tenths of a millimeter. Increasing the feed may reduce the possibility of the generation of this morphology. However, with the progress of tool wear, the melting process of the chips was observed. The cutting edge geometry was found to have a large influence on the production of the spheres.
a) Spherical chip b) microstructure of sphere Figure 3 Chips produced at 60,000 rprn. In general, serrated chips were produced at this high speed and the tendency of generating saw toothed chips was clear as shown in figure 4.
l O O x Magnification
Figure 1 Cross section showing radial chip thinning.
a) Top of chip b) Bottom of chip Figure 4 Morphology of chips produced at 60,000 rpm.
Straw coloured chips were produced when end milling workpieces (45 HR,) at a spindle speed of 10,000 rpm. On the other hand, dark blue chips were produced during the machining of the 55 HR, workpieces. For the cutting conditions of N =10,000rpm, f = 0.025 mdtooth, ADC = 0.625 mm and fp = 0.25 mm, the chips produced are shown in Figure 2. The optical microscope image shows that the chip deformed upon tool entrance. However, the chips
2. Modesof Tool failure In semi finish and finish machining of hardened steel, tool life is taken to be the time needed for the surface finish to deteriorate to a predetermined level. However, in finish machining of dies and molds, it would be misleading to rely solely on the surface roughness to evaluate the tool life of the PCBN tool [I]. A worn tool may impart a better surface
58
finish than a fresh one (fresh inserts cut well defined scallops resulting in a higher surface roughness). Since worn inserts also have an offset cutting edge, they leave stock on the surface that has to be removed manually [4]. In die and mold manufacturing, this type of surface defect should not exceed 0.025 mm and it mainly occurs when the maximum flank wear on the tool exceeds 0.05 mm [4]. Hence. the tool life in this study was ended when the level of the maximum flank wear reaches 0.075 mm. Different modes of tool failure were observed and are listed as follows; a) Modes of failure of High CBN tools: Grade 1 tools failed primarily by reaching maximum specified flank wear. However, it should be noted here that for a higher chip load such as feed of 0.1 mm/tooth, axial depth of cut of 2.0 mm, or when using honed and chamfered cutting edges. plastic deformation of !he tool nose has been observed repeatedly. This was found to accelerate the tool wear and cause a faster end to the tool life. An example of this plastic deformation can be observed from the SEM image in Figure 5. Tool chipping was only noticed when using a honed cutting edge at low feed, and catastrophic tool failure occurred when flood coolant was used.
develop. The tool nose in the vicinity of the cut deformed, and chipping of the cutting edge was initiated. Figures 7 a) and b) show two SEM images of the tool wear regions. Figure 7 a) corresponds to the wear developed on the cutting edge at a spindle speed of 10,000 rpm. The figure shows the evidence of material sticking and fine grooves formed on the rake face, and nose deformation. The very smooth surfaces on the rake face near the cutting edge shown in the SEM image may indicate that the binder material (cobalt) in high CBN tool diffuses into the chips produced. This may lower the bonding strength of the tool, allowing the pull out of some CBN particles causing abrasive action on the flank. In some cases notch wear was also observed. Notch wear is caused by both the sliding of chips and CBN grains. Figure 7 b) shows an SEM image of the tool wear produced during milling at a spindle speed of 60.000 rpm. A protective layer (liquid phasej is deposited on the cutting edge. This protective layer acts as a buffer between the cutting edge and the chip. Further adherence of the material has also been observed located at intermediate cutting velocities. At the highest contact point (highest velocity), material adhesion was not observed. At the extreme point of contact between the tool and the workpiece, the high temperatures generated during machining deforms the softer matrix of the workpiece. leaving the carbide material intact. Thus the possibility of abrasion wear may exist from these carbide particles.
Figure 5 Plastic deformation and chipping of grade 1 tool Modes of tool failure of low CBN tools: b) Grade 2 tools failed primarily by edge chipping. Chipping occurred irrespective of the type of edge preparation or cutting conditions used. For honed cutting edges and low feeds, massive chipping was observed after the first cutting pass. For sharp cutting edges, tool wear developed uniformly on the flank until its level was 0.04 mm. Then, plastic deformation started on the tool nose leading to tool chipping. Material sticking on the rake face side of the chamfered cutting edges always led to catastrophic failure of the tool at low feed, as shown in Figure 6. Severe depth of cut notch wear repeatedly occurred when a honed cutting edge was used with flood coolant.
a) material sticking b) chipping Figure 6 Material sticking and chipping of grade 2 tool. 3. Tool wear mechanisms In the early stages of cutting the tool started to wear at the point where the maximum cutting speed was achieved. With the progress of the cut, uniform tool wear developed along the whole length of contact with the workpiece. Depending on the cutting conditions and the cutting edge preparation, plastic deformation on the rake face may
a) wear at low cutting speed b) wear at high cutting speed Figure 7 Different tool wear mechanisms observed 4. Effect of Drocess variables on cuttina forces
Figure 8 shows the force components measured while end milling at 10,000 rpm spindle speed. As can be seen, the higher force component is in the thrust direction (FJ. Recalling that the workpiece was inclined at loo, Fz contains components from Fy. Irrespective of the resolution of forces, Fz is considerably higher than the others. This phenomenon can be related to the hard machining process where a pronounced increase in the thrust component is always observed when cutting at low feeds and depths of cut [17]. The use of negative rake angles and a large nose radius generates a ploughing force component which makes the magnitude of the thrust force component exceed that of the tangential component [18]. This is expected since the axial and radial depths of cut were very small and the ploughing process is dominant. Average resultant forces versus cutting time for different spindle speeds, feeds, axial depths of cut and edge preparation were investigated. The average force (within one tooth period) was calculated as the square root of the sum of the squares for the three cutting force components Fz, Fx and Fy. A reduction of the average cutting force occurred with the increase in cutting speed which can be related to chip thinning. The force component caused by the rate of change in momentum of chip material is negligible at the range of cutting speeds used in this study. The effect of chip material momentum on the cutting force signal was found to be significant only at cutting speeds above 9000 m/min [ 191. It should be noted here that the use of a magnetic bearing spindle for the higher speed machining induces large dynamic force components which were four times the actual
59
0.
h
5
h
v;
V'
LL a
500,
I
"0
0.005
0.015
0.01
0.02
Cutting time (sec) Figure 8 Force pattern generated during end milling. N=10,000 rpm. f=0.025mm/tooth, ADC=0.625mm, Grade 1
5.Effect of Drocess variables on tool life Figure 11 shows the effect of edge preparation, flood coolant, tool material grades and tool path on the performance of the PCBN tools. The best tool performance was obtained when using a high CBN content cutting tool material with a sharp edge preparation. The high CBN grade performed better than the low CBN grade under the given conditions except when flood coolant was used. Since the high CBN cutting tool has a higher apparent thermal conductivity (100 W/m/K), the heat generated during cutting can dissipate rapidly from the cutting zone, thereby diminishing the produced "hot machining effect" leaving the material difficult to cut [13]. Hence, using flood coolant, a 10fold reduction in tool life is observed. Further, CBN tools (as with most ceramic materials) have poor thermal shock resistance and, thus, flood coolant application results in rapid tool failure. 5000 y - ~ - - -
cutting force. This must be taken into consideration as it will affect the surface finish. The results from Figure 9 indicate that increasing the feed produces a more stable cutting process. On the other hand, the cutting edge preparation has a significant effect on the cutting forces generated (Figure 9).The cutting forces were the highest when honed edge was used. The large negative effective rake angle of honed edges result in more ploughing action instead of cutting, which increases the friction force component.
800
-L
I !
600
L
,
& 400
I jI
Chamfered, f=.025mm/tAoth rn Honed, f=.O25mm/tooth ' I I -
I
I
l
l
0 0
I
0.5
1
1.5
2
Cutting time (sec) Figure 9 Effect of feed and edge preparation on milling forces (N=10,000 rpm, Grade 1) Increasing the axial depth of cut increases the average resultant forces as the chip width increases (Figure 10). An increase in cutting force was observed for a decrease in hardness of 10 HR, (Figure 10).
800
-600 z
I I DADC=0.625
7
mm (45HR,), A ADC=0.625 mm/ ADC=1.25 mm, v ADG2.0 mm (55HR,)
L $, 400 ?!!
0
2 200 0 0.5 1 1.5 2 Cutting time (sec) Figure 10 Effect of axial depth of cut and workpiece hardness on milling forces (N=10,000 rpm, Grade 1) 0
60
h
4000 r ~3~~~ Y
1
4L,
!
1
2 Chamfered edge 3 Honed edge (Dry) 4. Honed edge (Wet) 5 Honededge 1 (Cut along direction of inclination)
2
3
4
5
Figure 11 Effect of process parameters on tool life. N= 10,000 rpm, f = 0.25 mm/tooth, ADC = 0.625 mm
i
!!? I
1
1 Sharpedge
The high CBN tool outperformed the low CBN tool because of its higher transverse rupture strength. The microstructures of the high CBN tools show that the void between the CBN particles is filled with a pool of liquid metal infiltrating from the molten tungsten carbide interface [14]. The liquid metal pool and the high strength carbide substrate provide additional mechanical support to the polycrystalline CBN layer, which gives high CBN tools favourable impact strength for use in interrupted cutting conditions. The honed cutting edges gave the worst tool performance. This was unexpected since this edge preparation is usually employed and recommended for finish cuts. This is consistent with the results obtained by Chou et al. [15]. This can be attributed to the large negative effective rake angle of honed edges which result in more ploughing action instead of cutting. This leads to higher friction and tool wear. The tool life of chamfered cutting edges was less than that of sharp edges. The effective negative rake angle in the case of chamfered cutting edges was -30°,which caused extensive wear on the rake face noted by a steepening of the tool edge, and chipping. A noticeable increase in the tool life was obtained when upward cutting was used where the tool axis was inclined in the feed direction with respect to the surface normal. These results were expected since this type of tool path gives the minimum tool-workpiece contact area, as shown in [15]. Also, this type of tool path prevents any engagement between the tip of the ball nose end mill and the workpiece surface. It is well known that in the centre of this tool the cutting speed equals to zero and thus, the effective chip velocity is very small. This leads to cutting edge chipping and a higher wear rate due to a ploughing effect. Figure 12 shows that wear on high CBN tools decreases as the workpiece hardness increases. Increasing the cutting speed has a limited effect on the tool life as a function of the cutting length when machining the harder H13
workpieces (refer to Figure 12). Conversely, when machining the less hard material, the tool life was reduced from 3750m to 1000 m lengths of cut. The variation of chip morphology as a function of workpiece hardness indicates that more heat is generated during the machining of 55 HR, tool steel. The higher the heat generation, the more the chip will be softened and the better the workpiece plastification. This will reduce the mechanical load imposed on the tool and, therefore, less tool wear will exist. 0.12 v
A c
45HRc 55HR,
- 10000 rpm
---
60000 rpm
0.08 Y
x
5
2 0.04 >
0
0
1000
2000 3000 Cutting length (m)
4000
5000
Figure 12 Effect of hardness and speed on tool wear. f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254, dry cut, sharp edge, Grade 1 tool Tool life was reduced as the axial depth of cut was increased due to larger chip load as shown in Figure 13 a). Increasing the axial depth of cut will increase the maximum cutting speed. The Increase in the cutting speed generated high thermal loads on the tool, which when combined with intermittent cuts, accelerated tool wear because of the thermal shock. Figure 13 b) shows the effect of feed on the tool wear for the cutting edge of both PCBN grades. In general, the increase in the feed reduced the level of tool wear. Since the sliding contact length between the tool and the workpiece reduces proportionally to the increase in feed, tool wear is also expected to be reduced. On the other hand, increasing the feed shortens the heating cycle which will be interrupted during its transient phase before reaching steady state. The peak temperature reached will be lower than the corresponding steady state value. However, the use of very
low feeds with the chamfered edge produces vibration on the system, which leads to a higher wear rate of high CBN and a catastrophic failure of the low CBN. 6.Effect of Drocess parameters on the surface intearity produced: The feed rates used combined with the high cutting speed and large nose radius consistently produced very good quality surface finishes. Geometric error was found to be in the range of 0.025 mm and it increases to 0.075 mm as the tool wears. The range of surface roughness varied from 0.2 pm ( fresh edge) to 0.6 pm (very worn tool) Ra in all cases of the 10,000 rpm speed (e.g. Figure 14 a). Rmax was found to be lower for the harder material (2.0 pm) as opposed to less than 4.0 pm for the 45 HR, workpiece. When the magnetic bearing spindle speed was used at 60,000 rpm, the maximum roughness obtained was .66 pm Ra, resulting from the vibration induced to the cutting system by the flexibility of the magnetic spindle. A noticeable increase in surface roughness was not realised as the cut off value of the flank wear is very low for finishing. Hence, the end of tool life was very difficult to gauge from the surface roughness. The value of the roughness when the tool catastrophically failed was obviously very high. At the high spindle speed, Rmax was found to vary contrary to what was observed at low spindle speeds i.e. increasing hardness increases the roughness (Figure 14 b). 2O I OOO rpm G - m 60000 rpm
1
h
.
3
I
0
0
.
-
I
1000 2000 3000 Cutting Length (m)
4000
5000
f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254 mm, dry cut, sharp edge Figure 14 a Effect of speed o n surface finish. h
f 4
h
E
v
1.27 mm ADC
0
1000
2000
3000
5000
4000
Cutting length (m) N=lO,OOO rpm, fz0.0254 mm/tooth, RWG0.254 mm a) effect of ADC
_~._. -.
5 T
3
u)
O
1000
2000
3000
4000
5000
Cutting Length (m)
.--,
,
N=lO,OOO rpm, f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254 mm, dry cut, sharp edge Figure 14 b Effect of hardness on surface finish.
E 0.04
03
> 0
1000
2000
3000
4000
5000
Cutting length (m) N=10,000 rpm, f=0.05 mm/tooth, RWG0.635 mm b) effect of feed Figure 13 Effect of axial depth of cut and feed on tool wear.
Metallographic examination of the workpiece subsurface revealed that a damaged layer exists a few microns under the cut surface and has a thickness of approximately 4-6 pm. This damaged layer is evident in all of the cutting conditions. It is metallurgically different than the fine martensitic structure (marked by a difference in contrast) of the bulk material refer to Figure 15 a). The workpiece hardness has a great influence on the observable
61
structure of the damaged layer. The 45 HR, workpiece revealed a relatively coarse martensitic structure far under the surface (>20 pm). However near the cut surface, it is apparent that the martensite grains are clearly deformed in the direction of chip flow, see Figure 15 b). Figure 16 shows SEM images of the microstructure of the subsurface damage produced at different cutting conditions. The thickness of the damaged layer was found to be dependant on edge preparation, and tool wear. Interestingly, increasing the axial depth of cut and the feed has no significant effect on the thickness of this layer (an example of which is shown in Figure 16 a) and b).
-
a) surface of 55 HR, b) surface of 45 HR, Figure 15 SEM images of workpiece subsurface damage.
a) ADG0.625 mm b) ADC = 2.0 mm Figure 16 Effect of axial depth of cut on subsurface damage.
Conclusions High speed hard milling tests using PCBN tools have been performed on H13 tool steel materials. From the analysis obtained, the following conclusions can be drawn: 1- Ultra high speed milling of hardened steel is feasible under the given cutting conditions. The geometric error was within acceptable limits. 2-A large chip load yields a more stable cutting process, and higher tool life. The surface quality and integrity are not significantly affected. 3- Contrary to the tool manufacturers specifications, sharp edge preparations outperform the honed and chamfered edges. 4- The experimental results yield that high volume fraction (goo/,) CBN tools are recommended for milling hardened tool steel. The main mode of tool failure for high CBN tools was classical flank wear. Consequently, low volume fraction CBN tools fail generally by chipping. 5- The higher the cutting speeds the thinner the chips produced. This leads to the production of spherical chips caused by material melting and recrystallizing.
References:
[ 11
[ 21
62
Altan, T., Lilly B. W., Kruth, J. P., Konig, W., Tonshoff, H. K., Van Luttervelt, C. A. and Khaity, A. B., 1993, “Advanced Techniques for Die and Mold Manufacturing”, Annals of CIRP, Vol. 42/4, pp 707. Takeuchi, Y., Idemura, T., Sata, T., 1991, “5-Axis Control Machining and Grinding Based on Solid
Model”, Annals of CIRP, Vol. 40/1, pp 455-458. Fallbohmer, P., Altan, T., Tonshoff, H. K. and Nakagawa, T., 1996, “Survey of the Die and Mold Manufacturing industry - Practices in Germany, Japan, and the United States”, Journal of Materials Processing Technology, Vol. 59, pp 158-168. Kruth, J. P. and Klewais, P., 1994, ”Optimization and Dynamic Adaptation of Cutter Inclination During Five-Axis Milling of Sculptured Surfaces”. Annals of CIRP, Vol. 43/1, pp 443-448. Schultz, H. and Hock, St.. 1995, “High-speed Milling of Dies and Molds Cutting Conditions and Technology”, Annals of CIRP. Vol. 44/1, pp . Ikeda, T., Takahashi. I., Matsuoka, T. and Nakagawa, T., 1992. “Ultra High Speed Milling of Die Steel with Ball-nose End mill“. Proc. Of the 2nd Int. Conf. On Die and Mold Technology, Singapore, Sept., pp 48-56. Winfough, W. R. and Smith, S., 1995, “Automatic selection of the Optimum Metal Removal Conditions for High Speed Milling”, Transactions of NAMRIISME. XXIII, pp 163-166. Kaldos, A., Dagiloke. I. F. and Boyle, A.. 1996. “Computer Aided Cutting Process Parameter Selection for High Speed Milling”, Journal of Materials Processing Technology, Vol. 61, pp 219224. Tlusty, J., Smith, S. and Winfough, W. R., 1996, “Techniques for the Use of Long Slender End Mills in High-speed Milling”, Annals of CIRP, Vol. 45/1, PP Yucesan, G.. and Altintas. Y.. 1996, “Prediction of Ball End Milling Forces”, Journal of Engineering for Industry, Vol.118, pp 95-103. Abrari, F. and Elbestawi, M. A., 1997,“ClosedForm Formulation of Cutting Forces For Ball and Flat End Mills”, Int. J. Mach. Tools Manufact. Vo1.37, No. 1, pp. 17-27. Rigby. P., 1993, “High Speed Milling in the Mold and Die Making Industries”, Diamond and CBN Ultra Hard Materials Symposium ‘93, Ontario, Canada. Konig, W., Komanduri, R., Tonshoff, H., K., and Ackershott, G., 1990, “Machining of Hard Materials”, Annals of CIRP, Vol. 39/1, pp. 417-427. Rai, G., 1993, “The Metallurgy of CBN and its Wear in High Speed Machining of Ferrous Materials”, Machining of Advanced Materials, Proceeding of InternationalConference on Machining of Advanced Materials, Gaithersburg, MD, July, pp. 501-514. Chou, Y. S., Barash, M. M., 1995, “Review on Hard Turning and CBN Cutting Tools”, 1st International Machining and Grinding Conference, Sept. Dearborn, Michigan, MR95-21, pp 951-962. Park, S., Jun, Y., Lee, C., Yang, M., 1993. “Determining the Cutting Conditions for Sculptured Surface Machining”, Int. J. Adv. Manuf. Technol., pp. 61-70. Konig, W., and Wand, Th,. 1987, “Turning Bearing Steel with Amborite and Ceramic” Indust. Diam. Rev., pp 117-120. El-Wardany, T.I., Boelhouwer, A.M., Elbestawi, M.A., 1995, “High Speed Machining of NickelBased Super Alloys with Silicon Carbide Whisker Reinforce Ceramics”, 1 International Machining and Grinding Conference, SME, pp 1-26. Komanduri, R., Flom, D.G., and Lee, M., 1984, “Highlights of the DARPA Advanced Machining Research Program”, Proceedings of the Symposium on High Speed Machining-PEDVol. 12, ASME WAM, New Orleans Louisiana, pp 287-307.
141
151
’‘
Abstract This paper presents an experimental investigation of high speed milling of dies and molds. Several critical issues involved with the high speed milling of H13 tool steel of hardness up to 55 HR have been studied and explained from a detailed analysis of experimental observations. The experiments were performed using several grades of PCBN ball-nose end mills with various edge preparations. The effect of different process parameters on the tool performance and the surface finish produced was also investigated. The cutting parameters involved were; cutting speeds in the range of 220 to 1320 m/min, feed variation from 0.0254 to 0.1 mm/tooth, axial depth of cut from 0.625 up to 2 mm, and radial width of cut of 0.254 mm. During the preliminary experimental investigation, the tilt angle was kept constant at 10 degrees. Several tests were conducted to study the effect of the different tool path directions on the cutting tool performance. Dry and wet cutting conditions were used and the effect of coolant on the tool life was also determined. The optimum cutting conditions have been specified based on the modes of tool failure, tool life and surface integriv produced. Keywords:
milling, dies and molds, Cubic Boron Nitride (CBN)
Introduction Dies and molds are utilized to produce a large number of discrete components to near net and final shape [l]. The functional parts of the die often include sculptured surfaces. The intricate geometry of these sculptured surfaces and the relatively high degree of hardness of the die materials used necessitates the use of advanced technology in their manufacture. This technology employs new NC tool path generation methods and high speed milling of heat treated die steels using CNC milling machines equipped with proper controlling sensors (21. With the development of super-hard cutting materials such as cubic boron nitride (CBN), poly crystalline cubic boron nitride (PCBN) and the like, the technology of high speed machining (HSM) of hardened steels has created considerable interest to several leading die and mold manufacturers. However, this technology is rather new and a lack of process knowledge and appropriate cutting edge geometry have impeded the broad application of HSM thus far [3]. It is believed that the use of HSM in the die and mold industry can reduce the machining time (4, 51, produce an improved workpiece surface quality and also provide longer tool life [6). Several issues involving high speed milling of dies and molds have been recently studied; optimization of the tool paths generated and cutting conditions, the metal removal efficiency, the tool life and process stability [2 to 91. Ball end milling cutters are used extensively in machining parts with sculptured surfaces. There has been significant research reported in modelling the mechanics of ball end milling [lo,I l l aiming for a longer tool life, where breakage of the tool shank, and chipping of ball shaped helical flutes are the critical factors in ball end milling of die alloys [lo1. PCBN ball end mill tools have been recently used in high speed milling of dies and molds. The workpiece materials were heat treated alloy steels of hardness 30 to 45 HR, [6,12].The suggested cutting speeds were between 500 to 1000 m/min. Also, ultra high spindle speeds (50K1 OOK rpm) allow higher feed rates of up to 10 m/min. This shortens the machine cycle or tightens the pick feeds to improve the surface finish produced. However, the integrity
Annals of the ClRP Vol. 46/1/1997
of the surface produced has yet to be investigated. At such high speeds, the nature of the chip formation is completely modified. The chip formation mechanism is expected to vary as the workpiece hardness increases. Also, different tool wear mechanisms may be observed at this high cutting speed due to the amount of heat generated and dissipated during cutting. This paper examines the cutting performance of PCBN tools during high speed semi finish and finish milling of H 13 tool steel of hardness up to 55 HR,. The type of tool wear, the dominant wear mechanism, and tool failure modes are studied. The effect of different process parameters on the cutting forces, chip morphology, and metallurgical surface integrity produced are also presented. Experimental Procedure . The workpiece material was H13 tool steel, heat treated to 45 and 55 HR,. The workpieces were prepared in the form of 50x100~300mm blocks. The cutting tools were 12.7 mm diameter single tooth ball nose end mills to avoid the influence of tool run-out on wear measurements. Two different grades of PCBN cutting edges brazed on solid carbide shanks were used. Grade 1 was a high volume fraction CBN (90%) with metallic binder and grade 2 was a low volume fraction CBN (65%) with ceramic binder. Three different edge preparations (sharp, honed with a 0.025 mm radius, and chamfered with a 20’ angle) were used. The rake angle was kept at -lo’,according to [6.12]. A 10’ tilt angle was used as recommended by several investigators [4,6]. The pick feed (fp) was kept constant for all cutting tests at 0.25 mm. The cutting speeds (V) were evaluated for the highest point of engagement on the ball end mill in the range of 220 to 1320 dmi n. Table 1 lists the selected variables and their respective values. The cutting force components (Fx. Fy, Fz) were measured using a force dynamometer. The length of each cutting path was approximately 300 mm and at the end of each 10” pass the tool was examined for chipping, breakage, notch wear, or material sticking on the tool rake face. The tool wear was measured using a tool maker’s microscope. The parameters measured to represent the progress of the wear were; average and maximum tool wear
57
Table 1 Cutting parameters used in the experiment Parameter Spindle speed (N) (revlmin) Feed (f) (mmltooth) Axial depth of cut (ADC) (mm)
Values
1 1 1 :3: 1 10,000
60,000
0.025
0.05
0.625
1.125
Type of cut
remain undeformed at the tool exit, because of the thinning phenomenon. Examining the chip cross section microstructure reveals an extensive shear zone with subsurface plastic flow. The expected saw tooth chips (characteristic of hard machining) were not observed because the cracks initiated at the free surface of the workpiece were terminated immediately due to plastification.
Flood coolant
PCBN grades Edge preparation ~
~~
I
sharp
Workpiece hardness HR, Cutting path
I
45 Q
honed
I
K-land
I
a) Top of chip b) Bottom of chip Figure 2 Chips produced at 10,000 rpm.
55
to fp
!I to fp
VB, and VB,, respectively. The position of the maximum tool wear on the cutting edge was also recorded. The surface roughness of the workpiece was measured at several locations along the length of cut using a portable stylus-type instrument. The deviation of the milled surface from the normal surface was measured for evaluating the effect of the tool wear on the geometry of the surface produced. To study the chip geometry, some chips were collected from each cutting test and mounted in an epoxy metallurgical mount, polished, etched, and examined by optical microscopy at magnifications lOOX and 1600X. In addition, some of these polished and mounted samples were also examined by Scanning Electron Microscopy (SEM) using back scattered electrons analysis to determine the region of plastic deformation. Samples of the machined surface at the start of cutting and at the end of tool life were taken from the block for examination of surface integrity under the optical and scanning electron microscope. Experimental Results and Discussion 1. ChiD formation durino hioh m e e d milling Owing to the limited period of engagement of the
cutting edge, the chip produced was short, completely segmented, and of variable thickness caused by the ball end mill geometry. The thinning of the chip in both axial and radial directions is illustrated in Figure 1.
figure 3 shows an SEM image of the chips produced when a spindle speed of 60,000 rpm is used under the same cutting conditions. The high heat generated at these speeds caused the thinner part of the chip to melt and form small spheres (refer to Figure 3). Fig. 3-b illustrates the recrystallization of this melt upon quenching. The diameter of !hese spheres are some tenths of a millimeter. Increasing the feed may reduce the possibility of the generation of this morphology. However, with the progress of tool wear, the melting process of the chips was observed. The cutting edge geometry was found to have a large influence on the production of the spheres.
a) Spherical chip b) microstructure of sphere Figure 3 Chips produced at 60,000 rprn. In general, serrated chips were produced at this high speed and the tendency of generating saw toothed chips was clear as shown in figure 4.
l O O x Magnification
Figure 1 Cross section showing radial chip thinning.
a) Top of chip b) Bottom of chip Figure 4 Morphology of chips produced at 60,000 rpm.
Straw coloured chips were produced when end milling workpieces (45 HR,) at a spindle speed of 10,000 rpm. On the other hand, dark blue chips were produced during the machining of the 55 HR, workpieces. For the cutting conditions of N =10,000rpm, f = 0.025 mdtooth, ADC = 0.625 mm and fp = 0.25 mm, the chips produced are shown in Figure 2. The optical microscope image shows that the chip deformed upon tool entrance. However, the chips
2. Modesof Tool failure In semi finish and finish machining of hardened steel, tool life is taken to be the time needed for the surface finish to deteriorate to a predetermined level. However, in finish machining of dies and molds, it would be misleading to rely solely on the surface roughness to evaluate the tool life of the PCBN tool [I]. A worn tool may impart a better surface
58
finish than a fresh one (fresh inserts cut well defined scallops resulting in a higher surface roughness). Since worn inserts also have an offset cutting edge, they leave stock on the surface that has to be removed manually [4]. In die and mold manufacturing, this type of surface defect should not exceed 0.025 mm and it mainly occurs when the maximum flank wear on the tool exceeds 0.05 mm [4]. Hence. the tool life in this study was ended when the level of the maximum flank wear reaches 0.075 mm. Different modes of tool failure were observed and are listed as follows; a) Modes of failure of High CBN tools: Grade 1 tools failed primarily by reaching maximum specified flank wear. However, it should be noted here that for a higher chip load such as feed of 0.1 mm/tooth, axial depth of cut of 2.0 mm, or when using honed and chamfered cutting edges. plastic deformation of !he tool nose has been observed repeatedly. This was found to accelerate the tool wear and cause a faster end to the tool life. An example of this plastic deformation can be observed from the SEM image in Figure 5. Tool chipping was only noticed when using a honed cutting edge at low feed, and catastrophic tool failure occurred when flood coolant was used.
develop. The tool nose in the vicinity of the cut deformed, and chipping of the cutting edge was initiated. Figures 7 a) and b) show two SEM images of the tool wear regions. Figure 7 a) corresponds to the wear developed on the cutting edge at a spindle speed of 10,000 rpm. The figure shows the evidence of material sticking and fine grooves formed on the rake face, and nose deformation. The very smooth surfaces on the rake face near the cutting edge shown in the SEM image may indicate that the binder material (cobalt) in high CBN tool diffuses into the chips produced. This may lower the bonding strength of the tool, allowing the pull out of some CBN particles causing abrasive action on the flank. In some cases notch wear was also observed. Notch wear is caused by both the sliding of chips and CBN grains. Figure 7 b) shows an SEM image of the tool wear produced during milling at a spindle speed of 60.000 rpm. A protective layer (liquid phasej is deposited on the cutting edge. This protective layer acts as a buffer between the cutting edge and the chip. Further adherence of the material has also been observed located at intermediate cutting velocities. At the highest contact point (highest velocity), material adhesion was not observed. At the extreme point of contact between the tool and the workpiece, the high temperatures generated during machining deforms the softer matrix of the workpiece. leaving the carbide material intact. Thus the possibility of abrasion wear may exist from these carbide particles.
Figure 5 Plastic deformation and chipping of grade 1 tool Modes of tool failure of low CBN tools: b) Grade 2 tools failed primarily by edge chipping. Chipping occurred irrespective of the type of edge preparation or cutting conditions used. For honed cutting edges and low feeds, massive chipping was observed after the first cutting pass. For sharp cutting edges, tool wear developed uniformly on the flank until its level was 0.04 mm. Then, plastic deformation started on the tool nose leading to tool chipping. Material sticking on the rake face side of the chamfered cutting edges always led to catastrophic failure of the tool at low feed, as shown in Figure 6. Severe depth of cut notch wear repeatedly occurred when a honed cutting edge was used with flood coolant.
a) material sticking b) chipping Figure 6 Material sticking and chipping of grade 2 tool. 3. Tool wear mechanisms In the early stages of cutting the tool started to wear at the point where the maximum cutting speed was achieved. With the progress of the cut, uniform tool wear developed along the whole length of contact with the workpiece. Depending on the cutting conditions and the cutting edge preparation, plastic deformation on the rake face may
a) wear at low cutting speed b) wear at high cutting speed Figure 7 Different tool wear mechanisms observed 4. Effect of Drocess variables on cuttina forces
Figure 8 shows the force components measured while end milling at 10,000 rpm spindle speed. As can be seen, the higher force component is in the thrust direction (FJ. Recalling that the workpiece was inclined at loo, Fz contains components from Fy. Irrespective of the resolution of forces, Fz is considerably higher than the others. This phenomenon can be related to the hard machining process where a pronounced increase in the thrust component is always observed when cutting at low feeds and depths of cut [17]. The use of negative rake angles and a large nose radius generates a ploughing force component which makes the magnitude of the thrust force component exceed that of the tangential component [18]. This is expected since the axial and radial depths of cut were very small and the ploughing process is dominant. Average resultant forces versus cutting time for different spindle speeds, feeds, axial depths of cut and edge preparation were investigated. The average force (within one tooth period) was calculated as the square root of the sum of the squares for the three cutting force components Fz, Fx and Fy. A reduction of the average cutting force occurred with the increase in cutting speed which can be related to chip thinning. The force component caused by the rate of change in momentum of chip material is negligible at the range of cutting speeds used in this study. The effect of chip material momentum on the cutting force signal was found to be significant only at cutting speeds above 9000 m/min [ 191. It should be noted here that the use of a magnetic bearing spindle for the higher speed machining induces large dynamic force components which were four times the actual
59
0.
h
5
h
v;
V'
LL a
500,
I
"0
0.005
0.015
0.01
0.02
Cutting time (sec) Figure 8 Force pattern generated during end milling. N=10,000 rpm. f=0.025mm/tooth, ADC=0.625mm, Grade 1
5.Effect of Drocess variables on tool life Figure 11 shows the effect of edge preparation, flood coolant, tool material grades and tool path on the performance of the PCBN tools. The best tool performance was obtained when using a high CBN content cutting tool material with a sharp edge preparation. The high CBN grade performed better than the low CBN grade under the given conditions except when flood coolant was used. Since the high CBN cutting tool has a higher apparent thermal conductivity (100 W/m/K), the heat generated during cutting can dissipate rapidly from the cutting zone, thereby diminishing the produced "hot machining effect" leaving the material difficult to cut [13]. Hence, using flood coolant, a 10fold reduction in tool life is observed. Further, CBN tools (as with most ceramic materials) have poor thermal shock resistance and, thus, flood coolant application results in rapid tool failure. 5000 y - ~ - - -
cutting force. This must be taken into consideration as it will affect the surface finish. The results from Figure 9 indicate that increasing the feed produces a more stable cutting process. On the other hand, the cutting edge preparation has a significant effect on the cutting forces generated (Figure 9).The cutting forces were the highest when honed edge was used. The large negative effective rake angle of honed edges result in more ploughing action instead of cutting, which increases the friction force component.
800
-L
I !
600
L
,
& 400
I jI
Chamfered, f=.025mm/tAoth rn Honed, f=.O25mm/tooth ' I I -
I
I
l
l
0 0
I
0.5
1
1.5
2
Cutting time (sec) Figure 9 Effect of feed and edge preparation on milling forces (N=10,000 rpm, Grade 1) Increasing the axial depth of cut increases the average resultant forces as the chip width increases (Figure 10). An increase in cutting force was observed for a decrease in hardness of 10 HR, (Figure 10).
800
-600 z
I I DADC=0.625
7
mm (45HR,), A ADC=0.625 mm/ ADC=1.25 mm, v ADG2.0 mm (55HR,)
L $, 400 ?!!
0
2 200 0 0.5 1 1.5 2 Cutting time (sec) Figure 10 Effect of axial depth of cut and workpiece hardness on milling forces (N=10,000 rpm, Grade 1) 0
60
h
4000 r ~3~~~ Y
1
4L,
!
1
2 Chamfered edge 3 Honed edge (Dry) 4. Honed edge (Wet) 5 Honededge 1 (Cut along direction of inclination)
2
3
4
5
Figure 11 Effect of process parameters on tool life. N= 10,000 rpm, f = 0.25 mm/tooth, ADC = 0.625 mm
i
!!? I
1
1 Sharpedge
The high CBN tool outperformed the low CBN tool because of its higher transverse rupture strength. The microstructures of the high CBN tools show that the void between the CBN particles is filled with a pool of liquid metal infiltrating from the molten tungsten carbide interface [14]. The liquid metal pool and the high strength carbide substrate provide additional mechanical support to the polycrystalline CBN layer, which gives high CBN tools favourable impact strength for use in interrupted cutting conditions. The honed cutting edges gave the worst tool performance. This was unexpected since this edge preparation is usually employed and recommended for finish cuts. This is consistent with the results obtained by Chou et al. [15]. This can be attributed to the large negative effective rake angle of honed edges which result in more ploughing action instead of cutting. This leads to higher friction and tool wear. The tool life of chamfered cutting edges was less than that of sharp edges. The effective negative rake angle in the case of chamfered cutting edges was -30°,which caused extensive wear on the rake face noted by a steepening of the tool edge, and chipping. A noticeable increase in the tool life was obtained when upward cutting was used where the tool axis was inclined in the feed direction with respect to the surface normal. These results were expected since this type of tool path gives the minimum tool-workpiece contact area, as shown in [15]. Also, this type of tool path prevents any engagement between the tip of the ball nose end mill and the workpiece surface. It is well known that in the centre of this tool the cutting speed equals to zero and thus, the effective chip velocity is very small. This leads to cutting edge chipping and a higher wear rate due to a ploughing effect. Figure 12 shows that wear on high CBN tools decreases as the workpiece hardness increases. Increasing the cutting speed has a limited effect on the tool life as a function of the cutting length when machining the harder H13
workpieces (refer to Figure 12). Conversely, when machining the less hard material, the tool life was reduced from 3750m to 1000 m lengths of cut. The variation of chip morphology as a function of workpiece hardness indicates that more heat is generated during the machining of 55 HR, tool steel. The higher the heat generation, the more the chip will be softened and the better the workpiece plastification. This will reduce the mechanical load imposed on the tool and, therefore, less tool wear will exist. 0.12 v
A c
45HRc 55HR,
- 10000 rpm
---
60000 rpm
0.08 Y
x
5
2 0.04 >
0
0
1000
2000 3000 Cutting length (m)
4000
5000
Figure 12 Effect of hardness and speed on tool wear. f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254, dry cut, sharp edge, Grade 1 tool Tool life was reduced as the axial depth of cut was increased due to larger chip load as shown in Figure 13 a). Increasing the axial depth of cut will increase the maximum cutting speed. The Increase in the cutting speed generated high thermal loads on the tool, which when combined with intermittent cuts, accelerated tool wear because of the thermal shock. Figure 13 b) shows the effect of feed on the tool wear for the cutting edge of both PCBN grades. In general, the increase in the feed reduced the level of tool wear. Since the sliding contact length between the tool and the workpiece reduces proportionally to the increase in feed, tool wear is also expected to be reduced. On the other hand, increasing the feed shortens the heating cycle which will be interrupted during its transient phase before reaching steady state. The peak temperature reached will be lower than the corresponding steady state value. However, the use of very
low feeds with the chamfered edge produces vibration on the system, which leads to a higher wear rate of high CBN and a catastrophic failure of the low CBN. 6.Effect of Drocess parameters on the surface intearity produced: The feed rates used combined with the high cutting speed and large nose radius consistently produced very good quality surface finishes. Geometric error was found to be in the range of 0.025 mm and it increases to 0.075 mm as the tool wears. The range of surface roughness varied from 0.2 pm ( fresh edge) to 0.6 pm (very worn tool) Ra in all cases of the 10,000 rpm speed (e.g. Figure 14 a). Rmax was found to be lower for the harder material (2.0 pm) as opposed to less than 4.0 pm for the 45 HR, workpiece. When the magnetic bearing spindle speed was used at 60,000 rpm, the maximum roughness obtained was .66 pm Ra, resulting from the vibration induced to the cutting system by the flexibility of the magnetic spindle. A noticeable increase in surface roughness was not realised as the cut off value of the flank wear is very low for finishing. Hence, the end of tool life was very difficult to gauge from the surface roughness. The value of the roughness when the tool catastrophically failed was obviously very high. At the high spindle speed, Rmax was found to vary contrary to what was observed at low spindle speeds i.e. increasing hardness increases the roughness (Figure 14 b). 2O I OOO rpm G - m 60000 rpm
1
h
.
3
I
0
0
.
-
I
1000 2000 3000 Cutting Length (m)
4000
5000
f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254 mm, dry cut, sharp edge Figure 14 a Effect of speed o n surface finish. h
f 4
h
E
v
1.27 mm ADC
0
1000
2000
3000
5000
4000
Cutting length (m) N=lO,OOO rpm, fz0.0254 mm/tooth, RWG0.254 mm a) effect of ADC
_~._. -.
5 T
3
u)
O
1000
2000
3000
4000
5000
Cutting Length (m)
.--,
,
N=lO,OOO rpm, f=0.0254 mm/tooth, ADG0.635 mm, RWC=0.254 mm, dry cut, sharp edge Figure 14 b Effect of hardness on surface finish.
E 0.04
03
> 0
1000
2000
3000
4000
5000
Cutting length (m) N=10,000 rpm, f=0.05 mm/tooth, RWG0.635 mm b) effect of feed Figure 13 Effect of axial depth of cut and feed on tool wear.
Metallographic examination of the workpiece subsurface revealed that a damaged layer exists a few microns under the cut surface and has a thickness of approximately 4-6 pm. This damaged layer is evident in all of the cutting conditions. It is metallurgically different than the fine martensitic structure (marked by a difference in contrast) of the bulk material refer to Figure 15 a). The workpiece hardness has a great influence on the observable
61
structure of the damaged layer. The 45 HR, workpiece revealed a relatively coarse martensitic structure far under the surface (>20 pm). However near the cut surface, it is apparent that the martensite grains are clearly deformed in the direction of chip flow, see Figure 15 b). Figure 16 shows SEM images of the microstructure of the subsurface damage produced at different cutting conditions. The thickness of the damaged layer was found to be dependant on edge preparation, and tool wear. Interestingly, increasing the axial depth of cut and the feed has no significant effect on the thickness of this layer (an example of which is shown in Figure 16 a) and b).
-
a) surface of 55 HR, b) surface of 45 HR, Figure 15 SEM images of workpiece subsurface damage.
a) ADG0.625 mm b) ADC = 2.0 mm Figure 16 Effect of axial depth of cut on subsurface damage.
Conclusions High speed hard milling tests using PCBN tools have been performed on H13 tool steel materials. From the analysis obtained, the following conclusions can be drawn: 1- Ultra high speed milling of hardened steel is feasible under the given cutting conditions. The geometric error was within acceptable limits. 2-A large chip load yields a more stable cutting process, and higher tool life. The surface quality and integrity are not significantly affected. 3- Contrary to the tool manufacturers specifications, sharp edge preparations outperform the honed and chamfered edges. 4- The experimental results yield that high volume fraction (goo/,) CBN tools are recommended for milling hardened tool steel. The main mode of tool failure for high CBN tools was classical flank wear. Consequently, low volume fraction CBN tools fail generally by chipping. 5- The higher the cutting speeds the thinner the chips produced. This leads to the production of spherical chips caused by material melting and recrystallizing.
References:
[ 11
[ 21
62
Altan, T., Lilly B. W., Kruth, J. P., Konig, W., Tonshoff, H. K., Van Luttervelt, C. A. and Khaity, A. B., 1993, “Advanced Techniques for Die and Mold Manufacturing”, Annals of CIRP, Vol. 42/4, pp 707. Takeuchi, Y., Idemura, T., Sata, T., 1991, “5-Axis Control Machining and Grinding Based on Solid
Model”, Annals of CIRP, Vol. 40/1, pp 455-458. Fallbohmer, P., Altan, T., Tonshoff, H. K. and Nakagawa, T., 1996, “Survey of the Die and Mold Manufacturing industry - Practices in Germany, Japan, and the United States”, Journal of Materials Processing Technology, Vol. 59, pp 158-168. Kruth, J. P. and Klewais, P., 1994, ”Optimization and Dynamic Adaptation of Cutter Inclination During Five-Axis Milling of Sculptured Surfaces”. Annals of CIRP, Vol. 43/1, pp 443-448. Schultz, H. and Hock, St.. 1995, “High-speed Milling of Dies and Molds Cutting Conditions and Technology”, Annals of CIRP. Vol. 44/1, pp . Ikeda, T., Takahashi. I., Matsuoka, T. and Nakagawa, T., 1992. “Ultra High Speed Milling of Die Steel with Ball-nose End mill“. Proc. Of the 2nd Int. Conf. On Die and Mold Technology, Singapore, Sept., pp 48-56. Winfough, W. R. and Smith, S., 1995, “Automatic selection of the Optimum Metal Removal Conditions for High Speed Milling”, Transactions of NAMRIISME. XXIII, pp 163-166. Kaldos, A., Dagiloke. I. F. and Boyle, A.. 1996. “Computer Aided Cutting Process Parameter Selection for High Speed Milling”, Journal of Materials Processing Technology, Vol. 61, pp 219224. Tlusty, J., Smith, S. and Winfough, W. R., 1996, “Techniques for the Use of Long Slender End Mills in High-speed Milling”, Annals of CIRP, Vol. 45/1, PP Yucesan, G.. and Altintas. Y.. 1996, “Prediction of Ball End Milling Forces”, Journal of Engineering for Industry, Vol.118, pp 95-103. Abrari, F. and Elbestawi, M. A., 1997,“ClosedForm Formulation of Cutting Forces For Ball and Flat End Mills”, Int. J. Mach. Tools Manufact. Vo1.37, No. 1, pp. 17-27. Rigby. P., 1993, “High Speed Milling in the Mold and Die Making Industries”, Diamond and CBN Ultra Hard Materials Symposium ‘93, Ontario, Canada. Konig, W., Komanduri, R., Tonshoff, H., K., and Ackershott, G., 1990, “Machining of Hard Materials”, Annals of CIRP, Vol. 39/1, pp. 417-427. Rai, G., 1993, “The Metallurgy of CBN and its Wear in High Speed Machining of Ferrous Materials”, Machining of Advanced Materials, Proceeding of InternationalConference on Machining of Advanced Materials, Gaithersburg, MD, July, pp. 501-514. Chou, Y. S., Barash, M. M., 1995, “Review on Hard Turning and CBN Cutting Tools”, 1st International Machining and Grinding Conference, Sept. Dearborn, Michigan, MR95-21, pp 951-962. Park, S., Jun, Y., Lee, C., Yang, M., 1993. “Determining the Cutting Conditions for Sculptured Surface Machining”, Int. J. Adv. Manuf. Technol., pp. 61-70. Konig, W., and Wand, Th,. 1987, “Turning Bearing Steel with Amborite and Ceramic” Indust. Diam. Rev., pp 117-120. El-Wardany, T.I., Boelhouwer, A.M., Elbestawi, M.A., 1995, “High Speed Machining of NickelBased Super Alloys with Silicon Carbide Whisker Reinforce Ceramics”, 1 International Machining and Grinding Conference, SME, pp 1-26. Komanduri, R., Flom, D.G., and Lee, M., 1984, “Highlights of the DARPA Advanced Machining Research Program”, Proceedings of the Symposium on High Speed Machining-PEDVol. 12, ASME WAM, New Orleans Louisiana, pp 287-307.
141
151
’‘