Post by stanmorek on Mar 15, 2008 23:23:46 GMT
TRAIN BASED INSPECTIONS
Track Recording Vehicle (TRV) – Dynamic inspection of track geometry under the load of a moving train.
Frequency of TRV runs are normally every 8 weeks (16 weeks some areas e.g. Depot roads, Olympia, Chesham, etc)
Reports of each TRV inspection run are sent to track managers listing the following parameters at noted chainage:
Left top of rail
Right top of rail
Track alignment
Gauge
Cant
Corrugation
2m twist – wheel spacing
10m twist – bogie spacing
Percentage wheel unloading (calculated from base parameters)
Reports Generated
Common Defects of Track Geometry
Poor Top
(Vertical alignment)
Major cause of rough ride. If accompanied by variations in cant, can contribute to derailment.
Poor Line
(Horizontal alignment)
Unlikely to cause derailment unless accompanied by catastrophic failure such as buckling. Major cause of rough ride which increases side cut.
Track Twist
Twist is most likely to occur at:
2m twist is critical when greater than 25mm over 2m.
Note - TRV runs are less likely to detect dipped joints if the vertical alignment of both joints on LHR & RHR both dip the same magnitude at the same chainage!
Not being an electronics expert, whilst taking a ride onboard the TRV, the inspection manager explained that rail top and alignment is measured relative to the train. Equipment fitted to the bogie generate electrical impulses that feed into on-board computers working out the position of the bogie relative to the car body. What I understand is that a plate is fixed on top of the bogie within an electromagnetic field and as the train travels along it is moving up down generating pulses that are being recorded and translated into a linear measurement. Output gives rail top and alignment in the range of +/- 20mm on a vertical scale.
Historical Note - Hallade Track Recorder
An early form of measuring track quality from a moving train was the use of the Hallade machine during the 1930s on the mainline railway. This instrument was carried in an ordinary carriage and relied on a system of pendulums attached to a pen scribe-paper roll that recorded the oscallation of the carriage. In recent times a modern equivalent based on a system of gyroscopes in a tool box sized container was also employed on the end motor car of A-stocks on the Metropolitan line.
ULTRASONIC RAIL INSPECTION ON LONDON UNDERGROUND TRACK
Introduction
The introduction of ultrasonic rail inspection to the industry was driven by the Hither Green Crash at Willesden in 1967 that killed 49 and injured 78. Train wheel and rail meet in a contact patch typically 50-150 square millimetres (about half the area of a five pence piece). With typically 7.5 tonnes in this small area, permanent deformation and metal fatigue to the rail is inevitable.
Definitions
Broken Rail
A running rail which is fractured through its full cross-section or has a piece broken out of it which renders it unserviceable. Needs to be replaced within 36 hours.
Defective Rail
A running rail, unbroken but which will require premature removal from the track for any reason except when the removal is due to head side wear or corrugation. Actions to take with defective rail depends on severity of the defect.
Three main categories of defect:
Inherent – created at time of manufacture
Handling Defects – Installing rails
Service Defects – Train driving defects
Common Causes of Rail Failure
Dipped joints
Corrosion & chair gall at rail foot
Defects left in rail after welding
Physical damage at the rail end caused during installation
Defects formed during rail manufacture
Common Rail Defects
Star Crack – at bolt holes caused by dipped joints
Tach Ovale – sand/air bubble starts inside the rail from manufacturing process
Horizontal Defect – usually associated with dipped joints or physical damage
Crack from surface defect
Squat Defect - created at time of manufacture
Wheelburn - Train driving defect, e.g. excessive traction & braking
Basic Principles
What is ultrasound?
Identification of Flaws
100% reflection of sound between steel and air when two materials are in intimate contact – this forms the basis for detection of flaws.
Rail testing frequency
2.5MHz, Wavelength = 2mm Minimum Defect size determined in rail web
Types of Waves Transmitted
Compression Waves –
0 Degrees normal to running surface of rail for detection of horizontal cracks
Shear Waves –
40 Degrees for detection of Star Cracks
70 Degrees for Tache Ovales and wheel burn cracks
Typical Problems with Ultrasonic Inspection
Permissible Bolt Holes at Rail Ends
Minimum distance of hole edge from rail end 25mm.
No holes permitted to have an edge less than 25mm from another hole.
Max hole diameter – BH(32mm) FB(35mm)
Equipment & Techniques
070 and 040 testing are currently carried out manually on LUL system.
Probe - Piezo crystal used to generate ultrasound
Couplant - Water to improve transmission of ultrasound between probe and rail.
Instrument - Screen, pulse generator, time base generator, delay circuit, receiver amplifier.
070 Trolley
Walking Stick – Sliding probe pack 0 & 70 degree & water bottle
Tache Ovales grow in direction of traffic. Bi-directional traffic rail is tested twice both ways.
Testing Frequency
Frequency dependent on Track category (tonnage, location) A, B, C
e.g. Sub-surface Track 6 months (070), Sub-surface 2 months (040)
Inspection Area
Inspection of 91 separate work areas on LUL system
Ultrasonic Inspections Regime
Routine Inspections
Mandatory inspection of running rails using U1, U2 and U3 procedures for defects at bolt holes, tache-ovales and horizontal cracks.
Special Inspections
Programmed at discretion of track manager - LUL standards require specials for possible squat defects (U5) and longitudinal/vertical split (U3).
0 Degree testing – for Horizontal defects at rail end
40 Degree testing – for star cracks
U3 070 test – for tache ovale defects
U5 070 Testing – for squat & tache ovale type defects
U8 Testing – in event of loss of rail bottom (hydrogen shatter cracking)
U3 testing – vertical splits
Testing of New Welds
U6 Test - 1 in 10 new welds required to be tested.
Procedure:
Head & Web – 0 degree probe used on 500mm either side of thermit weld for horizontal cracks
Web – Rig & 2no.45 degree probes
Bottom section – single 45 degree probe
Ankle & toe – small 70 degree probe
Gauge Corner Defect (see below - Post Hatfield)
U14 Testing – Offset probe to detect tache ovale defects initiated from gauge corner.
WPIT – Wheel Probe Inspection Trolley U15 (U3 & U14) see below
Rail Defect Report Sheet & Ultrasonic Operator Action Codes
Ultrasonic & Line Actions in event of Code 1 or 2 Defect Identified
FRC (Fault Reporting Centre) advised Class 1 or 2 Ultrasonic defect who will not permit surrender of line until emergency action is effected or defect removed.
Emergency Action will include the following for:
Code 1 Defects: Impose 20mph ESR & fit emergency clamped fishplates, broken rail clamps or temporary bolted fishplates as appropriate.
Code 2 Defects: Fit emergency clamped fishplates, broken rail clamps or temporary bolted fishplates as appropriate or impose 20mph ESR.
Crack extending into Rail Head or Foot: Impose 5mph ESR.
Minimum Actions – TSR and emergency clamp and then:
Code A: Replacement within 36 hours
Code B: Replacement within 7 days
Code C: Replacement within 3 months
Post Hatfield Accident
Rail Gauge Corner Cracking
Implications for LUL/Infracos
LUL reassured that no serious problems would occur
U14 Ultrasonic Test
Implementing New U15 Ultrasonic test (combined U3 & U14)
Historical Note - Sperry Rail Service Train
The high rate of rail breaks on the American railroads due to internal defects in the 1930s prompted an ingenious development in flaw detection. The Sperry Company built a fleet of specialised Sperry self propelled detector cars with onboard equipment. The principal method of detection was to pass a low voltage electric current through the railhead to set up a magnetic field around the railhead. Underside coil equipment could detect the distortions in flux as the magnetic field passed through fissures inside the railhead. The Sperry cars had onboard accommodation for the car, were 57ft in length and weighed 56 tons. The operating speed during inspection runs was 5-9 miles per hour.
www.barp.ca/rail/srs/index.html
In modern times Sperry were contracted by Metronet to trial a track trailer fitted with ultrasonic inpsection equipment on the London Underground network. The trailer was towed by a track mounted Land Rover vehicle.
NB Some notes personal notes of mine taken from various documents and presentations.
Update:
Hallade Track Recorder
Sperry Rail Service Train
Track Recording Vehicle (TRV) – Dynamic inspection of track geometry under the load of a moving train.
Frequency of TRV runs are normally every 8 weeks (16 weeks some areas e.g. Depot roads, Olympia, Chesham, etc)
Reports of each TRV inspection run are sent to track managers listing the following parameters at noted chainage:
Left top of rail
Right top of rail
Track alignment
Gauge
Cant
Corrugation
2m twist – wheel spacing
10m twist – bogie spacing
Percentage wheel unloading (calculated from base parameters)
Reports Generated
- Report on Level 1 exceedences of maintenance limits specified in LUL Engineering standards
- Report on Level 2 exceedences requiring priority maintenance
- Continuous trace/tachograph read out of recordings every 260mm as the TRV moves along the track
Common Defects of Track Geometry
Poor Top
(Vertical alignment)
Major cause of rough ride. If accompanied by variations in cant, can contribute to derailment.
Poor Line
(Horizontal alignment)
Unlikely to cause derailment unless accompanied by catastrophic failure such as buckling. Major cause of rough ride which increases side cut.
Track Twist
Twist is most likely to occur at:
- Dipped joints;
- Voids between chairs/baseplates and sleepers;
- Voids under sleepers e.g. wet beds;
- On junction work;
- On adjustment/expansion switches;
- Tube tunnel mouths i.e. change of track support;
2m twist is critical when greater than 25mm over 2m.
Note - TRV runs are less likely to detect dipped joints if the vertical alignment of both joints on LHR & RHR both dip the same magnitude at the same chainage!
Not being an electronics expert, whilst taking a ride onboard the TRV, the inspection manager explained that rail top and alignment is measured relative to the train. Equipment fitted to the bogie generate electrical impulses that feed into on-board computers working out the position of the bogie relative to the car body. What I understand is that a plate is fixed on top of the bogie within an electromagnetic field and as the train travels along it is moving up down generating pulses that are being recorded and translated into a linear measurement. Output gives rail top and alignment in the range of +/- 20mm on a vertical scale.
Historical Note - Hallade Track Recorder
An early form of measuring track quality from a moving train was the use of the Hallade machine during the 1930s on the mainline railway. This instrument was carried in an ordinary carriage and relied on a system of pendulums attached to a pen scribe-paper roll that recorded the oscallation of the carriage. In recent times a modern equivalent based on a system of gyroscopes in a tool box sized container was also employed on the end motor car of A-stocks on the Metropolitan line.
ULTRASONIC RAIL INSPECTION ON LONDON UNDERGROUND TRACK
Introduction
The introduction of ultrasonic rail inspection to the industry was driven by the Hither Green Crash at Willesden in 1967 that killed 49 and injured 78. Train wheel and rail meet in a contact patch typically 50-150 square millimetres (about half the area of a five pence piece). With typically 7.5 tonnes in this small area, permanent deformation and metal fatigue to the rail is inevitable.
Definitions
Broken Rail
A running rail which is fractured through its full cross-section or has a piece broken out of it which renders it unserviceable. Needs to be replaced within 36 hours.
Defective Rail
A running rail, unbroken but which will require premature removal from the track for any reason except when the removal is due to head side wear or corrugation. Actions to take with defective rail depends on severity of the defect.
Three main categories of defect:
Inherent – created at time of manufacture
Handling Defects – Installing rails
Service Defects – Train driving defects
Common Causes of Rail Failure
Dipped joints
Corrosion & chair gall at rail foot
Defects left in rail after welding
Physical damage at the rail end caused during installation
Defects formed during rail manufacture
Common Rail Defects
Star Crack – at bolt holes caused by dipped joints
Tach Ovale – sand/air bubble starts inside the rail from manufacturing process
Horizontal Defect – usually associated with dipped joints or physical damage
Crack from surface defect
Squat Defect - created at time of manufacture
Wheelburn - Train driving defect, e.g. excessive traction & braking
Basic Principles
What is ultrasound?
- Sound generated by mechanical vibrations (16000Hz to 10 MHz)
- Transmission & Reflection of Sound through a medium (steel rail in this case)
- The speed of transmission of sound is dependent on material and is constant.
Identification of Flaws
100% reflection of sound between steel and air when two materials are in intimate contact – this forms the basis for detection of flaws.
Rail testing frequency
2.5MHz, Wavelength = 2mm Minimum Defect size determined in rail web
Types of Waves Transmitted
Compression Waves –
0 Degrees normal to running surface of rail for detection of horizontal cracks
Shear Waves –
40 Degrees for detection of Star Cracks
70 Degrees for Tache Ovales and wheel burn cracks
Typical Problems with Ultrasonic Inspection
- Poorly cut rail ends will cause irregular reflections
- Step joints prevent cracks being identified until 5mm in length
- Bond holes can mask defects
Permissible Bolt Holes at Rail Ends
Minimum distance of hole edge from rail end 25mm.
No holes permitted to have an edge less than 25mm from another hole.
Max hole diameter – BH(32mm) FB(35mm)
Equipment & Techniques
070 and 040 testing are currently carried out manually on LUL system.
Probe - Piezo crystal used to generate ultrasound
Couplant - Water to improve transmission of ultrasound between probe and rail.
Instrument - Screen, pulse generator, time base generator, delay circuit, receiver amplifier.
070 Trolley
Walking Stick – Sliding probe pack 0 & 70 degree & water bottle
Tache Ovales grow in direction of traffic. Bi-directional traffic rail is tested twice both ways.
Testing Frequency
Frequency dependent on Track category (tonnage, location) A, B, C
e.g. Sub-surface Track 6 months (070), Sub-surface 2 months (040)
Inspection Area
Inspection of 91 separate work areas on LUL system
Ultrasonic Inspections Regime
Routine Inspections
Mandatory inspection of running rails using U1, U2 and U3 procedures for defects at bolt holes, tache-ovales and horizontal cracks.
Special Inspections
Programmed at discretion of track manager - LUL standards require specials for possible squat defects (U5) and longitudinal/vertical split (U3).
0 Degree testing – for Horizontal defects at rail end
40 Degree testing – for star cracks
U3 070 test – for tache ovale defects
U5 070 Testing – for squat & tache ovale type defects
U8 Testing – in event of loss of rail bottom (hydrogen shatter cracking)
U3 testing – vertical splits
Testing of New Welds
U6 Test - 1 in 10 new welds required to be tested.
Procedure:
Head & Web – 0 degree probe used on 500mm either side of thermit weld for horizontal cracks
Web – Rig & 2no.45 degree probes
Bottom section – single 45 degree probe
Ankle & toe – small 70 degree probe
Gauge Corner Defect (see below - Post Hatfield)
U14 Testing – Offset probe to detect tache ovale defects initiated from gauge corner.
WPIT – Wheel Probe Inspection Trolley U15 (U3 & U14) see below
Rail Defect Report Sheet & Ultrasonic Operator Action Codes
Ultrasonic & Line Actions in event of Code 1 or 2 Defect Identified
FRC (Fault Reporting Centre) advised Class 1 or 2 Ultrasonic defect who will not permit surrender of line until emergency action is effected or defect removed.
Emergency Action will include the following for:
Code 1 Defects: Impose 20mph ESR & fit emergency clamped fishplates, broken rail clamps or temporary bolted fishplates as appropriate.
Code 2 Defects: Fit emergency clamped fishplates, broken rail clamps or temporary bolted fishplates as appropriate or impose 20mph ESR.
Crack extending into Rail Head or Foot: Impose 5mph ESR.
Minimum Actions – TSR and emergency clamp and then:
Code A: Replacement within 36 hours
Code B: Replacement within 7 days
Code C: Replacement within 3 months
Post Hatfield Accident
Rail Gauge Corner Cracking
- Small cracks form at the side of the rail head and are not detected by older test probes until in an advanced stage propagating into the centre of the rail head
- Likely on curved rail. Root cause is believed to be lack of lubrication & excessive sidecut
- Shattered high rail on curve (Hatfield derailment 2000)
- Low priority in pre-2000 engineering standard (Minimum Action for Defects)
- Not previously considered a problem (GCC relatively untested for)
- Low priority for ultrasonically untestable rails
Implications for LUL/Infracos
- Outside events can influence management of LUL network (ensure risks ALARP).
- LUL investigated and reviewed existing standards re: rail gauge corner cracking.
- Each line was checked for worst areas.
Specification for examination of selected rails from sites drawn up:
Ultrasonic testing
Visual examination
Surface Cracking
Chemical analysis
LUL reassured that no serious problems would occur
- Maximum depth of cracking was found to be 1.5mm (8-10mm deep crack is required for problems).
- Conclusion is that no general TSR’s or re-railing required on LUL network.
- New ultrasonic U14 test developed to test head of rail (U3 old standard test).
U14 Ultrasonic Test
- Update/Formulation of new standards to incorporate new U14 testing
- Written notice by The Permanent Way Engineer to LUL Standard P8401 – replacement of attachment 3.
- Update of Defect Minimum Actions tables.
- Untestable rails (UR) to be re-profiled by grinding and re-tested
Implementing New U15 Ultrasonic test (combined U3 & U14)
- Wheel Probe Inspection Trolley – Plant Approval
- Compatibility/Asset Interface – Change Review Process
- Changes to Standards and minimum actions – Concession (written notice by Asset Engineer)
Historical Note - Sperry Rail Service Train
The high rate of rail breaks on the American railroads due to internal defects in the 1930s prompted an ingenious development in flaw detection. The Sperry Company built a fleet of specialised Sperry self propelled detector cars with onboard equipment. The principal method of detection was to pass a low voltage electric current through the railhead to set up a magnetic field around the railhead. Underside coil equipment could detect the distortions in flux as the magnetic field passed through fissures inside the railhead. The Sperry cars had onboard accommodation for the car, were 57ft in length and weighed 56 tons. The operating speed during inspection runs was 5-9 miles per hour.
www.barp.ca/rail/srs/index.html
In modern times Sperry were contracted by Metronet to trial a track trailer fitted with ultrasonic inpsection equipment on the London Underground network. The trailer was towed by a track mounted Land Rover vehicle.
NB Some notes personal notes of mine taken from various documents and presentations.
Update:
Hallade Track Recorder
Sperry Rail Service Train
