The ballastless rail fastening systems supplied by Edilon can play a large part in tunnel safety, says the company.

Its unique construction enables passengers to be evacuated more quickly in the case of accidents because it is easy to walk on the flat surface, which also enables ambulances, fire trucks and maintenance vehicles to drive along the tunnel. Materials meet the F1 classification for smoke and fires and maintenance is low. In addition, because the rails lie lower than with conventional sleepers, tunnel height can be reduced, saving on construction costs. And bridges can be made lighter because the weight of the embedded rail system is less than ballast track.
The Edilon Corkelast Block System comprises blocks with traditional rail fasteners. Together, they form an elastic and watertight rail connection system. The system offers a short installation time; reduction in noise and vibration; and an integrated drainage system in the concrete slabs that improves water run-off. The rails are glued in either metallic or concrete troughs which provide continuous support for the rail.
Edilon's Direct Fastening system is based on its Dex-R 2K adhesive, which ensures durable adhesion of the anchor in difficult circumstances while also providing electrical insulation.
The system comprises a specially designed anchor bolt, nylon washer, spring and security cap offered as a complete kit, including the adhesive in an easy-to-handle cartridge system. This fastening can be adapted to almost every situation and the rail can easily be re-aligned even after several years of service.

Custom solution

In Sweden, construction of the 1050 m long Chalmers double tunnel in Gothenburg required a customised solution because the tunnel runs under a hospital and a technical college. Vibrations had to be minimised because of the presence of sensitive apparatus and there was also the need to eliminate electromagnetic interference.
Using GPS technology and a laser guided slipform paver, reinforced floating slabs were constructed in the two tunnels including troughs for the rails. The rails were installed in the troughs and embedded using an Edilon Corkelast automatic application machine.
Edilon also supplies an Acoustic Webblock, a prefabricated block based on a special type of Corkelast. Blocks are glued in the web of the rail and isolate noise and vibration. Tests have proven a significant reduction of noise emission of the rail, in particular on tight curves. The webblocks are offered for every type of rail, do not pollute the subsoil and can be recycled afterwards.

But as our scientific knowledge increases, methods of investigation and analysis improve. This serves to remind us how little we know about some aspects of the world around us. Although railway engineering has continually developed since the very first train, there is still much more to learn according to Dr Michael Burrow, Research Fellow at Rail Research UK. This organisation has been established to help build a bigger, better and safer railway in Britain by providing a strong and coherent engineering science base for railway systems research. It is based on a number of universities that work closely together.
It brings together researchers, representatives from industry and policymakers to help define, prioritise and guide research, and use the results to challenge policy, inform policy development and improve industry practice.
At Leominster, in Herefordshire, it has installed a number of pieces of equipment to help monitor the state of the track over time. Leominster is a live railway line where operator Network Rail and maintenance company Carillion Rail, have enabled access. Equipment includes accelerometers, rail temperature gauges, void meters to measure sleeper movement, soil moisture probes and a weather station which helps monitor soil moisture deficit.

Changing outlook

Dr Burrow says it is now recognised that financial investment in railways must be accompanied by advances in infrastructure technology if the reliability and efficiency of rail services are to improve. But our current understanding of trackbed and subsoil performance is poor, particularly in the context of a high speed and heavily trafficked railway. And with the accounting move to price projects on whole life cost, the development of a proper scientific understanding of the dynamic load-deformation response and track/sub-base interactions would have enormous benefits in the design of new and replacement track systems and the development of remediation and maintenance strategies.
As an initial project, the University of Southampton will obtain field data by monitoring sections of new track. This will have the advantage of quality site investigation data for the characterisation of the sub-base materials at a site where the design is fully documented.
Against the data from this project, The University of Birmingham will monitor the Leominster site. This is a long-used site that has a clay sub-grade and has suffered considerable problems on the heavily loaded up-line, which carries steel from South Wales. Researchers at Birmingham have already used this site extensively for their ISERT project (Improving the Stiffness of Existing Railway Track), where they are trialling alternative techniques for improving sub-grade stiffness without the removal of track structure.
Monitoring will build on that carried out on the West Coast Line in Sweden and will include piezometers, TDR soil moisture probes, pressure cells, accelerometer strings andmulti-depth deflectometers. Analysis will include geophysical techniques such as the cross-hole shear wave method, which provides realistic measurements of the in-situ sub-base stiffness. The usual triaxial tests may underestimate this stiffness by an order of magnitude, says Burrow.
In addition, the effects of cyclic principal stress rotation, as opposed to cyclic compression, will be considered. Principal stress rotation results in a decrease in the resilient modulus and can significantly increase the rate of permanent deformation of thesub-base.
Another collaborative research project between the mechanical engineers at the University of Sheffield and metallurgists at the University of Birmingham within Rail Research UK will further develop the Whole Life model by incorporating detailed metallurgical data at themicro- and nano-scale into mechanical models. This will include predicting fatigue crack initiation.
Rail life is determined largely by the degradation processes, both wear and fatigue, that occur because of the cyclic loading from wheels. As the crack tip is advancing by fatigue, wear at the surface leads to crack truncation at the mouth. The net crack growth rate is the difference between crack tip propagation and crack mouth truncation rates. If the wear rate is high, this wear fatigue interaction ensures that with time the crack shortens and may even be completely removed.
The wear process is controlled primarily by the hardness of the rail surface and can be modelled using established empirical equations. Combining crack initiation, crack propagation by fatigue and crack truncation by wear into one single Whole Life model leads to a better insight into rail behaviour. This approach can be developed to provide a guide to maintenance and renewal strategies.
Current models show how failure in the surface leads to wear as well as crack-like features. Detailed micro and nano-hardness and ductility values from an actual surface are required along with microstructural characterisation of the steel both before service and during service. This information, along with the links to the micro-mechanisms of fatigue crack initiation, will form an input into the model of wear fatigue interaction during crack initiation and early growth.

As part of Rail Research UK, The Institute of Sound and Vibration Research at Southampton University and the Rail Technology Unit at Manchester Metropolitan University are collaborating to create a complete, validated model of curve squeal noise generation. This will analyse vehicle curving behaviour, friction characteristics and excitation due to unstable forces between wheel and rail. It will include wheel and track dynamic response and acoustic radiation. Investigations will also assess friction and wear characteristics for roughness growth and develop a preliminary model for roughness growth, friction and wear.
Curve squeal originates from the unstable response of a railway wheel that is subject to large creep forces between the rail and wheel when negotiating curved track. The wheel is excited particularly at the frequencies corresponding to its natural modes. The radiated noise has most of its energy between 250 Hz and 10k Hz and is often dominated by a single frequency. The level of sound can be intense and cause annoyance. Solutions have included adding damping to the wheel or using water sprays. However, it is desirable to have a fundamental understanding of the causes before further action.
Rolling noise is generated by surface roughness of the wheel and rail, with the most important wavelengths in the range 10 to 250 mm. A key outstanding question is how the roughness develops over time. In severe cases a periodic wear pattern - corrugation - may form, giving large increases in rolling noise. Although considerable previous research has been carried out on corrugation growth, the growth of broad-band roughness is much less well understood. The dynamic behaviour at the wheel-rail interface is vital for understanding both curve squeal and roughness growth.