The Tram-Train concept allows a railway vehicle to run in two operational modes; as an on-street tram serving city centres but also as a commuter train running on the existing local rail network. This dual operation concept provides great flexibility and efficient use of railway infrastructure and connections.

Tram-Train first became popular in Germany and is now spreading rapidly through Europe. The Sheffield-Rotherham Tram-Train scheme represents the UK’s first trial of the concept and has provided the project partners, the Department for Transport, Network Rail, Northern Rail, Stagecoach Supertram (SST) and South Yorkshire Passenger Transport Executive with many challenges.

Examining the interface

One of the key issues to be tackled in this project is the wheel/rail interface. In essence, this is how the wheels contact with the running rails, check rails, guard rails and switches and crossings (S&C) during operation. The wheel/rail interface is very important not only to the safe running of the vehicle but also to maximise wheelset life and to minimise wear and tear on the infrastructure. Getting this interface correct for a vehicle running on one rail system can be problematic but to optimise the interface for two very different systems is a major challenge.

Under contract by Network Rail, a team of engineers at the University of Huddersfield’s Institute of Railway Research (IRR) has been developing a new wheel profile for Tram- Train that can operate on both Network

Rail and SST track whilst minimising wear, rolling contact fatigue (RCF) and derailment risk and meeting all of the relevant industry standards.

David Crosbee, IRR Senior Research Fellow explains: “The first issue encountered when developing the new wheel profile was that SST and Network Rail use very different wheel and rail profiles. The SST trams use a modified DIN type wheel profile with a mix of BS80A vignole rail and 55G2 grooved rail. In contrast, Network Rail predominantly uses BR P8 wheel profiles running on BR113a rails. Add to this the fact that new rails wear to adopt a different shape, usually similar to the mean shape of the wheel profiles running on them, and we end up with an eclectic mix of different rail profiles that we have to work with.”

Worn rail profiles from both SST and Network Rail infrastructure were recorded using a MiniProf Rail instrument for use in the wheel profile development. Given that the Tram-Train traffic will form only a small proportion of the overall route traffic on both infrastructures, the wear on Network Rail and SST rails will be dominated by the existing vehicles. As a result, it was assumed that the existing rail profiles would remain a similar shape post Tram-Train introduction.

Due to the wide range of new and worn rail profiles present on the Tram-Train route, reducing the number of wheel/rail combinations prior to the wheel design stage was necessary. Given that new rails will wear to a worn shape in a relatively short period of time and that many of Tram Train Graphicsthe existing rail section types have already worn to a common shape, the SST aspect of the work employed a single representative worn rail shape. Likewise, a similar approach was adopted for the selection of a representative rail profile for Network Rail infrastructure.

Performance assessment

A combination of vehicle dynamic simulations and bespoke in-house software was used to assess the performance of the existing Network Rail and SST wheel profiles on the two infrastructures. The analysis included the calculation of contact patch stresses, rolling radius difference, contact angle and T-gamma – the energy dissipated in the contact patch, used to predict rail wear and RCF propagation rates throughout the route for a range of operating conditions.

Early on in the study, it was found that, due to the gauge corner of Network Rail rails having a considerably larger radius than the flange root radius of the SST wheel, severe two-point contact occurred. This condition creates two distinct contact regions, one on the rail head and one towards the rail gauge corner. The resulting net wheel-rail contact forces cause reduced wheelset steering and increased levels of wheel and rail wear. Therefore the SST DIN wheel was deemed unsuitable for use on Network Rail infrastructure. Conversely it was found that a P8 wheel profile on SST rail resulted in similar levels of wear to the standard SST wheel. As a result, a hybrid wheel profile based on a P8 profile with the addition of the SST flange tip and flangeback geometry was developed and further optimised for the two railway networks.

The back is important

When a rail vehicle negotiates S&C, and also tight curves with check rails, the back of the wheel flange provides guidance through contact with the check rail. The location of the check rail on the track is based on the back- to-back spacing of the wheelset and, although SST and Network Rail tracks have the same nominal track gauge of 1435mm, the current SST vehicle’s wheelsets have a larger wheelset back-to-back spacing of 1379mm when compared to the BR P8 back-to-back spacing of 1360mm. This is due to tramway wheels having a narrower flange for negotiating grooved rail on street running sections of track.

The difference in back-to-back spacing means that SST vehicles are not able to safely negotiate Network Rail infrastructure. It was therefore necessary to design the flangeback of the Tram-Train profile to provide the correct flange thickness and back-to-back spacing for running in grooved rail, whilst providing a checking surface at the correct back-to-back spacing for compatibility with Network Rail check rails and S&C. This was achieved by creating a stepped wheel flangeback, providing two sections with different effective back-to- back spacings.

As a result, the Network Rail checking surfaces are located some distance up the back of the wheel. To ensure on-going compatibility, Network Rail is required to raise the check rails throughout the Tram-Train route by 50mm. The adoption of a stepped wheel flangeback will also require that SST increase the lateral clearance to guard rails over bridges and viaducts to to prevent unwanted contact with the stepped region of the flangeback.

The design of the wheel flange was also a key part of the study. A particular constraint was the requirement to demonstrate that the new wheel profile could negotiate Network Rail switch tips with the permitted maximum level of residual switch opening. Analysis work found that the SST profile clashed with the switch tip in this condition and it was therefore necessary to adopt a flange angle which was similar to current approved Network Rail wheel profiles at 68°.

In relation to the existing SST wheel profile, the lower flange angle does reduce the absolute levels of flange climb protection, however vehicle dynamic simulations have proven acceptable derailment performance. In addition, precedents exist in other light systems such as Manchester Metrolink for running a heavy-rail flange angle within the tight curves of a tramway and the vehicle manufacturer will also optimise the vehicle suspension parameters to further mitigate any increased risk.

And the tips

A final key issue which was addressed in the study was the distinct differences in the flange tip geometries between SST and Network Rail profiles. On the SST system, as with many tramways, the wheel profile is designed to run on the flange tip when negotiating embedded crossings. The reason for this is to support the weight of the vehicle when passing over the discontinuities in the running rails. The flange tip of the Tram-Train profile therefore had to incorporate a flat flange tip design for compatibility with SST embedded track.

The prototype Tram-Train wheel profile will now be subject to scrutiny by the SST project team and the Rail Safety and Standards Board (RSSB). The IRR will also be collaborating with the vehicle manufacturer, Vossloh, to ensure that that the final vehicle design will be fully compatible with the Tram-Train route. Once the new wheel profile is approved for running on the mainline, Tram-Train test running can begin in earnest.

Report by Dr Paul Allen, assistant director of the University of Huddersfield’s Institute of Railway Research