Interviews

Tram-train: reaching a compromise

Dr Paul Allen, Assistant Director of the Institute of Rail Research at the University of Huddersfield and project manager for work on an optimal interface for the tram-train pilot in Sheffield, talks to RTM.

Among the many engineering challenges involved in the Sheffield tram-train pilot is the problem of developing a wheel that will run on both heavy and light rail infrastructure, with minimal wear and maintenance cost.

The University of Huddersfield’s Institute of Rail Research (IRR) has been working to develop a solution. RTM spoke to the IRR’s Assistant Director, Dr Paul Allen, who is project manager for this research.

The work aims to develop a compromise wheel profile for the two “very different infrastructures” of the two systems, Network Rail and Sheffield Supertram (SST). One is designed for light rail operations and has sections where the tracks are embedded in the street with a very flat rail head, whilst the heavy rail network is used for running typical Network Rail wheel profiles and has much smaller rail radius.

Dr Allen explained: “The geometric shape between the wheels and rails are very different for each system; you have to try to design a wheel which can work on both.”

This project is “a little more difficult” than the tram-train systems that have been running in Germany for decades, as the rail head sections of the track involved in the pilot are particularly diverse. This can make it more difficult to maintain an optimal wheel profile shape.

Optimised for worn rail

Because of the high curvature of the track, the Huddersfield team designed a wheelset based on the worn shape of the rail.

Various radii and geometric shapes were modelled using computer simulation to ensure “a reasonable conformality” between the rail and the wheel.

“We know that the rail section will wear to a certain shape, and as the tram-train will remain the minority fleet on both systems, we have designed the wheel profile to work with the worn rail profile shapes of each system,” he said.

“If the design is not optimised, the result can be severe two-point contact; one contact patch on the rail head and one on the rail gauge corner.

“This results in high levels of wheel-rail wear, you reduce the rail section life and the wheel life, combined with poor steering and higher forces between the wheel and rail.”

Striking a balance

A good initial design is essential to minimise the wear on both types of track, but Dr Allen said that no solution would ever be able to completely overcome the differences between the two systems.

“It’s about developing that initial design which offers the best compromise between the two section shapes.

“If you look at the shape of the rail sections from SST and Network Rail, it would not be possible to make a wheel profile which was optimal for both of them. They’re very different shapes.”

The modelling looked at the details of the contact between the wheel and the rail, including the location, size and shape of the contact patch on the railhead.

The level of conicity is also assessed, which governs how well the wheelset will steer around curves, “down to a reasonable radius”.

“We look at the distribution of curves on the route and we target the curve distribution on which we want the wheel to perform from a steering perspective.”

T-gamma – the energy dissipated in the contact patch – also allowed researchers to look at the likely wear rate of the wheel profile, in relative terms.

This output can then be compared to the existing wheel profiles which run on SST and on Network Rail to see if the profile is likely to wear more or less.

Check rails

But tram-train doesn’t just affect wheels. Check rails, which are put in place to guide the wheelset when the curve radius is quite small, must be raised to meet the different dimensions involved.

Network Rail and SST operate at different wheel back-to-back dimensions; on tram vehicles the spacing between the back of the two wheels is about 1380mm, but on heavy rail it’s 1360mm. It means the wheel profile has to be specially designed to be able to work at both flangebacks, and on Network Rail’s infrastructure the check-rails have to be raised to a new height to guide the wheel through tight curves. Obtaining gauging route clearance may be necessary on certain lines where tram-train is introduced to avoid gauge infringement.

“That will have to be looked at,” Dr Allen said.

Wider effects

There was “no doubt” that the pilot was having knock-on effects throughout the network, he added.

Sheffield city centre has had to be effectively re-railed to achieve the necessary wheel-flange design to guide vehicles safely through heavy rail tracks.

A taller wheel flange ensures it sufficiently overlaps the switchblade, otherwise there can be significant problems.

“The city centre track was significantly head worn which meant that the wheel-flange would have run on the bottom of the groove of the rail. That was quite a big problem.”

The new design could be used nationally, Dr Allen said, although complications with gauge infringements mean “you wouldn’t just be able to put it on any route”.

The design will now be submitted to the RSSB to be approved, before testing on the tracks can begin.

Tell us what you think – have your say below or email [email protected]