TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE) and
Jerrold W Kannel Battelle Columbus, Ohio

Editorial: STLE was very fortunate in obtaining permission from Dr Sheldon R Simon, Manager of B-TIP (Battelle
Technical Inputs to Planning) of Battelle, in Columbus, Ohio, to republish their report No. 26 entitled “Tribology: The
Science of Combating
Wear”. This 56-page report was copyrighted in 1981 but is now out of print. The contents are applicable today. The
report presents the basics of lubrication, friction and wear, in a very readable form with excellent sketches,
photographs, tables, graphs and information for practical use. The authors are William A Glaeser, Richard C Erickson,
Jerrold W Kannel, and Keith F Dufrane. Their biographies are presented in this issue. Mr Erickson is no longer at
Battelle. These authors are from the Tribology section of Battelle, which continues to be one of the nation’s best
Tribology research centers. Battelle contributes significantly to the solution of lubrications and wear problems in
industry.

The Editorial Committee has divided the report into ten sections, and will publish them in the “Back to Basics” section
of Lubrication Engineering. The whole report may be offered later as an STLE Special Publications. The Introduction
is presented as Part I in this issue.

We can all be grateful to Battelle for sharing this valuable report with the members of STLE. Mr Simon wrote,
“because of the educational nature of your endeavour, we are happy to co-operate.” The writing is a valuable
contribution to the Society’s obligation to disseminate Tribological information. The material will be most useful to
newcomers to the field, but also as a review for others, specially in related disciplines, such as additive chemistry.

There may be some duplication of previous basic tribology subjects, but we believe that the Battelle report is unique
and comprehensive, and that some repetition is justified.

The following is the “Preface” from the original report.

Douglas Godfrey
Contributing Editor

Many industries have long been plagued by profit-devouring problems resulting from wear: producing lines screech to
a halt when equipment malfunctions, maintenance costs soar, and output quotas are not met. In fact, wear ruins
hundreds of millions of dollars worth of equipment and products each year.
To curtail these losses, industry can use effective techniques resulting from tribology – the science of friction,
lubrication, and wear. By integrating fluid and machine dynamics, metallurgy, physical and surface chemistry, heat
transfer and stress analysis, tribology offers a systematic, broad based approach to wear control. Though still
developing, tribology already can save millions of dollars by upgrading design and protection of critical machine
elements – e.g., bearings, gears, cams and traction drives for such components as internal combustion engines, turbines
and other commonly used devices.

Ideally, tribology techniques should be applied in the equipment design stage. Here, service life can be predicted by
using recently developed computer models that can simulate operating conditions. Designers are better able to refine
geometries, select suitable lubricants and develop improved lubrication supply methods. These techniques can also be
beneficially applied during equipment operation for reducing noise, alleviating shock or cyclic loading, identifying
lubrication problems, and developing efficient maintenance programs – thus extending service life.

But to use these techniques effectively, companies may need to initiate a program to track wear costs. While a few
companies, particularly in Japan, are now instituting such programs, data gathering worldwide is sparse. With hard
data in hand, companies will be able to determine priorities for best applying their wear-control dollars. And, as the
science progresses, technological advances and better wear tracking methods will help companies improve equipment
and product resistance to wear, thus enhancing productivity in using materials and energy.
To date, most tribological research has focused on lubrication, primarily because it has been the simplest, most
available, and least expensive wear control method. Consequently, in this report, we will concentrate on the main
lubrication systems – hydrodynamic, elastohydrodynamic, and boundary: what principles govern their use, and where
and how they can be applied.

PART I – INTRODUCTION

Tribology: a modern science ……

Within recent years, the drive to minimize wear has focused on analysing and predicting the mechanisms that occur
between interacting surfaces in relative motion. Rigorous, basic studies begun only twenty years ago have evolved into
a dynamic science devoted to the study of lubrication, friction, and wear: tribology. A relatively new word in the
technical vocabulary – derived from the Greek root “tribein”, meaning “to rub” – the term “tribology was coined in
England in the mid 1960’s.

Today, by applying what has already been learned from basic research in tribology, wear control in our modern
industrial environment can be significantly advanced – and real cost savings ca be achieved. According to a 1966
report published by the British Lubrication Engineering Working Group, the application of established lubrication
principles in industry could save as much as $800 million (1) annually in Great Britain through reduced energy
consumption and machinery repair alone. And in the U.S., industry spends $2.8billion (2) a year in purchasing and
regrinding cutting tools – not including labour costs and machinery downtime for worn tool replacement. More recent
studies in the U.S. and elsewhere – prompted by product durability concerns and attempts to conserve energy and
materials - have resulted in similar conclusions.

Also, during the past two decades, the drive to develop advanced propulsion systems – e.g., high performance gas
turbines, nuclear engines and booster rockets – accelerated the need for more sophisticated lubrication, friction, and
wear technology. The nuclear and aerospace industries, in particular, faced critical problems – with ultra high-speed
bearings, gears, and seals operating at high temperature, and with erosion of supersonic air foils – that could not be
solved by conventional lubrication, bearing, or materials approaches to wear control. As a result, advanced research
conducted for these systems led to new principles, and to new lubricants, designs and materials that vastly enlarged the
existing store of tribological information.

…. responding to a pervasive need

An array of machine elements – without which most manufacturing processes and many products would screech to a
halt – depend on tribological principles and techniques to achieve acceptable performance levels and service life.
Among these are:
• Journal bearings
• Thrust bearings
• Rolling element bearings
• Ball bearings
• Gears
• Cams
• Traction drives
• Hypoid differentials
• Oscillating bearings
• Face seals

Incorporated into such vital components as internal combustion engines, turbines, compressors, and transmissions,
these elements can be found in a broad spectrum of industrial process machinery, transportation and aerospace
vehicles, and energy related equipment.
What tribology is
Tribology is a multidisciplinary field based on fluid and machine dynamics, metallurgy, physical and surface
chemistry, heat transfer, and stress analysis. The reason: contact between sliding or rolling surfaces creates highly
complex reactions that can only be understood by drawing from a range of disciplines. To better appreciate this
complexity, consider the next diagram, which depicts the constituents of sliding contact.

Various lubricants have long been used, almost casually, to prevent friction and wear between surfaces. However, we
now know it really isn’t that simple. Movement of the surfaces causes shearing within the lubricant, producing friction
and heat that are rejected through the lubricant and bearing, or other parts. Such heating also causes a reaction between
the machine element’s surfaces and the environment, resulting in a reaction product. Soluble phosphates in lubricant
additives, for example, will produce phosphate films on steel surfaces during sliding friction. Furthermore, fluctuations
in friction can lean to instability, vibration and noise. For example, surface-to-surface contact during start-up produces
high contact stresses at a few small high spots, or asperities, on the surface. Adhesion, a form of wear, can occur at
these contact points, causing momentary localised heavy flow of surface particles as the junctions shear.

Recognising the significance of all these processes, tribologists have incorporated them into a system capable of
controlling wear and predicting equipment performance.

In this systematic approach, critical performance factors that may cause a wear problem are first measured and
evaluated. Then, to dramatically reduce the detrimental effects of surface interaction – thereby enhancing service life
and reliability – more complex strategies are applied, based on:
• In-depth knowledge of the basic mechanisms of lubrication so that all relevant factors can be integrated into
the systematic approach
• Development of lubricants that perform under certain conditions of temperature, stress and pressure, and in
particular environments for the required service period
• Refinement of designs and geometries that minimize friction and wear, as well as the amount of lubrication
needed for components and assemblies
• Selection of materials that withstand a range of operating conditions

In essence, the science of tribology can be applied to

• Enhance understanding of what occurs between contacting surfaces in relative motion
• Optimise equipment performance through advanced lubrication technology – and, ultimately,
• Reduce wear and energy consumption.

Tribology systematics for industrial use

Since lubrication is the most available, understood, and frequently used key to wear reduction, the major thrust of this
report will be to describe advances in lubrication technology. It also forms the basis used here for classifying
tribological systems.

Processes involved in sliding contact

As body 1 approaches body 2 under a loading condition, the approaching surface and the space between them become
a closely interrelated system influencing the way one body slides over another. As sliding starts, any fluid lubricant
will shear and the viscous response to surface discontinuities (steps, tilt and taper) will produce pressure in the fluid. If
the pressure is sufficient to balance the applied load, sliding will occur with no solid contact between the surfaces.
Increasing the load will cause the fluid film to decrease in thickness and cause high spots on the surfaces to contact.

Tribological lubrication systems can be viewed – and, in part, differentiated – in terms of the amount of separation
between “contacting” surfaces of machine elements. In this classification scheme, the lubrication systems are:
• Hydrodynamic
• Elastohydrodynamic
• Boundary

As seen in the graph, a fourth condition – mixed film lubrication – can also occur.
When fluid dynamic processes are capable of causing complete separation of moving surfaces, hydrodynamic
conditions exist; and friction is proportional to viscosity and speed, and inversely proportional to the load.

As shown in the preceding graph, film thickness decreases with bearing pressure. For metallic bearing materials elastic
deformation begins to play a role in film support at about 10,000 psi. Lubricant viscosity is affected by pressure,
increasing with the increasing pressure. Thus film thickness becomes a function of the spreading of the contact area by
elastic deformation, and the increase in lubricant viscosity with increasing pressure. These conditions are called
“elastohydrodynamic”.

When the lubricating film becomes extremely thin, high spots on the surface break through the film and contact high
spots on the opposing surface. Here, contact pressures are very high and surface contact will occur principally on a
very thin coating of surface reaction products. If the coating has sufficient penetration resistance, it will prevent
intimate surface contact and adhesion, a form of wear. Boundary lubrication occurs under these conditions; and film
thickness and friction no longer are influenced by fluid viscosity.

Here, we will briefly look at these principal lubrication systems and their impact on friction and wear reduction when
applied to surfaces in rolling or sliding contact. In later sections of this report, each system – its purposes, principles,
industrial uses, and important technological advances – will be presented in more detail.

Hydrodynamic Lubrication
Hydrodynamic conditions exist when a viscous oil film completely separates moving surfaces within machine
elements – as in well operating journal bearings. Under these conditions, direct surface-to-surface contact does not
take place, and thus wear does not occur.

For elements operating with hydrodynamic lubrication, mathematical calculations have been formulated to predict
accurately
• Fluid film thickness
• Frictional losses
• Heat generation
• Safe operating limits of the lubrication system.

• Hydrodynamic – where surfaces are completely separated by a viscous film.

Although these idea conditions may no always be maintained throughout the entire equipment operating cycle,
hydrodynamic analysis, nonetheless, can yield workable solutions to a broad group of practical lubrication and wear
problems in internal combustion engines, turbines, compressors, and transmissions.

Elastohydrodynamic Lubrication

Commonly known as EHD, elastohydrodynamic lubrication also pertains to sliding surfaces, but more often involves
rolling surfaces separated by an oil film. The mechanisms for oil film formation in EHD lubrication are very much like
those in hydrodynamic lubrication; however, in elastohydrodynamic lubrication, the interface region between the
moving parts elastically deforms under the contact pressure. This deformation creates larger oil film areas, and
therefore, greater load carrying ability than that predicted by hydrodynamic theory. Elastohydrodynamics films are
very thin and require polished surfaces to prevent asperity contact.

Here too, mathematical calculation can be applied to predict film thickness, frictional losses, heat generation, and safe
operating limits of the lubrication system.

Elastohydrodynamic lubrication for wear control is typically used in rolling element bearings, gears, cams and traction
drives in such equipment as jet engines, precision gyroscopes for navigation, centrifugal blowers, and transmissions.

Elasohydrodynamic-where elastic deformation at the contact

interface allows greater load carrying ability.

Boundary Lubrication
For slowly moving elements under heavy load, hydrodynamic and elastohydrodynamic lubricant pressures are often
insufficient to separate surfaces completely. In these cases, surfaces can be protected from excessive friction and wear
by absorbed films created by chemical interactions between the surfaces and the lubricant. If this boundary film is
sufficiently thick, it will also prevent adhesion due to direct surface-to-surface contact. In this process – known as
boundary lubrication – lubricant thickness, friction, and heat generation predictions are empirically based.

Boundary lubrication reduces friction and wear in high-load, low-speed applications and permits start-up or shutdown
of hydrodynamic bearings when speeds are too slow to generate adequate films.

• Boundary-where adsorbed films protect contact

Surfaces exposed to show speeds or high loads.

Mixed Film Lubrication

Apart from hydrodynamic, elastohydrodynamic, and boundary lubrication, another condition, known as “mixed film”
lubrication may occur. Mixed film lubrication, as the term implies, is actually a combination of boundary and
hydrodynamic or elastohydrodynamic modes. This intermediate state is attained when the oil film becomes very thin,
causing asperities to penetrate the film and contact the opposing surface – i.e., the load is partially supported by a fluid
film, and partially by solid contact.
 • Mixed film lubrication-an intermediate condition-occurs
In hydrodynamic or elastohydrodynamic bearings wear.

This condition is created in hydrodynamic or elastohydrodynamic bearings when

• Operating speed is too low
• Load is too high
• Lubricant viscosity is too low

Consequently, mixed film lubrication results during:

• Start-up
• Slow-down to stop
• Overloading
• Overheating

Operation in this mode typically causes bearings to wear; yet realistically, this state cannot be totally avoided.
Consequently, bearing structural design must allow for operation during mixed film as well as during boundary film
conditions. In addition to meticulous geometrical design, careful selection of materials can also help to compensate for
operation in this mode.

In this report, mixed film lubrication will not be treated separately, but only as it relates to the total lubrication and
wear process.

The table found on the following pages lists the principal lubrication systems, and some important factors that
influence their use.

References:
(1) “Wear Control to Achieve Product Durability Workshop”, U.S. Office of Technology Assessment, Washington, D>C>,
February 23 through 25, 1976
(2) Ibid.

Reprinted with the permission of Dr Sheldon R Simon, Manager of Battelle Technical Inputs to Planning Program, Battelle,
Columbus, Ohio
TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE) and
Jerrold W Kannel Battelle Columbus, Ohio

Reprinted with permission of Dr Sheldon R Simon, Manager, Battelle Technical Inputs to Planning Program, Battelle,
Columbus, Ohio

PART 2 – FRICTION AND WEAR

FRICTION
The role of friction

Friction – which is actually the resistance to sliding of two contacting surfaces – is encountered in all machine
elements. It must be controlled in order to either:
• Minimise energy losses in mechanical systems – e.g., in internal combustion engines, or to
• Aid system operation by keeping the friction level high and constant – e.g. in brakes or wheel-rail traction
systems.

Its relation to wear

During operation of most tribological systems, friction, wear and noise are normally undesirable and good design
practices attempt to reduce or control these conditions. In some cases, reducing or controlling one will also alleviate or
eliminate another. For example, by controlling friction, wear and noise may also be reduced; controlling wear may
result in lower energy losses from friction. On the other hand, increased noise and friction in a bearing is usually
followed by excessive wear.

Friction and wear are related; both are products of solid contact between moving surfaces. But since friction is caused
by a somewhat different set of mechanisms than wear, devices can be designed for low wear and high friction or for
low wear and low friction.

How friction is produced

Actual solid-to-solid contact occurs only between high points in the microtopography of each surface. As these small
contact areas shear, friction results. Shearing may involve plowing of hard particles through a softer material or
scraping of soft surface layers. Thus, the real area of contact is only a fraction of the total surface. By increasing the
load, the real contact area is also increased, which in turn, increases friction force. (The coefficient of friction,
however, remains constant.)

Determining friction levels

In hydrodynamic and elastohydrodynamic bearings, where surfaces are separated by a viscous lubricant film, friction
levels can be estimated by calculating the amount of energy required to shear the film. In solid-sliding-on-solid
systems, however, where the basic friction process is not yet well understood, friction levels are more difficult to
predict. Under this type of contact, or during boundary lubrication conditions, friction levels must be determined by
experimentation.

How it is measured and predicted

Friction levels are expressed in terms of the friction coefficient, which is the ratio of frictional force and the normal
load between two contacting elements – or
Friction coefficient is measured by determining the tangential force resisting the sliding of two contacting bodies under
a known constant load. Thus, friction force – which resists sliding motion and requires a translation force to overcome
– generally is proportional to the load, while the friction coefficient is independent of load and total surface area.

Influence on friction
The amount of friction is a function of the area or volume being displaced, work hardened mechanical properties of the
near surface region, and surface composition. Although mechanical properties of material influence friction behaviour,
surface composition also is a significant factor, and can be effected by
• Environment – For the same material, the coefficient of friction for steel can be 0.2 in air but can reach 1.0 or
more in a hard vacuum.
• Temperature – The phenomenon of brake fade or reduction in gripping power of brakes during high speed
stops is the result of surface changes caused by frictional heating.
• Type and thickness of coatings – When a soft material like tin is coated on a steel surface, friction is reduced.
For example, a tin coating of just 0.0001 inch will substantially minimise the friction level.

Friction coefficient ranges

Expected friction coefficient ranges for various contact conditions can be summarised as follows:

Contact Hydro- Elastohydro- Boundary Dry Ultra clean

condition dynamic dynamic lubrication and surfaces in high
lubrication lubrication some plastics vacuum

Friction .001 to 0.1 .0005 to .003 .02 to .3 .2 to .8 1.0 to 100

coefficient
range

Energy loss in shearing a hydrodynamic oil film can also be expressed in terms of friction coefficient. While rolling
contact conditions do not allow for a totally “frictionless” bearing during EHD lubrication, very low friction is
produced, and arises from the inherent sliding action that also occurs between rolling elements.

In solid-sliding-on-solid systems, surfaces exposed to air become covered with thin oxide films, adsorbed gases, and
hydrocarbons. Though these “natural” films provide some reduction in sliding fiction via boundary lubrication,
“deliberate” boundary lubrication can further reduce friction by as much as an order of magnitude. Equipment
designers and operators have a relatively wide range of friction level control.

For example, in data on 230 measurements of friction for a wide variety of “dry” sliding systems, 80 percent of the
friction coefficients were between 0.4 and 0.6. Unlike wear coefficients which cover several orders of magnitude,
“dry” sliding friction coefficients measured in air cover a much smaller range. However, when measured in high
vacuum, the range is similarly dramatic. (1).
In the following sections, we will talk more about friction and the part it plays in tribological systems.
Growing impact and value of tribology
We have, indeed, pressed far beyond the era where the almost casual use of a lubricant was sufficient to alleviate wear
problems. In a report from the Hanniker Conference on National Materials Policy, wear is listed as one of three major
technical areas affecting materials utilisation and cost reductions. As a direct result of these and other studies, several
U.S. government agencies, and a number of universities and technology institutes worldwide are in the process of
establishing centres for co-ordinating and increasing practical applications of tribological principles to energy and
materials conservation in industry. And several research institutes including Battelle, are working intensively in this
area – both in the in the basic science of tribology and its practical applications. A universal conclusion can be drawn
from these studies. With current technology alone, significant savings can be achieved by applying accumulated
tribology know-how to control wear in today’s industrial processes and products.

WEAR
Impact of wear
Wear is a familiar phenomenon: tyres wear out, faucet gaskets deteriorate, lawn mower blades become dull, carpeting
wears down – many examples come to mind. In this context, wear is taken for granted; and its effect on the useful life
of our possessions is accepted.

In industry, however, the economic implications of wear are severe. In fact, according to recent estimates, billions of
dollars per year could be saved if the life of components could be increased.

Here is a particularly illustrative example of the detrimental effects – and waste – caused by wear. Wire rope in
dragline earth moving equipment used in strip mining must often be replaced after only a few months in operation. Re-
roping large 100 to 150 yard dragline is hard work, time-consuming and expensive – about $100,000 for the rope and
$30,000 in labour.

As shown in Fig. 1, wire notching reduces rope diameter. When the rope diameter is reduced by a given percentage,
the rope is taken out of service before it breaks.

Here, wearing away a relatively tiny amount of material – possibly two ounces of steel – makes it necessary to discard
nearly a ton of wire rope! (Used rope, unfortunately, has little scrap value). Other industrial examples abound, where
a small amount of wear destroys the usefulness of a critical mechanical component.

Fig 1 – Notching in wire rope worn during dragline operations.

Cost of wear
We intuitively recognise that any wear which reduces equipment precision and efficiency will be reflected as financial
loss. Bus as a prerequisite to analytically dealing with its economic effects, a value must be assigned to wear or its
costs must be documented. Unless wear can be quantified, a company cannot determine the level or extent to which it
is economically attractive to reduce wear costs. Yet, little has actually been done in this area. The published literature,
in fact, deals mainly with wear costs in a global sense, rather than on the company – or even on the industry – level.
In 1976, however, a major attempt to quantify wear costs was undertaken during a workshop on “Wear Control to
Achieve Product Durability”’ sponsored by the U.S. Office of Technology Assessment. (2). Here, tribology experts
and representatives from various industries concluded that very little effort on any level was being directed toward
monitoring wear costs. As a result, they attempted to estimate wear costs for six specific areas: metal cutting
machinery and tools, heavy construction equipment, automobiles, railroad equipment, naval aircraft structures, and
aircraft propulsion equipment. Some results are listed below:

Estimated Wear Costs

Category Costs*
Naval aircraft $243.87 per flight hour
Naval aircraft litres $1,853,200 per year
Naval ships $38.92 per hour
Cutting tool wear $9000,000,000 per year
Auto Maintenance and $40,000,000,000 per year
repair costs
*1976 dollars

These costs included:

• Replacement costs due to wear
• Scheduled maintenance costs
• Unscheduled maintenance or repair costs.

On the company level

General estimates of industry-wide costs and savings, however, provide little insight for individual companies
attempting to calculate wear costs. This lack of specific information is usually compensated for by early discard or
overdesign of components. On the other hand, a few standard techniques are available for conducting economic
appraisals that would aid in evaluating wear costs. One of the most common of these is life cycle costing. However,
few companies actually use this method; and those that do, tend to apply it inconsistently, reserving it only for key
phases in decision making.

A second appraisal technique employs the National Academy of Corrosion Engineers’ standards for corrosion
economics. (3).

Companies attempting to estimate wear can look at costs from three major viewpoints:
• Capital
• Labour
• Technology

Capital considerations can minimally include equipment replacement costs, and maximally, replacement costs plus all
downtime costs related to that replacement. Wear costs can thus be divided into direct and indirect costs. Direct costs
entail:
• Equipment replacement
• Equipment rental
• Standby maintenance.

Indirect costs include those incurred as a result of:

• Loss of production
• Product liability
• Incompleted work
• Damage to or loss of additional equipment.
Labour costs also should be considered in assessing wear costs. Maintenance programs, for example, often are labour
intensive. Even if only a few pieces of equipment require attention, a base level of labour is needed to perform that
maintenance. As long as interest rates remain high and short-term depreciation is allowed, long-term maintenance and
wear reduction will be measured in comparison to those labour costs.

Technological obsolescence reflects yet a third area of consideration in assessing wear costs. The cost of keeping wear
to a minimum on older equipment will be balanced against technological change for that equipment.

While it is true that most companies have not taken advantage of wear control benefits offered by tribology, one
notable exception is the British Steel Corporation. In the late 1960’s, this company established a tribology section
which cost approximately $200,000 during its first two years of operation – but produced a net savings of
approximately $4 million in hot mill roll production costs alone!

Similarly, the use of chemical vapour deposition (CVD) in the production of carbide cutting tools in the U.S. during the
same time period improved tool life by as much as 10 times. In an address to the wear control workshop, R.F.
Bunshah* estimated that the development of the CVD program cost $1-2 million over a three year period, but saved
$7.5 billion in machining costs and $30 million in tool costs (2).

(*”The Problem of Wear in Metal Cutting” Proceedings from Workshop on Wear Control to Achieve Products
Durability, OTS, U.S. Congress, 1976.)

Surface topography of wear Sub-surface deformation

Associated with wear

Fig. 2 – Surface and Sub-sirface defpr,atopm diromg wear.

Here we have highlighted several principle areas where costs are incurred, and identified two techniques for conducting
economic appraisals of wear. Clearly, companies will increasingly need to look to wear reduction as a direct,
immediate avenue for maintaining output quotas and for cutting production costs.

But to effectively cut costs of wear, we need to know how wear occurs.

Mechanisms of wear
Even though wear is a familiar occurrence and its impacts are well known, the mechanisms that cause wear are not yet
well understood. However, accumulated research results are steadily increasing our understanding of these
mechanisms and enhancing our ability to control wear.
When two surfaces rub together, the generation and release of wear debris is a complex process involving
• Surface chemistry
• Fracture mechanics
• Elastic and plastic deformation
• Heat generation
• Transfer
• Adhesion.

Because of the sequence of events that produces wear debris occurs at and below the interface of contacting asperities,
direct observation of the process has been virtually impossible. Most of what is known about wear has been inferred
by closely examining the products of wear:
• Wear scars
• Debris
• Microstructural alterations
• Noise
• Thermal effects
• Hardness changes

For instance, microscopic examination of wear scars has shown that most wear processes involve scratching and
plowing of the surface, with considerable plastic deformation as shown above in Fig.2. A material’s plastic strain
behaviour, therefore, is considered to be related to its wear properties.

But wear cannot be accurately predicted solely via some basic material property or group of properties. While
considerable research has been and continues to be directed at how various material properties influence wear, the
simplest approach thus far has been to use hardness as a rough indicator of wear resistance. Other properties that
appear to exert an influence include:
• Fracture toughness
• Elastic modulus
• Stacking fault energy*
• Strain hardening coefficient
• Thermal shock resistance

References
(1) Rabinowicz, E., “Friction – Especially Low Friction”, Fundamentals of Tribology, Proceedings of the International
Conference on the Fundamentals of Tribology, Edited by N.P Suh and N.Saka, the MIT Press, Cambridge, Massachusetts,
(1980).
(2) (2) “Wear Control to Achieve Product Durability Workshop”, U.S. Office of Technology Assessment, Washington,
D>C>, February 23 through 25, 1976.
(3) Economic Effects of Metallic Corrosion in the United States – A Report to NBS by Battelle Columbus Laboratories, U.S.
Depart. Of Commerce, (1979).

*Stacking fault energy (SFE) is a measure of the ease by which alterations are formed in the stacking sequence between crystal
planes, and influences the way in which plastic deformation proceeds.
TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE) and
Jerrold W Kannel Battelle Columbus, Ohio

Reprinted with permission of Dr Sheldon R Simon, Manager, Battelle Technical Inputs to Planning Program, Battelle,
Columbus, Ohio

PART 3 – WEAR (continued)

Predicting and analysing wear

Efforts in these directions are showing some success. Archard’s equation (7) is a basic technique that is used quite
often for predicting and analysing wear – but applies only to bearings, gears, and other sliding systems, and not to
rolling systems.

V = KWL
H
Where
V = Volume of wear, cubic inches
W= normal load, pounds
L= sliding distance, feet
H= penetration hardness of the softer material in the pair, psi
K= wear coefficient

The equation predicts that wear is a linear function of sliding distance and load, which agrees with general experience.
It enables the hardness characteristics of materials in one system to be compared to those in a control system. Instead
of having to run a test system for a given time in order to determine the wear rate, designers and failure analysts can
apply data from empirical studies on the control system to the equation.

For example, if several types of materials were being considered for use in a bearing, each type could be evaluated on a
standard wear tester to determine a wear coefficient. The wear coefficient is then useful for estimating service life
under a given loading and amount of sliding. The degree of tolerable wear must be determined; and from there,
predictions on component service life can be made, using Archard’s equation.

The equation, however, has many limitations. For example, it does not account for material properties other than
hardness and does not predict phenomena such as initial rapid wear during “wear-in”, and the dramatic increase in wear
rate with a relatively small increase in load. Also, it does not account for the effect of speed, lubrication, or contact
pressure* on sliding systems.

In addition, hardness is not a universal criterion for wear rate. When a metal is work hardened, for example, the
improvement in wear is negligible. Hardening steel by heat treatment, e.g., by induction heating, flame hardening, or
laser treatment – improves wear resistance, but not as significantly as changing to another material, such as chromium
plating or tungsten carbide.

An example of where a harder material has replaced a softer material: the heavily loaded bearings supporting the
control surfaces of aircraft. Bearings for the flaps, tail section, and some linkages in the airframe are now being
constructed of beryllium copper, which is not only harder, but also more corrosion-resistant than are traditional
aluminium bronze bearings.

*Some standardized contact pressure data have been compiled in the ASME Wear Control Handbook, 1980.
Frictional heating, or heat from other sources, can change the hardness of a sliding surface with unpredictable wear
results. Thus, using wear coefficients obtained by wear tests or from handbooks can cause misleading results unless
the limitations of Archard’s Equation are recognised.

But, even with these limitations, the formula is useful if the other influencing effects are recognised, and if
experimentally derived data are judiciously applied to compensate for limitations.

Wear Rate

Hardness

Fig: 1 – Effect of hardness on wear resistance of various metals and alloys.

Type of wear

Wear is often difficult to classify; yet identifying the type of wear that is occurring is a key factor in wear control. A
great deal of experience is generally required to analyse surface topography created by wear, and to determine if wear
is being produced by a chance encounter or by recurring contact.

Several recognised types of wear can be found in machinery:

• abrasive wear
• adhesive wear
• erosion
• fretting
• contact fatigue
• corrosive wear

Those that can be described by Archard’s equation are abrasive and adhesive wear; and those that cannot are erosion,
fretting, contact fatigue, and corrosive wear. Each type has a different mechanism, cause and effect.

Once the type of wear has been identified, and the cause pinpointed, control methods can then be implemented. These
methods can involve:
• lubrication technology
• materials substitution
• load reduction
• removal of impact conditions
Abrasive wear

In machinery, wear occurs most frequently as an abrasive process. We have pointed out that contacting surfaces really
touch at only a few high points – there are no atomically flat engineering surfaces! The high contact stresses developed
at these small contact areas cause localised plastic deformation. However, since the surfaces are moving tangentially,
this deformation results in microscopic grooves. The movement of the asperity over the opposing surface resembles
the action of a blunt tool with a negative rake angle trying to cut a surface. In fact, abrasive wear is much like material
removal by grinding. The scratch pattern produced by abrasive wear is shown in Fig. 2. Abrasive wear can be caused
by hard particles caught between sliding surfaces or by dragging metal parts over soil, concrete, or other rough areas.
In cases such as these, and especially if the abrasive particles are sharp, an occasional particle will become lodged so
that it cuts the surface much like a sharp tool. The result is curled, chip like wear debris.

Referring again to our mining industry example, we can picture a dragline bucket being worn by sliding contact as it is
pulled through soil and rocks. When abrasion is severe enough to scrape holes through the bucket, field repair often
requires welding plates over worn areas.

Abrasive wear also occurs under lubricated conditions – especially during boundary lubrication.

Adhesive wear
Adhesive wear entails metal transfer, which, to some extent, is a part of all wear. Even under lubricated conditions,
metal transfer is possible on a very limited scale. Minor adhesive wear can be seen, for example, as a slight
discoloration of a shaft that runs in a bronze journal bearing.

Fig. 2 - Photomicrograph of surface features caused by abrasive wear

Where metal transfers from one sliding surface to another in visible lumps,
adhesive wear predominates. An example of adhesive wear is shown below (fig 3.)

Fig. 3 - Photomicrograph of scar from adhesive wear.

Metal transfer in adhesive wear does not occur as chunks plucked out of one surface and attached to the other during
cold welding. Rather, it most often occurs as a build-up or agglomeration of small particles released from one surface
and transferred to the other surface in layers. The built-up edge on a cutting tool is an example of adhesive wear.

Under lubricated conditions, interruption of the lubricant film causing metal-to-metal contact can produce metal
transfer, adhesion, and surface damage. Piston ring scuffing is an example of adhesive wear during lubrication. The
metal transfer that results causes deep scratching and surface damage, leading to loss of compression. Aluminium in
the pistons is particularly sensitive to this process because it resits the formation of effective boundary lubrications
films.

Polymers

In some applications, polymeric materials can be used in place of metals. For example, they are being used
increasingly as piston rings in dry gas compressors where metal particle contamination cannot be tolerated. Any
number of low speed, low load devices are also using polymer gears, bearings and sliding surfaces. Polymers are also
appearing in medical applications where stainless steel sockets for hip joint prostheses are now being replaced by ultra-
high density polyethylene.

Polymers offer a number of advantages over metals. For example, they generally cost less to produce, weigh less,
resist corrosion, and distribute the load – thus resisting wear more effectively than metals because they have better
elastic deformation properties under contact pressure.

On the other hand, adhesive wear is usually more prevalent in polymers than in lubricated metal systems. Also,
polymeric materials have limited load-bearing capacity; and their low thermal conductivity and elastic modulus causes
them to over-heat and deform. These limitations thus tend to restrict their use – for the time being – to lighter weight,
more controlled applications. However, the wear resistance and strength of polymers can be enhanced by adding
ceramic or graphite fibres as reinforcements. In fact, composites of polymers, metal fibres and minerals are now being
used to fabricate such components as brakes.

Erosion

Erosive wear results from particle impingement or cavitation shocks against a surface, but not from the trapping of
loose particles between sliding contact surfaces. Wear of a dragline bucket pulled through gravel is abrasive; but wear
of an impeller moving through a slurry is erosive.

There are three types of erosion – recognition of each is important for determining appropriate erosion control:
• Solid particle erosion – surface wear by impingement of particles carried by a gas or fluid. Example: wear of
helicopter blade leading edges in dusty environments.
• Liquid drop erosion – surface wear by impingement of liquid drops.
Example: wear of centrifugal gas compressor blades by condensate droplets.
• Cavitation erosion – surface wear in a flowing liquid by the generation and implosive collapse of gas bubbles.
• Examples: Cavitation erosion of a ship’s propellers, and of components in fluid pumps.

Erosion is particularly insidious because it tends to concentrate its effect, and can cause damage quickly with little
warning – i.e. a hole punched through the elbow of a pipe carrying a slurry or gas suspension of particles.

Erosion is difficult to predict, and therefore, difficult to control. Because it is not as pervasive as abrasion, and more
confined to specific situations, it has not received the amount of scientific enquiry as has abrasive wear. Since fewer
studies have been undertaken to identify erosion causes and control methods, less general application experience is
available to draw on.
However, limited testing of material hardness has been conducted. Results reveal that when steels of various hardness
levels are tested in an erosion tester using silica as an eroder, their wear rate is not influenced by their hardness. A
comparison of the effects of hardness on erosion (Fig. 4) with the effects of hardness on other types of wear (Fig 1)
illustrates the difference in material behaviour.

Fig. 4 – Effect of steel hardness on erosion rate.

Eroding medium: Sand

Thus, the traditional methods of hardening materials by heat treatment or by adding hard particles to the microstructure
also do not affect erosion resistance. On the contrary, the more homogeneous a material is, the better the erosion
resistance.

The angle at which particles approach a surface is also important for wear control since there is a “critical impingement
angle” at which the maximum erosion rate occurs. In addition, the critical angle changes with velocity for some
materials making prediction of erosion severity difficult.

Though particle size and velocity will influence erosion rates, materials show varying sensitivity to these parameters.
For example, ductile materials have a different critical impingement angle from brittle materials.

When corrosive conditions are present, a combined erosion-corrosion wear process will occur. In this case, the
protective surface reaction products from corrosion are removed by erosion, thereby accelerating both processes and
producing a very rapid material loss. This condition creates severe problems that are particularly evident in coal
gasification and liquefaction plants. The critical problems arising here have added further incentive for developing
more refined erosion control strategies. Work in erosion control is progressing, much of it directed toward identifying
materials that will effectively withstand erosive environments.

References
(1) Archard, J.F., “Contact and Rubbing of Flat Surfaces,” Jour. Of Applied Physics, 24, p 981, (1953)

Fig. 5 – Erosion ripple pattern caused by sand impinging at a low angle on a rod from a sandblasting unit.
TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE) and
Jerrold W Kannel Battelle Columbus, Ohio

Reprinted with permission of Dr Sheldon R Simon, Manager, Battelle Technical Inputs to Planning Program, Battelle,
Columbus, Ohio

PART 4 – WEAR (Continued)

Fretting

Fretting occurs when two contacting surfaces rub during low amplitude oscillatory motion. This “scrubbing” action
produces fine debris, which generally oxidizes. A classic example: the fretting of steel or iron which produces a red
powder that exudes from the contact area. The debris developed by fretting can also cause mechanical or electrical
failure by fouling close-fitting parts, valves, and electrical contacts.

Although fretting does not cause rapid loss of material, it generally occurs in situations where no wear is anticipated
and, as a result, can initiate serious damage and failure. For example, flywheels that are force-fit onto crankshafts will
sometimes fret, with the loss of material eventually causing the wheel to loosen on the shaft. These and other “press
fits” are prone to fretting – especially under periodic load shift conditions.

The most serious problem associated with fretting is fatigue. The fretting process induces surface defects, which,
under cyclic loading, reduces the fatigue life and strength of a part. Thus, a part designed for safe life under known
fatigue loading can fail if exposed to fretting. Fatigue failure of turbine blades at the point of attachment to the wheel
is the most common example of fretting fatigue.

Fig. 1 Photograph of spall due to contact fatigue on a bearing race.

Contact fatigue

Ball and roller bearings and gears are subject to a surface pitting condition known as contact fatigue.

The contact fatigue process that results in pitting involves the high contact stress field developed during rolling contact.
This stress field interacts with surface and subsurface defects and hard inclusions to initiate cracking. The cracks
propagate below the surface and finally turn and emerge at the surface, producing a pit. The first indication of this
condition is rough running or noisy equipment operation. By this time, spalling (fig. 1) has often brought about
complete surface destruction, and fracture may be imminent. This condition usually does not occur until the bearing or
gear has been operating for a long period.
Production of very clean bearing steel by special vacuum melting processes has greatly reduced the incidences of
fatigue failures of bearings in many applications. Also, applying elastohydrodynamic design principles to the
manufacture and use of ball bearings has helped greatly to reduce this problem. Considerable research has been
directed toward analyzing the complex microstructural phenomena that occur in steel alloys subjected to rolling contact
stress.

Corrosive wear
During corrosion, the chemical attack forms a reaction product which usually does not adhere well to the surface. It
tends to break up easily in the sliding process, pulling off debris as well as uncorroded or abraded material.

Rubbing the surface removes protective films, increases surface activation, and lowers the energy required to initiate
surface reactions. And as it has already been noted, the wear process can be significantly accelerated when corrosion
occurs in an erosive environment. Thus, the general effect of combining corrosion and erosion is accelerated material
removal.

In a few special cases, the reaction product may form a surface diffusion barrier that inhibits growth of the film. The
severity of the entire corrosion and wear process then becomes self-limiting because the barrier slows down corrosion
by separating the reactants.

One such case: fuel rods in nuclear reactors are made of a zirconium alloy that forms a hard, wear-resistant oxide. The
oxide forms a self-replenishing barrier that protects the zirconium from wear.

Mechanical devices operating in marine environments or in seawater often experience corrosive wear. In one example,
a seawater pump for a two-occupant submersible ocean floor explorer developed a seal leak, permitting seawater to
mix with the lubricant. The sea-water-contaminated oil dissolved the copper from the motor windings. The copper
then plated out on the steel bearing surfaces, bringing on a galvanic corrosion process and, ultimately, bearing failure
(Fig.2.)

While this is merely on example of equipment failure, it is representative of what can occur in any device when a
corrosive fluid combines with a lubricant.

Fig. 2 – Corrosion of an outer rae from a seawater pump bearing contaminated by sea water

Theories of wear
In order to effectively predict and control wear, examining the theoretical causes and mechanisms of wear is needed.
Several theories have evolved, each contributing to better understanding and control of wear problems.

Adhesion theory: Briefly, this early theory contended that surface asperities contact at widely separated points.
Because actual contact occurs on only a small portion of the total surface, cold welding occurs at asperity contacts.
Tangential movement shears these junctions, causing the sheared asperities to come off as debris*.
While this theory was generally accepted for about 40 years, it has serious shortcomings. For example, it did not
account for the
• Wear-in process
• Large increase in wear resulting from even a small increase in load.

*Archard’s equation discussed in Part III, is based on the adhesive wear theory.

Also, when investigators began inspecting wear surfaces with the scanning electron microscope, they found results that
could not be explained by the adhesion theory – for example, lose debris and other wear phenomena that could not be
produced by adhesion. And so, a number of new theories have emerged within the past decade.

Diffusion theory: This theory, which has been effectively applied to wear of machine tools, claims that cratering of a
hard tool surface is caused by the very chips the tools is removing from the work piece. High frictional temperature
caused by chip contact with the tool surface theoretically induce diffusion of material from the tool into the chip.
Because a large surface area of chip material passes over the tool surface, material transfer on the atomic scale will
eventually wear away the tool surface.

In practice, selection of tool surface coatings based on materials with low diffusion and low solubility in iron has
dramatically improved tool cratering wear resistance. But still there are serious questions regarding the diffusion
theory – principally because calculated diffusion rates are much too slow to produce the wear rates observed, and the
required concentration gradient* does not exist.

Some researchers suggest that thermal decomposition of the tool surface material is a more likely process because
“good” materials used for anti-wear coatings also have high decomposition temperatures.

Delamination theory: Another theory, initially named delamination, is based on the concept of accumulation of near
surface plastic strain. The wear process is known to cause intense plastic deformation of a thin layer of the surface
(10 µm to 100 µm deep). This layer can be seen with electron microscopy by careful sectioning through wear scars. A
typical section is shown in Fig.3.
 Fig. 3 – Photomicrograph of deformation zones from the wear process.

Three deformation zones are clearly visible:

• Texturing
• Fragmentation
• Transfer

The transfer layer resides on the surface and is composed of fine particulate matter produced by asperity adhesion and
wear debris. The fragmented zone shows intense deformation: the familiar grain structure disappears and is replaced
by a region of maximum dislocation density taking the form of sub grains or cells. This zone has received considerable
research attention because of the assumption that cracks nucleate and propagate here, causing particles to break out
from the surface.

This concept explains how a piece of wear debris can be separated from a surface, while adhesive wear theories, on the
other hand, do not adequately predict separation of sheared asperities.

Though the physical process is not fully known, heavy deformation subsurface crack nucleation and propagation with
the release of flake-like debris is proposed as the process. Estimates of strain rates occurring during sliding contact
deformation indicate that subsurface fracture can occur by two, or possibly three, mechanisms:
• Cleavage
• Ductile fracture
• Intergranular fracture

The third mechanism is less likely because the material in which the cracks propagate no longer contains true grain
boundaries.

Delamination wear in which subsurface cracks propagate to the surface is not observed in pure metals and simple
alloys like bronze and carbon steel. Its occurrence is more likely in hardened alloys containing hard, non-metallic
inclusions.

The value of wear theories

The new wear theories provide more sophisticated approaches to wear control, and go beyond simply trying to make a
wear surface harder. These concepts have significant implications for wear control. For instance:
• Knowing the fracture characteristics of a material at strain rate associated with wear, scientists can apply
established approaches to inhibit crack initiation or to slow crack propagation rates.
• If a material is expected to exhibit cleavage-type fracture at a given surface temperature and contact strain
rate, fracture toughness properties would be important in considering methods for improving wear resistance.
• Ductile fractures might be reduced by considering concentrations of grain boundary inclusions in the
fragmented layer.
• More attention is now being paid to the microstructure as well as to the mechanical properties of metals
selected for wear resistance.

Wear testing and application of wear test data to real problems can now be more effective as tests are designed around
basic principles derived from new wear theory. The authors will explore these principles – and their significant to wear
control – in the succeeding sections. As mentioned previously, the main emphasis will be on applying lubrication
technology for wear reduction.
*the concentration gradient is the difference in concentration of diffusing constituents. The diffusion moves in the direction from
high concentration to low concentration.
TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE)
and Jerrold W Kannel Battelle Columbus, Ohio

Reprinted with permission of Dr Sheldon R Simon, Manager, Battelle Technical Inputs to Planning Program,
Battelle, Columbus, Ohio

PART 5 – HYDRODYNAMIC LUBRICATION HDL

In hydrodynamic lubrication – often referred to as “full film lubrication” – a film of oil, or other fluid, completely
separates moving components under load. Ideally, the fluid prevents solid-to-solid contact, thus eliminating wear
and maintaining low friction. If these ideal conditions could be indefinitely maintained, component life could be
extremely long.

Realistically, this perfect state cannot be sustained continuously during the service life of most devices – for
example, during start-up and stopping the relative sliding velocities of hydrodynamic bearings are inadequate to
form a full fluid film. However, today, new knowledge gained from advances in lubrication and bearing
technology, lubricants, materials, coatings and other critical components, coupled with data compiled from
extensive past research, enables designers to achieve full film lubrication during a high percentage of the service
cycle of many commonly used industrial equipment. Among these are:
• Internal combustion engines
• Turbines
• Compressors
• Transmissions

These machines depend on hydrodynamic bearings for proper functioning. Primarily, these bearings include:
• Journal or plain bearings
• Thrust bearings
• Porous bearings
• Bushings
• Gas lubricated bearings
• Piston rings
• Liquid metal bearings

Later, the authors will outline how hydrodynamic principles can be used to improve bearing performance, what
factors govern their operation, what lubrication techniques can be applied, and where these bearings can be most
effectively used.

Principles governing hydrodynamic lubrication

Before it is seen how HDL is applied, there needs to be an understanding of some of the factors that form the basis
of technology. These factors include:
• Oil film formation
• Viscosity
• Velocity
• Heat generation

Oil film formation

In HDL, a lubricating oil flows into a converging cavity or wedge that is formed by the geometry of the
moving parts. The lubricant forms a film between the members, preventing surface-to-surface contact.
The film’s characteristics – e.g., thickness and pressure – are determined by the viscosity and tenacity –
of the lubricant, the geometry of the moving surfaces, their relative velocity, and the load they are
supporting. (Fig. 1, shown below).

Fig: 1 – Wedge formed by converging geometrics of sliding components.

Viscosity

Since HDL moving components are completely separated by an oil film, oil viscosity plays a critical role
in determining equipment performance.

In an operating machine, if other relevant factors such as speed, load and temperature are not changed,
increasing the viscosity of a lubricant will increase the thickness of the lubricant film. Oil film thickness is
pro-proportional to the square root of viscosity.

However, fluid viscosity is also heavily dependent upon temperature. The effect of temperature on
viscosity is illustrated in Fig. 2.

The extent by which viscosity decreases with increases in temperature is described as “viscosity index”
(IV). Oils whose viscosity decreases dramatically between 37.8 and 98.9°C (100 and 210°F) are
assigned a viscosity index near zero while those with exhibit small changes in viscosity are assigned a
VI near 100.

Oils consisting of long-chain hydrocarbon molecules have a higher viscosity than those comprised of
short-chain molecules. Similarly, adding high molecular weight polymers to a base oil will increase its
viscosity at both low and high temperatures. However, the effect of these additives is greater at high
temperatures and results in an increase in VI. Such additives are called VI improvers. These and other
additives are routinely included in a broad spectrum of modern lubricating oils. Those listed in the chart
below are merely a sampling of additives on the market today; but they illustrate the ability of additives
to heighten lubricant effectiveness and enhance mechanical performance. (Table 1)
TABLE 1 – COMMONLY USED LUBRICATING OIL ADDITIVES
Lubricant Type of additive Purpose of additive
Turbine oil Oxidation and rust inhibitor Extends service life

Hydraulic fluid Rust inhibitor Prevents rust of ferrous alloys

Circulating oil Antifoam compound Minimizes foaming

Electric motor and spindle oils Anti-wear Reduces wear

Gear oils Anti-scuff Chemical action reduces scuffing

Worm gear oils Polar-fatty materials Reduces friction

(Lubricity)

Heavy duty motor, diesel Pour point depressant Promotes easy starting in cold weather
crankcase oils, and portable
compressor crankcase oil Viscosity index improver Reduces viscosity at low temperature
Increases viscosity at high temperature
Detergent-dispersant Oxidation- Maintains engine cleanliness
corrosion and antiwear Resists oil oxidation; prevents
corrosion; and extends life of cylinders,
rings and cams

Antifoam compound Minimizes foaming

Rolling element bearing greases Soaps produced from calcium, Gelling agents to provide consistency
lithium, and/or aluminum or synthetic to high dropping point
compounds and solids

Oxidation inhibitor Resists oxidation and molecular

change
Rust inhibitor
Prevents rusting in wet conditions
Metal deactivators
Prevents catalytic effect of metals

Wether high or low viscosity is required depends on the application; however, adequate viscosity is needed to
develop hydrodynamic lift for supporting and separating sliding surfaces. A viscosity level that is too high will
create excessive heat, while one that is too low will allow surfaces to touch, and hence cause wear.

The user should select a lubricant with just the right viscosity to develop lift without generating excessive heat from
internal shearing.

Viscosity, velocity and heat generation

These three factors are closely interrelated. As can be seen in Fig. 1, an equal volume of oil enters the large end of
the wedge and exits the small end of the wedge. Continuity of volume dictates that the linear velocity of the oil is
greatest at the exit end. Furthermore, since the oil is a tenacious substance, its velocity at both surfaces is very
close to zero. Therefore, a steep velocity gradient must exist between the oil in the center of the oil film and that
near the surfaces of the sliding components. This velocity gradient develops a large amount of shear strain in the
lubricant, particularly near the exit end of the wedge.

The effect of shear on viscosity is significant. Viscosity defines the lubricants resistance to shearing; and the
energy required to overcome this resistance is converted in the lubricant into heat. The generation of heat causes
the temperature of the oil film to increase. This higher temperature reduces the viscosity of the oil to a value where
subsequent frictional energy is equal to the heat dissipated to the environment and no further temperature increase
occurs within the lubricant.
Therefore, in certain applications, substituting a less viscous lubricant may cause the oil to operate at a lower
temperature, thus reducing the thermal degradation of the lubricant and possibly increasing the life of other
components in the system. For example, in an automotive internal combustio engine, use of lower viscosity oil can
result in lower viscous shear losses, causing less heat generation, and yielding higher mileage per gallon. However,
the lubricant must have sufficient viscosity to develop adequate hydrodynamic pressures and oil film thickness to
separate sliding components under all operating conditions expected during the equipment’s service life.

Where hydrodynamic principles can be applied in industry

When properly applied, hydrodynamic principles – which include lubricant properties, component design,
dynamics and operating conditions – can effectively help:
• Reduce fluid friction between components moving at high speed under high loads
• Reduce vibration and noise
• Minimize maintenance costs
• Extend – and predict – service life

Numerous types of industrial equipment depend on hydrodynamically lubricated components for optimum
performance. For example:

Internal combustion engines. Widely used in manufacturing and processing equipment, as well as in
automobiles, trucks and other vehicles, these engines contain journal bearings and piston rings.

Turbines and compressors. Typically required to operate steadily for long periods of time – in power plants, for
example – these machines depend on hydrodynamics to maintain low friction and provide high durability.

Transmissions. Hydrodynamically lubricated bushings and thrust washers for automatic transmissions support
rotating journals.

Bearings: fundamental hydrodynamic components

Following are the major types of bearings that depend on hydrodynamic lubrication, their lubrication needs, and the
factors that influence their performance.

Journal bearings
Journal bearings, the most widely used type which depend on HDL – can be found in such diverse applications as
clock motors and steam turbine generators. Journal bearings consist of a cylindrical shell with a rotating shaft or
journal extending through the bore. A journal bearing illustration is shown in Fig. 3.

Fig. 3 – Full of oil film formation in a journal bearing.

a. The journal rests at the lower part of the bearing under the applied load or the forces of gravity.
b. As the journal slowly rotates clockwise, friction between the contacting surfaces causes it to move to the
right. During this slow speed stage, boundary lubrication protects the two surfaces from excessive wear. A
wedge is then formed on the right side of the bearing
c. As oil is drawn into this wedge, hydrodynamic pressure begins to develop. While the journal speed
continues to increase, the lubricant is distributed.
d. When sufficient film pressure and velocity are created to overcome the applied load, the journal lifts off the
bearing surface.
The inside diameter of journal bearings is about 0.2 percent larger than the diameter of the journal, resulting in a
clearance cavity where the hydrodynamic wedge is formed.

To achieve effective HDL the film should be at least twice as thick as the sum of the height of asperities on the
surface of the journal and bearing. Oil film thickness and pressure, (Fig.4), can be computed from data on the oil
viscosity, applied load, journal rotational speed, and journal and bearing dimensions.

Fig: 4 Axial pressure distribution in a Fig. 5. Journal bearing. Lubricant enters the
journal bearing. bearing in an unloaded region and is carried
through the loaded region by the rotating
journal.

Another major consideration in designing a journal bearing system is the method by which the bearing is to be
lubricated (Fig.5). Several common methods are described in Fig.6.

Fig.6 – Lubrication techniques

Pressurized feeding - This technique is often used for bearings that operate under heavy loads and high speeds.
Under these conditions, the amount of heat generated in the lubricant may be quite high. In pressurized feeding, a
large volume of oil can pass through the bearing, transferring heat via some form of heat exchanger. This method
also assures an ample supply of oil during start-up, continuous operation, and stopping. Further, oil filtration prior
to bearing entry is simplified. In medium and large gasoline and diesel engines, a circulating pump delivers
pressurized oil to journal bearings and valve train components.

Oil wick feeding - A low-maintenance lubrication technique for bearings used in mild service applications, this
method consists of packing oil-soaked-felt or spun yarn around the outside of the bearing. A square hole or
window in the unloaded region of the bearing wall allows oil from the packing to enter and exit the bearing.
Periodic circulation of lubricant through the wicking material provides a mild degree of filtration. This oil is
absorbed back into the wick where it is stored when the bearing is not in service.
Dip feeding - This low-cost method is employed where an oil sump can be incorporated into the design.
Three variations of the dip feeding technique are: ring, disc and splash feeding.

Ring feeding - consists of fitting a ring or chain into a section in the unloaded portion of the bearing wall so that
the ring is in contact with the journal. The ring should be large enough to dip into a lubricant sump below the
bearing. As the journal rotates, the ring also rotates, carrying oil from the sump and depositing it on the journal.

Disc feeding - also referred to as the “oil collar technique”, employs a similar principle, but rather than a ring, a
disc or flange, projects radially from the journal and dips into the oil sump. In this case, two bearings support the
journal: one on each side of the disc. Wipers scrape the oil off the disc and direct it to the bearings.
Both the ring and disc feeding methods are limited by rotational speed, since centrifugal forces at high speeds tend
to sling oil out radially rather than deposit it on the journal.

Splash feeding also offers a relatively economical approach to lubricant delivery. It involves throwing oil at the
bearing by the motion of some form of striker through the oil sump. Once oil is delivered, grooves distribute it
through the bearing. The geometry of the groove patterns depends on the specific application. Smaller engines –
such as those typically used in lawn and garden equipment and small recreational vehicles – often use a splash feed
method to lubricate their moving parts.

Drip feeding - Fractional and integral horsepower electric motors often employ drip feed lubrication to assure that
an adequate supply of oil is available to each bearing. The reservoir of such units must be refilled at regular
maintenance intervals – e.g. in motors for forced air furnace blowers. Failure to replenish the oil in this family of
motors may result in damage to the bearings.
TRIBOLOGY: THE SCIENCE OF COMBATTING WEAR

By William A Glaeser (Member, STLE), Richard C Erickson (Member, STLE), Keith F Dufrane (Member, STLE) and
Jerrold W Kannel Battelle Columbus, Ohio

Reprinted with permission of Dr Sheldon R Simon, Manager, Battelle Technical Inputs to Planning Program, Battelle,
Columbus, Ohio

PART 6 – HYDRODYNAMIC LUBRICATION (Continued)

Automobile tires

A smooth automobile tire in contact with a wet road surface is somewhat analogous to a journal bearing system with an
extremely large clearance. The region of the tire just ahead of the contact forms a hydrodynamic wedge. If a sufficient
quantity of water accumulates in this region and the speed of the tire is high enough, a hydrodynamic water film may
develop with lifts the tire off the road and causes a dramatic loss of traction called hydroplaning. Fortunately, modern
treat designs and road grooving methods interrupt the hydrodynamic wedge, and impede the formation of
hydrodynamic film pressure.

Thrust bearings
These bearings are used to support an axial load on a rotating shaft or plate. Aircraft engines, turbine generators, and
most machines generally contain at least one thrust bearing.

The illustration in Fig. 1 shows the contacting face of a typical tapered pad thrust bearing system. In this drawing, 8
tapered pads and 2 circumferential oil grooves are shown; but other bearings may be flat, have 3 or more pads, and
contain fewer or more oil grooves. The hydrodynamic film pressure distribution for one pad from this bearing is
shown in Fig. 1.

Fig. 1 – Oil film pressure distribution in a tapered pad thrust bearing.

The range of operating conditions of a thrust bearing can sometimes be extended by supporting the pads on pivots.
These are called tilting pad thrust bearings.

For low-speed operating conditions – i.e. 50 to 700 rpm – thrust pads can be mounted on a series of preloaded springs.
These spring-mounted thrust pad bearings are used in applications where they can be immersed in lubricant to simplify
lubricant delivery and cooling, e.g. in very large machinery such as hydroelectric power generation turbines.
With proper lubrication, both the centrally pivot pad and spring-mounted thrust pad can be designed to carry high loads
and perform over a wide range of operating conditions, including both forward and reverse rotation.

Oil impregnated porous bearings

Produced by powder metallurgy, pores of the bearing are impregnated with lubricating oil prior to installation. Their
porosity serves as a built-in wick. For example, when the journal begins to rotate, oil is drawn out of the pores, passed
through the hydrodynamic wedge, and returned to the bearing walls by hydrodynamic pressure and capillary forces,
where it is filtered and heat is rejected. This cyclic action offers a low cost, low maintenance bearing system for such
equipment as sub fractional horsepower motors used, for example, in light duty exhaust fans.

Gas lubricated bearings

The use of a gas as a lubricant is sometimes referred to as “aerodynamic lubrication”. Gas bearings represent a special
kind of hydrodynamic lubrication that offers three major advantages:
• The viscosity of compressed gas is less sensitive than liquids to temperature extremes. The viscosity of many
gases actually increases with temperature, enabling gas bearings to perform well in extremely hot o cold
environments.
• The lower viscosity of compressed gases yields very low frictional resistance, and hence, low power loss at
very high speeds and high temperatures.
• The inertness of several common gases reduces the danger of chemical attack by the lubricant.

However, a major cost disadvantage is incurred when gas bearings are used in applications which can also use oil
lubricated components. The tight clearance required to form a hydrodynamic wedge for such a low viscosity substance
requires extremely close tolerances, and therefore, high manufacturing costs. Consequently, the use of gas bearings is
restricted to applications where less expensive alternatives won’t function adequately – e.g. long life gyroscopes.

Foil bearings
Foil bearings are relatively inexpensive gas bearings which use a thin metal foil as the load bearing element. The foil
conforms to the shaft and allows very small bearing clearances.

Piston rings
Piston rings are flexible components which slide along the bore of an engine cylinder to seal combustion gases inside
the chamber and guide the piston through its stroke. As the piston travels through the cylinder the ring tilts as shown in
Fig. 2.

This tilting action creates a wedge for accumulating a thin layer of oil between the ring surface and the cylinder wall,
thus generating hydrodynamic life. Lifting protects the cylinder and the piston ring from wear, and significantly
reduces friction through most of the stroke. Modern design techniques and materials selection have further reduced
piston ring and cylinder liner wear, and therefore, have increased engine life to within acceptable limits for all but the
most severe service conditions.
 Fig; 2 – Piston and ring in the bore of an engine cylinder.
Benefits and limitations of hydrodynamic lubrication systems
Hydrodynamic bearing systems have specific advantages and disadvantages, especially when compared with
elastohydrodynamic devices, e.g. rolling element bearings.

What hydrodynamic systems offer compared to rolling element bearings

Low noise. Since an oil film separates the moving components in a hydrodynamic bearing, very little noise is
produced. This factor is important to the saleability of appliances and other devices intended for commercial,
residential, and military markets. Rolling element bearings, on the other hand, because they consist of several balls or
rollers that can vibrate, often create unacceptable levels of noise.
Size. The advantage of the compact geometry of hydrodynamic bearings is illustrated in Fig. 3. Rolling element
bearings require greater space to house the rolling elements and races.
Conformability. The ability of hydrodynamic bearing materials to conform to minor misalignments resulting either
from assembly or changes during service provides a favourable degree of forgivingness to the components – e.g., main
bearings in diesel engines. Rolling element bearings are available which provide self-aligning characteristics, though
other performance capabilities may be limited and costs increased.
Embeddablity. Embeddability – which often accompanies conformability – describes the ease by which foreign
particles may embed below the bearing surface in the soft overlay, thus reducing abrasive damage to the mating
surface. Since rolling element bearing steels are too hard to provide embeddability, they must be protected from
contaminants through the use of shields or filtered lubricants.

Fig: 3 – Six comparison between a hydrodynamic bearing and a ball bearing.

Long Life. During extended operation under monotonic loading, rolling element bearings will often fail due to contact
fatigue. However, hydrodynamic bearings are not subject to contact fatigue under monotonic loading conditions.
Consequently, extremely long service life can be expected from well-designed and maintained hydrodynamic bearings.
Journal bearings subjected to cyclic loading as in compressors, out of balance shafting, or internal combustion engines
are subject to fatigue.
Shock load resistance. Hydrodynamic oil films can adjust their thickness and pressure distribution in response to
shock loading. Therefore, hydrodynamic bearings can generally withstand shock loading for extended periods without
failure. The bearings in a reciprocating engine are constantly subjected to repeated cyclic loads which are transmitted
through an oil film. Since the components of a rolling element bearing are in closer contact with each other, shock
loads must be absorbed by the elasticity of the bearing material. Such loadings can, therefore, significantly accelerate
bearing deterioration.

Limitations of hydrodynamic bearings

Limited low speed capability. HDL films break down at low operating speeds. This leads to the onset of mixed
hydrodynamic and boundary film lubrication in which somewhat higher friction and wear rates prevail. Consequently,
rolling element bearings or bearings designed for boundary lubrication perform best in applications involving slow
speed operation.
Lubricant maintenance. The performance characteristics of HDL components are dependent on the condition of the
lubricant employed. Lubricant maintenance recommendations must be carefully observed to avoid loss of adequate oil
supply or viscosity, chemical changes or excessive contamination which affect performance. In certain applications,
rolling element bearings often require less frequent or less extensive lubricant maintenance for comparable
performance.

Costs
Initial costs. The cost of a complete hydrodynamic bearing system depends, of course, on the requirements of each
application. Among the factors to be considered are:
• Lubricant delivery and filtration system
• Bearing material, geometry, and surface finish
• Precision requirements
• Cooling

Lubricant delivery and filtration system. As it has been shown in Part 5, hydrodynamic systems can be lubricated
by several techniques – some rather simple and others more complex. Choosing the most appropriate approach
depends mainly on the operating requirements and environment of each application.

Bearing material, geometry, and surface finish. The type of material selected will have a direct impact on costs.
While some materials are expensive – e.g. tin, Teflon and graphite fibers – others are comparatively less expensive –
e.g., lead and low tin bronze. The need for high tolerances will also boost costs, as will intricate designs and very
smooth surface finishes.
Precision requirements. In certain applications – e.g., in high-speed bearings – a high level of precision is often
required. Fabrication procedures to achieve a more precise bearing could involve grinding, lapping or burnishing
operations and/or individual matching of components, thus creating higher manufacturing costs. From all these basic
considerations, initial cost projections can be compiled.
Maintenance and replacement costs. The performance of hydrodynamically lubricated components depends on the
lubricant used. Thus, to achieve full benefit from the lubricant the system must be constantly maintained to avoid loss
of adequate supply oil supply, reduction in viscosity, and to prevent chemical changes that affect performance.

Inadequate system monitoring and improper lubricant maintenance are probably the main causes of repair and
replacement expenditures. Fluid levels, oil filters, and various other critical parts need to be inspected regularly. If
neglected or inadequately checked, interrupted lubricant supply or contamination can lead to wear – and eventually to
component failure. Thus a situation can be created where failure of one small bearing can shut down an entire
operation, resulting in untold financial losses – not only in repair or replacement but in production and labor as well.
Editor’s note: Further reading on HDL, see:
(1) Fuller, D.D., Theory and Practice of Lubrication for Engineers, 2nd Ed., John Wiley and Sons (1986)
(2) (2) Booser, E.R. and Wilcock, D.F., ”Hydrodynamic Lubrication”, Lubr. Eng., 47, 8, pp645-647 (1991)