Pergamon Int J tlech Sci Vol.

38, No, 3, p p 259 269, 1996

Elsevier Science Lid
Printed in Great Britain
0020 7403/96 $15.00 + 0.00

0020- 7403 05) 00057- 7

THREE DIMENSIONAL CUTTING FORCE ANALYSIS

IN END MILLING

LI ZHENG, YUN SHUN CHIOU and STEVEN Y. LIANG

George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology,
Atlanta, GA 30332-0405, U.S.A.

IReeit:ed 13 Septentber 1994; and in ret ised lorm 16 May 19951

Abstract The analysis of cutting forces plays an important part in the design of machine tool
systems as well as in the planning, optimization, and control of machining processes. This paper
presents a three-dimensional model of cutting forces in peripheral end milling in terms of material
properties, cutting parameters, machining configuration, and tool/work geometry. Based on the
relationship of the local cutting force and the chip load, the total cutting force model is established
via the angle domain convolution integration of the local forces in the feed, cross feed, and axial
directions. The integration is taken along the cutter axis and summarized across the cutting flutes.
The convolution integral leads to a periodic function of cutting forces in the angle domain and an
explicit expression of the dynamic cutting force components in the frequency domain. The closed-
form nature of the expressions allows the prediction and optimization of cutting forces to be
performed without the need of numerical iterations. To assess the fidelity of the analytical model,
experimental data from end milling tests are presented in the context of three dimensional time
waveforms, power spectra, and phase angles, in comparison to the values predicted by the model.

NOTATION

dr da radial and axial depths of cut

J~,l,.,J= total cutting forces in the feed, cross feed, and axial directions
Fx. F~, F: Fourier transform of J~,,j~..J~
.~,~,J~ local cutting forces in the tangential, radial, and axial directions
"[~,~,,~ local cutting forces in the feed. cross feed, and axial directions
.~'.~,,',.~' .~.~,.,~ on the ith flute
L',.I,L.i) cutting forces on the ith flute in the feed, cros~ feed. and axial directions
h as defined in Eqn I12)
k, tangential cutting pressure coefficient
k, radial to tangential cutting force ratio
k, axial to tangential cutting force ratio
N number of cutter flutes
tL. t~,. L- as defined in Eqn 19i
P1, Pt. P3 as defined in Eqns 117 lOj
R cutter radius
t~ undeformed chip thickness
f, average chip thickness
t~ feed per tooth
U as defined in Eqn (161
wlOi unit square window function
3~ cutter helix angle
angular position of any cutting point with respect to its flute tip
D angular position of the cutter with respect to a fixed reference
0 angular position of any cutting point with respect to a fixed reference
0~, 02 entry and exit angle
tO t tooth passing frequency in Hz; equivalent to spindle rotational frequency in rps multiplied by the
number of flutes

1 INTRODUCTION

T h e a n a l y s i s o f c u t t i n g forces in m a c h i n i n g p r o c e s s e s h a s b e e n t h e s u b j e c t o f m a n y r e s e a r c h
efforts in view o f its i m p o r t a n c e t o t h e d e s i g n o f c u t t i n g t o o l s a n d m a c h i n e r y . T h e
p e r f o r m a n c e a t t r i b u t e s of m a c h i n i n g p r o c e s s e s , s u c h as d y n a m i c s t a b i l i t y , t o o l - w o r k

259
260 Li Zhcng et al

positioning accuracy, tool condition, machined surface, and part form errors are often
studied on the basis of force analysis [1, 2]. At the advent of modern machine tools, the
fundamental understanding of cutting forces is critical to process diagnosis and servo
controls. Predictive models of forces in relation to process parameters often provides the
key guidelines to the planning, optimization, and automation of machining operations
[3, 4].
The analysis of cutting forces dates back perhaps to the work of Koenigsberger and
Sabberwal [5]. A relationship between chip width and local normal cutting force was
suggested, and on that basis, the force pulsations during slab and face millings were
investigated. Tlusty and MacNeil [2] related the local radial component of cutting force to
its tangential component by a pressure constant, which is found empirically to be tied to the
chip thickness. Yucesan et al. [6] modeled cutting forces using normal pressure coefficient,
friction coefficient, and chip flow angle. From the formulation of local cutting forces, the
total cutting forces were derived based on the integration of local cutting forces with respect
to the cutter orientation.
Most of the integration expressions in the literature have been formulated for single flute
cutting only. In the case of multi-flute cutting, the entry and exit conditions often require the
force expressions to take on different forms at different cutter locations thus rendering the
integration solutions inexplicit. Kline et al. [1] managed to carry out the integration
numerically by discretizing the cutter into small disks along the axis. The effect of cutter
runout was subsequently considered in terms of the cutting kinematics and forces [7, 8].
Sutherland and DeVor [9] further improved on the model by taking into account the effects
of tool- work system deflection.
In addition to the time domain modeling in coupling with numerical solutions, Fourier
series expansion has been utilized to give an analytical and closed-form representation of
the cutting forces. Yeilowley [10] derived an explicit expression of the force and torque with
multiple straight cutting edges. In the case of a single edge cutter, the expression was also
extended to consider the effect of tooth helix angle. Wang et al. [11] presented a convolu-
tion model that led to an algebraic solution for the dynamic force components in multi-flute
cutting. The closed-form nature of the solution facilitated the prediction of forces and the
optimization of cutting processes without resort to numerical iteration. However, the model
was only two dimensional without the contribution of axial force. Consequently, its
engineering application to the analysis of comprehensive machine tool behavior, including
the effect of axial runout and axial dynamic stability, is limited.
The scope of this paper is to establish a three dimensional cutting force model for
multi-flute peripheral end milling. This involves the integration of the local cutting forces
along the cutter axis and across the flutes. In this analysis the radial engagement is specified
by a unit square window function and the axial engagement by the integration limits. This
integration lends itself to a periodic force function in the cutter angle domain. Fourier
transform of the periodic function further yields an algebraic expression of the cutting force
frequency components in terms of the magnitude and the phase angle. In addition to the
consideration of axial force component, the approach taken in this study differentiates itself
from the convolution model suggested by [11] in that it provides a clearer view of the effects
of material constants, cutting parameters, and tool-work geometry on the dynamic cutting
forces. The validity of the model was evaluated in both the angle and the frequency domains
by comparing model-predicted cutting forces to experimentally measured values in a series
of end milling tests.

AN~',I.YTI(A[+ SFUI)Y
in the analytical modeling of cutting forces in end milling, a coordinate system was
chosen to have an origin located where the cutter axis intersects with its face plane. The
x coordinate designates the cutter feed direction with respect to the workpiece, z direction
points up along the cutter axis, and y direction lies across the feed while following the
right-hand convection. The coordinate system on a cutter section engaging in a workpiece is
depicted in Fig. 1.
 l b r e c dimensional cutting force analysis in end milling 261

vrkpiece7 ~(8) .. ".'." "~" t(8)

-~"[.../ ~ ~milling
(a) :It: /Z- ' -,y
d, I Z(e)/f[ o~/',,:~-----z l
:/ ~# t) - ' f x ~ . r_eferel~ze
line
+-I

A
t

(b) d~
i
i
J
J

Fig. I. Three dimensional cutting forces in (al :op view and (b) side view.

q.,

Fig. 2 The cutting forces acting on one flute.

2.1. Local cuttin9 forces

For small feed per tooth comparing to the cutter diameter, the chip thickness at any
cutting point on the cutter can be approximated by [12]
t+ = txsin0. (1)
The work of Koenigsberger and Sabberwal [5] suggested that the local tangential cutting
force at any point on the cutting edge can be related to the local chip thickness in the
following form:
.~(0) = k,t<(dz) = k,tx sin 0(dz) (2)
where dz is an incremental chip width at the cutting point as shown in Fig. 1. The tangential
cutting pressure constant k, can be determined experimentally, and it is often a function of
tool edge geometry, feed per tooth, and workpiece material properties. Although k, is
understood to vary with the instantaneous chip thickness, its value corresponding to the
average chip thickness has been commonly referenced for simplicity [13].
262 Li Zheng et al,

The engagement of the tool and the workpiece is intermittent in nature because of the
entry at 01 and the exit at 02 as seen in Fig. 1. With the use of a unit square window function
w(OI.
w ( 0 ) = l f o r 0 ~ ~<0~<02;0 otherwise. (3)
The local tangential cutting force can be expressed as the following to take into account the
effect of intermittent engagement
.~(0) = k,tcldzt = kd~ sin Ow(O)(dz). (4)
In the study of l-lusty and MacNeil [2] and Fu et al. [8], the local tangential cutting force
was related to the local radial and axial force components by
t: = k,~, f,, koL = (s)
In the rectangular coordinate syslem the forces can be expressed as

~,io)/= ~n,, -coso /~(o)/. (6)

;(otj ,> o L;(o)j
Non-dimensionalizing the above with the maximum possible tangential force acting on
any cutting point results in a set of local cutting force functions ~,, #y,/~ as follows

[~""'!
L ~:(o~,
k, tx(dz )
r-Pq
Lt:~,o)j
(7)

It can be shown from Eqns i4 7) lhai

,,[~'"~11
/~,"~)/=
i ~,
- k, l
o /~t)/
(8)
LO:(o)j o o ko Lp~m)j
where
/~1(01 = sin 20wO), #2t0) = tl -- cos 20)w(O), #3(0) = sin Ow(O). (9)

2.2. Total euning.thrces in the angle domain

The forces given in Eqn (6) are those acting on any local cutting point with an
infinitesimal area of t, by dz. Figure 2 shows one such point relative to other similar cutting
points on the cutter. The total forces acting on any one of the cutter flutes is the integration
of local forces over that entire flute.

':"~'/--- j(/,:ii/. (lO)

t:-,,,tj L/:,~o>j
Combining Eqns !71. f8L and 110). along with the geometric relationship 0 = 4> - 17 as seen
in Fig. 2 gives

/';"(q~)/= k,t.~
/':j(q~)J
k,
0
1
0
0
k,
/52( - fl) dz.
P 3 ( - fl)
] (11)

The axial increment, dz, is related to the angular increment, dfl, by the geometry of the
cutter flute. Figure 3 shows an unfolded view of a helical cutter flute. For a flute with
constant radius and constant helix angle it is seen that
R d~ tan
dz=h(fljdfi where h(fl) -- for0~<fl~<--; 0otherwise. (12)
tan ~ R
 Three dimensionalcutting force analysisin end milling 263

' -' an unvound helical

Rd~
Fig. 3. Relationship between axial depth of cut and chip width density function.

Substituting the above into Eqn [11) gives

.~(~b)[/ = k,G - k, 1 0 |/~2(q~ fl) h(fl)dfl. (13)

L'(~)J o o go L,O3(4, ~)
The cutting forces on different flutes are essentially the same, except for the angle delay
2rc/N between them. Therefore, the total forces taken by the whole cutter consisting of
N independent flutes can be expressed as

1I = [f~,,i I
If,,iI l I = k,'x
I l
- k, l 0 1~(4, /~)
L(4,I] '=' EL'f4,)] o o ko Lff~(4' t~)

[i=~h(fl-(i-1)~)ld ft. (14)

2.3. Total cuttin9 forces in the frequency domain

The cutting force model in the angle domain given by Eqn (14) is periodic in 2n/N, which
corresponds to the time period of tooth passing. The frequency components of cutting force
are then equivalent to the coefficient of their Fourier transforms at discrete multiples of the
tooth passing frequency. Taking the Fourier transform of Eqn (14) leads to
- F~(oJ,k )
Fr,(~o,kt = ~k,t~ - k, 1 ~ U(k) P2(k)| (15)
0 0 k~ P3(k)J
k=O, +_1, +_2, +_3...
in which
Nkd a tan
sin
-
2R
- e
-iNk d.ta.~
2R
(16)
i= 1 tan ~ Nk

Pl(k) =
( 2(4
1 _ (Nk)2) )[e_SNkO,(jNksin2Ol + 2cos201,

- e JNk<(jNk sin 202 + 2 cos 202)] (17)

~e jNkO~ _ e - JNkO~ 1
P2(k)
\ 2jNk- 2(4 --(Nk)2)/
[e JN~'(jNkcos2Oa - 2 s i n 2 O t ) - e JNk2(jNkcos202 + 2sin202)] (18)
1
P3(k) = _ _
(Nk) z ) [e- JNk'tjNk sin 01 + cos 011 -- e-JNk~(jNk sin 02 + cos 02)]. (19)
264 L! Zheng et al.

The a b o v e analytical m o d e l expresses the d y n a m i c cutting force c o m p o n e n t s in p e r i p h e r a l

end milling as closed-form functions of the cutting pressure constants, t o o l - w o r k g e o m e t r y ,
cutting p a r a m e t e r s , a n d m a c h i n i n g configuration. It p r o v i d e s b o t h the m a g n i t u d e a n d the
phase angle i n f o r m a t i o n of cutting force in all three dimensions. R e c o n s t r u c t i o n of the angle
d o m a i n waveform of cutting forces can be achieved from (19) via inverse F o u r i e r trans-
formation. Therefore a n e a r explicit solution is offered for time d o m a i n force analysis by
virtue of the algebraic n a t u r e of the frequency model.

3. EXPERIMENTAL V E R I F I C A T I O N

In verifying the a n a l y t i c a l model, a series of end milling tests were p e r f o r m e d over a range
of cutting c o n d i t i o n s a n d configurations. Specific cutting pressure c o n s t a n t s were first
e x p e r i m e n t a l l y identified, then t h r e e - d i m e n s i o n a l cutting forces were m e a s u r e d a n d c o m -
p a r e d to the m o d e l p r e d i c t e d values in the context of angle d o m a i n waveform, frequency
m a g n i t u d e , a n d frequency phase angle.
H i g h speed steel four-flute end milling cutter with 30 helix angle was used to m a c h i n e the
7075-T6 a l u m i n u m w o r k p i e c e on a vertical milling machine. A t h r e e - c o m p o n e n t p l a t f o r m
d y n a m o m e t e r was m o u n t e d between the w o r k a n d the m a c h i n i n g table to m e a s u r e cutting
forces. C u t t i n g tests were p e r f o r m e d over the range of c o n d i t i o n s listed in T a b l e 1. The first
fifteen set of d a t a (No. 1-15) in the Table were used for the identification of cutting pressure
c o n s t a n t s kt, k,, a n d ka. Setting k = 0 in E q n (15) results in the s o l u t i o n of pressure
c o n s t a n t s as

0 1- i011
ktk, P2(O) (20)
ktk.J
-

daNtx 0
- p,(o)
0
o I LF..(O)J
P3(O)

Table 1. Cutting conditions and average chip thickness (Workpiece: aluminum; cutter: HSS; D = 7/16 in.;
N = 4; helix angle = 30)

Spindle speed Infeed Cutting

Data No. (rpm) (in./'min) do(in.) d,(in.) configuration f~(in.)

1 210 1.80 0.20 0.40 up 0.000652

2 210 1.80 0.20 0.40 down 0.000652
3 210 1.80 0.20 0.09 up 0.000957
4 210 1.80 0.20 0.09 down 0.000957
5 210 1.80 0.20 0.14 up 0.001165
6 210 1.80 0.20 0.14 down 0.001165
7 210 1.80 0.20 0.19 up 0.001322
210 1.80 0.20 0.19 down 0.001322
9 210 1.80 0.20 0.24 up 0.001440
10 210 1.80 0.20 0.24 down 0.001440
11 210 1.80 0.20 0.29 up 0.001526
12 210 1.80 0.20 0.29 down 0.001526
13 210 1.80 0.20 0.34 up 0.001577
14 210 1.80 0.20 0.34 down 0.001577
15 210 1.80 0.20 / slot 0.001394
16 210 1.80 0.10 0.04 up 0.000652
17 210 1.80 0.10 0.04 down 0.000652
18 210 1.80 0.30 0.04 up 0.000652
19 210 1.80 0.30 0.04 down 0.000652
20 210 2.70 0.20 0.04 up 0.000978
2t 210 2.70 0.20 0.04 down 0.000978
22 210 3.60 0.20 0.04 up 0.001303
23 210 3.60 0.20 0.04 down 0.001303
24 325 1.80 0.20 0.04 up 0.000431
25 325 1.80 0.20 0.04 down 0.000431
26 135 1.80 0.20 0.04 up 0.001042
27 135 1.80 0.20 0.04 down 0.001042
 lhree dimensional cutting force anal)sis in end milling 265

where Fx(0), Fy(0), and /"_40) are the static components of the measured cutting forces.
Theoretically any integer value of k can be used for the solution, however, static force
components usually have larger magnitudes thereby offering a better resolution in calcu-
lation. The k,,kr, and k, found from the 15 data sets based on Eqn (20) were regressionally
related to the average chip thickness fc as
kt = 8.99([c) '259 10 5 l b / i n 2, k r = 3.00 x l 0 - 5 ( t ) 1.433, ka = 4.10 10-3(tc) 0.635

(21)

Kt

1.00E*06

1.00E+05
a.j

1.00E+04
0.001 0,01
Tc (1/3"1n.)

Kr

10

0.1
0.001 0.01
Tc (1/3"1n.)

Ka

10

1 i Exp.
i--- Ana. :

0,1
0001 0.01
Tc (1/3"1n.)

Fig. 4. Cutting pressure constants against average thickness for all the tested conditions listed in
Table 1.
266 Li Zheng et al.

in which
d~t:~
t, (22)
RIO 2 -- 01 )"

Figure 4 shows the pressure constants plotted against the average chip thickness in all
conditions listed in Table t. The relationship (21) as a straight line is also shown for
comparison. Note that the relationship fits well with the 27 tested conditions although it
was derived from the first 15 data sets only.
Results of the predicted magnitude at dominant harmonic frequencies are given in Table 2
in comparison to the experimentally measured forces. Table 3 gives the phase angle of the
harmonic frequency components from the model and the experiments. Note that these force
predictions are performed for cutting conditions independent to which the coefficients ka, k,
and k, were evaluated. The experimental data are seen to agree well with the analytical
predictions both in the magnitude and the phase angle. The identification process for k,, kr
and k, involved curve fitting which presents a source of prediction error.
Examples of cutting force power spectrum are given in Figs 5 and 6. They correspond to
the test conditions Nos 21 and 27 respectively. In an ideal cut, the force spectra ought to be
zero except at the harmonics of tooth passing frequency. It can be observed that there are
minor peaks in the experimental data at the spindle frequency and at one spindle frequency
above and below the harmonic frequency. These peaks are believed to be caused by the
disturbance of cutter axis offset and tilt [14, 15].
Model-predicted cutting force waveforms are also compared to experimental results in
Figs 5 and 6. The predicted values were inversely transformed to cutter angle domain from
the frequency domain solutions. The magnitude and pulsation period of the predicted forces
follow closely those of the experimental values. The effect of cutter runout is also readily
observable from the repeated pattern in the experimental waveforms with a period four
times of the tooth passing.

Table 2 ~, mnparison of the magnitude (Ib) of dynamic cutting forces from experimental measurement (M) and
model prediction (P)

IF~ ,'~,k)l IFr,(eo,k)l IFz(co,k)l

Data No ~, =:0 k= 1 k= 2 k=0 k= 1 k=2 k=0 k= 1 k=2

16(M) 414 3.00 1.10 2,17 1.63 0.65 1.98 1.06 0.42
161P! 4.51 3.67 193 2,14 1.70 0.78 1.46 1.17 0.59
i71Mt 215 1.71 0,76 5.81 4.57 2.01 1.38 1.03 0.50
17lP) 154 1.23 0,57 4.75 3.85 2.00 1.46 1.17 0.59
18!M! 1048 5.37 053 4.20 1.81 0.42 3.64 2.28 0.22
18IPI 1354 728 006 6.43 3.38 0.02 4.38 2.33 0.02
19iMI 701 3.32 0.30 16.82 7.82 0.84 4.27 2.17 0.12
19tP) 462 2.44 0.02 14.26 7.64 0.06 4.38 2.33 0.02
20 (M ! 1009 6.38 I 52 3.09 2.24 0.53 3.67 2.76 0.70
20(P) 1286 8.97 2,62 2.08 1.52 0.51 3.76 2.61 0.75
21 (M~ 6 78 4.45 (I,97 14.89 9.67 2.07 2.91 2.80 0.77
21 tPI 731 5.02 1.33 10.78 759 2.31 3.76 2.61 0.75
221M3 1282 8.19 2.00 3.35 2.52 0.60 5.05 3.79 1.00
22(P1 1712 11.92 3.43 0.14 1.08 0.72 4.50 3.12 0.89
23(M~ 870 5.69 1.26 17.46 tl.39 2.51 4.47 3.58 0.97
23~P1 I 1.85 8.15 2.20 12.36 8.76 2.73 4.50 3.12 0.89
24 IM I 5.74 3,68 0.93 3.09 1.86 0.45 1.29 1.32 0.33
24(P~ 670 4.74 l 47 593 408 1.10 2.22 1.54 0.44
25(Mr 3.16 2.11 ()49 8.69 5.74 1.43 1.87 1.30 0.40
254P~ 0.28 0.55 035 8.95 6.23 1.80 2.22 1.54 0.44
26{M~ 10.95 7.06 1 78 3.03 2.23 0.51 3.37 3.11 0.84
261P~ 13.31 9.29 2,71 1.83 1.39 0.51 3.84 2.67 0.76
27~M! 6.911 4.57 1.00 14.66 9.69 2.20 2.88 2.96 0.86
27[P! 780 5.36 1.42 10.94 7.71 2.36 3.84 2.67 0.76
 Table 3. C o m p a r i s o n of the phase angle !radian) of d.vnam,c cutting Ibrces from experimental m e a s u r e m e n t
(M) and model prediction (P)

l/ d-J, kll IFdo., , k)) {Fd(o , k))

D a t a No k = 11 k :: I k = 2 k =0 /, 1 k= 2 k=0 k= 1 k =2

16(M) OIIX) 215 157 3.14 174 0.19 0.00 2.28 1.46
16 ( Pl t).(X) 1.69 3 14 i !5 092 0.00 - 2.17 1.77
17IMi 3.14 2 30 1.28 O.(X) ~1.94 2.10 0.00 0.94 2.11
17{P) 3.14 2.~( 1.41 0.110 1.15 2.46 0.00 1.12 2.40
18 (M) 0.011 1 14 3.14 11.7b 2.02 0.00 2.93 1.08
18 (P) 0.(X1 2.?2 3.14 0.10 0.11 0.00 3.05 2.80
19(M) 3 14 113 t/.59 0.00 0 09 2.99 q.o0 (1,12 1,66
191P) :{ 14 293 0,38 0,1X1 00q 2 80 !00 0.06 2.86
21)(MI (I(X) 278 0.21 3.14 1.41 0.89 ')00 2.85 0.15
201P) O.(X) 069 3.14 <1,97 122 ~,00 2.70 0.72
21(M~ 3 14 " ~,9 163 0.00 0.66 1.72 0.00 0.62 1.60
21(P) 314 2 68 2.10 0.00 116"; 1,46 0.00 0.59 1.34
22 ( M ) 0 O0 278 0.27 3.14 155 0.67 0.(~) 2.85 0.27
22 (P) 0.(X) '71 i172 0.00 22 ~ 0.36 0.00 - 2.70 0.72
231M) 3 14 201 171 O.(XI (1~5 1.73 0.00 0.61 1.56
231PI 3 14 2 64 2.0l O.(Xl (L68 !.50 0.00 0.59 134
24 {M ) 0.00 282 0.17 3 14 I(72 1.50 0.00 --2.88 0.17
24(P) O.(X) 278 () 57 3 14 0.54 - 2,20 0.00 - 2.70 072
25(M) 314 ~s 8
.... ] 67 0.01/ O68 t.71 0.00 0.68 167
251P) ~14 235 2.60 0.0t) 1)6(i 135 O.(X) 0.59 1.34
26 ~M) o00 281 t1.17 3.14 141 096 ().(~1 -- 2.89 0.10
26 (Pl it(X) 272 (7.69 3.i4 106 I1([) O.(X) - 2.70 0.72
271M1 3 14 281 1.68 O.(X! (I.63 1o9 O.(X) 0.63 I61
27(P) 3.14 2.67 2.09 O.(X) o (m 1.47 0.00 0.59 1.34

X X
10v . . . . . . 20, - . . . . . . . . . . . . . . .

-40:
0 10 20 30 40 0 100 200 300 400
"fO Hz degree
Y Y
50,

150f . . . . . . . ~"
k
k /
#',\. //~\\
,I/ ' \ //7"\,'\ #
/~'

o
.~----~ 5 !~ 0
, ] i, ,~

-50
0 10 20 30 40 0 100 200 300 400
"f0 Hz degree
Z Z
41~ . . . . . . . . . . . Ii 20:
{

_'d lo [ i, /,~ /, ~',},, t~

2 ,., ,. 'q
2 0 *b,.~
?/v-.A "~",_ a
-10
0 10 20 30 40 100 200 300 400
"fO Hz degree
Fig. 5. C o m p a r i s o n of the m e a s u r e d and model-predicted cutting forces for data No. 21. In the
power spectra on the left. the circles are predictions and the solid lines represent measurements. In
the angle d o m a i n w a v e f o r m s on the right, the dotted lines are predictions and the solid lines
represent m e a s u r e m e n t s
268 Li Zheng et al.

X
20
A

~5
u_ k !

,t ~ -20
_3
o -40
o lO 20 3O 40 100 200 300 400
"fo Hz degree
Y Y
15 50
A
..Q
i

~
o
0

0 -50
0 10 20 30 4O 100 200 300 400
"fo H z degree
Z Z
2O
!
i

i ~ lO
52
=___ .9 0

o 10 20 30 40 0 100 200 300 400

*f0 Hz degree

Fig. 6. Comparison of the measured and model-predictedcutting forces for data No. 27. In the
power spectra on the left, the circles are predictionsand the solid lines representmeasurements.In
the angle domain waveforms on the right, the dotted lines are predictions and the solid lines
represent measurements.

4. CONCLUSIONS
An analytical model of three dimensional cutting forces in peripheral end milling is
developed in this paper. This involved the convolution integration of local cutting forces
with respect to the cutter orientation to yield the solution of cutting forces in the feed,
cross-feed, and axial directions. The effect of the intermittent engagement is modeled with
a rectangular window function in the cutter angle domain, and the effect of axial depth of
cut is described by the upper limit of the convolution integration. The resulting force
solutions in the angle domain are trains of pulses as functions of material cutting pressure
constants, tool geometry, machining configuration, and cutting parameters.
Through Laplace transformation, the frequency components of cutting forces are estab-
lished in closed form expressions at the harmonics of the tooth passing frequency. The
frequency model provides both the magnitude and the phase angle information in all three
directions. It can he used to reconstruct the angle domain waveform of forces via inverse
transformation, thereby offering a near-explicit solution for time/angle domain analysis by
virt ue of the algebraic nature of the frequency model. The model of this nature will allow the
prediction, interpretation, optimization, and control of cutting force systems to be per-
formed without the need of numerical iterations.
Peripheral end milling experiments were performed to verify the analytical model. This
included the identification of cutting pressure constants of the workpiece material, and
subsequently the comparison of model-predicted forces to experimental data over a range
of cutting parameters and machining configurations. Close agreement between experi-
 Three dimensional cutting force analysis in end milling 269

m e n t a l a n d m o d e l p r e d i c t e d c u t t i n g f o r c e s in t h e c o n t e x t o f t h e f r e q u e n c y m a g n i t u d e ,
frequency phase angle, and time waveform, has been concluded.

REFERENCES

1. W.A. Kline, R. E. DeVor and J. R. Lindberg, The prediction of cutting forces in end milling with application to
cornering cut. Int. J. Mach. Tool Manu. 27, 7 (1982).
2. J. Tlusty and P. MacNeiL Dynamics of cutting forces in end milling. A . C I R P 24, 21 (1975).
3. Y. Altintas and A. Spence, End milling force algorithms for CAD systems. A. C I R P 40, (1991).
4. S. Cetinkunt and R.-L. Tsai. Position error compensation of robotic contour end-milling, Int. J. Math. Tool
Manu. 30. 613 (1990).
5. F. Koenigsberger and A. J. P. Sabberwal, An investigation into the cutting force pulsations during milling
operations. Int. J. Mach. Tool Design Res. l, 15 33 (1961).
6. G. Yucesan, Q. Xie and A. E. Bayoumi, Determination of process parameter through a mechanistic force
model of milling operations. Int. d. Mach. Tool Manu. 33, 627 (19931.
7. W. A. Kline, R. E DeVor and I. A. Shareef, The prediction of surface accuracy in end milling A S M E ,L Engng
Ind. 104, 272 11982).
8. H. J. Fu, R. E. DeVor and S. G. Kapoor, A mechanistic model for the prediction of the force system in face
milling operation. A S M E d. Engng Ind. 106, 81 (1984).
9. J. W. Sutherland and R E. DeVor, An improved method for cutting force and surface error prediction in
flexible end milling systems. A S M E J. Engng Ind. 108, 269 (1986).
10. 1. Yellowley, A note on the significance of the quasi-mean resultant force and the modelling of instantaneous
torque and forces in peripheral milling operations. A S M E d. Engng Ind. ll0, 300 (1988).
11. J. J. Wang, S. Y. Liang and W. J. Book, Convolution analysis of milling force pulsation. A S M E d. Engng Ind.
116, 17 (1994j.
12. M. E. Martellotti, An analysis of the milling process, Trans. A S M E 63. 677 (1941).
13. I. Yellowley, Observations on the mean values of forces, torque and specific power in the peripheral milling
process. Int. d. Mach. Tool Manu. 25, 337 (1985).
14. W. A. Kline and R. E. DeVor, The effect of runout of cutting geometry and forces in end milling. Int. Y. Maeh.
Tool Manu. 23, 123 (1983).
15. S. Y. Liang and J. J. Wang, Milling force convolution modeling for identification of cutter axis offset. Int. J.
Math. Tool Manu. 34, 1177 (19941.