Abstract
This three day conference, concluding with plant visits to the Tata Steel Scunthorpe Medium Section Mill and Scunthorpe Rail Service Centre, attracted 51 delegates from 11 countries and was sponsored by Interfleet Technology, KNORR Technik, Portec Rail Group, Siemens VAI and Tata Steel. The conference, organised by Sue Parker (Tata Steel Long Products Business), was split into three main sessions: manufacturing, wheel/rail interface and innovation and maintenance.
In his opening keynote paper ‘Key challenges for our 21st century transport system’, Professor Andrew McNaughton (Chief Engineer, High Speed Two) stated that with the world’s population growing rapidly and an increasing aging population with greater expectations regarding the ease of use of transport systems, the move to a more sustainable railway system is required. In terms of energy efficiency, train travel, and even more so for high speed trains, compares favourably with both air travel and the private car. However, both airlines and car manufacturers are developing new, low impact designs with significant improvements. Electrification is an area in which railways are moving to greener energy sources, but consideration is needed on the method by which the electricity is generated. For example, the French railway is quoted as the most green, since a large proportion of electricity is from nuclear power stations. Looking to the future, railways will need to have greater reliability and safety with a steadily reduced downtime for maintenance, while being more affordable and environmentally friendly. Reducing the weight of vehicles leads to less impact on the track and therefore causes less damage and need for maintenance. A further benefit of reducing vehicle weight is better acceleration and breaking, reducing journey times and leading to more customers and revenue. However, over the past few decades, urban and regional passenger trains have become heavier. Should we rethink our rail systems around light rail solutions, for example, tram-train taking passengers to their homes rather than peripheral stations? One thing is for certain, due to the rapid population increase, gradual change in railways will not work. A fundamental change is needed and it will be the interface between rail and wheel that will be at the heart of the solution.
Session 1: Manufacturing
Session Chairs: Sue Parker and Kevin Sawley (Interfleet Technology)
In the keynote presentation for this session ‘Rail steel developments and their deployment to reduce maintenance costs’, Dr Jay Jaiswal (Director, Track Engineering and Technologies, Tata Steel Rail) stated that running steel wheels on steel rails leads to very arduous contact conditions that place challenging demands on the manufacturers, maintenance companies and the infrastructure managers responsible for the track. Although the rail industry has a very good record of safety, accidents such as that at Hatfield is a reminder of the consequences of improper management of the rail/wheel interface leading to the loss of rail integrity. Development of Tata Steel’s HPrail acknowledges that the pearlitic microstructure is a lamellae composite of pearlitic ferrite and cementite, with the hardness and strength properties of the two individual lamellae being remarkably different. In three dimensions, the orientation of the pearlite grains is random and the ferrite and cementite lamellae can be presented randomly at the rail/wheel contact. Thus, it is inappropriate to treat the pearlite as a single phase microstructure and assess its attributes using a macro-property test such as hardness. Furthermore, although refinement of the pearlite interlamellar spacing is hugely desirable, it is equally important to strengthen the constituent phases of the pearlitic microstructure. The composition of the newly developed steel was based on the metallurgical objectives of:
Increasing the volume fraction of cementite by an increase in carbon content, with the synergistic benefit of increased hardenability to refine the cementite lath thickness and the interlamellar spacing.
Increasing the strength of the pearlitic ferrite through solid solution strengthening by silicon additions and precipitation strengthening through vanadium additions. Additions of both silicon and vanadium have the synergistic effect of preventing/minimising the formation of deleterious grain boundary cementite networks.
Precise control of nitrogen and vanadium contents to capitalise on the hardenability effect of vanadium additions and to ensure the magnitude of lower temperature finer vanadium carbide precipitates within the pearlitic ferrite.
ReRail, a novel and patented technology was presented by its developer, Anders Sundgren, in the paper ‘ReRail a cost effective environmentally friendly rail alternative’. With ReRail, the rail is divided into two parts, the base made from a basic steel composition and a roll formed and hardened rail cap for the running surface. The base may be formed from the old rail with a milled head, custom profiled to fit the cap. The rail cap is the wearing part that can be replaced at speeds up to 10 m min−1. To increase the life of the rail, the cap is made with a hard surface and can have a lifetime that is more than doubled compared with standard rail. To optimise the system, the base should be manufactured with lower carbon content steel; this creates a rail very resistant to cracks and wear.
The specification for rail production stipulates that the rail must pass through two straighteners acting in two different planes to provide a straight rail. This has the unwanted side effect of introducing residual stresses in the foot of the rail due to the contact stresses generated between the rail foot and the rolls in the primary roller straightening machine. Producing rails with low residual stress is essential to improve resistance to crack initiation in service. The easiest way to do this is to use minimal loading through the straighteners, but this requires a rail with as uniform and straight a curvature as possible after cooling. Improvements to the camber profile and straightener set-up procedure that have enabled optimisation of the process and improved rail straightness at Tata Steel’s Scunthorpe Works were described by Ben Kay and Andrew Wilson in their paper ‘Producing consistently straight rails with consistently low residual stress’. It was stated that it is particularly difficult to maintain good straightness performance on short rail pieces, but fine tuning of the settings used in the straightener has since brought significant improvements to the short rail rework rate.
The new universal rail mill that forms part of a modernisation and capacity expansion programme for the Steel Authority of India (SAIL) at the of Bhilai Steel Plant was described in the paper by Ulrich Svejkovsky (SMS Meer GmbH) ‘The world’s most modern rail mill’. The new mill incorporates all state of the art technologies, such as universal rail rolling, tandem mill with CCS stands, straightening plant with CRS straightener, process models, profile gauge and a RailCool selective cooling line. The finishing end is especially designed for high throughput rates.
‘Intelligent track components’ by Paul Kitson (Tata Steel RD&T) described the current state of innovative components that have been developed under the seventh Framework Program, ‘Predictive Maintenance Employing Non-intrusive Inspection and Data Analysis (PM’n’IDEA). The components, fishplated joints, insulated block joints, breather switches, facing and trailing switches, transitions and the track itself, were selected because their integrity is assessed typically by manual inspection that is dependent on a number of factors, e.g. the competence of the track walking inspection team, which is time consuming, costly and in certain instances subjective. Development of the sensor technology was only part of the problem, the data produced by a particular measurement, and consequently the data analysis algorithms, that brings intelligence to the monitoring. The data analysis algorithms used are a combination of statistical analysis, Fourier transforms and wavelet analysis. It is the data analysis algorithm that produces a ‘signature tune’ for a particular component, the change of which over time is used to monitor the degradation of components and set limits for their use in service.
In recent years, there has been an increase in the number of requests to move to rail lengths over 100 m. This is not surprising as there are a number of advantages to railways with long rails, for example, fewer joints are required between rails, installation time is reduced and higher speed trains are possible. In the paper ‘From short rail to long rail’, Jim Hogg of Siemens VAI reported that many rail producers have been considering how to provide long rails within the confines of their existing production facility as each producer has local and specific considerations, whereas the equipment supplier aims to provide solutions with some commonality. In general, the major problems are not related to the rolling process, but to the finishing and despatch areas. This was typified in the TATA Scunthorpe Rail and Section Mill, where space was available at the side of the existing mill building but transfer of a long rail through to the bay and precambering of the long rails was difficult. The solution in this case uses a chicane where the hot rail deflects through a bend and passes through a single modified building column. This solution gave substantial benefit from reduced building modifications. There is no universal solution to changing from short to long rail, there is a solution for each application and the problem is agreeing the compromise.
Daniel Szablewski (Transportation Technology Centre, Inc.) presented the paper ‘Development and wear evaluation of rail steels in North American heavy haul environment’, a historical perspective on the tests conducted at TTCI over the past 20 years, and the developmental work carried out to prevent rolling contact fatigue (RCF) and increase in the rail life cycle in North America. In order to aid with this evaluation, TTCI developed a full scale facility for ‘accelerated service testing’ (FAST), which consists of a 4·3 km high tonnage loop track currently running a 110–115 car train with an average axle load of 35 t. The average accumulated daily tonnage is 1·55 mgt which allows for rapid evaluation of rail wear performance. Results indicate that gauge face (GF) lubrication has a significant impact on the high rail wear, as intermediate strength rails with an average head hardness of 335 HB and GF lubrication applied, wear ∼100% less than do the premium high strength rail steels with an average head hardness of 407 HB and a dry condition on the high rail. In addition, the softer rails do not exhibit severe gauge corner RCF, which is present in substantial amounts on nearly all premium rails.
‘Premium quality rails’, the paper presented by Danieli’s Director Process and Technology, Simon Wright, detailed how continuous efforts in research and development have made it possible to introduce significant innovations in the production of premium quality rails. Through the use of high quality steel feed stock, universal rolling process, inline heat treatment and an integrated process control system, rail manufacturers will be able to supply a superior quality product.
Rail quality is underpinned by state of the art rail inspection systems. Before a finished rail leaves the production plant, it has to be inspected according to its intended use based on the respective standards, as well as the customer’s requirements. Against this background rail inspection lines become more extensive, as described in the paper ‘State of the art rail inspection lines’ by Johann Knorr of Knorr Technik GmbH. Typically, inspection lines comprise brushing machines to remove loose scale from the rail surface before inspection, vision systems to inspect the rail surface for three-dimensional defects, profile measurement systems to measure the cross-sectional dimensions, straightness measurement systems, eddy current systems to inspect the rail surface for cracks, and ultrasonic systems to inspect the rail for interior defects.
Session 2: Wheel/rail interface/innovation
Session chairs Mick Steeper (Siemens VAI Metals Technologies) and Robert Carroll (Stagecoach Supertram).
‘The history of rails: from wood to steel’, was the topic of Rob Carroll’s keynote paper for the start of this session. Guided transport has been in existence since at least 600BC when the Diolkos waggonway used ruts cut into the limestone rocks to allow the transfer of boats across the Isthmus of Corinth. It was not until the Middle Ages that guided transport appeared in central Europe associated with mining. The first recorded use of rails and flanged wheels, in what we would now recognise as an early railway, was in 1604 on the Wollaton waggonway, Nottinghamshire but their use was soon taken up more widely in Northumberland and Shropshire. For over 150 years, these early railways used wooden rails until in 1767 cast iron plates were used in Coalbrookdale to reduce the wear of the wood. Steam locomotion only became a practical proposition with the introduction of rolled wrought iron rails in 1820.
R260 is currently the recommended rail steel in the UK, except in tight curves (<2500 m radius) where wear and RCF significantly reduce the rail life. Harder, premium grade rail steels can significantly reduce life cycle costs in critical curves, but the long term performance of these is not yet fully understood. Results from the European Framework project ‘InnoTrack’ were presented by Francis Franklin of Newcastle University in the paper ‘Plastic deformation and crack initiation in hard pearlitic rail steels’, in which R260, and four premium grades, were studied using twin disc tests in order to compare wear rates, strain hardening and shear strain accumulation. The data collected can feed into rail life prediction models, which can be used to optimise maintenance schedules.
Chris Hardwick (University of Sheffield) presented the paper ‘Rail wear, understanding the effect of third body materials’, in which modelling techniques were used to describe third body contamination. There are many forms of third body material contaminants present on the rail head. These include those occurring naturally (water, leaves, etc.), as a result of train passage over the rail (dust and small stones fom the ballast) or products applied deliberately for modifying the wheel/rail contact (grease, friction modifiers, lubricants and sand or traction gels). The work highlighted the deficiencies apparent in current multibody dynamic (MBD) modelling approaches, particularly those using the T-gamma approach with dry wear curves and fixed values of traction coefficient. It has indicated that accounting for third body materials simply by changing the traction coefficient value is insufficient as this will not account for the change in wear rate that will also occur. Continuing work is being carried out to produce a range of wear curves and creep maps that can be used to improve the accuracy of MBD simulations in the future.
During a rail’s service life, it is subjected to multiaxial rolling/sliding contact stresses that exceed the yield stress of the rail material, resulting in significant plastic deformation and work hardening of the surface and subsurface regions. George Roberts (Birmingham University) in his paper ‘Use of misorientation values to further understand deformation in rail steels’, described a model of a typical rail microstructure that allows the kernel average misorientation (KAM) value to be calculated. The microstructure in the model can be sheared, representing the deformation seen in rail steels below the contact patch, and the KAM value determined. The model results have been compared with experimental data and it has been found that macroscopic shear causes an increase in KAM value; however, changes to the low angle boundaries between ferrite lamellae also need to be taken into account. The model, along with experimental data, will be used to investigate the differences in deformation behaviour between different rail steels in the next stage of the work.
Residual stresses can be of significant influence on the structural integrity of railway components including rails, wheels and axles. These residual stresses can arise during manufacture or induced and modified through service conditions. Hence, it is of significant importance to characterise these stresses spatially and monitor their stability as a function of manufacturing processes and service usage. Alexander Evans (Institut Laue Langevin), described in the paper ‘Non destructive determination of residual stresses in railway components by Neutron diffraction’, a non-destructive technique for measuring the stress in a rail by neutron diffraction, which is able to map spatially the residual stress state from the near surface to deep within the bulk, and through diffraction peak shape analysis, inhomogeneous plastic strain distributions can also be mapped.
From early days of rail travel, high rail side wear has been recognised as a limiting parameter in rail life. With increasing tangential forces and peak contact stresses, RCF has also become a factor in rail asset life. On some sites, rapid corrugation growth can require excessive metal removal by grinding. The use of friction control to mitigate wear and RCF has been an evolving and developing technology that can significantly extend rail (and wheel) life. In his paper ‘Friction management and rail life extension in the UK, a review of the past and a look into the future’, Don Eadie (LB Foster Friction Management) examined how friction control developed and evolved in the UK and compared these with global developments. Anticipated future developments were given.
Professor Rod Smith of Imperial College London started the second part of session 2 with the keynote presentation, ‘Tracks into the 21st century’. The presentation focused on rail safety, focusing on a number of high profile rail accidents, concluding that there have been very few accidents caused by structural failure. The total number of fatalities since 1840 is around the same number on the roads in a single year and the trends for number and severity of accidents is falling.
A technique using ultrasonic guided waves in the kilohertz range (typically 20–300 kHz) to inspect for rail defects from single point of access for many metres with full volumetric coverage was described by Yousef Gharaibeh (TWI) in his paper ‘The application of long range ultrasonic testing (LRUT) to inspect railway tracks’. The findings show that ultrasonic guided waves can propagate in rails with the ability to detect common type of defects over a distance of 4 m located away from the ultrasonic excitation/reception region.
During rail manufacture, high soaking and rolling temperatures cause surface decarburisation. Recent work has shown that the depth of decarburisation increases both the total wear and wear rate, which affect the lifetime of rail. Wenquian Zhu (The University of Birmingham) in his paper ‘Evaluation of rail decarburisation depth using H-shaped electromagnetic sensor’, showed the possibility of using such a sensor to detect the different decarburisation depths in rail samples during offline testing at room temperature. This is a non-destructive test that does not require elaborate sample surface preparation, and therefore the measurement process can be carried out quickly. The sensor exploits the increase in relative permeability in the surface layers of the steel due to decarburisation.
Richard Stock from the rail manufacturer VoestAlpine presented their latest work on rail grade selection and development in the paper ‘Rail grade selection and friction management: a combined approach for optimising rail–wheel contact’. The combination of an adequate heat treated rail grade, an adjusted maintenance programme (grinding) and friction management will result in best possible rail performance. Current rail grade development at VoestAlpine focuses on the problems of wear and rolling contact fatigue. Bainitic rails with specially engineered microstructure (multiphase bainite) combine average wear resistance with excellent RCF resistance. For conditions with high wear rates, high strength pearlitic grades can offer significant rail life extension by providing excellent wear resistance combined with good fatigue resistance.
Alternating current field measurement (ACFM) is a non-destructive testing technique for the detection and sizing of surface breaking cracks in metallic components. The technique presented by Gemma Nicholson (The University of Birmingham) in the paper ‘Measurement and modelling of ACFM response to multiple RCF cracks in rail and wheels’, works by inducing locally uniform and unidirectional currents in the component under test. Identifiable defects cause disturbances in the associated magnetic field that are evaluated to produce an estimate of crack size (surface length and pocket depth). Fatigue cracks in rails and in wheels often occur in clusters, and it has been shown that the ACFM response is affected by the spacing and number of cracks present. With this method it is possible to distinguish individual cracks with spacing greater than 5 mm, but closer spacing leads to a single signal response.
The final paper of day two, ‘Detection of crack growth in rail steel using acoustic emission’, was presented by Andrii Kostryzhev of The University of Birmingham. Acoustic emission (AE) offers a technique for monitoring whether crack growth is occurring under traffic loading. It is known that fatigue cracks can grow slowly after onset and may not produce a great number of AE events that can be distinguished easily from the background noise, therefore, substantial analysis of the AE parameters is needed to distinguish crack signal events from noise. Results show that with this technique it is possible to distinguish between types of fracture, since the AE signal, associated with crack growth depends on the fracture mode. High duration low frequency signals result from ductile fracture, whereas low duration high frequency signals result from brittle fracture.
Sesison 3: Maintenance
Session Chair Francis Franklin (Newcastle University)
In the keynote paper for the final session, ‘Welds, the Achilles Heel of modern high performance railway?’, Richard Johnson of Thermit Welding GB, questioned whether welds are the Achilles heel of the high performance railway. For over 100 years aluminothermic welding has been an effective method for joining rails on site. The process still offers significant benefits in terms of capital cost, mobility and versatility, particularly in track maintenance. Modern manufacturing and SPC methods are now delivering high quality and welding consumables. By adopting new process control technology and coupling it to portable devices for data transmission, it is now possible to offer tools for aluminothermic welding that removes most of the uncertainty and lack of traceability out of the processes.
Network Rail have developed and implemented a number of key strategies for rail profile management over the last 10 years and these were discussed by Gareth Evans of Network Rail in the paper ‘Managing rail profile’. Rail grinding and lubrication, top of head friction modifiers and premium grades have been adopted to extend the serviceable life of rail in track. Significant cost benefit has been demonstrated with payback for grinding and lubrication after 3–4 years. However, in the long term, premium grades are predicted to give the greatest benefit in life cycle cost models.
On a similar theme, Rob Carroll of Stagecoach Supertram (Sheffield) discussed the latest techniques in tramway rail maintenance in the paper ‘Tramway rail management’. The Stagecoach Supertram system in Sheffield consists of 58 km of track of which 45% is ballasted track and 55% embedded street running. With an intensive service operating throughout the day carrying 15 million passengers per year, limited time for track maintenance at night combined with 25 m radius curves and gradients up to 10%, the duty requirements for rail are severe. Along with the use of premium grade products, developments such as infrastructure segmentation and weld repair are helping to target maintenance and extend rail life.
RCF is one of the main degradation mechanisms in rail track and has therefore been a significant subject of railway research in the UK over the last 10 years. Improving the resistance of the rail material to fatigue damage has been shown to reduce RFC. In the paper ‘Experience of premium grade rail steels to resist rolling contact fatigue (RCF) on GB network’, Mark Burstow of Network Rail described his involvement with trials at a number of sites on the GB network, where premium grade rail steels are monitored regularly for their performance with respect to both wear and RCF. The study shows that correct application of the premium grade material has improved greatly the rail life and eased the management of the rail. However, incorrect use of a premium grade can introduce RCF where it was not present before.
Rail profile management is typically undertaken by grinding. Paul Baker (Bakerail Services) has been working on a new milling solution described in the paper ‘Rail milling, an alternative to grinding or more a new tool in the tool box?’. The method offers a technique that closes the gap between rail grinding and rail renewal, offering the possibility of real rail life extension by returning the rail back to its original profile and free from defects. Whereas grinding may require many passes to recover the deepest defects, milling may only require two or three. Once treated the rail has many more millions of tonnes of rail life left, dependent on the subsequent management strategy, e.g. lubricators, preventative grinding, etc.
‘Rail grinding for the 21st century, taking the lead from the aerospace industry’, presented by Roger Singleton (AMRC) was the final paper of the conference. In order to make best use of the maintenance window, metal removal rates have been increased using best grinding practice techniques from the aerospace industry. Computer modelling was carried out to identify improvements in productivity by applying theory on power requirements and chip thickness of the grinding process. The work concluded that high efficiency deep grinding (HEDG), a novel abrasive machining process that readily achieves very high material removal rates, provides the best opportunity for performance increase in preventative grinding operations where high speed is desired. A theoretical three- to fourfold increase in grinding train speed is achievable by utilising the specific grinding energy associated with HEDG grinding of steel using current power constraints.
On the afternoon of the final day, conference delegates were given the opportunity to visit Tata Steel’s Scunthorpe Rail and Section Mill (SRSM). Andrew Wilson from the SRSM hosted the tour, starting at the reheating furnace, through the rolling process, then stamping, cooling and straightening. The tour then continued to the dedicated rail inspection centre, where rails are checked for internal and surface quality along with profile checks using state of the art equipment. This was an excellent opportunity for those not familiar with rail manufacture to see the process in action.
Overall this was an interesting, stimulating and enjoyable conference, held at an ideal venue in terms of facilities and transport access. The conference provided many opportunities for networking, especially at the conference dinner held in the Great Hall of the National Rail Museum alongside some of the great locomotives. Most delegates took the opportunity to look round the Museum during the lunch breaks providing a reminder of the past and inspiration for the future of this evolving industry.