Rolling Contact Bearings

Lecture #3

Course Name : DESIGN OF MACHINE ELEMENTS

Course Number: MET 214
Power transmission systems involve rotating shafts. As a consequence, the
techniques used in power transmission systems to support rotating shafts must
be different than the techniques used in construction technology to support
beams (non-rotating members), although the same general approach used to
categorize beam supports does apply to components used to support rotating
shafts.
The components used in a power transmission systems to support rotating
shafts must supply reaction forces and/or moments. The reaction forces and/or
moments are necessary to mount a shaft in a system. In addition, the
components that supply the reaction forces and/or moments to a shaft must
supply the reaction forces and/or moments to the shaft while permitting the
shaft to rotate. Accordingly, the supports used in power transmission systems
are fundamentally different that the supports used in construction technology.
In order to proceed with the design of a power transmission system, the designer must be
familiar with the different types of components that can be used to support rotating shafts. A
variety of components are available to support the rotating shafts. To assist in the
familiarization and/or categorization of the components used to support rotating shafts, a
brief review of beam supports used in construction technology will be presented.

As described in strength of material text books, beam supports can be classified according to
the types of reaction the supports provide. Different types of beam supports are shown in the
figure below.
A basic configuration that can satisfy the requirements to support rotating shafts used in
power transmission systems consists of a pair of rigid supports with a hole provided in
each support as shown below. The supports are separated by an amount that enables the
shaft to be supported by the hole of each support. If the diameter of the hole in each
support is slightly larger than the shaft diameter, the shaft will be able to rotate as the
supports supply reactions forces perpendicular to the axis of rotation of the shaft. The
supports illustrated below provide a radial reaction force capability while permitting the
shaft to rotate.
If a reaction force must be produced along the axis of rotation of the shaft, a stepped
shaft may be used with supports as shown in the side view below.

Figure #2: side view of shaft support configuration with axial reaction capability.

The configuration shown in Figure #2 has the disadvantage that very little margin is
provided for the shaft to expand. By relocating one of the supports and using a collar at
the other support, the shaft may be located laterally by the supports while permitting the
shaft to expand as shown in the figure on the next slide.
Figure #3: side view of shaft support configuration with axial reaction capability and
latitude for shaft expansion.

It should be noted that the thickness of the supports can be used to influence the amount
of reaction moment the support applies to the shaft . The reaction moment generated by
a support can be used to control the angular misalignment existing between the shaft and
the axis of the hole in the support when the shaft is experiencing transverse loading as
shown in the figures in the next slide.
Increasing the thickness of the supports and/or reducing the hole diameter in the supports enables
the supports to control the slope of the shaft at the supports, in essence providing for a moment
reaction.
A major drawback of the configurations illustrated previously is that the shaft experiences
a large amount of friction as the shaft rotates. This is due to the fact that as the shaft is
rotated, the shaft slides on the inside surface of the hole formed in support. To reduce the
amount of friction existing between the shaft and the support, a form of rolling friction
may be implemented. Rolling friction requires the use of rolling elements between the
shaft and the support as shown in the illustrations provided below.
In order to facilitate installation and/or standardization of shaft supports, the components necessary
to enable a shaft to rotate via rolling friction are assembled into self contained units, referred to as
rolling contact bearings as described in the illustrations provided below.

Although cylindrical rolling elements are shown in the figure above, a variety of rolling elements can
be used resulting in a variety of bearing types with different advantages/ disadvantages including
different reaction force and/or moment capabilities. Procedures for selecting bearings will be
discussed in a subsequent lecture.
Here we will discuss different types of rolling contact bearings and the applications in which each is
typically used. Many variations on the designs shown are available. As each is discussed, refer to Table
14 – 1 for a comparison of the performance relative to the others.
Radial loads act toward the center of the bearing along a radius. Such loads are typical of those
created by power transmission elements on shafts such as spur gears, V-belt drives, and chain drives.
Thrust loads are those that act parallel to the axis of the shaft. The axial components of the forces on
helical gears, worms and worm gears, and bevel gears are thrust loads. Also, bearings supporting
shafts with vertical axes are subjected to thrust loads due to the weight of the shaft and the elements
son the shaft as well as from axial operating forces. Misalignment refers to the angular deviation of the
axis of the shaft at the bearing from the true axis of the bearing itself. An excellent rating for
misalignment in Table 14 -1 indicates that the bearing can accommodate up to 4.00 of angular
deviation.
A bearing with a fair rating can withstand up to 0.150 , while a poor rating indicates that rigid shafts
with less than 0.050 of misalignment are required. Manufacturers catalogs should be consulted for
specific data.

Single-Row, Deep-Groove Ball Bearing

Sometimes called Conrad bearings, the single-row, deep-groove ball bearing is what most people think
of when the term ball bearing is used. The inner race is typically pressed on the shaft at the bearing
seat with a slight interference fit to ensure that it rotates with the shaft. The spherical rolling
elements, or balls, roll in a deep groove in both the inner and the outer races. The spacing of the balls
is maintained by retainer or “cages”. While designed primarily for radial load-carrying capacity, the
deep groove allows a fairly sizable thrust load to be carried. The thrust load would be applied to one
side of the inner race by a shoulder on the shaft. The load would pass across the side of the groove,
through the ball, to the opposite side of the outer race, and then to the housing. The radius of the ball
is slightly smaller than the radius of the groove to allow free rolling of the balls. The contact between a
ball and the race is theoretically at a point, but it is actually a small circular area because of the
deformation of the elements. Because the load is carried on a small area, very high local contact
stresses occur. To increase the capacity of a single-row bearing, a bearing with a greater number of
balls, or larger balls operating in larger-diameter races, should be used.

Double-Row, Deep-Groove Ball Bearing

Adding a second row of balls (Figure 14-2) increases the radial load-carrying capacity of the deep-
groove type of bearing compared with the single-row design because more balls share the load. Thus,
a greater load can be carried in the same space, or a given load can be carried in a smaller space. The
greater width of double-row bearings often adversely affects the misalignment capability.
Angular Contact Ball Bearing
One side of each race in an angular contact bearing is higher to allow the accommodation of greater
thrust loads compared with the standard single-row, deep-groove bearing. The sketch in Figure 14-3
shows the preferred angle of the resultant force (radial and thrust loads combined), with commercially
available bearings having angles of 150 to 400 .

Cylindrical Roller Bearing

Replacing the spherical balls with cylindrical balls (Figure 14-4), with corresponding changes in the
design of the races, gives a greater radial load capacity. The pattern of contact between a roller and its
race is theoretically a line, and it becomes a rectangular shape as the members deform under load.
The resulting contact stress levels are lower than for equivalent-sized ball bearings, allowing smaller
bearings to carry a given load or a given size bearing to carry a higher load. Thrust load capacity is
poor because any thrust load would be applied to the side of the rollers, causing rubbing, not true
rolling motion. It is recommended that no thrust load be applied. Roller bearings are often fairly wide,
giving the only fair ability to accommodate angular misalignment.
Needle Bearing
Needle bearing (Figure 14-5) are actually roller bearings, but they have much smaller-diameter rollers,
as you can see by comparing Figure 14-4 and 14-5. A smaller radial space is typically required for
needle bearings to carry a given load than fro any other type of rolling contact bearing. This makes it
easier to design them into many types of equipment and components such as pumps, universal joints,
precision instruments, and house hold appliances. The cam follower shown in Figure 14-5(b) is
another example in which the antifriction operation of needle bearings can be built-in with little radial
space require. As with other roller bearing, thrust and misalignment capabilities are poor.

Spherical Roller Bearing

The spherical roller bearing (Figure 14-6) is one form of self-aligning bearing, so called because there is
actual relative rotation of the outer race relative to the rollers and the inner race when angular
misalignments occur. This gives the excellent rating for misalignment capability while retaining
virtually the same ratings on radial load capacity.
Tapered Roller Bearing
Tapered roller bearings (Figure 14-7) are designed to take substantial thrust loads along with high
radial loads, resulting in excellent ratings on both. They are often used in wheel bearings for vehicles
and mobile equipment and in heavy-duty machinery having inherently high thrust loads. Section 14-12
gives additional information about their application. Figures 8-25, 9-36, 10-1, and 10-2 show tapered
roller bearings applied in gear-type speed reducers.
Thrust bearings:
The bearings discussed so far in this chapter have been designed to carry radial loads or a combination of
radial and thrust loads. Many machine design projects demand a bearing that resists only thrust loads,
and several types of standard thrust bearings are commercially available. The same types of rolling
elements are used: spherical balls, cylindrical rollers, and tapered rollers.
Most thrust bearings can take little or no radial load. Then the design and the selection of such bearings
are dependent only on the magnitude of the thrust load and the design life. The data rating and basic
static load rating are reported in manufacturers’ catalogs in the same way as they are for radial bearings.
Mounted Bearings:
In many types of heavy machines and special machines produced in small quantities, mounted bearings
rather than unmounted bearings are selected. The mounted bearings provide a means of attaching the
bearing unit directly to the frame of the machine with bolts rather than inserting it into a machined
recess in a housing as is required in unmounted bearings.
The figure shows the most common configuration for a mounted bearing: the pillow block. The housing
is made from formed steel, cast iron, or cast steel, with holes or slots provided for attachment during
assembly of the machine, at which time alignment of the bearing unit is adjusted. The bearings
themselves can be of virtually any of the types discuss in the preceding sections; ball, tapered roller, or
spherical roller is preferred. Misalignment capability is an important consideration because of the
conditions of use of such bearing. This capability is provided is provided either in the bearing
construction itself or in the housing.
Because the bearing itself is similar to those already discussed, the selection process is also similar.
Most catalogs provide extensive charts of data listing the load-carrying capacity at specified rated life
values. Internet site 7 includes an example.
Other forms of mounted bearings are shown in the figure. The flange units are designed to be mounted
on the vertical side frames of machines, holding horizontal shafts. Again, several bearing types and sizes
are available. The term take-up unit refers to a bearing mounted in housing, which in turn is mounted in
a frame that allows movement of the bearing with shaft in place. Used on conveyors, chain drives, belt
drives, and similar applications, the take-up unit permits adjustment of the center distance of the drive
components at the times of installation and during operation to accommodate wear or stretch of parts
of the assembly.
Bearing Materials:
The load on a rolling contact bearing is exerted on a small area . The resulting contact stresses are quite
high, regardless of types of bearing. Contact stresses of approximately 300000 psi are not uncommon in
commercially available bearings. To withstand such high stresses, the ball, rollers, and races are made
from a very hard, high-strength steel or ceramic.
The most widely used bearing material is AISI 52100 steel, which has a very high carbon content, 0.95%
to 1.10%, along with 1.30% to 1.60% chromium, 0.25% to 0.45% manganese, 0.20% to 0.35% silicon,
and others allowing elements with low but controlled amounts. Impurities are carefully minimized to
obtain a very clean steel. The material is through hardened to the range of 58-65 on the Rockwell c
scale to give it the ability to resist high contrast stress. Some tool steels, particularly M1 to M50, are
also used. Case hardening by carburizing is used with such steels, particularly as AISI 3310, 4620, and
8620 to achieve the high surface hardness required while retaining a tough core. Carefully control the
case depth is required because the stress occur in subsurface zones. Some more lightly loaded bearings
and those exposed to corrosive environments are AISI 440C stainless steel elements.
Rolling elements and other components can be made from ceramic materials such as silicon
nitrate(si3n4). Although their cost is higher than that of steel, ceramics offer significant advantages as
shown in table. Their light weight, high strength, and high temperature capability make them desirable
for aerospace, engine, military and other demanding applications.
In order to proceed with the design of a power transmission system, the designer
must understand how power is transferred through the system. In order to
develop an understanding of how power is transferred through a system,
consider the following situation.
A weight w is to be lifted off of the floor using the hoist shown below. Develop an
equation that relates the acceleration a of the weight w to the torque T m that a
motor applies to the cylinder of the hoist. In addition, determine the power
supplied by the motor to the hoist when the weight w is being hoisted at a
constant velocity.
Draw a free body diagram of the weight w to develop the equation of motion.
 F  ma
w
FT  w  a
g
w
FT  a  w
g

T 
Ftension
where in cable
w  being lifted by hoist
weight of object
accelerationaofobject being lifted by
hoist
g  acceleration due to gravity
Draw free body diagram of cylinder T  J m

TM  R0 FT  J m

Where T on cylinder

torques acting
R0 
radius of cylinder
torque produced by motor
Tm 
mass moment of 
J minertia of cylinder
angular acceleration
  of cylinder
To develop an expression for the motor torque in terms of the linear acceleration a of the
weight w, solve the previous equation for motor torque and substitute the expression for F T
into the equation for motor torque.
w  a
TM  R0  a  w  J M
g  R0
Where
R0α=a

Rearranging terms results in the following expression for motor torque TM as a function of
the acceleration a of the weight w and the load torque TW generated by the weight w.

J w J w
TM   M  R0 a  R0 w   M  R0 a  TW
 R0 g  R0 g

Where TW = R0 w

The acceleration a of the weight w may be expressed as a desired velocity divided by

the time to reach the desired velocity assuming the weight is hoisted from a rest position
with a constant acceleration a.

A generalized plot of the velocity verse time for the hoisting of the weight is shown on the
next slide. A de-acceleration interval has been provided to stop the hoist.
No values are shown for velocity on the graph in order to direct the focus toward the functional
relationships existing among the two graphs provided below.

A plot of the motor torque versus time for the above defined velocity profile is
shown below.
•  
Several important points are to be observed from the information presented
above.
1) Initially the weight w is at rest. In order to accelerate the weight w to a
desired velocity of V, a motor torque TMthat is larger than the value of the
load torque TW must be applied to the hoist. The amount of torque TM that
must be supplied from the motor in order for the hoist to lift the weight w for
the desired acceleration a can be determined from the equations provided
above.
2) After the weight has been accelerated to the desired velocity, the value of the
motor torque TM is reduced so that the motor torque will have the same value
as the load torque and as a result, the net torque applied to the hoist will be
zero. In order for the motor torque to have the same value as the load
torque, the acceleration a of the weight w must be zero. When the motor
torque is set equal to the load torque, a balance is established between the
load torque and the motor torque and according to Newton’s 1 st law, since
there is no longer any net torque applied to the hoist after the weight has
been hoisted to the desired velocity, the hoist will continue to lift the weight
at the desired velocity until the hoist is de-accelerated. The amount of power
supplied by the motor to the shaft while the weight is being hoisted at a
constant rate is equal to Pm = TM . The amount of power required to lift the
weight w is Pw = V w. Note Pm =Pw.
•3) Since
  the hoist is rotating in the same direction as the torque applied to the hoist by the
motor, the algebraic sign associated with the power the motor is supplying to the shaft is
always positive regardless of the sign convention used to define the direction of the motor
torque and/or the angular velocity of the hoist. This must be the case since under the
steady state conditions described above, the motor torque and the angular velocity of the
hoist will have the same orientation (sign), and as a consequence, the product of motor
torque with angular velocity will always produce a positive product. This must be the case
since the product of two positive terms is positive and the product of two negative terms is
also positive. The positive sign associated with indicates that the motor is supplying power
to the shaft.

4) Since attaching the weight w to the hoist impedes the motion of the hoist, the torque
applied to the hoist by the weight w acts in a direction opposite of the direction of rotation
of the hoist and/or the angular velocity of the hoist during the steady state. When the hoist
is rotating at a constant rate, the direction of the load torque is opposite the direction of
rotation and the algebraic sign associated with the power required to lift the weight w is
negative regardless of the sign convention used to define the direction of rotation of the
hoist and/or the angular velocity of the hoist. This must be the case since the product of
two terms having opposing signs results in a negative product. As a consequence, the sign
associated with the power transferred from the weight to the shaft will be negative
indicating the shaft is providing power to the weight in order to lift the weight.
 Pw prove to be very useful in
•  
5) The algebraic signs associated with and will
determining the power requirements of a motor based upon the power
requirements of the loads that the motor must drive. The algebraic signs
associated with and can be used to create a power balance for use in
determining the amount of power to be transferred through a power
transmission system. Due to the sign convention existing with the power
supplied by the motor and the power utilized by the load, the total power
supplied to the shaft will sum to zero.
6) + =0. The net power transferred to the shaft is zero. Since the algebraic sign
for is negative, the power balance for the shaft can be rewritten as follows: =
.
7) In general, when a component supplies or transfers power to a shaft, the
torque applied to the shaft by the component will always be in the same
direction as the direction of rotation of the shaft. When a component
removes power from a shaft or in other words, the shaft transfers power to
the component, the torque applied to the shaft by the component will be in a
direction opposite to the direction of rotation of the shaft.
8) Since all components affixed to the hoist have the same angular velocity, the
power balance equation can be extended to a torque balance equation.
•   =.

Where

Under steady state conditions, the net torque applied to a shaft must be zero. The
results of extending the power balance equation to a torque balance equation
produces consistency with the laws of statics indicating that the power balance
equations are credible. Relating the power balance concept to a torque balance
concept will prove useful in determining the direction of torque that a power
system component applies to a shaft. As will be evident after discussing torsional
shear stresses existing in a shaft, such considerations are fundamental to
identifying the appropriate diameter of a shaft used in a power transmission
system. When a shaft drives multiple components, the torque each component
applies to the shaft must be combined properly in order to identify the
appropriate diameter for the shaft.
The equation relating motor torque to the acceleration a of the weight w may be
rearranged to emphasize the assumptions that typically accompany the analysis of a
power transmission system in the steady state.

Recalling where mc = Mass of cylinder

J w
TM   wMc = Weight
R0 aof
Rcylinder
0w
R
 0 g 
1 1 wc 2
JM  mc R02  R0
2 2 g
1  aR
TM   wc  w 0  R0 w
2  g
 a 1 wc 
TM  R0 w1  1  
 g 2 w 

TM
The torque required from a motor to operate a hoist
TW contains two components
R0 w 
steady state component = load torque =
 1  aR0
w
 2 c  w  g 
dynamic component involving the acceleration a of the weight w.
 The Steady state component for load torque TWcan be factored out of the expression for T m
to obtain an expression for Tm that depends on the steady state component for load torque
and an adjustment factor, to be referred to as a Service Factor, that depends on the
dynamics associated with a particular application.
 a  1 wc  
TM  R0 w1  1     R0 w[ ServiceFactor ]
 g  2 w 
 a  1 wc  
TM  TW 1  1     TW [ ServiceFactor ]
 g  2 w 
a  1 wc 
ServiceFactor  1  1  2 w 
g  
Note:
Service Factor ≥1
when Service Factor=1 -> a=0
Service Factor>1 -> relevance depends upon a/g ratio
•  
The equation for motor torque can be multiplied on both sides by the angular velocity to
develop an expression for power.
 a  1 wc  
TM  TW 1  1     TW [ ServiceFactor ]
 g  2 w 

 a  1 wc  
TM   TW  1  1     TW [ ServiceFactor ]
 g  2 w 

 a  1 wc  
pM  pw 1  1     pw [ ServiceFactor ]
 g  2 w 

During steady State operations the following equation holds for power.

pM = pw

The above equation can be obtained from a power balance as noted earlier.
 Accordingly, to establish performance requirements for subsystem components so the
mechanical design process can proceed, the power and/or torque propagated through a power
transmission system can be assumed to be the steady state power and/or torque, and
adjustments to be made for dynamical influences can be incorporated by adjusting the steady
state values by a service factor consistent with the expected level of dynamic performance
anticipated for the application.

a  1 wc 
ServiceFactor  1  1  2 w 
g  

Note: Expression for service factor can be written in terms of final velocity and time to
accelerate to final velocity assuming constant acceleration.

 f  1 wc 
ServiceFactor  1  1
t f g  2 w 

 
where final velocity f
tf 
time to accelerate from zero velocity
•  
Recall power can be related to torque and customized to a particular set of units

P
where “steady state power”; (ft-lbs)/sec
Torque ft-lbs T
P  T 
angular velocity rads/sec 

In terms of customized units:

P= (T*n)/ 63,000 where P= horse power

T= torque in lbs
n= rpm

 
 
 1 ft   rev   2rads  1 min  1HP
P   in  lbs      
 12in   min 
  rev  60 sec   550
ft  lbs 

 sec 

2 Tn Tn
P  Tn  
(12)(60)(550) (12)(60)(550) 63,000
2

Using the system configuration associated with a design, it is possible to transfer power requirements from
one component to another and repeatedly use to determine the torque associated with each component.
After the torque loading is establish for a component, the stress level that must be accommodated by the
component can be determined and used to assist in the design of the component.
Accordingly, in essence, there are two different approaches that can be used to analyze the
torque that is being transmitted by a power transmission system. If an assumption is made
that a system is operating in the steady state, the equations governing the motion behavior
of the system are greatly simplified since the acceleration is zero. The results of a steady
state analysis can be adjusted by a service factor whose value depends on the application.
Alternatively, the mass moment of inertia of a system may be determined and a more
complete analysis performed. If the time to accelerate and/or the time to decelerate are
important issues to be analyzed for a system, then a complete analysis using the mass
moment of inertia would be required. However, if time to start and/or stop are not issues
that need to be analyzed, then a system can be analyzed using the steady state assumption
as compensated by a service factor. The method of analysis to be used in the initial stages of
the design process presented herein will be the steady state approach. As will be found, a
considerable amount of data supplied by manufacturers of components, including shaft
couplings, belt drive and/or chain drive systems, rely upon this approach to characterize
their products. Accordingly, the steady state approach will enable the designer to apply the
same basic principles used by component suppliers to assist in the selection of a variety of
components to be used in power transmission systems. After analyzing the basic building
blocks used in power transmission systems, the focus of the effort will be directed toward
conveyor systems and issues involving motor selection predicated on representing the
conveyor components undergoing motion in terms of an equivalent mass moment of
inertia.