Module 6 : Applications of Tribology

Introduction

Tribology is required for equipments/machines of petrochemical, railway, automotive, agriculture,

shipping, power generation, aerospace, military applications, electronic devices and almost all phase
of life where motion under load is encountered. Most common machine elements (rolling or sliding)
in those equipments/machines are :

• Bearings.

• Gears.

• Cams.

• Brakes.

• Seals.

The above tribo-elements have been classified based on their functionality.

Function of bearings is to support load, while the function of gears is to transmit power. Cams are
required to convert rotary motion in reciprocating motion. Brakes are required to stop the motion.
The function of seals is to minimize the leakage.

Fig. 6.1(a): Taper roller bearing.

Fig. 6.1(b): Ball bearings.

 Fig. 6.1(c): Sliding bearing.

At micro level; gears, cams, brakes, and seals are bearings. Bearings means system which supports
load. Gears, cams, brakes, and seals support load. Therefore they may be treated as bearing. When
we start imagining about bearing, very common type bearing “rolling element bearing” appears(Fig.
6.1(a) and 6.1(b). Other kind of bearings are sliding contact(Fig. 6.1(c)) and sliding non-contact
bearings.

Bearing Classifications based on Relative Motion :

• Rolling contact.

• Sliding.

Bearings are employed to separate rotating/sliding elements from relatively stationary components.

Bearing Classifications based on Direction of Load :

• Radial (Journal)

• Thrust.

• Conical/Taper.

Bearing Classifications based on lubrication system :

• Dry.

• Boundary lubricated.

• Elastohydrodynamic.

• Hydrostatic.

• Aerostatic.

• Hydrodynamic.

• Aerodynamic.

• Squeeze Film.

A group of elasto hydrodynamic, hydrostatic, hydrodynamic, aerostatic, aerodynamic and squeeze

film bearings are termed as “Fluid Film Lubricated Bearings”.
 Fig. 6.2: Spur gear.

GEARS: When we imagine gears, a picture of wheel with teeth appears(Fig. 6.2). The tooth profile
of these wheels is made to maintain the constant velocity ratio. The gears are used to amplify the
torque from 1:1 to 40:1 ratio (i.e. involute). Often multistage of gears is used to increase the torque
ratio.

Fig. 6.3: Various gear.

Spur gears, as shown in Fig. 6.2, are the simplest gears. Fig. 6.3 shows some gears.

Cams :

These components transmit rotary motion to reciprocating motion.

 Fig. 6.4: Camshaft and cam-follower mechanism.

Elastohydrodynamic lubrication occurs in cam follower mechanisms (Fig. 6.4). Maximum pressure
in these components may vary from 0.5 to 3 GPa.

Brakes: Basic principle of brakes(Fig. 6.5) depends on friction. So application of tribology is very
essential in brakes.

Fig. 6.5: Different types of brakes[1].

Seals :
There are mainly two types of seals :

1. Contact seals.

2. Non-contact seals.

Non - contact seals are prefered for high speed operation. Friction and wear play very important
role in contact seals.

References :

1. Hamrock B J, Jacobson B O & Schmid S R, Fundamentals of Machine Elements, McGraw-Hill Inc.,

1998.

Module 6 : Applications of Tribology

Rolling Contact Bearing

Rolling Element Bearing :

In rolling element bearings, friction and wear are low, so we require lesser quantity of lubricant.
These bearings work with elastohydrodynamic mechanism. Low friction & moderate lubricant
requirements are two important advantages of rolling bearings. Additionally, these bearings are
mass produced therefore, are cost effective. These bearings are purchased in standard sizes instead
of being customized by individual designer. The bearings are selected from catalogues of bearing.
But very large size bearings, which are specially designed, are often very costly; so selection of
bearing is important for proper life of bearing. One of the important bearing selection criterions is
retainer (Fig. 6.6), which is required to maintain constant angular distance between rolling elements.
But due to retainer, sliding (friction) increases and lesser number of rolling elements (means lesser
load carrying capacity) can be arranged.

Fig. 6.6: Retainer for rolling element bearing.

Bearing may fail due to following factors :

1. Sudden rise in bearing load.

2. Inappropriate selection of bearings.

3. Improper mounting.

4. Manufacturing faults.

Let us consider a case study to understand bearing failure.

Failure Of Four Row Cylindrical Roller Bearing :

• Two large roller bearings – (ID = 865 mm, OD = 1180 mm) failed in a cold rolling mill, each
bearing costed Rs. 35,00,000.
- One bearing failed within 105 hours (installed on 05/01/03 and failed completely on 10/01/03)
and
- Other failed within 300 hours of operation (installed on 05/01/03 and removed on 24/01/03 due
to detection of excessive vibration and metal particles).
- Expected life of bearings was approximately 40,000 operating hours. This means survival rate was
0.5% and 1.0%.

Fig. 6.7: Layout of a 4-high rolling mill stand.

Fig. 6.8 shows different types of failure due to corrosion, excessive loads, holes, fatigue load etc.

Fig. 6.8: Failed outer ring.

To analyze these two bearing failures, finite element study was performed. Outer ring was fixed(zero
degree of freedom) and assumption was made that all rolling element participate to bear the load.

Fig. 6.9: Finite element analysis of bearing.

Fig. 6.10: Factor of safety obtained by FEA.

Factor of safety, as shown in Fig. 6.10, comes out to be 100 at outer ring (which failed in operation),
which clearly indicates whole of outer ring does not participate in load support, but only a fraction
of ring supports the load.

In rolling mills, the load is of constant direction and due to clearence, only a quarter of the outer
race is under load. For this reason, the side faces of the outer races are divided into four zones,
indicated by I to IV. When the bearing is mounted first time, it is usual to position zone I along the
direction of the load. After a period of approximately 1000 operating hours (Ξ 2 months), outer race
is turned by 900. This means rated life of outer ring is four times of rated life of each load zone. In
other words, if fatigue life of outer ring is 40,000 hours, fatigue life of each load zone is 10,000
hours.
 Fig. 6.11: One quarter of outer ring under Fig. 6.12: Outer ring marked with four
load. load zones.

Two important observations from intial study were :

1. Management replaced pin type bearing with brass cage bearing, which was cheaper.

2. Load capacity as function of number of rolling elements.

Fig. 6.13(a): Brass cage.

Fig. 6.13(b): Pin cage.

It is interesting to note that brass cage takes more space and accommodates only 34 roller(per row)
compared to 38 roller(per row) in pin type bearing.
 Fig. 6.14: Load capacities of brass and pin type cage bearings.

Due to lesser number of rolling elements brass cage bearing load capacities are lesser than pin type
bearings. Following equation indicates lesser life of solid brass cage bearing compared to pin type
cage bearings.

Detailed survey of failed bearing indicated placement of hole along on the line of maximum load.
Four holes of 3/8” 10 UNC 3B of 45mm depth were drilled and tapped to facilitate the handling of
outer race, Fig. 6.15.

Fig. 6.15: Holes to handle the outer ring.

It is intreasting to note that stress level increases significantly if drilled hole directly comes under
load as shown in following figure. Following figure shows stress results of outer ring. On shear failure
of outer ring near lifting holes, tensile stress was experienced by remaining roller-outer-ring
interfaces. Under tensile stress, radial fracture of outer race occurs.
 Fig. 6.16: Tensile stress in outer ring.

After understanding the drawback of existing arrangement, new arrangement for outer ring was
suggested. With new arrangement as shown in following figure and no complaint was received.

Fig. 6.17

Bearing Terminology :

In previous paragraphs, we learnt bearing failure. To understand rolling bearing fully, we must under
terminology of bearing. Rolling element bearings, as shown in Fig. 6.18 consist of four main
components: inner ring, outer ring, balls, and the cage. Inner and outer rings are hollow cylinders.
Their thickness is decided based on stiffness criteria. Rolling elements, which may be balls, rollers,
needle rollers, etc. are solids with very good surface finish.

The purpose of cage is to maintain the rolling elements (i.e. balls or rollers) a fixed distance apart,
as they roll around the ring tracks. Some cages are made from steel pressing, in two halves which
are finally riveted together(Fig. 6.19), and are shaped to freely locate on the balls.
 Fig. 6.18: Bearing terminology.

Fig. 6.19: Splitted cage with balls.

For correct function, it is necessary for the ball to make point contact only with the rolling tracks in
the inner and outer rings and to achieve this, the track radius is approximately 4% greater than ball
radius. Cylindrical roller bearings, Fig. 6.20, are mainly designed to bear radial load.

Two different arrangements (arrangements A & B) are shown in Fig. 6.20. The basic difference
between two arrangements is the sketches of inner ring. In second arrangement, the inner ring has
a groove at outer surface and rollers are guided by this groove. With such arrangements, roller
bearings are able to support a part of thrust load.
 Fig. 6.20: Sectional view of cylindrical roller bearings.

Fig. 6.21 illustrates a fully complemented needle roller bearings. The word needle itself indicates
the large value of length to diameter ratio. The ends of these needles have smaller diameter, which
is required to locate the needle in outer/inner ring. Major disadvantage of this arrangement is higher
coefficient of friction because of relatively larger value of sliding to rolling motion. This arrangement
is advantageous from space point and load capacity point of view.

Fig. 6.21: Fully complemented needle bearing.

Other than straight roller and needle roller bearings, taper roller(Fig. 6.22), spherical roller(Fig.
6.24) and self aligning roller(Fig. 6.23) bearings are common. Taper roller bearing can sustain axial
+ radial loads with appropriate positioning of shaft. Spherical roller and self aligning roller bearings
can handle misalignment.
 Fig. 6.22: Sectional view of taper roller bearing.

Fig. 6.23: Self aligning roller bearing.

Fig. 6.24: Single-row spherical roller bearing.

Designation–International Organization for Standaradization :

Each rolling bearing is designed by a code that clearly indicates construction, dimensions, tolerances
and bearing clearance. Bearings codes comprising only the basic code without prefixes and suffixes
indicate normal bearings. Deviations from the normal construction are indicated by prefixes or
suffixes.

Fig. 6.25: Bearing designation as per DIN 623 standard.

Basic code consisits of "Series Code". Series code contains number(i.e, 0,1,2.....) or combination of
letter(i.e, BK, HK, .....) & numbers. Following table illustrates first part of "Series Code", which
indicates the type of bearing.
 Table 6.1: Series code.

A second and sometimes a third letter are used to identify the configuration of the flanges, e.g. NJ,
NU, NUP; double or multi-row cylindrical roller bearing designations always start with NN. It is
interesting to note that for same bore diameter, load capacity of rolling bearings can be increased
by increasing diameter of rolling elements(Fig. 6.26) and outer ring(Fig. 6.27).

Fig. 6.26: Diameter series.

 Fig. 6.27: Extra light to heavy series of rolling bearing[1].

• After defining bearing type, the width (B or T) or height (H or T) series and the second the
Diameter Series (D); is defined in "Series Code".

• Bearings "Bore diameter code" is defined by two numbers(i.e, 05, 06, 15, ...), which when
multiplied by 5, the bore diameter in millimetres is obtained.

For bearings, which have a bore diameter smaller than 10 mm and equal to or greater than 500
mm, the bore diameter is generally given in millimetres, e.g. 618/8 (d = 8 mm) or 511/530 (d =
530 mm). This is also true for bearings, which have bore diameters of 19, 28 and 33 mm, e.g.
322/28 (d = 28 mm). Bearings with bore diameters of 10, 12, 15 and 17 mm have the following
size identifications :

00 = 10 mm

01 = 12 mm

02 = 15 mm

03 = 17 mm

Series designations often include a suffix A, B, C, D or E or a combination of these letters e.g. CA.
These are used to identify differences in internal design, e.g. contact angle.

Contact angle α = 400 is designated by suffix B. Similarly, contact angles of 25 0 and 300 are
designated with suffixes AC and A respectively.

Bearing designation can be explained by considering deep groove bearing of 10 mm bore diameter.
Table 6.2 lists outer diameter (D), bearing length (B), dynamic load capacity (C), static laod capacity
(Co), Fatigue load capacity (Pu), rated speed for grease(as lubricant) and oil(as lubricant) and
designation.

If applied load is lesser than Pu, then bearing shall have infinite life. Limiting speed with liquid
lubricant is higher compared to bearings with grease lubricant, as liquid lubricant provides better
cooling and effective operating temperature shall be lesser. Bearing column of table 6.2 lists bearing
series(i.e, 6000, 6200) with suffix(i.e, 2RSH, 2Z).

Table 6.2: Deep groove ball bearing(bore dia = 10 mm)[2].

Suffix :

To explain suffix example of deep groove ball bearing (DGBB) has been considered. These bearings
possess deep grooves on both the rings(Fig. 6.28). These widely used bearings can support high
radial forces as well as some axial forces.

Fig. 6.28: Bearing dimensions.

 Fig. 6.29: Deep groove ball bearings, single row.

Fig. 6.29 illustrates single row deep groove ball bearing with no suffix(61804), with suffix for
shield(61804-27) and with suffix for seal(61804-2RS1). There are some advantages and
disadvantages of each configuration.

• Shield : Profiles sheet steel discs pressed into the grooves of outer ring and forming gap-type
seals with the inner-ring shoulders.

Advantages :
1. Low friction.
2. Recommended for high speed.

Disadvantages :
1. Possibility of water ingress in most environment.
2. To some extent grease may leak-out and dust may enter in dusty enviornment.

• Seals : It often made of elastic rubber. Bearings sealed on both sides are grease filled and in
normal working conditions, the grease filling lasts the entire service life of the bearings.

Advantages :
1. Very good grease retainability.
2. Does not allow dust to enter.

Disadvantages :
1. High pressure.
2. Limit on rotational speed due to temperature.

Other than deep groove ball bearings, angular contact ball bearing and cylindrical bearings are
widely used.

Angular Contact Ball Bearing (ACBB) :

Raceways are so arranged that forces are transmitted from one raceway to other under certain
contact angle-angle between line of action of the force & radial plane. Due to CA, ACBB are better
suited to sustain high axial loads than DGBB.

Cylindrical roller bearings :

 Fig. 6.30: Various configuration of cylindrical roller bearings[2].

Cylindrical roller bearings(Fig. 6.30) can accommodate heavy radial loads and can operate at high
speeds. The full complement bearings (without cage) can support particularly heavy radial loads,
but because of the special kinematic conditions in these bearings, they cannot be operated at such
high speeds as the caged bearings.

Classifications of rolling element bearings based on load.

Since most types of radial bearings can carry some thrust, there is no sharp distinction between
them; however, bearings having a contact angle α = 450 or smaller (α < 45 degrees) are considered
radial bearings and their ratings are given as radial load. Some of these bearings are shown in Fig.
6.31.

Bearings with a contact angle α greater than 45 0 (α > 450) are considered thrust bearings and are
rated axial bearings. Some of these bearings are shown in Fig. 6.32.

Fig. 6.31: Radial rolling bearings.

 Fig. 6.32: Thrust rolling bearings.

• Axial load capacity increases with increase in α.

• Radial load capacity decreases with increase in α.

Selection of Rolling Element Bearing :

Each type of bearing displays characteristic properties as listed in Table 6.3. For example, deep
groove ball bearings can accommodate moderate radial loads as well as axial loads. They have low
friction and can be produced with high precision in relatively samll sizes. Therefore, they are
preferred for small and medium-sized electric motors. Spherical roller bearings can carry very heavy
loads and are self-aligning. These properties make them popular in heavy engineering, where loads
are heavy and deformations and misalignments are produced by the loads. The information given
here should serve to indicate which are the most important of the following points to be considered
when selecting bearing type and thus facilitate an appropriate choice.

• Cylindrical & Needle roller – pure radial load.

• Thrust (cylindrical roller, ball), four point angular contact ball bearings – pure axial load.

• Taper roller, spherical roller, angular contact ball bearings – combined Load.

• Cylindrical roller, angular contact ball bearing– high speed.

• Deep groove, angular contact, and cylindrical roller bearing – high running accuracy.
 Table 6.3: Characteristic properties of rolling bearings.

The table 6.3 provides a relatively superficial classification of bearing types and facilitates an
appropriate choice. It should also be remembered that the total cost of a bearing arrangement and
inventory considerations also influence the final choice. Other important criteria to be observed
when designing a bearing arrangement including load carrying capacity and life, friction, permissible
speeds, bearing internal clearance or preload, lubrication and sealing are dealt with in depth in
separate sections.

Equivalent load :

Dynamic load capacity(as listed in table 6.2) is compared with applied radial load(axial load for
thrust rolling bearings) to estimate the bearing fatigue life. Often radial load is associated with axial
load and in those situations, concept of equivalent load is used.

• Generally vector sum(P = √(Fr2 + Fa2)) does not provide equivalent load. Amount of damage done
by thrust load is different than damage done by radial load, so it is important to find equivalent
radial load that cause same damage as combo of thrust & radial loads. Equivalent load is given by

P = V X Fr + Y Fa

Where

V - Rotation factor = 1 if inner ring rotates,

V = 1.2 if outer ring rotates,

X - radial factor,

Fr - applied radial load,

Y - thrust factor,

Fa - applied thrust load.

Fig. 6.33 shows that below certain limit, the effect of axial force should not be counted. In other
words, equivalent load P can be given by
Where e is a dimensionless ratio, indicating axial load lower than a certain limit does not affect total
load as shown in following Fig. 6.33. Value of e depends on arrangement & static load capacity (C 0)
of bearing as listed in table 6.4

Fig. 6.33: Effect of axial force on effective load.

 Table 6.4: X & Y factors for various bearings.

X & Y factors depend upon bearing geometry, number of balls and size of balls. These factors are
determined from tables and equivalent load is calculated based on those values. These tables
contain "C0", which is static load rating.
Static Load Rating C0 is defined as radial load causing permanent deflection greater than 0.01%
of ball dia.

In addition to C0, bearings are defined with dynamic load rating C, which is radial load (thrust load
for thrust bearings) that a group of identical bearings with stationary outer rings can theoretically
endure one million revolutions of inner ring.

Dynamic load rating is used to estimate bearing life using Lundberg Palmgren Approach.

Lundberg Palmgren Approach :

In ideal case, bearings fail by surface-fatigue(Fig. 6.34), which means bearing life can be estimated.

Dynamic load rating (catalogue C reading) is the load which 90% (reliability = 0.9) of a group of
identical bearings will sustain for minimum of 106 cycles.
 Fig. 6.34: Bearing failure due to surface pitting.

To estimate bearing life at given bearing load(i.e, P1, P2,....), following equations are used.

....Eq.(6.1)

• Example : Radial load = 2 224 N, Speed = 1500 rpm, Desired life = 8 hours/day, 5 day/weeks
for 5 years, Shock factor = 1.5. For shaft diameter of 25 mm, find suitable rolling bearing.

In the present example, shock factor(load amplification factor) is used. This factor depends on
machine in which bearings are employed.

• Smooth operation --> Motor, Machine tools (1.2), air-conditioners (1.2)

• Normal operation --> Automotive vehicle, paper making machine, elevator, crane. 1.2 – 1.5

• Operation with vibration --> Crusher, construction equipment, farming equipment --> 1.5 – 3.0

To select rolling bearings, first step is to estimate the required C.

C = Radial load * shock factor * (total revolutions)1/a.

• On substituting given values C ≥ 2224 * 1.5 * (10400 * 1500 * 60/10 6)1/a.

where a = 3 for ball bearings and a = 10/3 for roller bearings.

•C ≥ 32, 633 N for ball bearings.

•C ≥ 25, 978 N for roller bearings.

We can select bearing based on required "C" from catalogue.

 Fig. 6.35: Single row cylindrical roller bearings N205ECP.

Fig. 6.35 shows data of N205ECP roller bearing with basic dynamic rating of 28600 N ,which is
greater than the calculated value. So bearing runs desired life without failure. Another thing to be
noted from this figure is fatigue limit which is 3350N from catalogue. This value is smaller than the
radial load applied on the bearing, so bearing does not fail for proper lubrication.

Fig. 6.36: Deep groove ball barings, single row (6405).

We are able to find ball bearing(6405) which has dynamic load rating greater than the required
32,633 N as shown in Fig. 6.36.

But fatigue limit is lesser the applied radial load, so this bearing(6405) is going to fail by fatigue
load. So the best choice is roller bearing in terms of its dimensions and load aspects. Roller
bearing`s(N205ECP) outer diameter is 52 mm and length of bearing is 15 mm, which are lesser
than the corresponding dimension of ball bearing.

Example : Assume radial and axial loads on a bearing to be 7500N and 4500N respectively. Rotating
shaft dia = 70 mm. Select suitable single row deep groove ball bearing. To choose ball bearing,
which shall fit on shaft of 70 mm, we shall see bearing catalogue of deep groove ball bearings with
bearing series ending with 14, as given in Table 6.5. As per given data, shield or seal may not be
required, therefore, we can remove bearings, having shields or seals from selection Table 6.5. After
this intial screening 61814, 61914, 16014, 6014, 6214 and 6314 may be tested for static load
capacity.
 Table 6.5: Deep groove ball bearings of 70 mm bore size.

Using above value, calculate the ratio then decide the X, Y factors. Then calculate the equivalent
load (P).

Table 6.6: X & Y factors for deep groove bearing.

Based on axial load (Fa), radial load (Fr), and static capacity (C0),, factors X & Y can be evaluated
using Table 6.6. In present example (Fa/Fr) = 0.6. (Fa/C0) for bearing 6014 is 4500/31000. Using
table 6.6, X = 0.56 and Y = 1.37. Based on these data equivalent load = 10365. Similarly (F a/C0)
for 6314 bearing is 4500/68000. Using these data and table 6.6, X = 0.56 & Y = 1.65. Equivalent
load P = 11625 N.

Example : Assume radial and axial loads on a bearing are 7500N and 4500N respectively, shaft dia
= 70 mm and select a deep groove ball bearing. Consider shaft rotates at 1000 rpm and expected
bearing life = 3000 hours. In present example equivalent loads calculated in previous example may
be used to find bearing life.

Life consideration :

(C)a 106 = P1a L1 = P2a L2 = P3 a L 3

a = 3 for ball.

Table 6.7: Bearing data.

Bearing life in hours = (C/P)3 [1000,000/(60*speed)]

For present example, speed = 1000 rpm => Bearing life in hours = (C/P)3 (50/3)

Using table 6.7 and above equation; bearing life of 6014 shall be 937 hours, while life of 6314 shall
be 14,500 hours. This means bearing 6314 shall be selected.

Angular contact ball bearings :

In previous example we selected deep groove ball bearing for a combined load application. We may
have selected angular contact bearings. Angular contact ball bearings are often used in adjustable
bearing arrangement. Two popular arrangements are :

O arrangement.

X arrangement.

With O arrangement, the pressure cone apexes formed by the contact lines point outward, while
with X arrangement they point inward as shown in Fig. 6.37. With these two arrangements bearing
can take up axial load in both the directions. However, sometime larger load capacity is required
only in one direction, where tandem bearing are used.
 Fig. 6.37: Arrangements of angular contact bearings.

When arranged in tandem, the load lines are parallel and the radial and axial loads are equally
shared by the bearings. The bearing set can only take up axial loads acting in one direction. If axial
loads act in the opposite direction, as well as combined loads, these are taken up by a third bearing
which is adjusted against the tandem pair.

The major problem with these bearings is misalignment, which leads to increase ball loads and cage
stresses as well as a reduction in bearing life. Any misalignment of the bearing rings will also lead
to an appreciable increase in running noise. Special attention should also be paid to back-to-back
and face-to-face arrangements when the axial load acts predominantly in one direction. The
unfavourable rolling conditions for the balls of the unloaded bearing under such conditions can lead
to noise, rupture of the lubricant film and increased stressing of the cage. X and Y factors for angular
contact bearings are listed in Table 6.8. These factors can be used to estimate equivalent load and
bearing life as explained in next example.

Table 6.8: X & Y factors for angular contact bearings.

Example : A radial load of 3000N combined with thrust load of 2500N is to be carried on a 6214
ball bearing for 70 mm dia rotating shaft at 1000 rpm. Determine equivalent radial load to be used
for calculating fatigue life. Compare life of 6214 bearing with that for a 7214 (nominal contact angle
300).

Step 1: C0 for 6214 is 45kN and 7214 is 60 kN. C for 6214 is 63.7 kN and 7214 is 71.5 kN.

Step 2: Fa/VFr > e.

Step 3: Radial load for 6214 bearing is 5955N & for 7214 bearing radial load is 3070. Using these
data and following equation bearing life can be calculated.

Bearing life in hours = (C/P)3(1000,000/60 speed).

Step 4: Life for 6214 will be 7192 hours and 7214, life = 124,420 hours.

This example clearly demonstrates better life of angular contact bearing compared to deep groove
ball bearing.

Equivalent load under variable loading :

Often bearings operate at various rotational speeds. For example, bearing operates at 1000 rpm,
applied load of 500 N for 100 hours then bearing operates at 1200 rpm, 250 N for 250 hours. In
such situation it is advisable to find an equivalent load, using :
 as L = L1 + L2 + L3 + ......

Example : A ball bearing is run at to four piecewise load and speed conditions, as given in table 6.9.
Find its equivalent load.

Table 6.9: Data related to variable bearing load.

P = (p13 f1 + p23 f2 + p33 f3 + p43 f4)1/3,

P = (8.6636 X 109)1/3 = 2054 N.

Friction in Rolling Element Bearings

Bearing selection based on load and expected life was explained in previous subheadings. Sometime,
friction plays role in bearing selection. A rough estimation of frictional coefficients(f) is :

• Self-aligning ball bearings, f = 0.0010

• Cylindrical roller bearing, f = 0.0011

• Thrust ball bearings, f = 0.0013

• Single-row deep-groove ball bearings, f = 0.0015

• Tapered and spherical roller bearings, f = 0.0018

• Needle bearings, f = 0.0045

Coefficient of friction is represented by symbol f. The value of f is a system property and will change
with kind of lubricant, applied load, and mounting. The values quoted are idle values. The lowest
coefficient of friction is 0.001 and highest coefficient of friction is 0.0045. From this comparison it
can be concluded that self aligning bearings are preferable choice.

Why coefficient of friction in needle roller bearings is higher than other bearings ?

• Considerably greater length than dia.

• Rollers cannot be manufactured as accurately as other cylindrical rollers.

• Rollers cannot be guided well.

• Rubbing action against each other.

 Fig. 6.38: Needle roller bearings.

Anti-friction bearings is a common terminology used by a number of authors for rolling element
bearings. This means bearings which go against friction and almost negligible friction. But we are
trying dig out all possible sources of friction in these anti-friction bearings. One common cause of
friction in all rolling element bearing is hysteresis losses. Such losses occur due to loading and
unloading of rolling element on rings.

The lowest value of coefficient of friction due to hysteresis happen in chrome steel, therefore,
chrome steel is used as one of materials for rolling elements. It is necessary to understand ball/roller
are subjected from zero load (as shown in green color in Fig. 6.39(b)) to maximum (as shown in
red color in Fig. 6.39(b)) load. Due to such load variation, elastic deformation of rolling elements
change from zero to maximum as shown in Fig. 6.39(a).

Fig. 6.39: Loading and unloading of rolling elements.

Fig. 6.40: Hysteresis load.

Other than elastic hysteresis, relative speed (sliding between rolling elements and cage, sliding
between rolling elements and rings, lubricant shearing), seals and misalignment cause friction.

Fig. 6.41: Bearing seal.

• Total friction = Frictionload + Frictionspeed + Frictionseal

The total frictional moment of a bearing is obtained by adding the frictional moment M 0, which is
independent of the bearing load, and the load-dependent frictional moment M1. For sealed bearings
and axially loaded cylindrical roller bearings, additional components of the frictional moment must
be taken into account. Sliding friction occurs at guiding surfaces of rolling elements in the cage, at
the roller faces and the raceway lips, and between adjacent rolling elements in cage-less bearing.
Internal friction & churning action of lubricant cause friction.

Load dependent friction moment :

Under ideal conditions(P » 0.1 C, good lubrication, mean speed range ~0.3 to 0.7 permissible speed,
without seal) the frictional moment can be calculated from the following equation.

MP = f P (d/2)

where P = √(Fr2 + Fa2)

MP = Frictional moment due to external (+pre-) load, N.mm.

f = Coefficient of friction.

P = Resultant load, N.

d = Bore dia, mm.

Generally, f is a function of load. Data given in table are applicable for P = 0.1 C.
 Table 6.10: Friction coefficient.

Lubricant & Speed Dependent Friction Moment :

where;

ν = Operating viscosity of oil, mm2/s.

N = Rotational speed, rpm.

ML = Moment, N.mm.

dm = mean diameter, mm.

Table 6.11: Lubricant factor f . L

Seal Dependent Friction Moment :

In this component of friction moment, there is fixed and variable parts of friction. Where the bearing
is only sealed at one side, Ms/2 should be used instead of Ms. Where bearings are fitted with rubbing
seals, the frictional losses arising from the seal may exceed those arising from the bearing itself.

Table 6.12: Friction factors for seals.

• Total friction moment M = MP + ML + MS

Example : Estimate friction moment of 6214-2RS1 bearing running at 6,000 rpm under 5000 N
radial load when jet lubricated by synthetic ester jet engine oil having a viscosity of 6 mm 2/s (cSt)
at operating temperature.

Ans : 468.5 N.mm.

Bearing Temperature :

• Dependence on total friction moment, speed, extraneous heat source and heat dissipation
capability.
• Ideally, operating temprature needs to be atmospheric temp. because bearing friction is small.

• Excluding extraneous heat, steady state temperature of a bearing can be calculated by balancing
heat flow generated by bearing and heat flow dissipated into environment.

=> Heat dissipation rate QR = Rate of generation QL.

Heat flow dissipated to environment is calculated from the difference between the bearing
temperature and ambient temperature, size of heat transfer surfaces and heat flow density, which
depends on cooling conditions.

Example : Estimate bearing operating temperature of 6214-2RS1 bearing running at 6,000 rpm
under 5000 N radial load when jet lubricated by synthetic ester jet engine oil having a viscosity of
6 mm2/s (cSt) at operating temperature. Assume ambient temp = 30 0C and natural cooling of
bearing, OD = 125, B = 24.

Substituting values
In idle conditions, bearing temperature is well within limits, but due to installation faults, bearing
temperature increases significantly.

Bearing Clearance

In idle condition, friction loss and temperature rise are well within permissible range. But bearing
performance strongly depends on the bearing mounting, which affects the clearance between
elements of rolling bearing.

Fig. 6.42: Diametral clearence.

Clearence, before bearing mounting, is shown in Fig. 6.42 and tabulated in table 6.13. Bearing
clearences are often classified as normal(standard) clearence, C2 clearence, C3 clearence and
C4 clearence. C2 clearence, lesser than normal clearence, is recommended for very precise position
of shaft, such as spindle bearings. C3 and C4 clearences greater than standard clearences and are
recommended for high temperature applications. Generally reduction in bearing clearence occurs
after mounting.
 Table 6.13: Beraing clearences.

• Bearings are mounted on shaft and in housing with transition to interference fit.

• If interference fits exceed the internal radial clearance, the rolling elements become preloaded.

Fig. 6.43: Reduction in bearing clearence.

Operating/Running clearance in rolling bearings depends on the manufacturing processes(IT grade)
and bearing (light/medium) fit.

Manufacturing Processes :

Fig. 6.44: Tolerance range for various manufacturing processes.

IT Grade refers to the International Tolerance Grade of an industrial process. This grade identifies
what tolerances a given process can produce for a given dimension as listed in table 6.14.
 Table 6.14: IT grades and corresponding tolerances for various dimensions.

Bearing Fits :

Recommended bearing fits are :

• Light duty speed :

- H6 k5 fine.
- H7 k6 normal.
- H8 k7 coarse.

• Medium duty speed :

- H6 m5 fine.
- H7 m6 normal.
- H8 m7 coarse.

It is important to note that tolerances belong to one component but fit is referred to assembly of
components. For example H6k5 refers to role tolerance(bearing bore) and shaft tolerances. Here 6
& 5 are referred to IT6 and IT5 respectively. Letters associated to these numbers are in Fig. 6.44 &
Fig. 6.45.
 Fig. 6.44: Hole and shaft tolerance(μm).

Fig. 6.45 indicates smaller size of shaft (-ve sign) and larger size of bearing bore (+ve sign) provides
clearance fit, while larger shaft size (+ve sign) and smaller (-ve sign) make interference fit.

Fig. 6.45: Bearing fit.

Preload :

Due to negative clearance between rolling elements and bearing rings, bearing gets preloaded.
Preloading of bearings is required to prevent/ suppress shaft runout/vibration/noise. Preloading
improves the running accuracy and increases the bearing stiffness. But preloading increases the
friction loss and so running temperature.With increase in temperature, viscosity of grease/oil
decreases, which further increases the metallic friction. Therefore, preloaded bearings must be
designed with care.

Misalignment: Performance of rolling bearing is very sensitive towards misalignment. High

preloading on one side of the bearing happens due to misalignment.

• 0.10 misalignment results 52 μm preloading for 30 mm length.

• 0.010 misalignment results 5 μm preloading.

To understand the effect of preloading on bearing load, we can use concept of elastic
deformation(Fig. 6.46) and following formula.

Elastic Deformation suggested by Timoshenko & Goodier: :

Fig. 6.46: Elastic deformation during bearing mounting.

Example : Estimate bearing operating temperature of 6214-2RS1 bearing running at 6,000 rpm
under 5000 N radial load when jet lubricated by synthetic ester jet engine oil having a viscosity of
6 mm2/s (cSt) at operating temperature. Assume ambient temp = 300C, ball dia=12 mm, Deflection
due to improper mounting = 5 μm, and forced cooling of bearing.

Intial load on bearing due to preloading can be estimated using

E for bearing steel is 207 GPa and ν = 0.28. As per given data r = 6 mm & δ 1 = 0.005 mm. This
means Fintial = 42.337 kN. If we consider only applied force MP = μP(bore dia./2) => MP = 262.5
N.mm. On considering deflection MP = 2485 N.mm. This means significant increase in friction
moment occurs by 5 μm elastic deformation of rolling elements.
Ms = 105 N.mm.

Total moment = 2691 N.mm.

As we calculated in previous examples

=> t = 30 + 2.3.

This temperature rise (2.30C) occurs under forced cooling envioronment. If we assume normal
environment

The increase in temperature by 5.80C is acceptable, which means bearings are designed with some
capacity to handle some misalignment & interference fit.

Bearing Lubrication

To reduce friction, lubrication is needed. In addition lubrication is required to avoid rapid wear and
early failure. Lubricant minimizes metal to metal contact between surfaces in rolling contact and
sliding motion, thereby minimize friction and wear commonly used lubricants are :

• Greases :

- Used for most of rolling elements.

- Life long lubrication < 1.5 * 106 mm/min.

• Oils :

- An aid to heat dissipation.

• Solid Lubricants :

- Compensation for surface roughness.

- Life long lubrication, if < 1500 mm/min.

Grease has the advantage over oil that it is more easily retained in the bearing arrangement,
particularly where shafts are inclined or vertical, and it also contributes to sealing the arrangement
against contaminants, moisture and water. An excess of lubricant will cause the operating
temperature to rise rapidly, particularly when the bearings are running at high speeds. As a general
rule, the free space in the housing should be partly filled (30 to 50 %) with grease. Where bearings
are to operate at very low speeds and must be well protected against corrosion, it is advisable to
completely fill both bearings and housings with grease. Greases thickened with polyuria can soften
and harden reversibly depending on the shear rate in the application, i.e. they are relatively stiff at
low speeds and soft or semi-fluid above a given speed. For normal speeds and loads, use medium
soft grease, since it readily covers all the working parts and does not ridge up on the sides of tracks.
For small bearings functioning at high speeds under light loads, use a soft grease.

• NLGI grades nos. 1, 2, 3, or 4 are recommended for lubrication. Some of NLGI grades is given in
Table 6.15.

• NLGI 2 would normally be the first choice.

Table 6.15: NLGI grades.

• For better sealing and vertical shaft arrangements. Grease no. 3 or 4 is recommended.

• Shock load or vibration tends to cause grease to slump into moving components. A Harder grease
reduces problem.

• Centralized grease supply uses a softer grade.

Selection of lubricating oil :

• Bearing life may be extended by selecting an oil whose viscosity ν at the operating temperature
is somewhat higher than ν1. The values of ν1 can be obtained using Fig. 6.47.

• If the viscosity ratio, k = ν/ν1, is less than 1, an oil containing EP additives is recommended.
 Fig. 6.47: Required kinematic viscosity at operating temperature.

The viscosity of an oil is temperature dependent, becoming lower as the temperature rises. The
viscosity/temperature relationship of an oil is characterized by the viscosity index VI. For rolling
bearing lubrication, oils having a high viscosity index (little change with temperature) of at least 85
are recommended. Certain bearing types, e.g. spherical roller bearings, taper roller bearings, and
spherical roller thrust bearings, normally have a higher operating temperature than other bearing
types, e.g. deep groove ball bearings and cylindrical roller bearings, under comparable operating
conditions, therefore, high VI oil must be selected.

Example :

A bearing having a bore diameter d = 340 mm and outside diameter D = 420 mm is required to
operate at a speed n = 5000 r/min.

Since dm = 0.5 (d + D), dm = 380 mm. From Fig. 6.47, the minimum kinematic viscosity ν1 required
to give adequate lubrication at the operating temperature is 13 mm 2/s. From Fig. 6.48, assuming
that the operating temperature of the bearing is 700C, an oil having a viscosity ν at the reference
temperature of 400C of at least 39 mm2/s will be required.
 Fig. 6.48: Temperature cum viscosity.

Elastohydrodynamic lubrication of rolling contact :

In order to understand the role of a lubricant in rolling bearings, it is necessary to explore the
elastohydrodynamic lubrication. Under elastic deformation, effective film thickness increases as
shown in Fig. 6.49. A detailed analysis of EHL is out of scope, therefore widely used curve fit
equations will be discussed.

Fig. 6.49: Elastohydrodynamic lubrication in rolling bearings[3].

It is interesting to note (Fig. 6.50) that sesitivity of film thickness toward increase in load is very
low.
 Fig. 6.50: Effect of load on film thickness.

As lubricant is drawn into the leading edge of the film by the roller surface, the viscosity of the oil
increases rapidly and the lubricant film pressure rises to match the elastic Hertz contact pressure.
After passing the center of contact area, pressure of the lubricant film falls until it reaches the outlet
zone. Here the decreasing contact pressure relieves the compression of the roller surface, the gap
narrows to h0, and a distinctive pressure spike is reached at film exit as shown in Fig. 6.51. As speed
decreases or load increases, this pressure spike moves further towards the exit and eventually
disappears for very heavily loaded contacts.

Min. film thickness (h0) is often empirically represented in terms of material, speed and load
parameters.

Line Contact :

Curve fit equation for min film thickness in cylindrical bearings is given by;

Fig. 6.51: : Non-dimensional distance from center

of contact.

Point Contact:

Curve fit equation for min film thickness in spherical/ball bearings is given by;
Example : 6210 bearing has balls of 12 mm diameter and maximum load on each ball is 1500 N.
Young’s modulus of balls is 202 GPa Poisson`s ratio is 0.28 and pressure viscosity coefficient is 1.8
* 10-8 m2/N. Dynamic viscosity is 11.3 mPa.s. Radius of inner ring in direction of rolling is 28 mm
and its groove radius is 6.5 mm. Speed = 900 and estimate minimum film thickness.
To sustain EHL, roughness of rolling surfaces must be lesser than 0.125 μm.

References : :

1. Edwards K S & McKee R B, Fundamentals of Mechanical Component Design, McGraw-Hill Inc.,

1991.

2. http://www.skf.com/group/products/bearings-units-housings/index.htmlaccessed on 20th
February 2013.

3. http://www.maxxtorque.com/dieselcommunity/winter-2008/intro-to-lubrication?start=2
accesses on 20th February 2013.

Module 6 : Application Of Tribology

Tribology of gears

In this topic, we shall discuss about the geometry of gears and then consider three different aspects
of gearing namely elastohydrodynamic lubrication, tooth contact phenomena, and wear of gears.

Gears are machine elements, which are required to transmit power between shafts rotating at
different rotational speeds. By adding teeth of the proper shape on disk, power can be transmitted
without slip at uniform rate. These types of geometrics are known as external gears. Internal gears
are generally more efficient since the sliding velocity along the profile is lower than equivalent
external gears. It operates at closer center distance with its mating pinion than external gears of
the same size, which often permits a more compact design. The internal gears eliminates the use
of an idler gear, where it is necessary to have two parallel shafts rotate in the same direction. In
manufacturing point of view also, external gears are simpler than internal gears.
 Fig. 6.52: Internal & External gearing

Often gears are treated as pitch cylinder, which roll together without slip as shown in Fig. 6.53.
Smaller gear is known as pinion and larger mating gear is called gear. Generally, a gear pair acts
as a speed reducer aiming torque amplification at output shaft.

Fig. 6.53: Gear action .

Tooth Profile :

Generally tooth profile is designed so that velocity ratio does not change due to inaccuracies in
center distance. Involute profile may be visualized as the locus of points generated by the end of a
string, which is held in tension as it is wounded from a drum/cylinder, as shown in Fig. 6.54.
 Fig. 6.54: Concept of involute profile.

Tooth curves of the mating teeth need to be tangent to each other as shown in Fig. 6.55. Line of
action is tangent to both pinion & gear base circles. On changing center distance, line of action still
remains tangent to both base circles but slope changes.
 Fig. 6.55: Gears in action [1].

Pressure angle: Pressure angle is expressed as Φ1 = cos-1(Rb/RI). Nominal pressure angle is given
by :

The lower pressure angle has the advantage of smoother and quieter tooth action because of larger
profile contact ratio. In addition, lower loads are imposed on the support bearings because of a
decrease radial load component. Pressure angle at base circle is zero.

Backlash : Difference between tooth space and tooth thickness is known as backlash. It prevents
jamming of teeth and compensates for thermal expansion of teeth.
 Fig. 6.56: Backlash [1].

Velocity ratio :

The amount of speed reduction is simply the ratio of pitch diameters of the larger gear to the smaller
gear. There is no limit to the speed reduction ratio that can be achieved using gearing; but larger
ratio must be obtained using multi-stage reduction. In simple gear mesh, a maximum ratio in order
of 7:1 to 10:1 should not be exceeded. The limit on velocity ratio depends on gear pair for example
:

Speed reduction for a single pair of

• Spur gear < 7:1

• Helical 10:1
• Internal 4-8
• Bevel 1-8
• Cylindrical worm 3-80.

For high speed reduction, two stage or three stage construction are
preferred, otherwise gear wheel size increases, which increases the
gearbox size. For high speed reduction, compound gear trains are
required. In such trains(as shown in Fig. 6.57), at least one shaft
carries two gears.
Fig. 6.57: Compound gear
train.
 Fig. 6.58(a): Spur gear & Fig. 6.58(b): Helical gear.

An efficient method of achieving high reduction ratios in minimum space is the use of planetary
gearing. Helical gearing, in which the teeth are cut at an angle with the axis of rotation, was
developed after spur gearing and has the advantages that has the smoother action and tends to be
quieter(Fig. 6.58b). In helical angle greater than 15 degrees, the tooth bending capacity generally
begins to drop off due to the fact that the tooth thickness decreases rapidly. In addition, helical
gears causes axial thrust force and impose load on bearings.

Rolling and sliding in gears :

Ideally, rolling gears are required. In practice, sliding comes along rolling action. and therefore,
lubrication of gears is required. Typical gear pair having high sliding is shown in Fig. 6.59 to Fig.
6.61.

Fig. 6.59: Bevel gears

 Fig. 6.61: Worm gears [1].
Fig. 6.60: Helical crossed axis gears.

Worm gears have crossed axes(Fig. 6.61), line contact and a very large sliding component. Helical
crossed axes(Fig. 6.60) have point contact and large sliding component. Hypoid gears(Fig. 6.59)
are offset bevel gears. Hypoid gears are widely used in many power trains to transfer power between
two non-intersecting crossed axes. Their most common and highest-volume applications can be
found in front and rear axles of rear-wheel-drive or all-wheel-drive vehicles. It is interseting to note
that even spur gears experience sliding ,as depicted in Fig. 6.62.

Fig. 6.62: Sliding action in spur gears.

For rolling action, tangential velocities at point of contact must be equal to make sliding zero. This
happens at pitch point. At all other contact points(as shown in Fig. 6.63 at point I), radius of gear
and pinion will change (rotational speed remain constant) and that will introduce sliding.
Difference between VPI and VGI will provide positive/negative sliding speed.

Fig. 6.63: Sliding at point I

Friction & Lubrication Of Gears

Friction between gear pair occurs due to sliding between meshed teeth and churning of lubricant.
In absence of lubricant additives and antifriction coating, gears will be subjected to direct friction.
This is hypothetical situation which occurs rarely in extreme conditions. A worst situation from
tribology point of view i.e. high temperature, very high load, etc.

In normal atmospheric conditions, all engineering metallic surfaces are primarily coated with some
adsorbed gas(Fig. 6.64) and/or fluid films(Fig. 6.65). Shear strength of coating(τ i) is lower than that
of the base material(τy) as shown in following equation, they are continually rubbed off and
reformed. They thus protect the surface of the base material from excessive wear and subsequent
destruction. This favorable behavior is utilized by intentionally creating protective surface coating.
Following two coatings(as solid lubricants) are used to reduce friction.

Fig. 6.64: Gear surface with adsorbed gases.

• Phosphate layers, few microns.

• Graphite or molybdenum disulphide, 1-2 μ thin coating.

Fig. 6.65: Gear surfaces coated with boundary additive or antifriction coating.

Mixed Friction :

This friction process is aided by the presence of small quantities of lubricant, at the point of friction.
This friction process is characterized by solid-body friction as well as by fluid friction(Fig. 6.66),
therefore, it is called mixed friction. In this regime, friction and wear are influenced by the ability of
lubricant to create protective boundary films on gear tooth with chemical and physical reactions,
and by its viscous characteristics. Interface shear strength in mixed lubrication can be given by;
 Fig. 6.66: Mixed lubrication.

Fig. 6.67: Oils samples.

It is interesting to note that oil sample collected after 3 hours of operating gear at 500 rpm (no load
conition) show wear(Fig. 6.67). This means mild wear is bound to occur in gear operations.

Power Loss :

With involute profile of gears, only one contact position experiences pure rolling. As contact moves
towards or away from pitch point, sliding occurs. Due to sliding, power loss occurs and transmission
efficiency decreases. Typical values of gear efficiencies are listed in Table 6.16.
 Table 6.16: Gear efficiency.

Toothing load losses certainly account for most of the losses.

Toothing Power Loss :

In a geared system, the total power loss is comprised of two groups of losses: (i) load-dependent
(friction induced) mechanical power losses and (ii) load-independent (viscous) spin losses. Sliding
and rolling friction losses at the loaded gear meshes and at the bearings largely define the load-
dependent mechanical power losses. The total mechanical loss is then given as the sum of losses
from all gear meshes and bearings. The sliding friction losses are related to the coefficient of friction,
normal load and sliding velocity on the contact surfaces, while the rolling friction losses occur due
to the formation of an elastohydrodynamic (EHL) film. It is interesting to note that coefficient of
friction is variable and it depends on operating conditions. Friction losses can be divided into two
major categories: Loss due to load and loss due to speed. In high speed units, the churning losses
may exceed the friction losses; therefore, the type and amount of lubricant are critical.

• Load power loss.

Pload = Fr vg

Pload = μ Wn vg

Normal gear load (Wn) for a given application depends on pitch diameter and face width. These
dimensions are determined on the basis of tooth stresses, which are imposed by the transmitted
tooth load. The tooth load is simply the torque on a given gear divided by the gear pitch radius.
Torque is calculated from the horsepower transmitted and the speed of rotating component in
question.

Fig. 6.68: Normal load on tooth.

Coefficient of Friction :
μm = f(Wn, b, Ve, η, Rcomposite)

"Entraining velocity" - summation of rolling velocities.

Fig. 6.69: Higher entraining velocity increase "spin losses".

Composite roughness depends on gear manufaturing process as given in Table 6.17. The dynamic
viscosity η(νP) depends on operating temperature. The Walther`s relation relating kinematic
viscosity(ν) and absolute temperature is given by;

Table 6.17: R composite .

log10 log10 = (ν + 0.8) = Alog10T + B.

Gear life depends on effective lubrication, which can be quatified by minimum film thickness to
Rcomposite, as shown in Fig. 6.70.
 Fig. 6.70: Dependence of gear life on film thickness/surface roughness ratio.

Lubricant Film Thickness :

Curve fit EHL equation for minimum film thickness is given as;

Estimating effective temperature :

Following empirical formulae are used to estimate effective temperature(T F).

 Fig. 6.70: Contact stress.

where, σ is contact stress, V1 is tangential velocity of pinion, V2 is tangential velocity of gear and w
is width of contact patch as shown in Fig. 6.70.

Coefficient of thermal contact, β = thermal conduction * specific heat * density.

β = λCρ; On substituting value of contact patch, 2b and pmax from equation(12).

where

...........

Assuming same materials for gear & pinion.

where parameter ƒZ is decided based on number of teeth on pinion as given in Table 6.18 :
 Table 6.18: Parameter ƒ . Z

Higher value of Tf cause scuffing failure of gears. To reduce the value of Tf, effective lubrication that
maintains low friction coefficient is desirable. But sometime failure of lubrication (pump failure, filter
chocking, excessive leakage) occurs and gear materials must be able to handle such extreme
situations.

To understand this, let us consider two unlubricated gear pairs, one of Nylon/Nylon and other Nylon
steel pair. Nylon/Nylon gear pair friction is slightly more than Nylon/Steel gear pair.

Given data: Z = 17, m = 3 mm, b = 30 mm, ω = 150 rad/s, power = 850 W.

Table 6.19: Material parameters.

By using parameters shown in Table 6.19, calculate the value of temperature. For the N-N
temperature is 1610C but for N-S temperature is 19.50C ,which is very less value compared to the
N-N gear pair temperature. So N-S is prefferable.But for home applications, we use N-N gear pair,
where rate of heat generation is relatively low.

Surface fatigue of spur gears

To estimate the working life of gears, it is essential to analyze the destructive forces at work, and
knowledge of the ability of chosen gear materials to withstand those forces. Fig. 6.71 shows surface
pitting and bending failure of gear tooth. To design gear, we need to estimate Wn (Fig. 6.71) and
corresponding contact stresses. Bending & contact stresses must be within modified Goodman line
for material. Properly designed gear-sets should never fail but must be expected to eventually fail
by wear of one of surface. Insufficient backlash is sometimes the cause of excessive heat and wear.
If sufficient backlash has not been provided to take care of the differential thermal expansion, the
teeth will bind, with disastrous results. Inadequate lubrication may also be a source of excessive
heat and wear.

Fig. 6.71: Failures of gear.

Wear can be approximate using Archard`s equation.

Wear volume = [(K1 * load * sliding velocity)/(3 * hardness)]

Lesser value of K1 (i.e, better lubrication), lower load and high hardness reduce wear and enhance
gear life.

Surface/Contact stresses in spur gears :

Surface failure of gear tooth occurs due to very high local contact stresses. Maximum contact
pressure at the contact point between two cylinders is given by :

12

where pmax is max contact stress and b is half of contact patch.

d1 and d2 are curvatures of the profile at the point of contact.

On substituting, W = Wt/cos Φ, d1 = dp * sin Φ, L = F.

On substituting expression of b in pmax.

Maximum contact stress is equal to pmax. Therefore, contact stress is;

On rearranging;
 Table 6.20: Correction factors.

As per AGMA, we must include application factor, Ca (Table 6.20), load distribution factor, Cm (table
6.20) and velocity factor, Cv to estimate contact stress.

Calculation of Factor Cν :

Factor Cy depends on velocity and gear quality.

Table 6.21: Q vs geometric tolerance.

v

Example : A gear pair (ZP=23, θ = 200, Zg = 24, m = 1.75, F = 10.0 mm) transmits 8 N.m torque
from crankshaft (rotational speed 8000 rpm) of single cylinder IC engine to wheels. Bore diameter
of pinion is 17 mm, and bore dia of gear is 20 mm. Using AGMA pitting equation formula, determine
the maximum contact stress. Assume gears’ quality = 9, E = 2.e5 MPa, μ = 0.3
To avoid gear failure, σC must be lesser than material strength.

Contact Stress vs. Brinell Hardness :

Fig. 6.72: Effect of Brinell hardness on allowable contact stress for through-
hardness steel [2].

The Fig. 6.72 shows that strength of a gear tooth is proportional to the hardness of the steel. Most
gears are in the hardness range of approximately Rc 30 to Rc 38 or Rc 55 to Rc 64. The region from
Rc 30 to 38 is usually termed as “through hardened”, while the range Rc 55 to 64 is almost always
“surface hardened, where the tooth has a hard surface case and a softer inner core.

Lubrication :

Too much or too less lubricant is harmful for gear operation. Usually, following two types of
lubrication mechanisms are commonly used for gear lubrication.

• Splash lubrication, when power transmitted < 100 kW and Pitch_vel < 10 m/s.
• Pressurized lubrication (by oil jets) for large gear train transmitting power greater than 100 kW.

Following empirical formulae are availiable for splash lubrication system to minimize churning losses.

Table 6.22: Recommended oil viscosity [2].

With increase in pitch line velocity, lubricant used should be less viscous in order to min. power
losses. In case of heavily loaded gears, however, more viscous lubricant will be recommended. Table
6.22 provides an initial guidance for lubricant selection.

Long service life, free from wear problems depends on lubrication system; its ability to keep gear
cool, and to deliver lubricant free from hard particles (filter with 5 micron rating). Lack of lubricant,
may initiate scuffing failure.

References :

1. Hamrock B J, Jacobson B O & Schmid S R, Fundamentals of Machine Elements, McGraw-Hill Inc.,

1998.

2. Shigley J E, Mischke C R, Mechanical Engineering Design, Tata McGraw-Hill Publishing Company

Limited, 2003.
Module 6 : Application of Tribology
Journal Bearings

Bearings are needed to support the relatively moving components. Journal bearings support the
cylindrical rotating shaft. Journal bearing are designed based on lubricant and lubrication
mechanisms in following categories :

Fig. 6.74: Squeeze film

Fig. 6.73: Dry bearing impreganated with solid
lubrication.
lubricants.
 .........................(a) Hydrodynamic ................................ (b) Hydrostatic

Fig. 6.75: Journal bearing.

Dry bearings :

These bearings are made of self lubricated solid lubricant materials. Bearings operate dry or with
limited lubrication and experience solid to solid contact, which means relatively high friction
compared to full fluid film bearings..

For low temperature application, following materials can be utilized.

• Polymers: Thermoplastics --> PTFE, nylon, acetal resin.

• Thermosetting --> Phenolics, polyesters, and polyimides.

Following four material parameters/properties play important role in choosing material for dry
bearings.

• Maximum Temperature – Softening.

• PV limit– Heat generation, Wear rate.

• Maximum Pressure(P)

• Maximum velocity (V)

Most commonly used materials for dry bearings are listed in Table 6.23, along with the above four
properties.

Table 6.23: Materials for dry bearings.

Maximum temperature above which physical deterioration or abnormal softening happens is to be

avoided.
PV Approach :

For a given coefficient of friction, the PV product gives a measure of frictional heating and
temperature rise. PV values also gives a measure of wear rate. Material worn away is approximately
proportional to the normal load and distance travelled. P is simply the total force divided by the
bearing area. In dry bearings, actual contact between two rubbing surfaces occurs at sufficient high
asperities. Actual area of contact depends on the yield pressure of soft material. This imposes a limit
on the applied pressure and relative speed between two surfaces. Further, there is a limit for heat
dissipation from those surfaces. Product of loading pressure, sliding speed and coefficient of friction
identifies the rate energy release at the bearing surface. Since non-fluid film bearings are limited in
their ability to dissipate heat, the P-V approach is often employed to design dry bearing.

• PV factor : Limiting ‘PV’ above which wear increases rapidly either as a consequence of thermal
effects or of stresses approaching to elastic limit, is known as PV factor. Bearing design based on
PV factor is termed as PV approach.

• EX : A shaft running at 1000 rpm is supported on a porous bearing. Shaft dia = 1 inch and L/D =
1. Applied load = 1200 lbf.

- Velocity limit is 1180 ft/min.

- Pressure limit is 2000 psi.
- PV limit is 110,000 psi-ft/min.

Fig. 6.76: PV approach.

• Calculation of average pressure = 1200/(1*1) psi (8.273709 MPa) --> bearing is safe.

• Calculation of velocity = π.D.N. = 3.14*(1/12)* 1000 = 261 ft/min (1.326 m/s) --> bearing is
safe.

• Calculation of PV limit = 1200 psi * 261 = 313,000 psi-ft/min --> bearing will fail.

This example illustrates that bearing may be safe from pressure and velocity point of view but may
fail due to limited temperature dissipation.

Estimating bearing life: Bearing life may be evaluated based on the volume worn out from bearing
surface, which can be estimated using.

Volume of wear = specific wear rate * Applied load * distance of sliding.

ν = kWd

=> h = kPVt.

Wear constant k can be considered based on the material used, as shown in Table 6.24.

Table 6.24: Wear factor & friction coefficient.

EX : Estimate wear of 10mm long Nylon bushing supporting a 10-mm diameter and 5 kg shaft
running at 900 rpm.

Wear factors for polymeric bearings :

Velocity, V = Π * D * N/60 => V = 0.47124 m/s.

Pressure, p = W/(L * D) => 490500(N/m2).

Wear volume as per Archard`s equation

On substituting values of k for Nylon material from Table 6.24.

This value is too high. Wear can be reduced by appropriate filler in nylon with such material wear
thickness.

h = (0.24/4)0.32 ≈ 0.2 mm it is acceptable.

Table 6.25: Bearing materials.

Table 6.24 lists few materials that can be used under dry conditions. It is interesting to note that
velocity limit for iron copper is the lowest, while PV is the highest for same materials.

Fig. 6.77: Porous bearing.

Porous bearings :

Porous bearing may be treated as starved bearings. These bearings are modified form of “hydrostatic
bearings with orifice compensators”. Oil flows due to capillary action through the pores in the
unloaded region. The oil flows back through pores in the loaded part of bearing shell. These bearings
work satisfactory for light load and moderate speed. Porus bearing are of powdered metals(Fig.
6.77) which are pressed in dies. After compression, they are sintered at a high temperature in a
reducing atmosphere. Sintering causes powdered metal to fuse into a strong compact. After
sintering, bearings may be submerged in oil for impregnation. Voids may vary from 16% up to 36%
of the volume of the bearing. Bearings are finish-sized in punch press to close tolerances.
 Fig. 6.78: PV approach for porous bearing.

These bearings are often known as economic mean to feed lubricant to the bearing. It is interesting
to note that Costporous-bearing > Costlubricated-bearing, but overall cost of bearing + lubricant + lubrication
system may be higher than the cost of dry bearing. Operating temp puts a limit on working of porous
bearing as heat is dissipated through bounding solids and cause deterioration of mineral oils.
Therefore these bearings are recommended for small electric motors, household appliances,
automotive accessories, etc.

Hydrostatic Bearings

If pressurized lubricant is supplied continuously (without interruption), separation between bearing

surfaces with an adequate film thickness can be maintained even when speed is nil. Poor lubricants
such as air, water and liquid metals are usable. Oil film thickness varies as inverse of cube root of
load. Design of hydrostatic bearings commonly involve more complex lubrication systems, and
requires specialized design and application practices. High pressure supply, reliability & life of high
pressure oil lines are always in doubt.

Fig. 6.79: Hydrostatic bearing with orifice compensators.

Advantages of hydrostatic Bearing :

• Completely elimination of wear.

• Reduction in coefficient of friction may be 1/100 to 1/500.

• High stiffness.

•h α W-1/3

• Good for starting, stopping. Best for heavy load and extremly low speed.
 Fig. 6.80: Hydrostatic lift.

Estimating load capacity :

If the pressure and quantity of flow rate are in correct proportions, the shaft whether it is rotating
or not, will be raised and supported by an oil film. This arrangement of shaft(Fig. 6.80) support
shows as θ increases the oil film pressure decreases. Here the film thickness h is not constant, but
depends upon the angular position θ and eccentricity of journal in the bearing. When the shaft has
settled in the bearing and has made metal to metal contact, the eccentricity ratio is equal to 1.0.
When the shaft and bearing are concentric, the eccentricity ratio (ε) is zero. Assume a shaft of radius
r being floated in a bearing of radius R by oil, pumped through a slot at pressure PS, with;

where ε = e/Cr. On substituting

On - integrating;
where D is integration constant

Pressure P acts on area r.dθ.b and vertical component of force P.r.dθ.b.cos θ will balance the applied
load W :
Fig. 6.81 shows variation in load capacity of hydrostatic bearing on varying eccentricity ratio.

Fig. 6.81: Load versus eccentricity ratio.

Fig. 6.82: Multipocket bearing.

Negative value of eccentricity ratio describes the journal position when it is above bearing center.
This may be the case with double acting journal bearing or multi-pad journal bearings as shown in
Fig. 6.82.

Ex: A 101.6 mm diameter journal rests in a bearing of diameter 101.9 mm. Lubricant of 30 mPa.s
is supplied under pressure through a groove at the lowest point of the bearing. Length of bearing is
152.4 mm and applied load is 16 kN. What inlet pressure and flow rate are needed to raise the
journal by 0.0508 mm.
 Hydrodynamic journal bearings

Low cost, simple manufacturing, negligible wear and very low coefficient of frictions are main
features of hydrodynamic journal bearings. Two journal bearings, made of solid rubber and
lubricated by sea water are shown in Fig. 6.83. These bearings are known as long bearing as length
to diameter ratio is greater than two. These bearings are full (360 0) cylindrical bearings. In practice
nearly half bearing supports the loads. To understand this we fabricated journal bearing made of
acrylic material as shown in Fig. 6.84. We could observe the lubricated and cavilated regions of the
bearing. Two sketches of observed region is provided in Fig. 6.85.

Fig. 6.83: Long journal bearing.

Fig. 6.84: Acrylic journal bearing with temp measuring copper rivets.

Fig. 6.85(a): Journal bearing with cavitation.

 Fig. 6.85(b): Cavitated journal bearing.

A simplified solution to analyze a journal bearing, based on half Sommerfeld boundary condition is
given in table 6.25.1.

Table 6.25.1: Journal bearing design table.

Here, C is radial clearance (m), F is fluid force, D is journal diameter (m), n s is journal rotational
speed(rps),and Λ is length to diameter ratio. Here fric is friction force, L is length (m), ρ is oil density
(kg/m3), Co is specific heat of lubricant (J/kg.K), θ is circumferential coordinate, R is radius (m), U
is velocity in m/s, ΔT is temperature rise, κ is temperature coefficient (1/ 0C), Q is oil flow (cubic-
meter/sec), W is power loss (Watts), Φ is attitude angle (radians) and μ is average viscosity of
lubricant (Pa.s). The design of journal bearing is an inverse problem, where for a given load and
speed, the eccentricity ratio and attitude angle are determined. The angle between the load line and
line passing through bearing and journal centers is known as the attitude angle. The attitude angle
is a representative of bearing stability. Larger the attitude angle, lesser is the stability of bearing.
To solve bearing-inverse problem, initial guess of eccentricity ratio and attitude angle are made,
and an iterative procedure is used to match the given load using;

Finite difference mass conserving algorithm :

The eccentric position of journal in a hydrodynamic bearing forms a converging-diverging clearance

space. Due to unavailability of sufficient liquid lubricant, liquid-streamers separated by gas/vapor
space are formed in the divergent clearance space, as indicated in (Fig. 6.85(b)). This “cavitation”
phenomenon cannot be appropriately predicted using the Reynolds equation alone, or associating it
with Gumbel/Reynolds boundary condition. For example; using Gumbel conditions, to predict the
bearing performance simply neglects the cavitated zone, does not account recirculating flow (as
shown in below Fig. 6.85(c)), and assumes only half bearing.

Fig. 6.85(c): Recirculation of lubricating oil.

Such assumption predicts load capacity reasonably well (as cavitated pressure does not contribute
to load carrying capacity), but suffers inaccuracy in predicting oil flow and power loss. Therefore,
there is a need for a universal equation that describes both the full film and cavitation regions to
their full extent, such as:

where β is bulk modulus (Pa), Pc is cavitation pressure (Pa),

 ....Eq.(6.2)

The value of g = 1 makes Eq.(6.2) an elliptic partial differential equation (PDE), while makes Eq.(6.2)
a hyperbolic PDE. A robust convergent solution needs a central-finite-difference-scheme for elliptical
PDE, and upwind-finite-difference scheme for hyperbolic PDE. This can be achieved by adopting a
type difference scheme, such as:

....Eq.(6.3)

....Eq.(6.4)

Substituting Eq.(6.3 - 6.4) in Eq.(6.2) provides Eq.(6.5) with θ as unknown : f(θ) = 0.

....Eq.(6.5)

Values of θ can be evaluated using Newton iterative scheme :

where θ* is the available value of θ. The Newton method provides rapid convergence if initial guess
of the unknown is in the neighborhood of real solution. Here the analytical approach, discussed in
previous section, helps to determine reasonably good initial approximation. The oil leakage and
power loss can be evaluated by using the following expressions :

Locating Journal Position :

In hydrodynamic journal bearing, positions of journal floats and gets located based on operating
parameters(i.e, load, speed, viscosity etc). Fig. 6.86 shows three positions of journal in bearing.
First positive is corresponding to negligible load and relatively high speed. In this situation, journal
center coincides (ε = 0) with bearing center. Third position corresponds to high load and negligible
speed. In middle position, speed and load are measurable. To find ε and Φ, numerical solution of
Reynolds equation and integration of fluid pressure over effective area are required.

Fig. 6.86: Locations of journal in bearing.

How bearing gets lubricant ?

Effective full film (hydrodynamic) lubrication can be achieved by continously feeding lubricant (as
pressure greater than ambient pressure) through oil groove arrangement as shown in Fig. 6.87. In
other words, bearings are provided with feed hole and oil grooved to get lubricant.
Fig. 6.87: Hole and groove arrangement in bearing.

Fig. 6.88: Partial oil groove.

 Fig. 6.89: Oil supply path.

Oil can be supplied to journal bearing by oil hole(Fig. 6.89), partial groove(Fig. 6.88) or full
groove(Fig. 6.87) arrangement depends on required quantity of oil and load direction. If applied
load changes its direction then partial to full groove arrangement is essential to avoid starvation.

Flow rate under supply pressure Ps in bearing with oil hole(dh)[1] :

Flow rate under supply pressure Ps and lubricating full oil groove[1] :
Flow rate under supply pressure Ps and partial groove(Fig. 6.68)[1] :

Designing journal bearings :

Bearing is designed with bearing clearence(c) and length(L) which often is decided based on bearing
bore dia(D). Increasing bearing length increases load capacity of the bearing. Upper limit on bearing
length is decided based on space limitation. Increasing radial clearence decreases load capacity and
increases lubricant flow rate.
Pressure P, developed in bearing depends on geometric, lubricant and operating parameters.
Developed fluid pressure should not exceed the permissible value for bearing material(table 6.26).

Table 6.26: Bearing materials.

Pressure Estimation :

In earlier module advantages of hybrid approach were detailed. In the present case fluid pressure
is estimated using hybrid approach.
 Fig. 6.90: Long static bearing.

Fig. 6.91: Short static bearing.

P∞ is fluid pressure estimated by long bearing approximation(Fig. 6.90) and P 0 is fluid pressure
estimated by short bearing approximation(Fig. 6.91).

Geometric Configuration of Journal Bearing :

Fluid film thickness for geometry sketched in Fig. 6.92 is expressed as;

Fig. 6.92: Fluid film in journal bearing.

Deriving fluid film pressure considering short static bearing approximation :

On first integration and using symmetry condition dP/dz = 0 at z = 0.

Load capacity of short journal bearing :

Fig. 6.93: Load distribution.

Resultant load

Attitude angle :
 Fig. 6.94: Attitude angle vs eccentricity ratio for short bearing.

Friction force :

In hydrodynamic bearing friction depends on fluid shearing.

Oil flow :

• Flow due to velocity & feed pressure.

• Circumferential flow.

Above expression is based on Gumbel boundary condition and assumption that lubricant is partially
recirculated back (Fig. 6.95) in bearing.

On neglecting pressure gradient

QH = (U/2)c(1 + ε - 1 + ε)L.

or QH = U c ε L.

Mass flow rate, m = ρ U c ε L.

Fig. 6.95: Oil flow in journal bearing.

Temperature Rise :
Friction, due to shear of lubricant film, generates heat (F * U)) in lubricant oil and increases the
temperature of lubricant. Assuming that total generated heat is carried by the oil flowing through
bearing.

Rate of heat generated = Rate of heat convected by oil flow.

Design of Hydrodynamic Journal Bearing :

1. Guess eccentricity ratio.

2. Calculate load capacity, friction force, temperature rise.

3. Modify lubricant viscosity using η = ηin e-β Δt .

4. Repeat steps 1-3 so that average viscosity and load converge.

Example : Determine the minimum film thickness, maximum pressure, coefficient of friction for a
hydrodynamic journal bearing, which supports a 600 N laod at rotational speed of 2000 rpm. The
shaft dia. is 40 mm. Assume bearing length = 10 mm, oil viscosity at room temperature (300C) =
15 mPa.s, β = 0.029, and radial clearance 20 μm.

Given : U = 4.19 m/s. Factor 1 = U * L3 * π * 0.25/(c2) = 8227 m2/s, Factor 2 = 2 * U * L * R *

(π/c). These factors remain constant and do not change over iterations.

Load
STEP 1: Assume ε = 0.5 -> W = 118 N. This load capacity is very less.

assume ε = 0.75 -> W = 562 N.

STEP 2: Assume ε = 0.8 -> W = 900 N, F = 6.58 N, Δt = 8.65 0C.

STEP 3: Modify viscosity using η = ηin e-β Δt at Δt = 8.650C,

η = 0.0117.

STEP 4: For ε = 0.8 -> W = 703 N, F = 5.13 N, Δt = 6.75 0C,

η = 0.0123.

Now it is preferable to decrease ε. Let us assume ε = 0.78 and Δt = 7 0C

W = 599.1 N, F = 5.13 N,

Δt = 6.920C, η = 0.0122.

Answer ε = 0.78.

Φ = 32.20 using tan Φ = (π/4)[√(1 - ε2)/ε].

Minimum film thickness, hmin = c(1 - ε)

hmin = 20(1 - 0.78) => hmin = 4.4 micron.

On checking lubrication regime Λ = hmin/√(R2q1 + R2q2).

If Rq1 and Rq2 are approximately 0.4 μm, bearing operates in hydrodynamic regime.
Example : Same data which were defined in previous example, but bearing length = 20 mm.

STEP 1: Assume ε = 0.5 -> W = 943 N; Assume ε = 0.45 -> W = 741 N.

STEP 2: For ε = 0.45 -> F = 8.84 N, Δt = 10.33 0C.

STEP 3: Modify viscosity using η = ηin e-β Δt at Δt = 10.330C.

η = 0.0111.

STEP 4: For ε = 0.45 -> W = 548 N, F = 6.6 N, Δt = 8.00C, η = 0.0119. W = 588 N, F = 7 N, Δt =

8.20C.

STEP 5: For ε = 0.46 -> W = 606 N, F = 6.94.

Answer ε = 0.45.

Φ = 57.30 tan Φ = (Π/4) [√(1 - ε2)/ε.

pmax = 13 bar.

μ = F/W = 7/600 => μ = 0.0117.

This pressure and μ are higher than those estimated for 50% length of bearing.

pmax = 7.24 bar.

μ = F/W = 5.13/600 => μ = 0.0086.

Three examples of journal bearing were used to illustrate the accuracy of hybrid bearing approach.
Design procedure suggested in previous two examples can be used with hybrid pressure expression
in place of pressure by short bearing approximation.

Example 1 :

Example 2 :
Example 3 :
References :

1. Martin F A and Lee C S, “Feed pressure flow in plain journal bearings” ASLE Transactions, 26 (3),
381-392, 1983.