Proceedings of WTC2005

World Tribology Congress III

September 12-16, 2005, Washington, D.C., USA

WTC2005-63205

BEARING PERFORMANCE LIMITS WITH GREASE LUBRICATION

ABSTRACT
This paper has examined the influence of bearing design
and operation in controlling lubricant supply to the contact
zone. Grease lubricated contacts are liable to starvation and as
a result the film thickness is reduced, this can result in surface
damage or premature bearing failure. It is of obvious
importance to know when starvation occurs and the effect of
grease type, bearing design and operation on lubrication
replenishment. The aim therefore is to develop a starvation
parameter capable of predicting the operating limits for a
particular bearing/grease system.
A number of bearing design parameters are examined in
the paper, these include cage design, ball spin and bearing size.
Ball spin and cage effects can be efficient mechanisms for
maintaining the lubricant supply to the track. Increased bearing
size, line contact geometries and high load result in reduced
lubricant replenishment of the contact. Using this analysis it
will be possible to establish operating limits for families of
bearings
INTRODUCTION
The operation and life of rolling element bearings is highly
dependent on the lubricant performance. Both friction and
surface damage will be influenced by the properties and
thickness of the lubricant layer. The critical contacts in such
bearings operate under Elastohydrodynamic lubrication (EHL)
conditions. In this case film thickness is determined by
lubricant properties, bearing geometry and operating conditions
and can be predicted assuming fully flooded conditions [1].
However the actual film thickness can be much less than
predicted due to reduced availability of lubricant, this is the
starved regime and is often encountered with grease lubrication
[2]. In contrast to the established film thickness rules applying
to the fully flooded condition, the starved regime dependence is
less well understood. It is of obvious importance to know
under what conditions the transition from the fully flooded
regime to the starved regime occurs for a particular lubricant
and application. A typical film thickness/speed starvation
curve is shown in Figure 1. The speed at which this transition

Lubrecht, A.A., Laboratoire de Mcanique des

Contacts et des Solides UMR-CNRS 5514, INSA de
Lyon, France

occurs is identified as the starvation speed and this depends on

the lubricant properties, operating conditions and lubricant
layer thickness in the inlet [3]. The most important factor
determining lubrication level in bearings is the lubricant supply
to the contact zone. The traditional view has been that oil
bleeds from the grease reservoir to replenish the rolling track.
However this is too simplistic and ignores the role of
lubrication mechanisms specific to the bearing.
1000

Film thickness [nm]

Cann, P.M., Tribology, Department of Mechanical

Engineering, Imperial College London, UK

30C

58C

10
Starvation speed

0.01

0.10
Speed [m/s]

1.00

Figure 1 Starvation curve for two viscosities 0.59 pas (at 30C)
and 0.16 pas (at 58C)
Examination of used bearings has provided useful insights
into lubrication mechanisms. In these tests [4] the bearings
were dismantled and the lubricant distribution and chemical
composition assessed. Most of the grease was present on the
seals with a small amount in the cage pockets and a thin
lubricant layer on the raceways. This implied that a limited
amount of lubricant was available to the contact and that
efficient replenishment mechanisms are required to maintain an
adequate film thickness level.

Copyright 2005 by ASME

u
entrainment speed,
h oil oil layer height,

DEVELOPMENT OF LUBRICATION MODEL

In an earlier paper [3] a dimensionless parameter (SD)
was developed to define the fully flooded/ starved transition.
The relationship between film thickness and oil properties was
studied in an optical EHL device. Four parameters were varied:
oil volume, speed, contact dimensions (load) and viscosity., all
of which influence starvation speed and maxim film thickness
but in a different fashion. These results were replotted (Figure
2) as a relative film thickness (starved/fully flooded) against a
dimensionless starvation parameter (SD). The SD parameter
study was limited to a simple test device and base oil. The
challenge is now to extend the analysis to include both bearing
and grease parameters.
Fully flooded

1
Relative film

1000
Cage
100

Starved

SD =

0.4
0.2
0

0.5

0ua
h oil s
1

1.5

SD

2.5

3.5

0.01

0.10
Rolling speed [m/s]

1.00

Figure 3 Effect of a cage on starvation speed with grease

0.8
0.6

No cage

10

1.2

at the sides. The gap between the rolling element and the cage
also provides a site for shearing the grease and a reservoir of
lubricant for replenishment. The design of the cage will
obviously have strong influence on both of these roles.
Experimental measurements of film thickness in a modelbearing device have shown that the closely conforming cages
provide the most efficient relubrication. An example is shown
in Figure 3, which compares film thickness/speed behaviour
with and without a cage present.

Film thickness [nm]

NOMENCLATURE
0 viscosity,
a contact width,
s surface tension.

Figure 2 Relative film thickness (starved/fully flooded) as a

function of the SD parameter
The bearing parameters considered in this analysis
include size, type, clearance and cage design. The operating
parameters are speed, load, and temperature. Bearing
dimension affects lubrication level as it determines the distance
from lubricant reservoir at edge of track to the contact centre.
As such it influences the replenishment of the rolling track, thus
the maximum allowable speed is reduced for larger bearings.
Bearing type is determined by the rolling element
geometry such as rollers, needles, tapered and spherical that
will all give elongated contacts. Such geometries do not
facilitate replenishment because of the distance required.
Spherical rolling elements on the other hand have slightly
elongated or even circular contacts, which favour contact
replenishment. Furthermore such spherical bodies can spin,
causing a minor displacement of the rolling track; this
movement provides a substantial lubricant supply to the
contact. Tests with a model-bearing device have shown that
there is a significant increase in film thickness when spin is
imposed on the rolling element
Cage design will also influence lubricant redistribution
within a bearing; this can be both beneficial and detrimental.
In the worst case the cage acts as a scraper removing lubricant
from the already depleted track. In best case the cage
redistributes the lubricant supplying the track from the reservoir

CONCLUSIONS
This paper has examined the influence of bearing design
and operation in controlling lubricant supply to the contact
zone. The conclusions are as follows:
1. Efficient relubrication is the key to higher speed
operation and extended bearing life.
2. The design of the bearing plays a critical role in
determining operating limits of lubricating grease.
- Ball spin and cage effects can be efficient
mechanisms for maintaining the lubricant supply
to the track.
- Increased bearing size, line contact geometries
and high load result in reduced lubricant
replenishment of the contact.
3. Using this analysis it will be possible to establish
operating limits for families of bearings.
ACKNOWLEDGMENTS
Dr Cann would like to thank the Royal Swedish Academy
of Engineering Sciences and the Jacob Wallenberg Foundation
for the award of a research grant in support of this work.
REFERENCES
1. Dowson, D. and Higginson, G.R., Elastohydrodynamic
Lubrication, The Fundamentals of Roller and Gear
Lubrication, Pergamon Press Oxford, UK, (1966)

2. Cann, P.M., Trib Trans, 42, pp 867-873 (1999)

3. Cann, P.M., Damiens, B. and Lubrecht, A.A., Trib. Int.,
37, pp 859-864, (2004)
4. Cann, P.M., Doner, J.P, Webster, M.N. Lugt, P. and
Wikstrom, V., presentation at WTC2005, (2005)

Copyright 2005 by ASME