Unlike conventional train-mounted ultrasound devices, the technology is not limited to speeds of up to 30–35mph, which will allow it to be fitted to ordinary passenger trains, dramatically increasing the number of inspections that can be carried out, said Dr Steve Dixon, leader of Warwick University’s ultrasound group.
Conventional systems are limited in speed because they are based on a transducer that both generates and receives the ultrasound waves. This means the transducer must move no more than 1mm in the time it takes the wave to travel down into the rail, bounce off the defect and travel back to the transducer. Even with waves travelling at 3,000 or 6,000m/sec, this means the train carrying the device cannot travel very fast.
The new system is based on a separate transducer and detector, 20cm apart, which creates a time window long enough for the wave to travel along the surface of the rail and back up to be received by the detector — even when mounted on a train travelling at high speeds.
‘Once we can do things at speed, then we can start thinking about putting them on normal passenger trains.
‘You don’t need to put them on every train, as long as you ensure your section of track is inspected on a daily basis, that is still far more than is being done at the moment,’ said Dixon.
‘There is not only an improvement in safety to be had, but also a big gain in efficiency, and therefore cost savings.’
At present the team has only focused on detecting defects near the surface of the railhead such as gauge corner cracking, where a shallow crack can block the ultrasound from penetrating down to a deeper crack behind it. ‘In the case of surface defects, such as gauge corner cracking, we look at how much of the ultrasound and what frequencies propagate underneath the crack; that gives us a double check on the depth of crack,’ said Dixon.
While this in itself would be a significant benefit, the researchers also plan to investigate new approaches for inspecting the whole of the railhead at high speed, using ultrasound arrays.
These arrays could be built to generate a number of small beams scanning in different directions at once; or to use a number of different receivers in various positions, to form a complete picture of the inside of the railhead.
The EPSRC-funded project is based on previous laboratory work into the ultrasound technique carried out by the researchers. The team will initially use Birmingham University-based Rail Research
The project also includes Corus, Serco Railtest, ultrasonic testing specialist NDT Solutions and RWL (Roger West Laboratories).