5 DESIGN OF SPINDLES AND SPINDLE

SUPPORTS

5.1 FUNCTIONS OF SPINDLE UNIT AND REQUIREMENTS

Functions The spindle unit of a machine tool performs the following important functions:
1. Centreing the workpiece, e.g., in lathes, turrets, boring machines, etc., or the cutting tool, as in drilling
and milling machines.
2. Clamping the workpiece or cutting tool, as the case may be, such that the workpiece or cutting tool is
reliably held in position during the machining operation.
3. Imparting rotary motion (e.g., in lathes) or rotary cum translatory motion (e.g., in drilling machines)
to the cutting tool or workpiece.

determine the important design requirements to spindle units which are listed below.
Requirements
1. The spindle should rotate with a high degree of accuracy. The accuracy of rotation is determined
by the radial and axial run out of the spindle nose, and these must not exceed certain permissible

the front end.

2. The spindle unit must have high static stiffness. The stiffness of the unit is made up of the stiffness of

as well as torsional stiffness.

3. The spindle unit must have high dynamic stiffness and damping. Poor dynamic stability of the spindle
unit adversely affects the dynamic behaviour of the machine tool as a whole, resulting in poor surface

4. The mating surfaces that are liable to wear restrict the life of the spindle unit. These surfaces, such as
journals, quills (in drilling machines), etc., must be hardened to improve their wear resistance. The
spindle bearings must also be selected or designed to retain the initial accuracy during the service life
of the machine tool.
5. The deformation of the spindle due to heat transmitted to it by the bearings, cutting tool, work piece,
etc., should not be large, as this has an adverse effect on the machining accuracy. In case of spindles
running at high rotational speeds, particular care should be taken in selecting or designing the front
bearing as it is the major source of heat transmitted to the spindle.

cutting tool or workpiece. The centreing is achieved by means of an external or internal taper at the
 Design of Spindles and Spindle Supports 289

front end of the spindle. The spindle ends, including the taper have been standardised for the common
groups of machine tools and are shown in Table 5.1.

Table 5.1 Spindle ends

S. No. Figure Application Remark

7°7¢ 30¢¢
External taper 7°7¢30¢¢ is used
only in heavy-duty machine tools
Lathes, turrets, single spindle for centreing chuck or face plate.
1. automatic and semi-automatic Generally, chucks or face plates
lathes, etc. are centred by a cylindrical
surface.

Steep taper 7 : 24 is necessary so

that the milling arbor and cutter
2. Milling machines once located and clamped should
be able to hold their position
under the pulsating milling force.

Taper 7:24

Morse tapers come in 8 sizes

angle varies between 1°25¢27≤ and

3. Drilling and boring machines 1°30¢26≤. Small taper is essential
as rotation is transmitted from
Morse taper the spindle to the drill shank by
friction.

In grinding machines, the con-

centricity of location is extremely
important. Therefore, the grinding
4. Grinding machines
wheel is mounted on an adaptor

Taper 1:3 taper.

5.2 MATERIALS OF SPINDLES

The blank for a machine tool spindle may be
1. rolled stock in the case of spindles having diameter < 150 mm, and
290 Machine Tool Design and Numerical Control

2. casting (preferably obtained by centrifugal casting method) in the case of spindles having diameter >
150 mm.
It should be borne in mind that if the spindle blank is cut from rolled stock, the cutting must be done by a
cutting tool (shears, rotary saw, parting tool, etc.) to avoid additional distortions of the material microstruc-
ture.
In machine tool spindle design, the critical design parameter is not strength but stiffness. If we compare

although the strength of alloyed steels can be considerably greater than that of mild steel. Since stiffness (the
main design parameter) is primarily determined by the modulus of elasticity of the material, it may be con-

In the light of the preceding discussion and the requirements laid down in Sec. 5.1, the following recom-
mendations for selecting the spindle material may be formulated
1. for normal accuracy spindles, plain carbon steels C45 and C59 (AiSi C1045 and C1050) hardened and
tempered to RC = 30.
2. for above normal accuracy spindles—low alloy steel 40 Cr 1 Mn 60 Si 27 Ni25 (AiSi 5140) induction
hardened to RC =
Cr 1 Mn 60 Si 27 Ni 25 (AiSi 5147) is used with hardening to
RC = 55–60.
3. for spindles of precision machine tools, particularly those with sliding bearings—low alloyed steel
20 Cr 1 Mn 60 Si 27 Ni 25 (AiSi 5120) case hardened to RC = 56–60 or 38 Cr 1 A1 90 Mn 45 Si 27
Ni 25 Mo 20 (EN 41) nitrided to RC = 63–68, and
4. for hollow, heavy-duty spindles—grey cast iron or, spheroidal graphite iron.

EFFECT OF MACHINE TOOL COMPLIANCE ON

5.3 MACHINING ACCURACY
Consider a uniform shaft being machined between centres on a B≤
lathe (Fig. 5.1). Let KA be the stiffness of centre A and KB that of C≤
centre B. Due to radial component Py of the cutting force, centre A¢
A will be displaced by a distance C¢ B¢
A C B
PA
yA = (5.1)
KA
and centre B by
PB Py
yB = (5.2) x
KB
l
Here PA and PB are the forces of reaction at ends A and B,
Fig. 5.1 Schematic diagram of a simple
respectively. They can be determined from the following turning operation
equations of static equilibrium:
1. Moment of Forces about Point B = 0, i.e.,
PA l = Py (l – x)
 Design of Spindles and Spindle Supports 291

wherefrom,
Ê xˆ
PA = Py Á1- ˜ (5.3)
Ë l¯
2. Moment of Forces about Point A = 0, i.e.,
PB ◊ l = Py ◊ x
wherefrom,
x
PB = Py (5.4)
l
Substituting the values of PA and PB in Eqs (5.1) and (5.2), respectively, we obtain

Ê xˆ 1
yA = Py Á1 - ˜ (5.5)
Ë l ¯ KA
x 1
yB = Py ◊ (5.6)
l KB
Owing to the compliance of centres A and B, the workpiece occupies position A¢ B¢¢ (assuming KA > KB)
and its displacement at the cutting point can be found as
yx = yA + C¢C¢¢
from similar triangles A¢C ¢C ¢¢ and A¢B¢B¢¢, we have
C ¢C ¢¢ x
=
B ¢B ¢¢ l
since
B¢B¢¢ = yB – yA
x
C ¢C¢¢ = (yB – yA)
l
Therefore,
x
yx = yA + (yB – yA) (5.7)
l
Substituting the values of yA and yB from Eqs (5.5) and (5.6), Eq. (5.7) yields

ÈÊ xˆ 1 x Ïx 1 Ê x ˆ 1 ¸˘
yx = Py ÍÁ1 - ˜¯ ◊ + Ì ◊ - Á1 - ˜ ˝˙
ÍÎË l K A l Ó l KB Ë l ¯ K A ˛˙˚

Py È Ê xˆ
2
Ê xˆ ˘
2
or yx = Í K B Á1 - ˜ + K A ÁË ˜¯ ˙ (5.8)
K A ◊ KB Î Ë l¯ l ˚

If it is assumed that KA /KB = a,

P ÈÊ xˆ
2
Ê xˆ ˘
2
yx = y ÍÁ1 - ˜ + a ÁË ˜¯ ˙ (5.9)
K A ÎË l¯ l ˚
292 Machine Tool Design and Numerical Control

wherefrom, keeping in mind that Py/KA = yA max, we get

2 2
yx Ê xˆ Ê xˆ
= Á1 - ˜ + a Á ˜ (5.10)
y A max Ë l¯ Ël¯

The variation of yx/yAmax plotted as a function of x/l for different values of a is depicted in Fig. 5.2. From
the curves it may be concluded that
1. when a < 1, i.e., stiffness of the headstock centre is less than the stiffness of the tailstock centre,
maximum displacement of the workpiece occurs at the headstock; and
2. when a > 1, i.e., the stiffness of the tailstock centre is less than that of the headstock, maximum
displacement of the workpiece occurs at the tailstock.
yx /yAmax

a=4

a=2
1

a=1

a=0
0
x
0 0.2 0.4 0.6 0.8 1.0 I

Fig. 5.2 Variation of yx /yAmax as a function of x/l

and minimum displacements of the workpiece axis. As already stated, maximum displacement occurs at the
headstock or tailstock depending upon the value of
yx/Py is minimum. Therefore, the location of the point of

d Ê yx ˆ
=0
dx ÁË Py ˜¯
From Eq. (5.9),
d Ê yx ˆ 1 Ê 2x 2 ˆ 1 2x
= Á - ˜+
Á ˜
dx Ë Py ¯ K A Ë l 2 l ¯ KB l2
 Design of Spindles and Spindle Supports 293

which, upon equating to zero yields

KB ◊ l
x0 = (5.11)
K A + KB
x0

d 2 Ê yx ˆ 2Ê 1 1 ˆ
= Á + ˜ >0
dx 2 ÁË Py ˜¯ l 2 Ë K A K B ¯

Substituting the value of x0 in Eq. (5.10), we obtain

ÈÊ KB ˆ
2
Ê KB ˆ ˘
2
ymin = ÍÁ1 - +aÁ ˙ y A max
ÍÎË K A + K B ˜¯ Ë K A + K B ˜¯ ˙˚
a
or ymin = y A max (5.12)
1+a
For a < 1: The maximum displacement occurs at the headstock centre, i.e., ymax = yAmax. Therefore,
a
ymax – ymin = yA max – ◊ y A max
1+a
wherefrom,
ymax - ymin 1
= (5.13)
y A max 1+a

For a > 1: The maximum displacement occurs at the tailstock centre, i.e., ymax = yBmax. Therefore,
a
ymax – ymin = yB max – ◊ yA max = yA max È yB max - a ˘
1+a Íy ˙
Î A max 1 + a ˚
as yB max /yAmax = KA/KB = a, we get
ymax - ymin a2
= (5.14)
y A max 1+a

(5.13) for a < 1 and by Eq. (5.14) for a > 1. The relationships rep-
resented by Eqs (5.13) and (5.14) have been plotted as functions of 3 ymax = yB(a >1)
ymax – ymin /yAmax

a and shown in Fig. 5.3. The dotted portions of the curve represent
the range in which the curves are not valid.
It is evident from Fig. 5.3 that the difference ymax – ymin is 2
minimum at a = 1 and is equal to half the headstock centre
displacement, i.e., 1
y ymax = yA(a < 1)
(ymax – ymin)min = A max (5.15)
2 0 a
0 1.0 2.0 3.0 4.0
achieved when the stiffness of headstock and tailstock centre are y max - y min
Fig. 5.3 Variation of as a
equal. y A max
function of a
294 Machine Tool Design and Numerical Control

If the stiffness of the saddle is given by Ks, the displacement of the cutting edge due to saddle compliance
can be found as
Py
ys =
Ks

The total compliance of the machine tool can be found as the sum of compliances of the workpiece and
saddle, i.e.,
y y y
Cmt = mt = s + x
Py Py Py
2 2
1 1 Ê xˆ 1 Ê xˆ
or Cmt = + Á1 - ˜¯ + Á ˜ (5.16)
Ks K A Ë l KB Ë l ¯

the machine tool compliance as expressed through Eq. (5.16). The representative total compliance values for
some machine tools are given in Table 5.2.

Table 5.2 Values of compliance Cmt for different machine tools

S. No. Machine tool Description Compliance

(micron/kgf)
1. Lathe, workpiece clamped between centres Height of centres 200 mm 0.75
Height of centres 300 mm 0.52
Height of centres 400 mm 0.38

2. Lathe, workpiece clamped in chuck which is Height of centres 200 mm 1.5

screwed on spindle nose Height of centres 300 mm 1.2

3. Multiple-spindle vertical, semiautomatic lathes Height of centres 400 mm 0.5–0.7

4. Automatic bar lathes Screw type 0.2–0.3

Multiple-spindle type 0.3–0.4

5. Vertical milling machine Drive motor rating = 7 kW 0.4

Table size 320 ¥ 1250 mm

6. Plano-milling machine Table size 4.25 ¥ 1.5 m 0.4

7. Vertical turning and boring mill Table diameter 3 m 0.36

8. Centreless grinding machine 1.00

5.4 DESIGN CALCULATIONS OF SPINDLES

Figure 5.4 shows the schematic diagram of a spindle. As is evident from the diagram, the spindle represents
a shaft with
 Design of Spindles and Spindle Supports 295

1. supported length l acted upon by driving force P2, and P2 P1

2. cantilever of length c acted upon by external force P1.
The spindle is basically designed for bending stiffness which

l c

ymax £ [y] (5.17) Fig. 5.4 Schematic diagram of a

spindle

y1 of the spindle axis due to bending forces P1 and P2, and

y2 of the spindle axis due to compliance of the spindle supports.

y3 of the centre or cutting tool due to compliance of the tapered joint.

5.4.1 Deflection of Spindle Axis due to Bending

diagram. In this context, the following guidelines may be employed with success:
1. If the spindle is supported on a single anti-friction bearing at each end, it may be represented as a
simply supported beam.
2. If the spindle is supported in a sleeve bearing, the supported journal is analysed as a beam on an elastic
foundation; for purposes of the design diagram the sleeve bearing is replaced by a simple hinged
support and a reactive moment Mr acting at the middle of the sleeve bearing.
The reactive moment is given as
P2 P1
Mr = k ◊ M
where M = bending moment at the support,
k k = 0 at small
a b c
loads to k = 0.3–0.35
(a)
Consider, for example, the spindle shown schematically in P2 P1
Fig. 5.5a. By replacing the rear ball bearing by a hinge, and Mr
the front sleeve bearing by a hinge and reactive moment Mr,
the spindle can be reduced to the design diagram of Fig. 5.5b. (b)

determined by Macaulay’s method and is found to be

y1
1 È 2 Ê aˆ ˘
y1 = Í p1c (l + c) - 0.5 P2 abc ÁË1 - ˜¯ + M r lc ˙ (5.18) (c)
3EI Î l ˚
Fig. 5.5 (a) Schematic diagram of the
where E = modulus of elasticity of the spindle material
spindle (b) Design diagram of
I = average moment of inertia of the spindle section the spindle (c) Deflected axis of
the spindle
296 Machine Tool Design and Numerical Control

5.4.2 Deflection of Spindle Axis due to Compliance of Spindle Supports

Let dA and dB represent the displacement of the rear and
front supports, respectively. Owing to the compliance of
A¢
- dA O B
sign purposes we are considering the most unfavourable dB C
B¢ y2
case when bearing displacements are oppositely directed). C¢
From similar triangles OCC ¢ and OBB¢,
y2 d
= B Fig. 5.6 Deflection of spindle due to compliance
c+x x
wherefrom, of the supports

y2 = ÊÁ1 +
cˆ
˜ dB (5.19)
Ë x¯

Again, from similar triangles OAA¢ and OBB¢,

dB d
= A
x l-x
wherefrom,
l ◊ dB
x = (5.20)
d A + dB
Substituting this value of x in Eq. (5.19), we get
Ê Cˆ c
y2 = dB Á1 + ˜ + d A (5.21)
Ë l¯ l
It is evident from Eq. (5.21) that displacement dB
y2 of spindle nose than displacement dA of the rear bearing.
Displacements dA and dB can be determined from the following expressions:
RA
dA = (5.22)
KA
R
dB = B (5.23)
KB
where RA and RB are reactions at supports A and B, respectively, while KA and KB is their respective stiffness.
Reactions RA and RB can be found from the following equilibrium conditions applied to the design dia-
gram of Fig. 5.5b:
1. Moment of All Forces about Point A = 0, i.e.,
SMA = 0 or RB ◊ l – P2 ◊ a + Mr – P1(c + l) = 0
wherefrom,
P2 a - M r + P1 (l + c)
RB = (5.24)
l
 Design of Spindles and Spindle Supports 297

2. Moment of All Forces about Point B = 0, i.e.,

SMB = 0 or RA ◊ l – P2 ◊ b – Mr + P1c = 0
wherefrom,
P2 b + M r - P1c
RA = (5.25)
l
y2 may be written as follows:

P2 a - M r + P1 (l + c) Ê c ˆ P2 b + M r - P1c c
y2 = ÁË1 + ˜¯ + x (5.26)
l ◊ KB l l ◊ KA l

The total deflection of the spindle nose can be

determined as the sum of y1 (Eq. (5.18)) and y2 (Eq.
(5.26)), i.e.,
y2
y = y1 + y2 (5.26a) y
y1

Fig. 5.7.
Fig. 5.7 Total deflection of the spindle axis

5.4.3 Optimum Spacing between Spindle Supports

An important parameter in spindle design is the ratio l = l/c. The y1 y2,y
optimum value of this ratio is the one that ensures minimum total
y; it can be determined from the condition dy/dl = 0.
The qualitative variations of y1, y2 and y1 + y2, for constant
values of forces P1 and P2 are depicted in Fig. 5.8 as functions of y=y1 ,+y2
ratio l/c.
The point of minimum of the y1 + y2 curve yields the optimum
value of ratio l/c which generally lies between 3 and 5. The value y1
of lopt depends upon
1. ratio
KB
a= y2
KA
of the stiffness of the front and rear bearings, and
l
2. factor
lopt
K I
F= B c
Kc It Fig. 5.8 Effect of l/c ratio on y1, y2 and
3EI y1 + y2
where Kc = 3 c = bending stiffness of the cantilever,
c
Ic = average moment of inertia of the spindle over the cantilever length, and
Il = average moment of inertia of the spindle over the supported length.
298 Machine Tool Design and Numerical Control

The variation of l opt as a function of F is depicted in Fig. 5.9 l

for a = 1 and a = 10.
When the spindle is mounted on anti-friction bearings, an ad- 3
ditional check is necessary to ensure that the constraint lopt ≥ l min
= 2.5 is not violated because for l < 2.5 the bearing play leads to a =10
2
considerable radial run out of the spindle nose.
a =1
An opposite constraint on maximum span stems from the re-
1
quirement that for normal functioning of the spindle driving gear,
the stiffness of the span should not be less than 25–50 kgf/micron.
This constraint is expressed through the following relationship: 0 F
0 1 2 3
Dl4/3 Fig. 5.9 Effect of spindle and supports
l£ 1/3
(5.27)
k stiffness on lopt
where Dl = average diameter of the supported length of the spindle
k = 0.05 in the case of normal accuracy machine tools and 0.1 for precision machine tools

the span should satisfy the condition,

y l max £ 10–4l (5.28)
and the maximum span length lmax should be limited by the above constraint. This constraint is based upon

of the bearing gap.

5.4.4 Deflection due to Compliance of the Tapered Joint

The spindle ends of most machine tools have a tapered hole for accommodating a centre (as in lathes) or
d
1
from the spindle axis (Fig. 5.10), where a force P is acting, may be represented as
y3 = d + qd (5.29)

q
P0

t d

Fig. 5.10 Schematic diagram of a tapered joint

 Design of Spindles and Spindle Supports 299

where d = displacement of the shank or centre at the edge of taper due to contact compliance and
q = angle of slope of the shank or centre at the edge of the taper
If the manufacturing errors of the taper are ignored, d and q can be determined from the following
expressions:
4 Db C1
d = (bdC2 + C3) micron (5.30)
pD
4 Pb 2 C1
q= (2bdC4 + C2) (5.31)
pD
In these expressions,
C1 = C1 = 0.03–0.06, while for the 7 : 24 taper
C1 = 0.02 micron cm2/kgf
C2, C3, C4
tapers C2 = C3 = C4 = 1, while for the 7 : 24 taper C2 = C3 = 1.35 and C4 = 1
1/4
Ê 1 ˆ
b= cm–1
ÁË 2.3C D ˜¯
4
1
D and d are expressed in cm
Generally, displacement d due to contact compliance can be ignored in comparison with the displacement
due to bending of the shank or centre. The expression for y3 can then be written as

4 Pb 2 C1
y3 = (2bdC4 + C2)d micron (5.32)
pD
If it is assumed that 2bdC4 is much greater than C2 (which is true), the
for y3 is obtained:
Pd 2 3
y3 = 2.55 b C1C4 micron (5.33)
D

Keeping in mind the expression for b and substituting C4 = 1 and C1 = 0.02 micron◊cm2/kgf, the expres-
sion for displacement of a shank or centre mounted in a 7 : 24 taper can be written as follows:
Pd 2
y3 = micron (5.34)
2D4
Consequently, the stiffness of the 7 : 24 taper can be written as
P 2 D 4 kgf
K7 : 24 = = 2 (5.35)
y3 d micron
As the d/D ratio increases, the stiffness of the tapered joint decreases.
Equation (5.34) has been derived on the assumption that manufacturing errors of the taper are negligible.
As a matter of fact, any difference in the taper angle of the hole and shank severely affects the stiffness of
the tapered joint. A difference of 30–40¢ in taper angles can easily reduce the stiffness of a 7 : 24 taper by
10–15 times. The difference in taper angles should normally not exceed 1¢ and the shank should have the
larger taper angle.
300 Machine Tool Design and Numerical Control

raised by applying an axial tightening force on the centre or shank. The tightening force should be such as to
produce a pressure of p = 15–25 kgf/cm2 on the tapered surfaces. The required magnitude of the tightening
force can be determined from the expression,
P0 = p ◊ p(D – t tan a) ◊ t tan (a + r) (5.36)
where r = angle of friction, generally r = 12–14°
D and t are expressed in cm
In large machine tools mechanical tightening devices must be employed to

2. ensure that the larger tightening force is not transmitted to the thrust bearings of the spindle.
The principle of operation of such a device may be explained with the help of Fig. 5.11.2 The device con-
sists of bolt 2 screwed into tapered shank 1, spring-loaded levers 3, pulling rod 4, strong spring-loaded heli-
cal spring 5 and hydraulic cylinder 6 (Fig. 5.11a). When the hydraulic cylinder applies pressure, spring 5 is
compressed, rod 4 occupies the lowered position and levers 3 are tilted inwards due to the pressure of spring
7 (Fig. 5.11b). The cutting tool (e.g., face milling cutter) assembled on the adaptor with a tapered shank is

of springs 7 and open the levers, which engulf the bolt head. The shank now hangs freely, supported by the
levers.
For tightening the shank, the pressure of the hydraulic cylinder is released. Consequently, spring 5 begins
to expand; pulling rod 4 upwards. The tapered shank is also pulled upwards and tightened inside the tapered
hole. This position is shown in Fig. 5.11c.

3
2

(a) (b) (c) (d)

Fig. 5.11 Schematic diagram of a quick acting mechanical tightening device

 Design of Spindles and Spindle Supports 301

For unloosening the shank, rod 3 is pushed downwards by means of the hydraulic cylinder. In their down-
ward movement the levers push at the face of the tapered shank and unloosen it. The downward movement
of rod 3 is stopped immediately as the shank becomes free. The shank along with the cutting tool now again
hangs freely, supported only by spring-loaded levers. It can be easily removed from the spindle by a slight
downward pull of the hand.
This mechanised device not only serves the two functions listed above but has the added advantage of be-

in numerically controlled milling machines with automatic tool change provision. In some designs, a collet is
used instead of spring-loaded levers and a packet of leaf springs in place of spring 5.

Permissible Deflection and Design for Stiffness:

5.4.5 Additional Check for Strength
y1, y2 and y3 from Eqs (5.18), (5.26) from (5.32), respectively, the

y] depends upon the

machining accuracy required of the machine tool being designed. In general, it should be less than one-third
of the maximum permissible tolerance on radial run out of the spindle nose.
Machining accuracy depends not only upon the radial stiffness of the spindle, but also upon its axial and
torsional stiffness.
The axial displacement of the spindle unit consists of the axial deformation of the spindle plus the defor-
mation of the spindle thrust bearings. For a majority of machine tools, the latter component is predominant.
Therefore, the axial run out of the spindle can be kept within reasonable limits only by a proper selection of
the thrust bearings.

removal operations, such as gear and thread cutting in which the feed and primary cutting motions are ki-
nematically linked. The torsional deformation of the spindle unit consists of the deformation of the spindle

design calculations.
As stated earlier in Sec. 5.4, spindles are designed for stiffness, primarily radial. However, in heavily
loaded spindles the stiffness design must be substantiated by a strength check against fatigue failure. The
strength check requires that
n ≥ nmin (5.37)
where n = factor of safety against fatigue failure
nmin = minimum value of safety factor, generally equal to 1.3 to 1.5
For spindles subjected to combined bending and torsion, the factor of safety n is calculated from the ex-
pression,
(1 - a 4 )d e3s -1
n = (5.38)
10 (aM b ) 2 + (bM t ) 2

where a = di /de = ratio of the internal diameter to the external; de is in cm

s–1 = endurance limit of the spindle material, kgf/cm2
302 Machine Tool Design and Numerical Control

Mb = mean value of the bending moment acting on the spindle, kgf ◊ cm

Mt = mean value of the torque acting on the spindle, kgf ◊ cm
a
b
a can be determined from the following expression:
a = K s (1 + C) (5.39)
where Ks Ks = 1.7–2
C = Mba /Mb = ratio of amplitude of the bending moment to its average value
b can be calculated from the following formula:
s -1
b= + Kt Ct (5.40)
sy

where sy = yield stress of the spindle material, kgf/cm2

kt = Kt = 1.7–2
Ct = Mta /Mt = ratio of amplitude of torque to its average value
The values of C and Ct in Eqs (5.39) and (5.40), respectively depend upon the machining conditions. For

Mba = Mta = 0, and therefore, C = Ct = 0. In simple turning and drilling operations, distinguished by forma-
tion of continuous chips, the characteristic values are C = Ct= 0.1 – 0.2. The variation of bending moment
and torque is greater in intermittent cutting operations such as milling; for such operations C = Ct= 0.3–0.5.
The most important parameter in spindle design is the diameter of the front bearing journal. Typical values
of this diameter are given in Table 5.3.3

Table 5.3 Diameter (in mm) of front bearing journal3

Machine Tool
Power rating kW
Lathe Milling Cylindrical Grinding
1.5–2.5 60–80 50–90 —

2.5–3.5 70–90 60–90 50–60

3.5–5.5 70–105 60–95 55–70

5.5–7.0 95–130 75–100 70–80

7.5–11.0 110–145 90–105 75–90

11.0–14.5 140–165 100–115 75–100

14.5–18.0 150–190 — 90–100

18.0–22.0 220 — 105

22.0–30.0 230 — 105

 Design of Spindles and Spindle Supports 303

1. Additional Supports The radial stiffness of a spindle can be improved by supporting it at more than
two points. If an intermediate support is provided, the spindle constraints change from free supports to
clamped supports, thereby increasing its stiffness. However, in multiple bearing spindles the support journals
should be machined in one setting, as otherwise skewing and jamming of the spindle can occur due to large
misalignment.
2. Location of Bearings and Drive Elements The bearings should be located as near as possible to the

between the spindle nose and the point where the drive element transmits torque is subjected to bending and
torsion. It is, therefore, desirable that the drive element transmitting maximum torque to the spindle should
be located as near as possible to the front bearing.
3. Balancing The satisfactory performance of high-speed spindles is possible only if the spindle unit (after
mounting the gears, clutches, etc.) is dynamically balanced before assembly in the spindle head or headstock.
For general-purpose machine tools the permissible value of disbalance is 25 g ◊ cm at 2000 rpm.

5.5

compliance of the front and rear spindle supports. The rotational accuracy, which is one of the basic func-

operating conditions of spindles, anti-friction, hydrodynamic, hydrostatic, and lately, air-lubricated bearings
are used as spindle supports in different machine tools. Irrespective of the type of bearing, the common re-

1. guiding accuracy,
2. ability to perform satisfactorily under various conditions of spindle operation,
3. high stiffness,
4. minimum heating, as this can lead to additional spindle deformation, and
5. vibration stability, which is governed mainly by the damping.
Anti-friction bearings are one of the most widely standardised elements in industry and are manufactured
on a mass scale throughout the world. The distinguishing features of anti-friction bearings as compared to
sliding bearings are:
1. Low frictional moments and heat generation.
2. Low starting resistance.
3. High load capacity per unit width of the bearing.
4. Easy maintenance and less consumption of lubricants.

bearings. Ball bearings are less prone to heating, and therefore, permit larger rotational speeds. They are
also cheaper than roller bearings and less sensitive to small alignment errors. However, roller bearings have
higher load capacity.
304 Machine Tool Design and Numerical Control

A machine tool spindle experiences both axial and radial loads. These loads can be balanced either by
bearings that take up radial and axial loads separately or by bearings that take up both. In this regard, it should
be recalled that
1. cylindrical roller bearings can take up only radial load,
2. simple radial ball bearings are basically meant for taking up radial loads, but are generally able to
support axial loads too,
3. angular contact ball bearings can take up radial loads as well as relatively large axial loads in one
direction,
4. taper roller bearings can take up large radial and axial loads with equal ease,
5. ball thrust bearings are useful for supporting purely axial loads only, their maximum rotational speed
is just about 60% of a radial ball bearing of equal size, and
6. cylindrical roller thrust bearings are not recommended for general use on account of sliding between
rollers and races.
The number of possible combinations of various anti-friction bearings that can be employed in machine
-
ever, the viability of each combination must be assessed vis-a-vis the following parameters:
1. Radial stiffness of the spindle unit.
2. Axial stiffness of the spindle unit.
3. Radial run out of the spindle.
4. Axial run out of the spindle.
5. Heat generation.
6. Maximum permissible rotational speed, restricted by bearing wear and its heating.
7. Thermal deformation of the spindle.
8. Ease of manufacture and assembly of the spindle unit.
The relative performance and technological indices for eight combinations are given in Table 5.4. 4 It
may be pointed that these combinations are typical of a majority of spindle units in small- and medium-size
machine tools.
Based upon Table 5.4, some important operational features of different types of bearings can be summed
up as under:
1. Use of taper roller bearings considerably increases both radial as well as axial run outs.
2. Use of only cylindrical roller bearings at the front support greatly enhances axial thermal deformation
of the spindle nose.
3. Use of angular contact ball bearings at the front support results in low heat generation. This coupled
with a fairly satisfactory axial stiffness permits high rotational speeds.
It is evident from Table 5.4 that no single combination of bearings is ideal for all performance indices. In

on the functional accuracy of the spindle unit. Table 5.55 contains this information for some of the major
groups of machine tools.
The procedure for bearing selection is not described here and the reader should refer to a standard text on
Machine Design or a manufacturer’s catalogue for this purpose. However, some important features related to
the performance of bearings are discussed as follows.
 Design of Spindles and Spindle Supports 305

Table 5.4 Relative comparison of performance and technological indices of spindle supports

Heat general

S. Sketch Radial Axial Radial Axial Front Common Permissible Axial Ease of
No. stiffness stiffness run out run out bearing rpm thermal manufacture
deformation and
assembly

1. 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 A

2. 0.95 0.71 1.0 1.0 0.5 0.6 1.0 3.0 B

3. 1.0 3.0 1.0 1.0 1.2 1.15 0.75 1.0 A

4. 0.78 3.0 1.0 1.0 1.2 1.15 0.75 0.6 A

5. 1.0 2.5 1.0 1.0 0.5 0.75 0.8 3.0 B

6. 0.93 1.0 2.0 1.5 0.75 0.8 0.6 0.8 A

7. 1.05 1.0 2.0 1.5 1.4 1.4 0.6 0.8 A

8. 0.7 1.0 1.0 1.0 0.5 0.7 1.2 0.8 C

Notation: A—Very complicated Cylindrical roller bearing Angular contact ball bearing

B—Complicated Taper roller bearing Ball thrust bearing

C—Simple

Table 5.5

S. No. Machine tool Radial Axial Radial Axial Heat Permissible Thermal
stiffness stiffness run out run out generation rpm deformation
1. Lathes

(a) small-sized I I D D D D I

(b) medium-sized D I D D I I I

(c) single spindle D I D D I I D

automatic

Contd.
306 Machine Tool Design and Numerical Control

Table 5.5 Contd.

2. Work head spindle

of universal and
internal grinding
machine

(a) small-sized D U D D U U U

(b) medium-sized D I D D U U U
and heavy-duty

3. Universal milling D D D D I I D
machine

Notation: D—of decisive importance; I—of average importance; U—of relatively minor importance

5.5.1 Preloading of Anti-friction Bearings

as well as the stiffness (compliance) of the spindle supports. The discussion which follows will reveal the

The variation of spindle deformation d due to a radial force P is depicted in Fig. 5.12. If the bearing is
assembled with a clearance, a reversal of the direction of the applied force results in an abrupt change of
deformation (Fig. 5.12a). This is highly undesirable from the point of view of machining accuracy. Bearings
assembled with interference are free of this shortcoming and are distinguished by a smooth d-P curve (Fig
5.12b). It can be seen from Fig. 5.12 that the rate of deformation is initially high but later on decreases and
at large loads becomes virtually constant. This is due to the fact that as the load increases, its distribution
between the rolling members becomes more uniform. Since a larger number of rolling members support the

stiffness of the spindle unit.

d d

P P

(a) (b)

Fig. 5.12 Effect of radial force on spindle deflection when (a) bearing is assembled with a clearance
(b) bearing is assembled with an interference
 Design of Spindles and Spindle Supports 307

Table 5.6 Effect of bearing play on spindle unit stiffness6

S. No. Play of front bearing,

microns spindle nose, microns
bending, microns compliance of
support, microns

1. 15 (clearance) 14 16 30

2. – 5 (interference) 13 6 19

3. – 15 (interference) 11 5 16

It is evident from Table 5.6 that the assembly of a bearing with a small interference sharply reduces sup-

not so perceptibly. The reason for such behaviour is that as the interference increases, the end constraints
change from a simply supported beam to those of a clamped beam. This is equivalent to providing an addi-

nose.
A glance at Table 5.6 revels that increase of interference
from 5 to 15 microns barely yields a 1-micron reduction of
. However, a large inter-
ference is accompanied by excessive heating and also reduces
D
the bearing life on account of large contact deformation. Ob-
viously, optimum interference is one which precludes clear-
ance but does not result in excessive heating of the bearing.
Interference in the assembly of rolling elements is achieved by
preloading them.
Preloading of a bearing involves relative axial displacement
of the inner and outer races by a small amount (Fig. 5.13). The
methods of applying preloading in radial and angular contact
ball bearings that are generally mounted in pairs are shown in
Fig. 5.14. A constant preloading is achieved either by grinding Fig. 5.13 Schematic diagram depicting
off the faces of the inner races (Fig. 5.14a) or by inserting spac- preloading by relative axial
ing rings of different widths between the inner and outer races displacement of the bearing
(Fig. 5.14b). If the bearing rotates at high rpm, the initial pre- races
load has a tendency to weaken. In such cases, especially when
bearings are small, the preloading can be applied by means of springs which ensure a constant preload that
can be accurately adjusted (Fig. 5.14c). This method is adopted in precision bearings.
308 Machine Tool Design and Numerical Control

a a

a a

(a) (b) (c)

Fig. 5.14 Methods of preloading ball bearings

Cylindrical double roller bearings are generally mounted on tapered journals. The preloading is obtained
through axial displacement of the inner ring with an adjusting nut (Fig. 5.15). The utility of this arrangement
can be considerably improved by using a split nut. As the bearing wears, the preset value of preloading de-
creases. When this occurs, the split nut is removed, slightly ground and mounted again. The initial preloading
is then reinstated by additional axial displacement of the adjusting nut by a distance equal to the ground layer
removed from the split nut. However, this procedure should be discontinued after a few regrindings of the
split nut, because when the bearing wear becomes large, the roller length is partially in contact with the worn
surface of the race way and partially with the unworn surface. This results in excessive heat generation and
non uniform rotation. Axial displacement of the bearing race by the threaded nut does not provide uniform
contact of the face and can result in deformation of the spindle. The non-uniformity of face contact may be
somewhat reduced by inserting a sleeve between the nut and bearing as shown in Fig. 5.15b. The spherical

inner race is seriously impeded by large static friction between the inner race bore and the spindle. It requires
an axial force of 2000–3000 kg to shift the race of a small bearing about 100 mm in diameter. Application of
-
ness) of the bearing race, resulting in poor rotational accuracy of the spindle. To reduce friction at the time of
adjustment of bearing play, oil is force fed into the interface between the race bore and spindle.

Split nut Sleeve Split nut

(a) (b)

Fig. 5.15 Methods of preloading cylindrical, double roller bearings

 Design of Spindles and Spindle Supports 309

The oil is supplied by an injector through a small oil hole in the spindle and an annular groove (Fig. 5.16).
In this way, the required axial force for bearing adjustment can be reduced ten-fold. After completing the

Annular Oil hole

grooves

Injector

Fig. 5.16 Schematic diagram describing the oiling arrangement to reduce friction at the bearing-spindle
interface during bearing play adjustment

Taper roller bearings are preloaded by the methods shown in Fig. 5.17. In the method shown in Fig. 5.17a,
the inner and outer races are axially displaced with the help of nuts. This method is applied only in non-
precision bearings because the axes of inner and outer races get skewed. In the Gamet bearing arrangement
(Fig. 5.17b), the outer race is axially displaced by means of springs, whereas in the Timken bearing (Fig.
5.17c), this is achieved by supplying oil or air under controlled pressure.

(a) (b) (c)

Fig. 5.17 Methods of preloading taper roller bearings

310 Machine Tool Design and Numerical Control

5.6 SLIDING BEARINGS

by using needle roller bearings. However, wide application of needle bearings in machine tool spindles is

an eccentric load. When there is a constraint of space, generally sliding bearings are preferred. Sliding bear-
ings are used when
1. rotational speeds are so high that anti-friction bearings become uneconomical due to their short service
life,
2. accuracy of spindle rotation is required to be very high, and
3. the bearings are subjected to shocks and vibrations; the inherent damping of sliding bearings is

Sliding bearing is a general term that covers all bearings that do not use rollers or balls. These bearings
operate under conditions of sliding friction between the bearing bore and spindle journal, which are separated

to-metal contact; friction conditions at the interface of mating surfaces are of semi-liquid type and
these bearings are known as sleeve bearings,

or hydrostatic journal bearings.

The Hersi–Shtribek diagram shown in Fig. 5.18 depicts f
f as a function of
the quantity
1
mw
l=
p
where m = absolute viscosity of the lubricant 3
w = angular velocity of rotation of the journal
2
p = average pressure per unit area of the support- fmin
ing surface mw
l= p
l1 l2 l3
At very low rotational speed when l < l 1, the lu-
bricant film is extremely thin—of the order of 0.1 Fig. 5.18 Hersi-Shtribek diagram

not change with l. This region which lies to the left of point

For higher values of l lying between l1 and l2, the friction conditions are of semi-liquid type and the
l > l2, liquid friction
conditions prevail and they represent the operating conditions of hydrodynamic bearings.
 Design of Spindles and Spindle Supports 311

The sketch of a sliding journal bearing is shown in Fig. 5.19. Diameter d of the journal is always less than
diameter D of the bearing. At zero rotational speed, the journal rests on the bearing and metal-to-metal con-
tact takes place at point A (Fig. 5.19a). As the journal begins to rotate in the anti-clockwise direction, it tends
to roll up the bearing surface due to the friction force and moves to a position B
consists of two parts—a converging wedge above line BE and a diverging wedge below it. Owing to the hy-
drodynamic effect a positive pressure builds up in the converging wedge. This hydrodynamic force increases
with increase of rotational speed and overcomes the frictional force. As a result, the point of contact moves
to point C (Fig. 5.19c). As long as l < l2, the metal-to-metal contact at point C persists. However, when the
rotational speed is such that l > l 2, the hydrodynamic force becomes large enough to lift the journal and a
D (Fig. 5.19d).

E F

x
x x

x
x

C D
e B
A
hmin

(a) (b) (c) (d)

Fig. 5.19 Schematic diagram describing the working principle of a sliding journal bearing

bearings have less friction than a full journal bearing, but can be used only where the load always acts in one
direction.

+
+ + +
+ +

120° 120°
(a) (b) (c)

Fig. 5.20 Types of sliding journal bearings: (a) Full (b) 120° partial (c) 120° fitted

5.6.1 Sleeve Bearings

Sleeve bearings are designed for wear resistance. The design conditions are
P
p = £ [p] (5.41)
d ◊l
312 Machine Tool Design and Numerical Control

pv £ [pv] (5.42)
where p = bearing pressure
P = load on the journal
d = diameter of the journal
l = length of the bearing
v = peripheral speed
[p] = permissible value of bearing pressure
[pv] = permissible value of the product of bearing pressure and peripheral speed
The permissible values of [p] and [pv] vary in a wide range depending upon factors such as bearing ma-
terial, sliding velocity, cooling and lubrication conditions, etc. These are given for some important sliding
bearing materials in Table 5.7.

Table 5.7 Permissible values of [p] and [pv] for some bearing materials

S. No. Material v, m/s [p], kgf/cm2 [pv] kgf, m/cm2 ◊ s

1. Grey iron 0.5 40

1.0 20

2.0 1.0

2. Anti-friction cast iron 0.2 90 18

2.0 0.5 1.0

3. 1.0 120 120

5.0 5.0 25

4. Bronze 3 50 100

5. Aluminium bronze 4 150 150

6. Tin bronze 10 150 150

7. Graphite bronze (20–25% porosity) 0.2 60

0.4 10

8. White metal 12 250 300

9. Aluminium alloy 12 250 300

10. Zinc alloy 10 120 120

The selection of the sliding bearing material is based upon the following considerations:
1. High wear resistance.
2. High fatigue strength.
 Design of Spindles and Spindle Supports 313

3. High compressive strength.

4. High thermal conductivity.
5. High conformability to accommodate spindle deformations and reduce edge pressures.
6. High corrosion resistance.
7. Low modulus of elasticity.
Typical values of p and v that occur in machine tool spindle units and the recommended sliding bearing
materials are given in Table 5.8.
A sliding bearing made from anti-friction cast iron has poor conformability, therefore, the spindle should
have high stiffness to avoid large pressures. Copper alloy compositions are used in the form of bimetallic
sleeves. A layer of approximately 1.0 mm thickness is deposited on a steel or cast iron sleeve by the centrifu-
gal casting method. Porous graphite bronze bearings are employed at low sliding speeds under conditions of
variable loading. Aluminium alloy is employed as a replacement for white metal and zinc alloy for babbits.

Table 5.8 Recommended compositions of sliding bearing materials

S. No. v, m/s p, kgf/m2 pv, kgf, m/cm2 ◊ s Recommended material

< <
1. 2 10 20 Anti-friction cast iron

2. 2.5 8–16 20–40 Zinc alloy (84.5–86.5% Zn,

9–11% Al, 4.5–5.5% Cu)

3. 5 8–12 40–60 Aluminium alloy (88–91% Al,

7.5–9.5% Cu, 1.5–2.5% Si)

4. (a) 5 12–16 60–80 Bronze (86% Cu, 6% Sn, 6%

Zn, 3% Pb)

(b) 5 12–16 60–80 Aluminium bronze (87% Cu,

9% Al, 4% Fe)

5. 10 8–10 80–100 White metal (68.5–71.5% Cu,

27.5–31.5% Pb)

6. 10 10–12 100–120 Tin bronze (89.5% Cu, 10%

Sn, 0.5%, Pb)

5.6.2 Hydrodynamic Journal Bearings

As already stated in Sec. 5.6.1, a sleeve bearing begins to operate as a hydrodynamic bearing when 2.
The angular velocity at which the transformation from semi-liquid friction to liquid friction occurs is given
by the expression,
p ◊ y 2 –1
w2 = ,s (5.43)
m ◊ S0
314 Machine Tool Design and Numerical Control

where p = average bearing pressure, kgf/m2

D-d
y = = relative diameteral clearance
d
m = absolute viscosity of lubricant, kgf ◊ s/m2
S0 = critical value of Sommerfeld number
Obviously, when w > w2, the conditions are those of liquid friction, whereas when w < w2, the conditions
are of semi-liquid type. The values of S0 are given in Table 5.9.
The theory of hydrodynamic lubrication has been successfully developed to yield design expressions for

parameters that must be determined are discussed below.

1. Length to Diameter (l/d) Ratio The optimum value of l/d ratio that is often recommended is l. A larger
l/d ratio can be taken when the rigidity and alignment conditions are favourable and cooling of the system is

Table 5.9 Critical values of the Sommerfeld number S0

Journal diameter d in mm
l/d
30 40 50 60 70 80 100 150 200
y = 0.001
0.6 0.28 0.35 0.42 0.53 0.65 0.8 1.0 2.0 3.0

0.8 0.44 0.54 0.64 0.80 0.95 1.2 1.5 2.7 4.0

1.0 0.58 0.72 0.85 1.0 1.2 1.5 1.9 3.3 4.5

1.2 0.70 0.80 1.0 1.2 1.4 1.7 2.2 3.7 5.0

y = 0.002
0.6 0.42 0.53 0.65 0.8 1.0 1.4 2.0 3.0 5.0

0.8 0.64 0.80 0.95 1.2 1.5 1.9 2.7 4.0 6.0

1.0 0.85 1.0 1.2 1.5 1.9 2.4 3.3 4.5 7.0

1.2 1.0 1.2 1.2 1.7 2.2 2.6 3.7 5.0 8.0

y = 0.003
0.6 0.65 0.8 1.0 1.4 2.0 3.0 4.0 5.0 6.0

0.8 0.96 1.2 1.5 1.9 2.7 4.0 5.0 6.0 8.0

1.0 1.2 1.5 1.9 2.4 3.3 4.5 6.0 7.0 9.0

1.2 1.4 1.7 2.2 2.6 3.7 5.0 6.5 8.0 10.0
 Design of Spindles and Spindle Supports 315

not a problem. A value of l/d

While selecting the l/d ratio, it should generally be borne in mind that a larger l/d ratio provides for a

Also, a bearing with a large l/d ratio is more susceptible to the metal-to-metal contact between the bearing
and journal surfaces.
2. Bearing Clearance
-
nal and bearing. This clearance, also known as critical clearance, can be found from the expression,
D - d yd
hcr = Rzb + Rzj + y0 = = (5.44)
2 2
where Rzb = height of micro irregularities on the bearing surface
Rzj = height of micro irregularities on the journal surface
y0 =
In design practise, the following approximate expression, which provides for a certain margin of safety,
can be used.
hcr = 2(Rzb + R zj) (5.45)

rate of the lubricant and hence brings down the operating temperature. The following values of y can be used
as a guide for preliminary selection for different bearing materials (Table 5.10).
In general, the greater the l/d ratio and w, and the lower the pressure p, the higher should be the bearing
clearance. The following two empirical relationships may be used for approximate calculation of the y value:

Table 5.10 Recommended value of y different bearing materials

Bearing materials y = (D – d) / d
Tin base babbit 0.0005

Cadmium silver 0.0008

Copper-lead 0.001

Silver-lead-indium 0.001

Aluminium 0.001
316 Machine Tool Design and Numerical Control

0.0508
y = 0.001 + (5.45)
d
y = 0.8 ¥ 10–3v0.25 (5.46)
where v is the peripheral speed in m/s
d is the journal diameter in mm
The recommended values of diametral clearance D – d for various machine tools are given below:

High precision machines 0.004–0.010 mm

Grinding machines 0.010–0.015 mm

Normal accuracy lathes 0.015–0.025 mm

Automatic and production lathes 0.020–0.025 mm

Normal accuracy milling and drilling machines 0.02–0.03 mm

3. Viscosity of Lubricant After having found y, the required viscosity of the lubricant that ensures hydro-
dynamic lubrication can be determined from the expression,
py 2
m≥ kgf ◊ s/m2 (5.47)
w S0
where S0 represents the critical Sommerfeld number. It is not desirable to select a lubricant with a viscosity
much greater than the minimum essential, because this leads to increase of frictional losses in the bearing.
4. The load capacity of a hydrodynamic bearing is given by the
expression,
mw
P = ◊ l ◊ d ◊ CL (5.48)
y2
here l and d are in m, and CL l/d ratio and
the eccentricity ratio e,
e
e= (5.49)
c
D - d dy
c= =
2 2
is known as the radial clearance.
CL is related to the Sommerfeld number S by the expression CL = 1/2pS. For different
values of d/l and e, CL can be determined from Table 5.11.
 Design of Spindles and Spindle Supports 317

Table 5.11 L for hydrodynamic journal bearings

l/d
0.4 0.5 0.6 0.7 0.8 0.9 0.975
360° Full bearing
0.25 0.0559 0.0887 0.1472 0.2697 0.5981 2.08 15.758

0.5 0.2023 0.3121 0.4972 0.8563 1.716 4.935 26.134

0.75 0.3905 0.5969 0.911 1.4846 2.7435 6.9167 —

1.00 0.6121 0.8896 1.315 2.0394 3.6171 8.272 33.576

1.25 0.8161 1.1617 1.6753 2.5182 4.1882 9.1468 —

1.5 1.0009 1.4011 1.9648 2.8754 4.6400 9.830 —

180° Partial bearing

0.5 0.209 0.317 0.655 0.819 1.572 4.261 25.62

0.7 0.361 0.538 0.816 1.312 2.399 6.029 31.88

1.00 0.589 0.853 1.263 1.929 3.372 7.772 37.00

1.20 0.723 1.033 1.489 2.247 3.787 8.533 39.04

1.5 0.891 1.248 1.763 2.60 4.266 9.304 41.07

120° Partial bearing

0.5 0.188 0.261 0.462 0.826 1.676 4.717 29.33

0.70 0.299 0.441 0.709 1.221 2.365 6.213 34.30

1.00 0.436 0.633 0.992 1.644 3.042 7.508 38.08

1.20 0.506 0.722 1.126 1.838 3.335 8.075 39.58

1.5 0.583 0.831 1.271 2.041 3.667 8.618 41.06

After having determined the l/d ratio, y and m, the designer will be faced with one of the two design prob-
lems which are discussed as follows.
Design Problem 1 The load on the bearing is known and it is required to check whether the bearing having
the l/d, y and m values as found above, will perform satisfactorily.
CL is determined from Eq. (5.48). Then for the known l/d ratio, the ec-
centricity ratio e is determined from Table 5.11. Now, knowing and radial clearance c
thickness is determined from the expression,
hmin = c(1 – e) (5.50)
318 Machine Tool Design and Numerical Control

If hmin > hcr, the bearing operates under liquid-friction conditions and the design may be accepted.
If, however, hmin < hcr, then corrective measures must be taken. These may be
1. reducing hcr
2. increasing hmin either by increasing c or by increasing the l/d ratio as this yields a smaller value of e.
Design Problem 2
having l/d, y and m values as found above will be able to take the operating load.
The recommended values of hmin for different conditions are given in Table 5.12. These values satisfy the
condition hmin > hcr.

Table 5.12 Recommended value of hmin for different conditions

S. No. Operating condition hmin, mm

1. In order to pass dirt particles and prevent scoring 0.0026

2. Finely bored bronze bearing 0.0026

3. 0.0026
and aircraft engines

4. Babbit bearings running at high speeds 0.02

5. General recommendation 0.0015–0.0002 mm per mm of

the bearing diameter

For the known value of c and the selected value of hmin, the eccentricity ratio is determined from Eq.
(5.50). Next, for the particular values of l/d ratio and , CL is found from Table 5.11.
-
ing capacity is less than the load acting on it, the former must be improved by taking appropriate corrective
measures, which include
1. reduction of y,

C L,
2. increase of m, i.e., application of a more viscous lubricant,

considerations, and therefore, only l may be increased, thus providing a greater l/d ratio. The l/d ratio

the l/d ratio.

Check for Thermal Equilibrium The rotation of a journal in a hydrodynamic bearing is resisted
by the lubricant. The viscous friction offered by the lubricant results in a frictional force which must be

check for hydrodynamic bearings is

tb £ [tb] (5.51)
 Design of Spindles and Spindle Supports 319

where tb = temperature of the bearing

[tb] = permissible value of bearing temperature; generally [tb] = 60–75°C
The heat is dissipated through the housing, bearing body and journal. If the thermal equilibrium between
heat generated and heat dissipated is established such that the bearing temperature is less than the permissi-
ble, then the design is accepted. However, if the reverse takes place, then either the design parameters should

The heat generated can be determined from the expression,

wd ¸
W = f ◊P◊ kgf ◊ m/s ÔÔ
2
˝ (5.52)
3600 f ◊ P ◊ w d
= ◊ k cal/h Ô
427 2 Ô˛

where P = load on the bearing

w = rotational speed of the journal

p mw
f = (5.53)
y p

p mw
f= + 0.55 y h (5.54)
y p
1.5
l Êdˆ
for short bearing having < 1, h = Á ˜
d Ël¯
l
for bearings having > 1, h = 1
d
The heat dissipation through the body and shaft is assumed to be proportional to the free surface area of
the bearing assembly. It is given by the expression,
W1 = kA(tb – ta)kcal/h (5.55)
2
where k = ◊ h ◊ °C
A = free surface area of the bearing assembly, m2
ta = ambient temperature, °C
tb = bearing temperature, °C
k can be found from the following expression:

k = 6 + 10 v (5.56)
320 Machine Tool Design and Numerical Control

where v
rotational speed of the journal and has a minimum value of 1 m/s
Generally, k, lies between 25–35 kcal/m2 ◊ h ◊ °C
The free surface area of the bearing assembly depends upon the design and size of the bearing. On an
average, it may be taken equal to 25d2 or 20dl. However, for very simple assemblies, it may reduce to 12dl,
whereas for bearings mounted in high housings it may go up to 40dl. An additional area of (5–8)d2 per
journal should be added to the above to account for heat dissipation through the journal. The lower value of
5d2 is recommended for small bearings (d < 100 mm), while the higher value of 8d 2 is for large bearings
(d > 100 mm).
For thermal equilibrium, Eqs (5.52) and (5.55) are equated and the bearing temperature tb is calculated.
If tb is found to be less than the permissible value, there is no need for forced circulation of the lubricant.
However, if tb exceeds the permissible value, forced lubricant circulation is essential to carry away the extra

expression:
W2 = 60CQg (t0 – ti), kcal/h (5.57)
where C ◊ °C
Q
g = density of the lubricant, g/cm3
to = oil temperature after circulation through the bearing, °C
ti = oil temperature when it enters the bearing, °C

is calculated from the following thermal equilibrium equation after substituting t b¢ = [tb] the expression for
W1.
W = W1 + W2 (5.58)
-
2
ticles more than 2–3 microns in size. The oil is supplied at a pressure of 0.1–0.2 kgf/cm which is enough to

Finite Bearing Design considerations have been discussed till now with reference to an idealized

leakage. In an actual bearing, there is always some leakage of the lubricant at the ends. This reduces the load

lubricant must be supplied to compensate the leakage loss. An important question that arises in this context
is where to introduce the lubricant. In Fig. 5.19d, the wedge below line DF is diverging, and therefore,

in this region.

np. l/d
 Design of Spindles and Spindle Supports 321

ratio and e. It is plotted in Fig. 5.21 as a function of pd/l hP hF

e = 0.2
for different values of e. Also plotted on the same dia-
nF). 0.9 0.4 0.95
f can be
0.8 0.6 0.90
determined as
n 0.7
nf = p (5.59) 0.8 0.82
nF 0.6 0.80
Thus the load capacity of a finite hydrodynamic
0.5
journal bearing can be expressed as
mw 0.4
P = l ◊ d ◊ CL ◊ nP kgf (5.60) e = 0.6
y2 0.3 0.8

0.2
bearing to determine the bearing temperature, heat
generated in the lubricant can be determined as follows: 0.1 e = 0.2
0.4
wd 0
W = nf ◊ f ◊ P ◊ (5.61) 0 1.0 2.0 3.0 pd
2 l

nP Fig. 5.21 Leakage factors for load and friction

nF have been presented force in 120° partial journal bearings
in Fig. 5.21 only for a 120° partial bearing, these values
can be used with satisfactory approximation for journal bearings having different angles.

Multiple-wedge Bearings The hydrodynamic journal

bearing discussed till now has only one converging wedge and is
called single-wedge bearing. One of the major shortcomings of
these bearings is that the position of the journal inside the bearing
varies with the load and rotational speed. This positional change b
leads to unstable running of the journal, which is impermissible
in spindles of high accuracy and precision machine tools. These
requirements are met in bearings having several converging wedges
B
which uniformly surround the journal on all sides and ensure that
during rotation the position of the journal changes very little or
not at all. These bearings are known as multiple-wedge bearings.
Figure 5.22 shows a self-adjusting tilting-pad journal bearing in
which the wedges are formed due to tilt of the pads. These bearings
generally have three to eight wedges.
The recommended values and expressions for calculating the Fig. 5.22 Schematic diagram of a
design parameters of multiple-wedge bearings are discussed be- multiple-wedge bearing
low.7
1. Ratio of pad length l to journal diameter d
l = 0.75d for grinding machines
l = (0.85–0.9) for high precision lathes and boring machines
322 Machine Tool Design and Numerical Control

2. Pad width B The pad width should be such that it subtends an angle of b = 50–60° at the centre. It should
generally satisfy the condition
B = (0.6 – 0.8 )l
3. Radial clearance c The minimum clearance between the pad and journal depends upon their surface
Rz = 0.8–1.6 microns), then
For d = 30–50 mm, c = 3–5 micron
For d = 50–100 mm, c = 5–10 micron
For d = 100 mm, c = 10–15 micron
4. Viscosity of lubricant In multiple-wedge journal bearings, low viscosity oils are used. The recommend-
ed kinematic viscosity of the oil is 4–8 cS at 50°C, which corresponds to an absolute viscosity of 4–8 cP. It

because the resultant mixture has poor lubricating properties and gradually becomes more viscous due to

5. Bearing capacity The bearing capacity is calculated approximately, assuming each wedge to be a slider
bearing. The load capacity per wedge of an unloaded bearing is given by the expression,
mn dB 2 ◊ l
P0 = 10–2 ◊ CL2, kgf (5.62)
(2c) 2
where m = absolute viscosity of the lubricant, cP
n = rpm of the journal
1.25
CL2 =
1 + ( B/l ) 2
d, B, l are in cm; c is in microns
It is evident from Eq. (5.72) that considerable hydrodynamic force is developed in each wedge even
if there is no load on the bearing. These forces balance each other and tend to retain the rotating journal

eccentricity e. External load P and eccentricity e are related through the expression,

È 1 1 ˘
P = P0 Í
2
- ˙ (5.63)
Î (1 - 0.5e ) (1 + e ) 2 ˚
where e = e/c is the eccentricity ratio.
Equations (5.62) and (5.63) can be used for solving the two general design problems which have been
earlier discussed with reference to single-wedge bearings.
As a rule, multiple-wedge hydrodynamic journal bearings are provided with forced circulation of the
lubricant. Assuming that heat dissipation through the bearing assembly and housing is negligible as compared
to the heat carried away by the circulating lubricant, the increase in temperature of the lubricant can be
calculated from the expression,
860N F
Dt = °C (5.64)
Cg Q
 Design of Spindles and Spindle Supports 323

where C = ◊ °C.
g = density of the lubricant, g/cm3
Q =
NF = power loss due to friction, kW
The frictional power loss can be found from the following relationship:
NF = 7.5 ¥ 10–10 m n2d3z kW (5.65)
where m = absolute viscosity of the lubricant, cP
z = number of wedges
d = diameter of the journal, cm

5.6.3 Hydrostatic Journal Bearings

of the order of 0.1–0.2 microns. Besides, the stiffness of hydrodynamic bearings changes with lubricant vis-
cosity, temperature and rotational speed of the journal. These shortcomings are absent in hydrostatic journal
bearings. The basic features and principle of operation of hydrostatic journal bearings are essentially similar
to those of hydrostatic pad bearings discussed in Sec. 4.4.1. Hydrostatic journal bearings may be single-pad
(Fig. 5.23a), multiple-pad (Fig. 5.23b) and multiple-recess (Fig. 5.23c) type.
Constant pressure
manifold
Restrictor
2
1
3

Pad domain
(a) (b) (c)

Fig. 5.23 Types of hydrostatic bearings: (a) Single-pad (b) Multiple-pad (c) Multiple-recess

The single-pad journal bearing has less than 180° included angle and normally supports only unidirec-
tional load akin to partial hydrodynamic bearings. Multiple-pad bearings are employed when the load to be
324 Machine Tool Design and Numerical Control

supported is not unidirectional and varies as much as ± 180°, e.g., an oscillating load, a reversing load, etc.
In multiple-pad bearings, the number and location of pads are dictated by the angular variation of the applied
radial load. For instance, a single-pad having the included angle > 60° is adequate to support a purely unidi-
rectional load. Two pads at an angle of 120° can be employed when the load varies within a maximum of 80°.
If a purely reversing radial load acts on the journal, a two-pad bearing having pads 180° apart can be used.
For the general case in which load variation can assume any arbitrary value, we can use multiple-pad bearings
having three or more pads distributed round the journal. Single-pad and multiple-pad bearings are rarely used
in machine tool spindles due to the fact that there is considerable variation in direction as well as the magni-
tude of the cutting force depending upon cutting conditions, type of machining operation and other factors.
Multiple-recess bearings are akin to full (360°) hydrodynamic journal bearings and they can support all
types of radial loads, including reversing and rotating loads. In common practise, the term hydrostatic journal
bearing is understood to stand for a multiple-recess journal bearing and, therefore, the simpler term will be
used in all subsequent discussion. The main difference between a multiple-pad and multiple-recess bearing
is the absence of pressure-relieving grooves between the pads in the latter. Due to this, the pressure between

Figure 5.23c shows a full hydrostatic journal bearing having four oil pockets located at 90° to each other.
The lubricant is supplied to each pocket through an individual restrictor. When there is no load on the journal,
the latter occupies a concentric position. The clearances between the journal and bearing are equal and so are
the pressures in the recesses. Now assume that an external load acts on the journal, moving it towards recess

p across the restrictor, resulting in higher recess pressure which tends to restore the journal to the initial

hydrostatic bearing pad when the journal is in a concentric position is known as pad preload force and it is
given by the expression,
Ppr = CL ◊ A ◊ p0 (5.66)
where CL =
A = pad area, m2
p0 = pocket pressure, kgf/m2
Flow through the pocket is
p0 h3 3
Qpr = m /s (5.67)
mCF¢
where h =
m = absolute viscosity of the lubricant, kgf ◊ s/m2
 Design of Spindles and Spindle Supports 325

1
CF¢

8
Rectangular
pads with rectangular pockets are of two types:

1. Pads with equal sill lengths.

2. Pads with sill lengths in axial and circumferential directions proportional to the pad length in the same
directions.

Strictly speaking, the multiple-recess hydrostatic journal bearings do not have individual pads. Pad coef-

circumferential direction. Thus, in Eq. (5.66), A is taken as the area of the domain of each pocket. When a ra-

the practical hydrostatic journal bearing

given in Figs 5.24–5.28 at supply pressure,
pp = 2p0 (5.68)

CL 1/CF¢
R
1.0 6.0

r CL 5.0
0.8

0.6 4.0

d 0.4 3.0
1/C¢F
0.2 2.0

0.0 r
0.0 0.2 0.4 0.6 0.8 1.0 R

Fig. 5.24 Load and flow coefficients for a cylindrical pad with circular pocket
326 Machine Tool Design and Numerical Control

1/C¢F l/L
20.0 ⎫
⎬ 0.9
⎭
I L
10.0
8.0 ⎫
⎬ 0.7
6.0 ⎭
⎫
4.0 ⎬ 0.5
⎭
⎫
⎬ 0.1
d ⎭
2.0

q2 2q2<60°
q1 1.0
2q2<180°
0.8
L
0.6 dq2
0.2 0.4 0.6 0.8 1.0 2.0 4.0 6.0

Fig. 5.25 Flow coefficient for a cylindrical pad with rectangular pocket having equal sill lengths (L – / =
2 1)

CL l
L
1.0 }0.9
⎫
0.7 ⎬
⎭
0.8 ⎫
0.5 ⎬
⎭
⎫
0.3 ⎬
0.6 ⎫⎭
⎪
⎬ 0.1
⎪
⎭
0.4

2q2=180°

0.2
2q2=60°

L
0
0.2 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 dq
6.0 2

Fig. 5.26 Load coefficient for a cylindrical pad with rectangular pocket having equal sill area
 Design of Spindles and Spindle Supports 327

1/C¢F l/L

20.0 ⎫
⎬ 0.9
⎭
I L
10.0
8.0
⎫
6.0 ⎬ 0.7
⎭
⎫
4.0 ⎬ 0.5
⎫⎭
⎪
⎬ 0.1
d ⎪
2.0 ⎭

q2
q1
1.0 2q2<60°
0.8 2q2=180° L
0.6 dq2
0.2 0.4 0.6 0.8 1.0 2.0 4.0 6.0

Fig. 5.27 Flow coefficient for a cylindrical pad with rectangular pocket having equally proportional sill lengths
1 2)

l
CL L
1.0
ü
ý 0.9
þ
0.8
ü
ý 0.7
þ
0.6
ü
ý 0.5
þ
0.4
ü
ý 0.3
þ
0.2
ü
2q2=180° ý 0.1
þ
2q2<60° L
0
0.10 0.2 0.3 0.4 0.6 0.8 1.0 1.5 2.0 3.0 4.0 6.0 dq 2

Fig. 5.28 Load coefficient for a cylindrical pad with rectangular pocket having proportional sill area

The load capacity P of a hydrostatic journal is found as P = Ppr. It may be alternately determined by the
procedure discussed below.
p = pp ◊ A ◊ CL ◊ Cp(e, k) kgf (5.69)
Cp(e, k) depends
upon
1. eccentricity ratio e = e/c, and
2. k,
328 Machine Tool Design and Numerical Control

In high accuracy and precision machine tools in which hydrostatic journal bearings are employed, the load
on the journal is small, and hence eccentricity e is also small. For e 0.4, the value of Cp(e, k) = (3/2)e can
l/d ratio of the bearing is taken equal
to unity, the effective area of all the pads (A ◊ CL) can be taken equal to 0.5d2. Upon substituting the above
values of A ◊ CL and Cp(e, k)
of a hydrostatic journal bearing:
e
P = 0.75 ◊ e ◊ d 2 ◊ pp = 0.75 ◊ d 2pp (5.70)
c
The stiffness k of a hydrostatic journal bearing is obtained from the expression K = dP/de. For small values
of eccentricity, we can assume dP/de ª P/e, and therefore, from Eq. (5.70),
P d 2 pp
K = = 0.75 (5.71)
e c

3P
k ª (5.72)
c
The viscosity m of the lubricant is determined from consideration of minimum frictional loss and it is given
by the expression,
2
8 Ê c ˆ pp
m = 2.2 ¥ 10 Á ˜ cP (5.73)
Ë r¯ n

where c/r = ratio of radial clearance to the journal radius; both c and r are in cm
n = rpm of the journal
pp = supply pressure, kgf/cm2

( p p - p0 )
Qc = (5.74)
Rrc

Q0 = k0 2( p p - p0 ) (5.75)

In Eqs (5.74) and (5.75), Rrc and k0

They depend upon the geometrical parameters of the restrictor and the density and absolute viscosity of the

Equations (5.66)–(5.75) are the basic design equations of hydrostatic journal bearings. They can be used
for determining pad geometry and dimensions, restrictor parameters, bearing load capacity, etc. depending
upon the particular design problem.

5.6.4 Air-lubricated Bearings

-
tion in the lubricant. Rotational speeds can be considerably raised by using a lubricant of lower viscosity. This
 Design of Spindles and Spindle Supports 329

explains the recent development of using air-lubricated bearings in spindles of some high-speed precision
machines. Air has a viscosity which is approximately 100 times less man that of kerosene and more than 1000
times less than that of industrial oil. The main shortcomings that restrict the application of air lubricated bear-
ings are low load capacity and extreme sensitivity to overloading. The slightest over loading breaks the air

an accident. Air-lubricated bearings are made of babbit or other similar anti-friction materials, which permit
operation under dry friction conditions, though for an extremely short duration. Air-lubricated bearings can
be of two types:
1. Aerodynamic
2. Aerostatic
1. Aerodynamic Bearings -
ing on an air wedge at high rotational speed. They are used in light spindles which rotate at exceptionally high
speeds of the order of w = 103–104 rad/s, but are subjected to low pressures—p ª 1 kgf/cm2. The examples
of application of aerodynamic bearings are spindles of internal grinding machines, centrifuges, gyroscopes,
gas turbines, etc.
The load capacity of an aerodynamic bearing can be determined as
P = 0.5pa ◊ l ◊ d ◊ K1 ◊ K2 kgf (5.76)
where pa = air supply pressure, kgf/m2; generally pa = (2–5) 104 kgf/m2
l = length of the bearing, m
d = diameter of the bearing, m
K1, K2 =
K1 is a function of the Sommerfeld number. It is plotted in Fig. 5.29 for different values of
eccentricity ratio e.
K1
0.715
5.0

4.0
0.625

3.0

0.05
2.0

x=0.33
1.0

0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 2 4 6 10 20

mwr 2
2 p s=
c 2 pa

Fig. 5.29 Design curves for computing K1 as a function of Sommerfeld number for different values of e
330 Machine Tool Design and Numerical Control

K2 depends upon the ratio l/hd, where h is, in K2

its turn, a function of the Sommerfeld number. The curves K2 0.9
= f(l/hd) and h = f (mw r2/c2pa) are given in Figs 5.30 and 5.31, 0.8
respectively.
0.7
The heat generation in aerodynamic bearings is negligible,
and therefore, the check for bearing temperature is not nec- 0.6
essary. The air which is supplied to aerodynamic bearings is 0.5
dehydrated and mixed with a small quantity of oil to avoid cor- 0.4
rosion. Also, to avoid jamming of the journal and bearing sur-
0.3
faces at the time of starting or stopping the spindle (when the
aerodynamic effect is negligible), air is supplied at a pressure 0.2
of (2–5) 104 kgf/m2. 0.1

2. Aerostatic Bearings As has already been stated above, 0

1 1.2 1.4 1.7 2 2.5 3 4 6 10 20
aerodynamic bearings are used only in spindles rotating at ex- l
ceptionally high speeds. In the spindles of precision machine ηd
tools, aerostatic bearings have found wider application as they
can operate satisfactorily at relatively lower rotational speeds Fig. 5.30 Design curve for computing K2
also. as a function of l/hd

0.8
e =0.175

0.6 0.625

0.5
0.33
0.4

0.2

0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.7 2 2.5 4 6 10 20
2
μω r
2π S =
c 2 ρa

Fig. 5.31 Variation of h as a function of Sommerfeld number

The principle of operation of an aerostatic journal bearing is similar to that of a hydrostatic bearing. The
schematic diagram of an aerostatic bearing is shown in Fig. 5.32.
Air is supplied at a supply pressure of ps = 3–4 kgf/cm2. Generally, air is supplied at the ends as shown in
Fig. 5.32. However, in short bearings it can be supplied at the middle of the bearing. The air pockets should
 Design of Spindles and Spindle Supports 331

be connected by an annular microgroove which is shown in Section A-A in Fig. 5.32. The size and shape of
the microgroove are selected from the same considerations which were discussed in Sec. 4.5 with reference
to aerostatic slideways. The air is fed into the clearance between the bearing and journal through 0.2–0.3 mm
holes.

Fig. 5.32 Schematic diagram of an aerostatic bearing

The radial clearance is found as

c = (0.001 – 0.002)d, m (5.77)
where d = diameter of the bearing, m
The number of holes through which air is supplied is tentatively found from the expression,
Z = 20 ◊ p ◊ d (5.78)
and then rounded off to the nearest higher whole number. Generally, Z 4.
332 Machine Tool Design and Numerical Control

The load capacity of an aerostatic bearing can be determined from the following relationship:
P = 0.12ps ◊ d(l – l0) ◊ Cp(e) (5.79)
where ps = air supply pressure, kgf/m2
l = length of the bearing, m
l0 = distance of the air supply hole from the bearing edge, m
Cp(e) =
Cp(e) is given by the following approximate relationship:
p
Cp(e) = [(1 – e2) –1/2 –1] (5.80)
e
Upon expanding the series,
1 2 3 4 5 6
(1 – e2)–1/2–1 = e + e + e +
2 8 16
e < 0.3), the following simple
expression for load capacity of aerostatic bearing can be obtained:
P = 0.2 ◊ e ◊ ps ◊ d(l – l0), kgf (5.81)
Stiffness K = dP/de of the aerostatic bearing can be obtained by substituting e = e/c in Eq. (5.81) and
differentiating it.
dP P 0.2 ps ◊ d (l - l0 ) kgf
K = ª = (5.82)
de e c m
The recommended l/d ratio is
l/d = 1 – 1.5 (5.83)
Distance l0 should be taken as
l0 = 0.1l (5.84)

Review Questions
5.1 While turning a 750 mm long workpiece of 100 mm diameter between centres, the radial cutting
force was found to be 150 kgf when the tool was 200 mm from the tailstock. Calculate the machine
tool and system compliances if the stiffness of the saddle, headstock and tailstock is 3000, 4000 and
2500 kgf/ mm, respectively.
5.2 During the turning operation on a workpiece held between centres, the tangential cutting force
component was 150 kgf and the radial – 80 kgf. If the workpiece is 500 mm long and has diameter

tailstock stiffness is 35,000 and 15,000 kgf/cm, respectively.

5.3 A solid steel spindle (see Fig. 5.5a) transmits a torque of 40 kgf ◊ m through a gear of module
3 mm and number of teeth 80. Ratio a/b = 2 and compliances of front and rear bearings are 0.1 and
0.15 micron/ kgf, respectively. Determine the optimum l/c
spindle nose if force P1 = 150 kgf.
 Design of Spindles and Spindle Supports 333

5.4 Determine the outer and inner diameters of a hollow T2 T2

P2 P2
lathe spindle if their ratio = 2. The power on the spindle
is 4.5 kW and it is rotating at 700 rpm. The peripheral A B
C
and radial forces on the spindle are P2 = 215 kgf and
T2 = 78 kgf. A horizontal force P1 = 210 kgf and a 50 80
vertical force T1 = 60 kgf are acting on the spindle
170
nose. The spindle dimensions and loading are shown
in Fig. 5.33. Fig. 5.33 Schematic diagram of a lathe
5.5 The spindle of a high-speed precision lathe has a spindle
diameter 80 mm and can run at a speed up to 2500 rpm.
Select a suitable sleeve-bearing material for the above spindle if the maximum load on the bearing =
30 kgf and the length-to-diameter ratio of the journal = 1.
5.6 Select the design parameters of a full hydrodynamic journal bearing of copper-lead alloy for a journal
of 60 mm diameter to support a load of 150 kgf at 3000 rpm. Assume hcr = 10 microns and hmin / hcr >
1.1.
5.7 Design a full hydrodynamic babbit bearing for a journal of diameter 80 mm rotating at 2000 rpm.
Determine its load capacity if the average bearing pressure is not to exceed 8 kgf/cm2. Check the
bearing against excessive heating if the maximum permissible bearing temperature is 70°C and the
maximum ambient temperature = 40°C.
5.8 Design a multiple-wedge hydrodynamic bearing for a grinding machine if the journal diameter =
60 mm and it rotates at 3000 rpm. What is the load capacity of the unloaded bearing? How much load

if the operating temperature of the lubricant is 85°C. The lubricant has a kinematic viscosity = 6 cS,
density = 0.86 g/cm3
5.9 A cylindrical pad hydrostatic bearing having r/R ratio = 0.7 (see Fig. 5.24) supports a load of 200

Determine the required restrictor opening if the pump pressure = 20 kgf/cm2

= 0.005 cm and absolute viscosity of the lubricant = 4 ¥ 10–7 2
.
5.10 Determine the length and diameter of an aerodynamic bearing to support a load of 15 kgf. Given l/d =
0.8, air supply pressure = 4 kgf/cm2, radial gap = 0.01 mm and absolute viscosity of air = 0.02 cP.
5.11 ¥
stock end at s = 0.23 mm/rev. Compliance of the head stock, saddle and tailstock is 0.25, 0.25 and 0.3
microns/kgf, respectively. Given Py = 88 t0.9s0.75 kgf. Calculate the true depth of cut.
5.12 A machine spindle is supported on two anti-friction bearings 20 cm apart and having radial clearances
of 4 and 6 microns, respectively. Determine the radial run out of the spindle nose if its overhang is
8 cm from the bearing having 6 micron clearance.

References
1. Levina, ZM, et al., “Investigation of the stiffness of tapered joints”, Stanki I Instrument, 1973, No. 10.
2. Kuvshinskii, VV, Milling Operation, Mashinostroenie Publishers, Moscow, 1977, p. 127
3. Mamet, OP, Concise Handbook of Machine Tool Design, Mashinostroenie Publishers, Moscow, 1964,
p. 186.
334 Machine Tool Design and Numerical Control

4. Elyashev, A, “Fundamentals of machine tool design”, Notes for UNIDO Course, Moscow, p. 80.
5. Ibid., p. 82.
6. Honrath, K, Werkzengmaschinenspindeln und Deren Langerungev, T.H. Aachen, Ind Anz, 1957,
7. Sokolov, Yu N, “Multiple-wedge hydrodynamic bearings of precision machine tool spindles”, Stanki
I Instrument, 1963, No. 8.
8. Rippel, HC, Cast Bronze Hydrostatic Bearing Design Manual, Cast Bronze Bearing Institute, Inc.,
1963.