Introduction
5
1.
In any situation where the resultant of tangential forces is smaller
than some force parameter specific to that particular situation, the
friction force will be equal and opposite to the resultant of the
applied forces and no tangential motion will occur.
2.
When tangential motion occurs, the friction force always acts in a
direction opposite to that of the relative velocity of the surfaces.
3.
The friction force is proportional to the normal load.
4.
The coefficient of friction
is
independent of the apparent contact
area.
5.
The static coefficient
is
greater than the kinetic coefficient.
6.
The coefficient of friction is independent of sliding speed.
Strictly speaking, none of these laws is entirely accurate. Moore indicated
that laws
(3),
(4),
(9,
and
(6)
are reasonably valid for dry friction under the

and Eisner
[21]
had shown that when the tangential force Fis first applied, a
very small displacement occurs almost instantaneously in the direction of
F
with a magnitude in the order of 10-5 or
10-6
cm.
Seireg and Weiter
[22]
conducted experiments to investigate the load-
displacement and displacement-time characteristics of friction contacts of a
ball between two parallel flats under low rates of tangential load application.
The tests showed that the frictional joint exhibited “creep” behavior at
room temperatures under loads below the gross slip values which could
be described by a Boltzmann model
of
viscoelasticity.
They also investigated the frictional behaviors under dynamic excitation
[23, 241.
They found that under sinusoidal tangential forces the “break-
away” coefficient
of
friction was the same as that determined under static
conditions. They also found that the static coefficient of friction in Hertzian
contacts was independent of the area of contact, the magnitude
of
the
normal force, the frequency of the oscillatory tangential load, or the ratio
6

the mechanical interlocking of surface roughness elements. This
theory gives an explanation for the existence of a static coefficient
of friction, and explains dynamic friction as the force required to lift
the asperities of the upper surface over those
of
the lower surface.
Molecular attraction.
This was proposed by Tomlinson in 1929 and
Hardy in 1936 and attributes frictional forces to energy dissipation
when the atoms of one material are “plucked” out of the attraction
range of their counterparts on the mating surface. Later work
attributed adhesional friction to a molecular-kinetic bond rupture
process in which energy is dissipated by the stretch, break, and
relaxation cycle of surface and subsurface molecules.
Efectrostatic forces.
This mechanism was presented in 1961 and explains
the stick-slip phenomena between rubbing metal surfaces by the
initiation of a net flow of electrons.
Welding, shearing and ploughing.
This mechanism was proposed by
Bowden in 1950.
It
suggests that the pressure developed at the
discrete contact spots causes local welding. The functions thus
formed are subsequently sheared by relative sliding of the surfaces.
Ploughing by the asperities of the harder surface through the matrix
Introduction
7
of the softer material contributes the deformation component of
friction.

established that welding occurs at
a
critical temperature which is reached by
frictional heating of the surfaces. The method of calculating such a tem-
perature was published by Blok
[31],
and his results were adapted to gears in
1952
[32]. Since then, some emphasis has been focused on boundary lubrica-
tion. Several studies are available in the literature which deal with the
boundary lubrication condition; some of them are briefly reviewed in the
following.
Sharma [33] used the Bowden-Leben apparatus to investigate the effects
of load and surface roughness
on
the frictional behavior of various steels
over a range of temperature and
of
additive concentration. The following
observations are reported:
Sharp rise in friction can occur but is not necessarily followed by scuff-
Load affects the critical temperature quite strongly.
ing and surface damage.
8
Chapter
I
Neither the smoother surface nor the rougher surface gives the max-
imum absorption of heat, but there exists an optimum surface
roughness.
Nemlekar and Cheng [34] investigated the traction in rough contacts by

spheric pressure and at the test temperature; neither pressure-viscosity nor
temperature-viscosity properties appeared to be important factors. On the
other hand, non-Newtonian fluids (polymer-thickened oils) gave more con-
tact than their mineral oil counterparts. This suggested that shear-viscosity
was important. However, no beneficial effects of viscoelastic properties were
observed with these oils. Friction generally decreased with increasing visc-
osity because the more viscous oils gave less metal-to-metal contact. The
coefficient of friction was rather high: 0.13 at low viscosity, dropping to
0.08
at high viscosity. The oils having higher
PVIs
(pressure-viscosity index
a)
gave somewhat more friction which cannot be solely attributed to differ-
ences in metallic contact.
Furey [37] also investigated the surface roughness effects on metallic
contact and friction in the transition zone between the hydrodynamic and
boundary lubrications. He found that very smooth and very rough surfaces
gave less metallic contact than surfaces with intermediate roughness.
Friction was low for the highly polished surfaces and rose with increasing
Introduction
9
surface roughness. The rise in friction continued up to a roughness of about
10
pin, the same general level at which metallic contact stopped increasing.
However, whereas further increases in surface roughness caused a reduction
in metallic contact, there was no significant effect on friction. Friction was
found to be independent of roughness in the range of lopin center line
average (CLA). He also used four different types of antiwearlantifriction
additives (including tricresyl phosphate) and found that they reduced metal-

In general the variables that influence dry friction also influence bound-
ary friction.
Friction and surface damage depends
on
the chemical composition of
the lubricant and/or the products of reaction between the lubricant
and the solid surface.
Lubricant layers only a few molecules thick can provide effective
boundary lubrication.
The frictional behavior may be influenced by surface roughness, tem-
perature, presence
of
moisture, oxygen or other surface contami-
nants. In general, the coefficient of friction tends to increase with
surface roughness.
1.4
FRICTION IN FLUID FILM LUBRICATION
Among the early investigations in fluid film lubrication, Tower’s experi-
ments in 1883-1884 were a breakthrough which led to the development
of
10
Chapter
I
lubrication theory
[6,
71.
Tower reported the results of a series of experi-
ments intended to determine the best methods to lubricate a railroad journal
bearing. Working with a partial journal bearing in an oil bath, he noticed
and later measured the pressure generated in the oil film.

the film thickness to the bearing geometry is in the order of 10-3, Reynolds
established the well-known theory using an order-of-magnitude analysis.
The assumptions on which the theory is based can be listed as follows.
The pressure is constant across the thickness of the film.
The radius of curvature of bearing surface is large compared with film
The lubricant behaves as a Newtonian fluid.
Inertia and body forces are small compared with viscous and pressure
There is no slip at the boundaries.
Both bearing surfaces are rigid and elastic deformations are neglected.
thickness.
terms in the equations of motion.
Since then the hydrodynamic theory based on Reynolds’ work has attracted
considerable attention because of its practical importance. Most initial
investigations assumed isoviscous conditions in the film to simplify the ana-
lysis. This assumption provided good correlation with pressure distribution
Introduction
I1
under a given load but generally failed to predict the stiffness and damping
behavior of the bearing.
A
model which predicts bearing performance based on appropriate
thermal boundaries on the stationary and moving surfaces and includes a
pointwise variation of the film viscosity with temperature is generally
referred to as the thermohydrodynamic (THD) model. The THD analyses
in the past three decades have drawn considerable attention to the thermal
aspects of lubrication. Many experimental and theoretical studies have been
undertaken to shed some light on the influence of the energy generated in
the film, and the heat transfer within the film and to the surroundings, on
the generated pressure.
In 1929, McKee and McKee [39] performed a series of experiments on a

tricity ratios in a journal bearing lubricated with a non-Newtonian oil. They
found that a non-Newtonian oil shows
a
lower friction than a corresponding
Newtonian fluid under the same operating conditions. However, this
phenomenon did not agree with their analytical work and could not be
explained.
12
Chapter
I
Maximum bearing temperature
is
an important parameter which,
together with the minimum film thickness, constitutes a failure mechanism
in fluid film bearings. Brown and Newman
[45]
reported that for lightly
loaded bearings
of
diameter
60
in. operating under
6000
rpm, failure due
to overheating of the bearing material (babbitt) occurred at about
340°F.
Booser et al.
[46]
observed a babbitt-limiting maximum temperature in the
range

perature is small and the shaft can be treated as an isothermal component.
The experiments also indicated that the axial temperature gradients within
the bushing are negligible.
Viscosity is generally considered
to
be
the single most important prop-
erty of lubricants, therefore, it represents the central parameter in recent
lubricant analyses. By far the easiest approach to the question of viscosity
variation within a fluid film bearing is to adopt a representative or mean
value viscosity. Studies have provided many suggestions for calculations
of
the effective viscosity in a bearing analysis
[48].
When the temperature rise
of the lubricant across the bearing is small, bearing performance calcula-
tions are customarily based on the classical, isoviscous theory. In other
cases, where the temperature rise across the bearing is significant, the
classical theory loses its usefulness
for
performance prediction. One
of
the early applications of the energy equation to hydrodynamic lubrication
was made by Cope
[49]
in
1948.
His model was based on the assumptions
of negligible temperature variation across the film and negligible heat
conduction within the lubrication film as well as into the neighboring

mentally may differ considerably from those predicted by the insoviscous
hydrodynamic theory. The isoviscous theory can either underestimate or
overestimate the results depending on the operating conditions.
It
was
observed, however, that the normalized pressure distribution in both the
circumferential and axial directions of the journal bearing are almost iden-
tical to those predicted by the isoviscous hydrodynamic theory. Under all
conditions tested, the magnitude
of
the peak pressure (or the average pres-
sure) in the film is approximately proportional to the square root of the
rotational speed
of
the journal. The same relationship between the peak
pressure and speed was observed by Wang and Seireg
[52]
in a series
of
tests on a reciprocating slider bearing with fixed film geometry.
A
compre-
hensive review of thermal effects in hydrodynamic bearings is given by
Khonsari [53] and deals with both journal and slider bearings.
In 1975, Seireg and
Doshi
[54] studied nonsteady state behavior of the
journal bearing performance. The transient bushing temperature distribu-
tion in journal bearing appears to be similar to the steady-state temperature
distribution. It was also found that the maximum bushing surface tempera-

journal bearing to study the performance of a plain bearing. The pressure
and the temperature distributions on the bearing wall were measured, along
with the eccentricity ratio and the flow rate, for different speeds and loads.
All measurements were performed under steady-state conditions when ther-
mal equilibrium was reached. Good agreement was found with measure-
ments reported for pressure and temperature, but a large discrepancy was
noted between the predicted and measured values of eccentricity ratios. In
1986, Boncompain et al. [59] showed good agreement between their theo-
retical and experimental work on a journal bearing analysis. However, the
measured journal locus and calculated values differ. They concluded that
the temperature gradient across and along the fluid film is the most impor-
tant parameter when evaluating the bearing performance.
1.5
FRICTIONAL RESISTANCE IN ELASTOHYDRODYNAMIC
CONTACTS
In many mechanical systems, load is transmitted through lubricated con-
centrated contacts where rolling and sliding can occur. For such conditions
the pressure is expected to be sufficiently high to cause appreciable deforma-
tion of the contacting bodies and consequently the surface geometry in the
loaded area is a function of the generated pressure. The study of the beha-
vior of the lubricant film with consideration of the change of film geometry
due to the elasticity of the contacting bodies has attracted considerable
attention from tribologists over the last half century. Some of the studies
related to frictional resistance in this elastohydrodynamic (EHD) regime
are briefly reviewed in the following with emphasis on effect of viscosity and
temperature in the film.
Dyson [60] interpreted some of the friction results in terms
of
a model of
viscoelastic liquid. He divided the experimental curves of frictional traction

erned by the characteristics of the lubricant film, which, in the case of a
sliding contact, depends strongly on the temperature in the contact. The
temperature field is in turn governed directly by the heating function.
Crook
[61]
studied the friction and the temperature in
oil
films theore-
tically. He used a Newtonian liquid (shear stress proportional to the velocity
gradient in the film) and an exponential relation between viscosity and tem-
perature and pressure. In pure rolling of two disks
it
has been found that
there
is
no temperature rise within the pressure zone; the temperature rise
occurs on the entry side ahead
of
that zone. When sliding is introduced,
it
has
been found that the temperature on the entry side remains small, but
it
does
have
a
very marked influence upon the temperatures within the pressure
zone, for instance, the introduction of
400
cm/sec sliding causes the effective

with a more rigorous analysis of temperature by using a two-dimensional
numerical method. The effect of the local pressure-temperature-dependent
viscosity, the compressibility of the lubricant, and the heat from compres-
sion of the lubricant were considered in the analysis.
A
Newtonian liquid
was used. He found that the temperature had major influence on friction
force. A slight change in temperature-viscosity exponent could cause great
changes in friction data. He also compared his theoretical results with
Crook’s
[62]
experimental results and found a high theoretical value at
low sliding speed. Thus he concluded that the assumption of a Newtonian
fluid in the vicinity of the pressure peak might cease to be valid.
One of the most important experimental studies in EHD was carried out
by Johnson and Cameron
[64].
In their experiments they found that at high
sliding speeds the friction coefficient approached a common ceiling, which
was largely independent of contact pressure, rolling speed and disk tempera-
ture. At high loads and sliding speeds variations in rolling speed, disk tem-
perature and contact pressure did not appear to affect the friction
coefficient. Below the ceiling the friction coefficient increased with pressure
and decreased with increasing rolling speed and temperature.
Dowson and Whitaker
[65]
developed a numerical procedure to solve
the EHD problem of rolling and sliding contacts lubricated by a Newtonian
fluid. It was found that sliding caused an increase in the film temperatures
within the zone, and the temperature rise was roughly proportional to the

4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
MacCurdy, E., Leonardo da Vinci Notebooks, Jonathan Cape, London,
England, 1938.
Amontons, G., Histoire de 1’AcadCmie Royale des Sciences avec Les Memoires
de Mathematique et de Physique, Paris, 1699.
Coulomb, C. A., Memoires de Mathematique et de Physique de 1’Academie
Royale des Sciences, Paris, 1785.
Reynolds,
O.,
“On the Theory
of
Lubrication and Its Application to Mr.
Beauchamp Tower’s Experiments Including an Experimental Determination

of
Tribology, Longman, New York, NY, 1979.
Bowden, F. P., and Tabor, D., The Friction and Lubrication of Solids, Oxford
University Press, New York, NY, 1950.
Pinkus,
O.,
“The Reynolds Centennial: A Brief History of the Theory of
Hydrodynamic Lubrication,” ASME J. Tribol., 1987, Vol. 109,
pp.
2-20.
Pinkus,
O.,
Thermal Aspects
of
Fluid Film Tribology, ASME Press, pp. 126-
131, 1990.
Ling,
F.
F., Editor, “Wear Life Prediction in Mechanical Components,”
Industrial Research Institute, New York NY, 1985.
Suzuki,
S.,
Matsuura, T., Uchizawa, M., Yura,
S.,
Shibata,
H.,
and Fujita,
H.,
“Friction and Wear Studies
on

18
Chapter
1
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
Seireg, A., and Weiter,
E.
J., “Viscoelastic Behavior of Frictional Hertzian
Contacts Under Ramp-Type Loads,” Proc. Inst. Mech. Engrs, 1966-67, Vol.
181, Pt. 30, pp. 200-206.
Seireg, A. and Weiter, E. J., “Frictional Interface Behavior Under Dynamic

Kelley, B. W.,
“A
New Look at the Scoring Phenomena of Gears,” SAE
Trans., 1953, Vol. 61,
p.
175.
Sharma, J. P., and Cameron, A., “Surface Roughness and Load in Boundary
Lubrication,” ASLE Trans., Vol. 16(4), pp. 258-266.
Nemlekar, P. R., and Cheng,
H.
S.,
“Traction in Rough Elastohydrodynamic
Contacts,” Surface Roughness Effects in Hydrodynamic and Mixed
Lubrication“, The Winter Annual Meeting of ASME, 1980.
Hirst,
W.,
and Stafford, J.
V.,
“Transion Temperatures in Boundary
Lubrication,” Proc. Instn. Mech. Engrs, 1972, Vol. 186( 15/72), 179.
Furey, M.
J.,
and Appeldoorn,
J.
K.,
“The Effect of Lubricant Viscosity on
Metallic Contact and Friction in a Sliding System,” ASLE Trans. 1962, Vol.
5,
Furey, M. J., “Surface Roughness on Metallic Contact and Friction,” ASLE
Trans., 1963, Vol. 6, pp. 49-59.

50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
Tao, F. and Phillipoff, W., “Hydrodynamic Behavior of Viscoelastic Liquids
in
a Simulated Journal Bearing,” ASLE Trans., 1967, Vol. 10(3), p. 307.
Dubois,
G.,
Ocvrik, F., and Wehe, R., “Study of Effect of a Newtonian Oil on
Friction and Eccentricity Ratio of a Plain Journal Bearing,” NASA Tech.
Note, D-427, 1960.
Brown, T., and Newman, A., “High-speed Highly Loaded Bearings and Their
Development,” Proc. Conf. on Lub. and Wear, Inst. Mech. Engrs., 1957.
Booser et al. “Performance of Large Steam Turbine Journal Bearings,” ASLE
Trans., Vol. 13, n.4, Oct. 1970, pp. 262-268. Also, “Maximum Temperature for
Hydrodynamic Bearings Under Steady Load,” Lubric. Eng., Vol. 26, n.7, July
1970, pp. 226-235.
Dowson, D., Hudson, J., Hunter, B., and March, C., “An Experimental
Investigation of the Thermal Equilibrium of Steadily Loaded Journal
Bearings,” Proc. Inst. Mech. Engrs, 1966-67, Vol. 101, 3B.
Cameron, A., The Principles of Lubrication, Longmans Green
&
Co., London,

Seireg, A., and Dandage
S., “Empirical Design Procedure for the
Thermohydrodynamic Behavior of Journal Bearings,” ASME J. Lubr.
Technol., 1982, pp. 135-148.
Barwell, F.T., and Lingard,
S., “The Thermal Equilibrium of Plain Journal
Bearings,” Proceedings of the 6th Leeds-Lyon Symposium on Tribology,
Dowson, D. et al., Editors, 1980, pp. 24-33.
Tonnesen, J., and Hansen, P.
K.,
“Some Experiments on the Steady State
Characteristics of a Cylinderical Fluid-Film Bearing Considering Thermal
Effects,” ASME J. Lubr. Technol., 1981, Vol. 103, pp. 107-1 14.
Ferron, J., Frene, J., and Boncompain,
R.,
“A Study
of
the
Thermohydrodynamic Performance of a Plain Journal Bearing, Comparison
Between Theory and Experiments,” ASME J. Lubr. Technol., 1983, Vol.
105,
pp. 422428.
Boncompain, R., Fillon, M., and Frene, J., “Analysis of Thermal Effects
in
Hydrodynamic Bearings,” J. Tribol., 1986, Vol.
108,
pp. 219-224.
1948, Vol. A197, pp. 201-216.
20
Chapter

Crook,
A.
W., “The Lubrication of Rollers,” Phil. Trans. Roy. Soc., Lond.,
1963,
Vol.
A255,
p.
281.
Cheng, H.S., “A Refined Solution to the Thermal Elastohydrodynamic
Lubrication of Rolling and Sliding Cylinders,”
ASLE
Trans.,
1965,
Vol.
8,
pp.
397410.
Johnson,
K.
L., and Cameron, R., “Shear Behavior of Elastohydrodynamic Oil
Films at High Rolling Contact Pressures,” Proc. Inst. Mech. Engrs,
1967-68,
Vol.
182,
Pt.
1,
No.
14.
Dowson, D., and Whitaker, A. V.,
“A

pp.
186-194.
Benedict, G.
H.,
and Kelley, B. W., “Instaneous Coefficients of Gear Tooth
Friction,” ASLE Trans.,
1961,
Vol.
4,
pp.
59-70.
Misharin,
J.
A., “Influence
of
the Friction Conditions on the Magnitude of the
Friction Coefficient in the Case of Rolling with Sliding,” International
Conference on Gearing, Proceedings, Sept.
1958.
Hirst, W
.,
and Moore, A.
J.,
“Non-Newtonian Behavior in Elasto-hydrody-
namic Lubrication,” Proc. Roy. Soc.,
1974,
Vol.
A337,
pp.
101-121.

Contacts for Two Synthesized Hydrocarbon Fluids,” ASLE Trans.,
19?,
Vol.
Winer,
W.
O.,
“Regimes
of
Traction
in
Concentrated Contact Lubrication,”
Trans. ASME,
1982,
Vol.
104,
p.
382.
Sasaki, T., Okamura,
K.,
and Isogal,
R.,
“Fundamental Research on Gear
Lubrication,” Bull. JSME,
1961,
Vol.
4(14),
p.
382.
266(
1

Conditions of Simultaneous Rolling and Sliding
of
Bodies,” Wear, 1968,
Vol. 11, p. 291.
Kelley, B. W., and Lemaski, A.J., “Lubrication
of
Involute Gearing,” Proc.
Inst. Mech. Engrs, 1967-68, Vol. 182, Pt. 3A, p. 173.
Dowson, D., “Elastohydrodynamic Lubrication,” Interdisciplinary Approach
to the Lubrication of Concentrated Contacts, Special Publication No. NASA-
SP-237, National Aeronautics and Space Administration, Washington, D.C.,
1970, p.
34.
Wilson, W.
R.
D., and Sheu,
S.,
“Effect
of
Inlet Shear Heating Due to Sliding
on EHD Film Thickness,” ASME J. Lubr. Technol., Apr. 1983, Vol.
105,
p. 187.
Greenwood, J. A., and Tripp, J. H., “The Elastic Contact of Rough Spheres,”
J. Appl. Mech., March 1967, p. 153.
Lindberg, R. A., “Processes and Materials
of
Manufacture,” Allyn and Bacon,
1977, pp. 628-637.
“Wear Control Handbook”, ASME, 1980.

Concentrated Normal Load on the Boundary
of
a Semi-Infinite
Solid
The fundamental problem in the field of surface mechanics is that
of
a
concentrated, normal force
P
acting on the boundary
of
a semi-infinite
body as shown in Fig. 2.1. The solution of the problem was given by
Boussinesq
[I]
as:
22
The Contact Between Smooth Surfaces
23
2
Figure
2.1
Concentrated load on a semi-infinite elastic solid.
a,
=
horizontal stress at any point
I
=
P((1
2n

v
=
Poisson’s ratio
The resultant principal stress passes through the origin and has a magnitude:
3P

3P
=
4-
=
249
+
22)
-
2nd2
The displacements produced in the semi-infinite solid can be calculated
from:
24
Chapter
2
U
=
horizontal displacement
and
w
=
vertical displacement
where
E
=

Uniform Pressure over a Circular Area on the Surface of a
Semi-Infinite Solid
The solution for this case can be obtained from the solution for the con-
centrated load by superposition. When a uniform pressure
q
is distributed
over a circular area of radius
a
(as shown in Fig.
2.2)
the stresses and
deflections are found to be:
(w)~=(,
=
deflection at the boundary
of
the loaded circle
-
4(
1
-
u2)qa
-
JrE
(w)~=”
=
deflection at the center
of
the loaded circle
-

5
(or
-
oz)r=o
=
maximum shear stress at any point on the Z-axis
From the above equation
it
can be
shown
that the maximum combined
shear stress occurs at
a
point
given
by:
and its value is:
26
Chapter
2
Case
3:
Uniform Pressure over
a
Rectangular Area on the Surface of
a
Semi-Infinite
Solid
In this case (Fig. 2.3) the average deflection under the uniform pressure
q

(6
=
a)
the maximum and minimum deflections are given by:
Figure
2.3
Uniform
pressure over a rectangular area.
The Contact Between Smooth Surfaces
Table
2.1
Values
of
Factor
k
1
1.5
2
3
5
10
100
0.95
0.94
0.92
0.88
0.82
0.71
0.37
27

=
radius
of
the cylinder
The pressure distribution under the cylinder is given by
which indicates that the maximum pressure occurs at the boundary
(Y
=
a)
where localized yielding is expected. The minimum pressure occurs at the
center
of
the contact area
(r
=
0)
and has half the value
of
the average
pressure.
Case
5:
Two
Spherical Bodies in Contact
In this case (Fig.
2.5)
the area of contact
is
circular with radius
a

111
_-
+-
E,
El
E2
E,,
E2
=
modulus
of
elasticity for the two materials
Case
6:
Two
Cylindrical Bodies in Contact
The area
of
contact in this case (Fig.
2.6)
is
a rectangle with width
b
and
length equal to the length
of
the cylinders. The design relationships in this
case are:
q
=