Handbook For Failure Analysis
Handbook For Failure Analysis
Handbook For Failure Analysis
Introduction
ROLLING ELEMENT FATIGUE has been recognized since the turn of the century, with fatigue life testing beginning in
the early 1900s by rolling bearing manufacturers. Because of the high stresses imposed on bearings, steel with a hard
matrix was required. However, the hard matrix, along with other factors to be discussed, meant that the matrix was very
sensitive to loading and other external conditions that could lead to incipient material cracking and thus to contact fatigue.
In the early tests, large scatter in fatigue life was experienced. This persists and indicates that close correlation of the
incipient cracks and running time should not be expected (Ref 1).
Internal stress raisers, such as nonmetallic inclusions or carbides, are the beginning sites for incipient cracking. Because
of the varying severity, the magnitude of the stress raisers is difficult to calculate, so that rolling fatigue, basically a
problem of stress concentrations, displays the scatter mentioned or shows a clear statistical nature.
The high stresses involved come about from the geometry of bearings, which produces a concentration of load in the
contacts in a small volume of the material within a bearing (Ref 2). Stresses were first computed based on the basic
bearing internal geometries, and the stress distribution within and beneath the contact area was determined using the basic
equations proposed by Hertz (Ref 3) in 1881. From such fundamental work, along with considerable fatigue test results
that established various fatigue and material "constants," the first bearing ratings and fundamental life theories were
proposed.
Following the work of Grubin in 1949 (Ref 4), Dowson and Higginson in 1966 (Ref 5), and Dowson and Toyoda in 1978
(Ref 6), to name just a few, the important role of lubricant in the contacts of bearings was next understood. It was
recognized that the type of lubricant and the thickness of the lubricant film impacted bearing fatigue life and even
influenced modes of failure. Considerable work has been done to relate failure modes, including fatigue, to bearing life.
As knowledge in all these areas grew, it was possible to understand the requirements for a bearing to develop longer life
and produce acceptable life in a variety of adverse environments, such as low speed, high speed, elevated temperatures,
misalignment, and contamination. This work, drawing heavily on much experimental data, is the basis for the present
level of bearing life prediction.
References
1. C.A. MOYER AND E.V. ZARETSKY, FAILURE MODES RELATED TO BEARING LIFE, STLE LIFE
FACTORS FOR ROLLING BEARINGS, E.V. ZARETSKY, ED., STLE SP-34, 1992, P 47-69
2. E.S. ROWLAND, RESISTANCE OF MATERIALS TO ROLLING LOADS, CHAP 6, HANDBOOK OF
MECHANICAL WEAR, C. LIPSON AND L.V. COLWELL, ED., UNIVERSITY OF MICHIGAN PRESS,
1961, P 108-130
3. H. HERTZ, GESAMMELTE WERKE, VOL 1, LEIBZIG, 1895, ENGLISH TRANSLATION,
MISCELLANEOUS PAPERS, MACMILLAN CO., LONDON, 1896
4. A.N. GRUBIN AND I.E. VINOGRADOVA, INVESTIGATION OF THE CONTACT OF MACHINE