Rating life is a key factor – although not the only consideration – when choosing bearings. Over the last 50 years and more, hybrid units have been shown to outperform and outlive all-steel bearings in many situations. For instance, they have become an obvious choice for use with high-speed machine tool spindles.
Ceramic materials are particularly useful under demanding lubrication and contamination conditions. Their lower boundary-lubrication coefficient of friction enables greater efficiency when lubrication is poor, and they tend to exhibit lower operating temperatures.
They also have a higher resistance to surface damage from particulate matter and they avoid the possibility of steel-to-steel surface welding. In addition, they have other useful properties such as high electrical resistance and low weight.
The range of uses for hybrid bearings has expanded, partly due to the lowering of costs as manufacturing technology has advanced, but they remain a higher-priced option. Engineers considering their use need to know whether the extra cost will be justified by performance benefits.
Conventional bearing life modelling limitations
Traditionally, engineers enter information on their application’s expected loads and speeds into an equation which determines the bearing design’s rating life. The dynamic load rating (C) of a bearing principally indicates its performance in relation to sub-surface fatigue. Over millions of loading and unloading cycles, fatigue accumulates and the bearing eventually fails.
For hybrid bearings however, this bearing life model’s focus on sub-surface fatigue presents problems. Ceramic rolling elements are stiffer than steel ones, so they deform less under load.
The result is a concentration of loading over a smaller area, which increases stress and accelerates sub-surface fatigue. This can give an unfavourable impression which is often at odds with evidence of hybrid bearing performance in the field.
SKF’s real-world experience shows that most bearing failures in service are actually due to problems at the surface rather than in the body of the material. This is true for both steel and hybrid bearings.
Damage caused by poor lubrication or by contamination is usually the root cause. Modern standards such as ISO 281 attempt to accommodate these effects with correction factors but they still fail to reflect the real-life behaviour of bearings.
The new approach
To address these calculation deficiencies, a team at SKF was assembled in 2012 to develop a new model. The aim was to build on the existing sub-surface fatigue model but also embrace modelling for failure at the surface and use data from endurance tests.
The team’s research has drawn on decades of experience in materials science and tribology – the science of interacting surfaces in relative motion. Every aspect of the behaviour of bearing surfaces has been studied in detail, from their friction characteristics to the way they are indented by dirt particles under load.
Generating the data required to calibrate and validate the model involved testing hundreds of bearings, including both all-steel and hybrid types. Operating life was measured over a wide range of loads and surface conditions, and for bearings in poorly lubricated or contaminated conditions. To build behaviour curves, around 30 bearings needed to be tested for each point – with several of them failing in each case.
The outcome of this work, in 2018, was a new Generalised Bearing Life Model for hybrid bearings. It has since been tested and approved by a distinguished group of application engineers. Prototype versions of the model were applied alongside conventional methods for bearing life estimation and outputs were compared against the engineers’ real-world project experience.
Application in real life
To demonstrate the kind of insight made possible by this model, SKF’s team recently applied its method to four typical real-world scenarios. Rating life for a hybrid pump bearing, lubricated by an oil bath whose diluted oil resulted in poor lubrication, was eight times longer than an equivalent steel unit.
In addition, a hybrid screw compressor bearing running with contaminated lubricant was modelled and shown to give a hundred times the rating life of a conventional steel product.
Calculations for two electric motor applications were also modelled, under varying loading regimes but with clean and well-lubricated conditions in both cases. Similar rating lives were indicated for hybrid and steel bearings. In such cases, it should be remembered that other hybrid properties – such as electrical resistance or extension of grease life – may still justify the price difference.
A version of the model has been incorporated into SKF Bearing Select – a web tool offered to customers online and through dedicated software applications. Meanwhile, SKF engineers are supporting customer projects by applying a more sophisticated version of the modelling tool.
In general, bearings operating under lighter loads but with problems of poor lubrication or contamination are good candidates for replacement with hybrid alternatives. If loads are heavy but the bearing’s environment is clean and well-lubricated, steel’s resistance to sub-surface fatigue may have the advantage.
Whatever the situation, the new model will not only identify the option with longer life but will quantify the difference.