INTRO: The cost to Britain’s national railway of damage from rolling contact fatigue has been conservatively estimated at £250m a year, and measures to limit RCF crack formation could cut that by 55%. Andy Doherty, who chairs the newly-established Vehicle/Track System Interface Committee, explains how co-operation between rail industry players can achieve this goal

BYLINE: Andrew Doherty

Director of Railway Systems & Vehicle EngineeringNetwork Rail

ROLLING CONTACT FATIGUE costs Network Rail £200m a year, and quite possibly more. This estimate embraces premature renewal of cracked rail, including switches and crossings, plus RFC-related operating costs such as monitoring crack growth and rail grinding.

Based on current knowledge, no-one has yet estimated accurately the cost of RCF in wheels. However, in his keynote address to the Vehicle/Track Interaction Conference in February, the Strategic Rail Authority’s Executive Director, Technical, David Waboso conservatively estimated the total cost to the UK railway industry at £250m per annum.

It is known that substantial numbers of rolling stock wheelsets are now experiencing RCF crack initiation and growth. This is likely to require higher metal removal rates at wheel-turning intervals, leading to shorter wheel life and thus higher costs. As the mechanisms that generate RCF on the running surface of wheels are similar to those acting on rails, it is likely that any action which reduces propagation of cracks in one will deliver benefits to the other.

However, the fragmented structure of our national railway with its many contractual interfaces makes it difficult for individual companies to address in isolation issues like RCF, which arises at and spans the wheel/rail interface.

In an effort to resolve this difficulty, the Wheel/Rail Interface System Authority (WRISA) was set up in 2001 in the aftermath of the Hatfield derailment in October 2000. This was caused by the break-up of a rail that was severely affected by RCF in the form of gauge-shoulder cracking. The role of WRISA was to examine the issues surrounding the wheel/rail interface, and to propose system-level solutions to the problems encountered. It had no executive powers to impose these solutions.

WRISA commissioned and supported many studies, the most important of which were on the Great Western and London - Shoeburyness lines. Vehicle dynamics modelling using Vampire allowed the root causes of RCF at each site to be identified. Conclusions from these studies have been assimilated into the RCF Hypothesis, described below, which has now achieved broad acceptance across the industry.

After more than two years of operation, during which many other significant research and development studies were undertaken, WRISA was forced to cease trading on March 31 2004 as it was unable to secure cost-effective insurance cover against personal liability for recommendations that might subsequently be challenged in court.

The Vehicle/Track System Interface Committee (V/T SIC) was established in order to fill the gap left by WRISA. As V/T SIC is not a company, it will not suffer from insurance problems.

V/T SIC’s modus operandi

At a seminar on April 13 2004, V/T SIC reviewed its purpose and scope (p405). In order to build upon the work of WRISA, but move at a faster pace, V/T SIC resolved to engage stakeholders proactively across the industry in a constructive dialogue. The objective is that lessons learnt from ongoing research and development studies can be cascaded out to the industry and translated expeditiously into measurable improvements to the system as a whole.

The Committee will establish a helpline service for the UK rail industry to deal with any questions concerning the wheel/rail interface. This service will be handled by the Rail Safety & Standards Board’s Helpline (+442079047518), which will forward wheel/rail questions to the appropriate member of the V/T Permanent Project Group for reply. This will not be a live on-line service, but it should provide greater visibility and access to important areas of research and development that have been ongoing for the past few years. These have answered most of the difficult questions, allowing real operational advice to be given.

RCF Hypothesis

WRISA sponsored research projects making use of Shakedown Theory, as developed by Bowers, Johnson et al at Cambridge University, to define conditions that are likely to produce RCF. In brief terms, any two objects in rolling contact will produce a contact patch that can be described by the pressure and shear forces in the plane of contact. A shakedown limit that separates non-RCF and RCF conditions can be determined that is a consequence of material properties and contact friction.

The shakedown diagram in Fig 1 depicts a change in RCF propensity with wheel condition for a specific vehicle type. The pressure and force data was generated from a vehicle/track simulation as the vehicle traversed a switch & crossing set. The ’notional limit’ was developed by matching many modelling outputs with very accurate field RCF surveys. It has proven reasonable over a variety of track and vehicle conditions.

The data from many simulations and field surveys have been compared as part of the WRISA projects. As a result, an RCF Hypothesis has been generated that is a useful tool to encapsulate the insights obtained to date. The Hypothesis is regarded as a ’working document’ and is expected to change as more knowledge is obtained. A reasonable hypothesis to explain the initiation of the majority of RCF in the UK system is as follows:

1. The British wheel/rail system operates (on average) in the vicinity of the notional shakedown limit.

2. In most cases where a wheel/rail pair exceeds the notional shakedown limit, contact patch shear force rather than contact patch pressure is the primary cause. Longitudinal rather than lateral shear force is the cardinal factor, since it is a direct consequence of excessive rolling radii differences resulting from sub-optimal steering in response to either curves and/or local track anomalies: