Application of technology improves wheelset performance

The wheel/rail interface is one of the most critical parts of an entire railway operation. Through a contact patch the size of a small coin, all the forces between the moving element (the train) and the static element (the track) are transmitted. These forces are immense, and if not properly managed, disaster may follow.

Rolling contact fatigue can propagate tiny cracks in steel, so that the cracks grow and penetrate the rail web, causing it to disintegrate. This happened at Hatfield in the UK in October 2000, resulting in a fatal derailment. Following this, an intensive rail grinding regime was instituted, taking cracks off the railhead before they have a chance to propagate.

Similarly, forces on the wheels have to be carefully monitored. The failure of a tyre on an ICE train at Eschede in 1998 resulted in the worst ever rail accident in the Federal Republic of Germany, as the subsequent derailment resulted in the train hitting a bridge abutment – causing over 100 deaths. Monobloc wheels were fitted to the ICEs after this.

Again in Germany, there was a derailment of a later generation ICE train leaving Cologne in July 2008, caused by a broken axle. There were no fatal consequences, as the train was only travelling slowly and it was possible to evacuate the passengers onto the platform at the station. However, the incident rang alarm bells in the German railway industry as evidently the consequences could have been much worse.

Rolling stock was taken out of service pending an investigation, resulting in serious disruption to the timetable. The Department of Public Prosecution charged the Federal Institute for Materials Research (Bundesanstalt für Materialforschung, BAM) with investigating the cause of the axle failure. According to the BAM report, manufacturing-related material inclusions in the drive shaft led to the initial crack and finally to the breaking of the drive shaft.

Incidents such as these underline the necessity for precise manufacture of rails and wheelsets and a comprehensive monitoring programme to ensure both are standing up to the pressures to which they are subjected.

Manufacture

Western Europe has traditionally been a centre of wheelset manufacture and retains several important companies such as Lucchini. This firm’s main factory is at Piombino in northern Italy, and the firm also has a forge in Manchester, UK.

In the years after the fall of the Iron Curtain, many ‘metal-bashing’ activities associated with railway manufacturing migrated to central and eastern European countries, where labour rates were lower. As a result, the Czech Republic in particular has become important in the production of railway wheelsets for European railways, with the Bonatrans company in Bohumin an important supplier.

This has resulted in the growth of a high tech support industry. The Starmans electronics company in Prague has been developing an automatic system for surface defect detection on newly manufactured railway wheels known as DIO5000. The main goal has been to develop a highly reliable system based on image processing algorithms that gives a warning of surface flaws to prevent possible future accidents.

The system is based on magnetic inspection, where the wheel is put into a magnetic field and any cracks cause a magnetic stray field. Using this technique, the cracks are visible: they can be recorded using a high-speed digital camera, which has also been developed by the Starmans company. The recorded pictures are then evaluated using Starmans’ image processing algorithm. The whole surface and different sizes of railway wheels can be inspected.

There are two issues in evaluating the data. First, the picture is corrupted with a high ‘noise’ level, rendering the detection process more difficult – this is the hardware part of the question. Secondly, the defects have to be recognised from the scratchy surface of the wheel. Only the flaws have to be detected: this is the software part. Both noise reduction and detection algorithms are based on simple mathematical equations which are implemented in the Starmans digital camera. This means that the pictures are processed on-line, during inspection of wheels.

Starmans says that using the DIO5000 system it is possible to detect the surface flaws on railway wheels automatically as part of the manufacturing process. The digital camera with adequate resolution has been developed for the surface scanning, and the processing algorithms allow analysis of the data. The firm says that it is possible to detect flaws with a length of 1mm and thickness of 0.3mm.

Defect detection in operation

Systems such as the Starmans DIO5000 method are useful in wheel manufacture, but monitoring of wheels once in traffic can also help prevent potentially serious derailments. In 1999, the American Association of Railroads (AAR) initiated a research programme to detect cracks in freight wagon wheels using ultrasonic testing methods. Under the direction of the AAR Railway Technology Working Committee, the Transportation Technology Centre (TTCI) in Pueblo, Colorado began research and development for this project.

In June 2005, TTCI brought the Dapco company in on the project. Dapco specialises in the design and development of nondestructive ultrasonic railway wheel inspection systems: these systems are used to inspect railway wheels/wheelsets for subsurface flaws utilising proprietary technology with the ultimate goal of preventing derailments and service failures.

TTCI began working with Dapco to deploy an Automated Cracked Wheel Detection (ACWD) system utilising Dapco’s proprietary product line. In the spring of 2006, a system was deployed and commissioned for inspection of all wheels on one side of a train.

In 2008, Dapco was contracted by the Union Pacific Railroad to install an Ultrasonic Cracked Wheel Detector at the Bailey Yard in North Platte, Nebraska. This system includes many design modifications and increased functionality since the installation of the prototype system at the Transportation Technology Centre.

The Dapco Wayside Cracked Wheel Detection System uses a conventional ultrasonic approach dynamically to inspect railway wheels at train speeds of up to 5mph (8km/h). The main function of the ultrasonic inspection is to determine the presence of internal defects. The system provides real-time assessment and reporting of flaw type, size, and location in and across the tread and rim of the wheel.

The system consists of four inspection stations per rail with up to eight transducers housed in each inspection station. Each inspection station will inspect one wheel in a four wheel sequence. For example, if inspection station No.1 inspects the first wheel in a train formation, then it will inspect the fifth, ninth and so on. Similarly the remaining three inspection stations will test their relative sequential number.

As the inspection elements are put in place, the ultrasound is coupled into the surface of the rail wheel from underneath via a fine water mist. The wheels are inspected ultrasonically and after the inspection is complete the station will lower the inspection elements and return them to their home position to stand ready for the next assigned wheel.

The system utilises flange-bearing track which exposes the tread of the wheel for ultrasonic inspection. The inspection probes are engaged to the tread surface of the wheel from underneath and a fine mist of water is used to couple the ultrasound into the steel as the train and the wheel travel at speeds of up to 8km/h.

Another strand of TTCI research has been to reduce the stresses on wheels through an improved suspension system. According to TTCI’s Mr. Harry Tournay, “shelling” of the wheel tread results from fatigue of the material. Cycles of high stress under rolling contact create cracks in the wheel tread. Typically, these cracks will form circumferential bands. As the cracks propagate and grow, they can connect and dislodge a patch of wheel tread, leaving a void in the process.

Wheel heating from application of tread brakes can reduce a wheel’s fatigue resistance in a process known as thermal mechanical shelling (TMS).

Mr. Tournay says that one method to reduce TMS is to manage the demand for surface tractions by the vehicle suspension, i.e. by improving the steering. Reductions in wheel shelling were seen with the introduction of steering bogies in Canada in the 1990s. In steering bogies, the two axles move independently from one another, allowing the bogie to ‘steer’ round a corner.

Other possible approaches are to manage top-of-rail (TOR) friction, reduce the thermal load on the wheel, and to use wheel steels with higher yield strengths.

Choice of wheel steel is being investigated on the Malmbanan iron ore line linking the mines at Kiruna in northern Sweden with the harbour at Narvik in Norway, where trains in excess of 8,000 tonnes operate. Here, the potential for wheels made of bainitic steel is being explored.

According to Anders Ekberg of Chalmers University of Technology in Sweden, bainitic steel grades generally have superior rolling contact and thermal fatigue resistances, but poorer wear characteristics when compared with ‘conventional’ pearlitic steel wheels. To improve the wear characteristics, the manufacturing process of bainitic materials needs to be carefully controlled. On the Malmbanan, the potential of a bainitic steel manufacturing process known as ‘Micralos’ is being explored.

Wheel turning

Regular wheel turning is a key requirement if wheels are to be kept truly round and thus minimise impacts at the wheel/rail interface. Two years ago Hegenscheidt-MFD, a wheel lathe and wheel diagnostic manufacturer, formed a partnership with Nexala to develop a wheelset management system. The application allows rail operators to achieve cost savings by planning, forecasting and managing their use of wheelsets.

The application monitors wheelset wear behaviour across the fleet in line with expected target wears. It notifies maintenance staff on deviations within specific tolerances over and above the target wears for immediate action and damage/cost limitation. In addition, it forecasts scrap, turn and last turn dates for each individual wheelset for maintenance planning.