Copyright © TWI Ltd 2011 Carmen Campos Castellanos Yousef Gharaibeh NDT Technology Group TWI The application of Long Range Ultrasonic Testing (LRUT) to inspect railway tracks
Copyright © TWI Ltd 2011
Carmen Campos CastellanosYousef Gharaibeh
NDT Technology GroupTWI
The application of Long Range Ultrasonic Testing (LRUT) to inspect railway tracks
Copyright © TWI Ltd 2011
Contents
• MonitoRail Project overview– Rail industry need and market potential– Limitation of current inspection methods. – Project challenges
• Long range ultrasonic testing (LRUT) • The application of guided waves as an inspection
technique. – Previous work– Deformation shape of guided waves.– Investigation of different excitation conditions.– Experimental trials
• Conclusion and Future work.
Copyright © TWI Ltd 2011
MonitoRail project overview
• MONITORAIL: “Long range inspection and condition monitoring of rails using guided waves”
• Partly funded by the FP7 programme (Research for th e benefit of SMEs) over two years
• Project manager: Carmen Campos Castellanos -TWI Ltd
Jackweld
Copyright © TWI Ltd 2011
Rail industry need & market potential
Rail breaks England, Wales and Scotland (source: Ne twork rail)
• Recent advances in inspection and NDT techniques have drastically reduced the incidence of rail breaks.
• However, a residual number of rail breaks still occurs
Copyright © TWI Ltd 2011
Limitations of the existing NDT techniques
• Limitation in terms of reliability of defect detect ion (e.g. internal defects)
• Inspection speed• Maintenance is carried out in difficult conditions and
often at night• Inspection can be risky and dangerous operation• Can not cover the whole section of the rail (constr aints
in detecting defects in the rail foot)
Copyright © TWI Ltd 2011
Current inspection method
70 DegreeProbe
Coverage
37 DegreeProbe
Coverage
0 DegreeProbes
Coverage
RSU Tyre
0 Degree Probe
37 Fw Degree Probe
37 Rev Degree Probe
70 Fw G, C & F Degree Probes
70 Rev G, C & F Degree Probes
Copyright © TWI Ltd 2011
Limited defect sensitivity in the foot
• Currently there is no method to detect foot defects other than those directly beneath the web of the rail.
• Detection of defects in the rail head and web will also be investigated in order to provide a cost effective s olution.
Possible NotPossible
NotPossible
Copyright © TWI Ltd 2011
Project objectives
• To inspect critical areas where the probability for defects is high and there is limited access to carry out the c onventional NDT techniques.
• To inspect long lengths of rail track from a limite d number of access points.
• To achieve full volumetric coverage of the rail.
• To develop a cost efficient techniquefor condition monitoring.
• To extend the life of the rail throughearly repairs of rail tracks.
Copyright © TWI Ltd 2011
Project challenges
• Accessibility
• Environmental conditions:– Rain/snow– Temperature -20 to 60 Celsius degrees.
• Interface to rail engineering/ operation staff• Existing features on the rail attenuates the signal
Copyright © TWI Ltd 2011
LRUT- Ultrasonic Guided Waves
• Much lower frequency than conventional ultrasonics
• Equivalent to Lamb waves • Use a ‘wave guide’ - a regular cross section• Complex due to large number of wavemodes
0 20 Hz 20 kHz 1GHz
Infra sound Audible sound Ultrasonic Hyper sonicFrequency
Copyright © TWI Ltd 2011
Conventional Vs LRUT
Weld
Metal loss
Metal loss
FlangeConventional Transducer
Weld
Metal loss
Metal loss
FlangeTeletest® Tool
Guided Wave
100% Inspection
Localised Inspection
Copyright © TWI Ltd 2011
Adopting Guided waves as Long Range Ultrasonic Inspection technique
Copyright © TWI Ltd 2011
Railway track cross sectional surface (BS113A)
158.75mm
11.11mm
139.7mm
35.9mm
86.7mm
69.9mm
Head
Web
Foot
Copyright © TWI Ltd 2011
Dispersion Curves (modelling results)
Y Gharaibeh, et all“Investigation of the behaviour of selected ultrasonic guided wave modes to inspect rails for long-range testing and monitoring ” Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, pp. 225: 311 (2011)
Copyright © TWI Ltd 2011
Different possible wave modes for different sections in the railway track.
• Sole existence in each section in the railway track.
• Similar vibration patterns.
• Displacement in the entire section suggests 100% coverage of the cross sectional surface of the railway track.
Y Gharaibeh, et all“Investigation of the behaviour of selected ultrasonic guided wave modes to inspect rails for long-range testing and monitoring ” Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, pp. 225: 311 (2011)
Copyright © TWI Ltd 2011
Wave mode characterisation (Dispersion Curves in the foot)
Copyright © TWI Ltd 2011
Wave mode (F7) characterisation
Displacement distribution across the width of the foot of F7 wave mode
Deformation shape of the F7 wave mode
Copyright © TWI Ltd 2011
Wave mode (F4) characterisation
Displacement distribution across the width of the foot of F4 wave mode
Deformation shape of the F4 wave mode
Copyright © TWI Ltd 2011
Wave mode (F2) characterisation
Displacement distribution across the width of the foot of F2 wave mode
Deformation shape of the F2 wave mode
Copyright © TWI Ltd 2011
Investigating different excitation conditions
Copyright © TWI Ltd 2011
Excitation of F2 scenario 1
Copyright © TWI Ltd 2011
Excitation of F2 scenario 2
Copyright © TWI Ltd 2011
Excitation of F2 scenario 3
Copyright © TWI Ltd 2011
Experimental trials
• TWI rail – feature free
• Birmingham University rail – features: weld and clips
Copyright © TWI Ltd 2011
TWI sample -Transducer arrangement
Copyright © TWI Ltd 2011
Surface preparation and Sensor attachment
Copyright © TWI Ltd 2011
Defect addition
Copyright © TWI Ltd 2011
Defect detection sensitivity
0 300 500 1,000 1,500 2,000 2,500 3,0000
1
2
3
4
5
6
Time (us)Am
plitu
de (m
v)
Time Domain Signal
DeadZone
(a)
0 300 500 1,000 1,500 2,000 2,500 3,0000
1
2
3
4
5
6
Time (us)
Am
plitu
de (m
v)
Time Domain Signal
DeadZone
(b)
0 300 500 1,000 1,500 2,000 2,500 3,0000
1
2
3
4
5
6
Time (us)
Am
plitu
de (m
v)
Time Domain Signal
DeadZone
defect 2mm
defect 4mm
Copyright © TWI Ltd 2011
Birmingham University rail sample
Copyright © TWI Ltd 2011
Work plan
• To determine the effect in the wave mode propagation caused by common rail features such as clips and welds.
• To identify responses due to the rail features and to monitor the signal over time in order to detect any significant change over time that might indicate the presence of a defect.
• This work is still in progress.
Copyright © TWI Ltd 2011
Conclusion
• The characteristics of ultrasonic guided waves in t he rail complex geometrical profile have been identifi ed
• A suitable wave mode with full volumetric coverage in has been identified for each section of the rail.– F2 has been selected as the wave mode most suitable to
inspect the foot
• An improved excitation/reception conditions has bee n proposed.
• Defect detection sensitivity test have been conduct ed
• Experimental validations of the models are in progr ess
Copyright © TWI Ltd 2011
Future work • Improving of the quality of the propagated wave by
using:– Minimise coherent noise. – Apply phase delay.– Apply signal weighting technique. – Enhanced signal to noise ratio.
• Further experimental validations using – Railway track with feature free specimen– Railway track with clamps mounted on the
specimen. • Further signal processing analysis is needed. • Investigate exisiting wave modes in the rail head wi th
respect to the problem definition.
Copyright © TWI Ltd 2011
MONITORAIL acknowledgement
MONITORAIL is collaboration between the following organis ations:TWI Ltd, Vermon SA, OpenPattern, Aerosoft S.p.A, Jackweld L td,Network Rail Infrastructure Ltd, Cereteth and Brunel Unive rsity. TheProject is co-ordinated and managed by TWI Ltd. and is partly fundedby the EC under the Collaborative project programme- Resear ch forSMEs & Research for SME Associations. Grant Agreement Numbe r.26219.
Jackweld
Copyright © TWI Ltd 2011
Thanks for your attention
Questions?