• Home »  Features »  Process Controls & Plant »  Industrial Services

Accurate modelling widens the use of hybrid bearings

Published:  27 March, 2020

Applications for hybrid bearings, which use ceramic rolling elements on steel raceways, continue to grow. Properties including low weight and good performance under demanding conditions have seen them used in niche applications such as high-speed machine tool spindles. More recently, their high electrical resistance has made them popular in applications such as variable speed drives, where electrical arcing can damage conventional bearings.

Ceramic rolling elements have other advantages over all-steel bearings. They generally run at lower operating temperatures, resist surface damage from particulate matter and do not suffer from the potential risk of steel-to-steel surface welding. Because they have a lower boundary-lubrication coefficient of friction, this allows them to function more efficiently in applications with poor lubrication.

However, unless the advantage of using a hybrid bearing is clear in advance, it can be difficult for engineers to justify their use. Will it outperform its steel equivalent in a particular application? Are the possible performance benefits worth the extra cost? The reason is because the equations used to calculate the rating life of a bearing do not account for the real-world performance of hybrid designs.

Sub-surface fatigue

The conventional bearing life model is based on sub-surface fatigue. Rotating bearing components are continually loaded and unloaded. Over many cycles, fatigue accumulates, leading to eventual failure. Fatigue behaviour is well-understood, so engineers can use data from their application, such as loads and speeds, in an equation. This helps them determine the rating life of a given bearing design. The dynamic load rating of a bearing, which is readily available, is mainly used to quantify its sub-surface performance.

This model is commonly used and incorporated into international standards. However, it does not show hybrid bearings in their best light. Ceramic rolling elements are stiffer than their steel counterparts and deform less under load. This means loads are concentrated over a smaller area of material, which increases stress and accelerates sub-surface fatigue.

More importantly, real-world experience does not always agree with this traditional model. In the field, most bearings fail due to problems at the surface, not in the body of the material. The underlying cause is usually down to poor lubrication or contamination. To account for this, modern standards such as ISO 281 include correction factors in an attempt to accommodate these effects. However, these correction factors still do not represent the real behaviour of bearings in service.

Improved model

PWE spoke to Guillermo Morales-Espejel, principal scientist at SKF Research and Technology Development, who said that in 2012, SKF set out to remedy this by creating a better bearing life model. To do this, they needed three things. The first was a model of sub-surface fatigue within the material. The second was a model for failure at the surface. The third was data from endurance tests, which SKF could use to calibrate and validate the new model.

The company spent two years working on the new model, drawing on decades of experience in materials science and tribology. It required a detailed understanding of the behaviour of bearing surfaces, from their friction characteristics to the way dirt particles affect them under load.

An initial concept model was presented as a Generalised Bearing Life Model (GBLM) in 2015 at Hannover Messe. However, this did not cover the modelling of hybrid bearings. For this, SKF still needed to collect large amounts of data, for calibration and validation purposes.

This required lots of work. Morales-Espejel said SKF needed data on the operating life of bearings over a wide range of loads and surface conditions, in order to build behaviour curves. For every point on the curves, they needed to test around 30 bearings, knowing that several of them would fail. The company also needed to compare bearings with steel and ceramic rolling elements, and those operating with poor lubrication or in contaminated environments.

In all, SKF tested hundreds of bearings. The test programme, and the adaptation of the concept model, required another four years of work at its facilities in the Netherlands and Austria.

Tested and approved

The company completed the new GBLM for hybrid bearings in mid-2018. It has since been tested and approved by an influential group of SKF’s application engineers. They used prototype versions of the model alongside conventional bearing life estimation techniques and compared outputs to actual results from customer projects.

A key feature of the model is that it separates surface fatigue from sub-surface fatigue. It applies classical rolling contact fatigue in the sub-surface region and a new tribologically-dependent surface degradation model for the ceramic-steel raceway interface. The fatigue resistance of the ceramic-steel interface can, in most cases, compensate for the extra stress present in the sub-surface region of the contact.

Two variants of the GBLM for hybrid bearings were developed. One will be incorporated into the SKF Bearing Select webtool, which is offered to customers on-line and through dedicated software applications. A second variant, which is more sophisticated and complex, will be used by its application engineers to support customer projects.

The new model offers real insight into when the use of hybrid bearings can be justified. Benefits are offered in many typical situations. For instance, when a bearing is heavily loaded, but runs in a clean, well-lubricated environment, sub-surface fatigue is likely to be the ultimate failure mode. Here, a steel bearing may perform better than a hybrid. However, many bearings operate under lighter loads and with a greater likelihood of poor lubrication or contamination. The model will reveal whether a hybrid solution offers a longer life in those applications and will also quantify the difference.

Paper proof

In a 2019 scientific paper, published in Proceedings of the Institution of Mechanical Engineers, Morales-Espejel highlights that SKF demonstrated how the model was used in four representative real-world applications. These were: a pump bearing; a screw compressor bearing; and two applications of a bearing in an electric motor.

For the pump bearing, which ran under poor lubrication conditions, the rating life of a hybrid bearing was eight times that of a steel equivalent. For a screw compressor bearing with contaminated lubricant, the hybrid had 100 times the life of its steel equivalent.

The other two cases looked at an electric motor operating in clean, well-lubricated conditions, under two different load regimes. In both cases, the rating life of the hybrid bearing was very similar to the conventional bearing. However, other potential benefits of hybrid technology, such as electrical resistance or a longer grease life, could be decisive when making a final decision.

Existing life theory models for hybrid bearings have many shortcomings and cannot precisely model ‘real’ conditions. Accurate prediction of bearing life will however help engineers choose bearings that are better matched to the specific needs of their application.

Morales-Espejel says that to SKF’s knowledge, there is no other publicly available life model for hybrid rolling bearings. There are only adaptations of models for classical all-steel rolling bearings, which tend to penalise hybrid bearings because of their higher stress under identical loads. This is despite clear evidence that under ‘real’ conditions, such as low load, high contamination and poor lubrication, hybrid bearings can offer superior performance.

Making the new GBLM widely available will help engineers do this in the context of hybrid bearings. Here, the properties of hybrid bearings can be modelled in many scenarios, such as low lubrication, to see whether their use can be justified.