Transverse crack detection in rail head using low frequency eddy currents

Information

  • Patent Grant
  • 6768298
  • Patent Number
    6,768,298
  • Date Filed
    Tuesday, July 17, 2001
    22 years ago
  • Date Issued
    Tuesday, July 27, 2004
    19 years ago
Abstract
A method and system for detecting transverse cracks beneath horizontal cracks in the rail way track. As a transporter moves over rail, a saturation magnetic field is generated into and across the rail head using a toroidal-shaped DC saturation magnet located a predetermined distance above the rail head. Any transverse cracks in the rail head are detected with a low frequency eddy current probe mounted centrally between the opposing pole ends of the DC magnet and over the rail head. A force is applied to the low frequency eddy current probe to hold the probe towards the rail head as the transporter moves on the rail so as to follow the wear pattern of the rail head and to minimize lift-off. A second sensor is used to sense the presence of non-relevant indications that falsely indicate possible transverse cracks by the low frequency eddy current probe.
Description




BACKGROUND OF THE INVENTION




1. Field of the Invention




The present invention relates to the detection of transverse cracking in rail heads and, more particularly, to using a low frequency eddy current system designed to detect transverse cracks in rail heads.




2. Statement of the Problem




It has been reported that 40% of rail failures are due to the occurrence of transverse cracking in the rail head. An important goal in the railroad industry is to detect such transverse cracks before such failure occurs. A transverse crack is a progressive transverse fracture occurring in the head of the rail and propagating under cyclic fatigue.




One conventional nondestructive approach for detecting transverse cracking is the use of conventional ultrasonic methods involving roller search units and contact transducers. The problem with such conventional ultrasonic approaches is that transverse cracks often occur under horizontal cracks (shelling/air gaps) which reflect ultrasound thereby masking the transverse crack. Horizontal cracking is a condition in which the cold worked layer of material above the transverse crack separates and flows over the remaining rail. This “shelling” of the rail head is a condition of the contact fatigue generated due to wheel/rail interaction.




A need exists to detect transverse cracks beneath horizontal cracks before the transverse crack grows or fails in service.




In Earnest, Katragadda, Si, and Garcia, “Transverse Crack Detection in Railroad Heads,” Oral Presentation at the 6


th


NDE Topical Conference, Apr. 19-22, 1999, San Antonio, Tex., initial results were disclosed of an experimental low frequency eddy current approach designed to detect transverse cracks in rail heads. While this presentation reported results from static (stationary) testing, it did verify the ability to detect transverse cracks under horizontal cracks in rail head. Horizontal cracks are parallel to the direction of the saturated magnetic field making the probe relatively insensitive to the horizontal cracks and sensitive to any underlying transverse cracks. The static tests not only were able to detect transverse cracks under horizontal cracks, but were able to provide output signals indicating the approximate size of the transverse crack. It was further observed that higher frequencies for the eddy current did not penetrate well into the rail head whereas lower frequencies of 70 to 100 hertz provided deeper penetration. The Earnest et al. presentation speculated that the static feasibility test observed could be implemented into a push cart wherein brushes could be added to the saturating magnetic poles to increase the contact with the rail head.




A need exists to provide a low frequency eddy current system that moves along the rail at a velocity sufficient to detect transverse cracks in rail heads of actual rail tracks, that provides an eddy current probe for such moving inspections, that follows a rail head having a wear profile, that minimizes any lift-off of the probe while moving, and that couples the magnetic saturation to the rail.




A need exists also to reject those signals from the low frequency eddy current probe that are sensed to be transverse cracks, but are from non-relevant indications such as from thermite welds, plant welds, rail end joints, etc.




SUMMARY OF THE INVENTION




The present invention solves the aforestated problems by providing a novel method and system for increasing the detection of transverse cracks beneath horizontal cracks in the rail. The novel method and system of the present invention provides a low frequency eddy current system that moves along the rail at a velocity sufficient to detect transverse cracks in rail heads, that provides a low frequency eddy current probe for such moving inspections, that follows the wear profile of the rail head, that minimizes any lift-off of the probe while moving, and that couples the saturation magnet to the rail. The novel method and system of the present invention also utilizes a separate sensor such as a Hall element sensor to distinguish the low frequency eddy current signals, obtained from a transverse crack, from non-relevant indications such as thermite welds, plant welds, rail end joints, etc.




The novel method of the present invention provides moving a transporter on the rail at a velocity, generating a saturation magnetic field into and across the rail head using a DC saturation magnet mounted to the transporter a predetermined distance above the rail head, inductively coupling the opposing pole ends of the DC saturation magnet with the rail head, detecting transverse cracks in the rail head with a low frequency eddy current probe mounted centrally between the opposing pole ends of the DC saturation magnet and over the rail head, applying a force on the low frequency eddy current probe against the rail head as the transporter moves on the rail at the velocity so as to follow the wear pattern of the rail head, and to control the lift-off of the probe from the rail head.




The novel system of the present invention uses a transporter that moves along the track. A toroidal-shaped DC magnet mounted to the transporter with its opposing ends inwardly directed towards each other and aligned over the rail head to increase saturation. An inductive coupling provided between each of the poles and the rail heads increases magnetic saturation of the rail head. A low frequency eddy current probe centrally located between the poles of the toroidal-shaped DC magnet senses the presence of transverse cracks in the rail head. A carriage mounted to the transporter forces the low frequency eddy current probe towards the rail head at the central location. Protective material on the probe abuts the rail head as the transporter moves thereby protecting the low frequency eddy current probe from damage while permitting the probe to follow the wear profile. At least one wheel located on the carriage engages the rail head for controlling lift-off of the low frequency eddy current probe from the rail head.











BRIEF DESCRIPTION OF THE DRAWINGS





FIG. 1

is a prior art illustration of a rail head cross section showing a horizontal crack and a transverse crack lying underneath.





FIG. 2

is a side view illustration of the transporter of the present invention carrying a saturation magnet and the low frequency eddy current probe of the present invention.





FIG. 3

sets forth the details of the toroidal-shaped DC saturation magnet of the present invention.





FIG. 4

sets forth the dimensions of the toroidal-shaped DC saturation magnet of FIG.


3


.





FIG. 5

is a perspective view of the low frequency eddy current probe of the present invention.





FIG. 6

is an illustration of the low frequency eddy current probe of

FIG. 5

moving on the surface of the rail head.





FIG. 7

is a cross section of the low frequency eddy current probe carriage of the present invention.





FIG. 8

shows the low frequency eddy current probe carriage of

FIG. 7

on a rail head.





FIGS. 9



a


and


9




b


set forth an alternate embodiment of the present invention wherein two probes are used on the rail head.





FIG. 10

is a side view of the carriage of

FIG. 7

adapted, in an alternate embodiment, to carry all sensors.





FIG. 11

is an output screen of the eddyscope showing a transverse crack detection signal.





FIGS. 12



a


,


12




b


, and


12




c


show the output signals from the Hall sensors for a joint in the rail head.





FIG. 13

is a flow chart showing the operation of the low frequency eddy current probe of the present invention.





FIG. 14

is a bottom view of another embodiment of the probe carriage of the present invention.





FIG. 15

is a side view of the probe carriage of FIG.


14


.





FIG. 16

is an end view of the probe carriage of FIG.


14


.











DETAILED DESCRIPTION OF THE INVENTION




1. Overview. In

FIG. 1

, a prior art rail head cross section


20


is shown. The rail head


10


thickness T is typically about 1.5 inches. An area of shelling


30


is shown to have a horizontal crack


40


. Under the horizontal crack (shelling/air gaps)


40


is a transverse crack


50


which is typically formed in a kidney shaped pattern. The transverse crack


50


is a defect that grows in spurts sometimes resulting in concentric growth rings


60


. The transverse crack


50


reaches a certain dimension and the remaining rail undergoes brittle failure.




In

FIG. 2

, the system


100


of the present invention is shown on rail


90


. The system


100


is mounted to a transporter


110


having conventional rail wheels


120


. The transporter


110


moves in the direction


130


powered in any of a number of conventional approaches. In

FIG. 2

, the transporter


110


is shown as a cart on rail


90


which is moved along the rail


90


at a velocity, V. Shown, by dotted lines, is a conventional means


200


for moving the transporter


110


. Such conventional means


200


includes a motor of any type, a coupling of any type for pushing or pulling the transporter


110


, a manually operated power source of any type, or any functional device that provides movement at velocity V in direction


130


. The term “transporter” is used to mean any structural means


110


for transporting the DC saturation magnet


140


and probe carriage


100


. This structural transporting means


110


includes a separate cart, as shown; incorporation into a vehicle such as a modified pick-up capable of traveling on the rail or on a road; incorporation into a train car; or incorporation into any suitable device for conveying the magnet


140


and the probe


150


along the track


90


.




The low frequency eddy current system


100


uses a toroidal-shaped DC saturation magnet


140


and a low frequency eddy current probe


150


in a carriage


154


. Brushes


160


are mounted at the pole ends


142


of the saturation magnet


140


. Brushes


160


provide an inductive coupling between the pole ends


142


and the rail head


10


. The probe carriage


154


is mounted to the cart


110


between the pole ends


142


with support hangers


152


. The probe


150


abuts the upper surface


12


of the rail head


10


to maintain contact with the upper surface


12


. The saturation magnet


140


is mounted to the transporter


110


with aluminum supports so as not to interfere with the saturation fields.




In operation, the DC saturation magnet


140


provides a saturated magnetic field


170


in the rail head


10


. The DC saturation magnet


140


has two purposes: first, to reduce the skin effect for the low frequency eddy current probe


150


by reducing the effective rail head material permeability and; second, to eliminate signal variations due to random variations of permeability in the rail head material.




The probe


150


, as will be discussed, induces a current


180


in the rail head


10


. Eddy currents are alternating electrical currents which are induced in the rail head by the probe


150


. The induced flow pattern is disturbed by the presence of a transverse crack


50


. The disturbance in the flow pattern in turn affects the associated magnetic field which is detected.




2. DC Saturation Magnet. In

FIG. 2

, the DC saturation magnet


140


is shown to be toroidal in shape. In

FIGS. 3 and 4

, the details of the toroidal-shaped saturation magnet


140


are shown. In this embodiment, the saturation magnet


140


has a toroidal-shaped core


400


formed of 11 mil laminated silicone iron layers which are glued together. The material is in this embodiment EL-COR No. T0803-009 or comparable material. The dimensions for the magnet


140


of

FIG. 3

are shown in

FIG. 4

are set forth in Table I:

















TABLE I












Dimension




Min.




Max.







Reference




(inches)




(inches)




(inches)





























400




24.00




23.875




24.125







410




12.00




11.813




12.187







420




18.00




17.938




18.25







430




3.00




3.00




3.031







440




3.00




2.969




3.031















In

FIG. 3

, layers of spiral windings


410


are shown. In the preferred embodiment, four layers of windings


410


are provided for a total of 1,082 turns. Ten gauge solid enamel wire is used. It was observed for this embodiment that approximately 1,000 turns is optimal. More turns caused heat to build up and less turns reduced the saturation effect.




The saturation magnet


410


is powered by a DC power source


420


which in the above embodiment provides 20 amps into the turns


410


in a voltage range of 40-50 volts DC. The pole ends


142


of the saturation magnet


140


are spaced


450


from the top surface


12


on the rail head


10


, in this embodiment, in a range of about 0.5 to 0.75 inches. The brushes


160


are longer than the spacing


450


and each brush


160


contains 5 mil diameter carbon steel bristles.




In

FIG. 2

, the brushes


160


structurally function as a means


160


affixed to the opposing pole ends


142


for inductively coupling the DC saturation magnetic field


170


with the rail head


10


. The inductive coupling allows for slide ability over the rail head


10


. It is to be understood any suitable design or structure for performing this function could be utilized under the teachings contained herein. The brushes


160


also act to clean the surface


12


of the rail head


10


to remove dust or grease before the probe


150


slides over.




In contrast to the rectangular design used in the static test of the aforesaid Earnest presentation, the toroidal shape for the DC saturation magnet


140


of the present invention provides greater magnetic density


440


in the rail head


10


near the surface


12


. The intensity of the field


170


lessens downwardly into the rail


10


. The toroidal shape provides a greater cross-section of its core


400


in contact with the rail head


10


. The windings


410


shown in

FIG. 3

are more evenly distributed over the core and a smaller distance is provided between the pole ends.




The high intensity field


440


near the surface


12


reduces the skin-effect for the probe


150


and minimizes signal variations in probe


150


due to the rail head material. In comparison to the Earnest presentation, the toroidal-shaped saturation magnet


140


of the present invention provides a much higher intensity field


440


to eliminate manufacturing effects and material history effects that would vary the permeability of or near the surface


12


such as the horizontal cracks


40


previously discussed. As shown in

FIG. 3

, the opposing pole ends


142


terminate in an acute angle


144


. It is to be understood that in

FIG. 3

, the spacing


450


is exaggerated to illustrate the effect of the brushes


160


. This concentration of the magnetic saturating field


440


represents an improvement over the Earnest presentation which only disclosed the use of a set of opposing poles oriented vertical to the rail head


10


. The greater intensity


440


obtained with the acute angles


144


of a toroidal-shaped magnet


400


provide greater saturation to obtain the above two stated goals.




3. Low Frequency Eddy Current Probe. In

FIG. 5

, the details of the low frequency eddy current probe


150


are shown. In this embodiment, the diameter


500


of the coil


150


is 1⅞ inch. The core


510


is air and has a diameter


520


of ¼ inch. The height


570


of the coil is ¾ inch. The coil is made up of wire turns


530


which, in this embodiment, are 900 turns of 24 gauge enamel wire. The coil is immersed in epoxy resin


540


and has a protective TEFLON cap


550


. The cap


550


has a thickness


590


of 50 mil.




In

FIG. 5

, the coil


150


of the present invention is shown. It is to be understood that a number of conventional eddy current probes are known including those set forth (i.e., pancake, ferrite-core cylindrical, ferrite pot-core probe, and gage side ferrite) in the aforesaid Earnest presentation which operates with varying degrees of sensitivity under the teachings of the present invention.




In one embodiment of the present invention, an eddyscope


560


provides power over lines


580


to the coil


150


. In this embodiment, the coil


150


is driven at a low frequency such as 80 hertz. The gain of the eddyscope


560


is preferably set at 36.5. The probe


150


is a single coil which acts both as a sensor and an exciter. The eddyscope


560


monitors the impedance of the probe


150


. This embodiment uses a Model MIZ-22 from Zetec Inc., 1370 N.W. Mall St., P.O. Box 140, Issaquah, Wash. 98027.




As shown in

FIG. 6

, the cap


550


of the probe


150


abuts the surface


12


of the rail head


10


as the probe


150


moves with a velocity V. A force F is applied to the probe


150


to ensure contact of the probe cap


550


to the surface


12


especially to follow the wear pattern of the rail head


10


.




In

FIG. 7

, the probe


150


is mounted into a probe carriage


154


. The probe


150


is set into a formed cavity


700


of the probe carriage


154


and is held in place by two brackets


710


which are bolted


712


to the carriage


154


. This firmly holds the probe


150


in the carriage


154


. The TEFLON cap


550


acts as a protective coating and extends slightly below the undersurface


702


of the carriage


154


. The probe carriage


154


is mounted to the transporter


110


as shown in

FIG. 2

in any of a number of conventional fashions. Springs


730


are provided about the mounting bolts


152


to provide the force F to engage the cap


550


with the upper surface


12


of the rail head


10


. The force F causes the probe


150


to follow the wear profile of the rail surface


12


as the transporter moves. The force F also minimizes lift-off of the probe


150


from the rail surface


12


.




In

FIGS. 2

,


7


, and


10


, a first embodiment of a carriage


154


is shown and in

FIGS. 14

,


15


, and


16


a second embodiment is shown. The carriage


154


can be any structural means not limited to that shown, mounted to the transporter


110


, for carrying the probe


150


centrally between the opposing pole ends


160


of the saturation magnet


140


and over the rail head


10


of rail


90


. For example, the shape of the carrying means could be entirely curved rather than linear or a combination of both linear and curved. For example, the carrying means could be solid, as shown, hollow, or similar to a basket or bracket design. The shape or design can be any suitable configuration to structurally support the probe between the pole ends


160


over the rail head


10


as the transporter


100


moves along the rail.




In

FIGS. 7 and 10

, springs


730


are used to provide force F to probe


150


. While this is one embodiment any conventional means for applying the force F can be used. For example, any type of springs such as leaf springs could be used and any number of springs could be used such as 1, 3, or 4 or more. The force F could be other than a spring such as a pressurized gas force or a pneumatic force. Any structural means can be utilized under the teachings of this invention to apply the force F to the carriage


154


or, in a variation, directly to the probe


150


.




The probe carriage


154


is constructed of aluminum so as not to affect the flux path of the saturation magnet


140


nor to affect the eddy current in the probe


150


.




In

FIG. 8

, the carriage


154


is located over the centerline


800


of the rail head


10


thereby centering the probe


150


. In another embodiment, two probes could be utilized such as probes P


1


and P


2


in FIG.


9


. In FIG.


9


(


a


), the two probes are offset from the center


800


of the rail head


10


so as to concentrate their detection at the opposing sides


10




a


and


10




b


of the rail head


10


. Probes P


1


and P


2


are offset by a predetermined distance


900


as shown in FIG.


9


(


b


) but overlap to provide full coverage of the rail head.




In

FIGS. 14-16

and in another embodiment of the present invention, the probe carriage


154


is shown to have a plurality of wheels


1400


which ride on the surface


12


of the rail head


10


. In this embodiment, the probe


150


is located off the centerline


1430


by a distance


1420


. As shown in

FIG. 16

, this positions the probe


150


over to one side of the rail head


10


. The use of the wheels


1400


allow the carriage


154


to follow the surface


12


under the bias force F while minimizing lift-off of the probe


150


from the surface


12


. This is especially true when the rail head


10


is worn as shown in FIG.


16


. In which case, the wheels actually follow along on the wear profile of the surface


12


of the rail head


10


. Yet, the TEFLON cap


550


slideably engages along the surface


12


.




As shown in

FIG. 14

, four wheels


1400


are each mounted in a wheel well


1410


on an axle


1440


. It is to be expressly understood that while wheels are utilized, any suitable support means for controlling lift-off could be used under the teachings of the present invention. For example, such supports rather than being rolling wheels could be TEFLON pads. Any suitable support means for controlling lift-off and/or following the wear profile surface can be utilized under the teachings of the present invention. The use of wheels


1400


can occur with the embodiment shown in FIG.


7


. The present invention includes a number of variations concerning the force F, the carriage


154


, the probe


150


, the supports


1400


and the protective cap


550


. In one embodiment, the force F pushes the cap


550


in the carriage


154


against the rail surface


12


. In this embodiment, supports


1400


(e.g., wheels) can also be used or not be used. In a second embodiment, the force F pushes the carriage


154


with supports


1400


towards the rail surface


12


, but the cap


550


does not contact the rail surface. In which case, the supports


1400


engage the rail surface


12


. In this embodiment, the cap


550


may or may not be used.




4. Separate Sensor. In

FIG. 10

, the carriage


154


has separate sensors


1000


such as Hall element sensors mounted a predetermined distance


1010


to the low frequency eddy current probe


150


. The Hall sensors


1000


are two perpendicular Hall sensors H


T


(tangential) and HN(normal). The Hall element sensors


1000


are spaced above the surface


12


of the rail head


10


by an amount shown as


1020


which is typically around 1 mil. The Hall element sensors


1000


are mounted


1030


to the carriage


154


so as to take advantage of the force F as previously discussed (and/or in another embodiment the supports


1440


). How the Hall element sensors


1000


are physically mounted to the carriage is immaterial to the teachings of the present invention and a variety of conventional approaches could be utilized for providing this mount. Furthermore, while one variation has the Hall element sensors


1000


mounted to the carriage


154


near the probe


150


, in other variations of the present invention, the Hall element sensors are mounted in different relationships to the probe


150


either on the carriage


154


, within the carriage


154


(as shown by dotted lines


1000




a


) or at other locations on the transporter


110


. In the shown embodiment of

FIG. 10

, the sensors


1000


are located near the probe


150


and separated by a known distance


1010


.




The Hall element sensors


1000


provide added signal information which helps classify the low frequency eddy current signals produced by the probe


150


. In particular, the Hall sensors


100


have been included to detect a voltage perpendicular to the direction of current (from the saturating magnet). When a change in the magnetic field occurs a change in the polarity also occurs allowing the Hall voltage to measure magnetic flux density changes between thermite welds, plant (electric flash butt) welds, and rail end joints, to that of the parent rail material. These are termed “non-relevant indications.” Use of the Hall element sensors


1000


supplements the eddy current signals for classification of rail head conditions.




The Hall sensors


1000


are commercially available such as those available from Bell Technologies Inc., a SYPRIS Company, 6120 Hanging Moss Road, Orlando, Fla. 32807 (800)775-2550.)




It is to be expressly understood that in one variation of the present invention more than one set of Hall sensors


1000


could be used near the probe


150


.




While one embodiment of the present invention uses Hall sensors, it is to be expressly understood that any other separate sensor(s) could be utilized in conjunction with the low frequency eddy current probe


150


to generate signals that classify the signals generated by the low frequency eddy current probe. Any type of secondary (and, perhaps tertiary) sensor or group of sensors can be used in conjunction with the system


100


of the present invention to aid in the classification of data obtained by the eddyscope.




5. Signal Analysis. In

FIG. 11

, a typical low frequency eddy current signal from the eddyscope


560


is shown. The received eddy current signal from the probe


150


is shown as signal


1100


and the flaw signal corresponding to a transverse crack


50


is shown as signal


1110


. Signals


1100


and


1110


show changes in the inductance of the probe


150


over lines


580


as sensed by the eddyscope


560


. Signal


1110


represents the possible detection of a transverse crack. The eddyscope screen shows a phase vector or phasor diagram comparing amplitude and phase relationship of current signals at a set frequency. The Y-axis is inductive reactance and the X-axis represents resistance.




In

FIG. 12

, the operation of the separate Hall sensors


1000


is shown. When a non-relevant indication


1200


appears in the rail head


10


, the first Flail sensor H


N


delivers a wave form


1210


and the second Hall sensor H


T


delivers a wave form


1220


. The presence of these signals would cause the system


10


of the present invention, as will be explained later, to classify the corresponding transverse crack signal


1110


as shown in FIG.


11


.




6. Operation. In

FIG. 13

, the operation of the present invention is shown and starts in step


1300


. The process saturates


1310


the rail head


10


with the DC saturation magnet


140


. The DC saturation magnet


140


is powered by power supply


420


. The transporter


110


moves


1320


forward at a given velocity V and, during this movement the low frequency eddy current probe


150


and eddyscope


560


, obtains


1330


transverse crack defect signals. The probe


150


is powered


1340


at a given frequency. These signals are then analyzed


1350


. Once analyzed, the results are formulated and delivered


1360


which shows the detection, approximate size, and location of the transverse crack on the track. A number of conventional software packages are available for analyzing signals from eddy current instruments. First, the signal to noise ratio can be conventionally processed through averaging or other standard signal processing techniques. The improved signal can then be further processed through conventional neural network software, etc.




In an alternate embodiment, the separate sensor signals (e.g., from Hall sensors) are obtained in stage


1370


. As shown in

FIGS. 11 and 12

, when the probe


150


passes over a non-relevant indication


1200


the resulting signal


1110


in the eddyscope appears to sense a transverse crack


50


. But, the Hall sensors


1000


issue signals


1210


and


1220


when they pass over the non-relevant indication


1200


. The occurrence of signals


1210


and


1220


classify the corresponding signal


1110


as one corresponding to a non-relevant indication


1200


and, hence, stage


1350


ignores or rejects signal


1110


. In a variation to the present invention, this can be classified as a non-relevant indication in stage


1360


. For all other signals


1110


when corresponding Hall signals


1210


and


1220


do not exist, they are classified as transverse cracks


50


in stage


1350


and outputted in stage


1360


.




It is well known how to use velocity, V, GPS readings, etc. to physically locate actual defects on track in stage


1360


based on locating hardware, algorithms and software.



Claims
  • 1. A system for detecting transverse cracks in rail head on railway track comprising:a transporter on the railway track, said transporter moving the system along the railway track, a toroidal-shaped DC magnet mounted to the transporter with its opposing pole ends inwardly directed towards each other and aligned over the rail head, an inductive coupling between each of the poles and the rail head to magnetically saturate the rail head, the inductive coupling slideably engaging the rail head, at least one low frequency eddy current probe, centrally located between the poles of the toroidal-shaped DC magnet, for sensing said transverse cracks in the rail head, protective material on the low frequency eddy current probe, a carriage mounted to the transporter forcing said low frequency eddy current probe against said rail head, the protective material abutting the rail head when the transporter moves on the railway track thereby protecting the low frequency eddy current probe from damage.
  • 2. The system of claim 1 further comprising:one of a plurality of supports on said carriage engaging the rail head for controlling lift-off of the low frequency eddy current probe from the rail head.
  • 3. The system of claim 2 wherein said plurality of supports are wheels.
  • 4. The system of claim 1 wherein the toroidal-shaped DC magnet has four layers of windings extending down towards the opposing ends.
  • 5. The system of claim 1 wherein the inductive coupling comprises a plurality of carbon steel bristles mounted to each of the pole ends.
  • 6. The system of claim 1 wherein the low frequency eddy current probe is an air core coil.
  • 7. The system of claim 1 wherein the protective material is polytetrafluoroethylene.
  • 8. The system of claim 1 wherein the carriage comprises two spring-loaded supports.
  • 9. The system of claim 1 wherein the plurality of wheels is four.
  • 10. The system of claim 1 wherein the carriage locates the low frequency eddy current probe off the center of the rail.
  • 11. The system of claim 1 wherein the at least one probe is one probe, the aforesaid one probe centered over the centerline of the rail head.
  • 12. The system of claim 1 wherein the at least one probe is a plurality of probes, each of the aforesaid plurality of probes located across the rail head.
  • 13. A system for detecting transverse cracks in rail head on railway track comprising:a transporter on the railway track, said transporter moving the system along the railway track, a toroidal-shaped DC magnet mounted to the transporter with its opposing pole ends inwardly directed towards each other and aligned over the rail head, an inductive coupling between each of the poles and the rail head to magnetically saturate the rail head, the inductive coupling slideably engaging the rail head, at least one low frequency eddy current probe, centrally located between the poles of the toroidal-shaped DC magnet, for sensing said transverse cracks in the rail head, protective material on the low frequency eddy current probe, a separate sensor near said low frequency eddy current probe for sensing non-relevant indications in the rail head, said separate sensor held a predetermined distance above said rail head, said system rejecting a sensed transverse crack when the separate sensor senses a non-relevant indication, a carnage mounted to the transporter forcing said low frequency eddy current probe against said rail head, the protective material abutting the rail head when the transporter moves on the railway track thereby protecting the low frequency eddy current probe from damage.
  • 14. The system of claim 13 wherein the separate sensor is a Hall element sensor.
  • 15. A system for detecting transverse cracks in rail head on railway track comprising:a transporter on the railway track, said transporter moving the system along the railway track, a DC magnet mounted to the transporter with its opposing pole ends aligned over the rail head, an inductive coupling between each of the poles and the rail head to magnetically saturate the rail head, the inductive coupling slideably engaging the rail head, at least one low frequency eddy current probe, centrally located between the poles of the DC magnet and over said rail head, for sensing said transverse cracks in the rail head, a separate sensor near said low frequency eddy current probe for sensing non-relevant indications in the rail head, said separate sensor held a predetermined distance above said rail head, said system rejecting a sensed transverse crack from the low frequency eddy current probe in an area of the rail head when the separate sensor senses a non-relevant indication at said area.
  • 16. A method for detecting transverse cracks in rail head of a rail comprising:moving a transporter on the rail, generating a saturated magnetic field into and across the rail head with a DC saturation magnet mounted to the transporter a predetermined distance above the rail head while the transporter is moving, the saturation magnet having a toroidal-shape with opposing pole ends inwardly directed towards each other over the rail head, inductively coupling the opposing pole ends of the DC saturation magnet with the rail head, detecting transverse cracks in the rail head with a low frequency eddy current probe mounted centrally between the opposing pole ends of the DC saturation magnetic and over the rail head, applying a force to the low frequency eddy current probe against the rail head as the transporter moves on the rail, controlling lift-off of the low frequency eddy current probe from the rail head as the transporter moves on the rail.
  • 17. A method for detecting transverse cracks in rail head of a rail comprising:moving a transporter on the rail, generating a saturated magnetic field into and across the rail head with a DC saturation magnet mounted to the transporter a predetermined distance above the rail head while the transporter is moving, the saturation magnet having a toroidal-shape with opposing pole ends inwardly directed towards each other over the rail head, inductively coupling the opposing pole ends of the DC saturation magnet with the rail head, detecting transverse cracks in the rail head with a low frequency eddy current probe mounted centrally between the opposing pole ends of the DC saturation magnetic and over the rail head, applying a force to the low frequency eddy current probe against the rail head as the transporter moves on the rail, controlling lift-off of the low frequency eddy current probe from the rail head as the transporter moves on the rail, sensing non-relevant indications in the rail head with at least one separate sensor, rejecting a detected transverse crack by the low frequency eddy current probe when it corresponds to a sensed non-relevant indication by the at least one separate sensor.
  • 18. A method for detecting a transverse crack in rail head of a rail comprising:moving a transporter on the rail, generating a saturation magnetic field into and across the rail head with a DC saturation magnet mounted to the transporter a predetermined distance above the rail head while the transporter is moving, inductively coupling the opposing pole ends of the DC saturation magnet with the rail head, detecting a possible transverse crack in an area of the rail head with a low frequency eddy current probe mounted centrally between the opposing pole ends of the DC saturation magnetic and over said area of the rail head, sensing a non-relevant indication in said area of the rail head with at least one separate sensor, rejecting said possible detected transverse crack by the low frequency eddy current probe only when said a non-relevant indication is sensed by the at least one separate sensor for said area.
US Referenced Citations (11)
Number Name Date Kind
2049764 Drake Aug 1936 A
2884592 Wilson Apr 1959 A
3952315 Cecco Apr 1976 A
4510447 Moyer Apr 1985 A
4631688 Boehm et al. Dec 1986 A
4932618 Davenport et al. Jun 1990 A
5429329 Wallace et al. Jul 1995 A
5446382 Flora Aug 1995 A
5751144 Weischedel May 1998 A
5786750 Cooper Jul 1998 A
6102340 Peek et al. Aug 2000 A
Non-Patent Literature Citations (10)
Entry
ZETEC, Inc., MIZ-22 Dual-Frequency Eddy Current Instrument, 1993, 1 page (front of page).
ZETEC, Inc., MIZ-22 Dual-Frequency Eddy Current Instrument, 1993, 1 page (back of page).
Zarembski et al., TTCI, Technology Digest, Review of Rail-Testing Practices and Technologies—Overseas, Jun. 1998, 4 pages.
Garcia and Reiff, TTCI, Test and Evaluation of Rail Defect Detection Technologies, Sep. 1998, 17 pages.
Reiff and Garcia, TTCI, Increasing Cost Effectiveness of Continuous Welded Rail Through Improved Rail Defect Inspection, Oct. 19, 1998, 13 pages.
Garcia and Reiff, TTCI, Technology Digest, Rail-Defect Test Facility: A Tool of Evaluating Defect-Detection Technologies, Jan. 1999, 4 pages.
Stone et al., TTCI, Rail and Wheel Inspection Systems, Pueblo, CO, Jan. 6, 1999, 9 pages.
Garcia and Reiff, TTCI, Technology Digest, Benchmarking Results of North American Rail-Flaw Detection Technologies, Feb. 1999, 4 pages.
Earnest et al., Transverse Crack Detection in Rail Heads, Proceedings of: 6th NDE Topical Conference Apr. 19-22, 1999, San Antonio, Texas, 8 pages.
Transverse Crack Detection in Railroad Rails, 18 pages of presentation material given at the Proceedings of the 6th NDE Topical Conference, Apr. 19-22, 1999, San Antonio, Texas by Douglas R. Earnest.