TSC (UK) has developed a prototype electromagnetic inspection technique, capable of detecting cracks, such as Rolling Contact Fatigue at high-speeds. Early research has shown this technique can detect such cracking at above 70km/h, and determine the severity.

TSC has adapted its world renowned alternating current field measurement (ACFM) technology to the inspection of cracking in tracks, by modifying their existing instrumentation, and using digital signal processing algorithms, to substantially increase the speed of detection. These advances in instrumentation, combined with a novel approach to electromagnetic probe design, have created a non-contact inspection system, capable of detecting cracking at high-speed.

ACFM is a non-contacting electromagnetic sensor technology used to detect and size, surface breaking cracks in metals. The ACFM probe induces a high-frequency current along the top surface of the inspection piece. This current is perturbed by the presence of the surface breaking defect, in such a way that the magnetic field above the surface of the metal changes. This change is registered by orthogonal sensors, close to the surface, and the measurements compared to numerical models to determine severity. This approach differs from standard eddy-current technique, in that the lift-off effect can be greatly reduced, which increases probability of detection (PoD). In the final version, any lift-off will be measured by the University of Birmingham using optical methods.

The technology is more typically used for the detection of fatigue cracks in large offshore structures, however, it has been adapted for the detection of rolling contact fatigue in train tracks. There are three main problems in achieving this goal. The first problem is the design of the sensor. It must be sensitive enough to detect rolling contact fatigue, but robust enough to withstand running close to the rail. The second is the speed of the instrumentation. This speed has been dramatically increased by the use of digital signal processing (DSP) boards, running an optimised algorithm which efficiently processes the data from the probe. Parallel processing has been proven to work at speeds up to 70 km/h.

The final problem, is that of deployment. The probes must be placed close to the rail, without being damaged at high-speeds. ISQ and the University of Lisbon are currently designing this mechanism, that will allow the probes to run along the rail, without damage or excessive vibration or lift-off.

Site trials are expected to take place early in 2012, in Portugal. This paper reports the initial results from laboratory trials, using real and artificial defects.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £110 +P&P)