Professional Documents
Culture Documents
a r t i c l e in fo
abstract
Article history:
Received 30 June 2009
Received in revised form
6 October 2009
Accepted 6 October 2009
Available online 14 October 2009
Angular contact ball bearings have been widely used in machine tool spindles, and the bearing preload
plays an important role on the performance of the spindle. With the development of high speed
machining, especially for high speed milling, heavy cutting at a low speed and light cutting at a high
speed are often performed on a single machine tool spindle, thus, high stiffness at low speed and low
temperature rise at high speed are required. The traditional constant pressure preload method cannot
meet the technical requirement of this kind of spindle any more. The variable preload technology is
systematically investigated in this paper. At high speed range, FEM method is used to analyze the
temperature distribution of the spindle, and the variable spindle preload is determined according to the
constraint of temperature rise of bearings. At low speed range, the spindle preload is resolved by the
fatigue life of bearings. The dynamic stiffness of the variable preload spindle is analyzed utilizing the
Transfer Matrix Method (TMM) and a nonlinear bearing model that includes the centrifugal force and
gyroscopic effects. An experimental set-up for the variable preload spindle is developed using hydraulic
pressure that can automatically adjust the bearing preload. The proposed method to determine variable
preload is veried experimentally by measuring the dynamic stiffness of the spindle and the
temperature rise of the test bearing. The results show that the variable preload spindle gives
outstanding behavior that the temperature rise at high speed is lower than that of the constant pressure
preload spindle, and the dynamic stiffness at low speed range is signicantly increased.
& 2009 Elsevier Ltd. All rights reserved.
Keywords:
Machine tool spindle
Variable bearing preload
Temperature rise
Fatigue life
Dynamic stiffness
1. Introduction
Initial preload of rolling bearings is widely applied to obtain
high stiffness of machine tool spindles, restrain vibration and
enhance rotational accuracy [13]. Proper preload of the angular
ball bearing is important to the speed, rigidity, accuracy and life of
the spindle. The proposed bearing load mechanisms mainly
include hydraulic instruments [3,4,7] and piezoelectric actuators
[8]. The controller using hydraulic pressure is a most popular
method for applying an automatic variable preload due to its
simple structure and maintenance.
Currently, with the development of high speed machining,
especially for high speed milling, heavy cutting at low speed and
light cutting at high speed are often performed on a single
machine tool spindle. For conventional spindle, the constant
pressure preload method is preferred because it is simple and easy
to maintain. However, the constant pressure preload is not
favored for a high speed machining spindle.
Literature review shows that some researchers have investigated the effect of the preload on the dynamic characteristics of
the spindle system [27], and found that high preload of the
bearing can enhance the stiffness and the natural frequency of the
ARTICLE IN PRESS
20
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
Nomenclature
A
a
a1
a2
a3
b
Cr
C0
cair
d
dm
Fa
Fr
Hf
hr
h0
kair
L10
L10a
M
M1
M2
N
n
Pr
P0
Q
R
rh
Th
Tr
T0
X
Y
Xs
Ys
p
ah
ar
Z
l
la
lr
mair
n
nair
n0
n1
x
Subscripts
air
h
r
refers to air
refers to housing
refers to bearing outer ring
ARTICLE IN PRESS
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
signal
acquisition unit
driving motor
(motorized spindle)
power
amplifier
D/A
conversion
21
proportional
electromagnetic
hydraulic
station
hydraulic
chamber
revolution
transmitter
disk
unbalanced
mass
coupling
eddy current
sensor
oil-air lubrication nozzle
rearing bearings
front bearings
signal
acquisition unit
computer
Fig. 1. Schematic of the variable preload spindle system.
p1 Fa 0:1Fr
Table 1
Values of Xs and Ys for angular contact ball bearings.
Contact angle
151
201
251
301
351
401
Xs
Ys
0.5
0.47
0.5
0.42
0.5
0.38
0.5
0.33
0.5
0.29
0.5
0.26
n0 n 4 2000
n0 n r 2000
6
7
3mQax
8
ARTICLE IN PRESS
22
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
Table 2
Parameters of the test spindle.
Parameter
Value
07000
256
85
435
Table 3
Parameters of ball bearings.
Bearing
Front bearing
Rear bearing
Type
Material
Inner diameter (mm)
Outer diameter (mm)
Width (mm)
Ball diameter (mm)
Number of balls
Contact angle (deg.)
7015
Steel
75
115
20
12.3
20
25
7012
Steel
60
95
18
11.0
18
15
Table 4
The base oil viscositytemperature relationship of IsotexTopasL30.
Temperature (1C)
25
40
60
35
18
10
0.020
3.1.2.1. Convection of the air between shaft and housing. For this
experiment condition, the air owing through the spindle system
is assumed to be forced turbulent air, and the convection coefcient can be calculated by [20]
0.016
0.012
Ms / M
0.024
0.008
0.004
0.000
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07
Fig. 2. The ratio of Ms to M (rotational speed, 7000 rpm; axial preload, 1000 N).
Nu kair
d
10
11
12
where uair is the mean velocity of the uid owing through the
shaft surface superimposed by axial and tangent speeds.
Fig. 3 shows the heat convection coefcients of different
diameters of the multi-diameter shaft with respect to spindle
speed, and those of the housing inner surface are listed in Table 5.
ARTICLE IN PRESS
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
d=27mm
d=34mm
d=50mm
d=65mm
d=85mm
d=100mm
100
Rb
50
4000
5000
Spindle speed (rpm)
6000
7000
13
c
4l1 a
17
4l2 a
Rb Rb =N
18
14
16
Table 5
Convection of the housing inner surface.
4la
dy
p
1 k2 sin2 y
Zp=2
150
0
3000
19
1.2
1.0
Thermal Permissivity (W/K)
250
200
23
0.8
0.6
Inner Ring/Ball Contact, Rear
Outer Ring/Ball Contact, Rear
Inner Ring/Ball Contact, Front
Outer Ring/Ball Contact, Front
0.4
0.2
0.0
0
1000
2000
6000
7000 8000
ARTICLE IN PRESS
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
L10
106 Cr e
60n Pr
20
v
v1
21
151
Fa
C0
0.015
0.029
0.058
0.087
0.12
0.17
0.29
0.44
0.58
Fa
iZD2
0.172
0.345
0.689
1.03
1.38
2.07
3.45
5.17
6.89
Fa
Fr
4 e FFar r e
0.44
1.47
1.40
1.30
1.23
1.19
1.12
1.02
1.00
1.00
100
0.38
0.40
0.43
0.46
0.47
0.50
0.55
0.56
0.56
Table 7
Thermal parameters of materials.
Material
45#Steel
40Cr
GCr15
Density (kg/m3)
Modulus of elasticity (GPa)
Poisson ratio
Specic heat capacity (J/kg K)
Conduction coefcient (W/m K)
7.85
205
0.28
480
50.2
7.85
205
0.29
460
60.5
7.85
208
0.29
460
50
90
Temperatures (C)
24
80
70
1000N
2000N
3000N
4000N
5000N
6000N
7000N
8000N
60
50
40
30
3000
4000
5000
6000
Spindle speed (rpm)
7000
Fig. 6. Steady state temperatures for different preloads and spindle speeds.
Fig. 5. Temperature distribution of the spindle (rotational speed, 7000 rpm; preload, 1000 N).
ARTICLE IN PRESS
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
Speed (rpm)
r 4000
r 5000
r 6000
r 7000
2000
1000
1000
1000
Table 9
Lives of the front bearing (h).
1000
2000
3000
4000
5000
6000
7000
8000
5000
6000
7000
579,137
72,392
21,449
9049
4633
2681
1688
1131
463,309
57913
17159
7239
3706
2144
1350
904
386,091
48,261
14,299
6032
3088
1787
1125
754
330935
41366
12256
5170
2647
1532
964
646
Table 10
Lives of the front bearing (h).
Preload (N)
1000
2000
3000
4000
5000
6000
7000
8000
Speed (rpm)
r 1000
r 2000
r 3000
8000
7000
6000
40
30
20
10
0
1000
2000
3000
4000
5000
Spindle speed (rpm)
6000
7000
Table 8
Variable preloads for high speed range.
Preload (N)
Table 11
Variable preloads at low speed range.
25
2000
3000
5,575,215
808,860
273,130
128,980
72,656
45,535
30,627
21,663
2,555,320
370,730
125,190
59,115
33,301
20,870
14,037
9929
1,935,833
280,850
94,838
44,784
25,228
15,811
10,634
7522
3,949,136
572,940
193,470
91,359
51,465
32,254
21,694
15,345
ARTICLE IN PRESS
26
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
200
6. Experimental verication
6.1. Temperature rise
To verify the proposed method for variable preload in this
paper, an experimental set-up for the spindle was developed, as
shown in Fig. 9. In this section, the temperature rises were
detected by two thermocouples.
The temperature rises of both the variable preload and the
constant preload spindle were measured during high speed range,
as shown in Fig. 10. It can be seen that the simulation results agree
with the experimental measurement, and the bearing
temperature of the variable spindle is decreased in the high
speed range compared with that of the constant pressure preload
spindle. Furthermore, the experimental temperatures were all
lower than the theoretical under the constant pressure preload
and the variable preload due to some assumptions introduced into
the thermal model. The difference of temperatures between
theory and experiment is not signicant for constant pressure
preload spindle. As to the variable preload spindle, the
experimental temperatures are slightly lower than the
theoretical from 3000 to 5000 rpm, but agree well from 6000 to
7000 rpm.
6.2. Dynamic stiffness
190
Experiments were carried out to verify the predicting dynamic
stiffness of the spindle according to the rotation speed. As shown
in Fig. 1, an unbalanced mass is mounted on the disk located at
spindle nose to generate an exiting force during the rotation of the
spindle. The displacement of the spindle nose is detected by an
eddy current sensor and the displacement signal is sampled by a
signal acquisition unit (AZ308R), then the sampled signal is
analyzed by the CRAS V7.0 software and the out-of-balance
response at the spindle nose is obtained. The dynamic stiffness at
the spindle nose is as follows:
180
170
160
150
140
130
constant preload (simulated)
variable preload (simulated)
120
110
100
0
1000
6000 7000
me o2
9x9
ARTICLE IN PRESS
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
27
40
35
Temperature rise (C)
30
25
20
15
Constant preload (theoretical)
Constant preload (experimental)
Variable preload (theoretical)
Variable preload (experimental)
10
5
0
3000
200
190
180
170
160
150
140
130
120
110
100
4000
5000
Spindle speed (rpm)
6000
7000
1000
6000
7000
ARTICLE IN PRESS
28
S. Jiang, H. Mao / International Journal of Machine Tools & Manufacture 50 (2010) 1928
7. Conclusions
This paper has developed a method to analyze the variable
preload with rotation speed for a high speed machining spindle. The
entire range of speed is divided into low and high speed sections; a
lower preload is applied for the high speed section within the limit
of the temperature rise and a higher preload is applied for the low
speed section within the limit of bearing fatigue life.
In the high speed section, the temperature distribution of the
spindle has been analyzed by aid of FEM, and the spindle preload
was determined according to the constraint of temperature rise of
bearings. In the low speed section, the spindle preload has been
resolved by the fatigue life of bearings. The dynamic stiffness of
the variable preload spindle was studied by utilizing TMM and a
nonlinear dynamic model of bearing. The proposed method for the
variable bearing preload has been veried by experiment study.
Compared with the traditional constant pressure preload
technology, the temperature rise of the variable preload spindle
bearing at high speed range is reduced due to a lower preload, and
the lower preload can be greatly helpful in lengthening the service
life of the bearing. Meanwhile, the dynamic stiffness of the
variable preload spindle is signicantly enhanced at low speed
range due to a larger preload. Furthermore, the dynamic stiffness
of the variable preload spindle is 86.0% of the constant pressure
preload spindle at a maximum speed of 7000 rpm, though it
decreases slightly at high speed. The above results are promising
considering the fact that heavy cutting at low speed and light
cutting at high speed are demanded on a single spindle of high
speed machining.
The hydraulic actuator chosen for the preload applied is based
on the fact that the hydraulic system is an essential subsystem of
a machining center, for example, the automatic tool changer is
usually actuated by a hydraulic cylinder installed at the rear end
of the spindle. So we think it is simple and cheap to apply the axial
preload by using the hydraulic system employed in the machine
tool.
Acknowledgements
The authors gratefully wish to acknowledge the supports of
National Science Foundation through Grant nos. 50475073,
50775036 and Jiangsu High Technology Research and Development Project through Grant nos. BK2002059, BG2006035,
BK2009612.
References
[1] T.A. Harris, in: Rolling Bearing Analysis, third ed., John Wiley and Sons, New
York, 1991.