Extending bearing service life

​Reducing overhaul periodicity can have significant economic benefits and reduce the failure risk from unnecessary intervention. So not surprisingly several of our rail customers have approached ESR Technology to ask for advice on extending rolling element bearing (REB) life beyond that of the original design or overhaul specification.

The focus of life extension has been on wheelsets which are an important cost item in the industry. Powered wheelsets have axlebox bearings which require regular maintenance, and in many cases a final drive gearbox which can have an array of different type of rolling element bearings. In principal the approach developed by ESR can be applied to any fleet of assets where the reliability is determined by rolling element bearings.

To understand the problem of justifying life extension it is necessary to know why and how rolling element bearings fail. The contact pressure beneath the rollers can be extremely high, of the order of a GPa or more, see Figure. The repeated passage of the rollers through the loaded zone produces cyclic stress in the surface and eventually surface fatigue develops; the bearing is then said to have failed or ‘worn out’ when the first surface spall occurs. Thus, bearings fail by a fatigue damage process which accumulates over many millions of stress cycles. The initiation of fatigue cracks is a probabilistic process and in many ways is analogous to structural fatigue where the extent of the fatigue damage cannot be readily determined until a crack is formed.

Examining a sample of bearings taken from service to assess the condition of the whole population is the approach often used to justify extending overhaul periodicity. However, whilst it is important to know that there are no externally induced damage processes arising such as water or debris ingress, electrical damage, high vibration level, degradation of the lubricant, etc. a visual inspection cannot tell you what fatigue life is left in the bearings.

In the future magnetic non-destructive (NDT) methods may become available to quantify the accumulated fatigue damage, but such methods of damage characterisation are still experimental and not yet available commercially. Quantifying the risk of failure for a fleet of bearings can only to be done using statistical methods, and bearing failure is best described by the Weibull distribution. Weibull is a powerful technique to assess and quantify risk and is considered the leading method in the world for fitting life data. Normally a significant number of failures (10 or 12) would be needed in order to characterise the distribution, but because of the extensive test data available for REB’s, the slope of the Weibull curve is approximately known. This means that even with only a few failures the parameters of the Weibull distribution can be determined so the future risk can be quantified.

If the bearing condition and the failure risk of failure is acceptable then extending the replacement interval or overhaul period can be justified and the economic savings made. Conversely if adverse results are obtained there may be a case to remove the bearings early or investigate why the expected reliability is not being achieved.

For further information call Dr Jon Beard on 01925 843406 or email at This email address is being protected from spambots. You need JavaScript enabled to view it..