Procedia Manufacturing
Volume XXX, 2015, Pages 1–11
43rd Proceedings of the North American Manufacturing Research
Institution of SME http://www.sme.org/namrc
High Speed Milling Processes with Long Oblique
Cutting Edges
Hideaki Onozuka, Koji Utsumi, Ippei Kono1*
Junichi Hirai2*, Yasuhiro Numata3* and Toshiyuki Obikawa4*
1
Yokohama Research Laboratory, Hitachi, Ltd., Yokohama-shi, Kanagawa, Japan
2
Hitachi Works, Hitachi, Ltd., Hitachi-shi, Ibaraki, Japan
3
Hitachi Works, Mitsubishi Hitachi Power Systems, Ltd. Hitachi-shi, Ibaraki, Japan
4
Institute of Industrial Science, The University of Tokyo, Meguro-ku, Tokyo, Japan
hideaki.onozuka.cf@hitachi,com
Abstract
High speed end milling processes using an inclined end mill with long oblique end teeth was
investigated to minimize the scallop height of the finished surface. The optimal cutting conditions
were found to improve the tool life and machining accuracy. According to the results of cutting tests,
the cutting force decreased with increasing cutting speed due to the change in chip formation.
Although the cutting temperature increased with cutting speed, it was found that the change in the tool
wear mechanism with cutting speed minimized the tool wear at a higher cutting speed. Moreover,
although the surface roughness of the machined surface deteriorates due to the tool wear when cutting
length is increased, it is found that the deflection of the tool increases due to the increase of cutting
forces and it improves the surface roughness.
Keywords: High speed end milling, Tool wear, Surface roughness, Cutting temperature, Cutting force
1 Introduction
Development of the machine tools with a high speed spindle and table and the improvement of the cutting
performance of tool materials have dramatically increased the cutting speed and material removal rate.
Furthermore, the widespread multi-axis machine tools permit the high efficiency and high precision
machining of products with complex geometry. For example, it is possible to machine free surface of a product
by controlling the trajectory and the posture of a ball end mill (Muraki and Yamamoto, 2012). Additionally,
turning with an end mill so called turn-milling is proposed instead of turning with a single point cutting tool
(Muraki and Yamamoto, 2003, Savas and Ozay, 2007, Schulz and Kneisel, 1994, Filho, 2001). This method
prevents troubles often caused by a long continuous chips produced in turning process and permits the
*
Masterminded EasyChair and created the first stable version of this document
Selection and peer-review under responsibility of the Scientific Programme Committee of NAMRI/SME
c The Authors. Published by Elsevier B.V.
1
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
automation of machining processes. However, height of cusps generated on the curved surface machined with
a ball end mill or a corner radius end mill is often larger than the height of feed marks in turning with a single
point tool.
For example, as shown in Figure 1, the cusp height of the surface machined by a ball end mill with the radius
of Rc is theoretically expressed as the following equation;
ܴ୲୦ f 2
(1)
8c
where pf is the pick feed of the end mill.
On the contrary, by inclining the tool revolution axis in order that the oblique linear cutting edge becomes
parallel to the pick feed direction, it is possible to reduce the cusp height and improve the surface roughness. In
Figure 2, the linear cutting edge is inclined at angle θ from the rotation surface. The tool revolution axis is
inclined at the same angle to the feed direction, and the pick feed pf is smaller than the cutting edge length l.
This kind of cutting tool can be used to reduce the surface roughness of a cylindrical surface. However, this
cutting tool generates the surface by transcribing the linear cutting edge, so a change in the cutting edge
geometry due to tool wear affects surface roughness or machining accuracy. Accordingly, the present research
aims to clarify the effects of cutting conditions on tool wear and the effects of tool wear on surface roughness
in the machining process by using an end mill with the long oblique end cutting edge in order to improve the
efficiency and accuracy of the machining of curved surface.
Figure 1. Cusp height of surface machined
with a ball end mill.
Figure 2. Machining of curved surface with an oblique
Edge cutting tool.
2 Experimental Configurations
Cutting tests were conducted by milling an oblique surface of a work piece. As shown in Figure 3, the
surface was inclined at angle θ =15 degrees and parallel to the oblique cutting edge of the end mill. The work
piece was mounted on a dynamometer, and cutting forces generated during the tests were measured. Depth of
cut normal to the surface is represented by d mm, and feed direction of the end mill is in the x direction. Fx, Fy,
and Fz are the cutting forces in the feed direction, cross-feed direction, and axial direction respectively. The
dynamometer used in this experiment is Kistler 9257B. Cutting force signals from the dynamometer was
amplified with a charge amplifier and analyzed with a digital recorder. The work piece material is 13 Cr steel
(hardness HB 241) with length of 110mm and width of 100mm. Diameter of the cutting tool was 25mm,
oblique angle of the linear cutting edge θ was 15 degrees, and approach angle ψ was 30 degrees. Tool wear of
the cutting edge after the cutting tests was observed by microscope and evaluated by VBc (namely, corner
wear), VBmax1 (namely, wear of the oblique linear cutting edge), and VBmax2 (wear of the principal cutting edge).
2
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
The machined surface is measured by a surface roughness tester Talysurf 5 manufactured by Taylor Hobson,
Ltd. The geometry of the cutting tool used for the cutting tests is shown in detail in Figure 4.
Inserts made from coated tungsten carbide were mounted in the body of a steel cutter body.
Cross point of the linear oblique cutting edge and principal edge is the position of the diameter
of 21 mm. The accuracy of the oblique angle of the cutting edge on the machine tool was within
15 degrees ± 1minute on machine tool. The length of the cutting edge is l=4.6mm and the
radius of the corner R is 0.2 mm.
Figure 3. Experimental setup for the measurement
of cutting forces and tool wear.
Figure 4. Geometry of cutting tool.
In Figure 5, since pick feed pf must be smaller than the length which the corner radius R subtracted from the
cutting edge length l , it follows that;
‫݌‬୤ ℓ tanሼሺ180 ሻ/2ሽ
(2)
From this equation, pf < 4.5 mm. The pick feed was set to 4 mm. Then, temperature of the cutting edge during
the cutting process was measured by using a covered constantan wire embedded in the work piece as shown in
Figure 6. A 0.15 mm wide slot was machined on the work piece, and the enamel covered constantan wire with
width of 0.076 mm was embedded in the slot. This wire was machined at the same time as the work piece, and
a thermoelectromotive force was generated by the conduction with the burr of work piece when the cutting
edge passes(Iwata, Moriwaki and Okuda, 1987, Shamoto, Watanabe, Tanabe and Moriwaki,2002, Fujimura,
Kawabata and Shintani, 1973, Dewes, Ng, Chua, Newton and Aspinwall, 1999).
Figure 5. Geometry of cutting tool.
3
Milling with Long Oblique Cutting Edges
Figure 6.
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
Experimental setup for measuring cutting temperature of oblique milling tool.
Since the thermoelectromotive force when the cutting edge passes is small, it is necessary to electrically
isolate the work piece and the wire from the machine tool in order to avoid the influence of the noise. The
work piece was thus put between ceramic plates and mounted in a machine vise. Furthermore, the vertical
machining center used for this experiment uses a ceramic bearing for the tool spindle, so the machine tool and
work piece do not conduct through the cutting tool.
The above described experimental configurations are specified in Table 1. Depth of cut was 0.5 mm normal
to the machine surface, feed rate of the end mill was 0.3 mm/tooth, and pick feed was 4 mm. The feed
direction was down milling and soluble cutting fluid was provided.
3 Experimental Results
3.1 Relationship between Cutting Conditions and Tool Wear
The relationship between cutting velocity and tool wear after cutting a length of 11 m is shown in Figure 7.
According to the figure, corner wear VBC is the largest. Generally, tool wear increases as cutting velocity
increases.
Table 1. Experimental configurations
Machine tool
Vertical machining center
Cutting tool
Oblique edge end mill
TiCN coated cemented carbide
Diameter φ 25 mm
Work piece
13 Cr Steel
Cutting conditions Cutting velocity 100-800 m/min
(Spindle revolution 1273-10191 min-1)
Feed rate 0.3 mm/tooth)
Pitch of helical feed 4 mm
Depth of cut 0.5mm
Down milling
Cutting fluid
4
Soluble
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
This result indicates that Vc is increased by the increase of the cutting velocity. However, VBmax2 and VBmax1
are minimum at cutting velocity of 200 m/min and 400 m/min, respectively. Additionally, they increases
steeply at higher cutting velocity (i.e., more than 600 m/min). Since the flank wear of the oblique cutting edge
VBmax1 is considered to affect the surface roughness directly, cutting velocity of 400 m/min is the optimal
condition. Observations of the tool wear taken with a scanning electron microscope are shown in Figure 8.
Flank wear mm
0.5
VBc
0.4
0.3
VBmax2
0.2
VBmax1
0.1
0
200
400
600
800
Cutting velocity V m/min
1000
Figure 7. Relationship between cutting velocity and
flank wear.
Figure 8. SEM observations of tool wear.
The cutting forces at the beginning of cutting are plotted in Figure 9. Fx (namely, the force in feed direction)
and Fy (namely, the perpendicular force to the feed direction) decreases with increasing cutting velocity. Fz
(namely, the force in the axial direction) decreases when cutting velocity increases from 200 m/min to 400
m/min, but it increases as cutting speed exceeds 600 m/min. The reason that the cutting force decreases due to
increasing velocity is that the material strength decreases with increasing temperature at high cutting velocity.
However, tool wear increases above 600 m/min, so Fz is increased by increased ploughing force (Senba,
Taguchi, Sakuma and Hozumi, 1991) at the principal cutting edge.
Cutting force N
800
600
Fx
Fz
400
200
Fy
0
200
400
600
800
Cutting velocity V m/min
1000
Figure 9. Relationship between cutting velocity and flank wear.
5
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
3.2 Relationship between Cutting Conditions and Cutting Temperature
The machined surface of test piece used for temperature measurement (shown in Figure 5) is shown in Figure
10. On the machined surface, work material flew toward the cutting direction. As shown in this figure, a hot
junction is formed at the boundary of the work piece and constantan wire. An example of the measured
electromotive force signal at depth of cut of 0.1mm, feed rate of 0.05mm/tooth, and cutting velocity of 600
m/min is shown in Figure 11. Output voltage of the test piece rises according to the passage of the cutting edge
and lowers gradually.
Figure 10. Hot junction on machined surface.
Figure 11. Measured electromotive force due to
passage of cutting edge.
Temperature during cutting when cutting velocity was varied is plotted in Figure 12. Cutting temperature is
almost 600 K at cutting velocity of 100 m/min, and it increases to around 1400 K with increasing cutting
velocity.
Profiles of the chips generated when cutting velocity was varied are shown in Figure 13. Although the
profile of the chip at cutting velocity 200 m/min is a continuously formed shape, serrated chips are formed at
cutting velocity above 400 m/min.
Temperature K
1600
1400
1200
1000
800
600
400
200
400
600
800
Cutting velocity V m/min
1000
Figure 12. Relationship between axial depth
of cut, feed rate, and cutting temperature.
Figure 13. Observations of the chip generated by
oblique milling tests.
3.3 Relationship between Cutting Velocity and Surface Roughness
Transition change of machined surface roughness when cutting velocity was changed from 200 m/min to
600 m/min is plotted in Figure 14. Initial surface roughness at cutting length of 2.2 m is much the same
(around 1.5 µm Ry) in all there cutting velocity cases, but it deteriorates to 6.0 µmRy at cutting velocity of 200
m/min and to 5.0 µmRy at the cutting velocity of 600 m/min. However, surface roughness after cutting a
6
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
length of 11 m at 400 m/min is 4.5 µmRy. The profile of the machined surface for each cutting velocity is
shown in Figure 15. The profile was measured in the pick feed direction with a surface roughness tester. This
measured surface profile indicates that curves with 4 mm pitch (which corresponds to the pick feed) are
formed on the surface. As shown in Figure 5, length of the linear oblique cutting edge is l = 4.6 mm. The 4
mm part of the cutting edge (which is the pick feed) retreats due to the wear. The overlapped part (with 0.6
mm length) that is not used for cutting does not retreat by the wear and it forms deep parts on the machined
surface. As shown in Figure 7, since the retreat of the linear oblique edge by wear is small at cutting velocity
of 400 m/min, a smooth surface is obtained. The change of surface profile with increasing cutting length is
shown in Figure 16.
Surface roughness µm
8
V=600m/min
V=200m/min
6
4
V=400m/min
2
0
2
4
6
8
Cutting length L m
10
12
Figure 14 Relationship between cutting length and surface roughness.
Figure 15. Relationship between cutting velocity
and surface profile.
Figure 16. Profile of machined surface at
cutting velocity 400 m/min.
7
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
At the beginning of cutting at length L = 4.4 m, the overlapped part of the cutting edge forms grooves on the
surface. However, the surface profile changes to a saw tooth shape due to the wear of the linear oblique edge
when cutting length increases. Since cutting velocity is relatively high at the part that is close to the corner of
the cutting edge, wear of the cutting edge is larger than that of the overlapped part.
4 Discussion
4.1 Relationship between Cutting Velocity and Tool Wear
Egawa et al. reported that serrated chips are generated by increasing of cutting velocity during the cutting
of hardened steel (Egawa, Ichizaki, Kuroda, Hiasa and Tsukamoto, 1995). In the present cutting tests, end
milling of 13 Cr steel generates serrated chips, and cutting forces decreases at cutting velocity above 400
m/min. On the other hand, cutting temperature rises due to increasing cutting velocity. An optimal cutting
velocity thus exists because the mechanical influence reduces but thermal effects increases due to the
increase of cutting velocity.
4.2 Relationship between Tool Wear and Surface Roughness
As shown in Figure 16, a saw teeth shape profile is formed on the machined surface. The reason is that the part
of the cutting edge close to the corner retreats more than other area because the cutting speed is high. The
maximum retreat of the cutting edge d is expressed as follows in terms of maximum flank wear VB and
clearance angle β as shown in Figure 17.
ߜ = ܸ‫ܤ‬୫ୟ୶ଵ ∙ tanߚ
(3)
Since the cutting force in y direction acts on the cutting edge, the end mill deforms. As shown in Figure 18,
deflection at angle ζ occurs due to the cutting force. Therefore, the corner of the cutting edge displaces by εz,
and it set off the retreat of tool wear. As a result, the generation mechanisms of surface roughness is explained
by the retreat of the cutting edge and the deformation of the tool due to the cutting forces.
As shown in Figure 19, displacement of the tool end in the y direction εy due to a load of 1.0 N is expressed
as;
ߝ୷ = ߞ୰ ∙ ‫ ܮ‬+
‫ܮ‬ଷ
3‫ܫܧ‬
(4)
where L is length of the cutting tool, ζr is bending angle at the root of the end mill. E is Young’s modulus of
the end mill material, and I is geometrical moment of inertia.
According to measurement results obtained when the root of the end mill was gripped by a tool holder on
the machine tool spindle, displacement of the tool end is εy = 1.16 x10-7 m/N. Diameter of tool is D = 25 mm,
Young’s modulus of the end mill is E = 2.1 GPa, and the moment of inertia is I = πD4/64.
8
Milling with Long Oblique Cutting Edges
Previous tooth
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
Tool
Pick
feed
pf
l
δ
Retreat of
wiper edge
Work piece Wiper edge
Figure 17 Retreat of wiper edge due to tool wear.
Figure 18 Deflection of wiper edge due to cutting forces.
Figure 19 Deflection angle of cutting tool
due to cutting force.
The bending angle at the root of the end mill is thus calculated as ζr = 4.93x10-8 rad/N. The bending angle
at the end of the tool is expressed as;
୲ ଶ
2
(5)
From these equations, εz as shown in Figure 18 is expressed as;
୸ ୷ ୰ ୲ ∙ ℓ
(6)
Surface roughness R’th is thus expressed as;
ᇱ
୲୦
୸
(7)
Cutting forces in the y direction Fy when cutting length is increased is plotted in Figure 20. As shown in
this figure, cutting force increases as cutting length increases. Moreover, although cutting length is the smallest
when cutting velocity is 600 m/min, cutting force at cutting velocity of 400 m/min is the smallest which tool
wear is the smallest. The relationship between tool wear VBmax1 and cutting force Fy is shown in Figure 21. As
shown in this figure, cutting force is very relevant to tool wear, and Fy is approximated as the following
equation;
୷ 1717 ∙ ୫ୟ୶ଵ 268.5
(8)
By using this equation, surface roughness R’th is calculated using equation (6) and (7) where clearance
angle is β = 6°.
9
Milling with Long Oblique Cutting Edges
V=200m/min
500
500
400
300
Cutting force Fy N
Cutting force Fy N
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
V=400m/min
V=600m/min
200
100
0
2
4
6
8
Cutting length m
10
12
Figure 20 Relationship between cutting length and
cutting force in pick feed direction.
V=200m/min
400
V=600m/min
V=400m/min
300
200
Fy = 1717・VBmax1 + 268.5 (N)
100
0
0.02
0.04 0.06 0.08 0.1
Flank wear VBmax1 mm
0.12
Figure 21 Relationship between flank wear and
cutting force in pick feed direction.
The relationship between tool wear VBmax1 and surface roughness is shown in Figure 22. The retreat of the
cutting edge calculated by equation (3) and the theoretical surface roughness calculated by equation (7) which
the bending of the end mill by cutting force is considered are plotted in this figure. From this result, although
the surface roughness deteriorates from 1.0 µm to 5.0 µm due to increasing cutting length, it is still smaller
than the retreat of the cutting edge which was calculated from flank wear. It is concluded that the displacement
of the cutting edge due to the cutting forces improves the surface roughness.
Surface roughness µm
12
Retreat and deflection
δ −εz
10
8
6
Retreat δ
4
V=600m/min
2
V=200m/min
V=400m/min
0
0.02
0.04 0.06 0.08 0.1
Flank wear VBmax1 mm
0.12
Figure 22 Relationship between flank wear and surface roughness in pick feed direction.
In this experiment, roughness of the machined surface is less than 5.0 µm. According to equation (1), the pick
feed of the ball end mill which has the same diameter 25 mm as presented experiment is 0.7 mm in order to
obtain the same surface roughness 5.0 µm. Therefore, it is possible to increase the pick feed by more than 5
times by using proposed method.
5 Conclusions
To improve machining efficiency and surface roughness of a curved surface such as a cylindrical
work piece, optimal cutting conditions for an end mill with a linear oblique cutting edge were
investigated. The results of cutting tests on an inclined surface are summarized as the following three
points.
10
Milling with Long Oblique Cutting Edges
Onozuka, Utsumi, Kono, Hirai, Numata, and Obikawa
(1)Tool wear of a linear oblique edge is minimum at cutting velocity of around 400 m/min when
cutting velocity is increased from 100 m/min to 800 m/min.
(2)Serrated chips are generated and cutting force decreases when cutting velocity is increased to over
400 m/min and cutting forces decreases. However, the temperature of the cutting edge rises due to
the increased cutting velocity, so an optimal cutting speed exists because of the difference between
the tool wear mechanisms at low and high velocity conditions.
(3)Roughness of the machined surface deteriorates due to the tool wear when cutting length is
increased. However, deflection of the tool increases due to the increased cutting forces, and it
reduces surface roughness.
References
Egawa, T., Ichikizaki, T., Kuroda, M., Hiasa, Y.,Tsukamoto, H., Wear mechanism of cBN tool in
cutting of hardened steel, Journal of the Japan Society for Precision Engineering, Vol.61, No.6,
1995, pp.809-813 (in Japanese).
Filho,J.M.C., Prediction of cutting forces in mill turning through process simulation using five-axis
machining center, International Journal of Manufacturing Technology, 03, 2011
Fujimura, Y., Kawabata, H. and Shintani, K. The temperature distribution at worn surface of tool flank,
Journal of the Japan Society for Precision Engineering, Vol.42, No.4, 1973, pp.268-274(in
Japanese)
Iwata, K., Moriwaki, T. and Okuda, K., Analysis of cutting temperature in ultra-high precision
diamond cutting of copper, Journal of the Japan Society for Precision Engineering, Vol.53, No.8,
1987, pp.1253-1258.
Muraki, T. and Yamamoto, H., Current state and outlook of the multi-tasking machine, Journal of the
Japan Society for Precision Engineering, Vol.78, No.9, 2012, pp.740-743 (in Japanese).
Muraki, T. and Yamamoto, H., Mill-turning by usin multi-tasking machine(lathe with milling
spindle) : The method of high speed dry turning, Journal of the Japan Society for Precision
Engineering, Vol.69, No.7, 2003, pp.965-969 (in Japanese).
R.C.Dewes, E.Ng, K.S. Chua, P.G. Newton, D.K.Aspinwall, Temperature measurement when high
speed machining hardened mould/die steel, Journal of Materials Processing Technology, Vol.9293, 1999, pp.293-301.
Savas, H. and Ozay, C., Analysis of the surface roughness of tangential turn-milling for machining
with end milling cutter, Journal of Materials Processing Technology, 186, 2007, pp.279-283.
Schulz, H., Kneisel, T., Turn-milling of hardened steel – an Alternative to Turning, CIRP Annals,
Vol.43, 1, 1994, pp.93-96.
Senba, T., Taguchi, K., Sakuma, K. and Hozumi, Y., Development and evaluation of a fine finishing
end mill with high rigidity : The effect of a helix angle on the cutting force, Transactions of the
Japan Society of Mechanical Engineers.C, Vol.57, No.533, 1991, pp.313-319.
Shamoto, E., Watanabe, J., Tanabe, T., and Moriwaki, T., Measurement of cutting temperature in
endmilling by utilizing thermocouple : Proposal of measured data rearrangement method and
measured examples, Proceedings of the JSME 77th Annual Kansai Division Meeting, 2002(77), 319 (in Japanese).
11